diff options
Diffstat (limited to 'fs')
392 files changed, 18340 insertions, 10119 deletions
diff --git a/fs/9p/Kconfig b/fs/9p/Kconfig index 814ac4e..0a93dc1cb 100644 --- a/fs/9p/Kconfig +++ b/fs/9p/Kconfig @@ -1,6 +1,6 @@ config 9P_FS - tristate "Plan 9 Resource Sharing Support (9P2000) (Experimental)" - depends on INET && NET_9P && EXPERIMENTAL + tristate "Plan 9 Resource Sharing Support (9P2000)" + depends on INET && NET_9P help If you say Y here, you will get experimental support for Plan 9 resource sharing via the 9P2000 protocol. @@ -10,7 +10,6 @@ config 9P_FS If unsure, say N. if 9P_FS - config 9P_FSCACHE bool "Enable 9P client caching support (EXPERIMENTAL)" depends on EXPERIMENTAL diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 82a7c38..691c78f 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -259,7 +259,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, if (IS_ERR(inode_fid)) { err = PTR_ERR(inode_fid); mutex_unlock(&v9inode->v_mutex); - goto error; + goto err_clunk_old_fid; } v9inode->writeback_fid = (void *) inode_fid; } @@ -267,8 +267,8 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, /* Since we are opening a file, assign the open fid to the file */ filp = lookup_instantiate_filp(nd, dentry, generic_file_open); if (IS_ERR(filp)) { - p9_client_clunk(ofid); - return PTR_ERR(filp); + err = PTR_ERR(filp); + goto err_clunk_old_fid; } filp->private_data = ofid; #ifdef CONFIG_9P_FSCACHE @@ -278,10 +278,11 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, return 0; error: - if (ofid) - p9_client_clunk(ofid); if (fid) p9_client_clunk(fid); +err_clunk_old_fid: + if (ofid) + p9_client_clunk(ofid); return err; } @@ -47,7 +47,7 @@ config FS_POSIX_ACL def_bool n config EXPORTFS - bool + tristate config FILE_LOCKING bool "Enable POSIX file locking API" if EXPERT @@ -124,6 +124,7 @@ config TMPFS config TMPFS_POSIX_ACL bool "Tmpfs POSIX Access Control Lists" depends on TMPFS + select TMPFS_XATTR select GENERIC_ACL help POSIX Access Control Lists (ACLs) support permissions for users and @@ -134,6 +135,22 @@ config TMPFS_POSIX_ACL If you don't know what Access Control Lists are, say N. +config TMPFS_XATTR + bool "Tmpfs extended attributes" + depends on TMPFS + default n + help + Extended attributes are name:value pairs associated with inodes by + the kernel or by users (see the attr(5) manual page, or visit + <http://acl.bestbits.at/> for details). + + Currently this enables support for the trusted.* and + security.* namespaces. + + You need this for POSIX ACL support on tmpfs. + + If unsure, say N. + config HUGETLBFS bool "HugeTLB file system support" depends on X86 || IA64 || SPARC64 || (S390 && 64BIT) || \ diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 20c106f..1b0b195 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -584,11 +584,11 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, success: d_add(dentry, inode); - _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%llu }", + _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%u }", fid.vnode, fid.unique, dentry->d_inode->i_ino, - (unsigned long long)dentry->d_inode->i_version); + dentry->d_inode->i_generation); return NULL; } @@ -671,10 +671,10 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) * been deleted and replaced, and the original vnode ID has * been reused */ if (fid.unique != vnode->fid.unique) { - _debug("%s: file deleted (uq %u -> %u I:%llu)", + _debug("%s: file deleted (uq %u -> %u I:%u)", dentry->d_name.name, fid.unique, vnode->fid.unique, - (unsigned long long)dentry->d_inode->i_version); + dentry->d_inode->i_generation); spin_lock(&vnode->lock); set_bit(AFS_VNODE_DELETED, &vnode->flags); spin_unlock(&vnode->lock); diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index 4bd0218..346e328 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -89,7 +89,7 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, i_size_write(&vnode->vfs_inode, size); vnode->vfs_inode.i_uid = status->owner; vnode->vfs_inode.i_gid = status->group; - vnode->vfs_inode.i_version = vnode->fid.unique; + vnode->vfs_inode.i_generation = vnode->fid.unique; vnode->vfs_inode.i_nlink = status->nlink; mode = vnode->vfs_inode.i_mode; @@ -102,6 +102,7 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, vnode->vfs_inode.i_ctime.tv_sec = status->mtime_server; vnode->vfs_inode.i_mtime = vnode->vfs_inode.i_ctime; vnode->vfs_inode.i_atime = vnode->vfs_inode.i_ctime; + vnode->vfs_inode.i_version = data_version; } expected_version = status->data_version; diff --git a/fs/afs/inode.c b/fs/afs/inode.c index db66c52..0fdab6e 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -75,7 +75,8 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key) inode->i_ctime.tv_nsec = 0; inode->i_atime = inode->i_mtime = inode->i_ctime; inode->i_blocks = 0; - inode->i_version = vnode->fid.unique; + inode->i_generation = vnode->fid.unique; + inode->i_version = vnode->status.data_version; inode->i_mapping->a_ops = &afs_fs_aops; /* check to see whether a symbolic link is really a mountpoint */ @@ -100,7 +101,7 @@ static int afs_iget5_test(struct inode *inode, void *opaque) struct afs_iget_data *data = opaque; return inode->i_ino == data->fid.vnode && - inode->i_version == data->fid.unique; + inode->i_generation == data->fid.unique; } /* @@ -122,7 +123,7 @@ static int afs_iget5_set(struct inode *inode, void *opaque) struct afs_vnode *vnode = AFS_FS_I(inode); inode->i_ino = data->fid.vnode; - inode->i_version = data->fid.unique; + inode->i_generation = data->fid.unique; vnode->fid = data->fid; vnode->volume = data->volume; @@ -380,8 +381,7 @@ int afs_getattr(struct vfsmount *mnt, struct dentry *dentry, inode = dentry->d_inode; - _enter("{ ino=%lu v=%llu }", inode->i_ino, - (unsigned long long)inode->i_version); + _enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation); generic_fillattr(inode, stat); return 0; diff --git a/fs/afs/super.c b/fs/afs/super.c index fb240e8..356dcf0 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -31,8 +31,8 @@ static void afs_i_init_once(void *foo); static struct dentry *afs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data); +static void afs_kill_super(struct super_block *sb); static struct inode *afs_alloc_inode(struct super_block *sb); -static void afs_put_super(struct super_block *sb); static void afs_destroy_inode(struct inode *inode); static int afs_statfs(struct dentry *dentry, struct kstatfs *buf); @@ -40,7 +40,7 @@ struct file_system_type afs_fs_type = { .owner = THIS_MODULE, .name = "afs", .mount = afs_mount, - .kill_sb = kill_anon_super, + .kill_sb = afs_kill_super, .fs_flags = 0, }; @@ -50,7 +50,6 @@ static const struct super_operations afs_super_ops = { .drop_inode = afs_drop_inode, .destroy_inode = afs_destroy_inode, .evict_inode = afs_evict_inode, - .put_super = afs_put_super, .show_options = generic_show_options, }; @@ -282,19 +281,25 @@ static int afs_parse_device_name(struct afs_mount_params *params, */ static int afs_test_super(struct super_block *sb, void *data) { - struct afs_mount_params *params = data; + struct afs_super_info *as1 = data; struct afs_super_info *as = sb->s_fs_info; - return as->volume == params->volume; + return as->volume == as1->volume; +} + +static int afs_set_super(struct super_block *sb, void *data) +{ + sb->s_fs_info = data; + return set_anon_super(sb, NULL); } /* * fill in the superblock */ -static int afs_fill_super(struct super_block *sb, void *data) +static int afs_fill_super(struct super_block *sb, + struct afs_mount_params *params) { - struct afs_mount_params *params = data; - struct afs_super_info *as = NULL; + struct afs_super_info *as = sb->s_fs_info; struct afs_fid fid; struct dentry *root = NULL; struct inode *inode = NULL; @@ -302,23 +307,13 @@ static int afs_fill_super(struct super_block *sb, void *data) _enter(""); - /* allocate a superblock info record */ - as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); - if (!as) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } - - afs_get_volume(params->volume); - as->volume = params->volume; - /* fill in the superblock */ sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = AFS_FS_MAGIC; sb->s_op = &afs_super_ops; - sb->s_fs_info = as; sb->s_bdi = &as->volume->bdi; + strlcpy(sb->s_id, as->volume->vlocation->vldb.name, sizeof(sb->s_id)); /* allocate the root inode and dentry */ fid.vid = as->volume->vid; @@ -326,7 +321,7 @@ static int afs_fill_super(struct super_block *sb, void *data) fid.unique = 1; inode = afs_iget(sb, params->key, &fid, NULL, NULL); if (IS_ERR(inode)) - goto error_inode; + return PTR_ERR(inode); if (params->autocell) set_bit(AFS_VNODE_AUTOCELL, &AFS_FS_I(inode)->flags); @@ -342,16 +337,8 @@ static int afs_fill_super(struct super_block *sb, void *data) _leave(" = 0"); return 0; -error_inode: - ret = PTR_ERR(inode); - inode = NULL; error: iput(inode); - afs_put_volume(as->volume); - kfree(as); - - sb->s_fs_info = NULL; - _leave(" = %d", ret); return ret; } @@ -367,6 +354,7 @@ static struct dentry *afs_mount(struct file_system_type *fs_type, struct afs_volume *vol; struct key *key; char *new_opts = kstrdup(options, GFP_KERNEL); + struct afs_super_info *as; int ret; _enter(",,%s,%p", dev_name, options); @@ -399,12 +387,22 @@ static struct dentry *afs_mount(struct file_system_type *fs_type, ret = PTR_ERR(vol); goto error; } - params.volume = vol; + + /* allocate a superblock info record */ + as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); + if (!as) { + ret = -ENOMEM; + afs_put_volume(vol); + goto error; + } + as->volume = vol; /* allocate a deviceless superblock */ - sb = sget(fs_type, afs_test_super, set_anon_super, ¶ms); + sb = sget(fs_type, afs_test_super, afs_set_super, as); if (IS_ERR(sb)) { ret = PTR_ERR(sb); + afs_put_volume(vol); + kfree(as); goto error; } @@ -422,16 +420,16 @@ static struct dentry *afs_mount(struct file_system_type *fs_type, } else { _debug("reuse"); ASSERTCMP(sb->s_flags, &, MS_ACTIVE); + afs_put_volume(vol); + kfree(as); } - afs_put_volume(params.volume); afs_put_cell(params.cell); kfree(new_opts); _leave(" = 0 [%p]", sb); return dget(sb->s_root); error: - afs_put_volume(params.volume); afs_put_cell(params.cell); key_put(params.key); kfree(new_opts); @@ -439,18 +437,12 @@ error: return ERR_PTR(ret); } -/* - * finish the unmounting process on the superblock - */ -static void afs_put_super(struct super_block *sb) +static void afs_kill_super(struct super_block *sb) { struct afs_super_info *as = sb->s_fs_info; - - _enter(""); - + kill_anon_super(sb); afs_put_volume(as->volume); - - _leave(""); + kfree(as); } /* diff --git a/fs/afs/write.c b/fs/afs/write.c index 789b3af..b806285 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -84,23 +84,21 @@ void afs_put_writeback(struct afs_writeback *wb) * partly or wholly fill a page that's under preparation for writing */ static int afs_fill_page(struct afs_vnode *vnode, struct key *key, - loff_t pos, unsigned len, struct page *page) + loff_t pos, struct page *page) { loff_t i_size; - unsigned eof; int ret; + int len; - _enter(",,%llu,%u", (unsigned long long)pos, len); - - ASSERTCMP(len, <=, PAGE_CACHE_SIZE); + _enter(",,%llu", (unsigned long long)pos); i_size = i_size_read(&vnode->vfs_inode); - if (pos + len > i_size) - eof = i_size; + if (pos + PAGE_CACHE_SIZE > i_size) + len = i_size - pos; else - eof = PAGE_CACHE_SIZE; + len = PAGE_CACHE_SIZE; - ret = afs_vnode_fetch_data(vnode, key, 0, eof, page); + ret = afs_vnode_fetch_data(vnode, key, pos, len, page); if (ret < 0) { if (ret == -ENOENT) { _debug("got NOENT from server" @@ -153,9 +151,8 @@ int afs_write_begin(struct file *file, struct address_space *mapping, *pagep = page; /* page won't leak in error case: it eventually gets cleaned off LRU */ - if (!PageUptodate(page)) { - _debug("not up to date"); - ret = afs_fill_page(vnode, key, pos, len, page); + if (!PageUptodate(page) && len != PAGE_CACHE_SIZE) { + ret = afs_fill_page(vnode, key, index << PAGE_CACHE_SHIFT, page); if (ret < 0) { kfree(candidate); _leave(" = %d [prep]", ret); @@ -175,6 +175,13 @@ int notify_change(struct dentry * dentry, struct iattr * attr) return -EPERM; } + if ((ia_valid & ATTR_MODE)) { + mode_t amode = attr->ia_mode; + /* Flag setting protected by i_mutex */ + if (is_sxid(amode)) + inode->i_flags &= ~S_NOSEC; + } + now = current_fs_time(inode->i_sb); attr->ia_ctime = now; diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 9ad2369..bfcb18f 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -231,9 +231,6 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer, static int bad_inode_permission(struct inode *inode, int mask, unsigned int flags) { - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - return -EIO; } diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 397d305..1bffbe0 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -820,6 +820,8 @@ static int load_flat_shared_library(int id, struct lib_info *libs) int res; char buf[16]; + memset(&bprm, 0, sizeof(bprm)); + /* Create the file name */ sprintf(buf, "/lib/lib%d.so", id); @@ -835,6 +837,12 @@ static int load_flat_shared_library(int id, struct lib_info *libs) if (!bprm.cred) goto out; + /* We don't really care about recalculating credentials at this point + * as we're past the point of no return and are dealing with shared + * libraries. + */ + bprm.cred_prepared = 1; + res = prepare_binprm(&bprm); if (!IS_ERR_VALUE(res)) @@ -638,10 +638,11 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page * @offset: vec entry offset * * Attempt to add a page to the bio_vec maplist. This can fail for a - * number of reasons, such as the bio being full or target block - * device limitations. The target block device must allow bio's - * smaller than PAGE_SIZE, so it is always possible to add a single - * page to an empty bio. This should only be used by REQ_PC bios. + * number of reasons, such as the bio being full or target block device + * limitations. The target block device must allow bio's up to PAGE_SIZE, + * so it is always possible to add a single page to an empty bio. + * + * This should only be used by REQ_PC bios. */ int bio_add_pc_page(struct request_queue *q, struct bio *bio, struct page *page, unsigned int len, unsigned int offset) @@ -659,10 +660,9 @@ EXPORT_SYMBOL(bio_add_pc_page); * @offset: vec entry offset * * Attempt to add a page to the bio_vec maplist. This can fail for a - * number of reasons, such as the bio being full or target block - * device limitations. The target block device must allow bio's - * smaller than PAGE_SIZE, so it is always possible to add a single - * page to an empty bio. + * number of reasons, such as the bio being full or target block device + * limitations. The target block device must allow bio's up to PAGE_SIZE, + * so it is always possible to add a single page to an empty bio. */ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, unsigned int offset) diff --git a/fs/block_dev.c b/fs/block_dev.c index 257b00e..610e8e0 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -762,7 +762,19 @@ static struct block_device *bd_start_claiming(struct block_device *bdev, if (!disk) return ERR_PTR(-ENXIO); - whole = bdget_disk(disk, 0); + /* + * Normally, @bdev should equal what's returned from bdget_disk() + * if partno is 0; however, some drivers (floppy) use multiple + * bdev's for the same physical device and @bdev may be one of the + * aliases. Keep @bdev if partno is 0. This means claimer + * tracking is broken for those devices but it has always been that + * way. + */ + if (partno) + whole = bdget_disk(disk, 0); + else + whole = bdgrab(bdev); + module_put(disk->fops->owner); put_disk(disk); if (!whole) @@ -1120,6 +1132,15 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) goto restart; } } + + if (!ret && !bdev->bd_openers) { + bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); + bdi = blk_get_backing_dev_info(bdev); + if (bdi == NULL) + bdi = &default_backing_dev_info; + bdev_inode_switch_bdi(bdev->bd_inode, bdi); + } + /* * If the device is invalidated, rescan partition * if open succeeded or failed with -ENOMEDIUM. @@ -1130,14 +1151,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) rescan_partitions(disk, bdev); if (ret) goto out_clear; - - if (!bdev->bd_openers) { - bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); - bdi = blk_get_backing_dev_info(bdev); - if (bdi == NULL) - bdi = &default_backing_dev_info; - bdev_inode_switch_bdi(bdev->bd_inode, bdi); - } } else { struct block_device *whole; whole = bdget_disk(disk, 0); @@ -1237,6 +1250,8 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) res = __blkdev_get(bdev, mode, 0); if (whole) { + struct gendisk *disk = whole->bd_disk; + /* finish claiming */ mutex_lock(&bdev->bd_mutex); spin_lock(&bdev_lock); @@ -1263,15 +1278,16 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) spin_unlock(&bdev_lock); /* - * Block event polling for write claims. Any write - * holder makes the write_holder state stick until all - * are released. This is good enough and tracking - * individual writeable reference is too fragile given - * the way @mode is used in blkdev_get/put(). + * Block event polling for write claims if requested. Any + * write holder makes the write_holder state stick until + * all are released. This is good enough and tracking + * individual writeable reference is too fragile given the + * way @mode is used in blkdev_get/put(). */ - if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder) { + if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder && + (disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE)) { bdev->bd_write_holder = true; - disk_block_events(bdev->bd_disk); + disk_block_events(disk); } mutex_unlock(&bdev->bd_mutex); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 8e948ec..f30ac05 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -19,7 +19,6 @@ #ifndef __BTRFS_CTREE__ #define __BTRFS_CTREE__ -#include <linux/version.h> #include <linux/mm.h> #include <linux/highmem.h> #include <linux/fs.h> @@ -2549,7 +2548,7 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); int btrfs_readpage(struct file *file, struct page *page); void btrfs_evict_inode(struct inode *inode); int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc); -void btrfs_dirty_inode(struct inode *inode); +void btrfs_dirty_inode(struct inode *inode, int flags); struct inode *btrfs_alloc_inode(struct super_block *sb); void btrfs_destroy_inode(struct inode *inode); int btrfs_drop_inode(struct inode *inode); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index b181a94..7055d11 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -10,6 +10,8 @@ #include <linux/swap.h> #include <linux/writeback.h> #include <linux/pagevec.h> +#include <linux/prefetch.h> +#include <linux/cleancache.h> #include "extent_io.h" #include "extent_map.h" #include "compat.h" @@ -1969,6 +1971,13 @@ static int __extent_read_full_page(struct extent_io_tree *tree, set_page_extent_mapped(page); + if (!PageUptodate(page)) { + if (cleancache_get_page(page) == 0) { + BUG_ON(blocksize != PAGE_SIZE); + goto out; + } + } + end = page_end; while (1) { lock_extent(tree, start, end, GFP_NOFS); @@ -2102,6 +2111,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree, cur = cur + iosize; pg_offset += iosize; } +out: if (!nr) { if (!PageError(page)) SetPageUptodate(page); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 447612d..d340f63 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4299,7 +4299,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) * FIXME, needs more benchmarking...there are no reasons other than performance * to keep or drop this code. */ -void btrfs_dirty_inode(struct inode *inode) +void btrfs_dirty_inode(struct inode *inode, int flags) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 086b1e6..5e0a3dc 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -712,7 +712,7 @@ again: WARN_ON(cur->checked); if (!list_empty(&cur->upper)) { /* - * the backref was added previously when processsing + * the backref was added previously when processing * backref of type BTRFS_TREE_BLOCK_REF_KEY */ BUG_ON(!list_is_singular(&cur->upper)); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 3559d0b..0bb4ebb 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -39,6 +39,7 @@ #include <linux/miscdevice.h> #include <linux/magic.h> #include <linux/slab.h> +#include <linux/cleancache.h> #include "compat.h" #include "delayed-inode.h" #include "ctree.h" @@ -638,6 +639,7 @@ static int btrfs_fill_super(struct super_block *sb, sb->s_root = root_dentry; save_mount_options(sb, data); + cleancache_init_fs(sb); return 0; fail_close: @@ -823,7 +825,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, } else { char b[BDEVNAME_SIZE]; - s->s_flags = flags; + s->s_flags = flags | MS_NOSEC; strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); error = btrfs_fill_super(s, fs_devices, data, flags & MS_SILENT ? 1 : 0); diff --git a/fs/buffer.c b/fs/buffer.c index a08bb8e..1a80b04 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -41,6 +41,7 @@ #include <linux/bitops.h> #include <linux/mpage.h> #include <linux/bit_spinlock.h> +#include <linux/cleancache.h> static int fsync_buffers_list(spinlock_t *lock, struct list_head *list); @@ -269,6 +270,10 @@ void invalidate_bdev(struct block_device *bdev) invalidate_bh_lrus(); lru_add_drain_all(); /* make sure all lru add caches are flushed */ invalidate_mapping_pages(mapping, 0, -1); + /* 99% of the time, we don't need to flush the cleancache on the bdev. + * But, for the strange corners, lets be cautious + */ + cleancache_flush_inode(mapping); } EXPORT_SYMBOL(invalidate_bdev); @@ -1897,10 +1902,8 @@ int __block_write_begin(struct page *page, loff_t pos, unsigned len, if (!buffer_uptodate(*wait_bh)) err = -EIO; } - if (unlikely(err)) { + if (unlikely(err)) page_zero_new_buffers(page, from, to); - ClearPageUptodate(page); - } return err; } EXPORT_SYMBOL(__block_write_begin); @@ -2331,24 +2334,26 @@ EXPORT_SYMBOL(block_commit_write); * page lock we can determine safely if the page is beyond EOF. If it is not * beyond EOF, then the page is guaranteed safe against truncation until we * unlock the page. + * + * Direct callers of this function should call vfs_check_frozen() so that page + * fault does not busyloop until the fs is thawed. */ -int -block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, - get_block_t get_block) +int __block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, + get_block_t get_block) { struct page *page = vmf->page; struct inode *inode = vma->vm_file->f_path.dentry->d_inode; unsigned long end; loff_t size; - int ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */ + int ret; lock_page(page); size = i_size_read(inode); if ((page->mapping != inode->i_mapping) || (page_offset(page) > size)) { - /* page got truncated out from underneath us */ - unlock_page(page); - goto out; + /* We overload EFAULT to mean page got truncated */ + ret = -EFAULT; + goto out_unlock; } /* page is wholly or partially inside EOF */ @@ -2361,18 +2366,42 @@ block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, if (!ret) ret = block_commit_write(page, 0, end); - if (unlikely(ret)) { - unlock_page(page); - if (ret == -ENOMEM) - ret = VM_FAULT_OOM; - else /* -ENOSPC, -EIO, etc */ - ret = VM_FAULT_SIGBUS; - } else - ret = VM_FAULT_LOCKED; - -out: + if (unlikely(ret < 0)) + goto out_unlock; + /* + * Freezing in progress? We check after the page is marked dirty and + * with page lock held so if the test here fails, we are sure freezing + * code will wait during syncing until the page fault is done - at that + * point page will be dirty and unlocked so freezing code will write it + * and writeprotect it again. + */ + set_page_dirty(page); + if (inode->i_sb->s_frozen != SB_UNFROZEN) { + ret = -EAGAIN; + goto out_unlock; + } + wait_on_page_writeback(page); + return 0; +out_unlock: + unlock_page(page); return ret; } +EXPORT_SYMBOL(__block_page_mkwrite); + +int block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf, + get_block_t get_block) +{ + int ret; + struct super_block *sb = vma->vm_file->f_path.dentry->d_inode->i_sb; + + /* + * This check is racy but catches the common case. The check in + * __block_page_mkwrite() is reliable. + */ + vfs_check_frozen(sb, SB_FREEZE_WRITE); + ret = __block_page_mkwrite(vma, vmf, get_block); + return block_page_mkwrite_return(ret); +} EXPORT_SYMBOL(block_page_mkwrite); /* diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 38b8ab5..5a3953d 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -453,7 +453,7 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc) int err; struct inode *inode = page->mapping->host; BUG_ON(!inode); - igrab(inode); + ihold(inode); err = writepage_nounlock(page, wbc); unlock_page(page); iput(inode); @@ -848,7 +848,8 @@ get_more_pages: op->payload_len = cpu_to_le32(len); req->r_request->hdr.data_len = cpu_to_le32(len); - ceph_osdc_start_request(&fsc->client->osdc, req, true); + rc = ceph_osdc_start_request(&fsc->client->osdc, req, true); + BUG_ON(rc); req = NULL; /* continue? */ @@ -880,8 +881,6 @@ release_pvec_pages: out: if (req) ceph_osdc_put_request(req); - if (rc > 0) - rc = 0; /* vfs expects us to return 0 */ ceph_put_snap_context(snapc); dout("writepages done, rc = %d\n", rc); return rc; diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 2a5404c..f605753 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -569,7 +569,8 @@ retry: list_add_tail(&cap->session_caps, &session->s_caps); session->s_nr_caps++; spin_unlock(&session->s_cap_lock); - } + } else if (new_cap) + ceph_put_cap(mdsc, new_cap); if (!ci->i_snap_realm) { /* @@ -2634,6 +2635,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, struct ceph_mds_session *session, int *open_target_sessions) { + struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; unsigned mseq = le32_to_cpu(ex->migrate_seq); @@ -2670,6 +2672,19 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, * export targets, so that we get the matching IMPORT */ *open_target_sessions = 1; + + /* + * we can't flush dirty caps that we've seen the + * EXPORT but no IMPORT for + */ + spin_lock(&mdsc->cap_dirty_lock); + if (!list_empty(&ci->i_dirty_item)) { + dout(" moving %p to cap_dirty_migrating\n", + inode); + list_move(&ci->i_dirty_item, + &mdsc->cap_dirty_migrating); + } + spin_unlock(&mdsc->cap_dirty_lock); } __ceph_remove_cap(cap); } @@ -2707,6 +2722,13 @@ static void handle_cap_import(struct ceph_mds_client *mdsc, ci->i_cap_exporting_issued = 0; ci->i_cap_exporting_mseq = 0; ci->i_cap_exporting_mds = -1; + + spin_lock(&mdsc->cap_dirty_lock); + if (!list_empty(&ci->i_dirty_item)) { + dout(" moving %p back to cap_dirty\n", inode); + list_move(&ci->i_dirty_item, &mdsc->cap_dirty); + } + spin_unlock(&mdsc->cap_dirty_lock); } else { dout("handle_cap_import inode %p ci %p mds%d mseq %d\n", inode, ci, mds, mseq); @@ -2910,47 +2932,24 @@ void ceph_check_delayed_caps(struct ceph_mds_client *mdsc) */ void ceph_flush_dirty_caps(struct ceph_mds_client *mdsc) { - struct ceph_inode_info *ci, *nci = NULL; - struct inode *inode, *ninode = NULL; - struct list_head *p, *n; + struct ceph_inode_info *ci; + struct inode *inode; dout("flush_dirty_caps\n"); spin_lock(&mdsc->cap_dirty_lock); - list_for_each_safe(p, n, &mdsc->cap_dirty) { - if (nci) { - ci = nci; - inode = ninode; - ci->i_ceph_flags &= ~CEPH_I_NOFLUSH; - dout("flush_dirty_caps inode %p (was next inode)\n", - inode); - } else { - ci = list_entry(p, struct ceph_inode_info, - i_dirty_item); - inode = igrab(&ci->vfs_inode); - BUG_ON(!inode); - dout("flush_dirty_caps inode %p\n", inode); - } - if (n != &mdsc->cap_dirty) { - nci = list_entry(n, struct ceph_inode_info, - i_dirty_item); - ninode = igrab(&nci->vfs_inode); - BUG_ON(!ninode); - nci->i_ceph_flags |= CEPH_I_NOFLUSH; - dout("flush_dirty_caps next inode %p, noflush\n", - ninode); - } else { - nci = NULL; - ninode = NULL; - } + while (!list_empty(&mdsc->cap_dirty)) { + ci = list_first_entry(&mdsc->cap_dirty, struct ceph_inode_info, + i_dirty_item); + inode = &ci->vfs_inode; + ihold(inode); + dout("flush_dirty_caps %p\n", inode); spin_unlock(&mdsc->cap_dirty_lock); - if (inode) { - ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH, - NULL); - iput(inode); - } + ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH, NULL); + iput(inode); spin_lock(&mdsc->cap_dirty_lock); } spin_unlock(&mdsc->cap_dirty_lock); + dout("flush_dirty_caps done\n"); } /* diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 1a867a3..ef8f08c 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -308,7 +308,8 @@ more: req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_dentry = dget(filp->f_dentry); /* hints to request -> mds selection code */ req->r_direct_mode = USE_AUTH_MDS; @@ -360,7 +361,7 @@ more: rinfo = &fi->last_readdir->r_reply_info; dout("readdir frag %x num %d off %d chunkoff %d\n", frag, rinfo->dir_nr, off, fi->offset); - while (off - fi->offset >= 0 && off - fi->offset < rinfo->dir_nr) { + while (off >= fi->offset && off - fi->offset < rinfo->dir_nr) { u64 pos = ceph_make_fpos(frag, off); struct ceph_mds_reply_inode *in = rinfo->dir_in[off - fi->offset].in; @@ -787,10 +788,12 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir, req->r_dentry_drop = CEPH_CAP_FILE_SHARED; req->r_dentry_unless = CEPH_CAP_FILE_EXCL; err = ceph_mdsc_do_request(mdsc, dir, req); - if (err) + if (err) { d_drop(dentry); - else if (!req->r_reply_info.head->is_dentry) - d_instantiate(dentry, igrab(old_dentry->d_inode)); + } else if (!req->r_reply_info.head->is_dentry) { + ihold(old_dentry->d_inode); + d_instantiate(dentry, old_dentry->d_inode); + } ceph_mdsc_put_request(req); return err; } @@ -1066,16 +1069,17 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size, struct inode *inode = file->f_dentry->d_inode; struct ceph_inode_info *ci = ceph_inode(inode); int left; + const int bufsize = 1024; if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT)) return -EISDIR; if (!cf->dir_info) { - cf->dir_info = kmalloc(1024, GFP_NOFS); + cf->dir_info = kmalloc(bufsize, GFP_NOFS); if (!cf->dir_info) return -ENOMEM; cf->dir_info_len = - sprintf(cf->dir_info, + snprintf(cf->dir_info, bufsize, "entries: %20lld\n" " files: %20lld\n" " subdirs: %20lld\n" diff --git a/fs/ceph/export.c b/fs/ceph/export.c index e410561..f67b687 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -86,6 +86,7 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len, static struct dentry *__fh_to_dentry(struct super_block *sb, struct ceph_nfs_fh *fh) { + struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc; struct inode *inode; struct dentry *dentry; struct ceph_vino vino; @@ -95,8 +96,24 @@ static struct dentry *__fh_to_dentry(struct super_block *sb, vino.ino = fh->ino; vino.snap = CEPH_NOSNAP; inode = ceph_find_inode(sb, vino); - if (!inode) - return ERR_PTR(-ESTALE); + if (!inode) { + struct ceph_mds_request *req; + + req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPINO, + USE_ANY_MDS); + if (IS_ERR(req)) + return ERR_CAST(req); + + req->r_ino1 = vino; + req->r_num_caps = 1; + err = ceph_mdsc_do_request(mdsc, NULL, req); + inode = req->r_target_inode; + if (inode) + ihold(inode); + ceph_mdsc_put_request(req); + if (!inode) + return ERR_PTR(-ESTALE); + } dentry = d_obtain_alias(inode); if (IS_ERR(dentry)) { @@ -148,8 +165,10 @@ static struct dentry *__cfh_to_dentry(struct super_block *sb, snprintf(req->r_path2, 16, "%d", cfh->parent_name_hash); req->r_num_caps = 1; err = ceph_mdsc_do_request(mdsc, NULL, req); + inode = req->r_target_inode; + if (inode) + ihold(inode); ceph_mdsc_put_request(req); - inode = ceph_find_inode(sb, vino); if (!inode) return ERR_PTR(err ? err : -ESTALE); } diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 203252d..9542f07 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -191,7 +191,8 @@ int ceph_open(struct inode *inode, struct file *file) err = PTR_ERR(req); goto out; } - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_num_caps = 1; err = ceph_mdsc_do_request(mdsc, parent_inode, req); if (!err) @@ -282,7 +283,7 @@ int ceph_release(struct inode *inode, struct file *file) static int striped_read(struct inode *inode, u64 off, u64 len, struct page **pages, int num_pages, - int *checkeof, bool align_to_pages, + int *checkeof, bool o_direct, unsigned long buf_align) { struct ceph_fs_client *fsc = ceph_inode_to_client(inode); @@ -307,7 +308,7 @@ static int striped_read(struct inode *inode, io_align = off & ~PAGE_MASK; more: - if (align_to_pages) + if (o_direct) page_align = (pos - io_align + buf_align) & ~PAGE_MASK; else page_align = pos & ~PAGE_MASK; @@ -317,10 +318,10 @@ more: ci->i_truncate_seq, ci->i_truncate_size, page_pos, pages_left, page_align); - hit_stripe = this_len < left; - was_short = ret >= 0 && ret < this_len; if (ret == -ENOENT) ret = 0; + hit_stripe = this_len < left; + was_short = ret >= 0 && ret < this_len; dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read, ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : ""); @@ -345,20 +346,22 @@ more: } if (was_short) { - /* was original extent fully inside i_size? */ - if (pos + left <= inode->i_size) { - dout("zero tail\n"); - ceph_zero_page_vector_range(page_off + read, len - read, + /* did we bounce off eof? */ + if (pos + left > inode->i_size) + *checkeof = 1; + + /* zero trailing bytes (inside i_size) */ + if (left > 0 && pos < inode->i_size) { + if (pos + left > inode->i_size) + left = inode->i_size - pos; + + dout("zero tail %d\n", left); + ceph_zero_page_vector_range(page_off + read, left, pages); - read = len; - goto out; + read += left; } - - /* check i_size */ - *checkeof = 1; } -out: if (ret >= 0) ret = read; dout("striped_read returns %d\n", ret); @@ -658,7 +661,7 @@ out: /* hit EOF or hole? */ if (statret == 0 && *ppos < inode->i_size) { - dout("aio_read sync_read hit hole, reading more\n"); + dout("aio_read sync_read hit hole, ppos %lld < size %lld, reading more\n", *ppos, inode->i_size); read += ret; base += ret; len -= ret; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 70b6a48..d8858e9 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -1101,10 +1101,10 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, goto done; } req->r_dentry = dn; /* may have spliced */ - igrab(in); + ihold(in); } else if (ceph_ino(in) == vino.ino && ceph_snap(in) == vino.snap) { - igrab(in); + ihold(in); } else { dout(" %p links to %p %llx.%llx, not %llx.%llx\n", dn, in, ceph_ino(in), ceph_snap(in), @@ -1144,7 +1144,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, goto done; } req->r_dentry = dn; /* may have spliced */ - igrab(in); + ihold(in); rinfo->head->is_dentry = 1; /* fool notrace handlers */ } @@ -1328,7 +1328,7 @@ void ceph_queue_writeback(struct inode *inode) if (queue_work(ceph_inode_to_client(inode)->wb_wq, &ceph_inode(inode)->i_wb_work)) { dout("ceph_queue_writeback %p\n", inode); - igrab(inode); + ihold(inode); } else { dout("ceph_queue_writeback %p failed\n", inode); } @@ -1353,7 +1353,7 @@ void ceph_queue_invalidate(struct inode *inode) if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq, &ceph_inode(inode)->i_pg_inv_work)) { dout("ceph_queue_invalidate %p\n", inode); - igrab(inode); + ihold(inode); } else { dout("ceph_queue_invalidate %p failed\n", inode); } @@ -1477,7 +1477,7 @@ void ceph_queue_vmtruncate(struct inode *inode) if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq, &ci->i_vmtruncate_work)) { dout("ceph_queue_vmtruncate %p\n", inode); - igrab(inode); + ihold(inode); } else { dout("ceph_queue_vmtruncate %p failed, pending=%d\n", inode, ci->i_truncate_pending); @@ -1738,7 +1738,8 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) __mark_inode_dirty(inode, inode_dirty_flags); if (mask) { - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_inode_drop = release; req->r_args.setattr.mask = cpu_to_le32(mask); req->r_num_caps = 1; @@ -1779,7 +1780,8 @@ int ceph_do_getattr(struct inode *inode, int mask) req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS); if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_num_caps = 1; req->r_args.getattr.mask = cpu_to_le32(mask); err = ceph_mdsc_do_request(mdsc, NULL, req); diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index 8888c9ba..ef0b5f4 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -73,7 +73,8 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg) USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_inode_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_EXCL; req->r_args.setlayout.layout.fl_stripe_unit = @@ -135,7 +136,8 @@ static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg) if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_args.setlayout.layout.fl_stripe_unit = cpu_to_le32(l.stripe_unit); diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index 476b329..80576d05 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -23,7 +23,8 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); /* mds requires start and length rather than start and end */ if (LLONG_MAX == fl->fl_end) @@ -32,11 +33,10 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, length = fl->fl_end - fl->fl_start + 1; dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, " - "length: %llu, wait: %d, type`: %d", (int)lock_type, + "length: %llu, wait: %d, type: %d", (int)lock_type, (int)operation, (u64)fl->fl_pid, fl->fl_start, length, wait, fl->fl_type); - req->r_args.filelock_change.rule = lock_type; req->r_args.filelock_change.type = cmd; req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid); @@ -70,7 +70,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, } ceph_mdsc_put_request(req); dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, " - "length: %llu, wait: %d, type`: %d, err code %d", (int)lock_type, + "length: %llu, wait: %d, type: %d, err code %d", (int)lock_type, (int)operation, (u64)fl->fl_pid, fl->fl_start, length, wait, fl->fl_type, err); return err; @@ -109,16 +109,20 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) dout("mds locked, locking locally"); err = posix_lock_file(file, fl, NULL); if (err && (CEPH_MDS_OP_SETFILELOCK == op)) { - /* undo! This should only happen if the kernel detects - * local deadlock. */ + /* undo! This should only happen if + * the kernel detects local + * deadlock. */ ceph_lock_message(CEPH_LOCK_FCNTL, op, file, CEPH_LOCK_UNLOCK, 0, fl); - dout("got %d on posix_lock_file, undid lock", err); + dout("got %d on posix_lock_file, undid lock", + err); } } - } else { - dout("mds returned error code %d", err); + } else if (err == -ERESTARTSYS) { + dout("undoing lock\n"); + ceph_lock_message(CEPH_LOCK_FCNTL, op, file, + CEPH_LOCK_UNLOCK, 0, fl); } return err; } @@ -155,8 +159,11 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) file, CEPH_LOCK_UNLOCK, 0, fl); dout("got %d on flock_lock_file_wait, undid lock", err); } - } else { - dout("mds error code %d", err); + } else if (err == -ERESTARTSYS) { + dout("undoing lock\n"); + ceph_lock_message(CEPH_LOCK_FLOCK, + CEPH_MDS_OP_SETFILELOCK, + file, CEPH_LOCK_UNLOCK, 0, fl); } return err; } diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index d0fae4c..79743d1 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -578,6 +578,7 @@ static void __register_request(struct ceph_mds_client *mdsc, if (dir) { struct ceph_inode_info *ci = ceph_inode(dir); + ihold(dir); spin_lock(&ci->i_unsafe_lock); req->r_unsafe_dir = dir; list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops); @@ -598,6 +599,9 @@ static void __unregister_request(struct ceph_mds_client *mdsc, spin_lock(&ci->i_unsafe_lock); list_del_init(&req->r_unsafe_dir_item); spin_unlock(&ci->i_unsafe_lock); + + iput(req->r_unsafe_dir); + req->r_unsafe_dir = NULL; } ceph_mdsc_put_request(req); @@ -2691,7 +2695,6 @@ static void handle_lease(struct ceph_mds_client *mdsc, { struct super_block *sb = mdsc->fsc->sb; struct inode *inode; - struct ceph_inode_info *ci; struct dentry *parent, *dentry; struct ceph_dentry_info *di; int mds = session->s_mds; @@ -2728,7 +2731,6 @@ static void handle_lease(struct ceph_mds_client *mdsc, dout("handle_lease no inode %llx\n", vino.ino); goto release; } - ci = ceph_inode(inode); /* dentry */ parent = d_find_alias(inode); @@ -3002,6 +3004,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) spin_lock_init(&mdsc->snap_flush_lock); mdsc->cap_flush_seq = 0; INIT_LIST_HEAD(&mdsc->cap_dirty); + INIT_LIST_HEAD(&mdsc->cap_dirty_migrating); mdsc->num_cap_flushing = 0; spin_lock_init(&mdsc->cap_dirty_lock); init_waitqueue_head(&mdsc->cap_flushing_wq); diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 4e3a9cc..7d8a0d6 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -278,6 +278,7 @@ struct ceph_mds_client { u64 cap_flush_seq; struct list_head cap_dirty; /* inodes with dirty caps */ + struct list_head cap_dirty_migrating; /* ...that are migration... */ int num_cap_flushing; /* # caps we are flushing */ spinlock_t cap_dirty_lock; /* protects above items */ wait_queue_head_t cap_flushing_wq; diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index 24067d6..54b14de 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -722,7 +722,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc) ci = list_first_entry(&mdsc->snap_flush_list, struct ceph_inode_info, i_snap_flush_item); inode = &ci->vfs_inode; - igrab(inode); + ihold(inode); spin_unlock(&mdsc->snap_flush_lock); spin_lock(&inode->i_lock); __ceph_flush_snaps(ci, &session, 0); diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index f2b6286..f42d730 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -665,7 +665,8 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name, err = PTR_ERR(req); goto out; } - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_inode_drop = CEPH_CAP_XATTR_SHARED; req->r_num_caps = 1; req->r_args.setxattr.flags = cpu_to_le32(flags); @@ -795,7 +796,8 @@ static int ceph_send_removexattr(struct dentry *dentry, const char *name) USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); - req->r_inode = igrab(inode); + req->r_inode = inode; + ihold(inode); req->r_inode_drop = CEPH_CAP_XATTR_SHARED; req->r_num_caps = 1; req->r_path2 = kstrdup(name, GFP_NOFS); diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 7cb0f7f..f66cc16 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -7,6 +7,8 @@ config CIFS select CRYPTO_MD5 select CRYPTO_HMAC select CRYPTO_ARC4 + select CRYPTO_ECB + select CRYPTO_DES help This is the client VFS module for the Common Internet File System (CIFS) protocol which is the successor to the Server Message Block @@ -147,21 +149,13 @@ config CIFS_FSCACHE config CIFS_ACL bool "Provide CIFS ACL support (EXPERIMENTAL)" - depends on EXPERIMENTAL && CIFS_XATTR + depends on EXPERIMENTAL && CIFS_XATTR && KEYS help Allows to fetch CIFS/NTFS ACL from the server. The DACL blob is handed over to the application/caller. -config CIFS_EXPERIMENTAL - bool "CIFS Experimental Features (EXPERIMENTAL)" - depends on CIFS && EXPERIMENTAL +config CIFS_NFSD_EXPORT + bool "Allow nfsd to export CIFS file system (EXPERIMENTAL)" + depends on CIFS && EXPERIMENTAL && BROKEN help - Enables cifs features under testing. These features are - experimental and currently include DFS support and directory - change notification ie fcntl(F_DNOTIFY), as well as the upcall - mechanism which will be used for Kerberos session negotiation - and uid remapping. Some of these features also may depend on - setting a value of 1 to the pseudo-file /proc/fs/cifs/Experimental - (which is disabled by default). See the file fs/cifs/README - for more details. If unsure, say N. - + Allows NFS server to export a CIFS mounted share (nfsd over cifs) diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index d875584..005d524 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -4,7 +4,7 @@ obj-$(CONFIG_CIFS) += cifs.o cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \ - link.o misc.o netmisc.o smbdes.o smbencrypt.o transport.o asn1.o \ + link.o misc.o netmisc.o smbencrypt.o transport.o asn1.o \ cifs_unicode.o nterr.o xattr.o cifsencrypt.o \ readdir.o ioctl.o sess.o export.o diff --git a/fs/cifs/README b/fs/cifs/README index 74ab165f..c5c2c5e 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -457,6 +457,9 @@ A partial list of the supported mount options follows: otherwise - read from the server. All written data are stored in the cache, but if the client doesn't have Exclusive Oplock, it writes the data to the server. + rwpidforward Forward pid of a process who opened a file to any read or write + operation on that file. This prevent applications like WINE + from failing on read and write if we use mandatory brlock style. acl Allow setfacl and getfacl to manage posix ACLs if server supports them. (default) noacl Do not allow setfacl and getfacl calls on this mount @@ -704,18 +707,6 @@ the start of smb requests and responses can be enabled via: echo 1 > /proc/fs/cifs/traceSMB -Two other experimental features are under development. To test these -requires enabling CONFIG_CIFS_EXPERIMENTAL - - cifsacl support needed to retrieve approximated mode bits based on - the contents on the CIFS ACL. - - lease support: cifs will check the oplock state before calling into - the vfs to see if we can grant a lease on a file. - - DNOTIFY fcntl: needed for support of directory change - notification and perhaps later for file leases) - Per share (per client mount) statistics are available in /proc/fs/cifs/Stats if the kernel was configured with cifs statistics enabled. The statistics represent the number of successful (ie non-zero return code from the server) diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c index 53d57a3..545509c 100644 --- a/fs/cifs/cache.c +++ b/fs/cifs/cache.c @@ -92,7 +92,7 @@ static uint16_t cifs_server_get_key(const void *cookie_netfs_data, break; default: - cERROR(1, "CIFS: Unknown network family '%d'", sa->sa_family); + cERROR(1, "Unknown network family '%d'", sa->sa_family); key_len = 0; break; } @@ -146,13 +146,13 @@ static char *extract_sharename(const char *treename) static uint16_t cifs_super_get_key(const void *cookie_netfs_data, void *buffer, uint16_t maxbuf) { - const struct cifsTconInfo *tcon = cookie_netfs_data; + const struct cifs_tcon *tcon = cookie_netfs_data; char *sharename; uint16_t len; sharename = extract_sharename(tcon->treeName); if (IS_ERR(sharename)) { - cFYI(1, "CIFS: couldn't extract sharename\n"); + cFYI(1, "%s: couldn't extract sharename\n", __func__); sharename = NULL; return 0; } @@ -173,7 +173,7 @@ cifs_fscache_super_get_aux(const void *cookie_netfs_data, void *buffer, uint16_t maxbuf) { struct cifs_fscache_super_auxdata auxdata; - const struct cifsTconInfo *tcon = cookie_netfs_data; + const struct cifs_tcon *tcon = cookie_netfs_data; memset(&auxdata, 0, sizeof(auxdata)); auxdata.resource_id = tcon->resource_id; @@ -192,7 +192,7 @@ fscache_checkaux cifs_fscache_super_check_aux(void *cookie_netfs_data, uint16_t datalen) { struct cifs_fscache_super_auxdata auxdata; - const struct cifsTconInfo *tcon = cookie_netfs_data; + const struct cifs_tcon *tcon = cookie_netfs_data; if (datalen != sizeof(auxdata)) return FSCACHE_CHECKAUX_OBSOLETE; @@ -302,7 +302,7 @@ static void cifs_fscache_inode_now_uncached(void *cookie_netfs_data) pagevec_init(&pvec, 0); first = 0; - cFYI(1, "cifs inode 0x%p now uncached", cifsi); + cFYI(1, "%s: cifs inode 0x%p now uncached", __func__, cifsi); for (;;) { nr_pages = pagevec_lookup(&pvec, diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 30d01bc..2fe3cf1 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -63,7 +63,7 @@ void cifs_dump_detail(struct smb_hdr *smb) cERROR(1, "Cmd: %d Err: 0x%x Flags: 0x%x Flgs2: 0x%x Mid: %d Pid: %d", smb->Command, smb->Status.CifsError, smb->Flags, smb->Flags2, smb->Mid, smb->Pid); - cERROR(1, "smb buf %p len %d", smb, smbCalcSize_LE(smb)); + cERROR(1, "smb buf %p len %d", smb, smbCalcSize(smb)); } @@ -110,8 +110,8 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) struct list_head *tmp1, *tmp2, *tmp3; struct mid_q_entry *mid_entry; struct TCP_Server_Info *server; - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon; + struct cifs_ses *ses; + struct cifs_tcon *tcon; int i, j; __u32 dev_type; @@ -152,7 +152,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) tcp_ses_list); i++; list_for_each(tmp2, &server->smb_ses_list) { - ses = list_entry(tmp2, struct cifsSesInfo, + ses = list_entry(tmp2, struct cifs_ses, smb_ses_list); if ((ses->serverDomain == NULL) || (ses->serverOS == NULL) || @@ -171,7 +171,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) seq_printf(m, "TCP status: %d\n\tLocal Users To " "Server: %d SecMode: 0x%x Req On Wire: %d", server->tcpStatus, server->srv_count, - server->secMode, + server->sec_mode, atomic_read(&server->inFlight)); #ifdef CONFIG_CIFS_STATS2 @@ -183,7 +183,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) seq_puts(m, "\n\tShares:"); j = 0; list_for_each(tmp3, &ses->tcon_list) { - tcon = list_entry(tmp3, struct cifsTconInfo, + tcon = list_entry(tmp3, struct cifs_tcon, tcon_list); ++j; dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType); @@ -256,8 +256,8 @@ static ssize_t cifs_stats_proc_write(struct file *file, int rc; struct list_head *tmp1, *tmp2, *tmp3; struct TCP_Server_Info *server; - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon; + struct cifs_ses *ses; + struct cifs_tcon *tcon; rc = get_user(c, buffer); if (rc) @@ -273,11 +273,11 @@ static ssize_t cifs_stats_proc_write(struct file *file, server = list_entry(tmp1, struct TCP_Server_Info, tcp_ses_list); list_for_each(tmp2, &server->smb_ses_list) { - ses = list_entry(tmp2, struct cifsSesInfo, + ses = list_entry(tmp2, struct cifs_ses, smb_ses_list); list_for_each(tmp3, &ses->tcon_list) { tcon = list_entry(tmp3, - struct cifsTconInfo, + struct cifs_tcon, tcon_list); atomic_set(&tcon->num_smbs_sent, 0); atomic_set(&tcon->num_writes, 0); @@ -312,8 +312,8 @@ static int cifs_stats_proc_show(struct seq_file *m, void *v) int i; struct list_head *tmp1, *tmp2, *tmp3; struct TCP_Server_Info *server; - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon; + struct cifs_ses *ses; + struct cifs_tcon *tcon; seq_printf(m, "Resources in use\nCIFS Session: %d\n", @@ -346,11 +346,11 @@ static int cifs_stats_proc_show(struct seq_file *m, void *v) server = list_entry(tmp1, struct TCP_Server_Info, tcp_ses_list); list_for_each(tmp2, &server->smb_ses_list) { - ses = list_entry(tmp2, struct cifsSesInfo, + ses = list_entry(tmp2, struct cifs_ses, smb_ses_list); list_for_each(tmp3, &ses->tcon_list) { tcon = list_entry(tmp3, - struct cifsTconInfo, + struct cifs_tcon, tcon_list); i++; seq_printf(m, "\n%d) %s", i, tcon->treeName); diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 2b68ac5..8d8f28c 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -272,7 +272,7 @@ static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt) struct dfs_info3_param *referrals = NULL; unsigned int num_referrals = 0; struct cifs_sb_info *cifs_sb; - struct cifsSesInfo *ses; + struct cifs_ses *ses; char *full_path; int xid, i; int rc; diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index ac51cd2..7260e11 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -41,6 +41,8 @@ #define CIFS_MOUNT_MF_SYMLINKS 0x10000 /* Minshall+French Symlinks enabled */ #define CIFS_MOUNT_MULTIUSER 0x20000 /* multiuser mount */ #define CIFS_MOUNT_STRICT_IO 0x40000 /* strict cache mode */ +#define CIFS_MOUNT_RWPIDFORWARD 0x80000 /* use pid forwarding for rw */ +#define CIFS_MOUNT_POSIXACL 0x100000 /* mirror of MS_POSIXACL in mnt_cifs_flags */ struct cifs_sb_info { struct rb_root tlink_tree; @@ -56,11 +58,7 @@ struct cifs_sb_info { mode_t mnt_file_mode; mode_t mnt_dir_mode; unsigned int mnt_cifs_flags; - int prepathlen; - char *prepath; /* relative path under the share to mount to */ -#ifdef CONFIG_CIFS_DFS_UPCALL - char *mountdata; /* mount options received at mount time */ -#endif + char *mountdata; /* options received at mount time or via DFS refs */ struct backing_dev_info bdi; struct delayed_work prune_tlinks; }; diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 33d2213..2272fd5 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -95,7 +95,7 @@ struct key_type cifs_spnego_key_type = { /* get a key struct with a SPNEGO security blob, suitable for session setup */ struct key * -cifs_get_spnego_key(struct cifsSesInfo *sesInfo) +cifs_get_spnego_key(struct cifs_ses *sesInfo) { struct TCP_Server_Info *server = sesInfo->server; struct sockaddr_in *sa = (struct sockaddr_in *) &server->dstaddr; diff --git a/fs/cifs/cifs_spnego.h b/fs/cifs/cifs_spnego.h index e4041ec..31bef9e 100644 --- a/fs/cifs/cifs_spnego.h +++ b/fs/cifs/cifs_spnego.h @@ -41,7 +41,7 @@ struct cifs_spnego_msg { #ifdef __KERNEL__ extern struct key_type cifs_spnego_key_type; -extern struct key *cifs_get_spnego_key(struct cifsSesInfo *sesInfo); +extern struct key *cifs_get_spnego_key(struct cifs_ses *sesInfo); #endif /* KERNEL */ #endif /* _CIFS_SPNEGO_H */ diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h index 644dd882..6d02fd5 100644 --- a/fs/cifs/cifs_unicode.h +++ b/fs/cifs/cifs_unicode.h @@ -82,6 +82,9 @@ int cifs_strtoUCS(__le16 *, const char *, int, const struct nls_table *); char *cifs_strndup_from_ucs(const char *src, const int maxlen, const bool is_unicode, const struct nls_table *codepage); +extern int cifsConvertToUCS(__le16 *target, const char *source, int maxlen, + const struct nls_table *cp, int mapChars); + #endif /* diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index beeebf1..21de1d6 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -23,77 +23,405 @@ #include <linux/fs.h> #include <linux/slab.h> +#include <linux/string.h> +#include <linux/keyctl.h> +#include <linux/key-type.h> +#include <keys/user-type.h> #include "cifspdu.h" #include "cifsglob.h" #include "cifsacl.h" #include "cifsproto.h" #include "cifs_debug.h" - -static struct cifs_wksid wksidarr[NUM_WK_SIDS] = { - {{1, 0, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0, 0} }, "null user"}, - {{1, 1, {0, 0, 0, 0, 0, 1}, {0, 0, 0, 0, 0} }, "nobody"}, - {{1, 1, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(11), 0, 0, 0, 0} }, "net-users"}, - {{1, 1, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(18), 0, 0, 0, 0} }, "sys"}, - {{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(544), 0, 0, 0} }, "root"}, - {{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(545), 0, 0, 0} }, "users"}, - {{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(546), 0, 0, 0} }, "guest"} } -; - - /* security id for everyone/world system group */ static const struct cifs_sid sid_everyone = { 1, 1, {0, 0, 0, 0, 0, 1}, {0} }; /* security id for Authenticated Users system group */ static const struct cifs_sid sid_authusers = { - 1, 1, {0, 0, 0, 0, 0, 5}, {11} }; + 1, 1, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(11)} }; /* group users */ static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} }; +const struct cred *root_cred; -int match_sid(struct cifs_sid *ctsid) +static void +shrink_idmap_tree(struct rb_root *root, int nr_to_scan, int *nr_rem, + int *nr_del) { - int i, j; - int num_subauth, num_sat, num_saw; - struct cifs_sid *cwsid; + struct rb_node *node; + struct rb_node *tmp; + struct cifs_sid_id *psidid; + + node = rb_first(root); + while (node) { + tmp = node; + node = rb_next(tmp); + psidid = rb_entry(tmp, struct cifs_sid_id, rbnode); + if (nr_to_scan == 0 || *nr_del == nr_to_scan) + ++(*nr_rem); + else { + if (time_after(jiffies, psidid->time + SID_MAP_EXPIRE) + && psidid->refcount == 0) { + rb_erase(tmp, root); + ++(*nr_del); + } else + ++(*nr_rem); + } + } +} - if (!ctsid) - return -1; +/* + * Run idmap cache shrinker. + */ +static int +cifs_idmap_shrinker(struct shrinker *shrink, struct shrink_control *sc) +{ + int nr_to_scan = sc->nr_to_scan; + int nr_del = 0; + int nr_rem = 0; + struct rb_root *root; + + root = &uidtree; + spin_lock(&siduidlock); + shrink_idmap_tree(root, nr_to_scan, &nr_rem, &nr_del); + spin_unlock(&siduidlock); + + root = &gidtree; + spin_lock(&sidgidlock); + shrink_idmap_tree(root, nr_to_scan, &nr_rem, &nr_del); + spin_unlock(&sidgidlock); + + return nr_rem; +} - for (i = 0; i < NUM_WK_SIDS; ++i) { - cwsid = &(wksidarr[i].cifssid); +static struct shrinker cifs_shrinker = { + .shrink = cifs_idmap_shrinker, + .seeks = DEFAULT_SEEKS, +}; - /* compare the revision */ - if (ctsid->revision != cwsid->revision) - continue; +static int +cifs_idmap_key_instantiate(struct key *key, const void *data, size_t datalen) +{ + char *payload; + + payload = kmalloc(datalen, GFP_KERNEL); + if (!payload) + return -ENOMEM; - /* compare all of the six auth values */ - for (j = 0; j < 6; ++j) { - if (ctsid->authority[j] != cwsid->authority[j]) - break; + memcpy(payload, data, datalen); + key->payload.data = payload; + return 0; +} + +static inline void +cifs_idmap_key_destroy(struct key *key) +{ + kfree(key->payload.data); +} + +struct key_type cifs_idmap_key_type = { + .name = "cifs.idmap", + .instantiate = cifs_idmap_key_instantiate, + .destroy = cifs_idmap_key_destroy, + .describe = user_describe, + .match = user_match, +}; + +static void +sid_to_str(struct cifs_sid *sidptr, char *sidstr) +{ + int i; + unsigned long saval; + char *strptr; + + strptr = sidstr; + + sprintf(strptr, "%s", "S"); + strptr = sidstr + strlen(sidstr); + + sprintf(strptr, "-%d", sidptr->revision); + strptr = sidstr + strlen(sidstr); + + for (i = 0; i < 6; ++i) { + if (sidptr->authority[i]) { + sprintf(strptr, "-%d", sidptr->authority[i]); + strptr = sidstr + strlen(sidstr); } - if (j < 6) - continue; /* all of the auth values did not match */ - - /* compare all of the subauth values if any */ - num_sat = ctsid->num_subauth; - num_saw = cwsid->num_subauth; - num_subauth = num_sat < num_saw ? num_sat : num_saw; - if (num_subauth) { - for (j = 0; j < num_subauth; ++j) { - if (ctsid->sub_auth[j] != cwsid->sub_auth[j]) - break; - } - if (j < num_subauth) - continue; /* all sub_auth values do not match */ + } + + for (i = 0; i < sidptr->num_subauth; ++i) { + saval = le32_to_cpu(sidptr->sub_auth[i]); + sprintf(strptr, "-%ld", saval); + strptr = sidstr + strlen(sidstr); + } +} + +static void +id_rb_insert(struct rb_root *root, struct cifs_sid *sidptr, + struct cifs_sid_id **psidid, char *typestr) +{ + int rc; + char *strptr; + struct rb_node *node = root->rb_node; + struct rb_node *parent = NULL; + struct rb_node **linkto = &(root->rb_node); + struct cifs_sid_id *lsidid; + + while (node) { + lsidid = rb_entry(node, struct cifs_sid_id, rbnode); + parent = node; + rc = compare_sids(sidptr, &((lsidid)->sid)); + if (rc > 0) { + linkto = &(node->rb_left); + node = node->rb_left; + } else if (rc < 0) { + linkto = &(node->rb_right); + node = node->rb_right; } + } + + memcpy(&(*psidid)->sid, sidptr, sizeof(struct cifs_sid)); + (*psidid)->time = jiffies - (SID_MAP_RETRY + 1); + (*psidid)->refcount = 0; + + sprintf((*psidid)->sidstr, "%s", typestr); + strptr = (*psidid)->sidstr + strlen((*psidid)->sidstr); + sid_to_str(&(*psidid)->sid, strptr); + + clear_bit(SID_ID_PENDING, &(*psidid)->state); + clear_bit(SID_ID_MAPPED, &(*psidid)->state); + + rb_link_node(&(*psidid)->rbnode, parent, linkto); + rb_insert_color(&(*psidid)->rbnode, root); +} - cFYI(1, "matching sid: %s\n", wksidarr[i].sidname); - return 0; /* sids compare/match */ +static struct cifs_sid_id * +id_rb_search(struct rb_root *root, struct cifs_sid *sidptr) +{ + int rc; + struct rb_node *node = root->rb_node; + struct cifs_sid_id *lsidid; + + while (node) { + lsidid = rb_entry(node, struct cifs_sid_id, rbnode); + rc = compare_sids(sidptr, &((lsidid)->sid)); + if (rc > 0) { + node = node->rb_left; + } else if (rc < 0) { + node = node->rb_right; + } else /* node found */ + return lsidid; } - cFYI(1, "No matching sid"); - return -1; + return NULL; +} + +static int +sidid_pending_wait(void *unused) +{ + schedule(); + return signal_pending(current) ? -ERESTARTSYS : 0; +} + +static int +sid_to_id(struct cifs_sb_info *cifs_sb, struct cifs_sid *psid, + struct cifs_fattr *fattr, uint sidtype) +{ + int rc; + unsigned long cid; + struct key *idkey; + const struct cred *saved_cred; + struct cifs_sid_id *psidid, *npsidid; + struct rb_root *cidtree; + spinlock_t *cidlock; + + if (sidtype == SIDOWNER) { + cid = cifs_sb->mnt_uid; /* default uid, in case upcall fails */ + cidlock = &siduidlock; + cidtree = &uidtree; + } else if (sidtype == SIDGROUP) { + cid = cifs_sb->mnt_gid; /* default gid, in case upcall fails */ + cidlock = &sidgidlock; + cidtree = &gidtree; + } else + return -ENOENT; + + spin_lock(cidlock); + psidid = id_rb_search(cidtree, psid); + + if (!psidid) { /* node does not exist, allocate one & attempt adding */ + spin_unlock(cidlock); + npsidid = kzalloc(sizeof(struct cifs_sid_id), GFP_KERNEL); + if (!npsidid) + return -ENOMEM; + + npsidid->sidstr = kmalloc(SIDLEN, GFP_KERNEL); + if (!npsidid->sidstr) { + kfree(npsidid); + return -ENOMEM; + } + + spin_lock(cidlock); + psidid = id_rb_search(cidtree, psid); + if (psidid) { /* node happened to get inserted meanwhile */ + ++psidid->refcount; + spin_unlock(cidlock); + kfree(npsidid->sidstr); + kfree(npsidid); + } else { + psidid = npsidid; + id_rb_insert(cidtree, psid, &psidid, + sidtype == SIDOWNER ? "os:" : "gs:"); + ++psidid->refcount; + spin_unlock(cidlock); + } + } else { + ++psidid->refcount; + spin_unlock(cidlock); + } + + /* + * If we are here, it is safe to access psidid and its fields + * since a reference was taken earlier while holding the spinlock. + * A reference on the node is put without holding the spinlock + * and it is OK to do so in this case, shrinker will not erase + * this node until all references are put and we do not access + * any fields of the node after a reference is put . + */ + if (test_bit(SID_ID_MAPPED, &psidid->state)) { + cid = psidid->id; + psidid->time = jiffies; /* update ts for accessing */ + goto sid_to_id_out; + } + + if (time_after(psidid->time + SID_MAP_RETRY, jiffies)) + goto sid_to_id_out; + + if (!test_and_set_bit(SID_ID_PENDING, &psidid->state)) { + saved_cred = override_creds(root_cred); + idkey = request_key(&cifs_idmap_key_type, psidid->sidstr, ""); + if (IS_ERR(idkey)) + cFYI(1, "%s: Can't map SID to an id", __func__); + else { + cid = *(unsigned long *)idkey->payload.value; + psidid->id = cid; + set_bit(SID_ID_MAPPED, &psidid->state); + key_put(idkey); + kfree(psidid->sidstr); + } + revert_creds(saved_cred); + psidid->time = jiffies; /* update ts for accessing */ + clear_bit(SID_ID_PENDING, &psidid->state); + wake_up_bit(&psidid->state, SID_ID_PENDING); + } else { + rc = wait_on_bit(&psidid->state, SID_ID_PENDING, + sidid_pending_wait, TASK_INTERRUPTIBLE); + if (rc) { + cFYI(1, "%s: sidid_pending_wait interrupted %d", + __func__, rc); + --psidid->refcount; /* decremented without spinlock */ + return rc; + } + if (test_bit(SID_ID_MAPPED, &psidid->state)) + cid = psidid->id; + } + +sid_to_id_out: + --psidid->refcount; /* decremented without spinlock */ + if (sidtype == SIDOWNER) + fattr->cf_uid = cid; + else + fattr->cf_gid = cid; + + return 0; +} + +int +init_cifs_idmap(void) +{ + struct cred *cred; + struct key *keyring; + int ret; + + cFYI(1, "Registering the %s key type\n", cifs_idmap_key_type.name); + + /* create an override credential set with a special thread keyring in + * which requests are cached + * + * this is used to prevent malicious redirections from being installed + * with add_key(). + */ + cred = prepare_kernel_cred(NULL); + if (!cred) + return -ENOMEM; + + keyring = key_alloc(&key_type_keyring, ".cifs_idmap", 0, 0, cred, + (KEY_POS_ALL & ~KEY_POS_SETATTR) | + KEY_USR_VIEW | KEY_USR_READ, + KEY_ALLOC_NOT_IN_QUOTA); + if (IS_ERR(keyring)) { + ret = PTR_ERR(keyring); + goto failed_put_cred; + } + + ret = key_instantiate_and_link(keyring, NULL, 0, NULL, NULL); + if (ret < 0) + goto failed_put_key; + + ret = register_key_type(&cifs_idmap_key_type); + if (ret < 0) + goto failed_put_key; + + /* instruct request_key() to use this special keyring as a cache for + * the results it looks up */ + cred->thread_keyring = keyring; + cred->jit_keyring = KEY_REQKEY_DEFL_THREAD_KEYRING; + root_cred = cred; + + spin_lock_init(&siduidlock); + uidtree = RB_ROOT; + spin_lock_init(&sidgidlock); + gidtree = RB_ROOT; + + register_shrinker(&cifs_shrinker); + + cFYI(1, "cifs idmap keyring: %d\n", key_serial(keyring)); + return 0; + +failed_put_key: + key_put(keyring); +failed_put_cred: + put_cred(cred); + return ret; +} + +void +exit_cifs_idmap(void) +{ + key_revoke(root_cred->thread_keyring); + unregister_key_type(&cifs_idmap_key_type); + put_cred(root_cred); + unregister_shrinker(&cifs_shrinker); + cFYI(1, "Unregistered %s key type\n", cifs_idmap_key_type.name); +} + +void +cifs_destroy_idmaptrees(void) +{ + struct rb_root *root; + struct rb_node *node; + + root = &uidtree; + spin_lock(&siduidlock); + while ((node = rb_first(root))) + rb_erase(node, root); + spin_unlock(&siduidlock); + + root = &gidtree; + spin_lock(&sidgidlock); + while ((node = rb_first(root))) + rb_erase(node, root); + spin_unlock(&sidgidlock); } /* if the two SIDs (roughly equivalent to a UUID for a user or group) are @@ -104,16 +432,24 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) int num_subauth, num_sat, num_saw; if ((!ctsid) || (!cwsid)) - return 0; + return 1; /* compare the revision */ - if (ctsid->revision != cwsid->revision) - return 0; + if (ctsid->revision != cwsid->revision) { + if (ctsid->revision > cwsid->revision) + return 1; + else + return -1; + } /* compare all of the six auth values */ for (i = 0; i < 6; ++i) { - if (ctsid->authority[i] != cwsid->authority[i]) - return 0; + if (ctsid->authority[i] != cwsid->authority[i]) { + if (ctsid->authority[i] > cwsid->authority[i]) + return 1; + else + return -1; + } } /* compare all of the subauth values if any */ @@ -122,12 +458,17 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) num_subauth = num_sat < num_saw ? num_sat : num_saw; if (num_subauth) { for (i = 0; i < num_subauth; ++i) { - if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) - return 0; + if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) { + if (le32_to_cpu(ctsid->sub_auth[i]) > + le32_to_cpu(cwsid->sub_auth[i])) + return 1; + else + return -1; + } } } - return 1; /* sids compare/match */ + return 0; /* sids compare/match */ } @@ -382,22 +723,22 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl, #ifdef CONFIG_CIFS_DEBUG2 dump_ace(ppace[i], end_of_acl); #endif - if (compare_sids(&(ppace[i]->sid), pownersid)) + if (compare_sids(&(ppace[i]->sid), pownersid) == 0) access_flags_to_mode(ppace[i]->access_req, ppace[i]->type, &fattr->cf_mode, &user_mask); - if (compare_sids(&(ppace[i]->sid), pgrpsid)) + if (compare_sids(&(ppace[i]->sid), pgrpsid) == 0) access_flags_to_mode(ppace[i]->access_req, ppace[i]->type, &fattr->cf_mode, &group_mask); - if (compare_sids(&(ppace[i]->sid), &sid_everyone)) + if (compare_sids(&(ppace[i]->sid), &sid_everyone) == 0) access_flags_to_mode(ppace[i]->access_req, ppace[i]->type, &fattr->cf_mode, &other_mask); - if (compare_sids(&(ppace[i]->sid), &sid_authusers)) + if (compare_sids(&(ppace[i]->sid), &sid_authusers) == 0) access_flags_to_mode(ppace[i]->access_req, ppace[i]->type, &fattr->cf_mode, @@ -475,10 +816,10 @@ static int parse_sid(struct cifs_sid *psid, char *end_of_acl) /* Convert CIFS ACL to POSIX form */ -static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, - struct cifs_fattr *fattr) +static int parse_sec_desc(struct cifs_sb_info *cifs_sb, + struct cifs_ntsd *pntsd, int acl_len, struct cifs_fattr *fattr) { - int rc; + int rc = 0; struct cifs_sid *owner_sid_ptr, *group_sid_ptr; struct cifs_acl *dacl_ptr; /* no need for SACL ptr */ char *end_of_acl = ((char *)pntsd) + acl_len; @@ -500,12 +841,26 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, le32_to_cpu(pntsd->sacloffset), dacloffset); /* cifs_dump_mem("owner_sid: ", owner_sid_ptr, 64); */ rc = parse_sid(owner_sid_ptr, end_of_acl); - if (rc) + if (rc) { + cFYI(1, "%s: Error %d parsing Owner SID", __func__, rc); + return rc; + } + rc = sid_to_id(cifs_sb, owner_sid_ptr, fattr, SIDOWNER); + if (rc) { + cFYI(1, "%s: Error %d mapping Owner SID to uid", __func__, rc); return rc; + } rc = parse_sid(group_sid_ptr, end_of_acl); - if (rc) + if (rc) { + cFYI(1, "%s: Error %d mapping Owner SID to gid", __func__, rc); return rc; + } + rc = sid_to_id(cifs_sb, group_sid_ptr, fattr, SIDGROUP); + if (rc) { + cFYI(1, "%s: Error %d mapping Group SID to gid", __func__, rc); + return rc; + } if (dacloffset) parse_dacl(dacl_ptr, end_of_acl, owner_sid_ptr, @@ -520,7 +875,7 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, memcpy((void *)(&(cifscred->gsid)), (void *)group_sid_ptr, sizeof(struct cifs_sid)); */ - return 0; + return rc; } @@ -592,7 +947,7 @@ static struct cifs_ntsd *get_cifs_acl_by_path(struct cifs_sb_info *cifs_sb, int oplock = 0; int xid, rc; __u16 fid; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct tcon_link *tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) @@ -660,7 +1015,7 @@ static int set_cifs_acl_by_path(struct cifs_sb_info *cifs_sb, const char *path, int oplock = 0; int xid, rc; __u16 fid; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct tcon_link *tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) @@ -688,7 +1043,7 @@ out: } /* Set an ACL on the server */ -static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, +int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, struct inode *inode, const char *path) { struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); @@ -727,7 +1082,7 @@ cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr, rc = PTR_ERR(pntsd); cERROR(1, "%s: error %d getting sec desc", __func__, rc); } else { - rc = parse_sec_desc(pntsd, acllen, fattr); + rc = parse_sec_desc(cifs_sb, pntsd, acllen, fattr); kfree(pntsd); if (rc) cERROR(1, "parse sec desc failed rc = %d", rc); diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index c4ae7d0..5c902c7 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -39,6 +39,15 @@ #define ACCESS_ALLOWED 0 #define ACCESS_DENIED 1 +#define SIDOWNER 1 +#define SIDGROUP 2 +#define SIDLEN 150 /* S- 1 revision- 6 authorities- max 5 sub authorities */ + +#define SID_ID_MAPPED 0 +#define SID_ID_PENDING 1 +#define SID_MAP_EXPIRE (3600 * HZ) /* map entry expires after one hour */ +#define SID_MAP_RETRY (300 * HZ) /* wait 5 minutes for next attempt to map */ + struct cifs_ntsd { __le16 revision; /* revision level */ __le16 type; @@ -74,7 +83,21 @@ struct cifs_wksid { char sidname[SIDNAMELENGTH]; } __attribute__((packed)); -extern int match_sid(struct cifs_sid *); +struct cifs_sid_id { + unsigned int refcount; /* increment with spinlock, decrement without */ + unsigned long id; + unsigned long time; + unsigned long state; + char *sidstr; + struct rb_node rbnode; + struct cifs_sid sid; +}; + +#ifdef __KERNEL__ +extern struct key_type cifs_idmap_key_type; +extern const struct cred *root_cred; +#endif /* KERNEL */ + extern int compare_sids(const struct cifs_sid *, const struct cifs_sid *); #endif /* _CIFSACL_H */ diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index d1a016b..5a0ee7f 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -60,7 +60,7 @@ static int cifs_calculate_signature(const struct smb_hdr *cifs_pdu, server->session_key.response, server->session_key.len); crypto_shash_update(&server->secmech.sdescmd5->shash, - cifs_pdu->Protocol, cifs_pdu->smb_buf_length); + cifs_pdu->Protocol, be32_to_cpu(cifs_pdu->smb_buf_length)); rc = crypto_shash_final(&server->secmech.sdescmd5->shash, signature); @@ -184,7 +184,7 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu, if (cifs_pdu == NULL || server == NULL) return -EINVAL; - if (cifs_pdu->Command == SMB_COM_NEGOTIATE) + if (!server->session_estab) return 0; if (cifs_pdu->Command == SMB_COM_LOCKING_ANDX) { @@ -229,7 +229,7 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu, } /* first calculate 24 bytes ntlm response and then 16 byte session key */ -int setup_ntlm_response(struct cifsSesInfo *ses) +int setup_ntlm_response(struct cifs_ses *ses) { int rc = 0; unsigned int temp_len = CIFS_SESS_KEY_SIZE + CIFS_AUTH_RESP_SIZE; @@ -268,10 +268,11 @@ int setup_ntlm_response(struct cifsSesInfo *ses) } #ifdef CONFIG_CIFS_WEAK_PW_HASH -void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, +int calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, char *lnm_session_key) { int i; + int rc; char password_with_pad[CIFS_ENCPWD_SIZE]; memset(password_with_pad, 0, CIFS_ENCPWD_SIZE); @@ -282,7 +283,7 @@ void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, memset(lnm_session_key, 0, CIFS_SESS_KEY_SIZE); memcpy(lnm_session_key, password_with_pad, CIFS_ENCPWD_SIZE); - return; + return 0; } /* calculate old style session key */ @@ -299,10 +300,9 @@ void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, for (i = 0; i < CIFS_ENCPWD_SIZE; i++) password_with_pad[i] = toupper(password_with_pad[i]); - SMBencrypt(password_with_pad, cryptkey, lnm_session_key); + rc = SMBencrypt(password_with_pad, cryptkey, lnm_session_key); - /* clear password before we return/free memory */ - memset(password_with_pad, 0, CIFS_ENCPWD_SIZE); + return rc; } #endif /* CIFS_WEAK_PW_HASH */ @@ -312,7 +312,7 @@ void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, * Allocate domain name which gets freed when session struct is deallocated. */ static int -build_avpair_blob(struct cifsSesInfo *ses, const struct nls_table *nls_cp) +build_avpair_blob(struct cifs_ses *ses, const struct nls_table *nls_cp) { unsigned int dlen; unsigned int wlen; @@ -400,7 +400,7 @@ build_avpair_blob(struct cifsSesInfo *ses, const struct nls_table *nls_cp) * about target string i.e. for some, just user name might suffice. */ static int -find_domain_name(struct cifsSesInfo *ses, const struct nls_table *nls_cp) +find_domain_name(struct cifs_ses *ses, const struct nls_table *nls_cp) { unsigned int attrsize; unsigned int type; @@ -445,7 +445,7 @@ find_domain_name(struct cifsSesInfo *ses, const struct nls_table *nls_cp) return 0; } -static int calc_ntlmv2_hash(struct cifsSesInfo *ses, char *ntlmv2_hash, +static int calc_ntlmv2_hash(struct cifs_ses *ses, char *ntlmv2_hash, const struct nls_table *nls_cp) { int rc = 0; @@ -527,7 +527,7 @@ calc_exit_2: } static int -CalcNTLMv2_response(const struct cifsSesInfo *ses, char *ntlmv2_hash) +CalcNTLMv2_response(const struct cifs_ses *ses, char *ntlmv2_hash) { int rc; unsigned int offset = CIFS_SESS_KEY_SIZE + 8; @@ -563,7 +563,7 @@ CalcNTLMv2_response(const struct cifsSesInfo *ses, char *ntlmv2_hash) int -setup_ntlmv2_rsp(struct cifsSesInfo *ses, const struct nls_table *nls_cp) +setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp) { int rc; int baselen; @@ -649,7 +649,7 @@ setup_ntlmv2_rsp_ret: } int -calc_seckey(struct cifsSesInfo *ses) +calc_seckey(struct cifs_ses *ses) { int rc; struct crypto_blkcipher *tfm_arc4; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 5c412b3..35f9154 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -104,66 +104,31 @@ cifs_sb_deactive(struct super_block *sb) } static int -cifs_read_super(struct super_block *sb, void *data, - const char *devname, int silent) +cifs_read_super(struct super_block *sb) { struct inode *inode; struct cifs_sb_info *cifs_sb; int rc = 0; - /* BB should we make this contingent on mount parm? */ - sb->s_flags |= MS_NODIRATIME | MS_NOATIME; - sb->s_fs_info = kzalloc(sizeof(struct cifs_sb_info), GFP_KERNEL); cifs_sb = CIFS_SB(sb); - if (cifs_sb == NULL) - return -ENOMEM; - - spin_lock_init(&cifs_sb->tlink_tree_lock); - cifs_sb->tlink_tree = RB_ROOT; - - rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY); - if (rc) { - kfree(cifs_sb); - return rc; - } - cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages; -#ifdef CONFIG_CIFS_DFS_UPCALL - /* copy mount params to sb for use in submounts */ - /* BB: should we move this after the mount so we - * do not have to do the copy on failed mounts? - * BB: May be it is better to do simple copy before - * complex operation (mount), and in case of fail - * just exit instead of doing mount and attempting - * undo it if this copy fails?*/ - if (data) { - int len = strlen(data); - cifs_sb->mountdata = kzalloc(len + 1, GFP_KERNEL); - if (cifs_sb->mountdata == NULL) { - bdi_destroy(&cifs_sb->bdi); - kfree(sb->s_fs_info); - sb->s_fs_info = NULL; - return -ENOMEM; - } - strncpy(cifs_sb->mountdata, data, len + 1); - cifs_sb->mountdata[len] = '\0'; - } -#endif + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIXACL) + sb->s_flags |= MS_POSIXACL; - rc = cifs_mount(sb, cifs_sb, data, devname); + if (cifs_sb_master_tcon(cifs_sb)->ses->capabilities & CAP_LARGE_FILES) + sb->s_maxbytes = MAX_LFS_FILESIZE; + else + sb->s_maxbytes = MAX_NON_LFS; - if (rc) { - if (!silent) - cERROR(1, "cifs_mount failed w/return code = %d", rc); - goto out_mount_failed; - } + /* BB FIXME fix time_gran to be larger for LANMAN sessions */ + sb->s_time_gran = 100; sb->s_magic = CIFS_MAGIC_NUMBER; sb->s_op = &cifs_super_ops; sb->s_bdi = &cifs_sb->bdi; sb->s_blocksize = CIFS_MAX_MSGSIZE; sb->s_blocksize_bits = 14; /* default 2**14 = CIFS_MAX_MSGSIZE */ - inode = cifs_root_iget(sb, ROOT_I); + inode = cifs_root_iget(sb); if (IS_ERR(inode)) { rc = PTR_ERR(inode); @@ -184,12 +149,12 @@ cifs_read_super(struct super_block *sb, void *data, else sb->s_d_op = &cifs_dentry_ops; -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CIFS_NFSD_EXPORT if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) { cFYI(1, "export ops supported"); sb->s_export_op = &cifs_export_ops; } -#endif /* EXPERIMENTAL */ +#endif /* CIFS_NFSD_EXPORT */ return 0; @@ -198,49 +163,14 @@ out_no_root: if (inode) iput(inode); - cifs_umount(sb, cifs_sb); - -out_mount_failed: - if (cifs_sb) { -#ifdef CONFIG_CIFS_DFS_UPCALL - if (cifs_sb->mountdata) { - kfree(cifs_sb->mountdata); - cifs_sb->mountdata = NULL; - } -#endif - unload_nls(cifs_sb->local_nls); - bdi_destroy(&cifs_sb->bdi); - kfree(cifs_sb); - } return rc; } -static void -cifs_put_super(struct super_block *sb) +static void cifs_kill_sb(struct super_block *sb) { - int rc = 0; - struct cifs_sb_info *cifs_sb; - - cFYI(1, "In cifs_put_super"); - cifs_sb = CIFS_SB(sb); - if (cifs_sb == NULL) { - cFYI(1, "Empty cifs superblock info passed to unmount"); - return; - } - - rc = cifs_umount(sb, cifs_sb); - if (rc) - cERROR(1, "cifs_umount failed with return code %d", rc); -#ifdef CONFIG_CIFS_DFS_UPCALL - if (cifs_sb->mountdata) { - kfree(cifs_sb->mountdata); - cifs_sb->mountdata = NULL; - } -#endif - - unload_nls(cifs_sb->local_nls); - bdi_destroy(&cifs_sb->bdi); - kfree(cifs_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); + kill_anon_super(sb); + cifs_umount(cifs_sb); } static int @@ -248,7 +178,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); int rc = -EOPNOTSUPP; int xid; @@ -297,9 +227,6 @@ static int cifs_permission(struct inode *inode, int mask, unsigned int flags) { struct cifs_sb_info *cifs_sb; - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - cifs_sb = CIFS_SB(inode->i_sb); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) { @@ -392,6 +319,37 @@ cifs_show_address(struct seq_file *s, struct TCP_Server_Info *server) } } +static void +cifs_show_security(struct seq_file *s, struct TCP_Server_Info *server) +{ + seq_printf(s, ",sec="); + + switch (server->secType) { + case LANMAN: + seq_printf(s, "lanman"); + break; + case NTLMv2: + seq_printf(s, "ntlmv2"); + break; + case NTLM: + seq_printf(s, "ntlm"); + break; + case Kerberos: + seq_printf(s, "krb5"); + break; + case RawNTLMSSP: + seq_printf(s, "ntlmssp"); + break; + default: + /* shouldn't ever happen */ + seq_printf(s, "unknown"); + break; + } + + if (server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + seq_printf(s, "i"); +} + /* * cifs_show_options() is for displaying mount options in /proc/mounts. * Not all settable options are displayed but most of the important @@ -401,10 +359,12 @@ static int cifs_show_options(struct seq_file *s, struct vfsmount *m) { struct cifs_sb_info *cifs_sb = CIFS_SB(m->mnt_sb); - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct sockaddr *srcaddr; srcaddr = (struct sockaddr *)&tcon->ses->server->srcaddr; + cifs_show_security(s, tcon->ses->server); + seq_printf(s, ",unc=%s", tcon->treeName); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) @@ -455,14 +415,20 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) seq_printf(s, ",nocase"); if (tcon->retry) seq_printf(s, ",hard"); - if (cifs_sb->prepath) - seq_printf(s, ",prepath=%s", cifs_sb->prepath); + if (tcon->unix_ext) + seq_printf(s, ",unix"); + else + seq_printf(s, ",nounix"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) seq_printf(s, ",posixpaths"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) seq_printf(s, ",setuids"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) seq_printf(s, ",serverino"); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + seq_printf(s, ",rwpidforward"); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) + seq_printf(s, ",forcemand"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) seq_printf(s, ",directio"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) @@ -495,7 +461,7 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) static void cifs_umount_begin(struct super_block *sb) { struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; if (cifs_sb == NULL) return; @@ -552,7 +518,6 @@ static int cifs_drop_inode(struct inode *inode) } static const struct super_operations cifs_super_ops = { - .put_super = cifs_put_super, .statfs = cifs_statfs, .alloc_inode = cifs_alloc_inode, .destroy_inode = cifs_destroy_inode, @@ -570,29 +535,194 @@ static const struct super_operations cifs_super_ops = { #endif }; +/* + * Get root dentry from superblock according to prefix path mount option. + * Return dentry with refcount + 1 on success and NULL otherwise. + */ +static struct dentry * +cifs_get_root(struct smb_vol *vol, struct super_block *sb) +{ + int xid, rc; + struct inode *inode; + struct qstr name; + struct dentry *dparent = NULL, *dchild = NULL, *alias; + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); + unsigned int i, full_len, len; + char *full_path = NULL, *pstart; + char sep; + + full_path = cifs_build_path_to_root(vol, cifs_sb, + cifs_sb_master_tcon(cifs_sb)); + if (full_path == NULL) + return ERR_PTR(-ENOMEM); + + cFYI(1, "Get root dentry for %s", full_path); + + xid = GetXid(); + sep = CIFS_DIR_SEP(cifs_sb); + dparent = dget(sb->s_root); + full_len = strlen(full_path); + full_path[full_len] = sep; + pstart = full_path + 1; + + for (i = 1, len = 0; i <= full_len; i++) { + if (full_path[i] != sep || !len) { + len++; + continue; + } + + full_path[i] = 0; + cFYI(1, "get dentry for %s", pstart); + + name.name = pstart; + name.len = len; + name.hash = full_name_hash(pstart, len); + dchild = d_lookup(dparent, &name); + if (dchild == NULL) { + cFYI(1, "not exists"); + dchild = d_alloc(dparent, &name); + if (dchild == NULL) { + dput(dparent); + dparent = ERR_PTR(-ENOMEM); + goto out; + } + } + + cFYI(1, "get inode"); + if (dchild->d_inode == NULL) { + cFYI(1, "not exists"); + inode = NULL; + if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext) + rc = cifs_get_inode_info_unix(&inode, full_path, + sb, xid); + else + rc = cifs_get_inode_info(&inode, full_path, + NULL, sb, xid, NULL); + if (rc) { + dput(dchild); + dput(dparent); + dparent = ERR_PTR(rc); + goto out; + } + alias = d_materialise_unique(dchild, inode); + if (alias != NULL) { + dput(dchild); + if (IS_ERR(alias)) { + dput(dparent); + dparent = ERR_PTR(-EINVAL); /* XXX */ + goto out; + } + dchild = alias; + } + } + cFYI(1, "parent %p, child %p", dparent, dchild); + + dput(dparent); + dparent = dchild; + len = 0; + pstart = full_path + i + 1; + full_path[i] = sep; + } +out: + _FreeXid(xid); + kfree(full_path); + return dparent; +} + +static int cifs_set_super(struct super_block *sb, void *data) +{ + struct cifs_mnt_data *mnt_data = data; + sb->s_fs_info = mnt_data->cifs_sb; + return set_anon_super(sb, NULL); +} + static struct dentry * cifs_do_mount(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data) + int flags, const char *dev_name, void *data) { int rc; struct super_block *sb; - - sb = sget(fs_type, NULL, set_anon_super, NULL); + struct cifs_sb_info *cifs_sb; + struct smb_vol *volume_info; + struct cifs_mnt_data mnt_data; + struct dentry *root; cFYI(1, "Devname: %s flags: %d ", dev_name, flags); - if (IS_ERR(sb)) - return ERR_CAST(sb); + rc = cifs_setup_volume_info(&volume_info, (char *)data, dev_name); + if (rc) + return ERR_PTR(rc); + + cifs_sb = kzalloc(sizeof(struct cifs_sb_info), GFP_KERNEL); + if (cifs_sb == NULL) { + root = ERR_PTR(-ENOMEM); + goto out_nls; + } - sb->s_flags = flags; + cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL); + if (cifs_sb->mountdata == NULL) { + root = ERR_PTR(-ENOMEM); + goto out_cifs_sb; + } + + cifs_setup_cifs_sb(volume_info, cifs_sb); - rc = cifs_read_super(sb, data, dev_name, flags & MS_SILENT ? 1 : 0); + rc = cifs_mount(cifs_sb, volume_info); if (rc) { - deactivate_locked_super(sb); - return ERR_PTR(rc); + if (!(flags & MS_SILENT)) + cERROR(1, "cifs_mount failed w/return code = %d", rc); + root = ERR_PTR(rc); + goto out_mountdata; } - sb->s_flags |= MS_ACTIVE; - return dget(sb->s_root); + + mnt_data.vol = volume_info; + mnt_data.cifs_sb = cifs_sb; + mnt_data.flags = flags; + + sb = sget(fs_type, cifs_match_super, cifs_set_super, &mnt_data); + if (IS_ERR(sb)) { + root = ERR_CAST(sb); + cifs_umount(cifs_sb); + goto out; + } + + if (sb->s_root) { + cFYI(1, "Use existing superblock"); + cifs_umount(cifs_sb); + } else { + sb->s_flags = flags; + /* BB should we make this contingent on mount parm? */ + sb->s_flags |= MS_NODIRATIME | MS_NOATIME; + + rc = cifs_read_super(sb); + if (rc) { + root = ERR_PTR(rc); + goto out_super; + } + + sb->s_flags |= MS_ACTIVE; + } + + root = cifs_get_root(volume_info, sb); + if (IS_ERR(root)) + goto out_super; + + cFYI(1, "dentry root is: %p", root); + goto out; + +out_super: + deactivate_locked_super(sb); +out: + cifs_cleanup_volume_info(&volume_info); + return root; + +out_mountdata: + kfree(cifs_sb->mountdata); +out_cifs_sb: + kfree(cifs_sb); +out_nls: + unload_nls(volume_info->local_nls); + goto out; } static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, @@ -618,16 +748,31 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int origin) { /* origin == SEEK_END => we must revalidate the cached file length */ if (origin == SEEK_END) { - int retval; - - /* some applications poll for the file length in this strange - way so we must seek to end on non-oplocked files by - setting the revalidate time to zero */ - CIFS_I(file->f_path.dentry->d_inode)->time = 0; - - retval = cifs_revalidate_file(file); - if (retval < 0) - return (loff_t)retval; + int rc; + struct inode *inode = file->f_path.dentry->d_inode; + + /* + * We need to be sure that all dirty pages are written and the + * server has the newest file length. + */ + if (!CIFS_I(inode)->clientCanCacheRead && inode->i_mapping && + inode->i_mapping->nrpages != 0) { + rc = filemap_fdatawait(inode->i_mapping); + if (rc) { + mapping_set_error(inode->i_mapping, rc); + return rc; + } + } + /* + * Some applications poll for the file length in this strange + * way so we must seek to end on non-oplocked files by + * setting the revalidate time to zero. + */ + CIFS_I(inode)->time = 0; + + rc = cifs_revalidate_file_attr(file); + if (rc < 0) + return (loff_t)rc; } return generic_file_llseek_unlocked(file, offset, origin); } @@ -666,7 +811,7 @@ struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", .mount = cifs_do_mount, - .kill_sb = kill_anon_super, + .kill_sb = cifs_kill_sb, /* .fs_flags */ }; const struct inode_operations cifs_dir_inode_ops = { @@ -760,10 +905,11 @@ const struct file_operations cifs_file_strict_ops = { }; const struct file_operations cifs_file_direct_ops = { - /* no aio, no readv - - BB reevaluate whether they can be done with directio, no cache */ - .read = cifs_user_read, - .write = cifs_user_write, + /* BB reevaluate whether they can be done with directio, no cache */ + .read = do_sync_read, + .write = do_sync_write, + .aio_read = cifs_user_readv, + .aio_write = cifs_user_writev, .open = cifs_open, .release = cifs_close, .lock = cifs_lock, @@ -815,10 +961,11 @@ const struct file_operations cifs_file_strict_nobrl_ops = { }; const struct file_operations cifs_file_direct_nobrl_ops = { - /* no mmap, no aio, no readv - - BB reevaluate whether they can be done with directio, no cache */ - .read = cifs_user_read, - .write = cifs_user_write, + /* BB reevaluate whether they can be done with directio, no cache */ + .read = do_sync_read, + .write = do_sync_write, + .aio_read = cifs_user_readv, + .aio_write = cifs_user_writev, .open = cifs_open, .release = cifs_close, .fsync = cifs_fsync, @@ -981,10 +1128,10 @@ init_cifs(void) int rc = 0; cifs_proc_init(); INIT_LIST_HEAD(&cifs_tcp_ses_list); -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */ INIT_LIST_HEAD(&GlobalDnotifyReqList); INIT_LIST_HEAD(&GlobalDnotifyRsp_Q); -#endif +#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */ /* * Initialize Global counters */ @@ -1033,22 +1180,33 @@ init_cifs(void) if (rc) goto out_destroy_mids; - rc = register_filesystem(&cifs_fs_type); - if (rc) - goto out_destroy_request_bufs; #ifdef CONFIG_CIFS_UPCALL rc = register_key_type(&cifs_spnego_key_type); if (rc) - goto out_unregister_filesystem; -#endif + goto out_destroy_request_bufs; +#endif /* CONFIG_CIFS_UPCALL */ + +#ifdef CONFIG_CIFS_ACL + rc = init_cifs_idmap(); + if (rc) + goto out_register_key_type; +#endif /* CONFIG_CIFS_ACL */ + + rc = register_filesystem(&cifs_fs_type); + if (rc) + goto out_init_cifs_idmap; return 0; -#ifdef CONFIG_CIFS_UPCALL -out_unregister_filesystem: - unregister_filesystem(&cifs_fs_type); +out_init_cifs_idmap: +#ifdef CONFIG_CIFS_ACL + exit_cifs_idmap(); +out_register_key_type: #endif +#ifdef CONFIG_CIFS_UPCALL + unregister_key_type(&cifs_spnego_key_type); out_destroy_request_bufs: +#endif cifs_destroy_request_bufs(); out_destroy_mids: cifs_destroy_mids(); @@ -1070,6 +1228,10 @@ exit_cifs(void) #ifdef CONFIG_CIFS_DFS_UPCALL cifs_dfs_release_automount_timer(); #endif +#ifdef CONFIG_CIFS_ACL + cifs_destroy_idmaptrees(); + exit_cifs_idmap(); +#endif #ifdef CONFIG_CIFS_UPCALL unregister_key_type(&cifs_spnego_key_type); #endif diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index a9371b6..0900e16 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -47,7 +47,7 @@ extern void cifs_sb_deactive(struct super_block *sb); /* Functions related to inodes */ extern const struct inode_operations cifs_dir_inode_ops; -extern struct inode *cifs_root_iget(struct super_block *, unsigned long); +extern struct inode *cifs_root_iget(struct super_block *); extern int cifs_create(struct inode *, struct dentry *, int, struct nameidata *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, @@ -59,9 +59,11 @@ extern int cifs_mkdir(struct inode *, struct dentry *, int); extern int cifs_rmdir(struct inode *, struct dentry *); extern int cifs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); +extern int cifs_revalidate_file_attr(struct file *filp); +extern int cifs_revalidate_dentry_attr(struct dentry *); extern int cifs_revalidate_file(struct file *filp); extern int cifs_revalidate_dentry(struct dentry *); -extern void cifs_invalidate_mapping(struct inode *inode); +extern int cifs_invalidate_mapping(struct inode *inode); extern int cifs_getattr(struct vfsmount *, struct dentry *, struct kstat *); extern int cifs_setattr(struct dentry *, struct iattr *); @@ -80,12 +82,12 @@ extern const struct file_operations cifs_file_strict_nobrl_ops; extern int cifs_open(struct inode *inode, struct file *file); extern int cifs_close(struct inode *inode, struct file *file); extern int cifs_closedir(struct inode *inode, struct file *file); -extern ssize_t cifs_user_read(struct file *file, char __user *read_data, - size_t read_size, loff_t *poffset); +extern ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos); extern ssize_t cifs_strict_readv(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos); -extern ssize_t cifs_user_write(struct file *file, const char __user *write_data, - size_t write_size, loff_t *poffset); +extern ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos); extern ssize_t cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos); extern int cifs_lock(struct file *, int, struct file_lock *); @@ -123,9 +125,9 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CIFS_NFSD_EXPORT extern const struct export_operations cifs_export_ops; -#endif /* EXPERIMENTAL */ +#endif /* CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "1.71" +#define CIFS_VERSION "1.73" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index a5d1106..6255fa8 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -155,6 +155,81 @@ struct cifs_cred { ***************************************************************** */ +struct smb_vol { + char *username; + char *password; + char *domainname; + char *UNC; + char *UNCip; + char *iocharset; /* local code page for mapping to and from Unicode */ + char source_rfc1001_name[RFC1001_NAME_LEN_WITH_NULL]; /* clnt nb name */ + char target_rfc1001_name[RFC1001_NAME_LEN_WITH_NULL]; /* srvr nb name */ + uid_t cred_uid; + uid_t linux_uid; + gid_t linux_gid; + mode_t file_mode; + mode_t dir_mode; + unsigned secFlg; + bool retry:1; + bool intr:1; + bool setuids:1; + bool override_uid:1; + bool override_gid:1; + bool dynperm:1; + bool noperm:1; + bool no_psx_acl:1; /* set if posix acl support should be disabled */ + bool cifs_acl:1; + bool no_xattr:1; /* set if xattr (EA) support should be disabled*/ + bool server_ino:1; /* use inode numbers from server ie UniqueId */ + bool direct_io:1; + bool strict_io:1; /* strict cache behavior */ + bool remap:1; /* set to remap seven reserved chars in filenames */ + bool posix_paths:1; /* unset to not ask for posix pathnames. */ + bool no_linux_ext:1; + bool sfu_emul:1; + bool nullauth:1; /* attempt to authenticate with null user */ + bool nocase:1; /* request case insensitive filenames */ + bool nobrl:1; /* disable sending byte range locks to srv */ + bool mand_lock:1; /* send mandatory not posix byte range lock reqs */ + bool seal:1; /* request transport encryption on share */ + bool nodfs:1; /* Do not request DFS, even if available */ + bool local_lease:1; /* check leases only on local system, not remote */ + bool noblocksnd:1; + bool noautotune:1; + bool nostrictsync:1; /* do not force expensive SMBflush on every sync */ + bool fsc:1; /* enable fscache */ + bool mfsymlinks:1; /* use Minshall+French Symlinks */ + bool multiuser:1; + bool rwpidforward:1; /* pid forward for read/write operations */ + unsigned int rsize; + unsigned int wsize; + bool sockopt_tcp_nodelay:1; + unsigned short int port; + unsigned long actimeo; /* attribute cache timeout (jiffies) */ + char *prepath; + struct sockaddr_storage srcaddr; /* allow binding to a local IP */ + struct nls_table *local_nls; +}; + +#define CIFS_MOUNT_MASK (CIFS_MOUNT_NO_PERM | CIFS_MOUNT_SET_UID | \ + CIFS_MOUNT_SERVER_INUM | CIFS_MOUNT_DIRECT_IO | \ + CIFS_MOUNT_NO_XATTR | CIFS_MOUNT_MAP_SPECIAL_CHR | \ + CIFS_MOUNT_UNX_EMUL | CIFS_MOUNT_NO_BRL | \ + CIFS_MOUNT_CIFS_ACL | CIFS_MOUNT_OVERR_UID | \ + CIFS_MOUNT_OVERR_GID | CIFS_MOUNT_DYNPERM | \ + CIFS_MOUNT_NOPOSIXBRL | CIFS_MOUNT_NOSSYNC | \ + CIFS_MOUNT_FSCACHE | CIFS_MOUNT_MF_SYMLINKS | \ + CIFS_MOUNT_MULTIUSER | CIFS_MOUNT_STRICT_IO) + +#define CIFS_MS_MASK (MS_RDONLY | MS_MANDLOCK | MS_NOEXEC | MS_NOSUID | \ + MS_NODEV | MS_SYNCHRONOUS) + +struct cifs_mnt_data { + struct cifs_sb_info *cifs_sb; + struct smb_vol *vol; + int flags; +}; + struct TCP_Server_Info { struct list_head tcp_ses_list; struct list_head smb_ses_list; @@ -179,7 +254,7 @@ struct TCP_Server_Info { struct mutex srv_mutex; struct task_struct *tsk; char server_GUID[16]; - char secMode; + char sec_mode; bool session_estab; /* mark when very first sess is established */ u16 dialect; /* dialect index that server chose */ enum securityEnum secType; @@ -254,7 +329,7 @@ static inline void cifs_set_net_ns(struct TCP_Server_Info *srv, struct net *net) /* * Session structure. One of these for each uid session with a particular host */ -struct cifsSesInfo { +struct cifs_ses { struct list_head smb_ses_list; struct list_head tcon_list; struct mutex session_mutex; @@ -274,7 +349,8 @@ struct cifsSesInfo { int capabilities; char serverName[SERVER_NAME_LEN_WITH_NULL * 2]; /* BB make bigger for TCP names - will ipv6 and sctp addresses fit? */ - char *user_name; + char *user_name; /* must not be null except during init of sess + and after mount option parsing we fill it */ char *domainName; char *password; struct session_key auth_key; @@ -293,11 +369,11 @@ struct cifsSesInfo { * there is one of these for each connection to a resource on a particular * session */ -struct cifsTconInfo { +struct cifs_tcon { struct list_head tcon_list; int tc_count; struct list_head openFileList; - struct cifsSesInfo *ses; /* pointer to session associated with */ + struct cifs_ses *ses; /* pointer to session associated with */ char treeName[MAX_TREE_SIZE + 1]; /* UNC name of resource in ASCII */ char *nativeFileSystem; char *password; /* for share-level security */ @@ -379,12 +455,12 @@ struct tcon_link { #define TCON_LINK_IN_TREE 2 unsigned long tl_time; atomic_t tl_count; - struct cifsTconInfo *tl_tcon; + struct cifs_tcon *tl_tcon; }; extern struct tcon_link *cifs_sb_tlink(struct cifs_sb_info *cifs_sb); -static inline struct cifsTconInfo * +static inline struct cifs_tcon * tlink_tcon(struct tcon_link *tlink) { return tlink->tl_tcon; @@ -401,7 +477,7 @@ cifs_get_tlink(struct tcon_link *tlink) } /* This function is always expected to succeed */ -extern struct cifsTconInfo *cifs_sb_master_tcon(struct cifs_sb_info *cifs_sb); +extern struct cifs_tcon *cifs_sb_master_tcon(struct cifs_sb_info *cifs_sb); /* * This info hangs off the cifsFileInfo structure, pointed to by llist. @@ -454,6 +530,14 @@ struct cifsFileInfo { struct work_struct oplock_break; /* work for oplock breaks */ }; +struct cifs_io_parms { + __u16 netfid; + __u32 pid; + __u64 offset; + unsigned int length; + struct cifs_tcon *tcon; +}; + /* * Take a reference on the file private data. Must be called with * cifs_file_list_lock held. @@ -508,10 +592,30 @@ static inline char CIFS_DIR_SEP(const struct cifs_sb_info *cifs_sb) return '\\'; } +static inline void +convert_delimiter(char *path, char delim) +{ + int i; + char old_delim; + + if (path == NULL) + return; + + if (delim == '/') + old_delim = '\\'; + else + old_delim = '/'; + + for (i = 0; path[i] != '\0'; i++) { + if (path[i] == old_delim) + path[i] = delim; + } +} + #ifdef CONFIG_CIFS_STATS #define cifs_stats_inc atomic_inc -static inline void cifs_stats_bytes_written(struct cifsTconInfo *tcon, +static inline void cifs_stats_bytes_written(struct cifs_tcon *tcon, unsigned int bytes) { if (bytes) { @@ -521,7 +625,7 @@ static inline void cifs_stats_bytes_written(struct cifsTconInfo *tcon, } } -static inline void cifs_stats_bytes_read(struct cifsTconInfo *tcon, +static inline void cifs_stats_bytes_read(struct cifs_tcon *tcon, unsigned int bytes) { spin_lock(&tcon->stat_lock); @@ -542,9 +646,8 @@ struct mid_q_entry; * This is the prototype for the mid callback function. When creating one, * take special care to avoid deadlocks. Things to bear in mind: * - * - it will be called by cifsd - * - the GlobalMid_Lock will be held - * - the mid will be removed from the pending_mid_q list + * - it will be called by cifsd, with no locks held + * - the mid will be removed from any lists */ typedef void (mid_callback_t)(struct mid_q_entry *mid); @@ -572,7 +675,7 @@ struct mid_q_entry { struct oplock_q_entry { struct list_head qhead; struct inode *pinode; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; __u16 netfid; }; @@ -655,6 +758,7 @@ static inline void free_dfs_info_array(struct dfs_info3_param *param, #define MID_RESPONSE_RECEIVED 4 #define MID_RETRY_NEEDED 8 /* session closed while this request out */ #define MID_RESPONSE_MALFORMED 0x10 +#define MID_SHUTDOWN 0x20 /* Types of response buffer returned from SendReceive2 */ #define CIFS_NO_BUFFER 0 /* Response buffer not returned */ @@ -780,10 +884,12 @@ GLOBAL_EXTERN spinlock_t cifs_tcp_ses_lock; */ GLOBAL_EXTERN spinlock_t cifs_file_list_lock; +#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */ /* Outstanding dir notify requests */ GLOBAL_EXTERN struct list_head GlobalDnotifyReqList; /* DirNotify response queue */ GLOBAL_EXTERN struct list_head GlobalDnotifyRsp_Q; +#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */ /* * Global transaction id (XID) information @@ -830,6 +936,11 @@ GLOBAL_EXTERN unsigned int cifs_max_pending; /* MAX requests at once to server*/ /* reconnect after this many failed echo attempts */ GLOBAL_EXTERN unsigned short echo_retries; +GLOBAL_EXTERN struct rb_root uidtree; +GLOBAL_EXTERN struct rb_root gidtree; +GLOBAL_EXTERN spinlock_t siduidlock; +GLOBAL_EXTERN spinlock_t sidgidlock; + void cifs_oplock_break(struct work_struct *work); void cifs_oplock_break_get(struct cifsFileInfo *cfile); void cifs_oplock_break_put(struct cifsFileInfo *cfile); diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index b5c8cc5..de3aa28 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -397,9 +397,9 @@ #define GETU32(var) (*((__u32 *)var)) /* BB check for endian issues */ struct smb_hdr { - __u32 smb_buf_length; /* big endian on wire *//* BB length is only two - or three bytes - with one or two byte type preceding it that are - zero - we could mask the type byte off just in case BB */ + __be32 smb_buf_length; /* BB length is only two (rarely three) bytes, + with one or two byte "type" preceding it that will be + zero - we could mask the type byte off */ __u8 Protocol[4]; __u8 Command; union { @@ -428,43 +428,28 @@ struct smb_hdr { __u8 WordCount; } __attribute__((packed)); -/* given a pointer to an smb_hdr retrieve a char pointer to the byte count */ -#define BCC(smb_var) ((unsigned char *)(smb_var) + sizeof(struct smb_hdr) + \ - (2 * (smb_var)->WordCount)) +/* given a pointer to an smb_hdr, retrieve a void pointer to the ByteCount */ +static inline void * +BCC(struct smb_hdr *smb) +{ + return (void *)smb + sizeof(*smb) + 2 * smb->WordCount; +} /* given a pointer to an smb_hdr retrieve the pointer to the byte area */ #define pByteArea(smb_var) (BCC(smb_var) + 2) -/* get the converted ByteCount for a SMB packet and return it */ -static inline __u16 -get_bcc(struct smb_hdr *hdr) -{ - __u16 *bc_ptr = (__u16 *)BCC(hdr); - - return get_unaligned(bc_ptr); -} - /* get the unconverted ByteCount for a SMB packet and return it */ static inline __u16 -get_bcc_le(struct smb_hdr *hdr) +get_bcc(struct smb_hdr *hdr) { __le16 *bc_ptr = (__le16 *)BCC(hdr); return get_unaligned_le16(bc_ptr); } -/* set the ByteCount for a SMB packet in host-byte order */ -static inline void -put_bcc(__u16 count, struct smb_hdr *hdr) -{ - __u16 *bc_ptr = (__u16 *)BCC(hdr); - - put_unaligned(count, bc_ptr); -} - /* set the ByteCount for a SMB packet in little-endian */ static inline void -put_bcc_le(__u16 count, struct smb_hdr *hdr) +put_bcc(__u16 count, struct smb_hdr *hdr) { __le16 *bc_ptr = (__le16 *)BCC(hdr); diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 8096f27..257f312 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -53,9 +53,13 @@ do { \ cFYI(1, "CIFS VFS: leaving %s (xid = %d) rc = %d", \ __func__, curr_xid, (int)rc); \ } while (0) +extern int init_cifs_idmap(void); +extern void exit_cifs_idmap(void); +extern void cifs_destroy_idmaptrees(void); extern char *build_path_from_dentry(struct dentry *); -extern char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb, - struct cifsTconInfo *tcon); +extern char *cifs_build_path_to_root(struct smb_vol *vol, + struct cifs_sb_info *cifs_sb, + struct cifs_tcon *tcon); extern char *build_wildcard_path_from_dentry(struct dentry *direntry); extern char *cifs_compose_mount_options(const char *sb_mountdata, const char *fullpath, const struct dfs_info3_param *ref, @@ -64,20 +68,22 @@ extern char *cifs_compose_mount_options(const char *sb_mountdata, extern struct mid_q_entry *AllocMidQEntry(const struct smb_hdr *smb_buffer, struct TCP_Server_Info *server); extern void DeleteMidQEntry(struct mid_q_entry *midEntry); -extern int cifs_call_async(struct TCP_Server_Info *server, - struct smb_hdr *in_buf, mid_callback_t *callback, - void *cbdata); -extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, +extern int cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, + unsigned int nvec, mid_callback_t *callback, + void *cbdata, bool ignore_pend); +extern int SendReceive(const unsigned int /* xid */ , struct cifs_ses *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */ , const int long_op); -extern int SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, +extern int SendReceiveNoRsp(const unsigned int xid, struct cifs_ses *ses, struct smb_hdr *in_buf, int flags); -extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, +extern int cifs_check_receive(struct mid_q_entry *mid, + struct TCP_Server_Info *server, bool log_error); +extern int SendReceive2(const unsigned int /* xid */ , struct cifs_ses *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int flags); extern int SendReceiveBlockingLock(const unsigned int xid, - struct cifsTconInfo *ptcon, + struct cifs_tcon *ptcon, struct smb_hdr *in_buf , struct smb_hdr *out_buf, int *bytes_returned); @@ -90,21 +96,20 @@ extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool); extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool); extern unsigned int smbCalcSize(struct smb_hdr *ptr); -extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr); extern int decode_negTokenInit(unsigned char *security_blob, int length, struct TCP_Server_Info *server); extern int cifs_convert_address(struct sockaddr *dst, const char *src, int len); extern int cifs_set_port(struct sockaddr *addr, const unsigned short int port); extern int cifs_fill_sockaddr(struct sockaddr *dst, const char *src, int len, const unsigned short int port); -extern int map_smb_to_linux_error(struct smb_hdr *smb, int logErr); +extern int map_smb_to_linux_error(struct smb_hdr *smb, bool logErr); extern void header_assemble(struct smb_hdr *, char /* command */ , - const struct cifsTconInfo *, int /* length of + const struct cifs_tcon *, int /* length of fixed section (word count) in two byte units */); extern int small_smb_init_no_tc(const int smb_cmd, const int wct, - struct cifsSesInfo *ses, + struct cifs_ses *ses, void **request_buf); -extern int CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, +extern int CIFS_SessSetup(unsigned int xid, struct cifs_ses *ses, const struct nls_table *nls_cp); extern __u16 GetNextMid(struct TCP_Server_Info *server); extern struct timespec cifs_NTtimeToUnix(__le64 utc_nanoseconds_since_1601); @@ -143,103 +148,111 @@ extern int cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, extern int mode_to_cifs_acl(struct inode *inode, const char *path, __u64); extern struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *, struct inode *, const char *, u32 *); +extern int set_cifs_acl(struct cifs_ntsd *, __u32, struct inode *, + const char *); -extern int cifs_mount(struct super_block *, struct cifs_sb_info *, char *, - const char *); -extern int cifs_umount(struct super_block *, struct cifs_sb_info *); +extern void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, + struct cifs_sb_info *cifs_sb); +extern int cifs_match_super(struct super_block *, void *); +extern void cifs_cleanup_volume_info(struct smb_vol **pvolume_info); +extern int cifs_setup_volume_info(struct smb_vol **pvolume_info, + char *mount_data, const char *devname); +extern int cifs_mount(struct cifs_sb_info *, struct smb_vol *); +extern void cifs_umount(struct cifs_sb_info *); extern void cifs_dfs_release_automount_timer(void); void cifs_proc_init(void); void cifs_proc_clean(void); extern int cifs_negotiate_protocol(unsigned int xid, - struct cifsSesInfo *ses); -extern int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, + struct cifs_ses *ses); +extern int cifs_setup_session(unsigned int xid, struct cifs_ses *ses, struct nls_table *nls_info); -extern int CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses); +extern int CIFSSMBNegotiate(unsigned int xid, struct cifs_ses *ses); -extern int CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, - const char *tree, struct cifsTconInfo *tcon, +extern int CIFSTCon(unsigned int xid, struct cifs_ses *ses, + const char *tree, struct cifs_tcon *tcon, const struct nls_table *); -extern int CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, +extern int CIFSFindFirst(const int xid, struct cifs_tcon *tcon, const char *searchName, const struct nls_table *nls_codepage, __u16 *searchHandle, struct cifs_search_info *psrch_inf, int map, const char dirsep); -extern int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, +extern int CIFSFindNext(const int xid, struct cifs_tcon *tcon, __u16 searchHandle, struct cifs_search_info *psrch_inf); -extern int CIFSFindClose(const int, struct cifsTconInfo *tcon, +extern int CIFSFindClose(const int, struct cifs_tcon *tcon, const __u16 search_handle); -extern int CIFSSMBQFileInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBQFileInfo(const int xid, struct cifs_tcon *tcon, u16 netfid, FILE_ALL_INFO *pFindData); -extern int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBQPathInfo(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, FILE_ALL_INFO *findData, int legacy /* whether to use old info level */, const struct nls_table *nls_codepage, int remap); -extern int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, +extern int SMBQueryInformation(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, FILE_ALL_INFO *findData, const struct nls_table *nls_codepage, int remap); -extern int CIFSSMBUnixQFileInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBUnixQFileInfo(const int xid, struct cifs_tcon *tcon, u16 netfid, FILE_UNIX_BASIC_INFO *pFindData); extern int CIFSSMBUnixQPathInfo(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const unsigned char *searchName, FILE_UNIX_BASIC_INFO *pFindData, const struct nls_table *nls_codepage, int remap); -extern int CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, +extern int CIFSGetDFSRefer(const int xid, struct cifs_ses *ses, const unsigned char *searchName, struct dfs_info3_param **target_nodes, unsigned int *number_of_nodes_in_array, const struct nls_table *nls_codepage, int remap); -extern int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, +extern int get_dfs_path(int xid, struct cifs_ses *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, unsigned int *pnum_referrals, struct dfs_info3_param **preferrals, int remap); -extern void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, - struct super_block *sb, struct smb_vol *vol); -extern int CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, +extern void reset_cifs_unix_caps(int xid, struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + struct smb_vol *vol); +extern int CIFSSMBQFSInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData); -extern int SMBOldQFSInfo(const int xid, struct cifsTconInfo *tcon, +extern int SMBOldQFSInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData); -extern int CIFSSMBSetFSUnixInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetFSUnixInfo(const int xid, struct cifs_tcon *tcon, __u64 cap); extern int CIFSSMBQFSAttributeInfo(const int xid, - struct cifsTconInfo *tcon); -extern int CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon); -extern int CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon); -extern int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, + struct cifs_tcon *tcon); +extern int CIFSSMBQFSDeviceInfo(const int xid, struct cifs_tcon *tcon); +extern int CIFSSMBQFSUnixInfo(const int xid, struct cifs_tcon *tcon); +extern int CIFSSMBQFSPosixInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData); -extern int CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetPathInfo(const int xid, struct cifs_tcon *tcon, const char *fileName, const FILE_BASIC_INFO *data, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetFileInfo(const int xid, struct cifs_tcon *tcon, const FILE_BASIC_INFO *data, __u16 fid, __u32 pid_of_opener); -extern int CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetFileDisposition(const int xid, struct cifs_tcon *tcon, bool delete_file, __u16 fid, __u32 pid_of_opener); #if 0 -extern int CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetAttrLegacy(int xid, struct cifs_tcon *tcon, char *fileName, __u16 dos_attributes, const struct nls_table *nls_codepage); #endif /* possibly unneeded function */ -extern int CIFSSMBSetEOF(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetEOF(const int xid, struct cifs_tcon *tcon, const char *fileName, __u64 size, bool setAllocationSizeFlag, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetFileSize(const int xid, struct cifs_tcon *tcon, __u64 size, __u16 fileHandle, __u32 opener_pid, bool AllocSizeFlag); @@ -253,121 +266,116 @@ struct cifs_unix_set_info_args { dev_t device; }; -extern int CIFSSMBUnixSetFileInfo(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBUnixSetFileInfo(const int xid, struct cifs_tcon *tcon, const struct cifs_unix_set_info_args *args, u16 fid, u32 pid_of_opener); -extern int CIFSSMBUnixSetPathInfo(const int xid, struct cifsTconInfo *pTcon, +extern int CIFSSMBUnixSetPathInfo(const int xid, struct cifs_tcon *pTcon, char *fileName, const struct cifs_unix_set_info_args *args, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBMkDir(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBMkDir(const int xid, struct cifs_tcon *tcon, const char *newName, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBRmDir(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBRmDir(const int xid, struct cifs_tcon *tcon, const char *name, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSPOSIXDelFile(const int xid, struct cifsTconInfo *tcon, +extern int CIFSPOSIXDelFile(const int xid, struct cifs_tcon *tcon, const char *name, __u16 type, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBDelFile(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBDelFile(const int xid, struct cifs_tcon *tcon, const char *name, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBRename(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, +extern int CIFSSMBRenameOpenFile(const int xid, struct cifs_tcon *pTcon, int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSCreateHardLink(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSUnixCreateHardLink(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSUnixCreateSymLink(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage); extern int CIFSSMBUnixQuerySymLink(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const unsigned char *searchName, char **syminfo, const struct nls_table *nls_codepage); +#ifdef CONFIG_CIFS_SYMLINK_EXPERIMENTAL extern int CIFSSMBQueryReparseLinkInfo(const int xid, - struct cifsTconInfo *tcon, + struct cifs_tcon *tcon, const unsigned char *searchName, char *symlinkinfo, const int buflen, __u16 fid, const struct nls_table *nls_codepage); - -extern int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, +#endif /* temporarily unused until cifs_symlink fixed */ +extern int CIFSSMBOpen(const int xid, struct cifs_tcon *tcon, const char *fileName, const int disposition, const int access_flags, const int omode, __u16 *netfid, int *pOplock, FILE_ALL_INFO *, const struct nls_table *nls_codepage, int remap); -extern int SMBLegacyOpen(const int xid, struct cifsTconInfo *tcon, +extern int SMBLegacyOpen(const int xid, struct cifs_tcon *tcon, const char *fileName, const int disposition, const int access_flags, const int omode, __u16 *netfid, int *pOplock, FILE_ALL_INFO *, const struct nls_table *nls_codepage, int remap); -extern int CIFSPOSIXCreate(const int xid, struct cifsTconInfo *tcon, +extern int CIFSPOSIXCreate(const int xid, struct cifs_tcon *tcon, u32 posix_flags, __u64 mode, __u16 *netfid, FILE_UNIX_BASIC_INFO *pRetData, __u32 *pOplock, const char *name, const struct nls_table *nls_codepage, int remap); -extern int CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBClose(const int xid, struct cifs_tcon *tcon, const int smb_file_id); -extern int CIFSSMBFlush(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBFlush(const int xid, struct cifs_tcon *tcon, const int smb_file_id); -extern int CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, - const int netfid, unsigned int count, - const __u64 lseek, unsigned int *nbytes, char **buf, +extern int CIFSSMBRead(const int xid, struct cifs_io_parms *io_parms, + unsigned int *nbytes, char **buf, int *return_buf_type); -extern int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, - const int netfid, const unsigned int count, - const __u64 lseek, unsigned int *nbytes, - const char *buf, const char __user *ubuf, +extern int CIFSSMBWrite(const int xid, struct cifs_io_parms *io_parms, + unsigned int *nbytes, const char *buf, + const char __user *ubuf, const int long_op); +extern int CIFSSMBWrite2(const int xid, struct cifs_io_parms *io_parms, + unsigned int *nbytes, struct kvec *iov, const int nvec, const int long_op); -extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, - const int netfid, const unsigned int count, - const __u64 offset, unsigned int *nbytes, - struct kvec *iov, const int nvec, const int long_op); -extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, +extern int CIFSGetSrvInodeNumber(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, __u64 *inode_number, const struct nls_table *nls_codepage, int remap_special_chars); -extern int cifsConvertToUCS(__le16 *target, const char *source, int maxlen, - const struct nls_table *cp, int mapChars); -extern int CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBLock(const int xid, struct cifs_tcon *tcon, const __u16 netfid, const __u64 len, const __u64 offset, const __u32 numUnlock, const __u32 numLock, const __u8 lockType, const bool waitFlag, const __u8 oplock_level); -extern int CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBPosixLock(const int xid, struct cifs_tcon *tcon, const __u16 smb_file_id, const int get_flag, const __u64 len, struct file_lock *, const __u16 lock_type, const bool waitFlag); -extern int CIFSSMBTDis(const int xid, struct cifsTconInfo *tcon); +extern int CIFSSMBTDis(const int xid, struct cifs_tcon *tcon); extern int CIFSSMBEcho(struct TCP_Server_Info *server); -extern int CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses); +extern int CIFSSMBLogoff(const int xid, struct cifs_ses *ses); -extern struct cifsSesInfo *sesInfoAlloc(void); -extern void sesInfoFree(struct cifsSesInfo *); -extern struct cifsTconInfo *tconInfoAlloc(void); -extern void tconInfoFree(struct cifsTconInfo *); +extern struct cifs_ses *sesInfoAlloc(void); +extern void sesInfoFree(struct cifs_ses *); +extern struct cifs_tcon *tconInfoAlloc(void); +extern void tconInfoFree(struct cifs_tcon *); extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *, __u32 *); extern int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *, @@ -376,49 +384,51 @@ extern int cifs_verify_signature(struct smb_hdr *, struct TCP_Server_Info *server, __u32 expected_sequence_number); extern int SMBNTencrypt(unsigned char *, unsigned char *, unsigned char *); -extern int setup_ntlm_response(struct cifsSesInfo *); -extern int setup_ntlmv2_rsp(struct cifsSesInfo *, const struct nls_table *); +extern int setup_ntlm_response(struct cifs_ses *); +extern int setup_ntlmv2_rsp(struct cifs_ses *, const struct nls_table *); extern int cifs_crypto_shash_allocate(struct TCP_Server_Info *); extern void cifs_crypto_shash_release(struct TCP_Server_Info *); -extern int calc_seckey(struct cifsSesInfo *); +extern int calc_seckey(struct cifs_ses *); #ifdef CONFIG_CIFS_WEAK_PW_HASH -extern void calc_lanman_hash(const char *password, const char *cryptkey, +extern int calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, char *lnm_session_key); #endif /* CIFS_WEAK_PW_HASH */ +#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */ +extern int CIFSSMBNotify(const int xid, struct cifs_tcon *tcon, + const int notify_subdirs, const __u16 netfid, + __u32 filter, struct file *file, int multishot, + const struct nls_table *nls_codepage); +#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */ extern int CIFSSMBCopy(int xid, - struct cifsTconInfo *source_tcon, + struct cifs_tcon *source_tcon, const char *fromName, const __u16 target_tid, const char *toName, const int flags, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, - const int notify_subdirs, const __u16 netfid, - __u32 filter, struct file *file, int multishot, - const struct nls_table *nls_codepage); -extern ssize_t CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon, +extern ssize_t CIFSSMBQAllEAs(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, const unsigned char *ea_name, char *EAData, size_t bufsize, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetEA(const int xid, struct cifs_tcon *tcon, const char *fileName, const char *ea_name, const void *ea_value, const __u16 ea_value_len, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBGetCIFSACL(const int xid, struct cifs_tcon *tcon, __u16 fid, struct cifs_ntsd **acl_inf, __u32 *buflen); -extern int CIFSSMBSetCIFSACL(const int, struct cifsTconInfo *, __u16, +extern int CIFSSMBSetCIFSACL(const int, struct cifs_tcon *, __u16, struct cifs_ntsd *, __u32); -extern int CIFSSMBGetPosixACL(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBGetPosixACL(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, char *acl_inf, const int buflen, const int acl_type, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBSetPosixACL(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetPosixACL(const int xid, struct cifs_tcon *tcon, const unsigned char *fileName, const char *local_acl, const int buflen, const int acl_type, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon, +extern int CIFSGetExtAttr(const int xid, struct cifs_tcon *tcon, const int netfid, __u64 *pExtAttrBits, __u64 *pMask); extern void cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb); extern bool CIFSCouldBeMFSymlink(const struct cifs_fattr *fattr); @@ -427,9 +437,24 @@ extern int CIFSCheckMFSymlink(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb, int xid); extern int mdfour(unsigned char *, unsigned char *, int); extern int E_md4hash(const unsigned char *passwd, unsigned char *p16); -extern void SMBencrypt(unsigned char *passwd, const unsigned char *c8, - unsigned char *p24); -extern void E_P16(unsigned char *p14, unsigned char *p16); -extern void E_P24(unsigned char *p21, const unsigned char *c8, +extern int SMBencrypt(unsigned char *passwd, const unsigned char *c8, unsigned char *p24); + +/* asynchronous write support */ +struct cifs_writedata { + struct kref refcount; + enum writeback_sync_modes sync_mode; + struct work_struct work; + struct cifsFileInfo *cfile; + __u64 offset; + unsigned int bytes; + int result; + unsigned int nr_pages; + struct page *pages[1]; +}; + +int cifs_async_writev(struct cifs_writedata *wdata); +struct cifs_writedata *cifs_writedata_alloc(unsigned int nr_pages); +void cifs_writedata_release(struct kref *refcount); + #endif /* _CIFSPROTO_H */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index df959ba..1a9fe7f 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -32,6 +32,7 @@ #include <linux/vfs.h> #include <linux/slab.h> #include <linux/posix_acl_xattr.h> +#include <linux/pagemap.h> #include <asm/uaccess.h> #include "cifspdu.h" #include "cifsglob.h" @@ -84,7 +85,7 @@ static struct { /* Mark as invalid, all open files on tree connections since they were closed when session to server was lost */ -static void mark_open_files_invalid(struct cifsTconInfo *pTcon) +static void mark_open_files_invalid(struct cifs_tcon *pTcon) { struct cifsFileInfo *open_file = NULL; struct list_head *tmp; @@ -104,10 +105,10 @@ static void mark_open_files_invalid(struct cifsTconInfo *pTcon) /* reconnect the socket, tcon, and smb session if needed */ static int -cifs_reconnect_tcon(struct cifsTconInfo *tcon, int smb_command) +cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) { int rc = 0; - struct cifsSesInfo *ses; + struct cifs_ses *ses; struct TCP_Server_Info *server; struct nls_table *nls_codepage; @@ -226,7 +227,7 @@ out: SMB information in the SMB header. If the return code is zero, this function must have filled in request_buf pointer */ static int -small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, +small_smb_init(int smb_command, int wct, struct cifs_tcon *tcon, void **request_buf) { int rc; @@ -252,7 +253,7 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, int small_smb_init_no_tc(const int smb_command, const int wct, - struct cifsSesInfo *ses, void **request_buf) + struct cifs_ses *ses, void **request_buf) { int rc; struct smb_hdr *buffer; @@ -278,7 +279,7 @@ small_smb_init_no_tc(const int smb_command, const int wct, /* If the return code is zero, this function must fill in request_buf pointer */ static int -__smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, +__smb_init(int smb_command, int wct, struct cifs_tcon *tcon, void **request_buf, void **response_buf) { *request_buf = cifs_buf_get(); @@ -304,7 +305,7 @@ __smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, /* If the return code is zero, this function must fill in request_buf pointer */ static int -smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, +smb_init(int smb_command, int wct, struct cifs_tcon *tcon, void **request_buf, void **response_buf) { int rc; @@ -317,7 +318,7 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, } static int -smb_init_no_reconnect(int smb_command, int wct, struct cifsTconInfo *tcon, +smb_init_no_reconnect(int smb_command, int wct, struct cifs_tcon *tcon, void **request_buf, void **response_buf) { if (tcon->ses->need_reconnect || tcon->need_reconnect) @@ -339,12 +340,13 @@ static int validate_t2(struct smb_t2_rsp *pSMB) get_unaligned_le16(&pSMB->t2_rsp.DataOffset) > 1024) goto vt2_err; - /* check that bcc is at least as big as parms + data */ - /* check that bcc is less than negotiated smb buffer */ total_size = get_unaligned_le16(&pSMB->t2_rsp.ParameterCount); if (total_size >= 512) goto vt2_err; + /* check that bcc is at least as big as parms + data, and that it is + * less than negotiated smb buffer + */ total_size += get_unaligned_le16(&pSMB->t2_rsp.DataCount); if (total_size > get_bcc(&pSMB->hdr) || total_size >= CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) @@ -357,8 +359,15 @@ vt2_err: return -EINVAL; } +static inline void inc_rfc1001_len(void *pSMB, int count) +{ + struct smb_hdr *hdr = (struct smb_hdr *)pSMB; + + be32_add_cpu(&hdr->smb_buf_length, count); +} + int -CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) +CIFSSMBNegotiate(unsigned int xid, struct cifs_ses *ses) { NEGOTIATE_REQ *pSMB; NEGOTIATE_RSP *pSMBr; @@ -409,7 +418,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) count += strlen(protocols[i].name) + 1; /* null at end of source and target buffers anyway */ } - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); rc = SendReceive(xid, ses, (struct smb_hdr *) pSMB, @@ -442,7 +451,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) rc = -EOPNOTSUPP; goto neg_err_exit; } - server->secMode = (__u8)le16_to_cpu(rsp->SecurityMode); + server->sec_mode = (__u8)le16_to_cpu(rsp->SecurityMode); server->maxReq = le16_to_cpu(rsp->MaxMpxCount); server->maxBuf = min((__u32)le16_to_cpu(rsp->MaxBufSize), (__u32)CIFSMaxBufSize + MAX_CIFS_HDR_SIZE); @@ -496,7 +505,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { memcpy(ses->server->cryptkey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); - } else if (server->secMode & SECMODE_PW_ENCRYPT) { + } else if (server->sec_mode & SECMODE_PW_ENCRYPT) { rc = -EIO; /* need cryptkey unless plain text */ goto neg_err_exit; } @@ -518,11 +527,11 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) goto neg_err_exit; } /* else wct == 17 NTLM */ - server->secMode = pSMBr->SecurityMode; - if ((server->secMode & SECMODE_USER) == 0) + server->sec_mode = pSMBr->SecurityMode; + if ((server->sec_mode & SECMODE_USER) == 0) cFYI(1, "share mode security"); - if ((server->secMode & SECMODE_PW_ENCRYPT) == 0) + if ((server->sec_mode & SECMODE_PW_ENCRYPT) == 0) #ifdef CONFIG_CIFS_WEAK_PW_HASH if ((secFlags & CIFSSEC_MAY_PLNTXT) == 0) #endif /* CIFS_WEAK_PW_HASH */ @@ -541,10 +550,6 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->secType = RawNTLMSSP; else if (secFlags & CIFSSEC_MAY_LANMAN) server->secType = LANMAN; -/* #ifdef CONFIG_CIFS_EXPERIMENTAL - else if (secFlags & CIFSSEC_MAY_PLNTXT) - server->secType = ?? -#endif */ else { rc = -EOPNOTSUPP; cERROR(1, "Invalid security type"); @@ -566,19 +571,11 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) if (pSMBr->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { memcpy(ses->server->cryptkey, pSMBr->u.EncryptionKey, CIFS_CRYPTO_KEY_SIZE); - } else if ((pSMBr->hdr.Flags2 & SMBFLG2_EXT_SEC) - && (pSMBr->EncryptionKeyLength == 0)) { + } else if ((pSMBr->hdr.Flags2 & SMBFLG2_EXT_SEC || + server->capabilities & CAP_EXTENDED_SECURITY) && + (pSMBr->EncryptionKeyLength == 0)) { /* decode security blob */ - } else if (server->secMode & SECMODE_PW_ENCRYPT) { - rc = -EIO; /* no crypt key only if plain text pwd */ - goto neg_err_exit; - } - - /* BB might be helpful to save off the domain of server here */ - - if ((pSMBr->hdr.Flags2 & SMBFLG2_EXT_SEC) && - (server->capabilities & CAP_EXTENDED_SECURITY)) { - count = pSMBr->ByteCount; + count = get_bcc(&pSMBr->hdr); if (count < 16) { rc = -EIO; goto neg_err_exit; @@ -620,6 +617,9 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) } else rc = -EOPNOTSUPP; } + } else if (server->sec_mode & SECMODE_PW_ENCRYPT) { + rc = -EIO; /* no crypt key only if plain text pwd */ + goto neg_err_exit; } else server->capabilities &= ~CAP_EXTENDED_SECURITY; @@ -630,27 +630,27 @@ signing_check: /* MUST_SIGN already includes the MAY_SIGN FLAG so if this is zero it means that signing is disabled */ cFYI(1, "Signing disabled"); - if (server->secMode & SECMODE_SIGN_REQUIRED) { + if (server->sec_mode & SECMODE_SIGN_REQUIRED) { cERROR(1, "Server requires " "packet signing to be enabled in " "/proc/fs/cifs/SecurityFlags."); rc = -EOPNOTSUPP; } - server->secMode &= + server->sec_mode &= ~(SECMODE_SIGN_ENABLED | SECMODE_SIGN_REQUIRED); } else if ((secFlags & CIFSSEC_MUST_SIGN) == CIFSSEC_MUST_SIGN) { /* signing required */ cFYI(1, "Must sign - secFlags 0x%x", secFlags); - if ((server->secMode & + if ((server->sec_mode & (SECMODE_SIGN_ENABLED | SECMODE_SIGN_REQUIRED)) == 0) { cERROR(1, "signing required but server lacks support"); rc = -EOPNOTSUPP; } else - server->secMode |= SECMODE_SIGN_REQUIRED; + server->sec_mode |= SECMODE_SIGN_REQUIRED; } else { /* signing optional ie CIFSSEC_MAY_SIGN */ - if ((server->secMode & SECMODE_SIGN_REQUIRED) == 0) - server->secMode &= + if ((server->sec_mode & SECMODE_SIGN_REQUIRED) == 0) + server->sec_mode &= ~(SECMODE_SIGN_ENABLED | SECMODE_SIGN_REQUIRED); } @@ -662,7 +662,7 @@ neg_err_exit: } int -CIFSSMBTDis(const int xid, struct cifsTconInfo *tcon) +CIFSSMBTDis(const int xid, struct cifs_tcon *tcon) { struct smb_hdr *smb_buffer; int rc = 0; @@ -721,6 +721,7 @@ CIFSSMBEcho(struct TCP_Server_Info *server) { ECHO_REQ *smb; int rc = 0; + struct kvec iov; cFYI(1, "In echo request"); @@ -732,12 +733,13 @@ CIFSSMBEcho(struct TCP_Server_Info *server) smb->hdr.Tid = 0xffff; smb->hdr.WordCount = 1; put_unaligned_le16(1, &smb->EchoCount); - put_bcc_le(1, &smb->hdr); + put_bcc(1, &smb->hdr); smb->Data[0] = 'a'; - smb->hdr.smb_buf_length += 3; + inc_rfc1001_len(smb, 3); + iov.iov_base = smb; + iov.iov_len = be32_to_cpu(smb->hdr.smb_buf_length) + 4; - rc = cifs_call_async(server, (struct smb_hdr *)smb, - cifs_echo_callback, server); + rc = cifs_call_async(server, &iov, 1, cifs_echo_callback, server, true); if (rc) cFYI(1, "Echo request failed: %d", rc); @@ -747,7 +749,7 @@ CIFSSMBEcho(struct TCP_Server_Info *server) } int -CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses) +CIFSSMBLogoff(const int xid, struct cifs_ses *ses) { LOGOFF_ANDX_REQ *pSMB; int rc = 0; @@ -774,7 +776,7 @@ CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses) pSMB->hdr.Mid = GetNextMid(ses->server); - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) pSMB->hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; @@ -794,7 +796,7 @@ session_already_dead: } int -CIFSPOSIXDelFile(const int xid, struct cifsTconInfo *tcon, const char *fileName, +CIFSPOSIXDelFile(const int xid, struct cifs_tcon *tcon, const char *fileName, __u16 type, const struct nls_table *nls_codepage, int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; @@ -852,7 +854,7 @@ PsxDelete: pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_POSIX_UNLINK); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -869,7 +871,7 @@ PsxDelete: } int -CIFSSMBDelFile(const int xid, struct cifsTconInfo *tcon, const char *fileName, +CIFSSMBDelFile(const int xid, struct cifs_tcon *tcon, const char *fileName, const struct nls_table *nls_codepage, int remap) { DELETE_FILE_REQ *pSMB = NULL; @@ -898,7 +900,7 @@ DelFileRetry: pSMB->SearchAttributes = cpu_to_le16(ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM); pSMB->BufferFormat = 0x04; - pSMB->hdr.smb_buf_length += name_len + 1; + inc_rfc1001_len(pSMB, name_len + 1); pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -914,7 +916,7 @@ DelFileRetry: } int -CIFSSMBRmDir(const int xid, struct cifsTconInfo *tcon, const char *dirName, +CIFSSMBRmDir(const int xid, struct cifs_tcon *tcon, const char *dirName, const struct nls_table *nls_codepage, int remap) { DELETE_DIRECTORY_REQ *pSMB = NULL; @@ -942,7 +944,7 @@ RmDirRetry: } pSMB->BufferFormat = 0x04; - pSMB->hdr.smb_buf_length += name_len + 1; + inc_rfc1001_len(pSMB, name_len + 1); pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -957,7 +959,7 @@ RmDirRetry: } int -CIFSSMBMkDir(const int xid, struct cifsTconInfo *tcon, +CIFSSMBMkDir(const int xid, struct cifs_tcon *tcon, const char *name, const struct nls_table *nls_codepage, int remap) { int rc = 0; @@ -985,7 +987,7 @@ MkDirRetry: } pSMB->BufferFormat = 0x04; - pSMB->hdr.smb_buf_length += name_len + 1; + inc_rfc1001_len(pSMB, name_len + 1); pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -1000,7 +1002,7 @@ MkDirRetry: } int -CIFSPOSIXCreate(const int xid, struct cifsTconInfo *tcon, __u32 posix_flags, +CIFSPOSIXCreate(const int xid, struct cifs_tcon *tcon, __u32 posix_flags, __u64 mode, __u16 *netfid, FILE_UNIX_BASIC_INFO *pRetData, __u32 *pOplock, const char *name, const struct nls_table *nls_codepage, int remap) @@ -1063,7 +1065,7 @@ PsxCreat: pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_POSIX_OPEN); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -1075,7 +1077,7 @@ PsxCreat: cFYI(1, "copying inode info"); rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < sizeof(OPEN_PSX_RSP))) { + if (rc || get_bcc(&pSMBr->hdr) < sizeof(OPEN_PSX_RSP)) { rc = -EIO; /* bad smb */ goto psx_create_err; } @@ -1096,7 +1098,7 @@ PsxCreat: pRetData->Type = cpu_to_le32(-1); /* unknown */ cFYI(DBG2, "unknown type"); } else { - if (pSMBr->ByteCount < sizeof(OPEN_PSX_RSP) + if (get_bcc(&pSMBr->hdr) < sizeof(OPEN_PSX_RSP) + sizeof(FILE_UNIX_BASIC_INFO)) { cERROR(1, "Open response data too small"); pRetData->Type = cpu_to_le32(-1); @@ -1166,7 +1168,7 @@ access_flags_to_smbopen_mode(const int access_flags) } int -SMBLegacyOpen(const int xid, struct cifsTconInfo *tcon, +SMBLegacyOpen(const int xid, struct cifs_tcon *tcon, const char *fileName, const int openDisposition, const int access_flags, const int create_options, __u16 *netfid, int *pOplock, FILE_ALL_INFO *pfile_info, @@ -1228,7 +1230,7 @@ OldOpenRetry: pSMB->Sattr = cpu_to_le16(ATTR_HIDDEN | ATTR_SYSTEM | ATTR_DIRECTORY); pSMB->OpenFunction = cpu_to_le16(convert_disposition(openDisposition)); count += name_len; - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); /* long_op set to 1 to allow for oplock break timeouts */ @@ -1273,7 +1275,7 @@ OldOpenRetry: } int -CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, +CIFSSMBOpen(const int xid, struct cifs_tcon *tcon, const char *fileName, const int openDisposition, const int access_flags, const int create_options, __u16 *netfid, int *pOplock, FILE_ALL_INFO *pfile_info, @@ -1341,7 +1343,7 @@ openRetry: SECURITY_CONTEXT_TRACKING | SECURITY_EFFECTIVE_ONLY; count += name_len; - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); /* long_op set to 1 to allow for oplock break timeouts */ @@ -1375,8 +1377,7 @@ openRetry: } int -CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid, - const unsigned int count, const __u64 lseek, unsigned int *nbytes, +CIFSSMBRead(const int xid, struct cifs_io_parms *io_parms, unsigned int *nbytes, char **buf, int *pbuf_type) { int rc = -EACCES; @@ -1386,13 +1387,18 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid, int wct; int resp_buf_type = 0; struct kvec iov[1]; + __u32 pid = io_parms->pid; + __u16 netfid = io_parms->netfid; + __u64 offset = io_parms->offset; + struct cifs_tcon *tcon = io_parms->tcon; + unsigned int count = io_parms->length; cFYI(1, "Reading %d bytes on fid %d", count, netfid); if (tcon->ses->capabilities & CAP_LARGE_FILES) wct = 12; else { wct = 10; /* old style read */ - if ((lseek >> 32) > 0) { + if ((offset >> 32) > 0) { /* can not handle this big offset for old */ return -EIO; } @@ -1403,15 +1409,18 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid, if (rc) return rc; + pSMB->hdr.Pid = cpu_to_le16((__u16)pid); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid >> 16)); + /* tcon and ses pointer are checked in smb_init */ if (tcon->ses->server == NULL) return -ECONNABORTED; pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; - pSMB->OffsetLow = cpu_to_le32(lseek & 0xFFFFFFFF); + pSMB->OffsetLow = cpu_to_le32(offset & 0xFFFFFFFF); if (wct == 12) - pSMB->OffsetHigh = cpu_to_le32(lseek >> 32); + pSMB->OffsetHigh = cpu_to_le32(offset >> 32); pSMB->Remaining = 0; pSMB->MaxCount = cpu_to_le16(count & 0xFFFF); @@ -1426,7 +1435,7 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid, } iov[0].iov_base = (char *)pSMB; - iov[0].iov_len = pSMB->hdr.smb_buf_length + 4; + iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4; rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovecs */, &resp_buf_type, CIFS_LOG_ERROR); cifs_stats_inc(&tcon->num_reads); @@ -1480,9 +1489,8 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid, int -CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, - const int netfid, const unsigned int count, - const __u64 offset, unsigned int *nbytes, const char *buf, +CIFSSMBWrite(const int xid, struct cifs_io_parms *io_parms, + unsigned int *nbytes, const char *buf, const char __user *ubuf, const int long_op) { int rc = -EACCES; @@ -1491,6 +1499,11 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, int bytes_returned, wct; __u32 bytes_sent; __u16 byte_count; + __u32 pid = io_parms->pid; + __u16 netfid = io_parms->netfid; + __u64 offset = io_parms->offset; + struct cifs_tcon *tcon = io_parms->tcon; + unsigned int count = io_parms->length; *nbytes = 0; @@ -1512,6 +1525,10 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, (void **) &pSMBr); if (rc) return rc; + + pSMB->hdr.Pid = cpu_to_le16((__u16)pid); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid >> 16)); + /* tcon and ses pointer are checked in smb_init */ if (tcon->ses->server == NULL) return -ECONNABORTED; @@ -1560,7 +1577,7 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, pSMB->DataLengthLow = cpu_to_le16(bytes_sent & 0xFFFF); pSMB->DataLengthHigh = cpu_to_le16(bytes_sent >> 16); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); if (wct == 14) pSMB->ByteCount = cpu_to_le16(byte_count); @@ -1598,17 +1615,259 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, return rc; } +void +cifs_writedata_release(struct kref *refcount) +{ + struct cifs_writedata *wdata = container_of(refcount, + struct cifs_writedata, refcount); + + if (wdata->cfile) + cifsFileInfo_put(wdata->cfile); + + kfree(wdata); +} + +/* + * Write failed with a retryable error. Resend the write request. It's also + * possible that the page was redirtied so re-clean the page. + */ +static void +cifs_writev_requeue(struct cifs_writedata *wdata) +{ + int i, rc; + struct inode *inode = wdata->cfile->dentry->d_inode; + + for (i = 0; i < wdata->nr_pages; i++) { + lock_page(wdata->pages[i]); + clear_page_dirty_for_io(wdata->pages[i]); + } + + do { + rc = cifs_async_writev(wdata); + } while (rc == -EAGAIN); + + for (i = 0; i < wdata->nr_pages; i++) { + if (rc != 0) + SetPageError(wdata->pages[i]); + unlock_page(wdata->pages[i]); + } + + mapping_set_error(inode->i_mapping, rc); + kref_put(&wdata->refcount, cifs_writedata_release); +} + +static void +cifs_writev_complete(struct work_struct *work) +{ + struct cifs_writedata *wdata = container_of(work, + struct cifs_writedata, work); + struct inode *inode = wdata->cfile->dentry->d_inode; + int i = 0; + + if (wdata->result == 0) { + cifs_update_eof(CIFS_I(inode), wdata->offset, wdata->bytes); + cifs_stats_bytes_written(tlink_tcon(wdata->cfile->tlink), + wdata->bytes); + } else if (wdata->sync_mode == WB_SYNC_ALL && wdata->result == -EAGAIN) + return cifs_writev_requeue(wdata); + + for (i = 0; i < wdata->nr_pages; i++) { + struct page *page = wdata->pages[i]; + if (wdata->result == -EAGAIN) + __set_page_dirty_nobuffers(page); + else if (wdata->result < 0) + SetPageError(page); + end_page_writeback(page); + page_cache_release(page); + } + if (wdata->result != -EAGAIN) + mapping_set_error(inode->i_mapping, wdata->result); + kref_put(&wdata->refcount, cifs_writedata_release); +} + +struct cifs_writedata * +cifs_writedata_alloc(unsigned int nr_pages) +{ + struct cifs_writedata *wdata; + + /* this would overflow */ + if (nr_pages == 0) { + cERROR(1, "%s: called with nr_pages == 0!", __func__); + return NULL; + } + + /* writedata + number of page pointers */ + wdata = kzalloc(sizeof(*wdata) + + sizeof(struct page *) * (nr_pages - 1), GFP_NOFS); + if (wdata != NULL) { + INIT_WORK(&wdata->work, cifs_writev_complete); + kref_init(&wdata->refcount); + } + return wdata; +} + +/* + * Check the midState and signature on received buffer (if any), and queue the + * workqueue completion task. + */ +static void +cifs_writev_callback(struct mid_q_entry *mid) +{ + struct cifs_writedata *wdata = mid->callback_data; + struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); + unsigned int written; + WRITE_RSP *smb = (WRITE_RSP *)mid->resp_buf; + + switch (mid->midState) { + case MID_RESPONSE_RECEIVED: + wdata->result = cifs_check_receive(mid, tcon->ses->server, 0); + if (wdata->result != 0) + break; + + written = le16_to_cpu(smb->CountHigh); + written <<= 16; + written += le16_to_cpu(smb->Count); + /* + * Mask off high 16 bits when bytes written as returned + * by the server is greater than bytes requested by the + * client. OS/2 servers are known to set incorrect + * CountHigh values. + */ + if (written > wdata->bytes) + written &= 0xFFFF; + + if (written < wdata->bytes) + wdata->result = -ENOSPC; + else + wdata->bytes = written; + break; + case MID_REQUEST_SUBMITTED: + case MID_RETRY_NEEDED: + wdata->result = -EAGAIN; + break; + default: + wdata->result = -EIO; + break; + } + + queue_work(system_nrt_wq, &wdata->work); + DeleteMidQEntry(mid); + atomic_dec(&tcon->ses->server->inFlight); + wake_up(&tcon->ses->server->request_q); +} + +/* cifs_async_writev - send an async write, and set up mid to handle result */ +int +cifs_async_writev(struct cifs_writedata *wdata) +{ + int i, rc = -EACCES; + WRITE_REQ *smb = NULL; + int wct; + struct cifs_tcon *tcon = tlink_tcon(wdata->cfile->tlink); + struct inode *inode = wdata->cfile->dentry->d_inode; + struct kvec *iov = NULL; + + if (tcon->ses->capabilities & CAP_LARGE_FILES) { + wct = 14; + } else { + wct = 12; + if (wdata->offset >> 32 > 0) { + /* can not handle big offset for old srv */ + return -EIO; + } + } + + rc = small_smb_init(SMB_COM_WRITE_ANDX, wct, tcon, (void **)&smb); + if (rc) + goto async_writev_out; + + /* 1 iov per page + 1 for header */ + iov = kzalloc((wdata->nr_pages + 1) * sizeof(*iov), GFP_NOFS); + if (iov == NULL) { + rc = -ENOMEM; + goto async_writev_out; + } + + smb->hdr.Pid = cpu_to_le16((__u16)wdata->cfile->pid); + smb->hdr.PidHigh = cpu_to_le16((__u16)(wdata->cfile->pid >> 16)); + + smb->AndXCommand = 0xFF; /* none */ + smb->Fid = wdata->cfile->netfid; + smb->OffsetLow = cpu_to_le32(wdata->offset & 0xFFFFFFFF); + if (wct == 14) + smb->OffsetHigh = cpu_to_le32(wdata->offset >> 32); + smb->Reserved = 0xFFFFFFFF; + smb->WriteMode = 0; + smb->Remaining = 0; + + smb->DataOffset = + cpu_to_le16(offsetof(struct smb_com_write_req, Data) - 4); + + /* 4 for RFC1001 length + 1 for BCC */ + iov[0].iov_len = be32_to_cpu(smb->hdr.smb_buf_length) + 4 + 1; + iov[0].iov_base = smb; + + /* marshal up the pages into iov array */ + wdata->bytes = 0; + for (i = 0; i < wdata->nr_pages; i++) { + iov[i + 1].iov_len = min(inode->i_size - + page_offset(wdata->pages[i]), + (loff_t)PAGE_CACHE_SIZE); + iov[i + 1].iov_base = kmap(wdata->pages[i]); + wdata->bytes += iov[i + 1].iov_len; + } + + cFYI(1, "async write at %llu %u bytes", wdata->offset, wdata->bytes); + + smb->DataLengthLow = cpu_to_le16(wdata->bytes & 0xFFFF); + smb->DataLengthHigh = cpu_to_le16(wdata->bytes >> 16); + + if (wct == 14) { + inc_rfc1001_len(&smb->hdr, wdata->bytes + 1); + put_bcc(wdata->bytes + 1, &smb->hdr); + } else { + /* wct == 12 */ + struct smb_com_writex_req *smbw = + (struct smb_com_writex_req *)smb; + inc_rfc1001_len(&smbw->hdr, wdata->bytes + 5); + put_bcc(wdata->bytes + 5, &smbw->hdr); + iov[0].iov_len += 4; /* pad bigger by four bytes */ + } + + kref_get(&wdata->refcount); + rc = cifs_call_async(tcon->ses->server, iov, wdata->nr_pages + 1, + cifs_writev_callback, wdata, false); + + if (rc == 0) + cifs_stats_inc(&tcon->num_writes); + else + kref_put(&wdata->refcount, cifs_writedata_release); + + /* send is done, unmap pages */ + for (i = 0; i < wdata->nr_pages; i++) + kunmap(wdata->pages[i]); + +async_writev_out: + cifs_small_buf_release(smb); + kfree(iov); + return rc; +} + int -CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, - const int netfid, const unsigned int count, - const __u64 offset, unsigned int *nbytes, struct kvec *iov, - int n_vec, const int long_op) +CIFSSMBWrite2(const int xid, struct cifs_io_parms *io_parms, + unsigned int *nbytes, struct kvec *iov, int n_vec, + const int long_op) { int rc = -EACCES; WRITE_REQ *pSMB = NULL; int wct; int smb_hdr_len; int resp_buf_type = 0; + __u32 pid = io_parms->pid; + __u16 netfid = io_parms->netfid; + __u64 offset = io_parms->offset; + struct cifs_tcon *tcon = io_parms->tcon; + unsigned int count = io_parms->length; *nbytes = 0; @@ -1626,6 +1885,10 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, rc = small_smb_init(SMB_COM_WRITE_ANDX, wct, tcon, (void **) &pSMB); if (rc) return rc; + + pSMB->hdr.Pid = cpu_to_le16((__u16)pid); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid >> 16)); + /* tcon and ses pointer are checked in smb_init */ if (tcon->ses->server == NULL) return -ECONNABORTED; @@ -1644,11 +1907,12 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, pSMB->DataLengthLow = cpu_to_le16(count & 0xFFFF); pSMB->DataLengthHigh = cpu_to_le16(count >> 16); - smb_hdr_len = pSMB->hdr.smb_buf_length + 1; /* hdr + 1 byte pad */ + /* header + 1 byte pad */ + smb_hdr_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 1; if (wct == 14) - pSMB->hdr.smb_buf_length += count+1; + inc_rfc1001_len(pSMB, count + 1); else /* wct == 12 */ - pSMB->hdr.smb_buf_length += count+5; /* smb data starts later */ + inc_rfc1001_len(pSMB, count + 5); /* smb data starts later */ if (wct == 14) pSMB->ByteCount = cpu_to_le16(count + 1); else /* wct == 12 */ /* bigger pad, smaller smb hdr, keep offset ok */ { @@ -1700,7 +1964,7 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, int -CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, +CIFSSMBLock(const int xid, struct cifs_tcon *tcon, const __u16 smb_file_id, const __u64 len, const __u64 offset, const __u32 numUnlock, const __u32 numLock, const __u8 lockType, @@ -1748,7 +2012,7 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, /* oplock break */ count = 0; } - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); if (waitFlag) { @@ -1770,7 +2034,7 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, } int -CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, +CIFSSMBPosixLock(const int xid, struct cifs_tcon *tcon, const __u16 smb_file_id, const int get_flag, const __u64 len, struct file_lock *pLockData, const __u16 lock_type, const bool waitFlag) @@ -1839,14 +2103,14 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, pSMB->Fid = smb_file_id; pSMB->InformationLevel = cpu_to_le16(SMB_SET_POSIX_LOCK); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); if (waitFlag) { rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned); } else { iov[0].iov_base = (char *)pSMB; - iov[0].iov_len = pSMB->hdr.smb_buf_length + 4; + iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4; rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovecs */, &resp_buf_type, timeout); pSMB = NULL; /* request buf already freed by SendReceive2. Do @@ -1862,7 +2126,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, __u16 data_count; rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < sizeof(struct cifs_posix_lock))) { + if (rc || get_bcc(&pSMBr->hdr) < sizeof(*parm_data)) { rc = -EIO; /* bad smb */ goto plk_err_exit; } @@ -1908,7 +2172,7 @@ plk_err_exit: int -CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) +CIFSSMBClose(const int xid, struct cifs_tcon *tcon, int smb_file_id) { int rc = 0; CLOSE_REQ *pSMB = NULL; @@ -1941,7 +2205,7 @@ CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) } int -CIFSSMBFlush(const int xid, struct cifsTconInfo *tcon, int smb_file_id) +CIFSSMBFlush(const int xid, struct cifs_tcon *tcon, int smb_file_id) { int rc = 0; FLUSH_REQ *pSMB = NULL; @@ -1962,7 +2226,7 @@ CIFSSMBFlush(const int xid, struct cifsTconInfo *tcon, int smb_file_id) } int -CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, +CIFSSMBRename(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap) { @@ -2012,7 +2276,7 @@ renameRetry: } count = 1 /* 1st signature byte */ + name_len + name_len2; - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2029,7 +2293,7 @@ renameRetry: return rc; } -int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, +int CIFSSMBRenameOpenFile(const int xid, struct cifs_tcon *pTcon, int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap) { @@ -2092,7 +2356,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_RENAME_INFORMATION); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, pTcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -2109,7 +2373,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, } int -CIFSSMBCopy(const int xid, struct cifsTconInfo *tcon, const char *fromName, +CIFSSMBCopy(const int xid, struct cifs_tcon *tcon, const char *fromName, const __u16 target_tid, const char *toName, const int flags, const struct nls_table *nls_codepage, int remap) { @@ -2159,7 +2423,7 @@ copyRetry: } count = 1 /* 1st signature byte */ + name_len + name_len2; - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2177,7 +2441,7 @@ copyRetry: } int -CIFSUnixCreateSymLink(const int xid, struct cifsTconInfo *tcon, +CIFSUnixCreateSymLink(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage) { @@ -2249,7 +2513,7 @@ createSymLinkRetry: pSMB->DataOffset = cpu_to_le16(offset); pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_LINK); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -2266,7 +2530,7 @@ createSymLinkRetry: } int -CIFSUnixCreateHardLink(const int xid, struct cifsTconInfo *tcon, +CIFSUnixCreateHardLink(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap) { @@ -2335,7 +2599,7 @@ createHardLinkRetry: pSMB->DataOffset = cpu_to_le16(offset); pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_HLINK); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -2351,7 +2615,7 @@ createHardLinkRetry: } int -CIFSCreateHardLink(const int xid, struct cifsTconInfo *tcon, +CIFSCreateHardLink(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap) { @@ -2406,7 +2670,7 @@ winCreateHardLinkRetry: } count = 1 /* string type byte */ + name_len + name_len2; - pSMB->hdr.smb_buf_length += count; + inc_rfc1001_len(pSMB, count); pSMB->ByteCount = cpu_to_le16(count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2423,7 +2687,7 @@ winCreateHardLinkRetry: } int -CIFSSMBUnixQuerySymLink(const int xid, struct cifsTconInfo *tcon, +CIFSSMBUnixQuerySymLink(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, char **symlinkinfo, const struct nls_table *nls_codepage) { @@ -2477,7 +2741,7 @@ querySymLinkRetry: pSMB->ParameterCount = pSMB->TotalParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_LINK); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2489,7 +2753,7 @@ querySymLinkRetry: rc = validate_t2((struct smb_t2_rsp *)pSMBr); /* BB also check enough total bytes returned */ - if (rc || (pSMBr->ByteCount < 2)) + if (rc || get_bcc(&pSMBr->hdr) < 2) rc = -EIO; else { bool is_unicode; @@ -2516,9 +2780,19 @@ querySymLinkRetry: return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CONFIG_CIFS_SYMLINK_EXPERIMENTAL +/* + * Recent Windows versions now create symlinks more frequently + * and they use the "reparse point" mechanism below. We can of course + * do symlinks nicely to Samba and other servers which support the + * CIFS Unix Extensions and we can also do SFU symlinks and "client only" + * "MF" symlinks optionally, but for recent Windows we really need to + * reenable the code below and fix the cifs_symlink callers to handle this. + * In the interim this code has been moved to its own config option so + * it is not compiled in by default until callers fixed up and more tested. + */ int -CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBQueryReparseLinkInfo(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, char *symlinkinfo, const int buflen, __u16 fid, const struct nls_table *nls_codepage) @@ -2561,14 +2835,14 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, } else { /* decode response */ __u32 data_offset = le32_to_cpu(pSMBr->DataOffset); __u32 data_count = le32_to_cpu(pSMBr->DataCount); - if ((pSMBr->ByteCount < 2) || (data_offset > 512)) { - /* BB also check enough total bytes returned */ + if (get_bcc(&pSMBr->hdr) < 2 || data_offset > 512) { + /* BB also check enough total bytes returned */ rc = -EIO; /* bad smb */ goto qreparse_out; } if (data_count && (data_count < 2048)) { char *end_of_smb = 2 /* sizeof byte count */ + - pSMBr->ByteCount + (char *)&pSMBr->ByteCount; + get_bcc(&pSMBr->hdr) + (char *)&pSMBr->ByteCount; struct reparse_data *reparse_buf = (struct reparse_data *) @@ -2618,7 +2892,7 @@ qreparse_out: return rc; } -#endif /* CIFS_EXPERIMENTAL */ +#endif /* CIFS_SYMLINK_EXPERIMENTAL */ /* BB temporarily unused */ #ifdef CONFIG_CIFS_POSIX @@ -2756,7 +3030,7 @@ static __u16 ACL_to_cifs_posix(char *parm_data, const char *pACL, } int -CIFSSMBGetPosixACL(const int xid, struct cifsTconInfo *tcon, +CIFSSMBGetPosixACL(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, char *acl_inf, const int buflen, const int acl_type, const struct nls_table *nls_codepage, int remap) @@ -2814,7 +3088,7 @@ queryAclRetry: pSMB->ParameterCount = pSMB->TotalParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_POSIX_ACL); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2826,8 +3100,8 @@ queryAclRetry: /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 2)) /* BB also check enough total bytes returned */ + if (rc || get_bcc(&pSMBr->hdr) < 2) rc = -EIO; /* bad smb */ else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -2844,7 +3118,7 @@ queryAclRetry: } int -CIFSSMBSetPosixACL(const int xid, struct cifsTconInfo *tcon, +CIFSSMBSetPosixACL(const int xid, struct cifs_tcon *tcon, const unsigned char *fileName, const char *local_acl, const int buflen, const int acl_type, @@ -2908,7 +3182,7 @@ setAclRetry: pSMB->ParameterCount = cpu_to_le16(params); pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -2924,7 +3198,7 @@ setACLerrorExit: /* BB fix tabs in this function FIXME BB */ int -CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon, +CIFSGetExtAttr(const int xid, struct cifs_tcon *tcon, const int netfid, __u64 *pExtAttrBits, __u64 *pMask) { int rc = 0; @@ -2966,7 +3240,7 @@ GetExtAttrRetry: pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_ATTR_FLAGS); pSMB->Pad = 0; pSMB->Fid = netfid; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->t2.ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -2976,8 +3250,8 @@ GetExtAttrRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 2)) /* BB also check enough total bytes returned */ + if (rc || get_bcc(&pSMBr->hdr) < 2) /* If rc should we check for EOPNOSUPP and disable the srvino flag? or in caller? */ rc = -EIO; /* bad smb */ @@ -3017,7 +3291,7 @@ GetExtAttrOut: */ static int smb_init_nttransact(const __u16 sub_command, const int setup_count, - const int parm_len, struct cifsTconInfo *tcon, + const int parm_len, struct cifs_tcon *tcon, void **ret_buf) { int rc; @@ -3052,6 +3326,7 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata, char *end_of_smb; __u32 data_count, data_offset, parm_count, parm_offset; struct smb_com_ntransact_rsp *pSMBr; + u16 bcc; *pdatalen = 0; *pparmlen = 0; @@ -3061,8 +3336,8 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata, pSMBr = (struct smb_com_ntransact_rsp *)buf; - /* ByteCount was converted from little endian in SendReceive */ - end_of_smb = 2 /* sizeof byte count */ + pSMBr->ByteCount + + bcc = get_bcc(&pSMBr->hdr); + end_of_smb = 2 /* sizeof byte count */ + bcc + (char *)&pSMBr->ByteCount; data_offset = le32_to_cpu(pSMBr->DataOffset); @@ -3088,7 +3363,7 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata, *ppdata, data_count, (data_count + *ppdata), end_of_smb, pSMBr); return -EINVAL; - } else if (parm_count + data_count > pSMBr->ByteCount) { + } else if (parm_count + data_count > bcc) { cFYI(1, "parm count and data count larger than SMB"); return -EINVAL; } @@ -3099,7 +3374,7 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata, /* Get Security Descriptor (by handle) from remote server for a file or dir */ int -CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, +CIFSSMBGetCIFSACL(const int xid, struct cifs_tcon *tcon, __u16 fid, struct cifs_ntsd **acl_inf, __u32 *pbuflen) { int rc = 0; @@ -3124,9 +3399,9 @@ CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, pSMB->AclFlags = cpu_to_le32(CIFS_ACL_OWNER | CIFS_ACL_GROUP | CIFS_ACL_DACL); pSMB->ByteCount = cpu_to_le16(11); /* 3 bytes pad + 8 bytes parm */ - pSMB->hdr.smb_buf_length += 11; + inc_rfc1001_len(pSMB, 11); iov[0].iov_base = (char *)pSMB; - iov[0].iov_len = pSMB->hdr.smb_buf_length + 4; + iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4; rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovec */, &buf_type, 0); @@ -3191,7 +3466,7 @@ qsec_out: } int -CIFSSMBSetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, +CIFSSMBSetCIFSACL(const int xid, struct cifs_tcon *tcon, __u16 fid, struct cifs_ntsd *pntsd, __u32 acllen) { __u16 byte_count, param_count, data_count, param_offset, data_offset; @@ -3235,10 +3510,9 @@ setCifsAclRetry: memcpy((char *) &pSMBr->hdr.Protocol + data_offset, (char *) pntsd, acllen); - pSMB->hdr.smb_buf_length += (byte_count + data_count); - + inc_rfc1001_len(pSMB, byte_count + data_count); } else - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -3258,7 +3532,7 @@ setCifsAclRetry: /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ -int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, +int SMBQueryInformation(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, FILE_ALL_INFO *pFinfo, const struct nls_table *nls_codepage, int remap) @@ -3289,7 +3563,7 @@ QInfRetry: } pSMB->BufferFormat = 0x04; name_len++; /* account for buffer type byte */ - pSMB->hdr.smb_buf_length += (__u16) name_len; + inc_rfc1001_len(pSMB, (__u16)name_len); pSMB->ByteCount = cpu_to_le16(name_len); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -3326,7 +3600,7 @@ QInfRetry: } int -CIFSSMBQFileInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBQFileInfo(const int xid, struct cifs_tcon *tcon, u16 netfid, FILE_ALL_INFO *pFindData) { struct smb_t2_qfi_req *pSMB = NULL; @@ -3364,7 +3638,7 @@ QFileInfoRetry: pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); pSMB->Pad = 0; pSMB->Fid = netfid; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -3375,7 +3649,7 @@ QFileInfoRetry: if (rc) /* BB add auto retry on EOPNOTSUPP? */ rc = -EIO; - else if (pSMBr->ByteCount < 40) + else if (get_bcc(&pSMBr->hdr) < 40) rc = -EIO; /* bad smb */ else if (pFindData) { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -3393,7 +3667,7 @@ QFileInfoRetry: } int -CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBQPathInfo(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, FILE_ALL_INFO *pFindData, int legacy /* old style infolevel */, @@ -3451,7 +3725,7 @@ QPathInfoRetry: else pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -3463,9 +3737,9 @@ QPathInfoRetry: if (rc) /* BB add auto retry on EOPNOTSUPP? */ rc = -EIO; - else if (!legacy && (pSMBr->ByteCount < 40)) + else if (!legacy && get_bcc(&pSMBr->hdr) < 40) rc = -EIO; /* bad smb */ - else if (legacy && (pSMBr->ByteCount < 24)) + else if (legacy && get_bcc(&pSMBr->hdr) < 24) rc = -EIO; /* 24 or 26 expected but we do not read last field */ else if (pFindData) { @@ -3494,7 +3768,7 @@ QPathInfoRetry: } int -CIFSSMBUnixQFileInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBUnixQFileInfo(const int xid, struct cifs_tcon *tcon, u16 netfid, FILE_UNIX_BASIC_INFO *pFindData) { struct smb_t2_qfi_req *pSMB = NULL; @@ -3532,7 +3806,7 @@ UnixQFileInfoRetry: pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_BASIC); pSMB->Pad = 0; pSMB->Fid = netfid; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -3541,7 +3815,7 @@ UnixQFileInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < sizeof(FILE_UNIX_BASIC_INFO))) { + if (rc || get_bcc(&pSMBr->hdr) < sizeof(FILE_UNIX_BASIC_INFO)) { cERROR(1, "Malformed FILE_UNIX_BASIC_INFO response.\n" "Unix Extensions can be disabled on mount " "by specifying the nosfu mount option."); @@ -3563,7 +3837,7 @@ UnixQFileInfoRetry: } int -CIFSSMBUnixQPathInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBUnixQPathInfo(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, FILE_UNIX_BASIC_INFO *pFindData, const struct nls_table *nls_codepage, int remap) @@ -3617,7 +3891,7 @@ UnixQPathInfoRetry: pSMB->ParameterCount = pSMB->TotalParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_BASIC); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -3627,7 +3901,7 @@ UnixQPathInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < sizeof(FILE_UNIX_BASIC_INFO))) { + if (rc || get_bcc(&pSMBr->hdr) < sizeof(FILE_UNIX_BASIC_INFO)) { cERROR(1, "Malformed FILE_UNIX_BASIC_INFO response.\n" "Unix Extensions can be disabled on mount " "by specifying the nosfu mount option."); @@ -3649,7 +3923,7 @@ UnixQPathInfoRetry: /* xid, tcon, searchName and codepage are input parms, rest are returned */ int -CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, +CIFSFindFirst(const int xid, struct cifs_tcon *tcon, const char *searchName, const struct nls_table *nls_codepage, __u16 *pnetfid, @@ -3731,7 +4005,7 @@ findFirstRetry: /* BB what should we set StorageType to? Does it matter? BB */ pSMB->SearchStorageType = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -3797,7 +4071,7 @@ findFirstRetry: return rc; } -int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, +int CIFSFindNext(const int xid, struct cifs_tcon *tcon, __u16 searchHandle, struct cifs_search_info *psrch_inf) { TRANSACTION2_FNEXT_REQ *pSMB = NULL; @@ -3860,7 +4134,7 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, byte_count = params + 1 /* pad */ ; pSMB->TotalParameterCount = cpu_to_le16(params); pSMB->ParameterCount = pSMB->TotalParameterCount; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -3935,7 +4209,7 @@ FNext2_err_exit: } int -CIFSFindClose(const int xid, struct cifsTconInfo *tcon, +CIFSFindClose(const int xid, struct cifs_tcon *tcon, const __u16 searchHandle) { int rc = 0; @@ -3967,7 +4241,7 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, } int -CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, +CIFSGetSrvInodeNumber(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, __u64 *inode_number, const struct nls_table *nls_codepage, int remap) @@ -4022,7 +4296,7 @@ GetInodeNumberRetry: pSMB->ParameterCount = pSMB->TotalParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_INTERNAL_INFO); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4032,8 +4306,8 @@ GetInodeNumberRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 2)) /* BB also check enough total bytes returned */ + if (rc || get_bcc(&pSMBr->hdr) < 2) /* If rc should we check for EOPNOSUPP and disable the srvino flag? or in caller? */ rc = -EIO; /* bad smb */ @@ -4169,7 +4443,7 @@ parse_DFS_referrals_exit: } int -CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, +CIFSGetDFSRefer(const int xid, struct cifs_ses *ses, const unsigned char *searchName, struct dfs_info3_param **target_nodes, unsigned int *num_of_nodes, @@ -4218,7 +4492,7 @@ getDFSRetry: } if (ses->server) { - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) pSMB->hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; } @@ -4246,7 +4520,7 @@ getDFSRetry: pSMB->ParameterCount = cpu_to_le16(params); pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->MaxReferralLevel = cpu_to_le16(3); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, ses, (struct smb_hdr *) pSMB, @@ -4258,13 +4532,13 @@ getDFSRetry: rc = validate_t2((struct smb_t2_rsp *)pSMBr); /* BB Also check if enough total bytes returned? */ - if (rc || (pSMBr->ByteCount < 17)) { + if (rc || get_bcc(&pSMBr->hdr) < 17) { rc = -EIO; /* bad smb */ goto GetDFSRefExit; } cFYI(1, "Decoding GetDFSRefer response BCC: %d Offset %d", - pSMBr->ByteCount, + get_bcc(&pSMBr->hdr), le16_to_cpu(pSMBr->t2.DataOffset)); /* parse returned result into more usable form */ @@ -4283,7 +4557,7 @@ GetDFSRefExit: /* Query File System Info such as free space to old servers such as Win 9x */ int -SMBOldQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData) +SMBOldQFSInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData) { /* level 0x01 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -4320,7 +4594,7 @@ oldQFSInfoRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_INFO_ALLOCATION); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4330,12 +4604,12 @@ oldQFSInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 18)) + if (rc || get_bcc(&pSMBr->hdr) < 18) rc = -EIO; /* bad smb */ else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); cFYI(1, "qfsinf resp BCC: %d Offset %d", - pSMBr->ByteCount, data_offset); + get_bcc(&pSMBr->hdr), data_offset); response_data = (FILE_SYSTEM_ALLOC_INFO *) (((char *) &pSMBr->hdr.Protocol) + data_offset); @@ -4362,7 +4636,7 @@ oldQFSInfoRetry: } int -CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData) +CIFSSMBQFSInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData) { /* level 0x103 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -4399,7 +4673,7 @@ QFSInfoRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_SIZE_INFO); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4409,7 +4683,7 @@ QFSInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 24)) + if (rc || get_bcc(&pSMBr->hdr) < 24) rc = -EIO; /* bad smb */ else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -4441,7 +4715,7 @@ QFSInfoRetry: } int -CIFSSMBQFSAttributeInfo(const int xid, struct cifsTconInfo *tcon) +CIFSSMBQFSAttributeInfo(const int xid, struct cifs_tcon *tcon) { /* level 0x105 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -4479,7 +4753,7 @@ QFSAttributeRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_ATTRIBUTE_INFO); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4489,7 +4763,7 @@ QFSAttributeRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 13)) { + if (rc || get_bcc(&pSMBr->hdr) < 13) { /* BB also check if enough bytes returned */ rc = -EIO; /* bad smb */ } else { @@ -4511,7 +4785,7 @@ QFSAttributeRetry: } int -CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon) +CIFSSMBQFSDeviceInfo(const int xid, struct cifs_tcon *tcon) { /* level 0x104 SMB_QUERY_FILE_SYSTEM_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -4550,7 +4824,7 @@ QFSDeviceRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_DEVICE_INFO); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4560,7 +4834,8 @@ QFSDeviceRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < sizeof(FILE_SYSTEM_DEVICE_INFO))) + if (rc || get_bcc(&pSMBr->hdr) < + sizeof(FILE_SYSTEM_DEVICE_INFO)) rc = -EIO; /* bad smb */ else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -4581,7 +4856,7 @@ QFSDeviceRetry: } int -CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon) +CIFSSMBQFSUnixInfo(const int xid, struct cifs_tcon *tcon) { /* level 0x200 SMB_QUERY_CIFS_UNIX_INFO */ TRANSACTION2_QFSI_REQ *pSMB = NULL; @@ -4619,7 +4894,7 @@ QFSUnixRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_CIFS_UNIX_INFO); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4629,7 +4904,7 @@ QFSUnixRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 13)) { + if (rc || get_bcc(&pSMBr->hdr) < 13) { rc = -EIO; /* bad smb */ } else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -4651,7 +4926,7 @@ QFSUnixRetry: } int -CIFSSMBSetFSUnixInfo(const int xid, struct cifsTconInfo *tcon, __u64 cap) +CIFSSMBSetFSUnixInfo(const int xid, struct cifs_tcon *tcon, __u64 cap) { /* level 0x200 SMB_SET_CIFS_UNIX_INFO */ TRANSACTION2_SETFSI_REQ *pSMB = NULL; @@ -4702,7 +4977,7 @@ SETFSUnixRetry: pSMB->ClientUnixMinor = cpu_to_le16(CIFS_UNIX_MINOR_VERSION); pSMB->ClientUnixCap = cpu_to_le64(cap); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4725,7 +5000,7 @@ SETFSUnixRetry: int -CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBQFSPosixInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData) { /* level 0x201 SMB_QUERY_CIFS_POSIX_INFO */ @@ -4764,7 +5039,7 @@ QFSPosixRetry: pSMB->Reserved3 = 0; pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_POSIX_FS_INFO); - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4774,7 +5049,7 @@ QFSPosixRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 13)) { + if (rc || get_bcc(&pSMBr->hdr) < 13) { rc = -EIO; /* bad smb */ } else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); @@ -4818,7 +5093,7 @@ QFSPosixRetry: in Samba which this routine can run into */ int -CIFSSMBSetEOF(const int xid, struct cifsTconInfo *tcon, const char *fileName, +CIFSSMBSetEOF(const int xid, struct cifs_tcon *tcon, const char *fileName, __u64 size, bool SetAllocation, const struct nls_table *nls_codepage, int remap) { @@ -4890,7 +5165,7 @@ SetEOFRetry: pSMB->ParameterCount = cpu_to_le16(params); pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); parm_data->FileSize = cpu_to_le64(size); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -4907,7 +5182,7 @@ SetEOFRetry: } int -CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, +CIFSSMBSetFileSize(const int xid, struct cifs_tcon *tcon, __u64 size, __u16 fid, __u32 pid_of_opener, bool SetAllocation) { struct smb_com_transaction2_sfi_req *pSMB = NULL; @@ -4969,7 +5244,7 @@ CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, cpu_to_le16(SMB_SET_FILE_END_OF_FILE_INFO); } pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); if (rc) { @@ -4989,7 +5264,7 @@ CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, time and resort to the original setpathinfo level which takes the ancient DOS time format with 2 second granularity */ int -CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBSetFileInfo(const int xid, struct cifs_tcon *tcon, const FILE_BASIC_INFO *data, __u16 fid, __u32 pid_of_opener) { struct smb_com_transaction2_sfi_req *pSMB = NULL; @@ -5037,7 +5312,7 @@ CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, else pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_BASIC_INFO); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); memcpy(data_offset, data, sizeof(FILE_BASIC_INFO)); rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); @@ -5051,7 +5326,7 @@ CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, } int -CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, +CIFSSMBSetFileDisposition(const int xid, struct cifs_tcon *tcon, bool delete_file, __u16 fid, __u32 pid_of_opener) { struct smb_com_transaction2_sfi_req *pSMB = NULL; @@ -5096,7 +5371,7 @@ CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, pSMB->Fid = fid; pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_DISPOSITION_INFO); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); *data_offset = delete_file ? 1 : 0; rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0); @@ -5107,7 +5382,7 @@ CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, } int -CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBSetPathInfo(const int xid, struct cifs_tcon *tcon, const char *fileName, const FILE_BASIC_INFO *data, const struct nls_table *nls_codepage, int remap) { @@ -5169,7 +5444,7 @@ SetTimesRetry: else pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_BASIC_INFO); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); memcpy(data_offset, data, sizeof(FILE_BASIC_INFO)); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -5191,7 +5466,7 @@ SetTimesRetry: handling it anyway and NT4 was what we thought it would be needed for Do not delete it until we prove whether needed for Win9x though */ int -CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, char *fileName, +CIFSSMBSetAttrLegacy(int xid, struct cifs_tcon *tcon, char *fileName, __u16 dos_attrs, const struct nls_table *nls_codepage) { SETATTR_REQ *pSMB = NULL; @@ -5221,7 +5496,7 @@ SetAttrLgcyRetry: } pSMB->attr = cpu_to_le16(dos_attrs); pSMB->BufferFormat = 0x04; - pSMB->hdr.smb_buf_length += name_len + 1; + inc_rfc1001_len(pSMB, name_len + 1); pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -5279,7 +5554,7 @@ cifs_fill_unix_set_info(FILE_UNIX_BASIC_INFO *data_offset, } int -CIFSSMBUnixSetFileInfo(const int xid, struct cifsTconInfo *tcon, +CIFSSMBUnixSetFileInfo(const int xid, struct cifs_tcon *tcon, const struct cifs_unix_set_info_args *args, u16 fid, u32 pid_of_opener) { @@ -5326,7 +5601,7 @@ CIFSSMBUnixSetFileInfo(const int xid, struct cifsTconInfo *tcon, pSMB->Fid = fid; pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_BASIC); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); cifs_fill_unix_set_info(data_offset, args); @@ -5342,7 +5617,7 @@ CIFSSMBUnixSetFileInfo(const int xid, struct cifsTconInfo *tcon, } int -CIFSSMBUnixSetPathInfo(const int xid, struct cifsTconInfo *tcon, char *fileName, +CIFSSMBUnixSetPathInfo(const int xid, struct cifs_tcon *tcon, char *fileName, const struct cifs_unix_set_info_args *args, const struct nls_table *nls_codepage, int remap) { @@ -5402,7 +5677,7 @@ setPermsRetry: pSMB->TotalDataCount = pSMB->DataCount; pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_BASIC); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); cifs_fill_unix_set_info(data_offset, args); @@ -5418,79 +5693,6 @@ setPermsRetry: return rc; } -int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, - const int notify_subdirs, const __u16 netfid, - __u32 filter, struct file *pfile, int multishot, - const struct nls_table *nls_codepage) -{ - int rc = 0; - struct smb_com_transaction_change_notify_req *pSMB = NULL; - struct smb_com_ntransaction_change_notify_rsp *pSMBr = NULL; - struct dir_notify_req *dnotify_req; - int bytes_returned; - - cFYI(1, "In CIFSSMBNotify for file handle %d", (int)netfid); - rc = smb_init(SMB_COM_NT_TRANSACT, 23, tcon, (void **) &pSMB, - (void **) &pSMBr); - if (rc) - return rc; - - pSMB->TotalParameterCount = 0 ; - pSMB->TotalDataCount = 0; - pSMB->MaxParameterCount = cpu_to_le32(2); - /* BB find exact data count max from sess structure BB */ - pSMB->MaxDataCount = 0; /* same in little endian or be */ -/* BB VERIFY verify which is correct for above BB */ - pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - - MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); - - pSMB->MaxSetupCount = 4; - pSMB->Reserved = 0; - pSMB->ParameterOffset = 0; - pSMB->DataCount = 0; - pSMB->DataOffset = 0; - pSMB->SetupCount = 4; /* single byte does not need le conversion */ - pSMB->SubCommand = cpu_to_le16(NT_TRANSACT_NOTIFY_CHANGE); - pSMB->ParameterCount = pSMB->TotalParameterCount; - if (notify_subdirs) - pSMB->WatchTree = 1; /* one byte - no le conversion needed */ - pSMB->Reserved2 = 0; - pSMB->CompletionFilter = cpu_to_le32(filter); - pSMB->Fid = netfid; /* file handle always le */ - pSMB->ByteCount = 0; - - rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *)pSMBr, &bytes_returned, - CIFS_ASYNC_OP); - if (rc) { - cFYI(1, "Error in Notify = %d", rc); - } else { - /* Add file to outstanding requests */ - /* BB change to kmem cache alloc */ - dnotify_req = kmalloc( - sizeof(struct dir_notify_req), - GFP_KERNEL); - if (dnotify_req) { - dnotify_req->Pid = pSMB->hdr.Pid; - dnotify_req->PidHigh = pSMB->hdr.PidHigh; - dnotify_req->Mid = pSMB->hdr.Mid; - dnotify_req->Tid = pSMB->hdr.Tid; - dnotify_req->Uid = pSMB->hdr.Uid; - dnotify_req->netfid = netfid; - dnotify_req->pfile = pfile; - dnotify_req->filter = filter; - dnotify_req->multishot = multishot; - spin_lock(&GlobalMid_Lock); - list_add_tail(&dnotify_req->lhead, - &GlobalDnotifyReqList); - spin_unlock(&GlobalMid_Lock); - } else - rc = -ENOMEM; - } - cifs_buf_release(pSMB); - return rc; -} - #ifdef CONFIG_CIFS_XATTR /* * Do a path-based QUERY_ALL_EAS call and parse the result. This is a common @@ -5502,7 +5704,7 @@ int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, * the data isn't copied to it, but the length is returned. */ ssize_t -CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon, +CIFSSMBQAllEAs(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, const unsigned char *ea_name, char *EAData, size_t buf_size, const struct nls_table *nls_codepage, int remap) @@ -5560,7 +5762,7 @@ QAllEAsRetry: pSMB->ParameterCount = pSMB->TotalParameterCount; pSMB->InformationLevel = cpu_to_le16(SMB_INFO_QUERY_ALL_EAS); pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, @@ -5576,7 +5778,7 @@ QAllEAsRetry: of these trans2 responses */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 4)) { + if (rc || get_bcc(&pSMBr->hdr) < 4) { rc = -EIO; /* bad smb */ goto QAllEAsOut; } @@ -5683,7 +5885,7 @@ QAllEAsOut: } int -CIFSSMBSetEA(const int xid, struct cifsTconInfo *tcon, const char *fileName, +CIFSSMBSetEA(const int xid, struct cifs_tcon *tcon, const char *fileName, const char *ea_name, const void *ea_value, const __u16 ea_value_len, const struct nls_table *nls_codepage, int remap) @@ -5773,7 +5975,7 @@ SetEARetry: pSMB->ParameterCount = cpu_to_le16(params); pSMB->TotalParameterCount = pSMB->ParameterCount; pSMB->Reserved4 = 0; - pSMB->hdr.smb_buf_length += byte_count; + inc_rfc1001_len(pSMB, byte_count); pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -5787,5 +5989,99 @@ SetEARetry: return rc; } - #endif + +#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* BB unused temporarily */ +/* + * Years ago the kernel added a "dnotify" function for Samba server, + * to allow network clients (such as Windows) to display updated + * lists of files in directory listings automatically when + * files are added by one user when another user has the + * same directory open on their desktop. The Linux cifs kernel + * client hooked into the kernel side of this interface for + * the same reason, but ironically when the VFS moved from + * "dnotify" to "inotify" it became harder to plug in Linux + * network file system clients (the most obvious use case + * for notify interfaces is when multiple users can update + * the contents of the same directory - exactly what network + * file systems can do) although the server (Samba) could + * still use it. For the short term we leave the worker + * function ifdeffed out (below) until inotify is fixed + * in the VFS to make it easier to plug in network file + * system clients. If inotify turns out to be permanently + * incompatible for network fs clients, we could instead simply + * expose this config flag by adding a future cifs (and smb2) notify ioctl. + */ +int CIFSSMBNotify(const int xid, struct cifs_tcon *tcon, + const int notify_subdirs, const __u16 netfid, + __u32 filter, struct file *pfile, int multishot, + const struct nls_table *nls_codepage) +{ + int rc = 0; + struct smb_com_transaction_change_notify_req *pSMB = NULL; + struct smb_com_ntransaction_change_notify_rsp *pSMBr = NULL; + struct dir_notify_req *dnotify_req; + int bytes_returned; + + cFYI(1, "In CIFSSMBNotify for file handle %d", (int)netfid); + rc = smb_init(SMB_COM_NT_TRANSACT, 23, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + pSMB->TotalParameterCount = 0 ; + pSMB->TotalDataCount = 0; + pSMB->MaxParameterCount = cpu_to_le32(2); + /* BB find exact data count max from sess structure BB */ + pSMB->MaxDataCount = 0; /* same in little endian or be */ +/* BB VERIFY verify which is correct for above BB */ + pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); + + pSMB->MaxSetupCount = 4; + pSMB->Reserved = 0; + pSMB->ParameterOffset = 0; + pSMB->DataCount = 0; + pSMB->DataOffset = 0; + pSMB->SetupCount = 4; /* single byte does not need le conversion */ + pSMB->SubCommand = cpu_to_le16(NT_TRANSACT_NOTIFY_CHANGE); + pSMB->ParameterCount = pSMB->TotalParameterCount; + if (notify_subdirs) + pSMB->WatchTree = 1; /* one byte - no le conversion needed */ + pSMB->Reserved2 = 0; + pSMB->CompletionFilter = cpu_to_le32(filter); + pSMB->Fid = netfid; /* file handle always le */ + pSMB->ByteCount = 0; + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *)pSMBr, &bytes_returned, + CIFS_ASYNC_OP); + if (rc) { + cFYI(1, "Error in Notify = %d", rc); + } else { + /* Add file to outstanding requests */ + /* BB change to kmem cache alloc */ + dnotify_req = kmalloc( + sizeof(struct dir_notify_req), + GFP_KERNEL); + if (dnotify_req) { + dnotify_req->Pid = pSMB->hdr.Pid; + dnotify_req->PidHigh = pSMB->hdr.PidHigh; + dnotify_req->Mid = pSMB->hdr.Mid; + dnotify_req->Tid = pSMB->hdr.Tid; + dnotify_req->Uid = pSMB->hdr.Uid; + dnotify_req->netfid = netfid; + dnotify_req->pfile = pfile; + dnotify_req->filter = filter; + dnotify_req->multishot = multishot; + spin_lock(&GlobalMid_Lock); + list_add_tail(&dnotify_req->lhead, + &GlobalDnotifyReqList); + spin_unlock(&GlobalMid_Lock); + } else + rc = -ENOMEM; + } + cifs_buf_release(pSMB); + return rc; +} +#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 277262a..7f540df 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -57,61 +57,6 @@ extern mempool_t *cifs_req_poolp; -struct smb_vol { - char *username; - char *password; - char *domainname; - char *UNC; - char *UNCip; - char *iocharset; /* local code page for mapping to and from Unicode */ - char source_rfc1001_name[RFC1001_NAME_LEN_WITH_NULL]; /* clnt nb name */ - char target_rfc1001_name[RFC1001_NAME_LEN_WITH_NULL]; /* srvr nb name */ - uid_t cred_uid; - uid_t linux_uid; - gid_t linux_gid; - mode_t file_mode; - mode_t dir_mode; - unsigned secFlg; - bool retry:1; - bool intr:1; - bool setuids:1; - bool override_uid:1; - bool override_gid:1; - bool dynperm:1; - bool noperm:1; - bool no_psx_acl:1; /* set if posix acl support should be disabled */ - bool cifs_acl:1; - bool no_xattr:1; /* set if xattr (EA) support should be disabled*/ - bool server_ino:1; /* use inode numbers from server ie UniqueId */ - bool direct_io:1; - bool strict_io:1; /* strict cache behavior */ - bool remap:1; /* set to remap seven reserved chars in filenames */ - bool posix_paths:1; /* unset to not ask for posix pathnames. */ - bool no_linux_ext:1; - bool sfu_emul:1; - bool nullauth:1; /* attempt to authenticate with null user */ - bool nocase:1; /* request case insensitive filenames */ - bool nobrl:1; /* disable sending byte range locks to srv */ - bool mand_lock:1; /* send mandatory not posix byte range lock reqs */ - bool seal:1; /* request transport encryption on share */ - bool nodfs:1; /* Do not request DFS, even if available */ - bool local_lease:1; /* check leases only on local system, not remote */ - bool noblocksnd:1; - bool noautotune:1; - bool nostrictsync:1; /* do not force expensive SMBflush on every sync */ - bool fsc:1; /* enable fscache */ - bool mfsymlinks:1; /* use Minshall+French Symlinks */ - bool multiuser:1; - unsigned int rsize; - unsigned int wsize; - bool sockopt_tcp_nodelay:1; - unsigned short int port; - unsigned long actimeo; /* attribute cache timeout (jiffies) */ - char *prepath; - struct sockaddr_storage srcaddr; /* allow binding to a local IP */ - struct nls_table *local_nls; -}; - /* FIXME: should these be tunable? */ #define TLINK_ERROR_EXPIRE (1 * HZ) #define TLINK_IDLE_EXPIRE (600 * HZ) @@ -134,9 +79,10 @@ cifs_reconnect(struct TCP_Server_Info *server) { int rc = 0; struct list_head *tmp, *tmp2; - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon; + struct cifs_ses *ses; + struct cifs_tcon *tcon; struct mid_q_entry *mid_entry; + struct list_head retry_list; spin_lock(&GlobalMid_Lock); if (server->tcpStatus == CifsExiting) { @@ -156,11 +102,11 @@ cifs_reconnect(struct TCP_Server_Info *server) cFYI(1, "%s: marking sessions and tcons for reconnect", __func__); spin_lock(&cifs_tcp_ses_lock); list_for_each(tmp, &server->smb_ses_list) { - ses = list_entry(tmp, struct cifsSesInfo, smb_ses_list); + ses = list_entry(tmp, struct cifs_ses, smb_ses_list); ses->need_reconnect = true; ses->ipc_tid = 0; list_for_each(tmp2, &ses->tcon_list) { - tcon = list_entry(tmp2, struct cifsTconInfo, tcon_list); + tcon = list_entry(tmp2, struct cifs_tcon, tcon_list); tcon->need_reconnect = true; } } @@ -188,18 +134,25 @@ cifs_reconnect(struct TCP_Server_Info *server) mutex_unlock(&server->srv_mutex); /* mark submitted MIDs for retry and issue callback */ - cFYI(1, "%s: issuing mid callbacks", __func__); + INIT_LIST_HEAD(&retry_list); + cFYI(1, "%s: moving mids to private list", __func__); spin_lock(&GlobalMid_Lock); list_for_each_safe(tmp, tmp2, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); if (mid_entry->midState == MID_REQUEST_SUBMITTED) mid_entry->midState = MID_RETRY_NEEDED; + list_move(&mid_entry->qhead, &retry_list); + } + spin_unlock(&GlobalMid_Lock); + + cFYI(1, "%s: issuing mid callbacks", __func__); + list_for_each_safe(tmp, tmp2, &retry_list) { + mid_entry = list_entry(tmp, struct mid_q_entry, qhead); list_del_init(&mid_entry->qhead); mid_entry->callback(mid_entry); } - spin_unlock(&GlobalMid_Lock); - while (server->tcpStatus == CifsNeedReconnect) { + do { try_to_freeze(); /* we should try only the port we connected to before */ @@ -214,7 +167,7 @@ cifs_reconnect(struct TCP_Server_Info *server) server->tcpStatus = CifsNeedNegotiate; spin_unlock(&GlobalMid_Lock); } - } + } while (server->tcpStatus == CifsNeedReconnect); return rc; } @@ -316,19 +269,19 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) put_unaligned_le16(total_in_buf, &pSMBt->t2_rsp.DataCount); /* fix up the BCC */ - byte_count = get_bcc_le(pTargetSMB); + byte_count = get_bcc(pTargetSMB); byte_count += total_in_buf2; /* is the result too big for the field? */ if (byte_count > USHRT_MAX) return -EPROTO; - put_bcc_le(byte_count, pTargetSMB); + put_bcc(byte_count, pTargetSMB); - byte_count = pTargetSMB->smb_buf_length; + byte_count = be32_to_cpu(pTargetSMB->smb_buf_length); byte_count += total_in_buf2; /* don't allow buffer to overflow */ if (byte_count > CIFSMaxBufSize) return -ENOBUFS; - pTargetSMB->smb_buf_length = byte_count; + pTargetSMB->smb_buf_length = cpu_to_be32(byte_count); memcpy(data_area_of_target, data_area_of_buf2, total_in_buf2); @@ -495,8 +448,7 @@ incomplete_rcv: /* Note that FC 1001 length is big endian on the wire, but we convert it here so it is always manipulated as host byte order */ - pdu_length = be32_to_cpu((__force __be32)smb_buffer->smb_buf_length); - smb_buffer->smb_buf_length = pdu_length; + pdu_length = be32_to_cpu(smb_buffer->smb_buf_length); cFYI(1, "rfc1002 length 0x%x", pdu_length+4); @@ -672,12 +624,12 @@ multi_t2_fnd: mid_entry->when_received = jiffies; #endif list_del_init(&mid_entry->qhead); - mid_entry->callback(mid_entry); break; } spin_unlock(&GlobalMid_Lock); if (mid_entry != NULL) { + mid_entry->callback(mid_entry); /* Was previous buf put in mpx struct for multi-rsp? */ if (!isMultiRsp) { /* smb buffer will be freed by user thread */ @@ -735,21 +687,31 @@ multi_t2_fnd: sock_release(csocket); server->ssocket = NULL; } - /* buffer usuallly freed in free_mid - need to free it here on exit */ + /* buffer usually freed in free_mid - need to free it here on exit */ cifs_buf_release(bigbuf); if (smallbuf) /* no sense logging a debug message if NULL */ cifs_small_buf_release(smallbuf); if (!list_empty(&server->pending_mid_q)) { + struct list_head dispose_list; + + INIT_LIST_HEAD(&dispose_list); spin_lock(&GlobalMid_Lock); list_for_each_safe(tmp, tmp2, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - cFYI(1, "Clearing Mid 0x%x - issuing callback", - mid_entry->mid); + cFYI(1, "Clearing mid 0x%x", mid_entry->mid); + mid_entry->midState = MID_SHUTDOWN; + list_move(&mid_entry->qhead, &dispose_list); + } + spin_unlock(&GlobalMid_Lock); + + /* now walk dispose list and issue callbacks */ + list_for_each_safe(tmp, tmp2, &dispose_list) { + mid_entry = list_entry(tmp, struct mid_q_entry, qhead); + cFYI(1, "Callback mid 0x%x", mid_entry->mid); list_del_init(&mid_entry->qhead); mid_entry->callback(mid_entry); } - spin_unlock(&GlobalMid_Lock); /* 1/8th of sec is more than enough time for them to exit */ msleep(125); } @@ -818,10 +780,11 @@ extract_hostname(const char *unc) } static int -cifs_parse_mount_options(char *options, const char *devname, +cifs_parse_mount_options(const char *mountdata, const char *devname, struct smb_vol *vol) { char *value, *data, *end; + char *mountdata_copy = NULL, *options; unsigned int temp_len, i, j; char separator[2]; short int override_uid = -1; @@ -861,9 +824,14 @@ cifs_parse_mount_options(char *options, const char *devname, vol->actimeo = CIFS_DEF_ACTIMEO; - if (!options) - return 1; + if (!mountdata) + goto cifs_parse_mount_err; + mountdata_copy = kstrndup(mountdata, PAGE_SIZE, GFP_KERNEL); + if (!mountdata_copy) + goto cifs_parse_mount_err; + + options = mountdata_copy; end = options + strlen(options); if (strncmp(options, "sep=", 4) == 0) { if (options[4] != 0) { @@ -889,17 +857,22 @@ cifs_parse_mount_options(char *options, const char *devname, if (!value) { printk(KERN_WARNING "CIFS: invalid or missing username\n"); - return 1; /* needs_arg; */ + goto cifs_parse_mount_err; } else if (!*value) { /* null user, ie anonymous, authentication */ vol->nullauth = 1; } if (strnlen(value, MAX_USERNAME_SIZE) < MAX_USERNAME_SIZE) { - vol->username = value; + vol->username = kstrdup(value, GFP_KERNEL); + if (!vol->username) { + printk(KERN_WARNING "CIFS: no memory " + "for username\n"); + goto cifs_parse_mount_err; + } } else { printk(KERN_WARNING "CIFS: username too long\n"); - return 1; + goto cifs_parse_mount_err; } } else if (strnicmp(data, "pass", 4) == 0) { if (!value) { @@ -963,7 +936,7 @@ cifs_parse_mount_options(char *options, const char *devname, if (vol->password == NULL) { printk(KERN_WARNING "CIFS: no memory " "for password\n"); - return 1; + goto cifs_parse_mount_err; } for (i = 0, j = 0; i < temp_len; i++, j++) { vol->password[j] = value[i]; @@ -979,7 +952,7 @@ cifs_parse_mount_options(char *options, const char *devname, if (vol->password == NULL) { printk(KERN_WARNING "CIFS: no memory " "for password\n"); - return 1; + goto cifs_parse_mount_err; } strcpy(vol->password, value); } @@ -989,11 +962,16 @@ cifs_parse_mount_options(char *options, const char *devname, vol->UNCip = NULL; } else if (strnlen(value, INET6_ADDRSTRLEN) < INET6_ADDRSTRLEN) { - vol->UNCip = value; + vol->UNCip = kstrdup(value, GFP_KERNEL); + if (!vol->UNCip) { + printk(KERN_WARNING "CIFS: no memory " + "for UNC IP\n"); + goto cifs_parse_mount_err; + } } else { printk(KERN_WARNING "CIFS: ip address " "too long\n"); - return 1; + goto cifs_parse_mount_err; } } else if (strnicmp(data, "sec", 3) == 0) { if (!value || !*value) { @@ -1006,7 +984,7 @@ cifs_parse_mount_options(char *options, const char *devname, /* vol->secFlg |= CIFSSEC_MUST_SEAL | CIFSSEC_MAY_KRB5; */ cERROR(1, "Krb5 cifs privacy not supported"); - return 1; + goto cifs_parse_mount_err; } else if (strnicmp(value, "krb5", 4) == 0) { vol->secFlg |= CIFSSEC_MAY_KRB5; } else if (strnicmp(value, "ntlmsspi", 8) == 0) { @@ -1036,7 +1014,16 @@ cifs_parse_mount_options(char *options, const char *devname, vol->nullauth = 1; } else { cERROR(1, "bad security option: %s", value); - return 1; + goto cifs_parse_mount_err; + } + } else if (strnicmp(data, "vers", 3) == 0) { + if (!value || !*value) { + cERROR(1, "no protocol version specified" + " after vers= mount option"); + } else if ((strnicmp(value, "cifs", 4) == 0) || + (strnicmp(value, "1", 1) == 0)) { + /* this is the default */ + continue; } } else if ((strnicmp(data, "unc", 3) == 0) || (strnicmp(data, "target", 6) == 0) @@ -1044,12 +1031,12 @@ cifs_parse_mount_options(char *options, const char *devname, if (!value || !*value) { printk(KERN_WARNING "CIFS: invalid path to " "network resource\n"); - return 1; /* needs_arg; */ + goto cifs_parse_mount_err; } if ((temp_len = strnlen(value, 300)) < 300) { vol->UNC = kmalloc(temp_len+1, GFP_KERNEL); if (vol->UNC == NULL) - return 1; + goto cifs_parse_mount_err; strcpy(vol->UNC, value); if (strncmp(vol->UNC, "//", 2) == 0) { vol->UNC[0] = '\\'; @@ -1058,27 +1045,32 @@ cifs_parse_mount_options(char *options, const char *devname, printk(KERN_WARNING "CIFS: UNC Path does not begin " "with // or \\\\ \n"); - return 1; + goto cifs_parse_mount_err; } } else { printk(KERN_WARNING "CIFS: UNC name too long\n"); - return 1; + goto cifs_parse_mount_err; } } else if ((strnicmp(data, "domain", 3) == 0) || (strnicmp(data, "workgroup", 5) == 0)) { if (!value || !*value) { printk(KERN_WARNING "CIFS: invalid domain name\n"); - return 1; /* needs_arg; */ + goto cifs_parse_mount_err; } /* BB are there cases in which a comma can be valid in a domain name and need special handling? */ if (strnlen(value, 256) < 256) { - vol->domainname = value; + vol->domainname = kstrdup(value, GFP_KERNEL); + if (!vol->domainname) { + printk(KERN_WARNING "CIFS: no memory " + "for domainname\n"); + goto cifs_parse_mount_err; + } cFYI(1, "Domain name set"); } else { printk(KERN_WARNING "CIFS: domain name too " "long\n"); - return 1; + goto cifs_parse_mount_err; } } else if (strnicmp(data, "srcaddr", 7) == 0) { vol->srcaddr.ss_family = AF_UNSPEC; @@ -1086,7 +1078,7 @@ cifs_parse_mount_options(char *options, const char *devname, if (!value || !*value) { printk(KERN_WARNING "CIFS: srcaddr value" " not specified.\n"); - return 1; /* needs_arg; */ + goto cifs_parse_mount_err; } i = cifs_convert_address((struct sockaddr *)&vol->srcaddr, value, strlen(value)); @@ -1094,20 +1086,20 @@ cifs_parse_mount_options(char *options, const char *devname, printk(KERN_WARNING "CIFS: Could not parse" " srcaddr: %s\n", value); - return 1; + goto cifs_parse_mount_err; } } else if (strnicmp(data, "prefixpath", 10) == 0) { if (!value || !*value) { printk(KERN_WARNING "CIFS: invalid path prefix\n"); - return 1; /* needs_argument */ + goto cifs_parse_mount_err; } if ((temp_len = strnlen(value, 1024)) < 1024) { if (value[0] != '/') temp_len++; /* missing leading slash */ vol->prepath = kmalloc(temp_len+1, GFP_KERNEL); if (vol->prepath == NULL) - return 1; + goto cifs_parse_mount_err; if (value[0] != '/') { vol->prepath[0] = '/'; strcpy(vol->prepath+1, value); @@ -1116,24 +1108,33 @@ cifs_parse_mount_options(char *options, const char *devname, cFYI(1, "prefix path %s", vol->prepath); } else { printk(KERN_WARNING "CIFS: prefix too long\n"); - return 1; + goto cifs_parse_mount_err; } } else if (strnicmp(data, "iocharset", 9) == 0) { if (!value || !*value) { printk(KERN_WARNING "CIFS: invalid iocharset " "specified\n"); - return 1; /* needs_arg; */ + goto cifs_parse_mount_err; } if (strnlen(value, 65) < 65) { - if (strnicmp(value, "default", 7)) - vol->iocharset = value; + if (strnicmp(value, "default", 7)) { + vol->iocharset = kstrdup(value, + GFP_KERNEL); + + if (!vol->iocharset) { + printk(KERN_WARNING "CIFS: no " + "memory for" + "charset\n"); + goto cifs_parse_mount_err; + } + } /* if iocharset not set then load_nls_default is used by caller */ cFYI(1, "iocharset set to %s", value); } else { printk(KERN_WARNING "CIFS: iocharset name " "too long.\n"); - return 1; + goto cifs_parse_mount_err; } } else if (!strnicmp(data, "uid", 3) && value && *value) { vol->linux_uid = simple_strtoul(value, &value, 0); @@ -1246,7 +1247,7 @@ cifs_parse_mount_options(char *options, const char *devname, if (vol->actimeo > CIFS_MAX_ACTIMEO) { cERROR(1, "CIFS: attribute cache" "timeout too large"); - return 1; + goto cifs_parse_mount_err; } } } else if (strnicmp(data, "credentials", 4) == 0) { @@ -1358,6 +1359,8 @@ cifs_parse_mount_options(char *options, const char *devname, vol->server_ino = 1; } else if (strnicmp(data, "noserverino", 9) == 0) { vol->server_ino = 0; + } else if (strnicmp(data, "rwpidforward", 4) == 0) { + vol->rwpidforward = 1; } else if (strnicmp(data, "cifsacl", 7) == 0) { vol->cifs_acl = 1; } else if (strnicmp(data, "nocifsacl", 9) == 0) { @@ -1388,9 +1391,9 @@ cifs_parse_mount_options(char *options, const char *devname, "/proc/fs/cifs/LookupCacheEnabled to 0\n"); } else if (strnicmp(data, "fsc", 3) == 0) { #ifndef CONFIG_CIFS_FSCACHE - cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE" + cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE " "kernel config option set"); - return 1; + goto cifs_parse_mount_err; #endif vol->fsc = true; } else if (strnicmp(data, "mfsymlinks", 10) == 0) { @@ -1405,12 +1408,12 @@ cifs_parse_mount_options(char *options, const char *devname, if (devname == NULL) { printk(KERN_WARNING "CIFS: Missing UNC name for mount " "target\n"); - return 1; + goto cifs_parse_mount_err; } if ((temp_len = strnlen(devname, 300)) < 300) { vol->UNC = kmalloc(temp_len+1, GFP_KERNEL); if (vol->UNC == NULL) - return 1; + goto cifs_parse_mount_err; strcpy(vol->UNC, devname); if (strncmp(vol->UNC, "//", 2) == 0) { vol->UNC[0] = '\\'; @@ -1418,21 +1421,21 @@ cifs_parse_mount_options(char *options, const char *devname, } else if (strncmp(vol->UNC, "\\\\", 2) != 0) { printk(KERN_WARNING "CIFS: UNC Path does not " "begin with // or \\\\ \n"); - return 1; + goto cifs_parse_mount_err; } value = strpbrk(vol->UNC+2, "/\\"); if (value) *value = '\\'; } else { printk(KERN_WARNING "CIFS: UNC name too long\n"); - return 1; + goto cifs_parse_mount_err; } } if (vol->multiuser && !(vol->secFlg & CIFSSEC_MAY_KRB5)) { cERROR(1, "Multiuser mounts currently require krb5 " "authentication!"); - return 1; + goto cifs_parse_mount_err; } if (vol->UNCip == NULL) @@ -1450,7 +1453,12 @@ cifs_parse_mount_options(char *options, const char *devname, printk(KERN_NOTICE "CIFS: ignoring forcegid mount option " "specified with no gid= option.\n"); + kfree(mountdata_copy); return 0; + +cifs_parse_mount_err: + kfree(mountdata_copy); + return 1; } /** Returns true if srcaddr isn't specified and rhs isn't @@ -1589,16 +1597,35 @@ match_security(struct TCP_Server_Info *server, struct smb_vol *vol) /* now check if signing mode is acceptable */ if ((secFlags & CIFSSEC_MAY_SIGN) == 0 && - (server->secMode & SECMODE_SIGN_REQUIRED)) + (server->sec_mode & SECMODE_SIGN_REQUIRED)) return false; else if (((secFlags & CIFSSEC_MUST_SIGN) == CIFSSEC_MUST_SIGN) && - (server->secMode & + (server->sec_mode & (SECMODE_SIGN_ENABLED|SECMODE_SIGN_REQUIRED)) == 0) return false; return true; } +static int match_server(struct TCP_Server_Info *server, struct sockaddr *addr, + struct smb_vol *vol) +{ + if (!net_eq(cifs_net_ns(server), current->nsproxy->net_ns)) + return 0; + + if (!match_address(server, addr, + (struct sockaddr *)&vol->srcaddr)) + return 0; + + if (!match_port(server, addr)) + return 0; + + if (!match_security(server, vol)) + return 0; + + return 1; +} + static struct TCP_Server_Info * cifs_find_tcp_session(struct sockaddr *addr, struct smb_vol *vol) { @@ -1606,17 +1633,7 @@ cifs_find_tcp_session(struct sockaddr *addr, struct smb_vol *vol) spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(server, &cifs_tcp_ses_list, tcp_ses_list) { - if (!net_eq(cifs_net_ns(server), current->nsproxy->net_ns)) - continue; - - if (!match_address(server, addr, - (struct sockaddr *)&vol->srcaddr)) - continue; - - if (!match_port(server, addr)) - continue; - - if (!match_security(server, vol)) + if (!match_server(server, addr, vol)) continue; ++server->srv_count; @@ -1810,32 +1827,39 @@ out_err: return ERR_PTR(rc); } -static struct cifsSesInfo * +static int match_session(struct cifs_ses *ses, struct smb_vol *vol) +{ + switch (ses->server->secType) { + case Kerberos: + if (vol->cred_uid != ses->cred_uid) + return 0; + break; + default: + /* anything else takes username/password */ + if (ses->user_name == NULL) + return 0; + if (strncmp(ses->user_name, vol->username, + MAX_USERNAME_SIZE)) + return 0; + if (strlen(vol->username) != 0 && + ses->password != NULL && + strncmp(ses->password, + vol->password ? vol->password : "", + MAX_PASSWORD_SIZE)) + return 0; + } + return 1; +} + +static struct cifs_ses * cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol) { - struct cifsSesInfo *ses; + struct cifs_ses *ses; spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { - switch (server->secType) { - case Kerberos: - if (vol->cred_uid != ses->cred_uid) - continue; - break; - default: - /* anything else takes username/password */ - if (ses->user_name == NULL) - continue; - if (strncmp(ses->user_name, vol->username, - MAX_USERNAME_SIZE)) - continue; - if (strlen(vol->username) != 0 && - ses->password != NULL && - strncmp(ses->password, - vol->password ? vol->password : "", - MAX_PASSWORD_SIZE)) - continue; - } + if (!match_session(ses, vol)) + continue; ++ses->ses_count; spin_unlock(&cifs_tcp_ses_lock); return ses; @@ -1845,7 +1869,7 @@ cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol) } static void -cifs_put_smb_ses(struct cifsSesInfo *ses) +cifs_put_smb_ses(struct cifs_ses *ses) { int xid; struct TCP_Server_Info *server = ses->server; @@ -1871,11 +1895,11 @@ cifs_put_smb_ses(struct cifsSesInfo *ses) static bool warned_on_ntlm; /* globals init to false automatically */ -static struct cifsSesInfo * +static struct cifs_ses * cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) { int rc = -ENOMEM, xid; - struct cifsSesInfo *ses; + struct cifs_ses *ses; struct sockaddr_in *addr = (struct sockaddr_in *)&server->dstaddr; struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)&server->dstaddr; @@ -1952,7 +1976,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) warned_on_ntlm = true; cERROR(1, "default security mechanism requested. The default " "security mechanism will be upgraded from ntlm to " - "ntlmv2 in kernel release 2.6.41"); + "ntlmv2 in kernel release 3.1"); } ses->overrideSecFlg = volume_info->secFlg; @@ -1978,20 +2002,26 @@ get_ses_fail: return ERR_PTR(rc); } -static struct cifsTconInfo * -cifs_find_tcon(struct cifsSesInfo *ses, const char *unc) +static int match_tcon(struct cifs_tcon *tcon, const char *unc) +{ + if (tcon->tidStatus == CifsExiting) + return 0; + if (strncmp(tcon->treeName, unc, MAX_TREE_SIZE)) + return 0; + return 1; +} + +static struct cifs_tcon * +cifs_find_tcon(struct cifs_ses *ses, const char *unc) { struct list_head *tmp; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; spin_lock(&cifs_tcp_ses_lock); list_for_each(tmp, &ses->tcon_list) { - tcon = list_entry(tmp, struct cifsTconInfo, tcon_list); - if (tcon->tidStatus == CifsExiting) + tcon = list_entry(tmp, struct cifs_tcon, tcon_list); + if (!match_tcon(tcon, unc)) continue; - if (strncmp(tcon->treeName, unc, MAX_TREE_SIZE)) - continue; - ++tcon->tc_count; spin_unlock(&cifs_tcp_ses_lock); return tcon; @@ -2001,10 +2031,10 @@ cifs_find_tcon(struct cifsSesInfo *ses, const char *unc) } static void -cifs_put_tcon(struct cifsTconInfo *tcon) +cifs_put_tcon(struct cifs_tcon *tcon) { int xid; - struct cifsSesInfo *ses = tcon->ses; + struct cifs_ses *ses = tcon->ses; cFYI(1, "%s: tc_count=%d\n", __func__, tcon->tc_count); spin_lock(&cifs_tcp_ses_lock); @@ -2025,11 +2055,11 @@ cifs_put_tcon(struct cifsTconInfo *tcon) cifs_put_smb_ses(ses); } -static struct cifsTconInfo * -cifs_get_tcon(struct cifsSesInfo *ses, struct smb_vol *volume_info) +static struct cifs_tcon * +cifs_get_tcon(struct cifs_ses *ses, struct smb_vol *volume_info) { int rc, xid; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; tcon = cifs_find_tcon(ses, volume_info->UNC); if (tcon) { @@ -2118,8 +2148,105 @@ cifs_put_tlink(struct tcon_link *tlink) return; } +static inline struct tcon_link * +cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb) +{ + return cifs_sb->master_tlink; +} + +static int +compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data) +{ + struct cifs_sb_info *old = CIFS_SB(sb); + struct cifs_sb_info *new = mnt_data->cifs_sb; + + if ((sb->s_flags & CIFS_MS_MASK) != (mnt_data->flags & CIFS_MS_MASK)) + return 0; + + if ((old->mnt_cifs_flags & CIFS_MOUNT_MASK) != + (new->mnt_cifs_flags & CIFS_MOUNT_MASK)) + return 0; + + if (old->rsize != new->rsize) + return 0; + + /* + * We want to share sb only if we don't specify wsize or specified wsize + * is greater or equal than existing one. + */ + if (new->wsize && new->wsize < old->wsize) + return 0; + + if (old->mnt_uid != new->mnt_uid || old->mnt_gid != new->mnt_gid) + return 0; + + if (old->mnt_file_mode != new->mnt_file_mode || + old->mnt_dir_mode != new->mnt_dir_mode) + return 0; + + if (strcmp(old->local_nls->charset, new->local_nls->charset)) + return 0; + + if (old->actimeo != new->actimeo) + return 0; + + return 1; +} + +int +cifs_match_super(struct super_block *sb, void *data) +{ + struct cifs_mnt_data *mnt_data = (struct cifs_mnt_data *)data; + struct smb_vol *volume_info; + struct cifs_sb_info *cifs_sb; + struct TCP_Server_Info *tcp_srv; + struct cifs_ses *ses; + struct cifs_tcon *tcon; + struct tcon_link *tlink; + struct sockaddr_storage addr; + int rc = 0; + + memset(&addr, 0, sizeof(struct sockaddr_storage)); + + spin_lock(&cifs_tcp_ses_lock); + cifs_sb = CIFS_SB(sb); + tlink = cifs_get_tlink(cifs_sb_master_tlink(cifs_sb)); + if (IS_ERR(tlink)) { + spin_unlock(&cifs_tcp_ses_lock); + return rc; + } + tcon = tlink_tcon(tlink); + ses = tcon->ses; + tcp_srv = ses->server; + + volume_info = mnt_data->vol; + + if (!volume_info->UNCip || !volume_info->UNC) + goto out; + + rc = cifs_fill_sockaddr((struct sockaddr *)&addr, + volume_info->UNCip, + strlen(volume_info->UNCip), + volume_info->port); + if (!rc) + goto out; + + if (!match_server(tcp_srv, (struct sockaddr *)&addr, volume_info) || + !match_session(ses, volume_info) || + !match_tcon(tcon, volume_info->UNC)) { + rc = 0; + goto out; + } + + rc = compare_mount_options(sb, mnt_data); +out: + cifs_put_tlink(tlink); + spin_unlock(&cifs_tcp_ses_lock); + return rc; +} + int -get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, +get_dfs_path(int xid, struct cifs_ses *pSesInfo, const char *old_path, const struct nls_table *nls_codepage, unsigned int *pnum_referrals, struct dfs_info3_param **preferrals, int remap) { @@ -2280,7 +2407,7 @@ ip_rfc1001_connect(struct TCP_Server_Info *server) smb_buf = (struct smb_hdr *)ses_init_buf; /* sizeof RFC1002_SESSION_REQUEST with no scope */ - smb_buf->smb_buf_length = 0x81000044; + smb_buf->smb_buf_length = cpu_to_be32(0x81000044); rc = smb_send(server, smb_buf, 0x44); kfree(ses_init_buf); /* @@ -2418,8 +2545,8 @@ ip_connect(struct TCP_Server_Info *server) return generic_ip_connect(server); } -void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, - struct super_block *sb, struct smb_vol *vol_info) +void reset_cifs_unix_caps(int xid, struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, struct smb_vol *vol_info) { /* if we are reconnecting then should we check to see if * any requested capabilities changed locally e.g. via @@ -2447,7 +2574,7 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, if (!CIFSSMBQFSUnixInfo(xid, tcon)) { __u64 cap = le64_to_cpu(tcon->fsUnixInfo.Capability); - + cFYI(1, "unix caps which server supports %lld", cap); /* check for reconnect case in which we do not want to change the mount behavior if we can avoid it */ if (vol_info == NULL) { @@ -2465,33 +2592,31 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, } } + if (cap & CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP) + cERROR(1, "per-share encryption not supported yet"); + cap &= CIFS_UNIX_CAP_MASK; if (vol_info && vol_info->no_psx_acl) cap &= ~CIFS_UNIX_POSIX_ACL_CAP; else if (CIFS_UNIX_POSIX_ACL_CAP & cap) { cFYI(1, "negotiated posix acl support"); - if (sb) - sb->s_flags |= MS_POSIXACL; + if (cifs_sb) + cifs_sb->mnt_cifs_flags |= + CIFS_MOUNT_POSIXACL; } if (vol_info && vol_info->posix_paths == 0) cap &= ~CIFS_UNIX_POSIX_PATHNAMES_CAP; else if (cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) { cFYI(1, "negotiate posix pathnames"); - if (sb) - CIFS_SB(sb)->mnt_cifs_flags |= + if (cifs_sb) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; } - /* We might be setting the path sep back to a different - form if we are reconnecting and the server switched its - posix path capability for this share */ - if (sb && (CIFS_SB(sb)->prepathlen > 0)) - CIFS_SB(sb)->prepath[0] = CIFS_DIR_SEP(CIFS_SB(sb)); - - if (sb && (CIFS_SB(sb)->rsize > 127 * 1024)) { + if (cifs_sb && (cifs_sb->rsize > 127 * 1024)) { if ((cap & CIFS_UNIX_LARGE_READ_CAP) == 0) { - CIFS_SB(sb)->rsize = 127 * 1024; + cifs_sb->rsize = 127 * 1024; cFYI(DBG2, "larger reads not supported by srv"); } } @@ -2513,6 +2638,10 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, cFYI(1, "very large read cap"); if (cap & CIFS_UNIX_LARGE_WRITE_CAP) cFYI(1, "very large write cap"); + if (cap & CIFS_UNIX_TRANSPORT_ENCRYPTION_CAP) + cFYI(1, "transport encryption cap"); + if (cap & CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP) + cFYI(1, "mandatory transport encryption cap"); #endif /* CIFS_DEBUG2 */ if (CIFSSMBSetFSUnixInfo(xid, tcon, cap)) { if (vol_info == NULL) { @@ -2529,31 +2658,14 @@ void reset_cifs_unix_caps(int xid, struct cifsTconInfo *tcon, } } -static void -convert_delimiter(char *path, char delim) -{ - int i; - char old_delim; - - if (path == NULL) - return; - - if (delim == '/') - old_delim = '\\'; - else - old_delim = '/'; - - for (i = 0; path[i] != '\0'; i++) { - if (path[i] == old_delim) - path[i] = delim; - } -} - -static void setup_cifs_sb(struct smb_vol *pvolume_info, - struct cifs_sb_info *cifs_sb) +void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, + struct cifs_sb_info *cifs_sb) { INIT_DELAYED_WORK(&cifs_sb->prune_tlinks, cifs_prune_tlinks); + spin_lock_init(&cifs_sb->tlink_tree_lock); + cifs_sb->tlink_tree = RB_ROOT; + if (pvolume_info->rsize > CIFSMaxBufSize) { cERROR(1, "rsize %d too large, using MaxBufSize", pvolume_info->rsize); @@ -2564,40 +2676,19 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info, else /* default */ cifs_sb->rsize = CIFSMaxBufSize; - if (pvolume_info->wsize > PAGEVEC_SIZE * PAGE_CACHE_SIZE) { - cERROR(1, "wsize %d too large, using 4096 instead", - pvolume_info->wsize); - cifs_sb->wsize = 4096; - } else if (pvolume_info->wsize) - cifs_sb->wsize = pvolume_info->wsize; - else - cifs_sb->wsize = min_t(const int, - PAGEVEC_SIZE * PAGE_CACHE_SIZE, - 127*1024); - /* old default of CIFSMaxBufSize was too small now - that SMB Write2 can send multiple pages in kvec. - RFC1001 does not describe what happens when frame - bigger than 128K is sent so use that as max in - conjunction with 52K kvec constraint on arch with 4K - page size */ - if (cifs_sb->rsize < 2048) { cifs_sb->rsize = 2048; /* Windows ME may prefer this */ cFYI(1, "readsize set to minimum: 2048"); } - /* calculate prepath */ - cifs_sb->prepath = pvolume_info->prepath; - if (cifs_sb->prepath) { - cifs_sb->prepathlen = strlen(cifs_sb->prepath); - /* we can not convert the / to \ in the path - separators in the prefixpath yet because we do not - know (until reset_cifs_unix_caps is called later) - whether POSIX PATH CAP is available. We normalize - the / to \ after reset_cifs_unix_caps is called */ - pvolume_info->prepath = NULL; - } else - cifs_sb->prepathlen = 0; + + /* + * Temporarily set wsize for matching superblock. If we end up using + * new sb then cifs_negotiate_wsize will later negotiate it downward + * if needed. + */ + cifs_sb->wsize = pvolume_info->wsize; + cifs_sb->mnt_uid = pvolume_info->linux_uid; cifs_sb->mnt_gid = pvolume_info->linux_gid; cifs_sb->mnt_file_mode = pvolume_info->file_mode; @@ -2606,6 +2697,7 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info, cifs_sb->mnt_file_mode, cifs_sb->mnt_dir_mode); cifs_sb->actimeo = pvolume_info->actimeo; + cifs_sb->local_nls = pvolume_info->local_nls; if (pvolume_info->noperm) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_PERM; @@ -2625,6 +2717,8 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOSSYNC; if (pvolume_info->mand_lock) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOPOSIXBRL; + if (pvolume_info->rwpidforward) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_RWPIDFORWARD; if (pvolume_info->cifs_acl) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_ACL; if (pvolume_info->override_uid) @@ -2658,8 +2752,62 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info, "mount option supported"); } +/* + * When the server supports very large writes via POSIX extensions, we can + * allow up to 2^24-1, minus the size of a WRITE_AND_X header, not including + * the RFC1001 length. + * + * Note that this might make for "interesting" allocation problems during + * writeback however as we have to allocate an array of pointers for the + * pages. A 16M write means ~32kb page array with PAGE_CACHE_SIZE == 4096. + */ +#define CIFS_MAX_WSIZE ((1<<24) - 1 - sizeof(WRITE_REQ) + 4) + +/* + * When the server doesn't allow large posix writes, only allow a wsize of + * 128k minus the size of the WRITE_AND_X header. That allows for a write up + * to the maximum size described by RFC1002. + */ +#define CIFS_MAX_RFC1002_WSIZE (128 * 1024 - sizeof(WRITE_REQ) + 4) + +/* + * The default wsize is 1M. find_get_pages seems to return a maximum of 256 + * pages in a single call. With PAGE_CACHE_SIZE == 4k, this means we can fill + * a single wsize request with a single call. + */ +#define CIFS_DEFAULT_WSIZE (1024 * 1024) + +static unsigned int +cifs_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *pvolume_info) +{ + __u64 unix_cap = le64_to_cpu(tcon->fsUnixInfo.Capability); + struct TCP_Server_Info *server = tcon->ses->server; + unsigned int wsize = pvolume_info->wsize ? pvolume_info->wsize : + CIFS_DEFAULT_WSIZE; + + /* can server support 24-bit write sizes? (via UNIX extensions) */ + if (!tcon->unix_ext || !(unix_cap & CIFS_UNIX_LARGE_WRITE_CAP)) + wsize = min_t(unsigned int, wsize, CIFS_MAX_RFC1002_WSIZE); + + /* + * no CAP_LARGE_WRITE_X or is signing enabled without CAP_UNIX set? + * Limit it to max buffer offered by the server, minus the size of the + * WRITEX header, not including the 4 byte RFC1001 length. + */ + if (!(server->capabilities & CAP_LARGE_WRITE_X) || + (!(server->capabilities & CAP_UNIX) && + (server->sec_mode & (SECMODE_SIGN_ENABLED|SECMODE_SIGN_REQUIRED)))) + wsize = min_t(unsigned int, wsize, + server->maxBuf - sizeof(WRITE_REQ) + 4); + + /* hard limit of CIFS_MAX_WSIZE */ + wsize = min_t(unsigned int, wsize, CIFS_MAX_WSIZE); + + return wsize; +} + static int -is_path_accessible(int xid, struct cifsTconInfo *tcon, +is_path_accessible(int xid, struct cifs_tcon *tcon, struct cifs_sb_info *cifs_sb, const char *full_path) { int rc; @@ -2682,8 +2830,8 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon, return rc; } -static void -cleanup_volume_info(struct smb_vol **pvolume_info) +void +cifs_cleanup_volume_info(struct smb_vol **pvolume_info) { struct smb_vol *volume_info; @@ -2691,8 +2839,12 @@ cleanup_volume_info(struct smb_vol **pvolume_info) return; volume_info = *pvolume_info; + kfree(volume_info->username); kzfree(volume_info->password); kfree(volume_info->UNC); + kfree(volume_info->UNCip); + kfree(volume_info->domainname); + kfree(volume_info->iocharset); kfree(volume_info->prepath); kfree(volume_info); *pvolume_info = NULL; @@ -2709,55 +2861,78 @@ build_unc_path_to_root(const struct smb_vol *volume_info, char *full_path; int unc_len = strnlen(volume_info->UNC, MAX_TREE_SIZE + 1); - full_path = kmalloc(unc_len + cifs_sb->prepathlen + 1, GFP_KERNEL); + full_path = kmalloc(unc_len + 1, GFP_KERNEL); if (full_path == NULL) return ERR_PTR(-ENOMEM); strncpy(full_path, volume_info->UNC, unc_len); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) { - int i; - for (i = 0; i < unc_len; i++) { - if (full_path[i] == '\\') - full_path[i] = '/'; - } - } - - if (cifs_sb->prepathlen) - strncpy(full_path + unc_len, cifs_sb->prepath, - cifs_sb->prepathlen); - - full_path[unc_len + cifs_sb->prepathlen] = 0; /* add trailing null */ + full_path[unc_len] = 0; /* add trailing null */ + convert_delimiter(full_path, CIFS_DIR_SEP(cifs_sb)); return full_path; } -#endif -int -cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, - char *mount_data_global, const char *devname) +/* + * Perform a dfs referral query for a share and (optionally) prefix + * + * If a referral is found, cifs_sb->mountdata will be (re-)allocated + * to a string containing updated options for the submount. Otherwise it + * will be left untouched. + * + * Returns the rc from get_dfs_path to the caller, which can be used to + * determine whether there were referrals. + */ +static int +expand_dfs_referral(int xid, struct cifs_ses *pSesInfo, + struct smb_vol *volume_info, struct cifs_sb_info *cifs_sb, + int check_prefix) { int rc; - int xid; - struct smb_vol *volume_info; - struct cifsSesInfo *pSesInfo; - struct cifsTconInfo *tcon; - struct TCP_Server_Info *srvTcp; - char *full_path; - char *mount_data = mount_data_global; - struct tcon_link *tlink; -#ifdef CONFIG_CIFS_DFS_UPCALL - struct dfs_info3_param *referrals = NULL; unsigned int num_referrals = 0; - int referral_walks_count = 0; -try_mount_again: + struct dfs_info3_param *referrals = NULL; + char *full_path = NULL, *ref_path = NULL, *mdata = NULL; + + full_path = build_unc_path_to_root(volume_info, cifs_sb); + if (IS_ERR(full_path)) + return PTR_ERR(full_path); + + /* For DFS paths, skip the first '\' of the UNC */ + ref_path = check_prefix ? full_path + 1 : volume_info->UNC + 1; + + rc = get_dfs_path(xid, pSesInfo , ref_path, cifs_sb->local_nls, + &num_referrals, &referrals, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (!rc && num_referrals > 0) { + char *fake_devname = NULL; + + mdata = cifs_compose_mount_options(cifs_sb->mountdata, + full_path + 1, referrals, + &fake_devname); + + free_dfs_info_array(referrals, num_referrals); + kfree(fake_devname); + + if (cifs_sb->mountdata != NULL) + kfree(cifs_sb->mountdata); + + if (IS_ERR(mdata)) { + rc = PTR_ERR(mdata); + mdata = NULL; + } + cifs_sb->mountdata = mdata; + } + kfree(full_path); + return rc; +} #endif - rc = 0; - tcon = NULL; - pSesInfo = NULL; - srvTcp = NULL; - full_path = NULL; - tlink = NULL; - xid = GetXid(); +int cifs_setup_volume_info(struct smb_vol **pvolume_info, char *mount_data, + const char *devname) +{ + struct smb_vol *volume_info; + int rc = 0; + + *pvolume_info = NULL; volume_info = kzalloc(sizeof(struct smb_vol), GFP_KERNEL); if (!volume_info) { @@ -2765,14 +2940,19 @@ try_mount_again: goto out; } - if (cifs_parse_mount_options(mount_data, devname, volume_info)) { + if (cifs_parse_mount_options(mount_data, devname, + volume_info)) { rc = -EINVAL; goto out; } if (volume_info->nullauth) { cFYI(1, "null user"); - volume_info->username = ""; + volume_info->username = kzalloc(1, GFP_KERNEL); + if (volume_info->username == NULL) { + rc = -ENOMEM; + goto out; + } } else if (volume_info->username) { /* BB fixme parse for domain name here */ cFYI(1, "Username: %s", volume_info->username); @@ -2797,12 +2977,58 @@ try_mount_again: goto out; } } - cifs_sb->local_nls = volume_info->local_nls; + + *pvolume_info = volume_info; + return rc; +out: + cifs_cleanup_volume_info(&volume_info); + return rc; +} + +int +cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *volume_info) +{ + int rc = 0; + int xid; + struct cifs_ses *pSesInfo; + struct cifs_tcon *tcon; + struct TCP_Server_Info *srvTcp; + char *full_path; + struct tcon_link *tlink; +#ifdef CONFIG_CIFS_DFS_UPCALL + int referral_walks_count = 0; + + rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY); + if (rc) + return rc; + + cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages; + +try_mount_again: + /* cleanup activities if we're chasing a referral */ + if (referral_walks_count) { + if (tcon) + cifs_put_tcon(tcon); + else if (pSesInfo) + cifs_put_smb_ses(pSesInfo); + + cifs_cleanup_volume_info(&volume_info); + FreeXid(xid); + } +#endif + tcon = NULL; + pSesInfo = NULL; + srvTcp = NULL; + full_path = NULL; + tlink = NULL; + + xid = GetXid(); /* get a reference to a tcp session */ srvTcp = cifs_get_tcp_session(volume_info); if (IS_ERR(srvTcp)) { rc = PTR_ERR(srvTcp); + bdi_destroy(&cifs_sb->bdi); goto out; } @@ -2814,15 +3040,6 @@ try_mount_again: goto mount_fail_check; } - setup_cifs_sb(volume_info, cifs_sb); - if (pSesInfo->capabilities & CAP_LARGE_FILES) - sb->s_maxbytes = MAX_LFS_FILESIZE; - else - sb->s_maxbytes = MAX_NON_LFS; - - /* BB FIXME fix time_gran to be larger for LANMAN sessions */ - sb->s_time_gran = 100; - /* search for existing tcon to this server share */ tcon = cifs_get_tcon(pSesInfo, volume_info); if (IS_ERR(tcon)) { @@ -2831,40 +3048,59 @@ try_mount_again: goto remote_path_check; } - /* do not care if following two calls succeed - informational */ - if (!tcon->ipc) { - CIFSSMBQFSDeviceInfo(xid, tcon); - CIFSSMBQFSAttributeInfo(xid, tcon); - } - /* tell server which Unix caps we support */ - if (tcon->ses->capabilities & CAP_UNIX) + if (tcon->ses->capabilities & CAP_UNIX) { /* reset of caps checks mount to see if unix extensions disabled for just this mount */ - reset_cifs_unix_caps(xid, tcon, sb, volume_info); - else + reset_cifs_unix_caps(xid, tcon, cifs_sb, volume_info); + if ((tcon->ses->server->tcpStatus == CifsNeedReconnect) && + (le64_to_cpu(tcon->fsUnixInfo.Capability) & + CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP)) { + rc = -EACCES; + goto mount_fail_check; + } + } else tcon->unix_ext = 0; /* server does not support them */ - /* convert forward to back slashes in prepath here if needed */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0) - convert_delimiter(cifs_sb->prepath, CIFS_DIR_SEP(cifs_sb)); + /* do not care if following two calls succeed - informational */ + if (!tcon->ipc) { + CIFSSMBQFSDeviceInfo(xid, tcon); + CIFSSMBQFSAttributeInfo(xid, tcon); + } if ((tcon->unix_ext == 0) && (cifs_sb->rsize > (1024 * 127))) { cifs_sb->rsize = 1024 * 127; cFYI(DBG2, "no very large read support, rsize now 127K"); } - if (!(tcon->ses->capabilities & CAP_LARGE_WRITE_X)) - cifs_sb->wsize = min(cifs_sb->wsize, - (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE)); if (!(tcon->ses->capabilities & CAP_LARGE_READ_X)) cifs_sb->rsize = min(cifs_sb->rsize, (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE)); + cifs_sb->wsize = cifs_negotiate_wsize(tcon, volume_info); + remote_path_check: - /* check if a whole path (including prepath) is not remote */ +#ifdef CONFIG_CIFS_DFS_UPCALL + /* + * Perform an unconditional check for whether there are DFS + * referrals for this path without prefix, to provide support + * for DFS referrals from w2k8 servers which don't seem to respond + * with PATH_NOT_COVERED to requests that include the prefix. + * Chase the referral if found, otherwise continue normally. + */ + if (referral_walks_count == 0) { + int refrc = expand_dfs_referral(xid, pSesInfo, volume_info, + cifs_sb, false); + if (!refrc) { + referral_walks_count++; + goto try_mount_again; + } + } +#endif + + /* check if a whole path is not remote */ if (!rc && tcon) { /* build_path_to_root works only when we have a valid tcon */ - full_path = cifs_build_path_to_root(cifs_sb, tcon); + full_path = cifs_build_path_to_root(volume_info, cifs_sb, tcon); if (full_path == NULL) { rc = -ENOMEM; goto mount_fail_check; @@ -2890,50 +3126,15 @@ remote_path_check: rc = -ELOOP; goto mount_fail_check; } - /* convert forward to back slashes in prepath here if needed */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0) - convert_delimiter(cifs_sb->prepath, - CIFS_DIR_SEP(cifs_sb)); - full_path = build_unc_path_to_root(volume_info, cifs_sb); - if (IS_ERR(full_path)) { - rc = PTR_ERR(full_path); - goto mount_fail_check; - } - - cFYI(1, "Getting referral for: %s", full_path); - rc = get_dfs_path(xid, pSesInfo , full_path + 1, - cifs_sb->local_nls, &num_referrals, &referrals, - cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - if (!rc && num_referrals > 0) { - char *fake_devname = NULL; - - if (mount_data != mount_data_global) - kfree(mount_data); - - mount_data = cifs_compose_mount_options( - cifs_sb->mountdata, full_path + 1, - referrals, &fake_devname); - free_dfs_info_array(referrals, num_referrals); - kfree(fake_devname); - kfree(full_path); - - if (IS_ERR(mount_data)) { - rc = PTR_ERR(mount_data); - mount_data = NULL; - goto mount_fail_check; - } + rc = expand_dfs_referral(xid, pSesInfo, volume_info, cifs_sb, + true); - if (tcon) - cifs_put_tcon(tcon); - else if (pSesInfo) - cifs_put_smb_ses(pSesInfo); - - cleanup_volume_info(&volume_info); + if (!rc) { referral_walks_count++; - FreeXid(xid); goto try_mount_again; } + goto mount_fail_check; #else /* No DFS support, return error on mount */ rc = -EOPNOTSUPP; #endif @@ -2966,8 +3167,6 @@ remote_path_check: mount_fail_check: /* on error free sesinfo and tcon struct if needed */ if (rc) { - if (mount_data != mount_data_global) - kfree(mount_data); /* If find_unc succeeded then rc == 0 so we can not end */ /* up accidentally freeing someone elses tcon struct */ if (tcon) @@ -2976,6 +3175,7 @@ mount_fail_check: cifs_put_smb_ses(pSesInfo); else cifs_put_tcp_session(srvTcp); + bdi_destroy(&cifs_sb->bdi); goto out; } @@ -2985,14 +3185,17 @@ mount_fail_check: password will be freed at unmount time) */ out: /* zero out password before freeing */ - cleanup_volume_info(&volume_info); FreeXid(xid); return rc; } +/* + * Issue a TREE_CONNECT request. Note that for IPC$ shares, that the tcon + * pointer may be NULL. + */ int -CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, - const char *tree, struct cifsTconInfo *tcon, +CIFSTCon(unsigned int xid, struct cifs_ses *ses, + const char *tree, struct cifs_tcon *tcon, const struct nls_table *nls_codepage) { struct smb_hdr *smb_buffer; @@ -3024,7 +3227,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, pSMB->AndXCommand = 0xFF; pSMB->Flags = cpu_to_le16(TCON_EXTENDED_SECINFO); bcc_ptr = &pSMB->Password[0]; - if ((ses->server->secMode) & SECMODE_USER) { + if (!tcon || (ses->server->sec_mode & SECMODE_USER)) { pSMB->PasswordLength = cpu_to_le16(1); /* minimum */ *bcc_ptr = 0; /* password is null byte */ bcc_ptr++; /* skip password */ @@ -3041,7 +3244,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if ((global_secflags & CIFSSEC_MAY_LANMAN) && (ses->server->secType == LANMAN)) calc_lanman_hash(tcon->password, ses->server->cryptkey, - ses->server->secMode & + ses->server->sec_mode & SECMODE_PW_ENCRYPT ? true : false, bcc_ptr); else @@ -3057,7 +3260,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, } } - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) smb_buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; @@ -3083,7 +3286,8 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr += strlen("?????"); bcc_ptr += 1; count = bcc_ptr - &pSMB->Password[0]; - pSMB->hdr.smb_buf_length += count; + pSMB->hdr.smb_buf_length = cpu_to_be32(be32_to_cpu( + pSMB->hdr.smb_buf_length) + count); pSMB->ByteCount = cpu_to_le16(count); rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, &length, @@ -3146,13 +3350,12 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, return rc; } -int -cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) +void +cifs_umount(struct cifs_sb_info *cifs_sb) { struct rb_root *root = &cifs_sb->tlink_tree; struct rb_node *node; struct tcon_link *tlink; - char *tmp; cancel_delayed_work_sync(&cifs_sb->prune_tlinks); @@ -3169,15 +3372,13 @@ cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) } spin_unlock(&cifs_sb->tlink_tree_lock); - tmp = cifs_sb->prepath; - cifs_sb->prepathlen = 0; - cifs_sb->prepath = NULL; - kfree(tmp); - - return 0; + bdi_destroy(&cifs_sb->bdi); + kfree(cifs_sb->mountdata); + unload_nls(cifs_sb->local_nls); + kfree(cifs_sb); } -int cifs_negotiate_protocol(unsigned int xid, struct cifsSesInfo *ses) +int cifs_negotiate_protocol(unsigned int xid, struct cifs_ses *ses) { int rc = 0; struct TCP_Server_Info *server = ses->server; @@ -3195,7 +3396,7 @@ int cifs_negotiate_protocol(unsigned int xid, struct cifsSesInfo *ses) } if (rc == 0) { spin_lock(&GlobalMid_Lock); - if (server->tcpStatus != CifsExiting) + if (server->tcpStatus == CifsNeedNegotiate) server->tcpStatus = CifsGood; else rc = -EHOSTDOWN; @@ -3207,7 +3408,7 @@ int cifs_negotiate_protocol(unsigned int xid, struct cifsSesInfo *ses) } -int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, +int cifs_setup_session(unsigned int xid, struct cifs_ses *ses, struct nls_table *nls_info) { int rc = 0; @@ -3219,7 +3420,7 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, ses->capabilities &= (~CAP_UNIX); cFYI(1, "Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", - server->secMode, server->capabilities, server->timeAdj); + server->sec_mode, server->capabilities, server->timeAdj); rc = CIFS_SessSetup(xid, ses, nls_info); if (rc) { @@ -3251,14 +3452,16 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, return rc; } -static struct cifsTconInfo * +static struct cifs_tcon * cifs_construct_tcon(struct cifs_sb_info *cifs_sb, uid_t fsuid) { - struct cifsTconInfo *master_tcon = cifs_sb_master_tcon(cifs_sb); - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon = NULL; + struct cifs_tcon *master_tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_ses *ses; + struct cifs_tcon *tcon = NULL; struct smb_vol *vol_info; - char username[MAX_USERNAME_SIZE + 1]; + char username[28]; /* big enough for "krb50x" + hex of ULONG_MAX 6+16 */ + /* We used to have this as MAX_USERNAME which is */ + /* way too big now (256 instead of 32) */ vol_info = kzalloc(sizeof(*vol_info), GFP_KERNEL); if (vol_info == NULL) { @@ -3287,7 +3490,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, uid_t fsuid) ses = cifs_get_smb_ses(master_tcon->ses->server, vol_info); if (IS_ERR(ses)) { - tcon = (struct cifsTconInfo *)ses; + tcon = (struct cifs_tcon *)ses; cifs_put_tcp_session(master_tcon->ses->server); goto out; } @@ -3306,13 +3509,7 @@ out: return tcon; } -static inline struct tcon_link * -cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb) -{ - return cifs_sb->master_tlink; -} - -struct cifsTconInfo * +struct cifs_tcon * cifs_sb_master_tcon(struct cifs_sb_info *cifs_sb) { return tlink_tcon(cifs_sb_master_tlink(cifs_sb)); diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 9ea65cf..81914df 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -50,12 +50,11 @@ build_path_from_dentry(struct dentry *direntry) { struct dentry *temp; int namelen; - int pplen; int dfsplen; char *full_path; char dirsep; struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); if (direntry == NULL) return NULL; /* not much we can do if dentry is freed and @@ -63,13 +62,12 @@ build_path_from_dentry(struct dentry *direntry) when the server crashed */ dirsep = CIFS_DIR_SEP(cifs_sb); - pplen = cifs_sb->prepathlen; if (tcon->Flags & SMB_SHARE_IS_IN_DFS) dfsplen = strnlen(tcon->treeName, MAX_TREE_SIZE + 1); else dfsplen = 0; cifs_bp_rename_retry: - namelen = pplen + dfsplen; + namelen = dfsplen; for (temp = direntry; !IS_ROOT(temp);) { namelen += (1 + temp->d_name.len); temp = temp->d_parent; @@ -100,7 +98,7 @@ cifs_bp_rename_retry: return NULL; } } - if (namelen != pplen + dfsplen) { + if (namelen != dfsplen) { cERROR(1, "did not end path lookup where expected namelen is %d", namelen); /* presumably this is only possible if racing with a rename @@ -126,7 +124,6 @@ cifs_bp_rename_retry: } } } - strncpy(full_path + dfsplen, CIFS_SB(direntry->d_sb)->prepath, pplen); return full_path; } @@ -152,7 +149,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, __u16 fileHandle; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; char *full_path = NULL; FILE_ALL_INFO *buf = NULL; struct inode *newinode = NULL; @@ -356,7 +353,8 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; + struct cifs_io_parms io_parms; char *full_path = NULL; struct inode *newinode = NULL; int oplock = 0; @@ -439,16 +437,19 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, * timestamps in, but we can reuse it safely */ pdev = (struct win_dev *)buf; + io_parms.netfid = fileHandle; + io_parms.pid = current->tgid; + io_parms.tcon = pTcon; + io_parms.offset = 0; + io_parms.length = sizeof(struct win_dev); if (S_ISCHR(mode)) { memcpy(pdev->type, "IntxCHR", 8); pdev->major = cpu_to_le64(MAJOR(device_number)); pdev->minor = cpu_to_le64(MINOR(device_number)); - rc = CIFSSMBWrite(xid, pTcon, - fileHandle, - sizeof(struct win_dev), - 0, &bytes_written, (char *)pdev, + rc = CIFSSMBWrite(xid, &io_parms, + &bytes_written, (char *)pdev, NULL, 0); } else if (S_ISBLK(mode)) { memcpy(pdev->type, "IntxBLK", 8); @@ -456,10 +457,8 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, cpu_to_le64(MAJOR(device_number)); pdev->minor = cpu_to_le64(MINOR(device_number)); - rc = CIFSSMBWrite(xid, pTcon, - fileHandle, - sizeof(struct win_dev), - 0, &bytes_written, (char *)pdev, + rc = CIFSSMBWrite(xid, &io_parms, + &bytes_written, (char *)pdev, NULL, 0); } /* else if (S_ISFIFO) */ CIFSSMBClose(xid, pTcon, fileHandle); @@ -486,7 +485,7 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, bool posix_open = false; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifsFileInfo *cfile; struct inode *newInode = NULL; char *full_path = NULL; diff --git a/fs/cifs/export.c b/fs/cifs/export.c index 993f820..55d87ac 100644 --- a/fs/cifs/export.c +++ b/fs/cifs/export.c @@ -45,7 +45,7 @@ #include "cifs_debug.h" #include "cifsfs.h" -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CIFS_NFSD_EXPORT static struct dentry *cifs_get_parent(struct dentry *dentry) { /* BB need to add code here eventually to enable export via NFSD */ @@ -63,5 +63,5 @@ const struct export_operations cifs_export_ops = { .encode_fs = */ }; -#endif /* EXPERIMENTAL */ +#endif /* CIFS_NFSD_EXPORT */ diff --git a/fs/cifs/file.c b/fs/cifs/file.c index faf5952..bb71471 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -114,7 +114,7 @@ int cifs_posix_open(char *full_path, struct inode **pinode, struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct cifs_fattr fattr; struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; cFYI(1, "posix open %s", full_path); @@ -168,7 +168,7 @@ posix_open_ret: static int cifs_nt_open(char *full_path, struct inode *inode, struct cifs_sb_info *cifs_sb, - struct cifsTconInfo *tcon, unsigned int f_flags, __u32 *poplock, + struct cifs_tcon *tcon, unsigned int f_flags, __u32 *poplock, __u16 *pnetfid, int xid) { int rc; @@ -285,7 +285,7 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, void cifsFileInfo_put(struct cifsFileInfo *cifs_file) { struct inode *inode = cifs_file->dentry->d_inode; - struct cifsTconInfo *tcon = tlink_tcon(cifs_file->tlink); + struct cifs_tcon *tcon = tlink_tcon(cifs_file->tlink); struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifsLockInfo *li, *tmp; @@ -343,7 +343,7 @@ int cifs_open(struct inode *inode, struct file *file) int xid; __u32 oplock; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct tcon_link *tlink; struct cifsFileInfo *pCifsFile = NULL; char *full_path = NULL; @@ -457,7 +457,7 @@ static int cifs_reopen_file(struct cifsFileInfo *pCifsFile, bool can_flush) int xid; __u32 oplock; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct cifsInodeInfo *pCifsInode; struct inode *inode; char *full_path = NULL; @@ -596,7 +596,7 @@ int cifs_closedir(struct inode *inode, struct file *file) xid = GetXid(); if (pCFileStruct) { - struct cifsTconInfo *pTcon = tlink_tcon(pCFileStruct->tlink); + struct cifs_tcon *pTcon = tlink_tcon(pCFileStruct->tlink); cFYI(1, "Freeing private data in close dir"); spin_lock(&cifs_file_list_lock); @@ -653,7 +653,7 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) __u64 length; bool wait_flag = false; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; __u16 netfid; __u8 lockType = LOCKING_ANDX_LARGE_FILES; bool posix_locking = 0; @@ -725,8 +725,8 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) else posix_lock_type = CIFS_WRLCK; rc = CIFSSMBPosixLock(xid, tcon, netfid, 1 /* get */, - length, pfLock, - posix_lock_type, wait_flag); + length, pfLock, posix_lock_type, + wait_flag); FreeXid(xid); return rc; } @@ -797,8 +797,8 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) posix_lock_type = CIFS_UNLCK; rc = CIFSSMBPosixLock(xid, tcon, netfid, 0 /* set */, - length, pfLock, - posix_lock_type, wait_flag); + length, pfLock, posix_lock_type, + wait_flag); } else { struct cifsFileInfo *fid = file->private_data; @@ -857,96 +857,7 @@ cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, cifsi->server_eof = end_of_write; } -ssize_t cifs_user_write(struct file *file, const char __user *write_data, - size_t write_size, loff_t *poffset) -{ - struct inode *inode = file->f_path.dentry->d_inode; - int rc = 0; - unsigned int bytes_written = 0; - unsigned int total_written; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; - int xid; - struct cifsFileInfo *open_file; - struct cifsInodeInfo *cifsi = CIFS_I(inode); - - cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); - - /* cFYI(1, " write %d bytes to offset %lld of %s", write_size, - *poffset, file->f_path.dentry->d_name.name); */ - - if (file->private_data == NULL) - return -EBADF; - - open_file = file->private_data; - pTcon = tlink_tcon(open_file->tlink); - - rc = generic_write_checks(file, poffset, &write_size, 0); - if (rc) - return rc; - - xid = GetXid(); - - for (total_written = 0; write_size > total_written; - total_written += bytes_written) { - rc = -EAGAIN; - while (rc == -EAGAIN) { - if (file->private_data == NULL) { - /* file has been closed on us */ - FreeXid(xid); - /* if we have gotten here we have written some data - and blocked, and the file has been freed on us while - we blocked so return what we managed to write */ - return total_written; - } - if (open_file->invalidHandle) { - /* we could deadlock if we called - filemap_fdatawait from here so tell - reopen_file not to flush data to server - now */ - rc = cifs_reopen_file(open_file, false); - if (rc != 0) - break; - } - - rc = CIFSSMBWrite(xid, pTcon, - open_file->netfid, - min_t(const int, cifs_sb->wsize, - write_size - total_written), - *poffset, &bytes_written, - NULL, write_data + total_written, 0); - } - if (rc || (bytes_written == 0)) { - if (total_written) - break; - else { - FreeXid(xid); - return rc; - } - } else { - cifs_update_eof(cifsi, *poffset, bytes_written); - *poffset += bytes_written; - } - } - - cifs_stats_bytes_written(pTcon, total_written); - -/* Do not update local mtime - server will set its actual value on write - * inode->i_ctime = inode->i_mtime = - * current_fs_time(inode->i_sb);*/ - if (total_written > 0) { - spin_lock(&inode->i_lock); - if (*poffset > inode->i_size) - i_size_write(inode, *poffset); - spin_unlock(&inode->i_lock); - } - mark_inode_dirty_sync(inode); - - FreeXid(xid); - return total_written; -} - -static ssize_t cifs_write(struct cifsFileInfo *open_file, +static ssize_t cifs_write(struct cifsFileInfo *open_file, __u32 pid, const char *write_data, size_t write_size, loff_t *poffset) { @@ -954,10 +865,11 @@ static ssize_t cifs_write(struct cifsFileInfo *open_file, unsigned int bytes_written = 0; unsigned int total_written; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; int xid; struct dentry *dentry = open_file->dentry; struct cifsInodeInfo *cifsi = CIFS_I(dentry->d_inode); + struct cifs_io_parms io_parms; cifs_sb = CIFS_SB(dentry->d_sb); @@ -990,8 +902,13 @@ static ssize_t cifs_write(struct cifsFileInfo *open_file, /* iov[0] is reserved for smb header */ iov[1].iov_base = (char *)write_data + total_written; iov[1].iov_len = len; - rc = CIFSSMBWrite2(xid, pTcon, open_file->netfid, len, - *poffset, &bytes_written, iov, 1, 0); + io_parms.netfid = open_file->netfid; + io_parms.pid = pid; + io_parms.tcon = pTcon; + io_parms.offset = *poffset; + io_parms.length = len; + rc = CIFSSMBWrite2(xid, &io_parms, &bytes_written, iov, + 1, 0); } if (rc || (bytes_written == 0)) { if (total_written) @@ -1160,8 +1077,8 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) open_file = find_writable_file(CIFS_I(mapping->host), false); if (open_file) { - bytes_written = cifs_write(open_file, write_data, - to - from, &offset); + bytes_written = cifs_write(open_file, open_file->pid, + write_data, to - from, &offset); cifsFileInfo_put(open_file); /* Does mm or vfs already set times? */ inode->i_atime = inode->i_mtime = current_fs_time(inode->i_sb); @@ -1181,58 +1098,20 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) static int cifs_writepages(struct address_space *mapping, struct writeback_control *wbc) { - unsigned int bytes_to_write; - unsigned int bytes_written; - struct cifs_sb_info *cifs_sb; - int done = 0; - pgoff_t end; - pgoff_t index; - int range_whole = 0; - struct kvec *iov; - int len; - int n_iov = 0; - pgoff_t next; - int nr_pages; - __u64 offset = 0; - struct cifsFileInfo *open_file; - struct cifsTconInfo *tcon; - struct cifsInodeInfo *cifsi = CIFS_I(mapping->host); + struct cifs_sb_info *cifs_sb = CIFS_SB(mapping->host->i_sb); + bool done = false, scanned = false, range_whole = false; + pgoff_t end, index; + struct cifs_writedata *wdata; struct page *page; - struct pagevec pvec; int rc = 0; - int scanned = 0; - int xid; - - cifs_sb = CIFS_SB(mapping->host->i_sb); /* - * If wsize is smaller that the page cache size, default to writing + * If wsize is smaller than the page cache size, default to writing * one page at a time via cifs_writepage */ if (cifs_sb->wsize < PAGE_CACHE_SIZE) return generic_writepages(mapping, wbc); - iov = kmalloc(32 * sizeof(struct kvec), GFP_KERNEL); - if (iov == NULL) - return generic_writepages(mapping, wbc); - - /* - * if there's no open file, then this is likely to fail too, - * but it'll at least handle the return. Maybe it should be - * a BUG() instead? - */ - open_file = find_writable_file(CIFS_I(mapping->host), false); - if (!open_file) { - kfree(iov); - return generic_writepages(mapping, wbc); - } - - tcon = tlink_tcon(open_file->tlink); - cifsFileInfo_put(open_file); - - xid = GetXid(); - - pagevec_init(&pvec, 0); if (wbc->range_cyclic) { index = mapping->writeback_index; /* Start from prev offset */ end = -1; @@ -1240,24 +1119,49 @@ static int cifs_writepages(struct address_space *mapping, index = wbc->range_start >> PAGE_CACHE_SHIFT; end = wbc->range_end >> PAGE_CACHE_SHIFT; if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) - range_whole = 1; - scanned = 1; + range_whole = true; + scanned = true; } retry: - while (!done && (index <= end) && - (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, - PAGECACHE_TAG_DIRTY, - min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1))) { - int first; - unsigned int i; - - first = -1; - next = 0; - n_iov = 0; - bytes_to_write = 0; - - for (i = 0; i < nr_pages; i++) { - page = pvec.pages[i]; + while (!done && index <= end) { + unsigned int i, nr_pages, found_pages; + pgoff_t next = 0, tofind; + struct page **pages; + + tofind = min((cifs_sb->wsize / PAGE_CACHE_SIZE) - 1, + end - index) + 1; + + wdata = cifs_writedata_alloc((unsigned int)tofind); + if (!wdata) { + rc = -ENOMEM; + break; + } + + /* + * find_get_pages_tag seems to return a max of 256 on each + * iteration, so we must call it several times in order to + * fill the array or the wsize is effectively limited to + * 256 * PAGE_CACHE_SIZE. + */ + found_pages = 0; + pages = wdata->pages; + do { + nr_pages = find_get_pages_tag(mapping, &index, + PAGECACHE_TAG_DIRTY, + tofind, pages); + found_pages += nr_pages; + tofind -= nr_pages; + pages += nr_pages; + } while (nr_pages && tofind && index <= end); + + if (found_pages == 0) { + kref_put(&wdata->refcount, cifs_writedata_release); + break; + } + + nr_pages = 0; + for (i = 0; i < found_pages; i++) { + page = wdata->pages[i]; /* * At this point we hold neither mapping->tree_lock nor * lock on the page itself: the page may be truncated or @@ -1266,7 +1170,7 @@ retry: * mapping */ - if (first < 0) + if (nr_pages == 0) lock_page(page); else if (!trylock_page(page)) break; @@ -1277,7 +1181,7 @@ retry: } if (!wbc->range_cyclic && page->index > end) { - done = 1; + done = true; unlock_page(page); break; } @@ -1304,125 +1208,96 @@ retry: set_page_writeback(page); if (page_offset(page) >= mapping->host->i_size) { - done = 1; + done = true; unlock_page(page); end_page_writeback(page); break; } - /* - * BB can we get rid of this? pages are held by pvec - */ - page_cache_get(page); + wdata->pages[i] = page; + next = page->index + 1; + ++nr_pages; + } - len = min(mapping->host->i_size - page_offset(page), - (loff_t)PAGE_CACHE_SIZE); + /* reset index to refind any pages skipped */ + if (nr_pages == 0) + index = wdata->pages[0]->index + 1; - /* reserve iov[0] for the smb header */ - n_iov++; - iov[n_iov].iov_base = kmap(page); - iov[n_iov].iov_len = len; - bytes_to_write += len; + /* put any pages we aren't going to use */ + for (i = nr_pages; i < found_pages; i++) { + page_cache_release(wdata->pages[i]); + wdata->pages[i] = NULL; + } - if (first < 0) { - first = i; - offset = page_offset(page); - } - next = page->index + 1; - if (bytes_to_write + PAGE_CACHE_SIZE > cifs_sb->wsize) - break; + /* nothing to write? */ + if (nr_pages == 0) { + kref_put(&wdata->refcount, cifs_writedata_release); + continue; } - if (n_iov) { -retry_write: - open_file = find_writable_file(CIFS_I(mapping->host), - false); - if (!open_file) { - cERROR(1, "No writable handles for inode"); - rc = -EBADF; - } else { - rc = CIFSSMBWrite2(xid, tcon, open_file->netfid, - bytes_to_write, offset, - &bytes_written, iov, n_iov, - 0); - cifsFileInfo_put(open_file); - } - cFYI(1, "Write2 rc=%d, wrote=%u", rc, bytes_written); + wdata->sync_mode = wbc->sync_mode; + wdata->nr_pages = nr_pages; + wdata->offset = page_offset(wdata->pages[0]); - /* - * For now, treat a short write as if nothing got - * written. A zero length write however indicates - * ENOSPC or EFBIG. We have no way to know which - * though, so call it ENOSPC for now. EFBIG would - * get translated to AS_EIO anyway. - * - * FIXME: make it take into account the data that did - * get written - */ - if (rc == 0) { - if (bytes_written == 0) - rc = -ENOSPC; - else if (bytes_written < bytes_to_write) - rc = -EAGAIN; + do { + if (wdata->cfile != NULL) + cifsFileInfo_put(wdata->cfile); + wdata->cfile = find_writable_file(CIFS_I(mapping->host), + false); + if (!wdata->cfile) { + cERROR(1, "No writable handles for inode"); + rc = -EBADF; + break; } + rc = cifs_async_writev(wdata); + } while (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN); - /* retry on data-integrity flush */ - if (wbc->sync_mode == WB_SYNC_ALL && rc == -EAGAIN) - goto retry_write; + for (i = 0; i < nr_pages; ++i) + unlock_page(wdata->pages[i]); - /* fix the stats and EOF */ - if (bytes_written > 0) { - cifs_stats_bytes_written(tcon, bytes_written); - cifs_update_eof(cifsi, offset, bytes_written); - } - - for (i = 0; i < n_iov; i++) { - page = pvec.pages[first + i]; - /* on retryable write error, redirty page */ + /* send failure -- clean up the mess */ + if (rc != 0) { + for (i = 0; i < nr_pages; ++i) { if (rc == -EAGAIN) - redirty_page_for_writepage(wbc, page); - else if (rc != 0) - SetPageError(page); - kunmap(page); - unlock_page(page); - end_page_writeback(page); - page_cache_release(page); + redirty_page_for_writepage(wbc, + wdata->pages[i]); + else + SetPageError(wdata->pages[i]); + end_page_writeback(wdata->pages[i]); + page_cache_release(wdata->pages[i]); } - if (rc != -EAGAIN) mapping_set_error(mapping, rc); - else - rc = 0; + } + kref_put(&wdata->refcount, cifs_writedata_release); - if ((wbc->nr_to_write -= n_iov) <= 0) - done = 1; - index = next; - } else - /* Need to re-find the pages we skipped */ - index = pvec.pages[0]->index + 1; + wbc->nr_to_write -= nr_pages; + if (wbc->nr_to_write <= 0) + done = true; - pagevec_release(&pvec); + index = next; } + if (!scanned && !done) { /* * We hit the last page and there is more work to be done: wrap * back to the start of the file */ - scanned = 1; + scanned = true; index = 0; goto retry; } + if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0)) mapping->writeback_index = index; - FreeXid(xid); - kfree(iov); return rc; } -static int cifs_writepage(struct page *page, struct writeback_control *wbc) +static int +cifs_writepage_locked(struct page *page, struct writeback_control *wbc) { - int rc = -EFAULT; + int rc; int xid; xid = GetXid(); @@ -1442,21 +1317,43 @@ static int cifs_writepage(struct page *page, struct writeback_control *wbc) * to fail to update with the state of the page correctly. */ set_page_writeback(page); +retry_write: rc = cifs_partialpagewrite(page, 0, PAGE_CACHE_SIZE); - SetPageUptodate(page); /* BB add check for error and Clearuptodate? */ - unlock_page(page); + if (rc == -EAGAIN && wbc->sync_mode == WB_SYNC_ALL) + goto retry_write; + else if (rc == -EAGAIN) + redirty_page_for_writepage(wbc, page); + else if (rc != 0) + SetPageError(page); + else + SetPageUptodate(page); end_page_writeback(page); page_cache_release(page); FreeXid(xid); return rc; } +static int cifs_writepage(struct page *page, struct writeback_control *wbc) +{ + int rc = cifs_writepage_locked(page, wbc); + unlock_page(page); + return rc; +} + static int cifs_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata) { int rc; struct inode *inode = mapping->host; + struct cifsFileInfo *cfile = file->private_data; + struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb); + __u32 pid; + + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = cfile->pid; + else + pid = current->tgid; cFYI(1, "write_end for page %p from pos %lld with %d bytes", page, pos, copied); @@ -1480,8 +1377,7 @@ static int cifs_write_end(struct file *file, struct address_space *mapping, /* BB check if anything else missing out of ppw such as updating last write time */ page_data = kmap(page); - rc = cifs_write(file->private_data, page_data + offset, - copied, &pos); + rc = cifs_write(cfile, pid, page_data + offset, copied, &pos); /* if (rc < 0) should we set writebehind rc? */ kunmap(page); @@ -1509,7 +1405,7 @@ int cifs_strict_fsync(struct file *file, int datasync) { int xid; int rc = 0; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct cifsFileInfo *smbfile = file->private_data; struct inode *inode = file->f_path.dentry->d_inode; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); @@ -1519,8 +1415,13 @@ int cifs_strict_fsync(struct file *file, int datasync) cFYI(1, "Sync file - name: %s datasync: 0x%x", file->f_path.dentry->d_name.name, datasync); - if (!CIFS_I(inode)->clientCanCacheRead) - cifs_invalidate_mapping(inode); + if (!CIFS_I(inode)->clientCanCacheRead) { + rc = cifs_invalidate_mapping(inode); + if (rc) { + cFYI(1, "rc: %d during invalidate phase", rc); + rc = 0; /* don't care about it in fsync */ + } + } tcon = tlink_tcon(smbfile->tlink); if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) @@ -1534,7 +1435,7 @@ int cifs_fsync(struct file *file, int datasync) { int xid; int rc = 0; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct cifsFileInfo *smbfile = file->private_data; struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); @@ -1625,9 +1526,11 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, struct iov_iter it; struct inode *inode; struct cifsFileInfo *open_file; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifs_sb_info *cifs_sb; + struct cifs_io_parms io_parms; int xid, rc; + __u32 pid; len = iov_length(iov, nr_segs); if (!len) @@ -1659,6 +1562,12 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, xid = GetXid(); open_file = file->private_data; + + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = open_file->pid; + else + pid = current->tgid; + pTcon = tlink_tcon(open_file->tlink); inode = file->f_path.dentry->d_inode; @@ -1685,9 +1594,13 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, if (rc != 0) break; } - rc = CIFSSMBWrite2(xid, pTcon, open_file->netfid, - cur_len, *poffset, &written, - to_send, npages, 0); + io_parms.netfid = open_file->netfid; + io_parms.pid = pid; + io_parms.tcon = pTcon; + io_parms.offset = *poffset; + io_parms.length = cur_len; + rc = CIFSSMBWrite2(xid, &io_parms, &written, to_send, + npages, 0); } while (rc == -EAGAIN); for (i = 0; i < npages; i++) @@ -1726,7 +1639,7 @@ cifs_iovec_write(struct file *file, const struct iovec *iov, return total_written; } -static ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov, +ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { ssize_t written; @@ -1780,10 +1693,12 @@ cifs_iovec_read(struct file *file, const struct iovec *iov, size_t len, cur_len; int iov_offset = 0; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifsFileInfo *open_file; struct smb_com_read_rsp *pSMBr; + struct cifs_io_parms io_parms; char *read_data; + __u32 pid; if (!nr_segs) return 0; @@ -1798,6 +1713,11 @@ cifs_iovec_read(struct file *file, const struct iovec *iov, open_file = file->private_data; pTcon = tlink_tcon(open_file->tlink); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = open_file->pid; + else + pid = current->tgid; + if ((file->f_flags & O_ACCMODE) == O_WRONLY) cFYI(1, "attempting read on write only file instance"); @@ -1813,8 +1733,12 @@ cifs_iovec_read(struct file *file, const struct iovec *iov, if (rc != 0) break; } - rc = CIFSSMBRead(xid, pTcon, open_file->netfid, - cur_len, *poffset, &bytes_read, + io_parms.netfid = open_file->netfid; + io_parms.pid = pid; + io_parms.tcon = pTcon; + io_parms.offset = *poffset; + io_parms.length = len; + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, &read_data, &buf_type); pSMBr = (struct smb_com_read_rsp *)read_data; if (read_data) { @@ -1849,17 +1773,7 @@ cifs_iovec_read(struct file *file, const struct iovec *iov, return total_read; } -ssize_t cifs_user_read(struct file *file, char __user *read_data, - size_t read_size, loff_t *poffset) -{ - struct iovec iov; - iov.iov_base = read_data; - iov.iov_len = read_size; - - return cifs_iovec_read(file, &iov, 1, poffset); -} - -static ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov, +ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t pos) { ssize_t read; @@ -1901,11 +1815,13 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, unsigned int total_read; unsigned int current_read_size; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; int xid; char *current_offset; struct cifsFileInfo *open_file; + struct cifs_io_parms io_parms; int buf_type = CIFS_NO_BUFFER; + __u32 pid; xid = GetXid(); cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); @@ -1918,6 +1834,11 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, open_file = file->private_data; pTcon = tlink_tcon(open_file->tlink); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = open_file->pid; + else + pid = current->tgid; + if ((file->f_flags & O_ACCMODE) == O_WRONLY) cFYI(1, "attempting read on write only file instance"); @@ -1940,11 +1861,13 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, if (rc != 0) break; } - rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - current_read_size, *poffset, - &bytes_read, ¤t_offset, - &buf_type); + io_parms.netfid = open_file->netfid; + io_parms.pid = pid; + io_parms.tcon = pTcon; + io_parms.offset = *poffset; + io_parms.length = current_read_size; + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, + ¤t_offset, &buf_type); } if (rc || (bytes_read == 0)) { if (total_read) { @@ -1987,8 +1910,11 @@ int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma) xid = GetXid(); - if (!CIFS_I(inode)->clientCanCacheRead) - cifs_invalidate_mapping(inode); + if (!CIFS_I(inode)->clientCanCacheRead) { + rc = cifs_invalidate_mapping(inode); + if (rc) + return rc; + } rc = generic_file_mmap(file, vma); if (rc == 0) @@ -2072,13 +1998,15 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, loff_t offset; struct page *page; struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; unsigned int bytes_read = 0; unsigned int read_size, i; char *smb_read_data = NULL; struct smb_com_read_rsp *pSMBr; struct cifsFileInfo *open_file; + struct cifs_io_parms io_parms; int buf_type = CIFS_NO_BUFFER; + __u32 pid; xid = GetXid(); if (file->private_data == NULL) { @@ -2100,6 +2028,11 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, goto read_complete; cFYI(DBG2, "rpages: num pages %d", num_pages); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + pid = open_file->pid; + else + pid = current->tgid; + for (i = 0; i < num_pages; ) { unsigned contig_pages; struct page *tmp_page; @@ -2141,12 +2074,13 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, if (rc != 0) break; } - - rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - read_size, offset, - &bytes_read, &smb_read_data, - &buf_type); + io_parms.netfid = open_file->netfid; + io_parms.pid = pid; + io_parms.tcon = pTcon; + io_parms.offset = offset; + io_parms.length = read_size; + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, + &smb_read_data, &buf_type); /* BB more RC checks ? */ if (rc == -EAGAIN) { if (smb_read_data) { @@ -2415,6 +2349,27 @@ static void cifs_invalidate_page(struct page *page, unsigned long offset) cifs_fscache_invalidate_page(page, &cifsi->vfs_inode); } +static int cifs_launder_page(struct page *page) +{ + int rc = 0; + loff_t range_start = page_offset(page); + loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1); + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, + .range_start = range_start, + .range_end = range_end, + }; + + cFYI(1, "Launder page: %p", page); + + if (clear_page_dirty_for_io(page)) + rc = cifs_writepage_locked(page, &wbc); + + cifs_fscache_invalidate_page(page, page->mapping->host); + return rc; +} + void cifs_oplock_break(struct work_struct *work) { struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo, @@ -2486,7 +2441,7 @@ const struct address_space_operations cifs_addr_ops = { .set_page_dirty = __set_page_dirty_nobuffers, .releasepage = cifs_release_page, .invalidatepage = cifs_invalidate_page, - /* .direct_IO = */ + .launder_page = cifs_launder_page, }; /* @@ -2503,5 +2458,5 @@ const struct address_space_operations cifs_addr_ops_smallbuf = { .set_page_dirty = __set_page_dirty_nobuffers, .releasepage = cifs_release_page, .invalidatepage = cifs_invalidate_page, - /* .direct_IO = */ + .launder_page = cifs_launder_page, }; diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index 297a43d..8166966 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -28,32 +28,32 @@ void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) server->fscache = fscache_acquire_cookie(cifs_fscache_netfs.primary_index, &cifs_fscache_server_index_def, server); - cFYI(1, "CIFS: get client cookie (0x%p/0x%p)", server, - server->fscache); + cFYI(1, "%s: (0x%p/0x%p)", __func__, server, + server->fscache); } void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) { - cFYI(1, "CIFS: release client cookie (0x%p/0x%p)", server, - server->fscache); + cFYI(1, "%s: (0x%p/0x%p)", __func__, server, + server->fscache); fscache_relinquish_cookie(server->fscache, 0); server->fscache = NULL; } -void cifs_fscache_get_super_cookie(struct cifsTconInfo *tcon) +void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) { struct TCP_Server_Info *server = tcon->ses->server; tcon->fscache = fscache_acquire_cookie(server->fscache, &cifs_fscache_super_index_def, tcon); - cFYI(1, "CIFS: get superblock cookie (0x%p/0x%p)", - server->fscache, tcon->fscache); + cFYI(1, "%s: (0x%p/0x%p)", __func__, server->fscache, + tcon->fscache); } -void cifs_fscache_release_super_cookie(struct cifsTconInfo *tcon) +void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) { - cFYI(1, "CIFS: releasing superblock cookie (0x%p)", tcon->fscache); + cFYI(1, "%s: (0x%p)", __func__, tcon->fscache); fscache_relinquish_cookie(tcon->fscache, 0); tcon->fscache = NULL; } @@ -62,7 +62,7 @@ static void cifs_fscache_enable_inode_cookie(struct inode *inode) { struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); if (cifsi->fscache) return; @@ -70,8 +70,8 @@ static void cifs_fscache_enable_inode_cookie(struct inode *inode) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) { cifsi->fscache = fscache_acquire_cookie(tcon->fscache, &cifs_fscache_inode_object_def, cifsi); - cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache, - cifsi->fscache); + cFYI(1, "%s: got FH cookie (0x%p/0x%p)", __func__, + tcon->fscache, cifsi->fscache); } } @@ -80,8 +80,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode) struct cifsInodeInfo *cifsi = CIFS_I(inode); if (cifsi->fscache) { - cFYI(1, "CIFS releasing inode cookie (0x%p)", - cifsi->fscache); + cFYI(1, "%s: (0x%p)", __func__, cifsi->fscache); fscache_relinquish_cookie(cifsi->fscache, 0); cifsi->fscache = NULL; } @@ -92,8 +91,7 @@ static void cifs_fscache_disable_inode_cookie(struct inode *inode) struct cifsInodeInfo *cifsi = CIFS_I(inode); if (cifsi->fscache) { - cFYI(1, "CIFS disabling inode cookie (0x%p)", - cifsi->fscache); + cFYI(1, "%s: (0x%p)", __func__, cifsi->fscache); fscache_relinquish_cookie(cifsi->fscache, 1); cifsi->fscache = NULL; } @@ -121,8 +119,8 @@ void cifs_fscache_reset_inode_cookie(struct inode *inode) cifs_sb_master_tcon(cifs_sb)->fscache, &cifs_fscache_inode_object_def, cifsi); - cFYI(1, "CIFS: new cookie 0x%p oldcookie 0x%p", - cifsi->fscache, old); + cFYI(1, "%s: new cookie 0x%p oldcookie 0x%p", + __func__, cifsi->fscache, old); } } @@ -132,8 +130,8 @@ int cifs_fscache_release_page(struct page *page, gfp_t gfp) struct inode *inode = page->mapping->host; struct cifsInodeInfo *cifsi = CIFS_I(inode); - cFYI(1, "CIFS: fscache release page (0x%p/0x%p)", - page, cifsi->fscache); + cFYI(1, "%s: (0x%p/0x%p)", __func__, page, + cifsi->fscache); if (!fscache_maybe_release_page(cifsi->fscache, page, gfp)) return 0; } @@ -144,8 +142,7 @@ int cifs_fscache_release_page(struct page *page, gfp_t gfp) static void cifs_readpage_from_fscache_complete(struct page *page, void *ctx, int error) { - cFYI(1, "CFS: readpage_from_fscache_complete (0x%p/%d)", - page, error); + cFYI(1, "%s: (0x%p/%d)", __func__, page, error); if (!error) SetPageUptodate(page); unlock_page(page); @@ -158,7 +155,7 @@ int __cifs_readpage_from_fscache(struct inode *inode, struct page *page) { int ret; - cFYI(1, "CIFS: readpage_from_fscache(fsc:%p, p:%p, i:0x%p", + cFYI(1, "%s: (fsc:%p, p:%p, i:0x%p", __func__, CIFS_I(inode)->fscache, page, inode); ret = fscache_read_or_alloc_page(CIFS_I(inode)->fscache, page, cifs_readpage_from_fscache_complete, @@ -167,11 +164,11 @@ int __cifs_readpage_from_fscache(struct inode *inode, struct page *page) switch (ret) { case 0: /* page found in fscache, read submitted */ - cFYI(1, "CIFS: readpage_from_fscache: submitted"); + cFYI(1, "%s: submitted", __func__); return ret; case -ENOBUFS: /* page won't be cached */ case -ENODATA: /* page not in cache */ - cFYI(1, "CIFS: readpage_from_fscache %d", ret); + cFYI(1, "%s: %d", __func__, ret); return 1; default: @@ -190,7 +187,7 @@ int __cifs_readpages_from_fscache(struct inode *inode, { int ret; - cFYI(1, "CIFS: __cifs_readpages_from_fscache (0x%p/%u/0x%p)", + cFYI(1, "%s: (0x%p/%u/0x%p)", __func__, CIFS_I(inode)->fscache, *nr_pages, inode); ret = fscache_read_or_alloc_pages(CIFS_I(inode)->fscache, mapping, pages, nr_pages, @@ -199,12 +196,12 @@ int __cifs_readpages_from_fscache(struct inode *inode, mapping_gfp_mask(mapping)); switch (ret) { case 0: /* read submitted to the cache for all pages */ - cFYI(1, "CIFS: readpages_from_fscache: submitted"); + cFYI(1, "%s: submitted", __func__); return ret; case -ENOBUFS: /* some pages are not cached and can't be */ case -ENODATA: /* some pages are not cached */ - cFYI(1, "CIFS: readpages_from_fscache: no page"); + cFYI(1, "%s: no page", __func__); return 1; default: @@ -218,7 +215,7 @@ void __cifs_readpage_to_fscache(struct inode *inode, struct page *page) { int ret; - cFYI(1, "CIFS: readpage_to_fscache(fsc: %p, p: %p, i: %p", + cFYI(1, "%s: (fsc: %p, p: %p, i: %p)", __func__, CIFS_I(inode)->fscache, page, inode); ret = fscache_write_page(CIFS_I(inode)->fscache, page, GFP_KERNEL); if (ret != 0) @@ -230,7 +227,7 @@ void __cifs_fscache_invalidate_page(struct page *page, struct inode *inode) struct cifsInodeInfo *cifsi = CIFS_I(inode); struct fscache_cookie *cookie = cifsi->fscache; - cFYI(1, "CIFS: fscache invalidatepage (0x%p/0x%p)", page, cookie); + cFYI(1, "%s: (0x%p/0x%p)", __func__, page, cookie); fscache_wait_on_page_write(cookie, page); fscache_uncache_page(cookie, page); } diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h index 31b88ec2..6353932 100644 --- a/fs/cifs/fscache.h +++ b/fs/cifs/fscache.h @@ -40,8 +40,8 @@ extern void cifs_fscache_unregister(void); */ extern void cifs_fscache_get_client_cookie(struct TCP_Server_Info *); extern void cifs_fscache_release_client_cookie(struct TCP_Server_Info *); -extern void cifs_fscache_get_super_cookie(struct cifsTconInfo *); -extern void cifs_fscache_release_super_cookie(struct cifsTconInfo *); +extern void cifs_fscache_get_super_cookie(struct cifs_tcon *); +extern void cifs_fscache_release_super_cookie(struct cifs_tcon *); extern void cifs_fscache_release_inode_cookie(struct inode *); extern void cifs_fscache_set_inode_cookie(struct inode *, struct file *); @@ -99,9 +99,9 @@ static inline void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) {} static inline void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) {} -static inline void cifs_fscache_get_super_cookie(struct cifsTconInfo *tcon) {} +static inline void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) {} static inline void -cifs_fscache_release_super_cookie(struct cifsTconInfo *tcon) {} +cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) {} static inline void cifs_fscache_release_inode_cookie(struct inode *inode) {} static inline void cifs_fscache_set_inode_cookie(struct inode *inode, diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 8852470..9b018c8 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -295,7 +295,7 @@ int cifs_get_file_info_unix(struct file *filp) struct inode *inode = filp->f_path.dentry->d_inode; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifsFileInfo *cfile = filp->private_data; - struct cifsTconInfo *tcon = tlink_tcon(cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); xid = GetXid(); rc = CIFSSMBUnixQFileInfo(xid, tcon, cfile->netfid, &find_data); @@ -318,7 +318,7 @@ int cifs_get_inode_info_unix(struct inode **pinode, int rc; FILE_UNIX_BASIC_INFO find_data; struct cifs_fattr fattr; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct tcon_link *tlink; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); @@ -373,7 +373,8 @@ cifs_sfu_type(struct cifs_fattr *fattr, const unsigned char *path, int oplock = 0; __u16 netfid; struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; + struct cifs_io_parms io_parms; char buf[24]; unsigned int bytes_read; char *pbuf; @@ -405,9 +406,13 @@ cifs_sfu_type(struct cifs_fattr *fattr, const unsigned char *path, if (rc == 0) { int buf_type = CIFS_NO_BUFFER; /* Read header */ - rc = CIFSSMBRead(xid, tcon, netfid, - 24 /* length */, 0 /* offset */, - &bytes_read, &pbuf, &buf_type); + io_parms.netfid = netfid; + io_parms.pid = current->tgid; + io_parms.tcon = tcon; + io_parms.offset = 0; + io_parms.length = 24; + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, &pbuf, + &buf_type); if ((rc == 0) && (bytes_read >= 8)) { if (memcmp("IntxBLK", pbuf, 8) == 0) { cFYI(1, "Block device"); @@ -468,7 +473,7 @@ static int cifs_sfu_mode(struct cifs_fattr *fattr, const unsigned char *path, char ea_value[4]; __u32 mode; struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) @@ -502,7 +507,7 @@ static void cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info, struct cifs_sb_info *cifs_sb, bool adjust_tz) { - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); memset(fattr, 0, sizeof(*fattr)); fattr->cf_cifsattrs = le32_to_cpu(info->Attributes); @@ -553,7 +558,7 @@ int cifs_get_file_info(struct file *filp) struct inode *inode = filp->f_path.dentry->d_inode; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifsFileInfo *cfile = filp->private_data; - struct cifsTconInfo *tcon = tlink_tcon(cfile->tlink); + struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); xid = GetXid(); rc = CIFSSMBQFileInfo(xid, tcon, cfile->netfid, &find_data); @@ -590,7 +595,7 @@ int cifs_get_inode_info(struct inode **pinode, struct super_block *sb, int xid, const __u16 *pfid) { int rc = 0, tmprc; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct tcon_link *tlink; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); char *buf = NULL; @@ -735,10 +740,10 @@ static const struct inode_operations cifs_ipc_inode_ops = { .lookup = cifs_lookup, }; -char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb, - struct cifsTconInfo *tcon) +char *cifs_build_path_to_root(struct smb_vol *vol, struct cifs_sb_info *cifs_sb, + struct cifs_tcon *tcon) { - int pplen = cifs_sb->prepathlen; + int pplen = vol->prepath ? strlen(vol->prepath) : 0; int dfsplen; char *full_path = NULL; @@ -772,7 +777,7 @@ char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb, } } } - strncpy(full_path + dfsplen, cifs_sb->prepath, pplen); + strncpy(full_path + dfsplen, vol->prepath, pplen); full_path[dfsplen + pplen] = 0; /* add trailing null */ return full_path; } @@ -878,25 +883,19 @@ retry_iget5_locked: } /* gets root inode */ -struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino) +struct inode *cifs_root_iget(struct super_block *sb) { int xid; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct inode *inode = NULL; long rc; - char *full_path; - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); - - full_path = cifs_build_path_to_root(cifs_sb, tcon); - if (full_path == NULL) - return ERR_PTR(-ENOMEM); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); xid = GetXid(); if (tcon->unix_ext) - rc = cifs_get_inode_info_unix(&inode, full_path, sb, xid); + rc = cifs_get_inode_info_unix(&inode, "", sb, xid); else - rc = cifs_get_inode_info(&inode, full_path, NULL, sb, - xid, NULL); + rc = cifs_get_inode_info(&inode, "", NULL, sb, xid, NULL); if (!inode) { inode = ERR_PTR(rc); @@ -922,7 +921,6 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino) } out: - kfree(full_path); /* can not call macro FreeXid here since in a void func * TODO: This is no longer true */ @@ -943,7 +941,7 @@ cifs_set_file_info(struct inode *inode, struct iattr *attrs, int xid, struct cifsInodeInfo *cifsInode = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink = NULL; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; FILE_BASIC_INFO info_buf; if (attrs == NULL) @@ -1061,7 +1059,7 @@ cifs_rename_pending_delete(char *full_path, struct dentry *dentry, int xid) struct cifsInodeInfo *cifsInode = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; __u32 dosattr, origattr; FILE_BASIC_INFO *info_buf = NULL; @@ -1179,7 +1177,7 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry) struct super_block *sb = dir->i_sb; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; struct iattr *attrs = NULL; __u32 dosattr = 0, origattr = 0; @@ -1277,7 +1275,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; char *full_path = NULL; struct inode *newinode = NULL; struct cifs_fattr fattr; @@ -1455,7 +1453,7 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; char *full_path = NULL; struct cifsInodeInfo *cifsInode; @@ -1512,7 +1510,7 @@ cifs_do_rename(int xid, struct dentry *from_dentry, const char *fromPath, { struct cifs_sb_info *cifs_sb = CIFS_SB(from_dentry->d_sb); struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; __u16 srcfid; int oplock, rc; @@ -1564,7 +1562,7 @@ int cifs_rename(struct inode *source_dir, struct dentry *source_dentry, char *toName = NULL; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; FILE_UNIX_BASIC_INFO *info_buf_source = NULL; FILE_UNIX_BASIC_INFO *info_buf_target; int xid, rc, tmprc; @@ -1683,71 +1681,70 @@ cifs_inode_needs_reval(struct inode *inode) /* * Zap the cache. Called when invalid_mapping flag is set. */ -void +int cifs_invalidate_mapping(struct inode *inode) { - int rc; + int rc = 0; struct cifsInodeInfo *cifs_i = CIFS_I(inode); cifs_i->invalid_mapping = false; - /* write back any cached data */ if (inode->i_mapping && inode->i_mapping->nrpages != 0) { - rc = filemap_write_and_wait(inode->i_mapping); - mapping_set_error(inode->i_mapping, rc); + rc = invalidate_inode_pages2(inode->i_mapping); + if (rc) { + cERROR(1, "%s: could not invalidate inode %p", __func__, + inode); + cifs_i->invalid_mapping = true; + } } - invalidate_remote_inode(inode); + cifs_fscache_reset_inode_cookie(inode); + return rc; } -int cifs_revalidate_file(struct file *filp) +int cifs_revalidate_file_attr(struct file *filp) { int rc = 0; struct inode *inode = filp->f_path.dentry->d_inode; struct cifsFileInfo *cfile = (struct cifsFileInfo *) filp->private_data; if (!cifs_inode_needs_reval(inode)) - goto check_inval; + return rc; if (tlink_tcon(cfile->tlink)->unix_ext) rc = cifs_get_file_info_unix(filp); else rc = cifs_get_file_info(filp); -check_inval: - if (CIFS_I(inode)->invalid_mapping) - cifs_invalidate_mapping(inode); - return rc; } -/* revalidate a dentry's inode attributes */ -int cifs_revalidate_dentry(struct dentry *dentry) +int cifs_revalidate_dentry_attr(struct dentry *dentry) { int xid; int rc = 0; - char *full_path = NULL; struct inode *inode = dentry->d_inode; struct super_block *sb = dentry->d_sb; + char *full_path = NULL; if (inode == NULL) return -ENOENT; - xid = GetXid(); - if (!cifs_inode_needs_reval(inode)) - goto check_inval; + return rc; + + xid = GetXid(); /* can not safely grab the rename sem here if rename calls revalidate since that would deadlock */ full_path = build_path_from_dentry(dentry); if (full_path == NULL) { rc = -ENOMEM; - goto check_inval; + goto out; } - cFYI(1, "Revalidate: %s inode 0x%p count %d dentry: 0x%p d_time %ld " - "jiffies %ld", full_path, inode, inode->i_count.counter, + cFYI(1, "Update attributes: %s inode 0x%p count %d dentry: 0x%p d_time " + "%ld jiffies %ld", full_path, inode, inode->i_count.counter, dentry, dentry->d_time, jiffies); if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext) @@ -1756,41 +1753,83 @@ int cifs_revalidate_dentry(struct dentry *dentry) rc = cifs_get_inode_info(&inode, full_path, NULL, sb, xid, NULL); -check_inval: - if (CIFS_I(inode)->invalid_mapping) - cifs_invalidate_mapping(inode); - +out: kfree(full_path); FreeXid(xid); return rc; } +int cifs_revalidate_file(struct file *filp) +{ + int rc; + struct inode *inode = filp->f_path.dentry->d_inode; + + rc = cifs_revalidate_file_attr(filp); + if (rc) + return rc; + + if (CIFS_I(inode)->invalid_mapping) + rc = cifs_invalidate_mapping(inode); + return rc; +} + +/* revalidate a dentry's inode attributes */ +int cifs_revalidate_dentry(struct dentry *dentry) +{ + int rc; + struct inode *inode = dentry->d_inode; + + rc = cifs_revalidate_dentry_attr(dentry); + if (rc) + return rc; + + if (CIFS_I(inode)->invalid_mapping) + rc = cifs_invalidate_mapping(inode); + return rc; +} + int cifs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct cifs_sb_info *cifs_sb = CIFS_SB(dentry->d_sb); - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); - int err = cifs_revalidate_dentry(dentry); - - if (!err) { - generic_fillattr(dentry->d_inode, stat); - stat->blksize = CIFS_MAX_MSGSIZE; - stat->ino = CIFS_I(dentry->d_inode)->uniqueid; + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + struct inode *inode = dentry->d_inode; + int rc; - /* - * If on a multiuser mount without unix extensions, and the - * admin hasn't overridden them, set the ownership to the - * fsuid/fsgid of the current process. - */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) && - !tcon->unix_ext) { - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)) - stat->uid = current_fsuid(); - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)) - stat->gid = current_fsgid(); + /* + * We need to be sure that all dirty pages are written and the server + * has actual ctime, mtime and file length. + */ + if (!CIFS_I(inode)->clientCanCacheRead && inode->i_mapping && + inode->i_mapping->nrpages != 0) { + rc = filemap_fdatawait(inode->i_mapping); + if (rc) { + mapping_set_error(inode->i_mapping, rc); + return rc; } } - return err; + + rc = cifs_revalidate_dentry_attr(dentry); + if (rc) + return rc; + + generic_fillattr(inode, stat); + stat->blksize = CIFS_MAX_MSGSIZE; + stat->ino = CIFS_I(inode)->uniqueid; + + /* + * If on a multiuser mount without unix extensions, and the admin hasn't + * overridden them, set the ownership to the fsuid/fsgid of the current + * process. + */ + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) && + !tcon->unix_ext) { + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)) + stat->uid = current_fsuid(); + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)) + stat->gid = current_fsgid(); + } + return rc; } static int cifs_truncate_page(struct address_space *mapping, loff_t from) @@ -1831,7 +1870,8 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, struct cifsInodeInfo *cifsInode = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink = NULL; - struct cifsTconInfo *pTcon = NULL; + struct cifs_tcon *pTcon = NULL; + struct cifs_io_parms io_parms; /* * To avoid spurious oplock breaks from server, in the case of @@ -1853,8 +1893,14 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, cFYI(1, "SetFSize for attrs rc = %d", rc); if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { unsigned int bytes_written; - rc = CIFSSMBWrite(xid, pTcon, nfid, 0, attrs->ia_size, - &bytes_written, NULL, NULL, 1); + + io_parms.netfid = nfid; + io_parms.pid = npid; + io_parms.tcon = pTcon; + io_parms.offset = 0; + io_parms.length = attrs->ia_size; + rc = CIFSSMBWrite(xid, &io_parms, &bytes_written, + NULL, NULL, 1); cFYI(1, "Wrt seteof rc %d", rc); } } else @@ -1889,10 +1935,15 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc == 0) { unsigned int bytes_written; - rc = CIFSSMBWrite(xid, pTcon, netfid, 0, - attrs->ia_size, - &bytes_written, NULL, - NULL, 1); + + io_parms.netfid = netfid; + io_parms.pid = current->tgid; + io_parms.tcon = pTcon; + io_parms.offset = 0; + io_parms.length = attrs->ia_size; + rc = CIFSSMBWrite(xid, &io_parms, + &bytes_written, + NULL, NULL, 1); cFYI(1, "wrt seteof rc %d", rc); CIFSSMBClose(xid, pTcon, netfid); } @@ -1920,7 +1971,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) struct cifsInodeInfo *cifsInode = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifs_unix_set_info_args *args = NULL; struct cifsFileInfo *open_file; @@ -2206,7 +2257,7 @@ cifs_setattr(struct dentry *direntry, struct iattr *attrs) { struct inode *inode = direntry->d_inode; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); - struct cifsTconInfo *pTcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *pTcon = cifs_sb_master_tcon(cifs_sb); if (pTcon->unix_ext) return cifs_setattr_unix(direntry, attrs); diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c index 0c98672..4221b5e 100644 --- a/fs/cifs/ioctl.c +++ b/fs/cifs/ioctl.c @@ -38,7 +38,7 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg) struct cifs_sb_info *cifs_sb; #ifdef CONFIG_CIFS_POSIX struct cifsFileInfo *pSMBFile = filep->private_data; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; __u64 ExtAttrBits = 0; __u64 ExtAttrMask = 0; __u64 caps; diff --git a/fs/cifs/link.c b/fs/cifs/link.c index ce417a9..556b1a0 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -175,7 +175,7 @@ CIFSFormatMFSymlink(u8 *buf, unsigned int buf_len, const char *link_str) } static int -CIFSCreateMFSymLink(const int xid, struct cifsTconInfo *tcon, +CIFSCreateMFSymLink(const int xid, struct cifs_tcon *tcon, const char *fromName, const char *toName, const struct nls_table *nls_codepage, int remap) { @@ -184,6 +184,7 @@ CIFSCreateMFSymLink(const int xid, struct cifsTconInfo *tcon, __u16 netfid = 0; u8 *buf; unsigned int bytes_written = 0; + struct cifs_io_parms io_parms; buf = kmalloc(CIFS_MF_SYMLINK_FILE_SIZE, GFP_KERNEL); if (!buf) @@ -203,10 +204,13 @@ CIFSCreateMFSymLink(const int xid, struct cifsTconInfo *tcon, return rc; } - rc = CIFSSMBWrite(xid, tcon, netfid, - CIFS_MF_SYMLINK_FILE_SIZE /* length */, - 0 /* offset */, - &bytes_written, buf, NULL, 0); + io_parms.netfid = netfid; + io_parms.pid = current->tgid; + io_parms.tcon = tcon; + io_parms.offset = 0; + io_parms.length = CIFS_MF_SYMLINK_FILE_SIZE; + + rc = CIFSSMBWrite(xid, &io_parms, &bytes_written, buf, NULL, 0); CIFSSMBClose(xid, tcon, netfid); kfree(buf); if (rc != 0) @@ -219,7 +223,7 @@ CIFSCreateMFSymLink(const int xid, struct cifsTconInfo *tcon, } static int -CIFSQueryMFSymLink(const int xid, struct cifsTconInfo *tcon, +CIFSQueryMFSymLink(const int xid, struct cifs_tcon *tcon, const unsigned char *searchName, char **symlinkinfo, const struct nls_table *nls_codepage, int remap) { @@ -231,6 +235,7 @@ CIFSQueryMFSymLink(const int xid, struct cifsTconInfo *tcon, unsigned int bytes_read = 0; int buf_type = CIFS_NO_BUFFER; unsigned int link_len = 0; + struct cifs_io_parms io_parms; FILE_ALL_INFO file_info; rc = CIFSSMBOpen(xid, tcon, searchName, FILE_OPEN, GENERIC_READ, @@ -249,11 +254,13 @@ CIFSQueryMFSymLink(const int xid, struct cifsTconInfo *tcon, if (!buf) return -ENOMEM; pbuf = buf; + io_parms.netfid = netfid; + io_parms.pid = current->tgid; + io_parms.tcon = tcon; + io_parms.offset = 0; + io_parms.length = CIFS_MF_SYMLINK_FILE_SIZE; - rc = CIFSSMBRead(xid, tcon, netfid, - CIFS_MF_SYMLINK_FILE_SIZE /* length */, - 0 /* offset */, - &bytes_read, &pbuf, &buf_type); + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, &pbuf, &buf_type); CIFSSMBClose(xid, tcon, netfid); if (rc != 0) { kfree(buf); @@ -291,7 +298,8 @@ CIFSCheckMFSymlink(struct cifs_fattr *fattr, int oplock = 0; __u16 netfid = 0; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; + struct cifs_io_parms io_parms; u8 *buf; char *pbuf; unsigned int bytes_read = 0; @@ -328,11 +336,13 @@ CIFSCheckMFSymlink(struct cifs_fattr *fattr, goto out; } pbuf = buf; + io_parms.netfid = netfid; + io_parms.pid = current->tgid; + io_parms.tcon = pTcon; + io_parms.offset = 0; + io_parms.length = CIFS_MF_SYMLINK_FILE_SIZE; - rc = CIFSSMBRead(xid, pTcon, netfid, - CIFS_MF_SYMLINK_FILE_SIZE /* length */, - 0 /* offset */, - &bytes_read, &pbuf, &buf_type); + rc = CIFSSMBRead(xid, &io_parms, &bytes_read, &pbuf, &buf_type); CIFSSMBClose(xid, pTcon, netfid); if (rc != 0) { kfree(buf); @@ -370,7 +380,7 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode, char *toName = NULL; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifsInodeInfo *cifsInode; tlink = cifs_sb_tlink(cifs_sb); @@ -445,7 +455,7 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd) char *target_path = NULL; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink = NULL; - struct cifsTconInfo *tcon; + struct cifs_tcon *tcon; xid = GetXid(); @@ -518,7 +528,7 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) int xid; struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; char *full_path = NULL; struct inode *newinode = NULL; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 0c684ae..03a1f491 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -67,12 +67,12 @@ _FreeXid(unsigned int xid) spin_unlock(&GlobalMid_Lock); } -struct cifsSesInfo * +struct cifs_ses * sesInfoAlloc(void) { - struct cifsSesInfo *ret_buf; + struct cifs_ses *ret_buf; - ret_buf = kzalloc(sizeof(struct cifsSesInfo), GFP_KERNEL); + ret_buf = kzalloc(sizeof(struct cifs_ses), GFP_KERNEL); if (ret_buf) { atomic_inc(&sesInfoAllocCount); ret_buf->status = CifsNew; @@ -85,7 +85,7 @@ sesInfoAlloc(void) } void -sesInfoFree(struct cifsSesInfo *buf_to_free) +sesInfoFree(struct cifs_ses *buf_to_free) { if (buf_to_free == NULL) { cFYI(1, "Null buffer passed to sesInfoFree"); @@ -105,11 +105,11 @@ sesInfoFree(struct cifsSesInfo *buf_to_free) kfree(buf_to_free); } -struct cifsTconInfo * +struct cifs_tcon * tconInfoAlloc(void) { - struct cifsTconInfo *ret_buf; - ret_buf = kzalloc(sizeof(struct cifsTconInfo), GFP_KERNEL); + struct cifs_tcon *ret_buf; + ret_buf = kzalloc(sizeof(struct cifs_tcon), GFP_KERNEL); if (ret_buf) { atomic_inc(&tconInfoAllocCount); ret_buf->tidStatus = CifsNew; @@ -124,7 +124,7 @@ tconInfoAlloc(void) } void -tconInfoFree(struct cifsTconInfo *buf_to_free) +tconInfoFree(struct cifs_tcon *buf_to_free) { if (buf_to_free == NULL) { cFYI(1, "Null buffer passed to tconInfoFree"); @@ -295,21 +295,19 @@ __u16 GetNextMid(struct TCP_Server_Info *server) case it is responsbility of caller to set the mid */ void header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , - const struct cifsTconInfo *treeCon, int word_count + const struct cifs_tcon *treeCon, int word_count /* length of fixed section (word count) in two byte units */) { struct list_head *temp_item; - struct cifsSesInfo *ses; + struct cifs_ses *ses; char *temp = (char *) buffer; memset(temp, 0, 256); /* bigger than MAX_CIFS_HDR_SIZE */ - buffer->smb_buf_length = + buffer->smb_buf_length = cpu_to_be32( (2 * word_count) + sizeof(struct smb_hdr) - 4 /* RFC 1001 length field does not count */ + - 2 /* for bcc field itself */ ; - /* Note that this is the only network field that has to be converted - to big endian and it is done just before we send it */ + 2 /* for bcc field itself */) ; buffer->Protocol[0] = 0xFF; buffer->Protocol[1] = 'S'; @@ -361,7 +359,7 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , "did not match tcon uid"); spin_lock(&cifs_tcp_ses_lock); list_for_each(temp_item, &treeCon->ses->server->smb_ses_list) { - ses = list_entry(temp_item, struct cifsSesInfo, smb_ses_list); + ses = list_entry(temp_item, struct cifs_ses, smb_ses_list); if (ses->linux_uid == current_fsuid()) { if (ses->server == treeCon->ses->server) { cFYI(1, "found matching uid substitute right smb_uid"); @@ -382,7 +380,7 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , if (treeCon->nocase) buffer->Flags |= SMBFLG_CASELESS; if ((treeCon->ses) && (treeCon->ses->server)) - if (treeCon->ses->server->secMode & + if (treeCon->ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) buffer->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; } @@ -424,7 +422,7 @@ check_smb_hdr(struct smb_hdr *smb, __u16 mid) int checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length) { - __u32 len = smb->smb_buf_length; + __u32 len = be32_to_cpu(smb->smb_buf_length); __u32 clc_len; /* calculated length */ cFYI(0, "checkSMB Length: 0x%x, smb_buf_length: 0x%x", length, len); @@ -464,7 +462,7 @@ checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length) if (check_smb_hdr(smb, mid)) return 1; - clc_len = smbCalcSize_LE(smb); + clc_len = smbCalcSize(smb); if (4 + len != length) { cERROR(1, "Length read does not match RFC1001 length %d", @@ -509,8 +507,8 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) { struct smb_com_lock_req *pSMB = (struct smb_com_lock_req *)buf; struct list_head *tmp, *tmp1, *tmp2; - struct cifsSesInfo *ses; - struct cifsTconInfo *tcon; + struct cifs_ses *ses; + struct cifs_tcon *tcon; struct cifsInodeInfo *pCifsInode; struct cifsFileInfo *netfile; @@ -521,7 +519,7 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) (struct smb_com_transaction_change_notify_rsp *)buf; struct file_notify_information *pnotify; __u32 data_offset = 0; - if (get_bcc_le(buf) > sizeof(struct file_notify_information)) { + if (get_bcc(buf) > sizeof(struct file_notify_information)) { data_offset = le32_to_cpu(pSMBr->DataOffset); pnotify = (struct file_notify_information *) @@ -568,9 +566,9 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) /* look up tcon based on tid & uid */ spin_lock(&cifs_tcp_ses_lock); list_for_each(tmp, &srv->smb_ses_list) { - ses = list_entry(tmp, struct cifsSesInfo, smb_ses_list); + ses = list_entry(tmp, struct cifs_ses, smb_ses_list); list_for_each(tmp1, &ses->tcon_list) { - tcon = list_entry(tmp1, struct cifsTconInfo, tcon_list); + tcon = list_entry(tmp1, struct cifs_tcon, tcon_list); if (tcon->tid != buf->Tid) continue; diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index 79f641e..73e47e8 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -836,7 +836,7 @@ ntstatus_to_dos(__u32 ntstatus, __u8 *eclass, __u16 *ecode) } int -map_smb_to_linux_error(struct smb_hdr *smb, int logErr) +map_smb_to_linux_error(struct smb_hdr *smb, bool logErr) { unsigned int i; int rc = -EIO; /* if transport error smb error may not be set */ @@ -919,13 +919,6 @@ smbCalcSize(struct smb_hdr *ptr) 2 /* size of the bcc field */ + get_bcc(ptr)); } -unsigned int -smbCalcSize_LE(struct smb_hdr *ptr) -{ - return (sizeof(struct smb_hdr) + (2 * ptr->WordCount) + - 2 /* size of the bcc field */ + get_bcc_le(ptr)); -} - /* The following are taken from fs/ntfs/util.c */ #define NTFS_TIME_OFFSET ((u64)(369*365 + 89) * 24 * 3600 * 10000000) diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index f8e4cd2..6751e74 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -195,7 +195,7 @@ int get_symlink_reparse_path(char *full_path, struct cifs_sb_info *cifs_sb, int len; int oplock = 0; int rc; - struct cifsTconInfo *ptcon = cifs_sb_tcon(cifs_sb); + struct cifs_tcon *ptcon = cifs_sb_tcon(cifs_sb); char *tmpbuffer; rc = CIFSSMBOpen(xid, ptcon, full_path, FILE_OPEN, GENERIC_READ, @@ -223,7 +223,7 @@ static int initiate_cifs_search(const int xid, struct file *file) struct cifsFileInfo *cifsFile; struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); struct tcon_link *tlink = NULL; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; if (file->private_data == NULL) { tlink = cifs_sb_tlink(cifs_sb); @@ -496,7 +496,7 @@ static int cifs_save_resume_key(const char *current_entry, assume that they are located in the findfirst return buffer.*/ /* We start counting in the buffer with entry 2 and increment for every entry (do not increment for . or .. entry) */ -static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, +static int find_cifs_entry(const int xid, struct cifs_tcon *pTcon, struct file *file, char **ppCurrentEntry, int *num_to_ret) { int rc = 0; @@ -764,7 +764,7 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) { int rc = 0; int xid, i; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct cifsFileInfo *cifsFile = NULL; char *current_entry; int num_to_fill = 0; diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 645114a..3892ab8 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -37,13 +37,13 @@ * the socket has been reestablished (so we know whether to use vc 0). * Called while holding the cifs_tcp_ses_lock, so do not block */ -static bool is_first_ses_reconnect(struct cifsSesInfo *ses) +static bool is_first_ses_reconnect(struct cifs_ses *ses) { struct list_head *tmp; - struct cifsSesInfo *tmp_ses; + struct cifs_ses *tmp_ses; list_for_each(tmp, &ses->server->smb_ses_list) { - tmp_ses = list_entry(tmp, struct cifsSesInfo, + tmp_ses = list_entry(tmp, struct cifs_ses, smb_ses_list); if (tmp_ses->need_reconnect == false) return false; @@ -61,11 +61,11 @@ static bool is_first_ses_reconnect(struct cifsSesInfo *ses) * any vc but zero (some servers reset the connection on vcnum zero) * */ -static __le16 get_next_vcnum(struct cifsSesInfo *ses) +static __le16 get_next_vcnum(struct cifs_ses *ses) { __u16 vcnum = 0; struct list_head *tmp; - struct cifsSesInfo *tmp_ses; + struct cifs_ses *tmp_ses; __u16 max_vcs = ses->server->max_vcs; __u16 i; int free_vc_found = 0; @@ -87,7 +87,7 @@ static __le16 get_next_vcnum(struct cifsSesInfo *ses) free_vc_found = 1; list_for_each(tmp, &ses->server->smb_ses_list) { - tmp_ses = list_entry(tmp, struct cifsSesInfo, + tmp_ses = list_entry(tmp, struct cifs_ses, smb_ses_list); if (tmp_ses->vcnum == i) { free_vc_found = 0; @@ -114,7 +114,7 @@ get_vc_num_exit: return cpu_to_le16(vcnum); } -static __u32 cifs_ssetup_hdr(struct cifsSesInfo *ses, SESSION_SETUP_ANDX *pSMB) +static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB) { __u32 capabilities = 0; @@ -136,7 +136,7 @@ static __u32 cifs_ssetup_hdr(struct cifsSesInfo *ses, SESSION_SETUP_ANDX *pSMB) capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | CAP_LARGE_WRITE_X | CAP_LARGE_READ_X; - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) pSMB->req.hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; @@ -181,7 +181,7 @@ unicode_oslm_strings(char **pbcc_area, const struct nls_table *nls_cp) *pbcc_area = bcc_ptr; } -static void unicode_domain_string(char **pbcc_area, struct cifsSesInfo *ses, +static void unicode_domain_string(char **pbcc_area, struct cifs_ses *ses, const struct nls_table *nls_cp) { char *bcc_ptr = *pbcc_area; @@ -204,7 +204,7 @@ static void unicode_domain_string(char **pbcc_area, struct cifsSesInfo *ses, } -static void unicode_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, +static void unicode_ssetup_strings(char **pbcc_area, struct cifs_ses *ses, const struct nls_table *nls_cp) { char *bcc_ptr = *pbcc_area; @@ -236,7 +236,7 @@ static void unicode_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, *pbcc_area = bcc_ptr; } -static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, +static void ascii_ssetup_strings(char **pbcc_area, struct cifs_ses *ses, const struct nls_table *nls_cp) { char *bcc_ptr = *pbcc_area; @@ -276,7 +276,7 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, } static void -decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifsSesInfo *ses, +decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifs_ses *ses, const struct nls_table *nls_cp) { int len; @@ -310,7 +310,7 @@ decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifsSesInfo *ses, } static int decode_ascii_ssetup(char **pbcc_area, __u16 bleft, - struct cifsSesInfo *ses, + struct cifs_ses *ses, const struct nls_table *nls_cp) { int rc = 0; @@ -364,7 +364,7 @@ static int decode_ascii_ssetup(char **pbcc_area, __u16 bleft, } static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, - struct cifsSesInfo *ses) + struct cifs_ses *ses) { unsigned int tioffset; /* challenge message target info area */ unsigned int tilen; /* challenge message target info area length */ @@ -411,7 +411,7 @@ static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, /* We do not malloc the blob, it is passed in pbuffer, because it is fixed size, and small, making this approach cleaner */ static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, - struct cifsSesInfo *ses) + struct cifs_ses *ses) { NEGOTIATE_MESSAGE *sec_blob = (NEGOTIATE_MESSAGE *)pbuffer; __u32 flags; @@ -424,7 +424,7 @@ static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, flags = NTLMSSP_NEGOTIATE_56 | NTLMSSP_REQUEST_TARGET | NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_EXTENDED_SEC; - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { flags |= NTLMSSP_NEGOTIATE_SIGN; if (!ses->server->session_estab) @@ -449,7 +449,7 @@ static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, This function returns the length of the data in the blob */ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, u16 *buflen, - struct cifsSesInfo *ses, + struct cifs_ses *ses, const struct nls_table *nls_cp) { int rc; @@ -464,10 +464,10 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, NTLMSSP_REQUEST_TARGET | NTLMSSP_NEGOTIATE_TARGET_INFO | NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_EXTENDED_SEC; - if (ses->server->secMode & + if (ses->server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) flags |= NTLMSSP_NEGOTIATE_SIGN; - if (ses->server->secMode & SECMODE_SIGN_REQUIRED) + if (ses->server->sec_mode & SECMODE_SIGN_REQUIRED) flags |= NTLMSSP_NEGOTIATE_ALWAYS_SIGN; tmp = pbuffer + sizeof(AUTHENTICATE_MESSAGE); @@ -551,7 +551,7 @@ setup_ntlmv2_ret: } int -CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, +CIFS_SessSetup(unsigned int xid, struct cifs_ses *ses, const struct nls_table *nls_cp) { int rc = 0; @@ -621,7 +621,7 @@ ssetup_ntlmssp_authenticate: and rest of bcc area. This allows us to avoid a large buffer 17K allocation */ iov[0].iov_base = (char *)pSMB; - iov[0].iov_len = smb_buf->smb_buf_length + 4; + iov[0].iov_len = be32_to_cpu(smb_buf->smb_buf_length) + 4; /* setting this here allows the code at the end of the function to free the request buffer if there's an error */ @@ -656,8 +656,8 @@ ssetup_ntlmssp_authenticate: * to use challenge/response method (i.e. Password bit is 1). */ - calc_lanman_hash(ses->password, ses->server->cryptkey, - ses->server->secMode & SECMODE_PW_ENCRYPT ? + rc = calc_lanman_hash(ses->password, ses->server->cryptkey, + ses->server->sec_mode & SECMODE_PW_ENCRYPT ? true : false, lnm_session_key); ses->flags |= CIFS_SES_LANMAN; @@ -859,9 +859,10 @@ ssetup_ntlmssp_authenticate: iov[2].iov_len = (long) bcc_ptr - (long) str_area; count = iov[1].iov_len + iov[2].iov_len; - smb_buf->smb_buf_length += count; + smb_buf->smb_buf_length = + cpu_to_be32(be32_to_cpu(smb_buf->smb_buf_length) + count); - put_bcc_le(count, smb_buf); + put_bcc(count, smb_buf); rc = SendReceive2(xid, ses, iov, 3 /* num_iovecs */, &resp_buf_type, CIFS_LOG_ERROR); diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c deleted file mode 100644 index 0472148..0000000 --- a/fs/cifs/smbdes.c +++ /dev/null @@ -1,418 +0,0 @@ -/* - Unix SMB/Netbios implementation. - Version 1.9. - - a partial implementation of DES designed for use in the - SMB authentication protocol - - Copyright (C) Andrew Tridgell 1998 - Modified by Steve French (sfrench@us.ibm.com) 2002,2004 - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. -*/ - -/* NOTES: - - This code makes no attempt to be fast! In fact, it is a very - slow implementation - - This code is NOT a complete DES implementation. It implements only - the minimum necessary for SMB authentication, as used by all SMB - products (including every copy of Microsoft Windows95 ever sold) - - In particular, it can only do a unchained forward DES pass. This - means it is not possible to use this code for encryption/decryption - of data, instead it is only useful as a "hash" algorithm. - - There is no entry point into this code that allows normal DES operation. - - I believe this means that this code does not come under ITAR - regulations but this is NOT a legal opinion. If you are concerned - about the applicability of ITAR regulations to this code then you - should confirm it for yourself (and maybe let me know if you come - up with a different answer to the one above) -*/ -#include <linux/slab.h> -#define uchar unsigned char - -static uchar perm1[56] = { 57, 49, 41, 33, 25, 17, 9, - 1, 58, 50, 42, 34, 26, 18, - 10, 2, 59, 51, 43, 35, 27, - 19, 11, 3, 60, 52, 44, 36, - 63, 55, 47, 39, 31, 23, 15, - 7, 62, 54, 46, 38, 30, 22, - 14, 6, 61, 53, 45, 37, 29, - 21, 13, 5, 28, 20, 12, 4 -}; - -static uchar perm2[48] = { 14, 17, 11, 24, 1, 5, - 3, 28, 15, 6, 21, 10, - 23, 19, 12, 4, 26, 8, - 16, 7, 27, 20, 13, 2, - 41, 52, 31, 37, 47, 55, - 30, 40, 51, 45, 33, 48, - 44, 49, 39, 56, 34, 53, - 46, 42, 50, 36, 29, 32 -}; - -static uchar perm3[64] = { 58, 50, 42, 34, 26, 18, 10, 2, - 60, 52, 44, 36, 28, 20, 12, 4, - 62, 54, 46, 38, 30, 22, 14, 6, - 64, 56, 48, 40, 32, 24, 16, 8, - 57, 49, 41, 33, 25, 17, 9, 1, - 59, 51, 43, 35, 27, 19, 11, 3, - 61, 53, 45, 37, 29, 21, 13, 5, - 63, 55, 47, 39, 31, 23, 15, 7 -}; - -static uchar perm4[48] = { 32, 1, 2, 3, 4, 5, - 4, 5, 6, 7, 8, 9, - 8, 9, 10, 11, 12, 13, - 12, 13, 14, 15, 16, 17, - 16, 17, 18, 19, 20, 21, - 20, 21, 22, 23, 24, 25, - 24, 25, 26, 27, 28, 29, - 28, 29, 30, 31, 32, 1 -}; - -static uchar perm5[32] = { 16, 7, 20, 21, - 29, 12, 28, 17, - 1, 15, 23, 26, - 5, 18, 31, 10, - 2, 8, 24, 14, - 32, 27, 3, 9, - 19, 13, 30, 6, - 22, 11, 4, 25 -}; - -static uchar perm6[64] = { 40, 8, 48, 16, 56, 24, 64, 32, - 39, 7, 47, 15, 55, 23, 63, 31, - 38, 6, 46, 14, 54, 22, 62, 30, - 37, 5, 45, 13, 53, 21, 61, 29, - 36, 4, 44, 12, 52, 20, 60, 28, - 35, 3, 43, 11, 51, 19, 59, 27, - 34, 2, 42, 10, 50, 18, 58, 26, - 33, 1, 41, 9, 49, 17, 57, 25 -}; - -static uchar sc[16] = { 1, 1, 2, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1 }; - -static uchar sbox[8][4][16] = { - {{14, 4, 13, 1, 2, 15, 11, 8, 3, 10, 6, 12, 5, 9, 0, 7}, - {0, 15, 7, 4, 14, 2, 13, 1, 10, 6, 12, 11, 9, 5, 3, 8}, - {4, 1, 14, 8, 13, 6, 2, 11, 15, 12, 9, 7, 3, 10, 5, 0}, - {15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13} }, - - {{15, 1, 8, 14, 6, 11, 3, 4, 9, 7, 2, 13, 12, 0, 5, 10}, - {3, 13, 4, 7, 15, 2, 8, 14, 12, 0, 1, 10, 6, 9, 11, 5}, - {0, 14, 7, 11, 10, 4, 13, 1, 5, 8, 12, 6, 9, 3, 2, 15}, - {13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9} }, - - {{10, 0, 9, 14, 6, 3, 15, 5, 1, 13, 12, 7, 11, 4, 2, 8}, - {13, 7, 0, 9, 3, 4, 6, 10, 2, 8, 5, 14, 12, 11, 15, 1}, - {13, 6, 4, 9, 8, 15, 3, 0, 11, 1, 2, 12, 5, 10, 14, 7}, - {1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12} }, - - {{7, 13, 14, 3, 0, 6, 9, 10, 1, 2, 8, 5, 11, 12, 4, 15}, - {13, 8, 11, 5, 6, 15, 0, 3, 4, 7, 2, 12, 1, 10, 14, 9}, - {10, 6, 9, 0, 12, 11, 7, 13, 15, 1, 3, 14, 5, 2, 8, 4}, - {3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14} }, - - {{2, 12, 4, 1, 7, 10, 11, 6, 8, 5, 3, 15, 13, 0, 14, 9}, - {14, 11, 2, 12, 4, 7, 13, 1, 5, 0, 15, 10, 3, 9, 8, 6}, - {4, 2, 1, 11, 10, 13, 7, 8, 15, 9, 12, 5, 6, 3, 0, 14}, - {11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3} }, - - {{12, 1, 10, 15, 9, 2, 6, 8, 0, 13, 3, 4, 14, 7, 5, 11}, - {10, 15, 4, 2, 7, 12, 9, 5, 6, 1, 13, 14, 0, 11, 3, 8}, - {9, 14, 15, 5, 2, 8, 12, 3, 7, 0, 4, 10, 1, 13, 11, 6}, - {4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13} }, - - {{4, 11, 2, 14, 15, 0, 8, 13, 3, 12, 9, 7, 5, 10, 6, 1}, - {13, 0, 11, 7, 4, 9, 1, 10, 14, 3, 5, 12, 2, 15, 8, 6}, - {1, 4, 11, 13, 12, 3, 7, 14, 10, 15, 6, 8, 0, 5, 9, 2}, - {6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12} }, - - {{13, 2, 8, 4, 6, 15, 11, 1, 10, 9, 3, 14, 5, 0, 12, 7}, - {1, 15, 13, 8, 10, 3, 7, 4, 12, 5, 6, 11, 0, 14, 9, 2}, - {7, 11, 4, 1, 9, 12, 14, 2, 0, 6, 10, 13, 15, 3, 5, 8}, - {2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11} } -}; - -static void -permute(char *out, char *in, uchar *p, int n) -{ - int i; - for (i = 0; i < n; i++) - out[i] = in[p[i] - 1]; -} - -static void -lshift(char *d, int count, int n) -{ - char out[64]; - int i; - for (i = 0; i < n; i++) - out[i] = d[(i + count) % n]; - for (i = 0; i < n; i++) - d[i] = out[i]; -} - -static void -concat(char *out, char *in1, char *in2, int l1, int l2) -{ - while (l1--) - *out++ = *in1++; - while (l2--) - *out++ = *in2++; -} - -static void -xor(char *out, char *in1, char *in2, int n) -{ - int i; - for (i = 0; i < n; i++) - out[i] = in1[i] ^ in2[i]; -} - -static void -dohash(char *out, char *in, char *key, int forw) -{ - int i, j, k; - char *pk1; - char c[28]; - char d[28]; - char *cd; - char (*ki)[48]; - char *pd1; - char l[32], r[32]; - char *rl; - - /* Have to reduce stack usage */ - pk1 = kmalloc(56+56+64+64, GFP_KERNEL); - if (pk1 == NULL) - return; - - ki = kmalloc(16*48, GFP_KERNEL); - if (ki == NULL) { - kfree(pk1); - return; - } - - cd = pk1 + 56; - pd1 = cd + 56; - rl = pd1 + 64; - - permute(pk1, key, perm1, 56); - - for (i = 0; i < 28; i++) - c[i] = pk1[i]; - for (i = 0; i < 28; i++) - d[i] = pk1[i + 28]; - - for (i = 0; i < 16; i++) { - lshift(c, sc[i], 28); - lshift(d, sc[i], 28); - - concat(cd, c, d, 28, 28); - permute(ki[i], cd, perm2, 48); - } - - permute(pd1, in, perm3, 64); - - for (j = 0; j < 32; j++) { - l[j] = pd1[j]; - r[j] = pd1[j + 32]; - } - - for (i = 0; i < 16; i++) { - char *er; /* er[48] */ - char *erk; /* erk[48] */ - char b[8][6]; - char *cb; /* cb[32] */ - char *pcb; /* pcb[32] */ - char *r2; /* r2[32] */ - - er = kmalloc(48+48+32+32+32, GFP_KERNEL); - if (er == NULL) { - kfree(pk1); - kfree(ki); - return; - } - erk = er+48; - cb = erk+48; - pcb = cb+32; - r2 = pcb+32; - - permute(er, r, perm4, 48); - - xor(erk, er, ki[forw ? i : 15 - i], 48); - - for (j = 0; j < 8; j++) - for (k = 0; k < 6; k++) - b[j][k] = erk[j * 6 + k]; - - for (j = 0; j < 8; j++) { - int m, n; - m = (b[j][0] << 1) | b[j][5]; - - n = (b[j][1] << 3) | (b[j][2] << 2) | (b[j][3] << - 1) | b[j][4]; - - for (k = 0; k < 4; k++) - b[j][k] = - (sbox[j][m][n] & (1 << (3 - k))) ? 1 : 0; - } - - for (j = 0; j < 8; j++) - for (k = 0; k < 4; k++) - cb[j * 4 + k] = b[j][k]; - permute(pcb, cb, perm5, 32); - - xor(r2, l, pcb, 32); - - for (j = 0; j < 32; j++) - l[j] = r[j]; - - for (j = 0; j < 32; j++) - r[j] = r2[j]; - - kfree(er); - } - - concat(rl, r, l, 32, 32); - - permute(out, rl, perm6, 64); - kfree(pk1); - kfree(ki); -} - -static void -str_to_key(unsigned char *str, unsigned char *key) -{ - int i; - - key[0] = str[0] >> 1; - key[1] = ((str[0] & 0x01) << 6) | (str[1] >> 2); - key[2] = ((str[1] & 0x03) << 5) | (str[2] >> 3); - key[3] = ((str[2] & 0x07) << 4) | (str[3] >> 4); - key[4] = ((str[3] & 0x0F) << 3) | (str[4] >> 5); - key[5] = ((str[4] & 0x1F) << 2) | (str[5] >> 6); - key[6] = ((str[5] & 0x3F) << 1) | (str[6] >> 7); - key[7] = str[6] & 0x7F; - for (i = 0; i < 8; i++) - key[i] = (key[i] << 1); -} - -static void -smbhash(unsigned char *out, const unsigned char *in, unsigned char *key, - int forw) -{ - int i; - char *outb; /* outb[64] */ - char *inb; /* inb[64] */ - char *keyb; /* keyb[64] */ - unsigned char key2[8]; - - outb = kmalloc(64 * 3, GFP_KERNEL); - if (outb == NULL) - return; - - inb = outb + 64; - keyb = inb + 64; - - str_to_key(key, key2); - - for (i = 0; i < 64; i++) { - inb[i] = (in[i / 8] & (1 << (7 - (i % 8)))) ? 1 : 0; - keyb[i] = (key2[i / 8] & (1 << (7 - (i % 8)))) ? 1 : 0; - outb[i] = 0; - } - - dohash(outb, inb, keyb, forw); - - for (i = 0; i < 8; i++) - out[i] = 0; - - for (i = 0; i < 64; i++) { - if (outb[i]) - out[i / 8] |= (1 << (7 - (i % 8))); - } - kfree(outb); -} - -void -E_P16(unsigned char *p14, unsigned char *p16) -{ - unsigned char sp8[8] = - { 0x4b, 0x47, 0x53, 0x21, 0x40, 0x23, 0x24, 0x25 }; - smbhash(p16, sp8, p14, 1); - smbhash(p16 + 8, sp8, p14 + 7, 1); -} - -void -E_P24(unsigned char *p21, const unsigned char *c8, unsigned char *p24) -{ - smbhash(p24, c8, p21, 1); - smbhash(p24 + 8, c8, p21 + 7, 1); - smbhash(p24 + 16, c8, p21 + 14, 1); -} - -#if 0 /* currently unused */ -static void -D_P16(unsigned char *p14, unsigned char *in, unsigned char *out) -{ - smbhash(out, in, p14, 0); - smbhash(out + 8, in + 8, p14 + 7, 0); -} - -static void -E_old_pw_hash(unsigned char *p14, unsigned char *in, unsigned char *out) -{ - smbhash(out, in, p14, 1); - smbhash(out + 8, in + 8, p14 + 7, 1); -} -/* these routines are currently unneeded, but may be - needed later */ -void -cred_hash1(unsigned char *out, unsigned char *in, unsigned char *key) -{ - unsigned char buf[8]; - - smbhash(buf, in, key, 1); - smbhash(out, buf, key + 9, 1); -} - -void -cred_hash2(unsigned char *out, unsigned char *in, unsigned char *key) -{ - unsigned char buf[8]; - static unsigned char key2[8]; - - smbhash(buf, in, key, 1); - key2[0] = key[7]; - smbhash(out, buf, key2, 1); -} - -void -cred_hash3(unsigned char *out, unsigned char *in, unsigned char *key, int forw) -{ - static unsigned char key2[8]; - - smbhash(out, in, key, forw); - key2[0] = key[7]; - smbhash(out + 8, in + 8, key2, forw); -} -#endif /* unneeded routines */ diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index b5041c8..1c5b770 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -47,6 +47,86 @@ #define SSVALX(buf,pos,val) (CVAL(buf,pos)=(val)&0xFF,CVAL(buf,pos+1)=(val)>>8) #define SSVAL(buf,pos,val) SSVALX((buf),(pos),((__u16)(val))) +static void +str_to_key(unsigned char *str, unsigned char *key) +{ + int i; + + key[0] = str[0] >> 1; + key[1] = ((str[0] & 0x01) << 6) | (str[1] >> 2); + key[2] = ((str[1] & 0x03) << 5) | (str[2] >> 3); + key[3] = ((str[2] & 0x07) << 4) | (str[3] >> 4); + key[4] = ((str[3] & 0x0F) << 3) | (str[4] >> 5); + key[5] = ((str[4] & 0x1F) << 2) | (str[5] >> 6); + key[6] = ((str[5] & 0x3F) << 1) | (str[6] >> 7); + key[7] = str[6] & 0x7F; + for (i = 0; i < 8; i++) + key[i] = (key[i] << 1); +} + +static int +smbhash(unsigned char *out, const unsigned char *in, unsigned char *key) +{ + int rc; + unsigned char key2[8]; + struct crypto_blkcipher *tfm_des; + struct scatterlist sgin, sgout; + struct blkcipher_desc desc; + + str_to_key(key, key2); + + tfm_des = crypto_alloc_blkcipher("ecb(des)", 0, CRYPTO_ALG_ASYNC); + if (IS_ERR(tfm_des)) { + rc = PTR_ERR(tfm_des); + cERROR(1, "could not allocate des crypto API\n"); + goto smbhash_err; + } + + desc.tfm = tfm_des; + + crypto_blkcipher_setkey(tfm_des, key2, 8); + + sg_init_one(&sgin, in, 8); + sg_init_one(&sgout, out, 8); + + rc = crypto_blkcipher_encrypt(&desc, &sgout, &sgin, 8); + if (rc) + cERROR(1, "could not encrypt crypt key rc: %d\n", rc); + + crypto_free_blkcipher(tfm_des); +smbhash_err: + return rc; +} + +static int +E_P16(unsigned char *p14, unsigned char *p16) +{ + int rc; + unsigned char sp8[8] = + { 0x4b, 0x47, 0x53, 0x21, 0x40, 0x23, 0x24, 0x25 }; + + rc = smbhash(p16, sp8, p14); + if (rc) + return rc; + rc = smbhash(p16 + 8, sp8, p14 + 7); + return rc; +} + +static int +E_P24(unsigned char *p21, const unsigned char *c8, unsigned char *p24) +{ + int rc; + + rc = smbhash(p24, c8, p21); + if (rc) + return rc; + rc = smbhash(p24 + 8, c8, p21 + 7); + if (rc) + return rc; + rc = smbhash(p24 + 16, c8, p21 + 14); + return rc; +} + /* produce a md4 message digest from data of length n bytes */ int mdfour(unsigned char *md4_hash, unsigned char *link_str, int link_len) @@ -87,40 +167,30 @@ mdfour_err: return rc; } -/* Does the des encryption from the NT or LM MD4 hash. */ -static void -SMBOWFencrypt(unsigned char passwd[16], const unsigned char *c8, - unsigned char p24[24]) -{ - unsigned char p21[21]; - - memset(p21, '\0', 21); - - memcpy(p21, passwd, 16); - E_P24(p21, c8, p24); -} - /* This implements the X/Open SMB password encryption It takes a password, a 8 byte "crypt key" and puts 24 bytes of encrypted password into p24 */ /* Note that password must be uppercased and null terminated */ -void +int SMBencrypt(unsigned char *passwd, const unsigned char *c8, unsigned char *p24) { - unsigned char p14[15], p21[21]; + int rc; + unsigned char p14[14], p16[16], p21[21]; - memset(p21, '\0', 21); memset(p14, '\0', 14); - strncpy((char *) p14, (char *) passwd, 14); + memset(p16, '\0', 16); + memset(p21, '\0', 21); -/* strupper((char *)p14); *//* BB at least uppercase the easy range */ - E_P16(p14, p21); + memcpy(p14, passwd, 14); + rc = E_P16(p14, p16); + if (rc) + return rc; - SMBOWFencrypt(p21, c8, p24); + memcpy(p21, p16, 16); + rc = E_P24(p21, c8, p24); - memset(p14, 0, 15); - memset(p21, 0, 21); + return rc; } /* Routines for Windows NT MD4 Hash functions. */ @@ -279,16 +349,18 @@ int SMBNTencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24) { int rc; - unsigned char p21[21]; + unsigned char p16[16], p21[21]; + memset(p16, '\0', 16); memset(p21, '\0', 21); - rc = E_md4hash(passwd, p21); + rc = E_md4hash(passwd, p16); if (rc) { cFYI(1, "%s Can't generate NT hash, error: %d", __func__, rc); return rc; } - SMBOWFencrypt(p21, c8, p24); + memcpy(p21, p16, 16); + rc = E_P24(p21, c8, p24); return rc; } diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 46d8756f..147aa22 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -129,7 +129,7 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec) unsigned int len = iov[0].iov_len; unsigned int total_len; int first_vec = 0; - unsigned int smb_buf_length = smb_buffer->smb_buf_length; + unsigned int smb_buf_length = be32_to_cpu(smb_buffer->smb_buf_length); struct socket *ssocket = server->ssocket; if (ssocket == NULL) @@ -144,17 +144,10 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec) else smb_msg.msg_flags = MSG_NOSIGNAL; - /* smb header is converted in header_assemble. bcc and rest of SMB word - area, and byte area if necessary, is converted to littleendian in - cifssmb.c and RFC1001 len is converted to bigendian in smb_send - Flags2 is converted in SendReceive */ - - total_len = 0; for (i = 0; i < n_vec; i++) total_len += iov[i].iov_len; - smb_buffer->smb_buf_length = cpu_to_be32(smb_buffer->smb_buf_length); cFYI(1, "Sending smb: total_len %d", total_len); dump_smb(smb_buffer, len); @@ -243,7 +236,7 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec) /* Don't want to modify the buffer as a side effect of this call. */ - smb_buffer->smb_buf_length = smb_buf_length; + smb_buffer->smb_buf_length = cpu_to_be32(smb_buf_length); return rc; } @@ -302,7 +295,7 @@ static int wait_for_free_request(struct TCP_Server_Info *server, return 0; } -static int allocate_mid(struct cifsSesInfo *ses, struct smb_hdr *in_buf, +static int allocate_mid(struct cifs_ses *ses, struct smb_hdr *in_buf, struct mid_q_entry **ppmidQ) { if (ses->server->tcpStatus == CifsExiting) { @@ -349,22 +342,24 @@ wait_for_response(struct TCP_Server_Info *server, struct mid_q_entry *midQ) * the result. Caller is responsible for dealing with timeouts. */ int -cifs_call_async(struct TCP_Server_Info *server, struct smb_hdr *in_buf, - mid_callback_t *callback, void *cbdata) +cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, + unsigned int nvec, mid_callback_t *callback, void *cbdata, + bool ignore_pend) { int rc; struct mid_q_entry *mid; + struct smb_hdr *hdr = (struct smb_hdr *)iov[0].iov_base; - rc = wait_for_free_request(server, CIFS_ASYNC_OP); + rc = wait_for_free_request(server, ignore_pend ? CIFS_ASYNC_OP : 0); if (rc) return rc; /* enable signing if server requires it */ - if (server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) - in_buf->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; + if (server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + hdr->Flags2 |= SMBFLG2_SECURITY_SIGNATURE; mutex_lock(&server->srv_mutex); - mid = AllocMidQEntry(in_buf, server); + mid = AllocMidQEntry(hdr, server); if (mid == NULL) { mutex_unlock(&server->srv_mutex); return -ENOMEM; @@ -375,7 +370,7 @@ cifs_call_async(struct TCP_Server_Info *server, struct smb_hdr *in_buf, list_add_tail(&mid->qhead, &server->pending_mid_q); spin_unlock(&GlobalMid_Lock); - rc = cifs_sign_smb(in_buf, server, &mid->sequence_number); + rc = cifs_sign_smb2(iov, nvec, server, &mid->sequence_number); if (rc) { mutex_unlock(&server->srv_mutex); goto out_err; @@ -387,7 +382,7 @@ cifs_call_async(struct TCP_Server_Info *server, struct smb_hdr *in_buf, #ifdef CONFIG_CIFS_STATS2 atomic_inc(&server->inSend); #endif - rc = smb_send(server, in_buf, in_buf->smb_buf_length); + rc = smb_sendv(server, iov, nvec); #ifdef CONFIG_CIFS_STATS2 atomic_dec(&server->inSend); mid->when_sent = jiffies; @@ -414,7 +409,7 @@ out_err: * */ int -SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, +SendReceiveNoRsp(const unsigned int xid, struct cifs_ses *ses, struct smb_hdr *in_buf, int flags) { int rc; @@ -422,7 +417,7 @@ SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, int resp_buf_type; iov[0].iov_base = (char *)in_buf; - iov[0].iov_len = in_buf->smb_buf_length + 4; + iov[0].iov_len = be32_to_cpu(in_buf->smb_buf_length) + 4; flags |= CIFS_NO_RESP; rc = SendReceive2(xid, ses, iov, 1, &resp_buf_type, flags); cFYI(DBG2, "SendRcvNoRsp flags %d rc %d", flags, rc); @@ -431,7 +426,7 @@ SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses, } static int -sync_mid_result(struct mid_q_entry *mid, struct TCP_Server_Info *server) +cifs_sync_mid_result(struct mid_q_entry *mid, struct TCP_Server_Info *server) { int rc = 0; @@ -439,28 +434,21 @@ sync_mid_result(struct mid_q_entry *mid, struct TCP_Server_Info *server) mid->mid, mid->midState); spin_lock(&GlobalMid_Lock); - /* ensure that it's no longer on the pending_mid_q */ - list_del_init(&mid->qhead); - switch (mid->midState) { case MID_RESPONSE_RECEIVED: spin_unlock(&GlobalMid_Lock); return rc; - case MID_REQUEST_SUBMITTED: - /* socket is going down, reject all calls */ - if (server->tcpStatus == CifsExiting) { - cERROR(1, "%s: canceling mid=%d cmd=0x%x state=%d", - __func__, mid->mid, mid->command, mid->midState); - rc = -EHOSTDOWN; - break; - } case MID_RETRY_NEEDED: rc = -EAGAIN; break; case MID_RESPONSE_MALFORMED: rc = -EIO; break; + case MID_SHUTDOWN: + rc = -EHOSTDOWN; + break; default: + list_del_init(&mid->qhead); cERROR(1, "%s: invalid mid state mid=%d state=%d", __func__, mid->mid, mid->midState); rc = -EIO; @@ -488,10 +476,10 @@ send_nt_cancel(struct TCP_Server_Info *server, struct smb_hdr *in_buf, int rc = 0; /* -4 for RFC1001 length and +2 for BCC field */ - in_buf->smb_buf_length = sizeof(struct smb_hdr) - 4 + 2; + in_buf->smb_buf_length = cpu_to_be32(sizeof(struct smb_hdr) - 4 + 2); in_buf->Command = SMB_COM_NT_CANCEL; in_buf->WordCount = 0; - put_bcc_le(0, in_buf); + put_bcc(0, in_buf); mutex_lock(&server->srv_mutex); rc = cifs_sign_smb(in_buf, server, &mid->sequence_number); @@ -499,7 +487,7 @@ send_nt_cancel(struct TCP_Server_Info *server, struct smb_hdr *in_buf, mutex_unlock(&server->srv_mutex); return rc; } - rc = smb_send(server, in_buf, in_buf->smb_buf_length); + rc = smb_send(server, in_buf, be32_to_cpu(in_buf->smb_buf_length)); mutex_unlock(&server->srv_mutex); cFYI(1, "issued NT_CANCEL for mid %u, rc = %d", @@ -509,13 +497,31 @@ send_nt_cancel(struct TCP_Server_Info *server, struct smb_hdr *in_buf, } int -SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, +cifs_check_receive(struct mid_q_entry *mid, struct TCP_Server_Info *server, + bool log_error) +{ + dump_smb(mid->resp_buf, + min_t(u32, 92, be32_to_cpu(mid->resp_buf->smb_buf_length))); + + /* convert the length into a more usable form */ + if (server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { + /* FIXME: add code to kill session */ + if (cifs_verify_signature(mid->resp_buf, server, + mid->sequence_number + 1) != 0) + cERROR(1, "Unexpected SMB signature"); + } + + /* BB special case reconnect tid and uid here? */ + return map_smb_to_linux_error(mid->resp_buf, log_error); +} + +int +SendReceive2(const unsigned int xid, struct cifs_ses *ses, struct kvec *iov, int n_vec, int *pRespBufType /* ret */, const int flags) { int rc = 0; int long_op; - unsigned int receive_len; struct mid_q_entry *midQ; struct smb_hdr *in_buf = iov[0].iov_base; @@ -605,66 +611,31 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, cifs_small_buf_release(in_buf); - rc = sync_mid_result(midQ, ses->server); + rc = cifs_sync_mid_result(midQ, ses->server); if (rc != 0) { atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); return rc; } - receive_len = midQ->resp_buf->smb_buf_length; - - if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { - cERROR(1, "Frame too large received. Length: %d Xid: %d", - receive_len, xid); + if (!midQ->resp_buf || midQ->midState != MID_RESPONSE_RECEIVED) { rc = -EIO; + cFYI(1, "Bad MID state?"); goto out; } - /* rcvd frame is ok */ - - if (midQ->resp_buf && - (midQ->midState == MID_RESPONSE_RECEIVED)) { - - iov[0].iov_base = (char *)midQ->resp_buf; - if (midQ->largeBuf) - *pRespBufType = CIFS_LARGE_BUFFER; - else - *pRespBufType = CIFS_SMALL_BUFFER; - iov[0].iov_len = receive_len + 4; - - dump_smb(midQ->resp_buf, 80); - /* convert the length into a more usable form */ - if ((receive_len > 24) && - (ses->server->secMode & (SECMODE_SIGN_REQUIRED | - SECMODE_SIGN_ENABLED))) { - rc = cifs_verify_signature(midQ->resp_buf, - ses->server, - midQ->sequence_number+1); - if (rc) { - cERROR(1, "Unexpected SMB signature"); - /* BB FIXME add code to kill session */ - } - } + iov[0].iov_base = (char *)midQ->resp_buf; + iov[0].iov_len = be32_to_cpu(midQ->resp_buf->smb_buf_length) + 4; + if (midQ->largeBuf) + *pRespBufType = CIFS_LARGE_BUFFER; + else + *pRespBufType = CIFS_SMALL_BUFFER; - /* BB special case reconnect tid and uid here? */ - rc = map_smb_to_linux_error(midQ->resp_buf, - flags & CIFS_LOG_ERROR); - - /* convert ByteCount if necessary */ - if (receive_len >= sizeof(struct smb_hdr) - 4 - /* do not count RFC1001 header */ + - (2 * midQ->resp_buf->WordCount) + 2 /* bcc */ ) - put_bcc(get_bcc_le(midQ->resp_buf), midQ->resp_buf); - if ((flags & CIFS_NO_RESP) == 0) - midQ->resp_buf = NULL; /* mark it so buf will - not be freed by - delete_mid */ - } else { - rc = -EIO; - cFYI(1, "Bad MID state?"); - } + rc = cifs_check_receive(midQ, ses->server, flags & CIFS_LOG_ERROR); + /* mark it so buf will not be freed by delete_mid */ + if ((flags & CIFS_NO_RESP) == 0) + midQ->resp_buf = NULL; out: delete_mid(midQ); atomic_dec(&ses->server->inFlight); @@ -674,12 +645,11 @@ out: } int -SendReceive(const unsigned int xid, struct cifsSesInfo *ses, +SendReceive(const unsigned int xid, struct cifs_ses *ses, struct smb_hdr *in_buf, struct smb_hdr *out_buf, int *pbytes_returned, const int long_op) { int rc = 0; - unsigned int receive_len; struct mid_q_entry *midQ; if (ses == NULL) { @@ -698,9 +668,10 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, to the same server. We may make this configurable later or use ses->maxReq */ - if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + if (be32_to_cpu(in_buf->smb_buf_length) > CIFSMaxBufSize + + MAX_CIFS_HDR_SIZE - 4) { cERROR(1, "Illegal length, greater than maximum frame, %d", - in_buf->smb_buf_length); + be32_to_cpu(in_buf->smb_buf_length)); return -EIO; } @@ -733,7 +704,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, #ifdef CONFIG_CIFS_STATS2 atomic_inc(&ses->server->inSend); #endif - rc = smb_send(ses->server, in_buf, in_buf->smb_buf_length); + rc = smb_send(ses->server, in_buf, be32_to_cpu(in_buf->smb_buf_length)); #ifdef CONFIG_CIFS_STATS2 atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; @@ -761,60 +732,23 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, spin_unlock(&GlobalMid_Lock); } - rc = sync_mid_result(midQ, ses->server); + rc = cifs_sync_mid_result(midQ, ses->server); if (rc != 0) { atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); return rc; } - receive_len = midQ->resp_buf->smb_buf_length; - - if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { - cERROR(1, "Frame too large received. Length: %d Xid: %d", - receive_len, xid); - rc = -EIO; - goto out; - } - - /* rcvd frame is ok */ - - if (midQ->resp_buf && out_buf - && (midQ->midState == MID_RESPONSE_RECEIVED)) { - out_buf->smb_buf_length = receive_len; - memcpy((char *)out_buf + 4, - (char *)midQ->resp_buf + 4, - receive_len); - - dump_smb(out_buf, 92); - /* convert the length into a more usable form */ - if ((receive_len > 24) && - (ses->server->secMode & (SECMODE_SIGN_REQUIRED | - SECMODE_SIGN_ENABLED))) { - rc = cifs_verify_signature(out_buf, - ses->server, - midQ->sequence_number+1); - if (rc) { - cERROR(1, "Unexpected SMB signature"); - /* BB FIXME add code to kill session */ - } - } - - *pbytes_returned = out_buf->smb_buf_length; - - /* BB special case reconnect tid and uid here? */ - rc = map_smb_to_linux_error(out_buf, 0 /* no log */ ); - - /* convert ByteCount if necessary */ - if (receive_len >= sizeof(struct smb_hdr) - 4 - /* do not count RFC1001 header */ + - (2 * out_buf->WordCount) + 2 /* bcc */ ) - put_bcc(get_bcc_le(midQ->resp_buf), midQ->resp_buf); - } else { + if (!midQ->resp_buf || !out_buf || + midQ->midState != MID_RESPONSE_RECEIVED) { rc = -EIO; cERROR(1, "Bad MID state?"); + goto out; } + *pbytes_returned = be32_to_cpu(midQ->resp_buf->smb_buf_length); + memcpy(out_buf, midQ->resp_buf, *pbytes_returned + 4); + rc = cifs_check_receive(midQ, ses->server, 0); out: delete_mid(midQ); atomic_dec(&ses->server->inFlight); @@ -827,12 +761,12 @@ out: blocking lock to return. */ static int -send_lock_cancel(const unsigned int xid, struct cifsTconInfo *tcon, +send_lock_cancel(const unsigned int xid, struct cifs_tcon *tcon, struct smb_hdr *in_buf, struct smb_hdr *out_buf) { int bytes_returned; - struct cifsSesInfo *ses = tcon->ses; + struct cifs_ses *ses = tcon->ses; LOCK_REQ *pSMB = (LOCK_REQ *)in_buf; /* We just modify the current in_buf to change @@ -849,15 +783,14 @@ send_lock_cancel(const unsigned int xid, struct cifsTconInfo *tcon, } int -SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, +SendReceiveBlockingLock(const unsigned int xid, struct cifs_tcon *tcon, struct smb_hdr *in_buf, struct smb_hdr *out_buf, int *pbytes_returned) { int rc = 0; int rstart = 0; - unsigned int receive_len; struct mid_q_entry *midQ; - struct cifsSesInfo *ses; + struct cifs_ses *ses; if (tcon == NULL || tcon->ses == NULL) { cERROR(1, "Null smb session"); @@ -877,9 +810,10 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, to the same server. We may make this configurable later or use ses->maxReq */ - if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + if (be32_to_cpu(in_buf->smb_buf_length) > CIFSMaxBufSize + + MAX_CIFS_HDR_SIZE - 4) { cERROR(1, "Illegal length, greater than maximum frame, %d", - in_buf->smb_buf_length); + be32_to_cpu(in_buf->smb_buf_length)); return -EIO; } @@ -910,7 +844,7 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, #ifdef CONFIG_CIFS_STATS2 atomic_inc(&ses->server->inSend); #endif - rc = smb_send(ses->server, in_buf, in_buf->smb_buf_length); + rc = smb_send(ses->server, in_buf, be32_to_cpu(in_buf->smb_buf_length)); #ifdef CONFIG_CIFS_STATS2 atomic_dec(&ses->server->inSend); midQ->when_sent = jiffies; @@ -973,56 +907,20 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, rstart = 1; } - rc = sync_mid_result(midQ, ses->server); + rc = cifs_sync_mid_result(midQ, ses->server); if (rc != 0) return rc; - receive_len = midQ->resp_buf->smb_buf_length; - if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { - cERROR(1, "Frame too large received. Length: %d Xid: %d", - receive_len, xid); - rc = -EIO; - goto out; - } - /* rcvd frame is ok */ - - if ((out_buf == NULL) || (midQ->midState != MID_RESPONSE_RECEIVED)) { + if (out_buf == NULL || midQ->midState != MID_RESPONSE_RECEIVED) { rc = -EIO; cERROR(1, "Bad MID state?"); goto out; } - out_buf->smb_buf_length = receive_len; - memcpy((char *)out_buf + 4, - (char *)midQ->resp_buf + 4, - receive_len); - - dump_smb(out_buf, 92); - /* convert the length into a more usable form */ - if ((receive_len > 24) && - (ses->server->secMode & (SECMODE_SIGN_REQUIRED | - SECMODE_SIGN_ENABLED))) { - rc = cifs_verify_signature(out_buf, - ses->server, - midQ->sequence_number+1); - if (rc) { - cERROR(1, "Unexpected SMB signature"); - /* BB FIXME add code to kill session */ - } - } - - *pbytes_returned = out_buf->smb_buf_length; - - /* BB special case reconnect tid and uid here? */ - rc = map_smb_to_linux_error(out_buf, 0 /* no log */ ); - - /* convert ByteCount if necessary */ - if (receive_len >= sizeof(struct smb_hdr) - 4 - /* do not count RFC1001 header */ + - (2 * out_buf->WordCount) + 2 /* bcc */ ) - put_bcc(get_bcc_le(out_buf), out_buf); - + *pbytes_returned = be32_to_cpu(midQ->resp_buf->smb_buf_length); + memcpy(out_buf, midQ->resp_buf, *pbytes_returned + 4); + rc = cifs_check_receive(midQ, ses->server, 0); out: delete_mid(midQ); if (rstart && rc == -EACCES) diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index eae2a14..2a22fb2 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -49,7 +49,7 @@ int cifs_removexattr(struct dentry *direntry, const char *ea_name) int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct super_block *sb; char *full_path = NULL; @@ -109,9 +109,10 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct super_block *sb; char *full_path; + struct cifs_ntsd *pacl; if (direntry == NULL) return -EIO; @@ -166,6 +167,25 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name, rc = CIFSSMBSetEA(xid, pTcon, full_path, ea_name, ea_value, (__u16)value_size, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + } else if (strncmp(ea_name, CIFS_XATTR_CIFS_ACL, + strlen(CIFS_XATTR_CIFS_ACL)) == 0) { + pacl = kmalloc(value_size, GFP_KERNEL); + if (!pacl) { + cFYI(1, "%s: Can't allocate memory for ACL", + __func__); + rc = -ENOMEM; + } else { +#ifdef CONFIG_CIFS_ACL + memcpy(pacl, ea_value, value_size); + rc = set_cifs_acl(pacl, value_size, + direntry->d_inode, full_path); + if (rc == 0) /* force revalidate of the inode */ + CIFS_I(direntry->d_inode)->time = 0; + kfree(pacl); +#else + cFYI(1, "Set CIFS ACL not supported yet"); +#endif /* CONFIG_CIFS_ACL */ + } } else { int temp; temp = strncmp(ea_name, POSIX_ACL_XATTR_ACCESS, @@ -220,7 +240,7 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct super_block *sb; char *full_path; @@ -352,7 +372,7 @@ ssize_t cifs_listxattr(struct dentry *direntry, char *data, size_t buf_size) int xid; struct cifs_sb_info *cifs_sb; struct tcon_link *tlink; - struct cifsTconInfo *pTcon; + struct cifs_tcon *pTcon; struct super_block *sb; char *full_path; diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index 6cbb3af..cb140ef 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -43,8 +43,6 @@ const struct file_operations coda_ioctl_operations = { /* the coda pioctl inode ops */ static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags) { - if (flags & IPERM_FLAG_RCU) - return -ECHILD; return (mask & MAY_EXEC) ? -EACCES : 0; } diff --git a/fs/compat.c b/fs/compat.c index 72fe6cd..0ea00832 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1306,241 +1306,6 @@ compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, int return do_sys_open(dfd, filename, flags, mode); } -/* - * compat_count() counts the number of arguments/envelopes. It is basically - * a copy of count() from fs/exec.c, except that it works with 32 bit argv - * and envp pointers. - */ -static int compat_count(compat_uptr_t __user *argv, int max) -{ - int i = 0; - - if (argv != NULL) { - for (;;) { - compat_uptr_t p; - - if (get_user(p, argv)) - return -EFAULT; - if (!p) - break; - argv++; - if (i++ >= max) - return -E2BIG; - - if (fatal_signal_pending(current)) - return -ERESTARTNOHAND; - cond_resched(); - } - } - return i; -} - -/* - * compat_copy_strings() is basically a copy of copy_strings() from fs/exec.c - * except that it works with 32 bit argv and envp pointers. - */ -static int compat_copy_strings(int argc, compat_uptr_t __user *argv, - struct linux_binprm *bprm) -{ - struct page *kmapped_page = NULL; - char *kaddr = NULL; - unsigned long kpos = 0; - int ret; - - while (argc-- > 0) { - compat_uptr_t str; - int len; - unsigned long pos; - - if (get_user(str, argv+argc) || - !(len = strnlen_user(compat_ptr(str), MAX_ARG_STRLEN))) { - ret = -EFAULT; - goto out; - } - - if (len > MAX_ARG_STRLEN) { - ret = -E2BIG; - goto out; - } - - /* We're going to work our way backwords. */ - pos = bprm->p; - str += len; - bprm->p -= len; - - while (len > 0) { - int offset, bytes_to_copy; - - if (fatal_signal_pending(current)) { - ret = -ERESTARTNOHAND; - goto out; - } - cond_resched(); - - offset = pos % PAGE_SIZE; - if (offset == 0) - offset = PAGE_SIZE; - - bytes_to_copy = offset; - if (bytes_to_copy > len) - bytes_to_copy = len; - - offset -= bytes_to_copy; - pos -= bytes_to_copy; - str -= bytes_to_copy; - len -= bytes_to_copy; - - if (!kmapped_page || kpos != (pos & PAGE_MASK)) { - struct page *page; - - page = get_arg_page(bprm, pos, 1); - if (!page) { - ret = -E2BIG; - goto out; - } - - if (kmapped_page) { - flush_kernel_dcache_page(kmapped_page); - kunmap(kmapped_page); - put_page(kmapped_page); - } - kmapped_page = page; - kaddr = kmap(kmapped_page); - kpos = pos & PAGE_MASK; - flush_cache_page(bprm->vma, kpos, - page_to_pfn(kmapped_page)); - } - if (copy_from_user(kaddr+offset, compat_ptr(str), - bytes_to_copy)) { - ret = -EFAULT; - goto out; - } - } - } - ret = 0; -out: - if (kmapped_page) { - flush_kernel_dcache_page(kmapped_page); - kunmap(kmapped_page); - put_page(kmapped_page); - } - return ret; -} - -/* - * compat_do_execve() is mostly a copy of do_execve(), with the exception - * that it processes 32 bit argv and envp pointers. - */ -int compat_do_execve(char * filename, - compat_uptr_t __user *argv, - compat_uptr_t __user *envp, - struct pt_regs * regs) -{ - struct linux_binprm *bprm; - struct file *file; - struct files_struct *displaced; - bool clear_in_exec; - int retval; - - retval = unshare_files(&displaced); - if (retval) - goto out_ret; - - retval = -ENOMEM; - bprm = kzalloc(sizeof(*bprm), GFP_KERNEL); - if (!bprm) - goto out_files; - - retval = prepare_bprm_creds(bprm); - if (retval) - goto out_free; - - retval = check_unsafe_exec(bprm); - if (retval < 0) - goto out_free; - clear_in_exec = retval; - current->in_execve = 1; - - file = open_exec(filename); - retval = PTR_ERR(file); - if (IS_ERR(file)) - goto out_unmark; - - sched_exec(); - - bprm->file = file; - bprm->filename = filename; - bprm->interp = filename; - - retval = bprm_mm_init(bprm); - if (retval) - goto out_file; - - bprm->argc = compat_count(argv, MAX_ARG_STRINGS); - if ((retval = bprm->argc) < 0) - goto out; - - bprm->envc = compat_count(envp, MAX_ARG_STRINGS); - if ((retval = bprm->envc) < 0) - goto out; - - retval = prepare_binprm(bprm); - if (retval < 0) - goto out; - - retval = copy_strings_kernel(1, &bprm->filename, bprm); - if (retval < 0) - goto out; - - bprm->exec = bprm->p; - retval = compat_copy_strings(bprm->envc, envp, bprm); - if (retval < 0) - goto out; - - retval = compat_copy_strings(bprm->argc, argv, bprm); - if (retval < 0) - goto out; - - retval = search_binary_handler(bprm, regs); - if (retval < 0) - goto out; - - /* execve succeeded */ - current->fs->in_exec = 0; - current->in_execve = 0; - acct_update_integrals(current); - free_bprm(bprm); - if (displaced) - put_files_struct(displaced); - return retval; - -out: - if (bprm->mm) { - acct_arg_size(bprm, 0); - mmput(bprm->mm); - } - -out_file: - if (bprm->file) { - allow_write_access(bprm->file); - fput(bprm->file); - } - -out_unmark: - if (clear_in_exec) - current->fs->in_exec = 0; - current->in_execve = 0; - -out_free: - free_bprm(bprm); - -out_files: - if (displaced) - reset_files_struct(displaced); -out_ret: - return retval; -} - #define __COMPAT_NFDBITS (8 * sizeof(compat_ulong_t)) static int poll_select_copy_remaining(struct timespec *end_time, void __user *p, diff --git a/fs/dcache.c b/fs/dcache.c index 22a0ef4..37f72ee 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -35,6 +35,7 @@ #include <linux/hardirq.h> #include <linux/bit_spinlock.h> #include <linux/rculist_bl.h> +#include <linux/prefetch.h> #include "internal.h" /* @@ -1219,7 +1220,7 @@ void shrink_dcache_parent(struct dentry * parent) EXPORT_SYMBOL(shrink_dcache_parent); /* - * Scan `nr' dentries and return the number which remain. + * Scan `sc->nr_slab_to_reclaim' dentries and return the number which remain. * * We need to avoid reentering the filesystem if the caller is performing a * GFP_NOFS allocation attempt. One example deadlock is: @@ -1230,8 +1231,12 @@ EXPORT_SYMBOL(shrink_dcache_parent); * * In this case we return -1 to tell the caller that we baled. */ -static int shrink_dcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) +static int shrink_dcache_memory(struct shrinker *shrink, + struct shrink_control *sc) { + int nr = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; + if (nr) { if (!(gfp_mask & __GFP_FS)) return -1; diff --git a/fs/dcookies.c b/fs/dcookies.c index a21cabd..dda0dc7 100644 --- a/fs/dcookies.c +++ b/fs/dcookies.c @@ -178,6 +178,8 @@ SYSCALL_DEFINE(lookup_dcookie)(u64 cookie64, char __user * buf, size_t len) /* FIXME: (deleted) ? */ path = d_path(&dcs->path, kbuf, PAGE_SIZE); + mutex_unlock(&dcookie_mutex); + if (IS_ERR(path)) { err = PTR_ERR(path); goto out_free; @@ -194,6 +196,7 @@ SYSCALL_DEFINE(lookup_dcookie)(u64 cookie64, char __user * buf, size_t len) out_free: kfree(kbuf); + return err; out: mutex_unlock(&dcookie_mutex); return err; diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 89d394d..90f7657 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -428,26 +428,17 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { char buf[32]; - int buf_size; + size_t buf_size; + bool bv; u32 *val = file->private_data; buf_size = min(count, (sizeof(buf)-1)); if (copy_from_user(buf, user_buf, buf_size)) return -EFAULT; - switch (buf[0]) { - case 'y': - case 'Y': - case '1': - *val = 1; - break; - case 'n': - case 'N': - case '0': - *val = 0; - break; - } - + if (strtobool(buf, &bv) == 0) + *val = bv; + return count; } diff --git a/fs/dlm/config.c b/fs/dlm/config.c index 0d329ff..9b026ea 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -100,6 +100,7 @@ struct dlm_cluster { unsigned int cl_log_debug; unsigned int cl_protocol; unsigned int cl_timewarn_cs; + unsigned int cl_waitwarn_us; }; enum { @@ -114,6 +115,7 @@ enum { CLUSTER_ATTR_LOG_DEBUG, CLUSTER_ATTR_PROTOCOL, CLUSTER_ATTR_TIMEWARN_CS, + CLUSTER_ATTR_WAITWARN_US, }; struct cluster_attribute { @@ -166,6 +168,7 @@ CLUSTER_ATTR(scan_secs, 1); CLUSTER_ATTR(log_debug, 0); CLUSTER_ATTR(protocol, 0); CLUSTER_ATTR(timewarn_cs, 1); +CLUSTER_ATTR(waitwarn_us, 0); static struct configfs_attribute *cluster_attrs[] = { [CLUSTER_ATTR_TCP_PORT] = &cluster_attr_tcp_port.attr, @@ -179,6 +182,7 @@ static struct configfs_attribute *cluster_attrs[] = { [CLUSTER_ATTR_LOG_DEBUG] = &cluster_attr_log_debug.attr, [CLUSTER_ATTR_PROTOCOL] = &cluster_attr_protocol.attr, [CLUSTER_ATTR_TIMEWARN_CS] = &cluster_attr_timewarn_cs.attr, + [CLUSTER_ATTR_WAITWARN_US] = &cluster_attr_waitwarn_us.attr, NULL, }; @@ -439,6 +443,7 @@ static struct config_group *make_cluster(struct config_group *g, cl->cl_log_debug = dlm_config.ci_log_debug; cl->cl_protocol = dlm_config.ci_protocol; cl->cl_timewarn_cs = dlm_config.ci_timewarn_cs; + cl->cl_waitwarn_us = dlm_config.ci_waitwarn_us; space_list = &sps->ss_group; comm_list = &cms->cs_group; @@ -986,6 +991,7 @@ int dlm_our_addr(struct sockaddr_storage *addr, int num) #define DEFAULT_LOG_DEBUG 0 #define DEFAULT_PROTOCOL 0 #define DEFAULT_TIMEWARN_CS 500 /* 5 sec = 500 centiseconds */ +#define DEFAULT_WAITWARN_US 0 struct dlm_config_info dlm_config = { .ci_tcp_port = DEFAULT_TCP_PORT, @@ -998,6 +1004,7 @@ struct dlm_config_info dlm_config = { .ci_scan_secs = DEFAULT_SCAN_SECS, .ci_log_debug = DEFAULT_LOG_DEBUG, .ci_protocol = DEFAULT_PROTOCOL, - .ci_timewarn_cs = DEFAULT_TIMEWARN_CS + .ci_timewarn_cs = DEFAULT_TIMEWARN_CS, + .ci_waitwarn_us = DEFAULT_WAITWARN_US }; diff --git a/fs/dlm/config.h b/fs/dlm/config.h index 4f1d6fc..dd0ce24 100644 --- a/fs/dlm/config.h +++ b/fs/dlm/config.h @@ -28,6 +28,7 @@ struct dlm_config_info { int ci_log_debug; int ci_protocol; int ci_timewarn_cs; + int ci_waitwarn_us; }; extern struct dlm_config_info dlm_config; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index b942049..0262451 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -209,6 +209,7 @@ struct dlm_args { #define DLM_IFL_WATCH_TIMEWARN 0x00400000 #define DLM_IFL_TIMEOUT_CANCEL 0x00800000 #define DLM_IFL_DEADLOCK_CANCEL 0x01000000 +#define DLM_IFL_STUB_MS 0x02000000 /* magic number for m_flags */ #define DLM_IFL_USER 0x00000001 #define DLM_IFL_ORPHAN 0x00000002 @@ -245,6 +246,7 @@ struct dlm_lkb { int8_t lkb_wait_type; /* type of reply waiting for */ int8_t lkb_wait_count; + int lkb_wait_nodeid; /* for debugging */ struct list_head lkb_idtbl_list; /* lockspace lkbtbl */ struct list_head lkb_statequeue; /* rsb g/c/w list */ @@ -254,6 +256,7 @@ struct dlm_lkb { struct list_head lkb_ownqueue; /* list of locks for a process */ struct list_head lkb_time_list; ktime_t lkb_timestamp; + ktime_t lkb_wait_time; unsigned long lkb_timeout_cs; struct dlm_callback lkb_callbacks[DLM_CALLBACKS_SIZE]; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 56d6bfc..f71d0b5 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -799,10 +799,84 @@ static int msg_reply_type(int mstype) return -1; } +static int nodeid_warned(int nodeid, int num_nodes, int *warned) +{ + int i; + + for (i = 0; i < num_nodes; i++) { + if (!warned[i]) { + warned[i] = nodeid; + return 0; + } + if (warned[i] == nodeid) + return 1; + } + return 0; +} + +void dlm_scan_waiters(struct dlm_ls *ls) +{ + struct dlm_lkb *lkb; + ktime_t zero = ktime_set(0, 0); + s64 us; + s64 debug_maxus = 0; + u32 debug_scanned = 0; + u32 debug_expired = 0; + int num_nodes = 0; + int *warned = NULL; + + if (!dlm_config.ci_waitwarn_us) + return; + + mutex_lock(&ls->ls_waiters_mutex); + + list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) { + if (ktime_equal(lkb->lkb_wait_time, zero)) + continue; + + debug_scanned++; + + us = ktime_to_us(ktime_sub(ktime_get(), lkb->lkb_wait_time)); + + if (us < dlm_config.ci_waitwarn_us) + continue; + + lkb->lkb_wait_time = zero; + + debug_expired++; + if (us > debug_maxus) + debug_maxus = us; + + if (!num_nodes) { + num_nodes = ls->ls_num_nodes; + warned = kmalloc(GFP_KERNEL, num_nodes * sizeof(int)); + if (warned) + memset(warned, 0, num_nodes * sizeof(int)); + } + if (!warned) + continue; + if (nodeid_warned(lkb->lkb_wait_nodeid, num_nodes, warned)) + continue; + + log_error(ls, "waitwarn %x %lld %d us check connection to " + "node %d", lkb->lkb_id, (long long)us, + dlm_config.ci_waitwarn_us, lkb->lkb_wait_nodeid); + } + mutex_unlock(&ls->ls_waiters_mutex); + + if (warned) + kfree(warned); + + if (debug_expired) + log_debug(ls, "scan_waiters %u warn %u over %d us max %lld us", + debug_scanned, debug_expired, + dlm_config.ci_waitwarn_us, (long long)debug_maxus); +} + /* add/remove lkb from global waiters list of lkb's waiting for a reply from a remote node */ -static int add_to_waiters(struct dlm_lkb *lkb, int mstype) +static int add_to_waiters(struct dlm_lkb *lkb, int mstype, int to_nodeid) { struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error = 0; @@ -842,6 +916,8 @@ static int add_to_waiters(struct dlm_lkb *lkb, int mstype) lkb->lkb_wait_count++; lkb->lkb_wait_type = mstype; + lkb->lkb_wait_time = ktime_get(); + lkb->lkb_wait_nodeid = to_nodeid; /* for debugging */ hold_lkb(lkb); list_add(&lkb->lkb_wait_reply, &ls->ls_waiters); out: @@ -961,10 +1037,10 @@ static int remove_from_waiters_ms(struct dlm_lkb *lkb, struct dlm_message *ms) struct dlm_ls *ls = lkb->lkb_resource->res_ls; int error; - if (ms != &ls->ls_stub_ms) + if (ms->m_flags != DLM_IFL_STUB_MS) mutex_lock(&ls->ls_waiters_mutex); error = _remove_from_waiters(lkb, ms->m_type, ms); - if (ms != &ls->ls_stub_ms) + if (ms->m_flags != DLM_IFL_STUB_MS) mutex_unlock(&ls->ls_waiters_mutex); return error; } @@ -1157,6 +1233,16 @@ void dlm_adjust_timeouts(struct dlm_ls *ls) list_for_each_entry(lkb, &ls->ls_timeout, lkb_time_list) lkb->lkb_timestamp = ktime_add_us(lkb->lkb_timestamp, adj_us); mutex_unlock(&ls->ls_timeout_mutex); + + if (!dlm_config.ci_waitwarn_us) + return; + + mutex_lock(&ls->ls_waiters_mutex); + list_for_each_entry(lkb, &ls->ls_waiters, lkb_wait_reply) { + if (ktime_to_us(lkb->lkb_wait_time)) + lkb->lkb_wait_time = ktime_get(); + } + mutex_unlock(&ls->ls_waiters_mutex); } /* lkb is master or local copy */ @@ -1376,14 +1462,8 @@ static void grant_lock_pending(struct dlm_rsb *r, struct dlm_lkb *lkb) ALTPR/ALTCW: our rqmode may have been changed to PR or CW to become compatible with other granted locks */ -static void munge_demoted(struct dlm_lkb *lkb, struct dlm_message *ms) +static void munge_demoted(struct dlm_lkb *lkb) { - if (ms->m_type != DLM_MSG_CONVERT_REPLY) { - log_print("munge_demoted %x invalid reply type %d", - lkb->lkb_id, ms->m_type); - return; - } - if (lkb->lkb_rqmode == DLM_LOCK_IV || lkb->lkb_grmode == DLM_LOCK_IV) { log_print("munge_demoted %x invalid modes gr %d rq %d", lkb->lkb_id, lkb->lkb_grmode, lkb->lkb_rqmode); @@ -2844,12 +2924,12 @@ static int send_common(struct dlm_rsb *r, struct dlm_lkb *lkb, int mstype) struct dlm_mhandle *mh; int to_nodeid, error; - error = add_to_waiters(lkb, mstype); + to_nodeid = r->res_nodeid; + + error = add_to_waiters(lkb, mstype, to_nodeid); if (error) return error; - to_nodeid = r->res_nodeid; - error = create_message(r, lkb, to_nodeid, mstype, &ms, &mh); if (error) goto fail; @@ -2880,9 +2960,9 @@ static int send_convert(struct dlm_rsb *r, struct dlm_lkb *lkb) /* down conversions go without a reply from the master */ if (!error && down_conversion(lkb)) { remove_from_waiters(lkb, DLM_MSG_CONVERT_REPLY); + r->res_ls->ls_stub_ms.m_flags = DLM_IFL_STUB_MS; r->res_ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY; r->res_ls->ls_stub_ms.m_result = 0; - r->res_ls->ls_stub_ms.m_flags = lkb->lkb_flags; __receive_convert_reply(r, lkb, &r->res_ls->ls_stub_ms); } @@ -2951,12 +3031,12 @@ static int send_lookup(struct dlm_rsb *r, struct dlm_lkb *lkb) struct dlm_mhandle *mh; int to_nodeid, error; - error = add_to_waiters(lkb, DLM_MSG_LOOKUP); + to_nodeid = dlm_dir_nodeid(r); + + error = add_to_waiters(lkb, DLM_MSG_LOOKUP, to_nodeid); if (error) return error; - to_nodeid = dlm_dir_nodeid(r); - error = create_message(r, NULL, to_nodeid, DLM_MSG_LOOKUP, &ms, &mh); if (error) goto fail; @@ -3070,6 +3150,9 @@ static void receive_flags(struct dlm_lkb *lkb, struct dlm_message *ms) static void receive_flags_reply(struct dlm_lkb *lkb, struct dlm_message *ms) { + if (ms->m_flags == DLM_IFL_STUB_MS) + return; + lkb->lkb_sbflags = ms->m_sbflags; lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) | (ms->m_flags & 0x0000FFFF); @@ -3612,7 +3695,7 @@ static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, /* convert was queued on remote master */ receive_flags_reply(lkb, ms); if (is_demoted(lkb)) - munge_demoted(lkb, ms); + munge_demoted(lkb); del_lkb(r, lkb); add_lkb(r, lkb, DLM_LKSTS_CONVERT); add_timeout(lkb); @@ -3622,7 +3705,7 @@ static void __receive_convert_reply(struct dlm_rsb *r, struct dlm_lkb *lkb, /* convert was granted on remote master */ receive_flags_reply(lkb, ms); if (is_demoted(lkb)) - munge_demoted(lkb, ms); + munge_demoted(lkb); grant_lock_pc(r, lkb, ms); queue_cast(r, lkb, 0); break; @@ -3996,15 +4079,17 @@ void dlm_receive_buffer(union dlm_packet *p, int nodeid) dlm_put_lockspace(ls); } -static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb) +static void recover_convert_waiter(struct dlm_ls *ls, struct dlm_lkb *lkb, + struct dlm_message *ms_stub) { if (middle_conversion(lkb)) { hold_lkb(lkb); - ls->ls_stub_ms.m_type = DLM_MSG_CONVERT_REPLY; - ls->ls_stub_ms.m_result = -EINPROGRESS; - ls->ls_stub_ms.m_flags = lkb->lkb_flags; - ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; - _receive_convert_reply(lkb, &ls->ls_stub_ms); + memset(ms_stub, 0, sizeof(struct dlm_message)); + ms_stub->m_flags = DLM_IFL_STUB_MS; + ms_stub->m_type = DLM_MSG_CONVERT_REPLY; + ms_stub->m_result = -EINPROGRESS; + ms_stub->m_header.h_nodeid = lkb->lkb_nodeid; + _receive_convert_reply(lkb, ms_stub); /* Same special case as in receive_rcom_lock_args() */ lkb->lkb_grmode = DLM_LOCK_IV; @@ -4045,13 +4130,27 @@ static int waiter_needs_recovery(struct dlm_ls *ls, struct dlm_lkb *lkb) void dlm_recover_waiters_pre(struct dlm_ls *ls) { struct dlm_lkb *lkb, *safe; + struct dlm_message *ms_stub; int wait_type, stub_unlock_result, stub_cancel_result; + ms_stub = kmalloc(GFP_KERNEL, sizeof(struct dlm_message)); + if (!ms_stub) { + log_error(ls, "dlm_recover_waiters_pre no mem"); + return; + } + mutex_lock(&ls->ls_waiters_mutex); list_for_each_entry_safe(lkb, safe, &ls->ls_waiters, lkb_wait_reply) { - log_debug(ls, "pre recover waiter lkid %x type %d flags %x", - lkb->lkb_id, lkb->lkb_wait_type, lkb->lkb_flags); + + /* exclude debug messages about unlocks because there can be so + many and they aren't very interesting */ + + if (lkb->lkb_wait_type != DLM_MSG_UNLOCK) { + log_debug(ls, "recover_waiter %x nodeid %d " + "msg %d to %d", lkb->lkb_id, lkb->lkb_nodeid, + lkb->lkb_wait_type, lkb->lkb_wait_nodeid); + } /* all outstanding lookups, regardless of destination will be resent after recovery is done */ @@ -4097,26 +4196,28 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) break; case DLM_MSG_CONVERT: - recover_convert_waiter(ls, lkb); + recover_convert_waiter(ls, lkb, ms_stub); break; case DLM_MSG_UNLOCK: hold_lkb(lkb); - ls->ls_stub_ms.m_type = DLM_MSG_UNLOCK_REPLY; - ls->ls_stub_ms.m_result = stub_unlock_result; - ls->ls_stub_ms.m_flags = lkb->lkb_flags; - ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; - _receive_unlock_reply(lkb, &ls->ls_stub_ms); + memset(ms_stub, 0, sizeof(struct dlm_message)); + ms_stub->m_flags = DLM_IFL_STUB_MS; + ms_stub->m_type = DLM_MSG_UNLOCK_REPLY; + ms_stub->m_result = stub_unlock_result; + ms_stub->m_header.h_nodeid = lkb->lkb_nodeid; + _receive_unlock_reply(lkb, ms_stub); dlm_put_lkb(lkb); break; case DLM_MSG_CANCEL: hold_lkb(lkb); - ls->ls_stub_ms.m_type = DLM_MSG_CANCEL_REPLY; - ls->ls_stub_ms.m_result = stub_cancel_result; - ls->ls_stub_ms.m_flags = lkb->lkb_flags; - ls->ls_stub_ms.m_header.h_nodeid = lkb->lkb_nodeid; - _receive_cancel_reply(lkb, &ls->ls_stub_ms); + memset(ms_stub, 0, sizeof(struct dlm_message)); + ms_stub->m_flags = DLM_IFL_STUB_MS; + ms_stub->m_type = DLM_MSG_CANCEL_REPLY; + ms_stub->m_result = stub_cancel_result; + ms_stub->m_header.h_nodeid = lkb->lkb_nodeid; + _receive_cancel_reply(lkb, ms_stub); dlm_put_lkb(lkb); break; @@ -4127,6 +4228,7 @@ void dlm_recover_waiters_pre(struct dlm_ls *ls) schedule(); } mutex_unlock(&ls->ls_waiters_mutex); + kfree(ms_stub); } static struct dlm_lkb *find_resend_waiter(struct dlm_ls *ls) @@ -4191,8 +4293,8 @@ int dlm_recover_waiters_post(struct dlm_ls *ls) ou = is_overlap_unlock(lkb); err = 0; - log_debug(ls, "recover_waiters_post %x type %d flags %x %s", - lkb->lkb_id, mstype, lkb->lkb_flags, r->res_name); + log_debug(ls, "recover_waiter %x nodeid %d msg %d r_nodeid %d", + lkb->lkb_id, lkb->lkb_nodeid, mstype, r->res_nodeid); /* At this point we assume that we won't get a reply to any previous op or overlap op on this lock. First, do a big diff --git a/fs/dlm/lock.h b/fs/dlm/lock.h index 88e93c8..265017a 100644 --- a/fs/dlm/lock.h +++ b/fs/dlm/lock.h @@ -24,6 +24,7 @@ int dlm_put_lkb(struct dlm_lkb *lkb); void dlm_scan_rsbs(struct dlm_ls *ls); int dlm_lock_recovery_try(struct dlm_ls *ls); void dlm_unlock_recovery(struct dlm_ls *ls); +void dlm_scan_waiters(struct dlm_ls *ls); void dlm_scan_timeout(struct dlm_ls *ls); void dlm_adjust_timeouts(struct dlm_ls *ls); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index f994a7d..14cbf40 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -243,7 +243,6 @@ static struct dlm_ls *find_ls_to_scan(void) static int dlm_scand(void *data) { struct dlm_ls *ls; - int timeout_jiffies = dlm_config.ci_scan_secs * HZ; while (!kthread_should_stop()) { ls = find_ls_to_scan(); @@ -252,13 +251,14 @@ static int dlm_scand(void *data) ls->ls_scan_time = jiffies; dlm_scan_rsbs(ls); dlm_scan_timeout(ls); + dlm_scan_waiters(ls); dlm_unlock_recovery(ls); } else { ls->ls_scan_time += HZ; } - } else { - schedule_timeout_interruptible(timeout_jiffies); + continue; } + schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ); } return 0; } diff --git a/fs/dlm/main.c b/fs/dlm/main.c index b80e0aa..5a59efa 100644 --- a/fs/dlm/main.c +++ b/fs/dlm/main.c @@ -50,7 +50,7 @@ static int __init init_dlm(void) if (error) goto out_netlink; - printk("DLM (built %s %s) installed\n", __DATE__, __TIME__); + printk("DLM installed\n"); return 0; diff --git a/fs/dlm/plock.c b/fs/dlm/plock.c index 30d8b85..e2b8780 100644 --- a/fs/dlm/plock.c +++ b/fs/dlm/plock.c @@ -71,6 +71,36 @@ static void send_op(struct plock_op *op) wake_up(&send_wq); } +/* If a process was killed while waiting for the only plock on a file, + locks_remove_posix will not see any lock on the file so it won't + send an unlock-close to us to pass on to userspace to clean up the + abandoned waiter. So, we have to insert the unlock-close when the + lock call is interrupted. */ + +static void do_unlock_close(struct dlm_ls *ls, u64 number, + struct file *file, struct file_lock *fl) +{ + struct plock_op *op; + + op = kzalloc(sizeof(*op), GFP_NOFS); + if (!op) + return; + + op->info.optype = DLM_PLOCK_OP_UNLOCK; + op->info.pid = fl->fl_pid; + op->info.fsid = ls->ls_global_id; + op->info.number = number; + op->info.start = 0; + op->info.end = OFFSET_MAX; + if (fl->fl_lmops && fl->fl_lmops->fl_grant) + op->info.owner = (__u64) fl->fl_pid; + else + op->info.owner = (__u64)(long) fl->fl_owner; + + op->info.flags |= DLM_PLOCK_FL_CLOSE; + send_op(op); +} + int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file, int cmd, struct file_lock *fl) { @@ -114,9 +144,19 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file, send_op(op); - if (xop->callback == NULL) - wait_event(recv_wq, (op->done != 0)); - else { + if (xop->callback == NULL) { + rv = wait_event_killable(recv_wq, (op->done != 0)); + if (rv == -ERESTARTSYS) { + log_debug(ls, "dlm_posix_lock: wait killed %llx", + (unsigned long long)number); + spin_lock(&ops_lock); + list_del(&op->list); + spin_unlock(&ops_lock); + kfree(xop); + do_unlock_close(ls, number, file, fl); + goto out; + } + } else { rv = FILE_LOCK_DEFERRED; goto out; } @@ -233,6 +273,13 @@ int dlm_posix_unlock(dlm_lockspace_t *lockspace, u64 number, struct file *file, else op->info.owner = (__u64)(long) fl->fl_owner; + if (fl->fl_flags & FL_CLOSE) { + op->info.flags |= DLM_PLOCK_FL_CLOSE; + send_op(op); + rv = 0; + goto out; + } + send_op(op); wait_event(recv_wq, (op->done != 0)); @@ -334,7 +381,10 @@ static ssize_t dev_read(struct file *file, char __user *u, size_t count, spin_lock(&ops_lock); if (!list_empty(&send_list)) { op = list_entry(send_list.next, struct plock_op, list); - list_move(&op->list, &recv_list); + if (op->info.flags & DLM_PLOCK_FL_CLOSE) + list_del(&op->list); + else + list_move(&op->list, &recv_list); memcpy(&info, &op->info, sizeof(info)); } spin_unlock(&ops_lock); @@ -342,6 +392,13 @@ static ssize_t dev_read(struct file *file, char __user *u, size_t count, if (!op) return -EAGAIN; + /* there is no need to get a reply from userspace for unlocks + that were generated by the vfs cleaning up for a close + (the process did not make an unlock call). */ + + if (op->info.flags & DLM_PLOCK_FL_CLOSE) + kfree(op); + if (copy_to_user(u, &info, sizeof(info))) return -EFAULT; return sizeof(info); diff --git a/fs/dlm/user.c b/fs/dlm/user.c index d5ab3fe..e96bf3e 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -611,7 +611,6 @@ static ssize_t device_write(struct file *file, const char __user *buf, out_sig: sigprocmask(SIG_SETMASK, &tmpsig, NULL); - recalc_sigpending(); out_free: kfree(kbuf); return error; diff --git a/fs/drop_caches.c b/fs/drop_caches.c index 98b77c8..c00e055 100644 --- a/fs/drop_caches.c +++ b/fs/drop_caches.c @@ -40,9 +40,12 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused) static void drop_slab(void) { int nr_objects; + struct shrink_control shrink = { + .gfp_mask = GFP_KERNEL, + }; do { - nr_objects = shrink_slab(1000, GFP_KERNEL, 1000); + nr_objects = shrink_slab(&shrink, 1000, 1000); } while (nr_objects > 10); } diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index b8d5c80..58609bd 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1024,25 +1024,25 @@ out: } /** - * contains_ecryptfs_marker - check for the ecryptfs marker + * ecryptfs_validate_marker - check for the ecryptfs marker * @data: The data block in which to check * - * Returns one if marker found; zero if not found + * Returns zero if marker found; -EINVAL if not found */ -static int contains_ecryptfs_marker(char *data) +static int ecryptfs_validate_marker(char *data) { u32 m_1, m_2; m_1 = get_unaligned_be32(data); m_2 = get_unaligned_be32(data + 4); if ((m_1 ^ MAGIC_ECRYPTFS_MARKER) == m_2) - return 1; + return 0; ecryptfs_printk(KERN_DEBUG, "m_1 = [0x%.8x]; m_2 = [0x%.8x]; " "MAGIC_ECRYPTFS_MARKER = [0x%.8x]\n", m_1, m_2, MAGIC_ECRYPTFS_MARKER); ecryptfs_printk(KERN_DEBUG, "(m_1 ^ MAGIC_ECRYPTFS_MARKER) = " "[0x%.8x]\n", (m_1 ^ MAGIC_ECRYPTFS_MARKER)); - return 0; + return -EINVAL; } struct ecryptfs_flag_map_elem { @@ -1201,27 +1201,19 @@ int ecryptfs_cipher_code_to_string(char *str, u8 cipher_code) return rc; } -int ecryptfs_read_and_validate_header_region(char *data, - struct inode *ecryptfs_inode) +int ecryptfs_read_and_validate_header_region(struct inode *inode) { - struct ecryptfs_crypt_stat *crypt_stat = - &(ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat); + u8 file_size[ECRYPTFS_SIZE_AND_MARKER_BYTES]; + u8 *marker = file_size + ECRYPTFS_FILE_SIZE_BYTES; int rc; - if (crypt_stat->extent_size == 0) - crypt_stat->extent_size = ECRYPTFS_DEFAULT_EXTENT_SIZE; - rc = ecryptfs_read_lower(data, 0, crypt_stat->extent_size, - ecryptfs_inode); - if (rc < 0) { - printk(KERN_ERR "%s: Error reading header region; rc = [%d]\n", - __func__, rc); - goto out; - } - if (!contains_ecryptfs_marker(data + ECRYPTFS_FILE_SIZE_BYTES)) { - rc = -EINVAL; - } else - rc = 0; -out: + rc = ecryptfs_read_lower(file_size, 0, ECRYPTFS_SIZE_AND_MARKER_BYTES, + inode); + if (rc < ECRYPTFS_SIZE_AND_MARKER_BYTES) + return rc >= 0 ? -EINVAL : rc; + rc = ecryptfs_validate_marker(marker); + if (!rc) + ecryptfs_i_size_init(file_size, inode); return rc; } @@ -1242,8 +1234,7 @@ ecryptfs_write_header_metadata(char *virt, (*written) = 6; } -struct kmem_cache *ecryptfs_header_cache_1; -struct kmem_cache *ecryptfs_header_cache_2; +struct kmem_cache *ecryptfs_header_cache; /** * ecryptfs_write_headers_virt @@ -1496,11 +1487,9 @@ static int ecryptfs_read_headers_virt(char *page_virt, crypt_stat->mount_crypt_stat = &ecryptfs_superblock_to_private( ecryptfs_dentry->d_sb)->mount_crypt_stat; offset = ECRYPTFS_FILE_SIZE_BYTES; - rc = contains_ecryptfs_marker(page_virt + offset); - if (rc == 0) { - rc = -EINVAL; + rc = ecryptfs_validate_marker(page_virt + offset); + if (rc) goto out; - } if (!(crypt_stat->flags & ECRYPTFS_I_SIZE_INITIALIZED)) ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode); offset += MAGIC_ECRYPTFS_MARKER_SIZE_BYTES; @@ -1567,20 +1556,21 @@ out: return rc; } -int ecryptfs_read_and_validate_xattr_region(char *page_virt, - struct dentry *ecryptfs_dentry) +int ecryptfs_read_and_validate_xattr_region(struct dentry *dentry, + struct inode *inode) { + u8 file_size[ECRYPTFS_SIZE_AND_MARKER_BYTES]; + u8 *marker = file_size + ECRYPTFS_FILE_SIZE_BYTES; int rc; - rc = ecryptfs_read_xattr_region(page_virt, ecryptfs_dentry->d_inode); - if (rc) - goto out; - if (!contains_ecryptfs_marker(page_virt + ECRYPTFS_FILE_SIZE_BYTES)) { - printk(KERN_WARNING "Valid data found in [%s] xattr, but " - "the marker is invalid\n", ECRYPTFS_XATTR_NAME); - rc = -EINVAL; - } -out: + rc = ecryptfs_getxattr_lower(ecryptfs_dentry_to_lower(dentry), + ECRYPTFS_XATTR_NAME, file_size, + ECRYPTFS_SIZE_AND_MARKER_BYTES); + if (rc < ECRYPTFS_SIZE_AND_MARKER_BYTES) + return rc >= 0 ? -EINVAL : rc; + rc = ecryptfs_validate_marker(marker); + if (!rc) + ecryptfs_i_size_init(file_size, inode); return rc; } @@ -1610,7 +1600,7 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) ecryptfs_copy_mount_wide_flags_to_inode_flags(crypt_stat, mount_crypt_stat); /* Read the first page from the underlying file */ - page_virt = kmem_cache_alloc(ecryptfs_header_cache_1, GFP_USER); + page_virt = kmem_cache_alloc(ecryptfs_header_cache, GFP_USER); if (!page_virt) { rc = -ENOMEM; printk(KERN_ERR "%s: Unable to allocate page_virt\n", @@ -1655,7 +1645,7 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) out: if (page_virt) { memset(page_virt, 0, PAGE_CACHE_SIZE); - kmem_cache_free(ecryptfs_header_cache_1, page_virt); + kmem_cache_free(ecryptfs_header_cache, page_virt); } return rc; } diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index e702827..43c7c43 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -200,6 +200,8 @@ ecryptfs_get_key_payload_data(struct key *key) #define MAGIC_ECRYPTFS_MARKER 0x3c81b7f5 #define MAGIC_ECRYPTFS_MARKER_SIZE_BYTES 8 /* 4*2 */ #define ECRYPTFS_FILE_SIZE_BYTES (sizeof(u64)) +#define ECRYPTFS_SIZE_AND_MARKER_BYTES (ECRYPTFS_FILE_SIZE_BYTES \ + + MAGIC_ECRYPTFS_MARKER_SIZE_BYTES) #define ECRYPTFS_DEFAULT_CIPHER "aes" #define ECRYPTFS_DEFAULT_KEY_BYTES 16 #define ECRYPTFS_DEFAULT_HASH "md5" @@ -603,8 +605,7 @@ extern struct kmem_cache *ecryptfs_file_info_cache; extern struct kmem_cache *ecryptfs_dentry_info_cache; extern struct kmem_cache *ecryptfs_inode_info_cache; extern struct kmem_cache *ecryptfs_sb_info_cache; -extern struct kmem_cache *ecryptfs_header_cache_1; -extern struct kmem_cache *ecryptfs_header_cache_2; +extern struct kmem_cache *ecryptfs_header_cache; extern struct kmem_cache *ecryptfs_xattr_cache; extern struct kmem_cache *ecryptfs_key_record_cache; extern struct kmem_cache *ecryptfs_key_sig_cache; @@ -625,14 +626,9 @@ struct ecryptfs_open_req { struct list_head kthread_ctl_list; }; -#define ECRYPTFS_INTERPOSE_FLAG_D_ADD 0x00000001 -int ecryptfs_interpose(struct dentry *hidden_dentry, - struct dentry *this_dentry, struct super_block *sb, - u32 flags); +struct inode *ecryptfs_get_inode(struct inode *lower_inode, + struct super_block *sb); void ecryptfs_i_size_init(const char *page_virt, struct inode *inode); -int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, - struct dentry *lower_dentry, - struct inode *ecryptfs_dir_inode); int ecryptfs_decode_and_decrypt_filename(char **decrypted_name, size_t *decrypted_name_size, struct dentry *ecryptfs_dentry, @@ -664,10 +660,9 @@ int ecryptfs_new_file_context(struct dentry *ecryptfs_dentry); void ecryptfs_write_crypt_stat_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat, size_t *written); -int ecryptfs_read_and_validate_header_region(char *data, - struct inode *ecryptfs_inode); -int ecryptfs_read_and_validate_xattr_region(char *page_virt, - struct dentry *ecryptfs_dentry); +int ecryptfs_read_and_validate_header_region(struct inode *inode); +int ecryptfs_read_and_validate_xattr_region(struct dentry *dentry, + struct inode *inode); u8 ecryptfs_code_for_cipher_string(char *cipher_name, size_t key_bytes); int ecryptfs_cipher_code_to_string(char *str, u8 cipher_code); void ecryptfs_set_default_sizes(struct ecryptfs_crypt_stat *crypt_stat); @@ -679,9 +674,6 @@ int ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, unsigned char *src, struct dentry *ecryptfs_dentry); int ecryptfs_truncate(struct dentry *dentry, loff_t new_length); -int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); -int ecryptfs_inode_set(struct inode *inode, void *lower_inode); -void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode); ssize_t ecryptfs_getxattr_lower(struct dentry *lower_dentry, const char *name, void *value, size_t size); @@ -761,7 +753,7 @@ int ecryptfs_privileged_open(struct file **lower_file, struct dentry *lower_dentry, struct vfsmount *lower_mnt, const struct cred *cred); -int ecryptfs_get_lower_file(struct dentry *ecryptfs_dentry); +int ecryptfs_get_lower_file(struct dentry *dentry, struct inode *inode); void ecryptfs_put_lower_file(struct inode *inode); int ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 566e547..4ec9eb0 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -191,7 +191,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file) | ECRYPTFS_ENCRYPTED); } mutex_unlock(&crypt_stat->cs_mutex); - rc = ecryptfs_get_lower_file(ecryptfs_dentry); + rc = ecryptfs_get_lower_file(ecryptfs_dentry, inode); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " "the lower file for the dentry with name " diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 4d4cc6a..7349ade 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -51,6 +51,97 @@ static void unlock_dir(struct dentry *dir) dput(dir); } +static int ecryptfs_inode_test(struct inode *inode, void *lower_inode) +{ + if (ecryptfs_inode_to_lower(inode) == (struct inode *)lower_inode) + return 1; + return 0; +} + +static int ecryptfs_inode_set(struct inode *inode, void *opaque) +{ + struct inode *lower_inode = opaque; + + ecryptfs_set_inode_lower(inode, lower_inode); + fsstack_copy_attr_all(inode, lower_inode); + /* i_size will be overwritten for encrypted regular files */ + fsstack_copy_inode_size(inode, lower_inode); + inode->i_ino = lower_inode->i_ino; + inode->i_version++; + inode->i_mapping->a_ops = &ecryptfs_aops; + + if (S_ISLNK(inode->i_mode)) + inode->i_op = &ecryptfs_symlink_iops; + else if (S_ISDIR(inode->i_mode)) + inode->i_op = &ecryptfs_dir_iops; + else + inode->i_op = &ecryptfs_main_iops; + + if (S_ISDIR(inode->i_mode)) + inode->i_fop = &ecryptfs_dir_fops; + else if (special_file(inode->i_mode)) + init_special_inode(inode, inode->i_mode, inode->i_rdev); + else + inode->i_fop = &ecryptfs_main_fops; + + return 0; +} + +static struct inode *__ecryptfs_get_inode(struct inode *lower_inode, + struct super_block *sb) +{ + struct inode *inode; + + if (lower_inode->i_sb != ecryptfs_superblock_to_lower(sb)) + return ERR_PTR(-EXDEV); + if (!igrab(lower_inode)) + return ERR_PTR(-ESTALE); + inode = iget5_locked(sb, (unsigned long)lower_inode, + ecryptfs_inode_test, ecryptfs_inode_set, + lower_inode); + if (!inode) { + iput(lower_inode); + return ERR_PTR(-EACCES); + } + if (!(inode->i_state & I_NEW)) + iput(lower_inode); + + return inode; +} + +struct inode *ecryptfs_get_inode(struct inode *lower_inode, + struct super_block *sb) +{ + struct inode *inode = __ecryptfs_get_inode(lower_inode, sb); + + if (!IS_ERR(inode) && (inode->i_state & I_NEW)) + unlock_new_inode(inode); + + return inode; +} + +/** + * ecryptfs_interpose + * @lower_dentry: Existing dentry in the lower filesystem + * @dentry: ecryptfs' dentry + * @sb: ecryptfs's super_block + * + * Interposes upper and lower dentries. + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_interpose(struct dentry *lower_dentry, + struct dentry *dentry, struct super_block *sb) +{ + struct inode *inode = ecryptfs_get_inode(lower_dentry->d_inode, sb); + + if (IS_ERR(inode)) + return PTR_ERR(inode); + d_instantiate(dentry, inode); + + return 0; +} + /** * ecryptfs_create_underlying_file * @lower_dir_inode: inode of the parent in the lower fs of the new file @@ -129,7 +220,7 @@ ecryptfs_do_create(struct inode *directory_inode, goto out_lock; } rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry, - directory_inode->i_sb, 0); + directory_inode->i_sb); if (rc) { ecryptfs_printk(KERN_ERR, "Failure in ecryptfs_interpose\n"); goto out_lock; @@ -168,7 +259,8 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) "context; rc = [%d]\n", rc); goto out; } - rc = ecryptfs_get_lower_file(ecryptfs_dentry); + rc = ecryptfs_get_lower_file(ecryptfs_dentry, + ecryptfs_dentry->d_inode); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " "the lower file for the dentry with name " @@ -215,102 +307,90 @@ out: return rc; } +static int ecryptfs_i_size_read(struct dentry *dentry, struct inode *inode) +{ + struct ecryptfs_crypt_stat *crypt_stat; + int rc; + + rc = ecryptfs_get_lower_file(dentry, inode); + if (rc) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the lower file for the dentry with name " + "[%s]; rc = [%d]\n", __func__, + dentry->d_name.name, rc); + return rc; + } + + crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat; + /* TODO: lock for crypt_stat comparison */ + if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED)) + ecryptfs_set_default_sizes(crypt_stat); + + rc = ecryptfs_read_and_validate_header_region(inode); + ecryptfs_put_lower_file(inode); + if (rc) { + rc = ecryptfs_read_and_validate_xattr_region(dentry, inode); + if (!rc) + crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR; + } + + /* Must return 0 to allow non-eCryptfs files to be looked up, too */ + return 0; +} + /** - * ecryptfs_lookup_and_interpose_lower - Perform a lookup + * ecryptfs_lookup_interpose - Dentry interposition for a lookup */ -int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, - struct dentry *lower_dentry, - struct inode *ecryptfs_dir_inode) +static int ecryptfs_lookup_interpose(struct dentry *dentry, + struct dentry *lower_dentry, + struct inode *dir_inode) { - struct dentry *lower_dir_dentry; + struct inode *inode, *lower_inode = lower_dentry->d_inode; + struct ecryptfs_dentry_info *dentry_info; struct vfsmount *lower_mnt; - struct inode *lower_inode; - struct ecryptfs_crypt_stat *crypt_stat; - char *page_virt = NULL; - int put_lower = 0, rc = 0; - - lower_dir_dentry = lower_dentry->d_parent; - lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt( - ecryptfs_dentry->d_parent)); - lower_inode = lower_dentry->d_inode; - fsstack_copy_attr_atime(ecryptfs_dir_inode, lower_dir_dentry->d_inode); + int rc = 0; + + lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); + fsstack_copy_attr_atime(dir_inode, lower_dentry->d_parent->d_inode); BUG_ON(!lower_dentry->d_count); - ecryptfs_set_dentry_private(ecryptfs_dentry, - kmem_cache_alloc(ecryptfs_dentry_info_cache, - GFP_KERNEL)); - if (!ecryptfs_dentry_to_private(ecryptfs_dentry)) { - rc = -ENOMEM; + + dentry_info = kmem_cache_alloc(ecryptfs_dentry_info_cache, GFP_KERNEL); + ecryptfs_set_dentry_private(dentry, dentry_info); + if (!dentry_info) { printk(KERN_ERR "%s: Out of memory whilst attempting " "to allocate ecryptfs_dentry_info struct\n", __func__); - goto out_put; + dput(lower_dentry); + mntput(lower_mnt); + d_drop(dentry); + return -ENOMEM; } - ecryptfs_set_dentry_lower(ecryptfs_dentry, lower_dentry); - ecryptfs_set_dentry_lower_mnt(ecryptfs_dentry, lower_mnt); + ecryptfs_set_dentry_lower(dentry, lower_dentry); + ecryptfs_set_dentry_lower_mnt(dentry, lower_mnt); + if (!lower_dentry->d_inode) { /* We want to add because we couldn't find in lower */ - d_add(ecryptfs_dentry, NULL); - goto out; + d_add(dentry, NULL); + return 0; } - rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry, - ecryptfs_dir_inode->i_sb, - ECRYPTFS_INTERPOSE_FLAG_D_ADD); - if (rc) { - printk(KERN_ERR "%s: Error interposing; rc = [%d]\n", - __func__, rc); - goto out; + inode = __ecryptfs_get_inode(lower_inode, dir_inode->i_sb); + if (IS_ERR(inode)) { + printk(KERN_ERR "%s: Error interposing; rc = [%ld]\n", + __func__, PTR_ERR(inode)); + return PTR_ERR(inode); } - if (S_ISDIR(lower_inode->i_mode)) - goto out; - if (S_ISLNK(lower_inode->i_mode)) - goto out; - if (special_file(lower_inode->i_mode)) - goto out; - /* Released in this function */ - page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, GFP_USER); - if (!page_virt) { - printk(KERN_ERR "%s: Cannot kmem_cache_zalloc() a page\n", - __func__); - rc = -ENOMEM; - goto out; - } - rc = ecryptfs_get_lower_file(ecryptfs_dentry); - if (rc) { - printk(KERN_ERR "%s: Error attempting to initialize " - "the lower file for the dentry with name " - "[%s]; rc = [%d]\n", __func__, - ecryptfs_dentry->d_name.name, rc); - goto out_free_kmem; - } - put_lower = 1; - crypt_stat = &ecryptfs_inode_to_private( - ecryptfs_dentry->d_inode)->crypt_stat; - /* TODO: lock for crypt_stat comparison */ - if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED)) - ecryptfs_set_default_sizes(crypt_stat); - rc = ecryptfs_read_and_validate_header_region(page_virt, - ecryptfs_dentry->d_inode); - if (rc) { - memset(page_virt, 0, PAGE_CACHE_SIZE); - rc = ecryptfs_read_and_validate_xattr_region(page_virt, - ecryptfs_dentry); + if (S_ISREG(inode->i_mode)) { + rc = ecryptfs_i_size_read(dentry, inode); if (rc) { - rc = 0; - goto out_free_kmem; + make_bad_inode(inode); + return rc; } - crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR; } - ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode); -out_free_kmem: - kmem_cache_free(ecryptfs_header_cache_2, page_virt); - goto out; -out_put: - dput(lower_dentry); - mntput(lower_mnt); - d_drop(ecryptfs_dentry); -out: - if (put_lower) - ecryptfs_put_lower_file(ecryptfs_dentry->d_inode); + + if (inode->i_state & I_NEW) + unlock_new_inode(inode); + d_add(dentry, inode); + return rc; } @@ -353,12 +433,12 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, goto out_d_drop; } if (lower_dentry->d_inode) - goto lookup_and_interpose; + goto interpose; mount_crypt_stat = &ecryptfs_superblock_to_private( ecryptfs_dentry->d_sb)->mount_crypt_stat; if (!(mount_crypt_stat && (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES))) - goto lookup_and_interpose; + goto interpose; dput(lower_dentry); rc = ecryptfs_encrypt_and_encode_filename( &encrypted_and_encoded_name, &encrypted_and_encoded_name_size, @@ -381,9 +461,9 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, encrypted_and_encoded_name); goto out_d_drop; } -lookup_and_interpose: - rc = ecryptfs_lookup_and_interpose_lower(ecryptfs_dentry, lower_dentry, - ecryptfs_dir_inode); +interpose: + rc = ecryptfs_lookup_interpose(ecryptfs_dentry, lower_dentry, + ecryptfs_dir_inode); goto out; out_d_drop: d_drop(ecryptfs_dentry); @@ -411,7 +491,7 @@ static int ecryptfs_link(struct dentry *old_dentry, struct inode *dir, lower_new_dentry); if (rc || !lower_new_dentry->d_inode) goto out_lock; - rc = ecryptfs_interpose(lower_new_dentry, new_dentry, dir->i_sb, 0); + rc = ecryptfs_interpose(lower_new_dentry, new_dentry, dir->i_sb); if (rc) goto out_lock; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); @@ -478,7 +558,7 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, kfree(encoded_symname); if (rc || !lower_dentry->d_inode) goto out_lock; - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb); if (rc) goto out_lock; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); @@ -502,7 +582,7 @@ static int ecryptfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) rc = vfs_mkdir(lower_dir_dentry->d_inode, lower_dentry, mode); if (rc || !lower_dentry->d_inode) goto out; - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb); if (rc) goto out; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); @@ -527,6 +607,8 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) dget(lower_dentry); rc = vfs_rmdir(lower_dir_dentry->d_inode, lower_dentry); dput(lower_dentry); + if (!rc && dentry->d_inode) + clear_nlink(dentry->d_inode); fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); dir->i_nlink = lower_dir_dentry->d_inode->i_nlink; unlock_dir(lower_dir_dentry); @@ -548,7 +630,7 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) rc = vfs_mknod(lower_dir_dentry->d_inode, lower_dentry, mode, dev); if (rc || !lower_dentry->d_inode) goto out; - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb); if (rc) goto out; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); @@ -748,7 +830,7 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia, lower_ia->ia_valid &= ~ATTR_SIZE; return 0; } - rc = ecryptfs_get_lower_file(dentry); + rc = ecryptfs_get_lower_file(dentry, inode); if (rc) return rc; crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; @@ -904,7 +986,7 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) mount_crypt_stat = &ecryptfs_superblock_to_private( dentry->d_sb)->mount_crypt_stat; - rc = ecryptfs_get_lower_file(dentry); + rc = ecryptfs_get_lower_file(dentry, inode); if (rc) { mutex_unlock(&crypt_stat->cs_mutex); goto out; @@ -1077,21 +1159,6 @@ out: return rc; } -int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode) -{ - if ((ecryptfs_inode_to_lower(inode) - == (struct inode *)candidate_lower_inode)) - return 1; - else - return 0; -} - -int ecryptfs_inode_set(struct inode *inode, void *lower_inode) -{ - ecryptfs_init_inode(inode, (struct inode *)lower_inode); - return 0; -} - const struct inode_operations ecryptfs_symlink_iops = { .readlink = ecryptfs_readlink, .follow_link = ecryptfs_follow_link, diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 03e609c..27a7fef 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -599,8 +599,8 @@ struct ecryptfs_write_tag_70_packet_silly_stack { struct mutex *tfm_mutex; char *block_aligned_filename; struct ecryptfs_auth_tok *auth_tok; - struct scatterlist src_sg; - struct scatterlist dst_sg; + struct scatterlist src_sg[2]; + struct scatterlist dst_sg[2]; struct blkcipher_desc desc; char iv[ECRYPTFS_MAX_IV_BYTES]; char hash[ECRYPTFS_TAG_70_DIGEST_SIZE]; @@ -816,23 +816,21 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, memcpy(&s->block_aligned_filename[s->num_rand_bytes], filename, filename_size); rc = virt_to_scatterlist(s->block_aligned_filename, - s->block_aligned_filename_size, &s->src_sg, 1); - if (rc != 1) { + s->block_aligned_filename_size, s->src_sg, 2); + if (rc < 1) { printk(KERN_ERR "%s: Internal error whilst attempting to " - "convert filename memory to scatterlist; " - "expected rc = 1; got rc = [%d]. " + "convert filename memory to scatterlist; rc = [%d]. " "block_aligned_filename_size = [%zd]\n", __func__, rc, s->block_aligned_filename_size); goto out_release_free_unlock; } rc = virt_to_scatterlist(&dest[s->i], s->block_aligned_filename_size, - &s->dst_sg, 1); - if (rc != 1) { + s->dst_sg, 2); + if (rc < 1) { printk(KERN_ERR "%s: Internal error whilst attempting to " "convert encrypted filename memory to scatterlist; " - "expected rc = 1; got rc = [%d]. " - "block_aligned_filename_size = [%zd]\n", __func__, rc, - s->block_aligned_filename_size); + "rc = [%d]. block_aligned_filename_size = [%zd]\n", + __func__, rc, s->block_aligned_filename_size); goto out_release_free_unlock; } /* The characters in the first block effectively do the job @@ -855,7 +853,7 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, mount_crypt_stat->global_default_fn_cipher_key_bytes); goto out_release_free_unlock; } - rc = crypto_blkcipher_encrypt_iv(&s->desc, &s->dst_sg, &s->src_sg, + rc = crypto_blkcipher_encrypt_iv(&s->desc, s->dst_sg, s->src_sg, s->block_aligned_filename_size); if (rc) { printk(KERN_ERR "%s: Error attempting to encrypt filename; " @@ -891,8 +889,8 @@ struct ecryptfs_parse_tag_70_packet_silly_stack { struct mutex *tfm_mutex; char *decrypted_filename; struct ecryptfs_auth_tok *auth_tok; - struct scatterlist src_sg; - struct scatterlist dst_sg; + struct scatterlist src_sg[2]; + struct scatterlist dst_sg[2]; struct blkcipher_desc desc; char fnek_sig_hex[ECRYPTFS_SIG_SIZE_HEX + 1]; char iv[ECRYPTFS_MAX_IV_BYTES]; @@ -1008,13 +1006,12 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, } mutex_lock(s->tfm_mutex); rc = virt_to_scatterlist(&data[(*packet_size)], - s->block_aligned_filename_size, &s->src_sg, 1); - if (rc != 1) { + s->block_aligned_filename_size, s->src_sg, 2); + if (rc < 1) { printk(KERN_ERR "%s: Internal error whilst attempting to " "convert encrypted filename memory to scatterlist; " - "expected rc = 1; got rc = [%d]. " - "block_aligned_filename_size = [%zd]\n", __func__, rc, - s->block_aligned_filename_size); + "rc = [%d]. block_aligned_filename_size = [%zd]\n", + __func__, rc, s->block_aligned_filename_size); goto out_unlock; } (*packet_size) += s->block_aligned_filename_size; @@ -1028,13 +1025,12 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, goto out_unlock; } rc = virt_to_scatterlist(s->decrypted_filename, - s->block_aligned_filename_size, &s->dst_sg, 1); - if (rc != 1) { + s->block_aligned_filename_size, s->dst_sg, 2); + if (rc < 1) { printk(KERN_ERR "%s: Internal error whilst attempting to " "convert decrypted filename memory to scatterlist; " - "expected rc = 1; got rc = [%d]. " - "block_aligned_filename_size = [%zd]\n", __func__, rc, - s->block_aligned_filename_size); + "rc = [%d]. block_aligned_filename_size = [%zd]\n", + __func__, rc, s->block_aligned_filename_size); goto out_free_unlock; } /* The characters in the first block effectively do the job of @@ -1065,7 +1061,7 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, mount_crypt_stat->global_default_fn_cipher_key_bytes); goto out_free_unlock; } - rc = crypto_blkcipher_decrypt_iv(&s->desc, &s->dst_sg, &s->src_sg, + rc = crypto_blkcipher_decrypt_iv(&s->desc, s->dst_sg, s->src_sg, s->block_aligned_filename_size); if (rc) { printk(KERN_ERR "%s: Error attempting to decrypt filename; " diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 89b9338..9f1bb74 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -135,12 +135,12 @@ static int ecryptfs_init_lower_file(struct dentry *dentry, return rc; } -int ecryptfs_get_lower_file(struct dentry *dentry) +int ecryptfs_get_lower_file(struct dentry *dentry, struct inode *inode) { - struct ecryptfs_inode_info *inode_info = - ecryptfs_inode_to_private(dentry->d_inode); + struct ecryptfs_inode_info *inode_info; int count, rc = 0; + inode_info = ecryptfs_inode_to_private(inode); mutex_lock(&inode_info->lower_file_mutex); count = atomic_inc_return(&inode_info->lower_file_count); if (WARN_ON_ONCE(count < 1)) @@ -168,75 +168,6 @@ void ecryptfs_put_lower_file(struct inode *inode) } } -static struct inode *ecryptfs_get_inode(struct inode *lower_inode, - struct super_block *sb) -{ - struct inode *inode; - int rc = 0; - - if (lower_inode->i_sb != ecryptfs_superblock_to_lower(sb)) { - rc = -EXDEV; - goto out; - } - if (!igrab(lower_inode)) { - rc = -ESTALE; - goto out; - } - inode = iget5_locked(sb, (unsigned long)lower_inode, - ecryptfs_inode_test, ecryptfs_inode_set, - lower_inode); - if (!inode) { - rc = -EACCES; - iput(lower_inode); - goto out; - } - if (inode->i_state & I_NEW) - unlock_new_inode(inode); - else - iput(lower_inode); - if (S_ISLNK(lower_inode->i_mode)) - inode->i_op = &ecryptfs_symlink_iops; - else if (S_ISDIR(lower_inode->i_mode)) - inode->i_op = &ecryptfs_dir_iops; - if (S_ISDIR(lower_inode->i_mode)) - inode->i_fop = &ecryptfs_dir_fops; - if (special_file(lower_inode->i_mode)) - init_special_inode(inode, lower_inode->i_mode, - lower_inode->i_rdev); - fsstack_copy_attr_all(inode, lower_inode); - /* This size will be overwritten for real files w/ headers and - * other metadata */ - fsstack_copy_inode_size(inode, lower_inode); - return inode; -out: - return ERR_PTR(rc); -} - -/** - * ecryptfs_interpose - * @lower_dentry: Existing dentry in the lower filesystem - * @dentry: ecryptfs' dentry - * @sb: ecryptfs's super_block - * @flags: flags to govern behavior of interpose procedure - * - * Interposes upper and lower dentries. - * - * Returns zero on success; non-zero otherwise - */ -int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, - struct super_block *sb, u32 flags) -{ - struct inode *lower_inode = lower_dentry->d_inode; - struct inode *inode = ecryptfs_get_inode(lower_inode, sb); - if (IS_ERR(inode)) - return PTR_ERR(inode); - if (flags & ECRYPTFS_INTERPOSE_FLAG_D_ADD) - d_add(dentry, inode); - else - d_instantiate(dentry, inode); - return 0; -} - enum { ecryptfs_opt_sig, ecryptfs_opt_ecryptfs_sig, ecryptfs_opt_cipher, ecryptfs_opt_ecryptfs_cipher, ecryptfs_opt_ecryptfs_key_bytes, @@ -704,13 +635,8 @@ static struct ecryptfs_cache_info { .size = sizeof(struct ecryptfs_sb_info), }, { - .cache = &ecryptfs_header_cache_1, - .name = "ecryptfs_headers_1", - .size = PAGE_CACHE_SIZE, - }, - { - .cache = &ecryptfs_header_cache_2, - .name = "ecryptfs_headers_2", + .cache = &ecryptfs_header_cache, + .name = "ecryptfs_headers", .size = PAGE_CACHE_SIZE, }, { diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 245b517..dbd52d40 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -93,22 +93,6 @@ static void ecryptfs_destroy_inode(struct inode *inode) } /** - * ecryptfs_init_inode - * @inode: The ecryptfs inode - * - * Set up the ecryptfs inode. - */ -void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode) -{ - ecryptfs_set_inode_lower(inode, lower_inode); - inode->i_ino = lower_inode->i_ino; - inode->i_version++; - inode->i_op = &ecryptfs_main_iops; - inode->i_fop = &ecryptfs_main_fops; - inode->i_mapping->a_ops = &ecryptfs_aops; -} - -/** * ecryptfs_statfs * @sb: The ecryptfs super block * @buf: The struct kstatfs to fill in with stats @@ -42,7 +42,6 @@ #include <linux/pid_namespace.h> #include <linux/module.h> #include <linux/namei.h> -#include <linux/proc_fs.h> #include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> @@ -55,6 +54,7 @@ #include <linux/fs_struct.h> #include <linux/pipe_fs_i.h> #include <linux/oom.h> +#include <linux/compat.h> #include <asm/uaccess.h> #include <asm/mmu_context.h> @@ -166,8 +166,13 @@ out: } #ifdef CONFIG_MMU - -void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) +/* + * The nascent bprm->mm is not visible until exec_mmap() but it can + * use a lot of memory, account these pages in current->mm temporary + * for oom_badness()->get_mm_rss(). Once exec succeeds or fails, we + * change the counter back via acct_arg_size(0). + */ +static void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) { struct mm_struct *mm = current->mm; long diff = (long)(pages - bprm->vma_pages); @@ -186,7 +191,7 @@ void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) #endif } -struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, +static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, int write) { struct page *page; @@ -194,7 +199,7 @@ struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, #ifdef CONFIG_STACK_GROWSUP if (write) { - ret = expand_stack_downwards(bprm->vma, pos); + ret = expand_downwards(bprm->vma, pos); if (ret < 0) return NULL; } @@ -305,11 +310,11 @@ static bool valid_arg_len(struct linux_binprm *bprm, long len) #else -void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) +static inline void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) { } -struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, +static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, int write) { struct page *page; @@ -398,22 +403,56 @@ err: return err; } +struct user_arg_ptr { +#ifdef CONFIG_COMPAT + bool is_compat; +#endif + union { + const char __user *const __user *native; +#ifdef CONFIG_COMPAT + compat_uptr_t __user *compat; +#endif + } ptr; +}; + +static const char __user *get_user_arg_ptr(struct user_arg_ptr argv, int nr) +{ + const char __user *native; + +#ifdef CONFIG_COMPAT + if (unlikely(argv.is_compat)) { + compat_uptr_t compat; + + if (get_user(compat, argv.ptr.compat + nr)) + return ERR_PTR(-EFAULT); + + return compat_ptr(compat); + } +#endif + + if (get_user(native, argv.ptr.native + nr)) + return ERR_PTR(-EFAULT); + + return native; +} + /* * count() counts the number of strings in array ARGV. */ -static int count(const char __user * const __user * argv, int max) +static int count(struct user_arg_ptr argv, int max) { int i = 0; - if (argv != NULL) { + if (argv.ptr.native != NULL) { for (;;) { - const char __user * p; + const char __user *p = get_user_arg_ptr(argv, i); - if (get_user(p, argv)) - return -EFAULT; if (!p) break; - argv++; + + if (IS_ERR(p)) + return -EFAULT; + if (i++ >= max) return -E2BIG; @@ -430,7 +469,7 @@ static int count(const char __user * const __user * argv, int max) * processes's memory to the new process's stack. The call to get_user_pages() * ensures the destination page is created and not swapped out. */ -static int copy_strings(int argc, const char __user *const __user *argv, +static int copy_strings(int argc, struct user_arg_ptr argv, struct linux_binprm *bprm) { struct page *kmapped_page = NULL; @@ -443,16 +482,18 @@ static int copy_strings(int argc, const char __user *const __user *argv, int len; unsigned long pos; - if (get_user(str, argv+argc) || - !(len = strnlen_user(str, MAX_ARG_STRLEN))) { - ret = -EFAULT; + ret = -EFAULT; + str = get_user_arg_ptr(argv, argc); + if (IS_ERR(str)) goto out; - } - if (!valid_arg_len(bprm, len)) { - ret = -E2BIG; + len = strnlen_user(str, MAX_ARG_STRLEN); + if (!len) + goto out; + + ret = -E2BIG; + if (!valid_arg_len(bprm, len)) goto out; - } /* We're going to work our way backwords. */ pos = bprm->p; @@ -519,14 +560,19 @@ out: /* * Like copy_strings, but get argv and its values from kernel memory. */ -int copy_strings_kernel(int argc, const char *const *argv, +int copy_strings_kernel(int argc, const char *const *__argv, struct linux_binprm *bprm) { int r; mm_segment_t oldfs = get_fs(); + struct user_arg_ptr argv = { + .ptr.native = (const char __user *const __user *)__argv, + }; + set_fs(KERNEL_DS); - r = copy_strings(argc, (const char __user *const __user *)argv, bprm); + r = copy_strings(argc, argv, bprm); set_fs(oldfs); + return r; } EXPORT_SYMBOL(copy_strings_kernel); @@ -553,7 +599,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) unsigned long length = old_end - old_start; unsigned long new_start = old_start - shift; unsigned long new_end = old_end - shift; - struct mmu_gather *tlb; + struct mmu_gather tlb; BUG_ON(new_start > new_end); @@ -579,12 +625,12 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) return -ENOMEM; lru_add_drain(); - tlb = tlb_gather_mmu(mm, 0); + tlb_gather_mmu(&tlb, mm, 0); if (new_end > old_start) { /* * when the old and new regions overlap clear from new_end. */ - free_pgd_range(tlb, new_end, old_end, new_end, + free_pgd_range(&tlb, new_end, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } else { /* @@ -593,10 +639,10 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * have constraints on va-space that make this illegal (IA64) - * for the others its just a little faster. */ - free_pgd_range(tlb, old_start, old_end, new_end, + free_pgd_range(&tlb, old_start, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } - tlb_finish_mmu(tlb, new_end, old_end); + tlb_finish_mmu(&tlb, new_end, old_end); /* * Shrink the vma to just the new range. Always succeeds. @@ -1004,6 +1050,7 @@ char *get_task_comm(char *buf, struct task_struct *tsk) task_unlock(tsk); return buf; } +EXPORT_SYMBOL_GPL(get_task_comm); void set_task_comm(struct task_struct *tsk, char *buf) { @@ -1046,6 +1093,7 @@ int flush_old_exec(struct linux_binprm * bprm) bprm->mm = NULL; /* We're using it now */ + set_fs(USER_DS); current->flags &= ~(PF_RANDOMIZE | PF_KTHREAD); flush_thread(); current->personality &= ~bprm->per_clear; @@ -1310,10 +1358,6 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) if (retval) return retval; - /* kernel module loader fixup */ - /* so we don't try to load run modprobe in kernel space. */ - set_fs(USER_DS); - retval = audit_bprm(bprm); if (retval) return retval; @@ -1379,10 +1423,10 @@ EXPORT_SYMBOL(search_binary_handler); /* * sys_execve() executes a new program. */ -int do_execve(const char * filename, - const char __user *const __user *argv, - const char __user *const __user *envp, - struct pt_regs * regs) +static int do_execve_common(const char *filename, + struct user_arg_ptr argv, + struct user_arg_ptr envp, + struct pt_regs *regs) { struct linux_binprm *bprm; struct file *file; @@ -1489,6 +1533,34 @@ out_ret: return retval; } +int do_execve(const char *filename, + const char __user *const __user *__argv, + const char __user *const __user *__envp, + struct pt_regs *regs) +{ + struct user_arg_ptr argv = { .ptr.native = __argv }; + struct user_arg_ptr envp = { .ptr.native = __envp }; + return do_execve_common(filename, argv, envp, regs); +} + +#ifdef CONFIG_COMPAT +int compat_do_execve(char *filename, + compat_uptr_t __user *__argv, + compat_uptr_t __user *__envp, + struct pt_regs *regs) +{ + struct user_arg_ptr argv = { + .is_compat = true, + .ptr.compat = __argv, + }; + struct user_arg_ptr envp = { + .is_compat = true, + .ptr.compat = __envp, + }; + return do_execve_common(filename, argv, envp, regs); +} +#endif + void set_binfmt(struct linux_binfmt *new) { struct mm_struct *mm = current->mm; @@ -1548,6 +1620,41 @@ expand_fail: return ret; } +static int cn_print_exe_file(struct core_name *cn) +{ + struct file *exe_file; + char *pathbuf, *path, *p; + int ret; + + exe_file = get_mm_exe_file(current->mm); + if (!exe_file) + return cn_printf(cn, "(unknown)"); + + pathbuf = kmalloc(PATH_MAX, GFP_TEMPORARY); + if (!pathbuf) { + ret = -ENOMEM; + goto put_exe_file; + } + + path = d_path(&exe_file->f_path, pathbuf, PATH_MAX); + if (IS_ERR(path)) { + ret = PTR_ERR(path); + goto free_buf; + } + + for (p = path; *p; p++) + if (*p == '/') + *p = '!'; + + ret = cn_printf(cn, "%s", path); + +free_buf: + kfree(pathbuf); +put_exe_file: + fput(exe_file); + return ret; +} + /* format_corename will inspect the pattern parameter, and output a * name into corename, which must have space for at least * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator. @@ -1619,6 +1726,9 @@ static int format_corename(struct core_name *cn, long signr) case 'e': err = cn_printf(cn, "%s", current->comm); break; + case 'E': + err = cn_print_exe_file(cn); + break; /* core limit size */ case 'c': err = cn_printf(cn, "%lu", @@ -1659,6 +1769,7 @@ static int zap_process(struct task_struct *start, int exit_code) t = start; do { + task_clear_group_stop_pending(t); if (t != current && t->mm) { sigaddset(&t->pending.signal, SIGKILL); signal_wake_up(t, 1); @@ -1885,7 +1996,7 @@ static void wait_for_dump_helpers(struct file *file) * is a special value that we use to trap recursive * core dumps */ -static int umh_pipe_setup(struct subprocess_info *info) +static int umh_pipe_setup(struct subprocess_info *info, struct cred *new) { struct file *rp, *wp; struct fdtable *fdt; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 0a78dae..1dd62ed 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -898,7 +898,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) brelse(bh); if (!sb_set_blocksize(sb, blocksize)) { - ext2_msg(sb, KERN_ERR, "error: blocksize is too small"); + ext2_msg(sb, KERN_ERR, + "error: bad blocksize %d", blocksize); goto failed_sbi; } diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 68b2e43..3451d23 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -3392,7 +3392,7 @@ int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode) * so would cause a commit on atime updates, which we don't bother doing. * We handle synchronous inodes at the highest possible level. */ -void ext3_dirty_inode(struct inode *inode) +void ext3_dirty_inode(struct inode *inode, int flags) { handle_t *current_handle = ext3_journal_current_handle(); handle_t *handle; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 32f3b86..34b6d9b 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1416,10 +1416,19 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, frame->at = entries; frame->bh = bh; bh = bh2; + /* + * Mark buffers dirty here so that if do_split() fails we write a + * consistent set of buffers to disk. + */ + ext3_journal_dirty_metadata(handle, frame->bh); + ext3_journal_dirty_metadata(handle, bh); de = do_split(handle,dir, &bh, frame, &hinfo, &retval); - dx_release (frames); - if (!(de)) + if (!de) { + ext3_mark_inode_dirty(handle, dir); + dx_release(frames); return retval; + } + dx_release(frames); return add_dirent_to_buf(handle, dentry, inode, de, bh); } @@ -2189,6 +2198,7 @@ static int ext3_symlink (struct inode * dir, handle_t *handle; struct inode * inode; int l, err, retries = 0; + int credits; l = strlen(symname)+1; if (l > dir->i_sb->s_blocksize) @@ -2196,10 +2206,26 @@ static int ext3_symlink (struct inode * dir, dquot_initialize(dir); + if (l > EXT3_N_BLOCKS * 4) { + /* + * For non-fast symlinks, we just allocate inode and put it on + * orphan list in the first transaction => we need bitmap, + * group descriptor, sb, inode block, quota blocks. + */ + credits = 4 + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb); + } else { + /* + * Fast symlink. We have to add entry to directory + * (EXT3_DATA_TRANS_BLOCKS + EXT3_INDEX_EXTRA_TRANS_BLOCKS), + * allocate new inode (bitmap, group descriptor, inode block, + * quota blocks, sb is already counted in previous macros). + */ + credits = EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb); + } retry: - handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + - EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); + handle = ext3_journal_start(dir, credits); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2211,21 +2237,45 @@ retry: if (IS_ERR(inode)) goto out_stop; - if (l > sizeof (EXT3_I(inode)->i_data)) { + if (l > EXT3_N_BLOCKS * 4) { inode->i_op = &ext3_symlink_inode_operations; ext3_set_aops(inode); /* - * page_symlink() calls into ext3_prepare/commit_write. - * We have a transaction open. All is sweetness. It also sets - * i_size in generic_commit_write(). + * We cannot call page_symlink() with transaction started + * because it calls into ext3_write_begin() which acquires page + * lock which ranks below transaction start (and it can also + * wait for journal commit if we are running out of space). So + * we have to stop transaction now and restart it when symlink + * contents is written. + * + * To keep fs consistent in case of crash, we have to put inode + * to orphan list in the mean time. */ + drop_nlink(inode); + err = ext3_orphan_add(handle, inode); + ext3_journal_stop(handle); + if (err) + goto err_drop_inode; err = __page_symlink(inode, symname, l, 1); + if (err) + goto err_drop_inode; + /* + * Now inode is being linked into dir (EXT3_DATA_TRANS_BLOCKS + * + EXT3_INDEX_EXTRA_TRANS_BLOCKS), inode is also modified + */ + handle = ext3_journal_start(dir, + EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 1); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto err_drop_inode; + } + inc_nlink(inode); + err = ext3_orphan_del(handle, inode); if (err) { + ext3_journal_stop(handle); drop_nlink(inode); - unlock_new_inode(inode); - ext3_mark_inode_dirty(handle, inode); - iput (inode); - goto out_stop; + goto err_drop_inode; } } else { inode->i_op = &ext3_fast_symlink_inode_operations; @@ -2239,6 +2289,10 @@ out_stop: if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; +err_drop_inode: + unlock_new_inode(inode); + iput(inode); + return err; } static int ext3_link (struct dentry * old_dentry, diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 3c6a9e0..aad153e 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -36,6 +36,7 @@ #include <linux/quotaops.h> #include <linux/seq_file.h> #include <linux/log2.h> +#include <linux/cleancache.h> #include <asm/uaccess.h> @@ -1367,6 +1368,7 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, } else { ext3_msg(sb, KERN_INFO, "using internal journal"); } + cleancache_init_fs(sb); return res; } diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index c947e36..04109460 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -6,7 +6,8 @@ obj-$(CONFIG_EXT4_FS) += ext4.o ext4-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o page-io.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ - ext4_jbd2.o migrate.o mballoc.o block_validity.o move_extent.o + ext4_jbd2.o migrate.o mballoc.o block_validity.o move_extent.o \ + mmp.o ext4-$(CONFIG_EXT4_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4-$(CONFIG_EXT4_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 1c67139..264f694 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -362,130 +362,6 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) } /** - * ext4_add_groupblocks() -- Add given blocks to an existing group - * @handle: handle to this transaction - * @sb: super block - * @block: start physcial block to add to the block group - * @count: number of blocks to free - * - * This marks the blocks as free in the bitmap. We ask the - * mballoc to reload the buddy after this by setting group - * EXT4_GROUP_INFO_NEED_INIT_BIT flag - */ -void ext4_add_groupblocks(handle_t *handle, struct super_block *sb, - ext4_fsblk_t block, unsigned long count) -{ - struct buffer_head *bitmap_bh = NULL; - struct buffer_head *gd_bh; - ext4_group_t block_group; - ext4_grpblk_t bit; - unsigned int i; - struct ext4_group_desc *desc; - struct ext4_sb_info *sbi = EXT4_SB(sb); - int err = 0, ret, blk_free_count; - ext4_grpblk_t blocks_freed; - struct ext4_group_info *grp; - - ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1); - - ext4_get_group_no_and_offset(sb, block, &block_group, &bit); - grp = ext4_get_group_info(sb, block_group); - /* - * Check to see if we are freeing blocks across a group - * boundary. - */ - if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) { - goto error_return; - } - bitmap_bh = ext4_read_block_bitmap(sb, block_group); - if (!bitmap_bh) - goto error_return; - desc = ext4_get_group_desc(sb, block_group, &gd_bh); - if (!desc) - goto error_return; - - if (in_range(ext4_block_bitmap(sb, desc), block, count) || - in_range(ext4_inode_bitmap(sb, desc), block, count) || - in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || - in_range(block + count - 1, ext4_inode_table(sb, desc), - sbi->s_itb_per_group)) { - ext4_error(sb, "Adding blocks in system zones - " - "Block = %llu, count = %lu", - block, count); - goto error_return; - } - - /* - * We are about to add blocks to the bitmap, - * so we need undo access. - */ - BUFFER_TRACE(bitmap_bh, "getting undo access"); - err = ext4_journal_get_undo_access(handle, bitmap_bh); - if (err) - goto error_return; - - /* - * We are about to modify some metadata. Call the journal APIs - * to unshare ->b_data if a currently-committing transaction is - * using it - */ - BUFFER_TRACE(gd_bh, "get_write_access"); - err = ext4_journal_get_write_access(handle, gd_bh); - if (err) - goto error_return; - /* - * make sure we don't allow a parallel init on other groups in the - * same buddy cache - */ - down_write(&grp->alloc_sem); - for (i = 0, blocks_freed = 0; i < count; i++) { - BUFFER_TRACE(bitmap_bh, "clear bit"); - if (!ext4_clear_bit_atomic(ext4_group_lock_ptr(sb, block_group), - bit + i, bitmap_bh->b_data)) { - ext4_error(sb, "bit already cleared for block %llu", - (ext4_fsblk_t)(block + i)); - BUFFER_TRACE(bitmap_bh, "bit already cleared"); - } else { - blocks_freed++; - } - } - ext4_lock_group(sb, block_group); - blk_free_count = blocks_freed + ext4_free_blks_count(sb, desc); - ext4_free_blks_set(sb, desc, blk_free_count); - desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc); - ext4_unlock_group(sb, block_group); - percpu_counter_add(&sbi->s_freeblocks_counter, blocks_freed); - - if (sbi->s_log_groups_per_flex) { - ext4_group_t flex_group = ext4_flex_group(sbi, block_group); - atomic_add(blocks_freed, - &sbi->s_flex_groups[flex_group].free_blocks); - } - /* - * request to reload the buddy with the - * new bitmap information - */ - set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state)); - grp->bb_free += blocks_freed; - up_write(&grp->alloc_sem); - - /* We dirtied the bitmap block */ - BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); - err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh); - - /* And the group descriptor block */ - BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); - ret = ext4_handle_dirty_metadata(handle, NULL, gd_bh); - if (!err) - err = ret; - -error_return: - brelse(bitmap_bh); - ext4_std_error(sb, err); - return; -} - -/** * ext4_has_free_blocks() * @sbi: in-core super block structure. * @nblocks: number of needed blocks @@ -493,7 +369,8 @@ error_return: * Check if filesystem has nblocks free & available for allocation. * On success return 1, return 0 on failure. */ -static int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks) +static int ext4_has_free_blocks(struct ext4_sb_info *sbi, + s64 nblocks, unsigned int flags) { s64 free_blocks, dirty_blocks, root_blocks; struct percpu_counter *fbc = &sbi->s_freeblocks_counter; @@ -507,11 +384,6 @@ static int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks) EXT4_FREEBLOCKS_WATERMARK) { free_blocks = percpu_counter_sum_positive(fbc); dirty_blocks = percpu_counter_sum_positive(dbc); - if (dirty_blocks < 0) { - printk(KERN_CRIT "Dirty block accounting " - "went wrong %lld\n", - (long long)dirty_blocks); - } } /* Check whether we have space after * accounting for current dirty blocks & root reserved blocks. @@ -522,7 +394,9 @@ static int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks) /* Hm, nope. Are (enough) root reserved blocks available? */ if (sbi->s_resuid == current_fsuid() || ((sbi->s_resgid != 0) && in_group_p(sbi->s_resgid)) || - capable(CAP_SYS_RESOURCE)) { + capable(CAP_SYS_RESOURCE) || + (flags & EXT4_MB_USE_ROOT_BLOCKS)) { + if (free_blocks >= (nblocks + dirty_blocks)) return 1; } @@ -531,9 +405,9 @@ static int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks) } int ext4_claim_free_blocks(struct ext4_sb_info *sbi, - s64 nblocks) + s64 nblocks, unsigned int flags) { - if (ext4_has_free_blocks(sbi, nblocks)) { + if (ext4_has_free_blocks(sbi, nblocks, flags)) { percpu_counter_add(&sbi->s_dirtyblocks_counter, nblocks); return 0; } else @@ -554,7 +428,7 @@ int ext4_claim_free_blocks(struct ext4_sb_info *sbi, */ int ext4_should_retry_alloc(struct super_block *sb, int *retries) { - if (!ext4_has_free_blocks(EXT4_SB(sb), 1) || + if (!ext4_has_free_blocks(EXT4_SB(sb), 1, 0) || (*retries)++ > 3 || !EXT4_SB(sb)->s_journal) return 0; @@ -577,7 +451,8 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) * error stores in errp pointer */ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp) + ext4_fsblk_t goal, unsigned int flags, + unsigned long *count, int *errp) { struct ext4_allocation_request ar; ext4_fsblk_t ret; @@ -587,6 +462,7 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, ar.inode = inode; ar.goal = goal; ar.len = count ? *count : 1; + ar.flags = flags; ret = ext4_mb_new_blocks(handle, &ar, errp); if (count) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 4daaf2b..1921392 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -108,7 +108,8 @@ typedef unsigned int ext4_group_t; #define EXT4_MB_DELALLOC_RESERVED 0x0400 /* We are doing stream allocation */ #define EXT4_MB_STREAM_ALLOC 0x0800 - +/* Use reserved root blocks if needed */ +#define EXT4_MB_USE_ROOT_BLOCKS 0x1000 struct ext4_allocation_request { /* target inode for block we're allocating */ @@ -209,6 +210,8 @@ struct ext4_io_submit { */ #define EXT4_BAD_INO 1 /* Bad blocks inode */ #define EXT4_ROOT_INO 2 /* Root inode */ +#define EXT4_USR_QUOTA_INO 3 /* User quota inode */ +#define EXT4_GRP_QUOTA_INO 4 /* Group quota inode */ #define EXT4_BOOT_LOADER_INO 5 /* Boot loader inode */ #define EXT4_UNDEL_DIR_INO 6 /* Undelete directory inode */ #define EXT4_RESIZE_INO 7 /* Reserved group descriptors inode */ @@ -512,6 +515,10 @@ struct ext4_new_group_data { /* Convert extent to initialized after IO complete */ #define EXT4_GET_BLOCKS_IO_CONVERT_EXT (EXT4_GET_BLOCKS_CONVERT|\ EXT4_GET_BLOCKS_CREATE_UNINIT_EXT) + /* Punch out blocks of an extent */ +#define EXT4_GET_BLOCKS_PUNCH_OUT_EXT 0x0020 + /* Don't normalize allocation size (used for fallocate) */ +#define EXT4_GET_BLOCKS_NO_NORMALIZE 0x0040 /* * Flags used by ext4_free_blocks @@ -1028,7 +1035,7 @@ struct ext4_super_block { __le16 s_want_extra_isize; /* New inodes should reserve # bytes */ __le32 s_flags; /* Miscellaneous flags */ __le16 s_raid_stride; /* RAID stride */ - __le16 s_mmp_interval; /* # seconds to wait in MMP checking */ + __le16 s_mmp_update_interval; /* # seconds to wait in MMP checking */ __le64 s_mmp_block; /* Block for multi-mount protection */ __le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/ __u8 s_log_groups_per_flex; /* FLEX_BG group size */ @@ -1144,6 +1151,9 @@ struct ext4_sb_info { unsigned long s_ext_blocks; unsigned long s_ext_extents; #endif + /* ext4 extent cache stats */ + unsigned long extent_cache_hits; + unsigned long extent_cache_misses; /* for buddy allocator */ struct ext4_group_info ***s_group_info; @@ -1201,6 +1211,9 @@ struct ext4_sb_info { struct ext4_li_request *s_li_request; /* Wait multiplier for lazy initialization thread */ unsigned int s_li_wait_mult; + + /* Kernel thread for multiple mount protection */ + struct task_struct *s_mmp_tsk; }; static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb) @@ -1338,6 +1351,7 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei) #define EXT4_FEATURE_RO_COMPAT_GDT_CSUM 0x0010 #define EXT4_FEATURE_RO_COMPAT_DIR_NLINK 0x0020 #define EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE 0x0040 +#define EXT4_FEATURE_RO_COMPAT_QUOTA 0x0100 #define EXT4_FEATURE_INCOMPAT_COMPRESSION 0x0001 #define EXT4_FEATURE_INCOMPAT_FILETYPE 0x0002 @@ -1351,13 +1365,29 @@ static inline void ext4_clear_state_flags(struct ext4_inode_info *ei) #define EXT4_FEATURE_INCOMPAT_EA_INODE 0x0400 /* EA in inode */ #define EXT4_FEATURE_INCOMPAT_DIRDATA 0x1000 /* data in dirent */ +#define EXT2_FEATURE_COMPAT_SUPP EXT4_FEATURE_COMPAT_EXT_ATTR +#define EXT2_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ + EXT4_FEATURE_INCOMPAT_META_BG) +#define EXT2_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ + EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ + EXT4_FEATURE_RO_COMPAT_BTREE_DIR) + +#define EXT3_FEATURE_COMPAT_SUPP EXT4_FEATURE_COMPAT_EXT_ATTR +#define EXT3_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ + EXT4_FEATURE_INCOMPAT_RECOVER| \ + EXT4_FEATURE_INCOMPAT_META_BG) +#define EXT3_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ + EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ + EXT4_FEATURE_RO_COMPAT_BTREE_DIR) + #define EXT4_FEATURE_COMPAT_SUPP EXT2_FEATURE_COMPAT_EXT_ATTR #define EXT4_FEATURE_INCOMPAT_SUPP (EXT4_FEATURE_INCOMPAT_FILETYPE| \ EXT4_FEATURE_INCOMPAT_RECOVER| \ EXT4_FEATURE_INCOMPAT_META_BG| \ EXT4_FEATURE_INCOMPAT_EXTENTS| \ EXT4_FEATURE_INCOMPAT_64BIT| \ - EXT4_FEATURE_INCOMPAT_FLEX_BG) + EXT4_FEATURE_INCOMPAT_FLEX_BG| \ + EXT4_FEATURE_INCOMPAT_MMP) #define EXT4_FEATURE_RO_COMPAT_SUPP (EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER| \ EXT4_FEATURE_RO_COMPAT_LARGE_FILE| \ EXT4_FEATURE_RO_COMPAT_GDT_CSUM| \ @@ -1590,12 +1620,6 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, */ struct ext4_lazy_init { unsigned long li_state; - - wait_queue_head_t li_wait_daemon; - wait_queue_head_t li_wait_task; - struct timer_list li_timer; - struct task_struct *li_task; - struct list_head li_request_list; struct mutex li_list_mtx; }; @@ -1615,6 +1639,67 @@ struct ext4_features { }; /* + * This structure will be used for multiple mount protection. It will be + * written into the block number saved in the s_mmp_block field in the + * superblock. Programs that check MMP should assume that if + * SEQ_FSCK (or any unknown code above SEQ_MAX) is present then it is NOT safe + * to use the filesystem, regardless of how old the timestamp is. + */ +#define EXT4_MMP_MAGIC 0x004D4D50U /* ASCII for MMP */ +#define EXT4_MMP_SEQ_CLEAN 0xFF4D4D50U /* mmp_seq value for clean unmount */ +#define EXT4_MMP_SEQ_FSCK 0xE24D4D50U /* mmp_seq value when being fscked */ +#define EXT4_MMP_SEQ_MAX 0xE24D4D4FU /* maximum valid mmp_seq value */ + +struct mmp_struct { + __le32 mmp_magic; /* Magic number for MMP */ + __le32 mmp_seq; /* Sequence no. updated periodically */ + + /* + * mmp_time, mmp_nodename & mmp_bdevname are only used for information + * purposes and do not affect the correctness of the algorithm + */ + __le64 mmp_time; /* Time last updated */ + char mmp_nodename[64]; /* Node which last updated MMP block */ + char mmp_bdevname[32]; /* Bdev which last updated MMP block */ + + /* + * mmp_check_interval is used to verify if the MMP block has been + * updated on the block device. The value is updated based on the + * maximum time to write the MMP block during an update cycle. + */ + __le16 mmp_check_interval; + + __le16 mmp_pad1; + __le32 mmp_pad2[227]; +}; + +/* arguments passed to the mmp thread */ +struct mmpd_data { + struct buffer_head *bh; /* bh from initial read_mmp_block() */ + struct super_block *sb; /* super block of the fs */ +}; + +/* + * Check interval multiplier + * The MMP block is written every update interval and initially checked every + * update interval x the multiplier (the value is then adapted based on the + * write latency). The reason is that writes can be delayed under load and we + * don't want readers to incorrectly assume that the filesystem is no longer + * in use. + */ +#define EXT4_MMP_CHECK_MULT 2UL + +/* + * Minimum interval for MMP checking in seconds. + */ +#define EXT4_MMP_MIN_CHECK_INTERVAL 5UL + +/* + * Maximum interval for MMP checking in seconds. + */ +#define EXT4_MMP_MAX_CHECK_INTERVAL 300UL + +/* * Function prototypes */ @@ -1638,10 +1723,12 @@ extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group); extern unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group); extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp); -extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks); -extern void ext4_add_groupblocks(handle_t *handle, struct super_block *sb, - ext4_fsblk_t block, unsigned long count); + ext4_fsblk_t goal, + unsigned int flags, + unsigned long *count, + int *errp); +extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, + s64 nblocks, unsigned int flags); extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *); extern void ext4_check_blocks_bitmap(struct super_block *); extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, @@ -1706,6 +1793,8 @@ extern void ext4_free_blocks(handle_t *handle, struct inode *inode, unsigned long count, int flags); extern int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t i, struct ext4_group_desc *desc); +extern void ext4_add_groupblocks(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count); extern int ext4_trim_fs(struct super_block *, struct fstrim_range *); /* inode.c */ @@ -1724,11 +1813,12 @@ extern int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, extern void ext4_evict_inode(struct inode *); extern void ext4_clear_inode(struct inode *); extern int ext4_sync_inode(handle_t *, struct inode *); -extern void ext4_dirty_inode(struct inode *); +extern void ext4_dirty_inode(struct inode *, int); extern int ext4_change_inode_journal_flag(struct inode *, int); extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); extern int ext4_can_truncate(struct inode *inode); extern void ext4_truncate(struct inode *); +extern int ext4_punch_hole(struct file *file, loff_t offset, loff_t length); extern int ext4_truncate_restart_trans(handle_t *, struct inode *, int nblocks); extern void ext4_set_inode_flags(struct inode *); extern void ext4_get_inode_flags(struct ext4_inode_info *); @@ -1738,6 +1828,8 @@ extern int ext4_writepage_trans_blocks(struct inode *); extern int ext4_chunk_trans_blocks(struct inode *, int nrblocks); extern int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from); +extern int ext4_block_zero_page_range(handle_t *handle, + struct address_space *mapping, loff_t from, loff_t length); extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); extern qsize_t *ext4_get_reserved_space(struct inode *inode); extern void ext4_da_update_reserve_space(struct inode *inode, @@ -1788,6 +1880,10 @@ extern void __ext4_warning(struct super_block *, const char *, unsigned int, __LINE__, ## message) extern void ext4_msg(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); +extern void __dump_mmp_msg(struct super_block *, struct mmp_struct *mmp, + const char *, unsigned int, const char *); +#define dump_mmp_msg(sb, mmp, msg) __dump_mmp_msg(sb, mmp, __func__, \ + __LINE__, msg) extern void __ext4_grp_locked_error(const char *, unsigned int, \ struct super_block *, ext4_group_t, \ unsigned long, ext4_fsblk_t, \ @@ -2064,6 +2160,8 @@ extern int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, struct ext4_map_blocks *map, int flags); extern void ext4_ext_truncate(struct inode *); +extern int ext4_ext_punch_hole(struct file *file, loff_t offset, + loff_t length); extern void ext4_ext_init(struct super_block *); extern void ext4_ext_release(struct super_block *); extern long ext4_fallocate(struct file *file, int mode, loff_t offset, @@ -2092,6 +2190,9 @@ extern int ext4_bio_write_page(struct ext4_io_submit *io, int len, struct writeback_control *wbc); +/* mmp.c */ +extern int ext4_multi_mount_protect(struct super_block *, ext4_fsblk_t); + /* BH_Uninit flag: blocks are allocated but uninitialized on disk */ enum ext4_state_bits { BH_Uninit /* blocks are allocated but uninitialized on disk */ diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 2e29abb..095c36f 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -125,7 +125,7 @@ struct ext4_ext_path { * positive retcode - signal for ext4_ext_walk_space(), see below * callback must return valid extent (passed or newly created) */ -typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, +typedef int (*ext_prepare_callback)(struct inode *, ext4_lblk_t, struct ext4_ext_cache *, struct ext4_extent *, void *); @@ -133,8 +133,11 @@ typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, #define EXT_BREAK 1 #define EXT_REPEAT 2 -/* Maximum logical block in a file; ext4_extent's ee_block is __le32 */ -#define EXT_MAX_BLOCK 0xffffffff +/* + * Maximum number of logical blocks in a file; ext4_extent's ee_block is + * __le32. + */ +#define EXT_MAX_BLOCKS 0xffffffff /* * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index 6e272ef..f5240aa 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -6,20 +6,6 @@ #include <trace/events/ext4.h> -int __ext4_journal_get_undo_access(const char *where, unsigned int line, - handle_t *handle, struct buffer_head *bh) -{ - int err = 0; - - if (ext4_handle_valid(handle)) { - err = jbd2_journal_get_undo_access(handle, bh); - if (err) - ext4_journal_abort_handle(where, line, __func__, bh, - handle, err); - } - return err; -} - int __ext4_journal_get_write_access(const char *where, unsigned int line, handle_t *handle, struct buffer_head *bh) { diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index d0f5353..bb85757 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -126,9 +126,6 @@ void ext4_journal_abort_handle(const char *caller, unsigned int line, const char *err_fn, struct buffer_head *bh, handle_t *handle, int err); -int __ext4_journal_get_undo_access(const char *where, unsigned int line, - handle_t *handle, struct buffer_head *bh); - int __ext4_journal_get_write_access(const char *where, unsigned int line, handle_t *handle, struct buffer_head *bh); @@ -146,8 +143,6 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line, int __ext4_handle_dirty_super(const char *where, unsigned int line, handle_t *handle, struct super_block *sb); -#define ext4_journal_get_undo_access(handle, bh) \ - __ext4_journal_get_undo_access(__func__, __LINE__, (handle), (bh)) #define ext4_journal_get_write_access(handle, bh) \ __ext4_journal_get_write_access(__func__, __LINE__, (handle), (bh)) #define ext4_forget(handle, is_metadata, inode, bh, block_nr) \ diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 4890d6f..f815cc8 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -46,6 +46,13 @@ #include <trace/events/ext4.h> +static int ext4_split_extent(handle_t *handle, + struct inode *inode, + struct ext4_ext_path *path, + struct ext4_map_blocks *map, + int split_flag, + int flags); + static int ext4_ext_truncate_extend_restart(handle_t *handle, struct inode *inode, int needed) @@ -192,12 +199,13 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, static ext4_fsblk_t ext4_ext_new_meta_block(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, - struct ext4_extent *ex, int *err) + struct ext4_extent *ex, int *err, unsigned int flags) { ext4_fsblk_t goal, newblock; goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); - newblock = ext4_new_meta_blocks(handle, inode, goal, NULL, err); + newblock = ext4_new_meta_blocks(handle, inode, goal, flags, + NULL, err); return newblock; } @@ -474,9 +482,43 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) } ext_debug("\n"); } + +static void ext4_ext_show_move(struct inode *inode, struct ext4_ext_path *path, + ext4_fsblk_t newblock, int level) +{ + int depth = ext_depth(inode); + struct ext4_extent *ex; + + if (depth != level) { + struct ext4_extent_idx *idx; + idx = path[level].p_idx; + while (idx <= EXT_MAX_INDEX(path[level].p_hdr)) { + ext_debug("%d: move %d:%llu in new index %llu\n", level, + le32_to_cpu(idx->ei_block), + ext4_idx_pblock(idx), + newblock); + idx++; + } + + return; + } + + ex = path[depth].p_ext; + while (ex <= EXT_MAX_EXTENT(path[depth].p_hdr)) { + ext_debug("move %d:%llu:[%d]%d in new leaf %llu\n", + le32_to_cpu(ex->ee_block), + ext4_ext_pblock(ex), + ext4_ext_is_uninitialized(ex), + ext4_ext_get_actual_len(ex), + newblock); + ex++; + } +} + #else #define ext4_ext_show_path(inode, path) #define ext4_ext_show_leaf(inode, path) +#define ext4_ext_show_move(inode, path, newblock, level) #endif void ext4_ext_drop_refs(struct ext4_ext_path *path) @@ -792,14 +834,14 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, * - initializes subtree */ static int ext4_ext_split(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, - struct ext4_extent *newext, int at) + unsigned int flags, + struct ext4_ext_path *path, + struct ext4_extent *newext, int at) { struct buffer_head *bh = NULL; int depth = ext_depth(inode); struct ext4_extent_header *neh; struct ext4_extent_idx *fidx; - struct ext4_extent *ex; int i = at, k, m, a; ext4_fsblk_t newblock, oldblock; __le32 border; @@ -847,7 +889,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); for (a = 0; a < depth - at; a++) { newblock = ext4_ext_new_meta_block(handle, inode, path, - newext, &err); + newext, &err, flags); if (newblock == 0) goto cleanup; ablocks[a] = newblock; @@ -876,7 +918,6 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode, 0)); neh->eh_magic = EXT4_EXT_MAGIC; neh->eh_depth = 0; - ex = EXT_FIRST_EXTENT(neh); /* move remainder of path[depth] to the new leaf */ if (unlikely(path[depth].p_hdr->eh_entries != @@ -888,25 +929,12 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, goto cleanup; } /* start copy from next extent */ - /* TODO: we could do it by single memmove */ - m = 0; - path[depth].p_ext++; - while (path[depth].p_ext <= - EXT_MAX_EXTENT(path[depth].p_hdr)) { - ext_debug("move %d:%llu:[%d]%d in new leaf %llu\n", - le32_to_cpu(path[depth].p_ext->ee_block), - ext4_ext_pblock(path[depth].p_ext), - ext4_ext_is_uninitialized(path[depth].p_ext), - ext4_ext_get_actual_len(path[depth].p_ext), - newblock); - /*memmove(ex++, path[depth].p_ext++, - sizeof(struct ext4_extent)); - neh->eh_entries++;*/ - path[depth].p_ext++; - m++; - } + m = EXT_MAX_EXTENT(path[depth].p_hdr) - path[depth].p_ext++; + ext4_ext_show_move(inode, path, newblock, depth); if (m) { - memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m); + struct ext4_extent *ex; + ex = EXT_FIRST_EXTENT(neh); + memmove(ex, path[depth].p_ext, sizeof(struct ext4_extent) * m); le16_add_cpu(&neh->eh_entries, m); } @@ -968,12 +996,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, ext_debug("int.index at %d (block %llu): %u -> %llu\n", i, newblock, le32_to_cpu(border), oldblock); - /* copy indexes */ - m = 0; - path[i].p_idx++; - ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx, - EXT_MAX_INDEX(path[i].p_hdr)); + /* move remainder of path[i] to the new index block */ if (unlikely(EXT_MAX_INDEX(path[i].p_hdr) != EXT_LAST_INDEX(path[i].p_hdr))) { EXT4_ERROR_INODE(inode, @@ -982,20 +1006,13 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, err = -EIO; goto cleanup; } - while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { - ext_debug("%d: move %d:%llu in new index %llu\n", i, - le32_to_cpu(path[i].p_idx->ei_block), - ext4_idx_pblock(path[i].p_idx), - newblock); - /*memmove(++fidx, path[i].p_idx++, - sizeof(struct ext4_extent_idx)); - neh->eh_entries++; - BUG_ON(neh->eh_entries > neh->eh_max);*/ - path[i].p_idx++; - m++; - } + /* start copy indexes */ + m = EXT_MAX_INDEX(path[i].p_hdr) - path[i].p_idx++; + ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx, + EXT_MAX_INDEX(path[i].p_hdr)); + ext4_ext_show_move(inode, path, newblock, i); if (m) { - memmove(++fidx, path[i].p_idx - m, + memmove(++fidx, path[i].p_idx, sizeof(struct ext4_extent_idx) * m); le16_add_cpu(&neh->eh_entries, m); } @@ -1056,8 +1073,9 @@ cleanup: * just created block */ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, - struct ext4_extent *newext) + unsigned int flags, + struct ext4_ext_path *path, + struct ext4_extent *newext) { struct ext4_ext_path *curp = path; struct ext4_extent_header *neh; @@ -1065,7 +1083,8 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, ext4_fsblk_t newblock; int err = 0; - newblock = ext4_ext_new_meta_block(handle, inode, path, newext, &err); + newblock = ext4_ext_new_meta_block(handle, inode, path, + newext, &err, flags); if (newblock == 0) return err; @@ -1140,8 +1159,9 @@ out: * if no free index is found, then it requests in-depth growing. */ static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, - struct ext4_extent *newext) + unsigned int flags, + struct ext4_ext_path *path, + struct ext4_extent *newext) { struct ext4_ext_path *curp; int depth, i, err = 0; @@ -1161,7 +1181,7 @@ repeat: if (EXT_HAS_FREE_INDEX(curp)) { /* if we found index with free entry, then use that * entry: create all needed subtree and add new leaf */ - err = ext4_ext_split(handle, inode, path, newext, i); + err = ext4_ext_split(handle, inode, flags, path, newext, i); if (err) goto out; @@ -1174,7 +1194,8 @@ repeat: err = PTR_ERR(path); } else { /* tree is full, time to grow in depth */ - err = ext4_ext_grow_indepth(handle, inode, path, newext); + err = ext4_ext_grow_indepth(handle, inode, flags, + path, newext); if (err) goto out; @@ -1387,7 +1408,7 @@ got_index: /* * ext4_ext_next_allocated_block: - * returns allocated block in subsequent extent or EXT_MAX_BLOCK. + * returns allocated block in subsequent extent or EXT_MAX_BLOCKS. * NOTE: it considers block number from index entry as * allocated block. Thus, index entries have to be consistent * with leaves. @@ -1401,7 +1422,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) depth = path->p_depth; if (depth == 0 && path->p_ext == NULL) - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; while (depth >= 0) { if (depth == path->p_depth) { @@ -1418,12 +1439,12 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) depth--; } - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; } /* * ext4_ext_next_leaf_block: - * returns first allocated block from next leaf or EXT_MAX_BLOCK + * returns first allocated block from next leaf or EXT_MAX_BLOCKS */ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, struct ext4_ext_path *path) @@ -1435,7 +1456,7 @@ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, /* zero-tree has no leaf blocks at all */ if (depth == 0) - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; /* go to index block */ depth--; @@ -1448,7 +1469,7 @@ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, depth--; } - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; } /* @@ -1563,7 +1584,7 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, * Returns 0 if the extents (ex and ex+1) were _not_ merged and returns * 1 if they got merged. */ -static int ext4_ext_try_to_merge(struct inode *inode, +static int ext4_ext_try_to_merge_right(struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *ex) { @@ -1603,6 +1624,31 @@ static int ext4_ext_try_to_merge(struct inode *inode, } /* + * This function tries to merge the @ex extent to neighbours in the tree. + * return 1 if merge left else 0. + */ +static int ext4_ext_try_to_merge(struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *ex) { + struct ext4_extent_header *eh; + unsigned int depth; + int merge_done = 0; + int ret = 0; + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + eh = path[depth].p_hdr; + + if (ex > EXT_FIRST_EXTENT(eh)) + merge_done = ext4_ext_try_to_merge_right(inode, path, ex - 1); + + if (!merge_done) + ret = ext4_ext_try_to_merge_right(inode, path, ex); + + return ret; +} + +/* * check if a portion of the "newext" extent overlaps with an * existing extent. * @@ -1631,13 +1677,13 @@ static unsigned int ext4_ext_check_overlap(struct inode *inode, */ if (b2 < b1) { b2 = ext4_ext_next_allocated_block(path); - if (b2 == EXT_MAX_BLOCK) + if (b2 == EXT_MAX_BLOCKS) goto out; } /* check for wrap through zero on extent logical start block*/ if (b1 + len1 < b1) { - len1 = EXT_MAX_BLOCK - b1; + len1 = EXT_MAX_BLOCKS - b1; newext->ee_len = cpu_to_le16(len1); ret = 1; } @@ -1668,6 +1714,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, int depth, len, err; ext4_lblk_t next; unsigned uninitialized = 0; + int flags = 0; if (unlikely(ext4_ext_get_actual_len(newext) == 0)) { EXT4_ERROR_INODE(inode, "ext4_ext_get_actual_len(newext) == 0"); @@ -1720,7 +1767,7 @@ repeat: fex = EXT_LAST_EXTENT(eh); next = ext4_ext_next_leaf_block(inode, path); if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block) - && next != EXT_MAX_BLOCK) { + && next != EXT_MAX_BLOCKS) { ext_debug("next leaf block - %d\n", next); BUG_ON(npath != NULL); npath = ext4_ext_find_extent(inode, next, NULL); @@ -1742,7 +1789,9 @@ repeat: * There is no free space in the found leaf. * We're gonna add a new leaf in the tree. */ - err = ext4_ext_create_new_leaf(handle, inode, path, newext); + if (flag & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) + flags = EXT4_MB_USE_ROOT_BLOCKS; + err = ext4_ext_create_new_leaf(handle, inode, flags, path, newext); if (err) goto cleanup; depth = ext_depth(inode); @@ -1838,7 +1887,7 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, BUG_ON(func == NULL); BUG_ON(inode == NULL); - while (block < last && block != EXT_MAX_BLOCK) { + while (block < last && block != EXT_MAX_BLOCKS) { num = last - block; /* find extent for this block */ down_read(&EXT4_I(inode)->i_data_sem); @@ -1909,7 +1958,7 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, err = -EIO; break; } - err = func(inode, path, &cbex, ex, cbdata); + err = func(inode, next, &cbex, ex, cbdata); ext4_ext_drop_refs(path); if (err < 0) @@ -1971,7 +2020,7 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, if (ex == NULL) { /* there is no extent yet, so gap is [0;-] */ lblock = 0; - len = EXT_MAX_BLOCK; + len = EXT_MAX_BLOCKS; ext_debug("cache gap(whole file):"); } else if (block < le32_to_cpu(ex->ee_block)) { lblock = block; @@ -2003,13 +2052,25 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, } /* + * ext4_ext_in_cache() + * Checks to see if the given block is in the cache. + * If it is, the cached extent is stored in the given + * cache extent pointer. If the cached extent is a hole, + * this routine should be used instead of + * ext4_ext_in_cache if the calling function needs to + * know the size of the hole. + * + * @inode: The files inode + * @block: The block to look for in the cache + * @ex: Pointer where the cached extent will be stored + * if it contains block + * * Return 0 if cache is invalid; 1 if the cache is valid */ -static int -ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, - struct ext4_extent *ex) -{ +static int ext4_ext_check_cache(struct inode *inode, ext4_lblk_t block, + struct ext4_ext_cache *ex){ struct ext4_ext_cache *cex; + struct ext4_sb_info *sbi; int ret = 0; /* @@ -2017,26 +2078,60 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, */ spin_lock(&EXT4_I(inode)->i_block_reservation_lock); cex = &EXT4_I(inode)->i_cached_extent; + sbi = EXT4_SB(inode->i_sb); /* has cache valid data? */ if (cex->ec_len == 0) goto errout; if (in_range(block, cex->ec_block, cex->ec_len)) { - ex->ee_block = cpu_to_le32(cex->ec_block); - ext4_ext_store_pblock(ex, cex->ec_start); - ex->ee_len = cpu_to_le16(cex->ec_len); + memcpy(ex, cex, sizeof(struct ext4_ext_cache)); ext_debug("%u cached by %u:%u:%llu\n", block, cex->ec_block, cex->ec_len, cex->ec_start); ret = 1; } errout: + if (!ret) + sbi->extent_cache_misses++; + else + sbi->extent_cache_hits++; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); return ret; } /* + * ext4_ext_in_cache() + * Checks to see if the given block is in the cache. + * If it is, the cached extent is stored in the given + * extent pointer. + * + * @inode: The files inode + * @block: The block to look for in the cache + * @ex: Pointer where the cached extent will be stored + * if it contains block + * + * Return 0 if cache is invalid; 1 if the cache is valid + */ +static int +ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block, + struct ext4_extent *ex) +{ + struct ext4_ext_cache cex; + int ret = 0; + + if (ext4_ext_check_cache(inode, block, &cex)) { + ex->ee_block = cpu_to_le32(cex.ec_block); + ext4_ext_store_pblock(ex, cex.ec_start); + ex->ee_len = cpu_to_le16(cex.ec_len); + ret = 1; + } + + return ret; +} + + +/* * ext4_ext_rm_idx: * removes index from the index block. * It's used in truncate case only, thus all requests are for @@ -2163,8 +2258,16 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, ext4_free_blocks(handle, inode, NULL, start, num, flags); } else if (from == le32_to_cpu(ex->ee_block) && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) { - printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n", - from, to, le32_to_cpu(ex->ee_block), ee_len); + /* head removal */ + ext4_lblk_t num; + ext4_fsblk_t start; + + num = to - from; + start = ext4_ext_pblock(ex); + + ext_debug("free first %u blocks starting %llu\n", num, start); + ext4_free_blocks(handle, inode, 0, start, num, flags); + } else { printk(KERN_INFO "strange request: removal(2) " "%u-%u from %u:%u\n", @@ -2173,9 +2276,22 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, return 0; } + +/* + * ext4_ext_rm_leaf() Removes the extents associated with the + * blocks appearing between "start" and "end", and splits the extents + * if "start" and "end" appear in the same extent + * + * @handle: The journal handle + * @inode: The files inode + * @path: The path to the leaf + * @start: The first block to remove + * @end: The last block to remove + */ static int ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, ext4_lblk_t start) + struct ext4_ext_path *path, ext4_lblk_t start, + ext4_lblk_t end) { int err = 0, correct_index = 0; int depth = ext_depth(inode), credits; @@ -2186,6 +2302,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, unsigned short ex_ee_len; unsigned uninitialized = 0; struct ext4_extent *ex; + struct ext4_map_blocks map; /* the header must be checked already in ext4_ext_remove_space() */ ext_debug("truncate since %u in leaf\n", start); @@ -2215,31 +2332,95 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, path[depth].p_ext = ex; a = ex_ee_block > start ? ex_ee_block : start; - b = ex_ee_block + ex_ee_len - 1 < EXT_MAX_BLOCK ? - ex_ee_block + ex_ee_len - 1 : EXT_MAX_BLOCK; + b = ex_ee_block+ex_ee_len - 1 < end ? + ex_ee_block+ex_ee_len - 1 : end; ext_debug(" border %u:%u\n", a, b); - if (a != ex_ee_block && b != ex_ee_block + ex_ee_len - 1) { - block = 0; - num = 0; - BUG(); + /* If this extent is beyond the end of the hole, skip it */ + if (end <= ex_ee_block) { + ex--; + ex_ee_block = le32_to_cpu(ex->ee_block); + ex_ee_len = ext4_ext_get_actual_len(ex); + continue; + } else if (a != ex_ee_block && + b != ex_ee_block + ex_ee_len - 1) { + /* + * If this is a truncate, then this condition should + * never happen because at least one of the end points + * needs to be on the edge of the extent. + */ + if (end == EXT_MAX_BLOCKS - 1) { + ext_debug(" bad truncate %u:%u\n", + start, end); + block = 0; + num = 0; + err = -EIO; + goto out; + } + /* + * else this is a hole punch, so the extent needs to + * be split since neither edge of the hole is on the + * extent edge + */ + else{ + map.m_pblk = ext4_ext_pblock(ex); + map.m_lblk = ex_ee_block; + map.m_len = b - ex_ee_block; + + err = ext4_split_extent(handle, + inode, path, &map, 0, + EXT4_GET_BLOCKS_PUNCH_OUT_EXT | + EXT4_GET_BLOCKS_PRE_IO); + + if (err < 0) + goto out; + + ex_ee_len = ext4_ext_get_actual_len(ex); + + b = ex_ee_block+ex_ee_len - 1 < end ? + ex_ee_block+ex_ee_len - 1 : end; + + /* Then remove tail of this extent */ + block = ex_ee_block; + num = a - block; + } } else if (a != ex_ee_block) { /* remove tail of the extent */ block = ex_ee_block; num = a - block; } else if (b != ex_ee_block + ex_ee_len - 1) { /* remove head of the extent */ - block = a; - num = b - a; - /* there is no "make a hole" API yet */ - BUG(); + block = b; + num = ex_ee_block + ex_ee_len - b; + + /* + * If this is a truncate, this condition + * should never happen + */ + if (end == EXT_MAX_BLOCKS - 1) { + ext_debug(" bad truncate %u:%u\n", + start, end); + err = -EIO; + goto out; + } } else { /* remove whole extent: excellent! */ block = ex_ee_block; num = 0; - BUG_ON(a != ex_ee_block); - BUG_ON(b != ex_ee_block + ex_ee_len - 1); + if (a != ex_ee_block) { + ext_debug(" bad truncate %u:%u\n", + start, end); + err = -EIO; + goto out; + } + + if (b != ex_ee_block + ex_ee_len - 1) { + ext_debug(" bad truncate %u:%u\n", + start, end); + err = -EIO; + goto out; + } } /* @@ -2270,7 +2451,13 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (num == 0) { /* this extent is removed; mark slot entirely unused */ ext4_ext_store_pblock(ex, 0); - le16_add_cpu(&eh->eh_entries, -1); + } else if (block != ex_ee_block) { + /* + * If this was a head removal, then we need to update + * the physical block since it is now at a different + * location + */ + ext4_ext_store_pblock(ex, ext4_ext_pblock(ex) + (b-a)); } ex->ee_block = cpu_to_le32(block); @@ -2286,6 +2473,27 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, if (err) goto out; + /* + * If the extent was completely released, + * we need to remove it from the leaf + */ + if (num == 0) { + if (end != EXT_MAX_BLOCKS - 1) { + /* + * For hole punching, we need to scoot all the + * extents up when an extent is removed so that + * we dont have blank extents in the middle + */ + memmove(ex, ex+1, (EXT_LAST_EXTENT(eh) - ex) * + sizeof(struct ext4_extent)); + + /* Now get rid of the one at the end */ + memset(EXT_LAST_EXTENT(eh), 0, + sizeof(struct ext4_extent)); + } + le16_add_cpu(&eh->eh_entries, -1); + } + ext_debug("new extent: %u:%u:%llu\n", block, num, ext4_ext_pblock(ex)); ex--; @@ -2326,7 +2534,8 @@ ext4_ext_more_to_rm(struct ext4_ext_path *path) return 1; } -static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start) +static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, + ext4_lblk_t end) { struct super_block *sb = inode->i_sb; int depth = ext_depth(inode); @@ -2365,7 +2574,8 @@ again: while (i >= 0 && err == 0) { if (i == depth) { /* this is leaf block */ - err = ext4_ext_rm_leaf(handle, inode, path, start); + err = ext4_ext_rm_leaf(handle, inode, path, + start, end); /* root level has p_bh == NULL, brelse() eats this */ brelse(path[i].p_bh); path[i].p_bh = NULL; @@ -2529,6 +2739,195 @@ static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) return ret; } +/* + * used by extent splitting. + */ +#define EXT4_EXT_MAY_ZEROOUT 0x1 /* safe to zeroout if split fails \ + due to ENOSPC */ +#define EXT4_EXT_MARK_UNINIT1 0x2 /* mark first half uninitialized */ +#define EXT4_EXT_MARK_UNINIT2 0x4 /* mark second half uninitialized */ + +/* + * ext4_split_extent_at() splits an extent at given block. + * + * @handle: the journal handle + * @inode: the file inode + * @path: the path to the extent + * @split: the logical block where the extent is splitted. + * @split_flags: indicates if the extent could be zeroout if split fails, and + * the states(init or uninit) of new extents. + * @flags: flags used to insert new extent to extent tree. + * + * + * Splits extent [a, b] into two extents [a, @split) and [@split, b], states + * of which are deterimined by split_flag. + * + * There are two cases: + * a> the extent are splitted into two extent. + * b> split is not needed, and just mark the extent. + * + * return 0 on success. + */ +static int ext4_split_extent_at(handle_t *handle, + struct inode *inode, + struct ext4_ext_path *path, + ext4_lblk_t split, + int split_flag, + int flags) +{ + ext4_fsblk_t newblock; + ext4_lblk_t ee_block; + struct ext4_extent *ex, newex, orig_ex; + struct ext4_extent *ex2 = NULL; + unsigned int ee_len, depth; + int err = 0; + + ext_debug("ext4_split_extents_at: inode %lu, logical" + "block %llu\n", inode->i_ino, (unsigned long long)split); + + ext4_ext_show_leaf(inode, path); + + depth = ext_depth(inode); + ex = path[depth].p_ext; + ee_block = le32_to_cpu(ex->ee_block); + ee_len = ext4_ext_get_actual_len(ex); + newblock = split - ee_block + ext4_ext_pblock(ex); + + BUG_ON(split < ee_block || split >= (ee_block + ee_len)); + + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) + goto out; + + if (split == ee_block) { + /* + * case b: block @split is the block that the extent begins with + * then we just change the state of the extent, and splitting + * is not needed. + */ + if (split_flag & EXT4_EXT_MARK_UNINIT2) + ext4_ext_mark_uninitialized(ex); + else + ext4_ext_mark_initialized(ex); + + if (!(flags & EXT4_GET_BLOCKS_PRE_IO)) + ext4_ext_try_to_merge(inode, path, ex); + + err = ext4_ext_dirty(handle, inode, path + depth); + goto out; + } + + /* case a */ + memcpy(&orig_ex, ex, sizeof(orig_ex)); + ex->ee_len = cpu_to_le16(split - ee_block); + if (split_flag & EXT4_EXT_MARK_UNINIT1) + ext4_ext_mark_uninitialized(ex); + + /* + * path may lead to new leaf, not to original leaf any more + * after ext4_ext_insert_extent() returns, + */ + err = ext4_ext_dirty(handle, inode, path + depth); + if (err) + goto fix_extent_len; + + ex2 = &newex; + ex2->ee_block = cpu_to_le32(split); + ex2->ee_len = cpu_to_le16(ee_len - (split - ee_block)); + ext4_ext_store_pblock(ex2, newblock); + if (split_flag & EXT4_EXT_MARK_UNINIT2) + ext4_ext_mark_uninitialized(ex2); + + err = ext4_ext_insert_extent(handle, inode, path, &newex, flags); + if (err == -ENOSPC && (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + err = ext4_ext_zeroout(inode, &orig_ex); + if (err) + goto fix_extent_len; + /* update the extent length and mark as initialized */ + ex->ee_len = cpu_to_le32(ee_len); + ext4_ext_try_to_merge(inode, path, ex); + err = ext4_ext_dirty(handle, inode, path + depth); + goto out; + } else if (err) + goto fix_extent_len; + +out: + ext4_ext_show_leaf(inode, path); + return err; + +fix_extent_len: + ex->ee_len = orig_ex.ee_len; + ext4_ext_dirty(handle, inode, path + depth); + return err; +} + +/* + * ext4_split_extents() splits an extent and mark extent which is covered + * by @map as split_flags indicates + * + * It may result in splitting the extent into multiple extents (upto three) + * There are three possibilities: + * a> There is no split required + * b> Splits in two extents: Split is happening at either end of the extent + * c> Splits in three extents: Somone is splitting in middle of the extent + * + */ +static int ext4_split_extent(handle_t *handle, + struct inode *inode, + struct ext4_ext_path *path, + struct ext4_map_blocks *map, + int split_flag, + int flags) +{ + ext4_lblk_t ee_block; + struct ext4_extent *ex; + unsigned int ee_len, depth; + int err = 0; + int uninitialized; + int split_flag1, flags1; + + depth = ext_depth(inode); + ex = path[depth].p_ext; + ee_block = le32_to_cpu(ex->ee_block); + ee_len = ext4_ext_get_actual_len(ex); + uninitialized = ext4_ext_is_uninitialized(ex); + + if (map->m_lblk + map->m_len < ee_block + ee_len) { + split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT ? + EXT4_EXT_MAY_ZEROOUT : 0; + flags1 = flags | EXT4_GET_BLOCKS_PRE_IO; + if (uninitialized) + split_flag1 |= EXT4_EXT_MARK_UNINIT1 | + EXT4_EXT_MARK_UNINIT2; + err = ext4_split_extent_at(handle, inode, path, + map->m_lblk + map->m_len, split_flag1, flags1); + if (err) + goto out; + } + + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, map->m_lblk, path); + if (IS_ERR(path)) + return PTR_ERR(path); + + if (map->m_lblk >= ee_block) { + split_flag1 = split_flag & EXT4_EXT_MAY_ZEROOUT ? + EXT4_EXT_MAY_ZEROOUT : 0; + if (uninitialized) + split_flag1 |= EXT4_EXT_MARK_UNINIT1; + if (split_flag & EXT4_EXT_MARK_UNINIT2) + split_flag1 |= EXT4_EXT_MARK_UNINIT2; + err = ext4_split_extent_at(handle, inode, path, + map->m_lblk, split_flag1, flags); + if (err) + goto out; + } + + ext4_ext_show_leaf(inode, path); +out: + return err ? err : map->m_len; +} + #define EXT4_EXT_ZERO_LEN 7 /* * This function is called by ext4_ext_map_blocks() if someone tries to write @@ -2545,17 +2944,13 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, struct ext4_map_blocks *map, struct ext4_ext_path *path) { - struct ext4_extent *ex, newex, orig_ex; - struct ext4_extent *ex1 = NULL; - struct ext4_extent *ex2 = NULL; - struct ext4_extent *ex3 = NULL; - struct ext4_extent_header *eh; + struct ext4_map_blocks split_map; + struct ext4_extent zero_ex; + struct ext4_extent *ex; ext4_lblk_t ee_block, eof_block; unsigned int allocated, ee_len, depth; - ext4_fsblk_t newblock; int err = 0; - int ret = 0; - int may_zeroout; + int split_flag = 0; ext_debug("ext4_ext_convert_to_initialized: inode %lu, logical" "block %llu, max_blocks %u\n", inode->i_ino, @@ -2567,280 +2962,86 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, eof_block = map->m_lblk + map->m_len; depth = ext_depth(inode); - eh = path[depth].p_hdr; ex = path[depth].p_ext; ee_block = le32_to_cpu(ex->ee_block); ee_len = ext4_ext_get_actual_len(ex); allocated = ee_len - (map->m_lblk - ee_block); - newblock = map->m_lblk - ee_block + ext4_ext_pblock(ex); - - ex2 = ex; - orig_ex.ee_block = ex->ee_block; - orig_ex.ee_len = cpu_to_le16(ee_len); - ext4_ext_store_pblock(&orig_ex, ext4_ext_pblock(ex)); + WARN_ON(map->m_lblk < ee_block); /* * It is safe to convert extent to initialized via explicit * zeroout only if extent is fully insde i_size or new_size. */ - may_zeroout = ee_block + ee_len <= eof_block; + split_flag |= ee_block + ee_len <= eof_block ? EXT4_EXT_MAY_ZEROOUT : 0; - err = ext4_ext_get_access(handle, inode, path + depth); - if (err) - goto out; /* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */ - if (ee_len <= 2*EXT4_EXT_ZERO_LEN && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); + if (ee_len <= 2*EXT4_EXT_ZERO_LEN && + (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + err = ext4_ext_zeroout(inode, ex); if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zeroed the full extent */ - return allocated; - } - - /* ex1: ee_block to map->m_lblk - 1 : uninitialized */ - if (map->m_lblk > ee_block) { - ex1 = ex; - ex1->ee_len = cpu_to_le16(map->m_lblk - ee_block); - ext4_ext_mark_uninitialized(ex1); - ex2 = &newex; - } - /* - * for sanity, update the length of the ex2 extent before - * we insert ex3, if ex1 is NULL. This is to avoid temporary - * overlap of blocks. - */ - if (!ex1 && allocated > map->m_len) - ex2->ee_len = cpu_to_le16(map->m_len); - /* ex3: to ee_block + ee_len : uninitialised */ - if (allocated > map->m_len) { - unsigned int newdepth; - /* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */ - if (allocated <= EXT4_EXT_ZERO_LEN && may_zeroout) { - /* - * map->m_lblk == ee_block is handled by the zerouout - * at the beginning. - * Mark first half uninitialized. - * Mark second half initialized and zero out the - * initialized extent - */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = cpu_to_le16(ee_len - allocated); - ext4_ext_mark_uninitialized(ex); - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - - ex3 = &newex; - ex3->ee_block = cpu_to_le32(map->m_lblk); - ext4_ext_store_pblock(ex3, newblock); - ex3->ee_len = cpu_to_le16(allocated); - err = ext4_ext_insert_extent(handle, inode, path, - ex3, 0); - if (err == -ENOSPC) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, - ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* blocks available from map->m_lblk */ - return allocated; - - } else if (err) - goto fix_extent_len; - - /* - * We need to zero out the second half because - * an fallocate request can update file size and - * converting the second half to initialized extent - * implies that we can leak some junk data to user - * space. - */ - err = ext4_ext_zeroout(inode, ex3); - if (err) { - /* - * We should actually mark the - * second half as uninit and return error - * Insert would have changed the extent - */ - depth = ext_depth(inode); - ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, map->m_lblk, - path); - if (IS_ERR(path)) { - err = PTR_ERR(path); - return err; - } - /* get the second half extent details */ - ex = path[depth].p_ext; - err = ext4_ext_get_access(handle, inode, - path + depth); - if (err) - return err; - ext4_ext_mark_uninitialized(ex); - ext4_ext_dirty(handle, inode, path + depth); - return err; - } - - /* zeroed the second half */ - return allocated; - } - ex3 = &newex; - ex3->ee_block = cpu_to_le32(map->m_lblk + map->m_len); - ext4_ext_store_pblock(ex3, newblock + map->m_len); - ex3->ee_len = cpu_to_le16(allocated - map->m_len); - ext4_ext_mark_uninitialized(ex3); - err = ext4_ext_insert_extent(handle, inode, path, ex3, 0); - if (err == -ENOSPC && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zeroed the full extent */ - /* blocks available from map->m_lblk */ - return allocated; - - } else if (err) - goto fix_extent_len; - /* - * The depth, and hence eh & ex might change - * as part of the insert above. - */ - newdepth = ext_depth(inode); - /* - * update the extent length after successful insert of the - * split extent - */ - ee_len -= ext4_ext_get_actual_len(ex3); - orig_ex.ee_len = cpu_to_le16(ee_len); - may_zeroout = ee_block + ee_len <= eof_block; - - depth = newdepth; - ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, map->m_lblk, path); - if (IS_ERR(path)) { - err = PTR_ERR(path); goto out; - } - eh = path[depth].p_hdr; - ex = path[depth].p_ext; - if (ex2 != &newex) - ex2 = ex; err = ext4_ext_get_access(handle, inode, path + depth); if (err) goto out; - - allocated = map->m_len; - - /* If extent has less than EXT4_EXT_ZERO_LEN and we are trying - * to insert a extent in the middle zerout directly - * otherwise give the extent a chance to merge to left - */ - if (le16_to_cpu(orig_ex.ee_len) <= EXT4_EXT_ZERO_LEN && - map->m_lblk != ee_block && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zero out the first half */ - /* blocks available from map->m_lblk */ - return allocated; - } - } - /* - * If there was a change of depth as part of the - * insertion of ex3 above, we need to update the length - * of the ex1 extent again here - */ - if (ex1 && ex1 != ex) { - ex1 = ex; - ex1->ee_len = cpu_to_le16(map->m_lblk - ee_block); - ext4_ext_mark_uninitialized(ex1); - ex2 = &newex; - } - /* ex2: map->m_lblk to map->m_lblk + maxblocks-1 : initialised */ - ex2->ee_block = cpu_to_le32(map->m_lblk); - ext4_ext_store_pblock(ex2, newblock); - ex2->ee_len = cpu_to_le16(allocated); - if (ex2 != ex) - goto insert; - /* - * New (initialized) extent starts from the first block - * in the current extent. i.e., ex2 == ex - * We have to see if it can be merged with the extent - * on the left. - */ - if (ex2 > EXT_FIRST_EXTENT(eh)) { - /* - * To merge left, pass "ex2 - 1" to try_to_merge(), - * since it merges towards right _only_. - */ - ret = ext4_ext_try_to_merge(inode, path, ex2 - 1); - if (ret) { - err = ext4_ext_correct_indexes(handle, inode, path); - if (err) - goto out; - depth = ext_depth(inode); - ex2--; - } + ext4_ext_mark_initialized(ex); + ext4_ext_try_to_merge(inode, path, ex); + err = ext4_ext_dirty(handle, inode, path + depth); + goto out; } + /* - * Try to Merge towards right. This might be required - * only when the whole extent is being written to. - * i.e. ex2 == ex and ex3 == NULL. + * four cases: + * 1. split the extent into three extents. + * 2. split the extent into two extents, zeroout the first half. + * 3. split the extent into two extents, zeroout the second half. + * 4. split the extent into two extents with out zeroout. */ - if (!ex3) { - ret = ext4_ext_try_to_merge(inode, path, ex2); - if (ret) { - err = ext4_ext_correct_indexes(handle, inode, path); + split_map.m_lblk = map->m_lblk; + split_map.m_len = map->m_len; + + if (allocated > map->m_len) { + if (allocated <= EXT4_EXT_ZERO_LEN && + (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + /* case 3 */ + zero_ex.ee_block = + cpu_to_le32(map->m_lblk); + zero_ex.ee_len = cpu_to_le16(allocated); + ext4_ext_store_pblock(&zero_ex, + ext4_ext_pblock(ex) + map->m_lblk - ee_block); + err = ext4_ext_zeroout(inode, &zero_ex); if (err) goto out; + split_map.m_lblk = map->m_lblk; + split_map.m_len = allocated; + } else if ((map->m_lblk - ee_block + map->m_len < + EXT4_EXT_ZERO_LEN) && + (EXT4_EXT_MAY_ZEROOUT & split_flag)) { + /* case 2 */ + if (map->m_lblk != ee_block) { + zero_ex.ee_block = ex->ee_block; + zero_ex.ee_len = cpu_to_le16(map->m_lblk - + ee_block); + ext4_ext_store_pblock(&zero_ex, + ext4_ext_pblock(ex)); + err = ext4_ext_zeroout(inode, &zero_ex); + if (err) + goto out; + } + + split_map.m_lblk = ee_block; + split_map.m_len = map->m_lblk - ee_block + map->m_len; + allocated = map->m_len; } } - /* Mark modified extent as dirty */ - err = ext4_ext_dirty(handle, inode, path + depth); - goto out; -insert: - err = ext4_ext_insert_extent(handle, inode, path, &newex, 0); - if (err == -ENOSPC && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zero out the first half */ - return allocated; - } else if (err) - goto fix_extent_len; + + allocated = ext4_split_extent(handle, inode, path, + &split_map, split_flag, 0); + if (allocated < 0) + err = allocated; + out: - ext4_ext_show_leaf(inode, path); return err ? err : allocated; - -fix_extent_len: - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_mark_uninitialized(ex); - ext4_ext_dirty(handle, inode, path + depth); - return err; } /* @@ -2871,15 +3072,11 @@ static int ext4_split_unwritten_extents(handle_t *handle, struct ext4_ext_path *path, int flags) { - struct ext4_extent *ex, newex, orig_ex; - struct ext4_extent *ex1 = NULL; - struct ext4_extent *ex2 = NULL; - struct ext4_extent *ex3 = NULL; - ext4_lblk_t ee_block, eof_block; - unsigned int allocated, ee_len, depth; - ext4_fsblk_t newblock; - int err = 0; - int may_zeroout; + ext4_lblk_t eof_block; + ext4_lblk_t ee_block; + struct ext4_extent *ex; + unsigned int ee_len; + int split_flag = 0, depth; ext_debug("ext4_split_unwritten_extents: inode %lu, logical" "block %llu, max_blocks %u\n", inode->i_ino, @@ -2889,156 +3086,22 @@ static int ext4_split_unwritten_extents(handle_t *handle, inode->i_sb->s_blocksize_bits; if (eof_block < map->m_lblk + map->m_len) eof_block = map->m_lblk + map->m_len; - - depth = ext_depth(inode); - ex = path[depth].p_ext; - ee_block = le32_to_cpu(ex->ee_block); - ee_len = ext4_ext_get_actual_len(ex); - allocated = ee_len - (map->m_lblk - ee_block); - newblock = map->m_lblk - ee_block + ext4_ext_pblock(ex); - - ex2 = ex; - orig_ex.ee_block = ex->ee_block; - orig_ex.ee_len = cpu_to_le16(ee_len); - ext4_ext_store_pblock(&orig_ex, ext4_ext_pblock(ex)); - /* * It is safe to convert extent to initialized via explicit * zeroout only if extent is fully insde i_size or new_size. */ - may_zeroout = ee_block + ee_len <= eof_block; - - /* - * If the uninitialized extent begins at the same logical - * block where the write begins, and the write completely - * covers the extent, then we don't need to split it. - */ - if ((map->m_lblk == ee_block) && (allocated <= map->m_len)) - return allocated; - - err = ext4_ext_get_access(handle, inode, path + depth); - if (err) - goto out; - /* ex1: ee_block to map->m_lblk - 1 : uninitialized */ - if (map->m_lblk > ee_block) { - ex1 = ex; - ex1->ee_len = cpu_to_le16(map->m_lblk - ee_block); - ext4_ext_mark_uninitialized(ex1); - ex2 = &newex; - } - /* - * for sanity, update the length of the ex2 extent before - * we insert ex3, if ex1 is NULL. This is to avoid temporary - * overlap of blocks. - */ - if (!ex1 && allocated > map->m_len) - ex2->ee_len = cpu_to_le16(map->m_len); - /* ex3: to ee_block + ee_len : uninitialised */ - if (allocated > map->m_len) { - unsigned int newdepth; - ex3 = &newex; - ex3->ee_block = cpu_to_le32(map->m_lblk + map->m_len); - ext4_ext_store_pblock(ex3, newblock + map->m_len); - ex3->ee_len = cpu_to_le16(allocated - map->m_len); - ext4_ext_mark_uninitialized(ex3); - err = ext4_ext_insert_extent(handle, inode, path, ex3, flags); - if (err == -ENOSPC && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zeroed the full extent */ - /* blocks available from map->m_lblk */ - return allocated; - - } else if (err) - goto fix_extent_len; - /* - * The depth, and hence eh & ex might change - * as part of the insert above. - */ - newdepth = ext_depth(inode); - /* - * update the extent length after successful insert of the - * split extent - */ - ee_len -= ext4_ext_get_actual_len(ex3); - orig_ex.ee_len = cpu_to_le16(ee_len); - may_zeroout = ee_block + ee_len <= eof_block; - - depth = newdepth; - ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, map->m_lblk, path); - if (IS_ERR(path)) { - err = PTR_ERR(path); - goto out; - } - ex = path[depth].p_ext; - if (ex2 != &newex) - ex2 = ex; + depth = ext_depth(inode); + ex = path[depth].p_ext; + ee_block = le32_to_cpu(ex->ee_block); + ee_len = ext4_ext_get_actual_len(ex); - err = ext4_ext_get_access(handle, inode, path + depth); - if (err) - goto out; + split_flag |= ee_block + ee_len <= eof_block ? EXT4_EXT_MAY_ZEROOUT : 0; + split_flag |= EXT4_EXT_MARK_UNINIT2; - allocated = map->m_len; - } - /* - * If there was a change of depth as part of the - * insertion of ex3 above, we need to update the length - * of the ex1 extent again here - */ - if (ex1 && ex1 != ex) { - ex1 = ex; - ex1->ee_len = cpu_to_le16(map->m_lblk - ee_block); - ext4_ext_mark_uninitialized(ex1); - ex2 = &newex; - } - /* - * ex2: map->m_lblk to map->m_lblk + map->m_len-1 : to be written - * using direct I/O, uninitialised still. - */ - ex2->ee_block = cpu_to_le32(map->m_lblk); - ext4_ext_store_pblock(ex2, newblock); - ex2->ee_len = cpu_to_le16(allocated); - ext4_ext_mark_uninitialized(ex2); - if (ex2 != ex) - goto insert; - /* Mark modified extent as dirty */ - err = ext4_ext_dirty(handle, inode, path + depth); - ext_debug("out here\n"); - goto out; -insert: - err = ext4_ext_insert_extent(handle, inode, path, &newex, flags); - if (err == -ENOSPC && may_zeroout) { - err = ext4_ext_zeroout(inode, &orig_ex); - if (err) - goto fix_extent_len; - /* update the extent length and mark as initialized */ - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_dirty(handle, inode, path + depth); - /* zero out the first half */ - return allocated; - } else if (err) - goto fix_extent_len; -out: - ext4_ext_show_leaf(inode, path); - return err ? err : allocated; - -fix_extent_len: - ex->ee_block = orig_ex.ee_block; - ex->ee_len = orig_ex.ee_len; - ext4_ext_store_pblock(ex, ext4_ext_pblock(&orig_ex)); - ext4_ext_mark_uninitialized(ex); - ext4_ext_dirty(handle, inode, path + depth); - return err; + flags |= EXT4_GET_BLOCKS_PRE_IO; + return ext4_split_extent(handle, inode, path, map, split_flag, flags); } + static int ext4_convert_unwritten_extents_endio(handle_t *handle, struct inode *inode, struct ext4_ext_path *path) @@ -3047,46 +3110,27 @@ static int ext4_convert_unwritten_extents_endio(handle_t *handle, struct ext4_extent_header *eh; int depth; int err = 0; - int ret = 0; depth = ext_depth(inode); eh = path[depth].p_hdr; ex = path[depth].p_ext; + ext_debug("ext4_convert_unwritten_extents_endio: inode %lu, logical" + "block %llu, max_blocks %u\n", inode->i_ino, + (unsigned long long)le32_to_cpu(ex->ee_block), + ext4_ext_get_actual_len(ex)); + err = ext4_ext_get_access(handle, inode, path + depth); if (err) goto out; /* first mark the extent as initialized */ ext4_ext_mark_initialized(ex); - /* - * We have to see if it can be merged with the extent - * on the left. - */ - if (ex > EXT_FIRST_EXTENT(eh)) { - /* - * To merge left, pass "ex - 1" to try_to_merge(), - * since it merges towards right _only_. - */ - ret = ext4_ext_try_to_merge(inode, path, ex - 1); - if (ret) { - err = ext4_ext_correct_indexes(handle, inode, path); - if (err) - goto out; - depth = ext_depth(inode); - ex--; - } - } - /* - * Try to Merge towards right. + /* note: ext4_ext_correct_indexes() isn't needed here because + * borders are not changed */ - ret = ext4_ext_try_to_merge(inode, path, ex); - if (ret) { - err = ext4_ext_correct_indexes(handle, inode, path); - if (err) - goto out; - depth = ext_depth(inode); - } + ext4_ext_try_to_merge(inode, path, ex); + /* Mark modified extent as dirty */ err = ext4_ext_dirty(handle, inode, path + depth); out: @@ -3302,15 +3346,19 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t newblock = 0; int err = 0, depth, ret; unsigned int allocated = 0; + unsigned int punched_out = 0; + unsigned int result = 0; struct ext4_allocation_request ar; ext4_io_end_t *io = EXT4_I(inode)->cur_aio_dio; + struct ext4_map_blocks punch_map; ext_debug("blocks %u/%u requested for inode %lu\n", map->m_lblk, map->m_len, inode->i_ino); trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags); /* check in cache */ - if (ext4_ext_in_cache(inode, map->m_lblk, &newex)) { + if (ext4_ext_in_cache(inode, map->m_lblk, &newex) && + ((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0)) { if (!newex.ee_start_lo && !newex.ee_start_hi) { if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) { /* @@ -3375,16 +3423,84 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, ext_debug("%u fit into %u:%d -> %llu\n", map->m_lblk, ee_block, ee_len, newblock); - /* Do not put uninitialized extent in the cache */ - if (!ext4_ext_is_uninitialized(ex)) { - ext4_ext_put_in_cache(inode, ee_block, - ee_len, ee_start); - goto out; + if ((flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) == 0) { + /* + * Do not put uninitialized extent + * in the cache + */ + if (!ext4_ext_is_uninitialized(ex)) { + ext4_ext_put_in_cache(inode, ee_block, + ee_len, ee_start); + goto out; + } + ret = ext4_ext_handle_uninitialized_extents( + handle, inode, map, path, flags, + allocated, newblock); + return ret; } - ret = ext4_ext_handle_uninitialized_extents(handle, - inode, map, path, flags, allocated, - newblock); - return ret; + + /* + * Punch out the map length, but only to the + * end of the extent + */ + punched_out = allocated < map->m_len ? + allocated : map->m_len; + + /* + * Sense extents need to be converted to + * uninitialized, they must fit in an + * uninitialized extent + */ + if (punched_out > EXT_UNINIT_MAX_LEN) + punched_out = EXT_UNINIT_MAX_LEN; + + punch_map.m_lblk = map->m_lblk; + punch_map.m_pblk = newblock; + punch_map.m_len = punched_out; + punch_map.m_flags = 0; + + /* Check to see if the extent needs to be split */ + if (punch_map.m_len != ee_len || + punch_map.m_lblk != ee_block) { + + ret = ext4_split_extent(handle, inode, + path, &punch_map, 0, + EXT4_GET_BLOCKS_PUNCH_OUT_EXT | + EXT4_GET_BLOCKS_PRE_IO); + + if (ret < 0) { + err = ret; + goto out2; + } + /* + * find extent for the block at + * the start of the hole + */ + ext4_ext_drop_refs(path); + kfree(path); + + path = ext4_ext_find_extent(inode, + map->m_lblk, NULL); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + goto out2; + } + + depth = ext_depth(inode); + ex = path[depth].p_ext; + ee_len = ext4_ext_get_actual_len(ex); + ee_block = le32_to_cpu(ex->ee_block); + ee_start = ext4_ext_pblock(ex); + + } + + ext4_ext_mark_uninitialized(ex); + + err = ext4_ext_remove_space(inode, map->m_lblk, + map->m_lblk + punched_out); + + goto out2; } } @@ -3446,6 +3562,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, else /* disable in-core preallocation for non-regular files */ ar.flags = 0; + if (flags & EXT4_GET_BLOCKS_NO_NORMALIZE) + ar.flags |= EXT4_MB_HINT_NOPREALLOC; newblock = ext4_mb_new_blocks(handle, &ar, &err); if (!newblock) goto out2; @@ -3529,7 +3647,11 @@ out2: } trace_ext4_ext_map_blocks_exit(inode, map->m_lblk, newblock, map->m_len, err ? err : allocated); - return err ? err : allocated; + + result = (flags & EXT4_GET_BLOCKS_PUNCH_OUT_EXT) ? + punched_out : allocated; + + return err ? err : result; } void ext4_ext_truncate(struct inode *inode) @@ -3577,7 +3699,7 @@ void ext4_ext_truncate(struct inode *inode) last_block = (inode->i_size + sb->s_blocksize - 1) >> EXT4_BLOCK_SIZE_BITS(sb); - err = ext4_ext_remove_space(inode, last_block); + err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1); /* In a multi-transaction truncate, we only make the final * transaction synchronous. @@ -3585,8 +3707,9 @@ void ext4_ext_truncate(struct inode *inode) if (IS_SYNC(inode)) ext4_handle_sync(handle); -out_stop: up_write(&EXT4_I(inode)->i_data_sem); + +out_stop: /* * If this was a simple ftruncate() and the file will remain alive, * then we need to clear up the orphan record which we created above. @@ -3651,10 +3774,6 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) struct ext4_map_blocks map; unsigned int credits, blkbits = inode->i_blkbits; - /* We only support the FALLOC_FL_KEEP_SIZE mode */ - if (mode & ~FALLOC_FL_KEEP_SIZE) - return -EOPNOTSUPP; - /* * currently supporting (pre)allocate mode for extent-based * files _only_ @@ -3662,6 +3781,13 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) return -EOPNOTSUPP; + /* Return error if mode is not supported */ + if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) + return -EOPNOTSUPP; + + if (mode & FALLOC_FL_PUNCH_HOLE) + return ext4_punch_hole(file, offset, len); + trace_ext4_fallocate_enter(inode, offset, len, mode); map.m_lblk = offset >> blkbits; /* @@ -3691,7 +3817,8 @@ retry: break; } ret = ext4_map_blocks(handle, inode, &map, - EXT4_GET_BLOCKS_CREATE_UNINIT_EXT); + EXT4_GET_BLOCKS_CREATE_UNINIT_EXT | + EXT4_GET_BLOCKS_NO_NORMALIZE); if (ret <= 0) { #ifdef EXT4FS_DEBUG WARN_ON(ret <= 0); @@ -3787,14 +3914,13 @@ int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, /* * Callback function called for each extent to gather FIEMAP information. */ -static int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path, +static int ext4_ext_fiemap_cb(struct inode *inode, ext4_lblk_t next, struct ext4_ext_cache *newex, struct ext4_extent *ex, void *data) { __u64 logical; __u64 physical; __u64 length; - loff_t size; __u32 flags = 0; int ret = 0; struct fiemap_extent_info *fieinfo = data; @@ -3822,6 +3948,7 @@ static int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path, pgoff_t last_offset; pgoff_t offset; pgoff_t index; + pgoff_t start_index = 0; struct page **pages = NULL; struct buffer_head *bh = NULL; struct buffer_head *head = NULL; @@ -3848,39 +3975,57 @@ out: kfree(pages); return EXT_CONTINUE; } + index = 0; +next_page: /* Try to find the 1st mapped buffer. */ - end = ((__u64)pages[0]->index << PAGE_SHIFT) >> + end = ((__u64)pages[index]->index << PAGE_SHIFT) >> blksize_bits; - if (!page_has_buffers(pages[0])) + if (!page_has_buffers(pages[index])) goto out; - head = page_buffers(pages[0]); + head = page_buffers(pages[index]); if (!head) goto out; + index++; bh = head; do { - if (buffer_mapped(bh)) { + if (end >= newex->ec_block + + newex->ec_len) + /* The buffer is out of + * the request range. + */ + goto out; + + if (buffer_mapped(bh) && + end >= newex->ec_block) { + start_index = index - 1; /* get the 1st mapped buffer. */ - if (end > newex->ec_block + - newex->ec_len) - /* The buffer is out of - * the request range. - */ - goto out; goto found_mapped_buffer; } + bh = bh->b_this_page; end++; } while (bh != head); - /* No mapped buffer found. */ - goto out; + /* No mapped buffer in the range found in this page, + * We need to look up next page. + */ + if (index >= ret) { + /* There is no page left, but we need to limit + * newex->ec_len. + */ + newex->ec_len = end - newex->ec_block; + goto out; + } + goto next_page; } else { /*Find contiguous delayed buffers. */ if (ret > 0 && pages[0]->index == last_offset) head = page_buffers(pages[0]); bh = head; + index = 1; + start_index = 0; } found_mapped_buffer: @@ -3903,7 +4048,7 @@ found_mapped_buffer: end++; } while (bh != head); - for (index = 1; index < ret; index++) { + for (; index < ret; index++) { if (!page_has_buffers(pages[index])) { bh = NULL; break; @@ -3913,8 +4058,10 @@ found_mapped_buffer: bh = NULL; break; } + if (pages[index]->index != - pages[0]->index + index) { + pages[start_index]->index + index + - start_index) { /* Blocks are not contiguous. */ bh = NULL; break; @@ -3955,8 +4102,7 @@ found_delayed_extent: if (ex && ext4_ext_is_uninitialized(ex)) flags |= FIEMAP_EXTENT_UNWRITTEN; - size = i_size_read(inode); - if (logical + length >= size) + if (next == EXT_MAX_BLOCKS) flags |= FIEMAP_EXTENT_LAST; ret = fiemap_fill_next_extent(fieinfo, logical, physical, @@ -4006,6 +4152,177 @@ static int ext4_xattr_fiemap(struct inode *inode, return (error < 0 ? error : 0); } +/* + * ext4_ext_punch_hole + * + * Punches a hole of "length" bytes in a file starting + * at byte "offset" + * + * @inode: The inode of the file to punch a hole in + * @offset: The starting byte offset of the hole + * @length: The length of the hole + * + * Returns the number of blocks removed or negative on err + */ +int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct super_block *sb = inode->i_sb; + struct ext4_ext_cache cache_ex; + ext4_lblk_t first_block, last_block, num_blocks, iblock, max_blocks; + struct address_space *mapping = inode->i_mapping; + struct ext4_map_blocks map; + handle_t *handle; + loff_t first_block_offset, last_block_offset, block_len; + loff_t first_page, last_page, first_page_offset, last_page_offset; + int ret, credits, blocks_released, err = 0; + + first_block = (offset + sb->s_blocksize - 1) >> + EXT4_BLOCK_SIZE_BITS(sb); + last_block = (offset + length) >> EXT4_BLOCK_SIZE_BITS(sb); + + first_block_offset = first_block << EXT4_BLOCK_SIZE_BITS(sb); + last_block_offset = last_block << EXT4_BLOCK_SIZE_BITS(sb); + + first_page = (offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + last_page = (offset + length) >> PAGE_CACHE_SHIFT; + + first_page_offset = first_page << PAGE_CACHE_SHIFT; + last_page_offset = last_page << PAGE_CACHE_SHIFT; + + /* + * Write out all dirty pages to avoid race conditions + * Then release them. + */ + if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) { + err = filemap_write_and_wait_range(mapping, + first_page_offset == 0 ? 0 : first_page_offset-1, + last_page_offset); + + if (err) + return err; + } + + /* Now release the pages */ + if (last_page_offset > first_page_offset) { + truncate_inode_pages_range(mapping, first_page_offset, + last_page_offset-1); + } + + /* finish any pending end_io work */ + ext4_flush_completed_IO(inode); + + credits = ext4_writepage_trans_blocks(inode); + handle = ext4_journal_start(inode, credits); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + err = ext4_orphan_add(handle, inode); + if (err) + goto out; + + /* + * Now we need to zero out the un block aligned data. + * If the file is smaller than a block, just + * zero out the middle + */ + if (first_block > last_block) + ext4_block_zero_page_range(handle, mapping, offset, length); + else { + /* zero out the head of the hole before the first block */ + block_len = first_block_offset - offset; + if (block_len > 0) + ext4_block_zero_page_range(handle, mapping, + offset, block_len); + + /* zero out the tail of the hole after the last block */ + block_len = offset + length - last_block_offset; + if (block_len > 0) { + ext4_block_zero_page_range(handle, mapping, + last_block_offset, block_len); + } + } + + /* If there are no blocks to remove, return now */ + if (first_block >= last_block) + goto out; + + down_write(&EXT4_I(inode)->i_data_sem); + ext4_ext_invalidate_cache(inode); + ext4_discard_preallocations(inode); + + /* + * Loop over all the blocks and identify blocks + * that need to be punched out + */ + iblock = first_block; + blocks_released = 0; + while (iblock < last_block) { + max_blocks = last_block - iblock; + num_blocks = 1; + memset(&map, 0, sizeof(map)); + map.m_lblk = iblock; + map.m_len = max_blocks; + ret = ext4_ext_map_blocks(handle, inode, &map, + EXT4_GET_BLOCKS_PUNCH_OUT_EXT); + + if (ret > 0) { + blocks_released += ret; + num_blocks = ret; + } else if (ret == 0) { + /* + * If map blocks could not find the block, + * then it is in a hole. If the hole was + * not already cached, then map blocks should + * put it in the cache. So we can get the hole + * out of the cache + */ + memset(&cache_ex, 0, sizeof(cache_ex)); + if ((ext4_ext_check_cache(inode, iblock, &cache_ex)) && + !cache_ex.ec_start) { + + /* The hole is cached */ + num_blocks = cache_ex.ec_block + + cache_ex.ec_len - iblock; + + } else { + /* The block could not be identified */ + err = -EIO; + break; + } + } else { + /* Map blocks error */ + err = ret; + break; + } + + if (num_blocks == 0) { + /* This condition should never happen */ + ext_debug("Block lookup failed"); + err = -EIO; + break; + } + + iblock += num_blocks; + } + + if (blocks_released > 0) { + ext4_ext_invalidate_cache(inode); + ext4_discard_preallocations(inode); + } + + if (IS_SYNC(inode)) + ext4_handle_sync(handle); + + up_write(&EXT4_I(inode)->i_data_sem); + +out: + ext4_orphan_del(handle, inode); + inode->i_mtime = inode->i_ctime = ext4_current_time(inode); + ext4_mark_inode_dirty(handle, inode); + ext4_journal_stop(handle); + return err; +} int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len) { @@ -4028,8 +4345,8 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, start_blk = start >> inode->i_sb->s_blocksize_bits; last_blk = (start + len - 1) >> inode->i_sb->s_blocksize_bits; - if (last_blk >= EXT_MAX_BLOCK) - last_blk = EXT_MAX_BLOCK-1; + if (last_blk >= EXT_MAX_BLOCKS) + last_blk = EXT_MAX_BLOCKS-1; len_blks = ((ext4_lblk_t) last_blk) - start_blk + 1; /* @@ -4042,4 +4359,3 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, return error; } - diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 7b80d54..2c09723 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -272,7 +272,6 @@ const struct file_operations ext4_file_operations = { }; const struct inode_operations ext4_file_inode_operations = { - .truncate = ext4_truncate, .setattr = ext4_setattr, .getattr = ext4_getattr, #ifdef CONFIG_EXT4_FS_XATTR diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index e9473cb..ce66d2f 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -36,7 +36,7 @@ static void dump_completed_IO(struct inode * inode) { -#ifdef EXT4_DEBUG +#ifdef EXT4FS_DEBUG struct list_head *cur, *before, *after; ext4_io_end_t *io, *io0, *io1; unsigned long flags; @@ -172,6 +172,7 @@ int ext4_sync_file(struct file *file, int datasync) journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; int ret; tid_t commit_tid; + bool needs_barrier = false; J_ASSERT(ext4_journal_current_handle() == NULL); @@ -211,22 +212,12 @@ int ext4_sync_file(struct file *file, int datasync) } commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid; - if (jbd2_log_start_commit(journal, commit_tid)) { - /* - * When the journal is on a different device than the - * fs data disk, we need to issue the barrier in - * writeback mode. (In ordered mode, the jbd2 layer - * will take care of issuing the barrier. In - * data=journal, all of the data blocks are written to - * the journal device.) - */ - if (ext4_should_writeback_data(inode) && - (journal->j_fs_dev != journal->j_dev) && - (journal->j_flags & JBD2_BARRIER)) - blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, - NULL); - ret = jbd2_log_wait_commit(journal, commit_tid); - } else if (journal->j_flags & JBD2_BARRIER) + if (journal->j_flags & JBD2_BARRIER && + !jbd2_trans_will_send_data_barrier(journal, commit_tid)) + needs_barrier = true; + jbd2_log_start_commit(journal, commit_tid); + ret = jbd2_log_wait_commit(journal, commit_tid); + if (needs_barrier) blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); out: trace_ext4_sync_file_exit(inode, ret); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index f2fa5e8..e3126c0 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -639,8 +639,8 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, while (target > 0) { count = target; /* allocating blocks for indirect blocks and direct blocks */ - current_block = ext4_new_meta_blocks(handle, inode, - goal, &count, err); + current_block = ext4_new_meta_blocks(handle, inode, goal, + 0, &count, err); if (*err) goto failed_out; @@ -1930,7 +1930,7 @@ repeat: * We do still charge estimated metadata to the sb though; * we cannot afford to run out of free blocks. */ - if (ext4_claim_free_blocks(sbi, md_needed + 1)) { + if (ext4_claim_free_blocks(sbi, md_needed + 1, 0)) { dquot_release_reservation_block(inode, 1); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { yield(); @@ -2634,7 +2634,7 @@ static int ext4_writepage(struct page *page, struct buffer_head *page_bufs = NULL; struct inode *inode = page->mapping->host; - trace_ext4_writepage(inode, page); + trace_ext4_writepage(page); size = i_size_read(inode); if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; @@ -2796,9 +2796,7 @@ static int write_cache_pages_da(struct address_space *mapping, continue; } - if (PageWriteback(page)) - wait_on_page_writeback(page); - + wait_on_page_writeback(page); BUG_ON(PageWriteback(page)); if (mpd->next_page != page->index) @@ -3513,7 +3511,7 @@ retry: loff_t end = offset + iov_length(iov, nr_segs); if (end > isize) - vmtruncate(inode, isize); + ext4_truncate_failed_write(inode); } } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -3916,9 +3914,30 @@ void ext4_set_aops(struct inode *inode) int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from) { + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned length; + unsigned blocksize; + struct inode *inode = mapping->host; + + blocksize = inode->i_sb->s_blocksize; + length = blocksize - (offset & (blocksize - 1)); + + return ext4_block_zero_page_range(handle, mapping, from, length); +} + +/* + * ext4_block_zero_page_range() zeros out a mapping of length 'length' + * starting from file offset 'from'. The range to be zero'd must + * be contained with in one block. If the specified range exceeds + * the end of the block it will be shortened to end of the block + * that cooresponds to 'from' + */ +int ext4_block_zero_page_range(handle_t *handle, + struct address_space *mapping, loff_t from, loff_t length) +{ ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; unsigned offset = from & (PAGE_CACHE_SIZE-1); - unsigned blocksize, length, pos; + unsigned blocksize, max, pos; ext4_lblk_t iblock; struct inode *inode = mapping->host; struct buffer_head *bh; @@ -3931,7 +3950,15 @@ int ext4_block_truncate_page(handle_t *handle, return -EINVAL; blocksize = inode->i_sb->s_blocksize; - length = blocksize - (offset & (blocksize - 1)); + max = blocksize - (offset & (blocksize - 1)); + + /* + * correct length if it does not fall between + * 'from' and the end of the block + */ + if (length > max || length < 0) + length = max; + iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); if (!page_has_buffers(page)) @@ -4380,8 +4407,6 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, int ext4_can_truncate(struct inode *inode) { - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - return 0; if (S_ISREG(inode->i_mode)) return 1; if (S_ISDIR(inode->i_mode)) @@ -4392,6 +4417,31 @@ int ext4_can_truncate(struct inode *inode) } /* + * ext4_punch_hole: punches a hole in a file by releaseing the blocks + * associated with the given offset and length + * + * @inode: File inode + * @offset: The offset where the hole will begin + * @len: The length of the hole + * + * Returns: 0 on sucess or negative on failure + */ + +int ext4_punch_hole(struct file *file, loff_t offset, loff_t length) +{ + struct inode *inode = file->f_path.dentry->d_inode; + if (!S_ISREG(inode->i_mode)) + return -ENOTSUPP; + + if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) { + /* TODO: Add support for non extent hole punching */ + return -ENOTSUPP; + } + + return ext4_ext_punch_hole(file, offset, length); +} + +/* * ext4_truncate() * * We block out ext4_get_block() block instantiations across the entire @@ -4617,7 +4667,7 @@ static int __ext4_get_inode_loc(struct inode *inode, /* * Figure out the offset within the block group inode table */ - inodes_per_block = (EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb)); + inodes_per_block = EXT4_SB(sb)->s_inodes_per_block; inode_offset = ((inode->i_ino - 1) % EXT4_INODES_PER_GROUP(sb)); block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block); @@ -5311,8 +5361,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) if (S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE && - (attr->ia_size < inode->i_size || - (ext4_test_inode_flag(inode, EXT4_INODE_EOFBLOCKS)))) { + (attr->ia_size < inode->i_size)) { handle_t *handle; handle = ext4_journal_start(inode, 3); @@ -5346,14 +5395,15 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) goto err_out; } } - /* ext4_truncate will clear the flag */ - if ((ext4_test_inode_flag(inode, EXT4_INODE_EOFBLOCKS))) - ext4_truncate(inode); } - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) - rc = vmtruncate(inode, attr->ia_size); + if (attr->ia_valid & ATTR_SIZE) { + if (attr->ia_size != i_size_read(inode)) { + truncate_setsize(inode, attr->ia_size); + ext4_truncate(inode); + } else if (ext4_test_inode_flag(inode, EXT4_INODE_EOFBLOCKS)) + ext4_truncate(inode); + } if (!rc) { setattr_copy(inode, attr); @@ -5683,7 +5733,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) * so would cause a commit on atime updates, which we don't bother doing. * We handle synchronous inodes at the highest possible level. */ -void ext4_dirty_inode(struct inode *inode) +void ext4_dirty_inode(struct inode *inode, int flags) { handle_t *handle; @@ -5811,15 +5861,19 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) goto out_unlock; } ret = 0; - if (PageMappedToDisk(page)) - goto out_unlock; + + lock_page(page); + wait_on_page_writeback(page); + if (PageMappedToDisk(page)) { + up_read(&inode->i_alloc_sem); + return VM_FAULT_LOCKED; + } if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; else len = PAGE_CACHE_SIZE; - lock_page(page); /* * return if we have all the buffers mapped. This avoid * the need to call write_begin/write_end which does a @@ -5829,8 +5883,8 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) if (page_has_buffers(page)) { if (!walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, ext4_bh_unmapped)) { - unlock_page(page); - goto out_unlock; + up_read(&inode->i_alloc_sem); + return VM_FAULT_LOCKED; } } unlock_page(page); @@ -5850,6 +5904,16 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) if (ret < 0) goto out_unlock; ret = 0; + + /* + * write_begin/end might have created a dirty page and someone + * could wander in and start the IO. Make sure that hasn't + * happened. + */ + lock_page(page); + wait_on_page_writeback(page); + up_read(&inode->i_alloc_sem); + return VM_FAULT_LOCKED; out_unlock: if (ret) ret = VM_FAULT_SIGBUS; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index d8a16ee..6ed859d 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -787,6 +787,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore) struct inode *inode; char *data; char *bitmap; + struct ext4_group_info *grinfo; mb_debug(1, "init page %lu\n", page->index); @@ -819,6 +820,18 @@ static int ext4_mb_init_cache(struct page *page, char *incore) if (first_group + i >= ngroups) break; + grinfo = ext4_get_group_info(sb, first_group + i); + /* + * If page is uptodate then we came here after online resize + * which added some new uninitialized group info structs, so + * we must skip all initialized uptodate buddies on the page, + * which may be currently in use by an allocating task. + */ + if (PageUptodate(page) && !EXT4_MB_GRP_NEED_INIT(grinfo)) { + bh[i] = NULL; + continue; + } + err = -EIO; desc = ext4_get_group_desc(sb, first_group + i, NULL); if (desc == NULL) @@ -871,26 +884,28 @@ static int ext4_mb_init_cache(struct page *page, char *incore) } /* wait for I/O completion */ - for (i = 0; i < groups_per_page && bh[i]; i++) - wait_on_buffer(bh[i]); + for (i = 0; i < groups_per_page; i++) + if (bh[i]) + wait_on_buffer(bh[i]); err = -EIO; - for (i = 0; i < groups_per_page && bh[i]; i++) - if (!buffer_uptodate(bh[i])) + for (i = 0; i < groups_per_page; i++) + if (bh[i] && !buffer_uptodate(bh[i])) goto out; err = 0; first_block = page->index * blocks_per_page; - /* init the page */ - memset(page_address(page), 0xff, PAGE_CACHE_SIZE); for (i = 0; i < blocks_per_page; i++) { int group; - struct ext4_group_info *grinfo; group = (first_block + i) >> 1; if (group >= ngroups) break; + if (!bh[group - first_group]) + /* skip initialized uptodate buddy */ + continue; + /* * data carry information regarding this * particular group in the format specified @@ -919,6 +934,8 @@ static int ext4_mb_init_cache(struct page *page, char *incore) * incore got set to the group block bitmap below */ ext4_lock_group(sb, group); + /* init the buddy */ + memset(data, 0xff, blocksize); ext4_mb_generate_buddy(sb, data, incore, group); ext4_unlock_group(sb, group); incore = NULL; @@ -948,7 +965,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore) out: if (bh) { - for (i = 0; i < groups_per_page && bh[i]; i++) + for (i = 0; i < groups_per_page; i++) brelse(bh[i]); if (bh != &bhs) kfree(bh); @@ -957,22 +974,21 @@ out: } /* - * lock the group_info alloc_sem of all the groups - * belonging to the same buddy cache page. This - * make sure other parallel operation on the buddy - * cache doesn't happen whild holding the buddy cache - * lock + * Lock the buddy and bitmap pages. This make sure other parallel init_group + * on the same buddy page doesn't happen whild holding the buddy page lock. + * Return locked buddy and bitmap pages on e4b struct. If buddy and bitmap + * are on the same page e4b->bd_buddy_page is NULL and return value is 0. */ -static int ext4_mb_get_buddy_cache_lock(struct super_block *sb, - ext4_group_t group) +static int ext4_mb_get_buddy_page_lock(struct super_block *sb, + ext4_group_t group, struct ext4_buddy *e4b) { - int i; - int block, pnum; + struct inode *inode = EXT4_SB(sb)->s_buddy_cache; + int block, pnum, poff; int blocks_per_page; - int groups_per_page; - ext4_group_t ngroups = ext4_get_groups_count(sb); - ext4_group_t first_group; - struct ext4_group_info *grp; + struct page *page; + + e4b->bd_buddy_page = NULL; + e4b->bd_bitmap_page = NULL; blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; /* @@ -982,57 +998,40 @@ static int ext4_mb_get_buddy_cache_lock(struct super_block *sb, */ block = group * 2; pnum = block / blocks_per_page; - first_group = pnum * blocks_per_page / 2; - - groups_per_page = blocks_per_page >> 1; - if (groups_per_page == 0) - groups_per_page = 1; - /* read all groups the page covers into the cache */ - for (i = 0; i < groups_per_page; i++) { + poff = block % blocks_per_page; + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (!page) + return -EIO; + BUG_ON(page->mapping != inode->i_mapping); + e4b->bd_bitmap_page = page; + e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); - if ((first_group + i) >= ngroups) - break; - grp = ext4_get_group_info(sb, first_group + i); - /* take all groups write allocation - * semaphore. This make sure there is - * no block allocation going on in any - * of that groups - */ - down_write_nested(&grp->alloc_sem, i); + if (blocks_per_page >= 2) { + /* buddy and bitmap are on the same page */ + return 0; } - return i; + + block++; + pnum = block / blocks_per_page; + poff = block % blocks_per_page; + page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); + if (!page) + return -EIO; + BUG_ON(page->mapping != inode->i_mapping); + e4b->bd_buddy_page = page; + return 0; } -static void ext4_mb_put_buddy_cache_lock(struct super_block *sb, - ext4_group_t group, int locked_group) +static void ext4_mb_put_buddy_page_lock(struct ext4_buddy *e4b) { - int i; - int block, pnum; - int blocks_per_page; - ext4_group_t first_group; - struct ext4_group_info *grp; - - blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; - /* - * the buddy cache inode stores the block bitmap - * and buddy information in consecutive blocks. - * So for each group we need two blocks. - */ - block = group * 2; - pnum = block / blocks_per_page; - first_group = pnum * blocks_per_page / 2; - /* release locks on all the groups */ - for (i = 0; i < locked_group; i++) { - - grp = ext4_get_group_info(sb, first_group + i); - /* take all groups write allocation - * semaphore. This make sure there is - * no block allocation going on in any - * of that groups - */ - up_write(&grp->alloc_sem); + if (e4b->bd_bitmap_page) { + unlock_page(e4b->bd_bitmap_page); + page_cache_release(e4b->bd_bitmap_page); + } + if (e4b->bd_buddy_page) { + unlock_page(e4b->bd_buddy_page); + page_cache_release(e4b->bd_buddy_page); } - } /* @@ -1044,93 +1043,60 @@ static noinline_for_stack int ext4_mb_init_group(struct super_block *sb, ext4_group_t group) { - int ret = 0; - void *bitmap; - int blocks_per_page; - int block, pnum, poff; - int num_grp_locked = 0; struct ext4_group_info *this_grp; - struct ext4_sb_info *sbi = EXT4_SB(sb); - struct inode *inode = sbi->s_buddy_cache; - struct page *page = NULL, *bitmap_page = NULL; + struct ext4_buddy e4b; + struct page *page; + int ret = 0; mb_debug(1, "init group %u\n", group); - blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; this_grp = ext4_get_group_info(sb, group); /* * This ensures that we don't reinit the buddy cache * page which map to the group from which we are already * allocating. If we are looking at the buddy cache we would * have taken a reference using ext4_mb_load_buddy and that - * would have taken the alloc_sem lock. + * would have pinned buddy page to page cache. */ - num_grp_locked = ext4_mb_get_buddy_cache_lock(sb, group); - if (!EXT4_MB_GRP_NEED_INIT(this_grp)) { + ret = ext4_mb_get_buddy_page_lock(sb, group, &e4b); + if (ret || !EXT4_MB_GRP_NEED_INIT(this_grp)) { /* * somebody initialized the group * return without doing anything */ - ret = 0; goto err; } - /* - * the buddy cache inode stores the block bitmap - * and buddy information in consecutive blocks. - * So for each group we need two blocks. - */ - block = group * 2; - pnum = block / blocks_per_page; - poff = block % blocks_per_page; - page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); - if (page) { - BUG_ON(page->mapping != inode->i_mapping); - ret = ext4_mb_init_cache(page, NULL); - if (ret) { - unlock_page(page); - goto err; - } - unlock_page(page); - } - if (page == NULL || !PageUptodate(page)) { + + page = e4b.bd_bitmap_page; + ret = ext4_mb_init_cache(page, NULL); + if (ret) + goto err; + if (!PageUptodate(page)) { ret = -EIO; goto err; } mark_page_accessed(page); - bitmap_page = page; - bitmap = page_address(page) + (poff * sb->s_blocksize); - /* init buddy cache */ - block++; - pnum = block / blocks_per_page; - poff = block % blocks_per_page; - page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); - if (page == bitmap_page) { + if (e4b.bd_buddy_page == NULL) { /* * If both the bitmap and buddy are in * the same page we don't need to force * init the buddy */ - unlock_page(page); - } else if (page) { - BUG_ON(page->mapping != inode->i_mapping); - ret = ext4_mb_init_cache(page, bitmap); - if (ret) { - unlock_page(page); - goto err; - } - unlock_page(page); + ret = 0; + goto err; } - if (page == NULL || !PageUptodate(page)) { + /* init buddy cache */ + page = e4b.bd_buddy_page; + ret = ext4_mb_init_cache(page, e4b.bd_bitmap); + if (ret) + goto err; + if (!PageUptodate(page)) { ret = -EIO; goto err; } mark_page_accessed(page); err: - ext4_mb_put_buddy_cache_lock(sb, group, num_grp_locked); - if (bitmap_page) - page_cache_release(bitmap_page); - if (page) - page_cache_release(page); + ext4_mb_put_buddy_page_lock(&e4b); return ret; } @@ -1164,24 +1130,8 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, e4b->bd_group = group; e4b->bd_buddy_page = NULL; e4b->bd_bitmap_page = NULL; - e4b->alloc_semp = &grp->alloc_sem; - - /* Take the read lock on the group alloc - * sem. This would make sure a parallel - * ext4_mb_init_group happening on other - * groups mapped by the page is blocked - * till we are done with allocation - */ -repeat_load_buddy: - down_read(e4b->alloc_semp); if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) { - /* we need to check for group need init flag - * with alloc_semp held so that we can be sure - * that new blocks didn't get added to the group - * when we are loading the buddy cache - */ - up_read(e4b->alloc_semp); /* * we need full data about the group * to make a good selection @@ -1189,7 +1139,6 @@ repeat_load_buddy: ret = ext4_mb_init_group(sb, group); if (ret) return ret; - goto repeat_load_buddy; } /* @@ -1273,15 +1222,14 @@ repeat_load_buddy: return 0; err: + if (page) + page_cache_release(page); if (e4b->bd_bitmap_page) page_cache_release(e4b->bd_bitmap_page); if (e4b->bd_buddy_page) page_cache_release(e4b->bd_buddy_page); e4b->bd_buddy = NULL; e4b->bd_bitmap = NULL; - - /* Done with the buddy cache */ - up_read(e4b->alloc_semp); return ret; } @@ -1291,9 +1239,6 @@ static void ext4_mb_unload_buddy(struct ext4_buddy *e4b) page_cache_release(e4b->bd_bitmap_page); if (e4b->bd_buddy_page) page_cache_release(e4b->bd_buddy_page); - /* Done with the buddy cache */ - if (e4b->alloc_semp) - up_read(e4b->alloc_semp); } @@ -1606,9 +1551,6 @@ static void ext4_mb_use_best_found(struct ext4_allocation_context *ac, get_page(ac->ac_bitmap_page); ac->ac_buddy_page = e4b->bd_buddy_page; get_page(ac->ac_buddy_page); - /* on allocation we use ac to track the held semaphore */ - ac->alloc_semp = e4b->alloc_semp; - e4b->alloc_semp = NULL; /* store last allocated for subsequent stream allocation */ if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) { spin_lock(&sbi->s_md_lock); @@ -2659,7 +2601,7 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn) struct super_block *sb = journal->j_private; struct ext4_buddy e4b; struct ext4_group_info *db; - int err, ret, count = 0, count2 = 0; + int err, count = 0, count2 = 0; struct ext4_free_data *entry; struct list_head *l, *ltmp; @@ -2669,15 +2611,9 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn) mb_debug(1, "gonna free %u blocks in group %u (0x%p):", entry->count, entry->group, entry); - if (test_opt(sb, DISCARD)) { - ret = ext4_issue_discard(sb, entry->group, - entry->start_blk, entry->count); - if (unlikely(ret == -EOPNOTSUPP)) { - ext4_warning(sb, "discard not supported, " - "disabling"); - clear_opt(sb, DISCARD); - } - } + if (test_opt(sb, DISCARD)) + ext4_issue_discard(sb, entry->group, + entry->start_blk, entry->count); err = ext4_mb_load_buddy(sb, entry->group, &e4b); /* we expect to find existing buddy because it's pinned */ @@ -3642,8 +3578,8 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, free += next - bit; trace_ext4_mballoc_discard(sb, NULL, group, bit, next - bit); - trace_ext4_mb_release_inode_pa(sb, pa->pa_inode, pa, - grp_blk_start + bit, next - bit); + trace_ext4_mb_release_inode_pa(pa, grp_blk_start + bit, + next - bit); mb_free_blocks(pa->pa_inode, e4b, bit, next - bit); bit = next + 1; } @@ -3672,7 +3608,7 @@ ext4_mb_release_group_pa(struct ext4_buddy *e4b, ext4_group_t group; ext4_grpblk_t bit; - trace_ext4_mb_release_group_pa(sb, pa); + trace_ext4_mb_release_group_pa(pa); BUG_ON(pa->pa_deleted == 0); ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); BUG_ON(group != e4b->bd_group && pa->pa_len != 0); @@ -4226,15 +4162,12 @@ static int ext4_mb_release_context(struct ext4_allocation_context *ac) spin_unlock(&pa->pa_lock); } } - if (ac->alloc_semp) - up_read(ac->alloc_semp); if (pa) { /* * We want to add the pa to the right bucket. * Remove it from the list and while adding * make sure the list to which we are adding - * doesn't grow big. We need to release - * alloc_semp before calling ext4_mb_add_n_trim() + * doesn't grow big. */ if ((pa->pa_type == MB_GROUP_PA) && likely(pa->pa_free)) { spin_lock(pa->pa_obj_lock); @@ -4303,7 +4236,9 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, * there is enough free blocks to do block allocation * and verify allocation doesn't exceed the quota limits. */ - while (ar->len && ext4_claim_free_blocks(sbi, ar->len)) { + while (ar->len && + ext4_claim_free_blocks(sbi, ar->len, ar->flags)) { + /* let others to free the space */ yield(); ar->len = ar->len >> 1; @@ -4313,9 +4248,15 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, return 0; } reserv_blks = ar->len; - while (ar->len && dquot_alloc_block(ar->inode, ar->len)) { - ar->flags |= EXT4_MB_HINT_NOPREALLOC; - ar->len--; + if (ar->flags & EXT4_MB_USE_ROOT_BLOCKS) { + dquot_alloc_block_nofail(ar->inode, ar->len); + } else { + while (ar->len && + dquot_alloc_block(ar->inode, ar->len)) { + + ar->flags |= EXT4_MB_HINT_NOPREALLOC; + ar->len--; + } } inquota = ar->len; if (ar->len == 0) { @@ -4507,7 +4448,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, * @inode: inode * @block: start physical block to free * @count: number of blocks to count - * @metadata: Are these metadata blocks + * @flags: flags used by ext4_free_blocks */ void ext4_free_blocks(handle_t *handle, struct inode *inode, struct buffer_head *bh, ext4_fsblk_t block, @@ -4704,6 +4645,127 @@ error_return: } /** + * ext4_add_groupblocks() -- Add given blocks to an existing group + * @handle: handle to this transaction + * @sb: super block + * @block: start physcial block to add to the block group + * @count: number of blocks to free + * + * This marks the blocks as free in the bitmap and buddy. + */ +void ext4_add_groupblocks(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count) +{ + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *gd_bh; + ext4_group_t block_group; + ext4_grpblk_t bit; + unsigned int i; + struct ext4_group_desc *desc; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_buddy e4b; + int err = 0, ret, blk_free_count; + ext4_grpblk_t blocks_freed; + struct ext4_group_info *grp; + + ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1); + + ext4_get_group_no_and_offset(sb, block, &block_group, &bit); + grp = ext4_get_group_info(sb, block_group); + /* + * Check to see if we are freeing blocks across a group + * boundary. + */ + if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) + goto error_return; + + bitmap_bh = ext4_read_block_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + desc = ext4_get_group_desc(sb, block_group, &gd_bh); + if (!desc) + goto error_return; + + if (in_range(ext4_block_bitmap(sb, desc), block, count) || + in_range(ext4_inode_bitmap(sb, desc), block, count) || + in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(sb, desc), + sbi->s_itb_per_group)) { + ext4_error(sb, "Adding blocks in system zones - " + "Block = %llu, count = %lu", + block, count); + goto error_return; + } + + BUFFER_TRACE(bitmap_bh, "getting write access"); + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto error_return; + + /* + * We are about to modify some metadata. Call the journal APIs + * to unshare ->b_data if a currently-committing transaction is + * using it + */ + BUFFER_TRACE(gd_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, gd_bh); + if (err) + goto error_return; + + for (i = 0, blocks_freed = 0; i < count; i++) { + BUFFER_TRACE(bitmap_bh, "clear bit"); + if (!mb_test_bit(bit + i, bitmap_bh->b_data)) { + ext4_error(sb, "bit already cleared for block %llu", + (ext4_fsblk_t)(block + i)); + BUFFER_TRACE(bitmap_bh, "bit already cleared"); + } else { + blocks_freed++; + } + } + + err = ext4_mb_load_buddy(sb, block_group, &e4b); + if (err) + goto error_return; + + /* + * need to update group_info->bb_free and bitmap + * with group lock held. generate_buddy look at + * them with group lock_held + */ + ext4_lock_group(sb, block_group); + mb_clear_bits(bitmap_bh->b_data, bit, count); + mb_free_blocks(NULL, &e4b, bit, count); + blk_free_count = blocks_freed + ext4_free_blks_count(sb, desc); + ext4_free_blks_set(sb, desc, blk_free_count); + desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc); + ext4_unlock_group(sb, block_group); + percpu_counter_add(&sbi->s_freeblocks_counter, blocks_freed); + + if (sbi->s_log_groups_per_flex) { + ext4_group_t flex_group = ext4_flex_group(sbi, block_group); + atomic_add(blocks_freed, + &sbi->s_flex_groups[flex_group].free_blocks); + } + + ext4_mb_unload_buddy(&e4b); + + /* We dirtied the bitmap block */ + BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); + err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh); + + /* And the group descriptor block */ + BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); + ret = ext4_handle_dirty_metadata(handle, NULL, gd_bh); + if (!err) + err = ret; + +error_return: + brelse(bitmap_bh); + ext4_std_error(sb, err); + return; +} + +/** * ext4_trim_extent -- function to TRIM one single free extent in the group * @sb: super block for the file system * @start: starting block of the free extent in the alloc. group @@ -4715,11 +4777,10 @@ error_return: * one will allocate those blocks, mark it as used in buddy bitmap. This must * be called with under the group lock. */ -static int ext4_trim_extent(struct super_block *sb, int start, int count, - ext4_group_t group, struct ext4_buddy *e4b) +static void ext4_trim_extent(struct super_block *sb, int start, int count, + ext4_group_t group, struct ext4_buddy *e4b) { struct ext4_free_extent ex; - int ret = 0; assert_spin_locked(ext4_group_lock_ptr(sb, group)); @@ -4733,12 +4794,9 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count, */ mb_mark_used(e4b, &ex); ext4_unlock_group(sb, group); - - ret = ext4_issue_discard(sb, group, start, count); - + ext4_issue_discard(sb, group, start, count); ext4_lock_group(sb, group); mb_free_blocks(NULL, e4b, start, ex.fe_len); - return ret; } /** @@ -4760,21 +4818,26 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count, * the group buddy bitmap. This is done until whole group is scanned. */ static ext4_grpblk_t -ext4_trim_all_free(struct super_block *sb, struct ext4_buddy *e4b, - ext4_grpblk_t start, ext4_grpblk_t max, ext4_grpblk_t minblocks) +ext4_trim_all_free(struct super_block *sb, ext4_group_t group, + ext4_grpblk_t start, ext4_grpblk_t max, + ext4_grpblk_t minblocks) { void *bitmap; ext4_grpblk_t next, count = 0; - ext4_group_t group; - int ret = 0; + struct ext4_buddy e4b; + int ret; - BUG_ON(e4b == NULL); + ret = ext4_mb_load_buddy(sb, group, &e4b); + if (ret) { + ext4_error(sb, "Error in loading buddy " + "information for %u", group); + return ret; + } + bitmap = e4b.bd_bitmap; - bitmap = e4b->bd_bitmap; - group = e4b->bd_group; - start = (e4b->bd_info->bb_first_free > start) ? - e4b->bd_info->bb_first_free : start; ext4_lock_group(sb, group); + start = (e4b.bd_info->bb_first_free > start) ? + e4b.bd_info->bb_first_free : start; while (start < max) { start = mb_find_next_zero_bit(bitmap, max, start); @@ -4783,10 +4846,8 @@ ext4_trim_all_free(struct super_block *sb, struct ext4_buddy *e4b, next = mb_find_next_bit(bitmap, max, start); if ((next - start) >= minblocks) { - ret = ext4_trim_extent(sb, start, - next - start, group, e4b); - if (ret < 0) - break; + ext4_trim_extent(sb, start, + next - start, group, &e4b); count += next - start; } start = next + 1; @@ -4802,17 +4863,15 @@ ext4_trim_all_free(struct super_block *sb, struct ext4_buddy *e4b, ext4_lock_group(sb, group); } - if ((e4b->bd_info->bb_free - count) < minblocks) + if ((e4b.bd_info->bb_free - count) < minblocks) break; } ext4_unlock_group(sb, group); + ext4_mb_unload_buddy(&e4b); ext4_debug("trimmed %d blocks in the group %d\n", count, group); - if (ret < 0) - count = ret; - return count; } @@ -4830,11 +4889,11 @@ ext4_trim_all_free(struct super_block *sb, struct ext4_buddy *e4b, */ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) { - struct ext4_buddy e4b; + struct ext4_group_info *grp; ext4_group_t first_group, last_group; ext4_group_t group, ngroups = ext4_get_groups_count(sb); ext4_grpblk_t cnt = 0, first_block, last_block; - uint64_t start, len, minlen, trimmed; + uint64_t start, len, minlen, trimmed = 0; ext4_fsblk_t first_data_blk = le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); int ret = 0; @@ -4842,7 +4901,6 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) start = range->start >> sb->s_blocksize_bits; len = range->len >> sb->s_blocksize_bits; minlen = range->minlen >> sb->s_blocksize_bits; - trimmed = 0; if (unlikely(minlen > EXT4_BLOCKS_PER_GROUP(sb))) return -EINVAL; @@ -4863,11 +4921,12 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) return -EINVAL; for (group = first_group; group <= last_group; group++) { - ret = ext4_mb_load_buddy(sb, group, &e4b); - if (ret) { - ext4_error(sb, "Error in loading buddy " - "information for %u", group); - break; + grp = ext4_get_group_info(sb, group); + /* We only do this if the grp has never been initialized */ + if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) { + ret = ext4_mb_init_group(sb, group); + if (ret) + break; } /* @@ -4880,16 +4939,14 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range) last_block = first_block + len; len -= last_block - first_block; - if (e4b.bd_info->bb_free >= minlen) { - cnt = ext4_trim_all_free(sb, &e4b, first_block, + if (grp->bb_free >= minlen) { + cnt = ext4_trim_all_free(sb, group, first_block, last_block, minlen); if (cnt < 0) { ret = cnt; - ext4_mb_unload_buddy(&e4b); break; } } - ext4_mb_unload_buddy(&e4b); trimmed += cnt; first_block = 0; } diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 22bd4d7..20b5e7b 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -193,11 +193,6 @@ struct ext4_allocation_context { __u8 ac_op; /* operation, for history only */ struct page *ac_bitmap_page; struct page *ac_buddy_page; - /* - * pointer to the held semaphore upon successful - * block allocation - */ - struct rw_semaphore *alloc_semp; struct ext4_prealloc_space *ac_pa; struct ext4_locality_group *ac_lg; }; @@ -215,7 +210,6 @@ struct ext4_buddy { struct super_block *bd_sb; __u16 bd_blkbits; ext4_group_t bd_group; - struct rw_semaphore *alloc_semp; }; #define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap) #define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy) diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 92816b4..b57b98f 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -376,7 +376,7 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode, * We have the extent map build with the tmp inode. * Now copy the i_data across */ - ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS); + ext4_set_inode_flag(inode, EXT4_INODE_EXTENTS); memcpy(ei->i_data, tmp_ei->i_data, sizeof(ei->i_data)); /* diff --git a/fs/ext4/mmp.c b/fs/ext4/mmp.c new file mode 100644 index 0000000..9bdef3f --- /dev/null +++ b/fs/ext4/mmp.c @@ -0,0 +1,351 @@ +#include <linux/fs.h> +#include <linux/random.h> +#include <linux/buffer_head.h> +#include <linux/utsname.h> +#include <linux/kthread.h> + +#include "ext4.h" + +/* + * Write the MMP block using WRITE_SYNC to try to get the block on-disk + * faster. + */ +static int write_mmp_block(struct buffer_head *bh) +{ + mark_buffer_dirty(bh); + lock_buffer(bh); + bh->b_end_io = end_buffer_write_sync; + get_bh(bh); + submit_bh(WRITE_SYNC, bh); + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) + return 1; + + return 0; +} + +/* + * Read the MMP block. It _must_ be read from disk and hence we clear the + * uptodate flag on the buffer. + */ +static int read_mmp_block(struct super_block *sb, struct buffer_head **bh, + ext4_fsblk_t mmp_block) +{ + struct mmp_struct *mmp; + + if (*bh) + clear_buffer_uptodate(*bh); + + /* This would be sb_bread(sb, mmp_block), except we need to be sure + * that the MD RAID device cache has been bypassed, and that the read + * is not blocked in the elevator. */ + if (!*bh) + *bh = sb_getblk(sb, mmp_block); + if (*bh) { + get_bh(*bh); + lock_buffer(*bh); + (*bh)->b_end_io = end_buffer_read_sync; + submit_bh(READ_SYNC, *bh); + wait_on_buffer(*bh); + if (!buffer_uptodate(*bh)) { + brelse(*bh); + *bh = NULL; + } + } + if (!*bh) { + ext4_warning(sb, "Error while reading MMP block %llu", + mmp_block); + return -EIO; + } + + mmp = (struct mmp_struct *)((*bh)->b_data); + if (le32_to_cpu(mmp->mmp_magic) != EXT4_MMP_MAGIC) + return -EINVAL; + + return 0; +} + +/* + * Dump as much information as possible to help the admin. + */ +void __dump_mmp_msg(struct super_block *sb, struct mmp_struct *mmp, + const char *function, unsigned int line, const char *msg) +{ + __ext4_warning(sb, function, line, msg); + __ext4_warning(sb, function, line, + "MMP failure info: last update time: %llu, last update " + "node: %s, last update device: %s\n", + (long long unsigned int) le64_to_cpu(mmp->mmp_time), + mmp->mmp_nodename, mmp->mmp_bdevname); +} + +/* + * kmmpd will update the MMP sequence every s_mmp_update_interval seconds + */ +static int kmmpd(void *data) +{ + struct super_block *sb = ((struct mmpd_data *) data)->sb; + struct buffer_head *bh = ((struct mmpd_data *) data)->bh; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + struct mmp_struct *mmp; + ext4_fsblk_t mmp_block; + u32 seq = 0; + unsigned long failed_writes = 0; + int mmp_update_interval = le16_to_cpu(es->s_mmp_update_interval); + unsigned mmp_check_interval; + unsigned long last_update_time; + unsigned long diff; + int retval; + + mmp_block = le64_to_cpu(es->s_mmp_block); + mmp = (struct mmp_struct *)(bh->b_data); + mmp->mmp_time = cpu_to_le64(get_seconds()); + /* + * Start with the higher mmp_check_interval and reduce it if + * the MMP block is being updated on time. + */ + mmp_check_interval = max(EXT4_MMP_CHECK_MULT * mmp_update_interval, + EXT4_MMP_MIN_CHECK_INTERVAL); + mmp->mmp_check_interval = cpu_to_le16(mmp_check_interval); + bdevname(bh->b_bdev, mmp->mmp_bdevname); + + memcpy(mmp->mmp_nodename, init_utsname()->sysname, + sizeof(mmp->mmp_nodename)); + + while (!kthread_should_stop()) { + if (++seq > EXT4_MMP_SEQ_MAX) + seq = 1; + + mmp->mmp_seq = cpu_to_le32(seq); + mmp->mmp_time = cpu_to_le64(get_seconds()); + last_update_time = jiffies; + + retval = write_mmp_block(bh); + /* + * Don't spew too many error messages. Print one every + * (s_mmp_update_interval * 60) seconds. + */ + if (retval && (failed_writes % 60) == 0) { + ext4_error(sb, "Error writing to MMP block"); + failed_writes++; + } + + if (!(le32_to_cpu(es->s_feature_incompat) & + EXT4_FEATURE_INCOMPAT_MMP)) { + ext4_warning(sb, "kmmpd being stopped since MMP feature" + " has been disabled."); + EXT4_SB(sb)->s_mmp_tsk = NULL; + goto failed; + } + + if (sb->s_flags & MS_RDONLY) { + ext4_warning(sb, "kmmpd being stopped since filesystem " + "has been remounted as readonly."); + EXT4_SB(sb)->s_mmp_tsk = NULL; + goto failed; + } + + diff = jiffies - last_update_time; + if (diff < mmp_update_interval * HZ) + schedule_timeout_interruptible(mmp_update_interval * + HZ - diff); + + /* + * We need to make sure that more than mmp_check_interval + * seconds have not passed since writing. If that has happened + * we need to check if the MMP block is as we left it. + */ + diff = jiffies - last_update_time; + if (diff > mmp_check_interval * HZ) { + struct buffer_head *bh_check = NULL; + struct mmp_struct *mmp_check; + + retval = read_mmp_block(sb, &bh_check, mmp_block); + if (retval) { + ext4_error(sb, "error reading MMP data: %d", + retval); + + EXT4_SB(sb)->s_mmp_tsk = NULL; + goto failed; + } + + mmp_check = (struct mmp_struct *)(bh_check->b_data); + if (mmp->mmp_seq != mmp_check->mmp_seq || + memcmp(mmp->mmp_nodename, mmp_check->mmp_nodename, + sizeof(mmp->mmp_nodename))) { + dump_mmp_msg(sb, mmp_check, + "Error while updating MMP info. " + "The filesystem seems to have been" + " multiply mounted."); + ext4_error(sb, "abort"); + goto failed; + } + put_bh(bh_check); + } + + /* + * Adjust the mmp_check_interval depending on how much time + * it took for the MMP block to be written. + */ + mmp_check_interval = max(min(EXT4_MMP_CHECK_MULT * diff / HZ, + EXT4_MMP_MAX_CHECK_INTERVAL), + EXT4_MMP_MIN_CHECK_INTERVAL); + mmp->mmp_check_interval = cpu_to_le16(mmp_check_interval); + } + + /* + * Unmount seems to be clean. + */ + mmp->mmp_seq = cpu_to_le32(EXT4_MMP_SEQ_CLEAN); + mmp->mmp_time = cpu_to_le64(get_seconds()); + + retval = write_mmp_block(bh); + +failed: + kfree(data); + brelse(bh); + return retval; +} + +/* + * Get a random new sequence number but make sure it is not greater than + * EXT4_MMP_SEQ_MAX. + */ +static unsigned int mmp_new_seq(void) +{ + u32 new_seq; + + do { + get_random_bytes(&new_seq, sizeof(u32)); + } while (new_seq > EXT4_MMP_SEQ_MAX); + + return new_seq; +} + +/* + * Protect the filesystem from being mounted more than once. + */ +int ext4_multi_mount_protect(struct super_block *sb, + ext4_fsblk_t mmp_block) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + struct buffer_head *bh = NULL; + struct mmp_struct *mmp = NULL; + struct mmpd_data *mmpd_data; + u32 seq; + unsigned int mmp_check_interval = le16_to_cpu(es->s_mmp_update_interval); + unsigned int wait_time = 0; + int retval; + + if (mmp_block < le32_to_cpu(es->s_first_data_block) || + mmp_block >= ext4_blocks_count(es)) { + ext4_warning(sb, "Invalid MMP block in superblock"); + goto failed; + } + + retval = read_mmp_block(sb, &bh, mmp_block); + if (retval) + goto failed; + + mmp = (struct mmp_struct *)(bh->b_data); + + if (mmp_check_interval < EXT4_MMP_MIN_CHECK_INTERVAL) + mmp_check_interval = EXT4_MMP_MIN_CHECK_INTERVAL; + + /* + * If check_interval in MMP block is larger, use that instead of + * update_interval from the superblock. + */ + if (mmp->mmp_check_interval > mmp_check_interval) + mmp_check_interval = mmp->mmp_check_interval; + + seq = le32_to_cpu(mmp->mmp_seq); + if (seq == EXT4_MMP_SEQ_CLEAN) + goto skip; + + if (seq == EXT4_MMP_SEQ_FSCK) { + dump_mmp_msg(sb, mmp, "fsck is running on the filesystem"); + goto failed; + } + + wait_time = min(mmp_check_interval * 2 + 1, + mmp_check_interval + 60); + + /* Print MMP interval if more than 20 secs. */ + if (wait_time > EXT4_MMP_MIN_CHECK_INTERVAL * 4) + ext4_warning(sb, "MMP interval %u higher than expected, please" + " wait.\n", wait_time * 2); + + if (schedule_timeout_interruptible(HZ * wait_time) != 0) { + ext4_warning(sb, "MMP startup interrupted, failing mount\n"); + goto failed; + } + + retval = read_mmp_block(sb, &bh, mmp_block); + if (retval) + goto failed; + mmp = (struct mmp_struct *)(bh->b_data); + if (seq != le32_to_cpu(mmp->mmp_seq)) { + dump_mmp_msg(sb, mmp, + "Device is already active on another node."); + goto failed; + } + +skip: + /* + * write a new random sequence number. + */ + mmp->mmp_seq = seq = cpu_to_le32(mmp_new_seq()); + + retval = write_mmp_block(bh); + if (retval) + goto failed; + + /* + * wait for MMP interval and check mmp_seq. + */ + if (schedule_timeout_interruptible(HZ * wait_time) != 0) { + ext4_warning(sb, "MMP startup interrupted, failing mount\n"); + goto failed; + } + + retval = read_mmp_block(sb, &bh, mmp_block); + if (retval) + goto failed; + mmp = (struct mmp_struct *)(bh->b_data); + if (seq != le32_to_cpu(mmp->mmp_seq)) { + dump_mmp_msg(sb, mmp, + "Device is already active on another node."); + goto failed; + } + + mmpd_data = kmalloc(sizeof(struct mmpd_data), GFP_KERNEL); + if (!mmpd_data) { + ext4_warning(sb, "not enough memory for mmpd_data"); + goto failed; + } + mmpd_data->sb = sb; + mmpd_data->bh = bh; + + /* + * Start a kernel thread to update the MMP block periodically. + */ + EXT4_SB(sb)->s_mmp_tsk = kthread_run(kmmpd, mmpd_data, "kmmpd-%s", + bdevname(bh->b_bdev, + mmp->mmp_bdevname)); + if (IS_ERR(EXT4_SB(sb)->s_mmp_tsk)) { + EXT4_SB(sb)->s_mmp_tsk = NULL; + kfree(mmpd_data); + ext4_warning(sb, "Unable to create kmmpd thread for %s.", + sb->s_id); + goto failed; + } + + return 0; + +failed: + brelse(bh); + return 1; +} + + diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c index b9f3e78..f57455a 100644 --- a/fs/ext4/move_extent.c +++ b/fs/ext4/move_extent.c @@ -876,8 +876,7 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode, * It needs to call wait_on_page_writeback() to wait for the * writeback of the page. */ - if (PageWriteback(page)) - wait_on_page_writeback(page); + wait_on_page_writeback(page); /* Release old bh and drop refs */ try_to_release_page(page, 0); @@ -1003,12 +1002,12 @@ mext_check_arguments(struct inode *orig_inode, return -EINVAL; } - if ((orig_start > EXT_MAX_BLOCK) || - (donor_start > EXT_MAX_BLOCK) || - (*len > EXT_MAX_BLOCK) || - (orig_start + *len > EXT_MAX_BLOCK)) { + if ((orig_start >= EXT_MAX_BLOCKS) || + (donor_start >= EXT_MAX_BLOCKS) || + (*len > EXT_MAX_BLOCKS) || + (orig_start + *len >= EXT_MAX_BLOCKS)) { ext4_debug("ext4 move extent: Can't handle over [%u] blocks " - "[ino:orig %lu, donor %lu]\n", EXT_MAX_BLOCK, + "[ino:orig %lu, donor %lu]\n", EXT_MAX_BLOCKS, orig_inode->i_ino, donor_inode->i_ino); return -EINVAL; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 67fd0b0..b754b77 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1413,10 +1413,22 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, frame->at = entries; frame->bh = bh; bh = bh2; + + ext4_handle_dirty_metadata(handle, dir, frame->bh); + ext4_handle_dirty_metadata(handle, dir, bh); + de = do_split(handle,dir, &bh, frame, &hinfo, &retval); - dx_release (frames); - if (!(de)) + if (!de) { + /* + * Even if the block split failed, we have to properly write + * out all the changes we did so far. Otherwise we can end up + * with corrupted filesystem. + */ + ext4_mark_inode_dirty(handle, dir); + dx_release(frames); return retval; + } + dx_release(frames); retval = add_dirent_to_buf(handle, dentry, inode, de, bh); brelse(bh); @@ -2240,6 +2252,7 @@ static int ext4_symlink(struct inode *dir, handle_t *handle; struct inode *inode; int l, err, retries = 0; + int credits; l = strlen(symname)+1; if (l > dir->i_sb->s_blocksize) @@ -2247,10 +2260,26 @@ static int ext4_symlink(struct inode *dir, dquot_initialize(dir); + if (l > EXT4_N_BLOCKS * 4) { + /* + * For non-fast symlinks, we just allocate inode and put it on + * orphan list in the first transaction => we need bitmap, + * group descriptor, sb, inode block, quota blocks. + */ + credits = 4 + EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb); + } else { + /* + * Fast symlink. We have to add entry to directory + * (EXT4_DATA_TRANS_BLOCKS + EXT4_INDEX_EXTRA_TRANS_BLOCKS), + * allocate new inode (bitmap, group descriptor, inode block, + * quota blocks, sb is already counted in previous macros). + */ + credits = EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb); + } retry: - handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT4_INDEX_EXTRA_TRANS_BLOCKS + 5 + - EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); + handle = ext4_journal_start(dir, credits); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2263,21 +2292,44 @@ retry: if (IS_ERR(inode)) goto out_stop; - if (l > sizeof(EXT4_I(inode)->i_data)) { + if (l > EXT4_N_BLOCKS * 4) { inode->i_op = &ext4_symlink_inode_operations; ext4_set_aops(inode); /* - * page_symlink() calls into ext4_prepare/commit_write. - * We have a transaction open. All is sweetness. It also sets - * i_size in generic_commit_write(). + * We cannot call page_symlink() with transaction started + * because it calls into ext4_write_begin() which can wait + * for transaction commit if we are running out of space + * and thus we deadlock. So we have to stop transaction now + * and restart it when symlink contents is written. + * + * To keep fs consistent in case of crash, we have to put inode + * to orphan list in the mean time. */ + drop_nlink(inode); + err = ext4_orphan_add(handle, inode); + ext4_journal_stop(handle); + if (err) + goto err_drop_inode; err = __page_symlink(inode, symname, l, 1); + if (err) + goto err_drop_inode; + /* + * Now inode is being linked into dir (EXT4_DATA_TRANS_BLOCKS + * + EXT4_INDEX_EXTRA_TRANS_BLOCKS), inode is also modified + */ + handle = ext4_journal_start(dir, + EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 1); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto err_drop_inode; + } + inc_nlink(inode); + err = ext4_orphan_del(handle, inode); if (err) { + ext4_journal_stop(handle); clear_nlink(inode); - unlock_new_inode(inode); - ext4_mark_inode_dirty(handle, inode); - iput(inode); - goto out_stop; + goto err_drop_inode; } } else { /* clear the extent format for fast symlink */ @@ -2293,6 +2345,10 @@ out_stop: if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) goto retry; return err; +err_drop_inode: + unlock_new_inode(inode); + iput(inode); + return err; } static int ext4_link(struct dentry *old_dentry, diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index b6dbd05..7bb8f76 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -203,46 +203,29 @@ static void ext4_end_bio(struct bio *bio, int error) for (i = 0; i < io_end->num_io_pages; i++) { struct page *page = io_end->pages[i]->p_page; struct buffer_head *bh, *head; - int partial_write = 0; + loff_t offset; + loff_t io_end_offset; - head = page_buffers(page); - if (error) + if (error) { SetPageError(page); - BUG_ON(!head); - if (head->b_size != PAGE_CACHE_SIZE) { - loff_t offset; - loff_t io_end_offset = io_end->offset + io_end->size; + set_bit(AS_EIO, &page->mapping->flags); + head = page_buffers(page); + BUG_ON(!head); + + io_end_offset = io_end->offset + io_end->size; offset = (sector_t) page->index << PAGE_CACHE_SHIFT; bh = head; do { if ((offset >= io_end->offset) && - (offset+bh->b_size <= io_end_offset)) { - if (error) - buffer_io_error(bh); - - } - if (buffer_delay(bh)) - partial_write = 1; - else if (!buffer_mapped(bh)) - clear_buffer_dirty(bh); - else if (buffer_dirty(bh)) - partial_write = 1; + (offset+bh->b_size <= io_end_offset)) + buffer_io_error(bh); + offset += bh->b_size; bh = bh->b_this_page; } while (bh != head); } - /* - * If this is a partial write which happened to make - * all buffers uptodate then we can optimize away a - * bogus readpage() for the next read(). Here we - * 'discover' whether the page went uptodate as a - * result of this (potentially partial) write. - */ - if (!partial_write) - SetPageUptodate(page); - put_io_page(io_end->pages[i]); } io_end->num_io_pages = 0; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 8553dfb..9ea71aa 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -38,6 +38,7 @@ #include <linux/ctype.h> #include <linux/log2.h> #include <linux/crc16.h> +#include <linux/cleancache.h> #include <asm/uaccess.h> #include <linux/kthread.h> @@ -75,11 +76,27 @@ static void ext4_write_super(struct super_block *sb); static int ext4_freeze(struct super_block *sb); static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data); +static inline int ext2_feature_set_ok(struct super_block *sb); +static inline int ext3_feature_set_ok(struct super_block *sb); static int ext4_feature_set_ok(struct super_block *sb, int readonly); static void ext4_destroy_lazyinit_thread(void); static void ext4_unregister_li_request(struct super_block *sb); static void ext4_clear_request_list(void); +#if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23) +static struct file_system_type ext2_fs_type = { + .owner = THIS_MODULE, + .name = "ext2", + .mount = ext4_mount, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; +#define IS_EXT2_SB(sb) ((sb)->s_bdev->bd_holder == &ext2_fs_type) +#else +#define IS_EXT2_SB(sb) (0) +#endif + + #if !defined(CONFIG_EXT3_FS) && !defined(CONFIG_EXT3_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23) static struct file_system_type ext3_fs_type = { .owner = THIS_MODULE, @@ -806,6 +823,8 @@ static void ext4_put_super(struct super_block *sb) invalidate_bdev(sbi->journal_bdev); ext4_blkdev_remove(sbi); } + if (sbi->s_mmp_tsk) + kthread_stop(sbi->s_mmp_tsk); sb->s_fs_info = NULL; /* * Now that we are completely done shutting down the @@ -1096,7 +1115,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) if (!test_opt(sb, INIT_INODE_TABLE)) seq_puts(seq, ",noinit_inode_table"); - else if (sbi->s_li_wait_mult) + else if (sbi->s_li_wait_mult != EXT4_DEF_LI_WAIT_MULT) seq_printf(seq, ",init_inode_table=%u", (unsigned) sbi->s_li_wait_mult); @@ -1187,9 +1206,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type, const char *data, size_t len, loff_t off); static const struct dquot_operations ext4_quota_operations = { -#ifdef CONFIG_QUOTA .get_reserved_space = ext4_get_reserved_space, -#endif .write_dquot = ext4_write_dquot, .acquire_dquot = ext4_acquire_dquot, .release_dquot = ext4_release_dquot, @@ -1900,7 +1917,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, ext4_msg(sb, KERN_WARNING, "warning: mounting fs with errors, " "running e2fsck is recommended"); - else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 && + else if ((__s16) le16_to_cpu(es->s_max_mnt_count) > 0 && le16_to_cpu(es->s_mnt_count) >= (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count)) ext4_msg(sb, KERN_WARNING, @@ -1932,6 +1949,7 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt, sbi->s_mount_opt2); + cleancache_init_fs(sb); return res; } @@ -2225,6 +2243,12 @@ static void ext4_orphan_cleanup(struct super_block *sb, * in the vfs. ext4 inode has 48 bits of i_block in fsblock units, * so that won't be a limiting factor. * + * However there is other limiting factor. We do store extents in the form + * of starting block and length, hence the resulting length of the extent + * covering maximum file size must fit into on-disk format containers as + * well. Given that length is always by 1 unit bigger than max unit (because + * we count 0 as well) we have to lower the s_maxbytes by one fs block. + * * Note, this does *not* consider any metadata overhead for vfs i_blocks. */ static loff_t ext4_max_size(int blkbits, int has_huge_files) @@ -2246,10 +2270,13 @@ static loff_t ext4_max_size(int blkbits, int has_huge_files) upper_limit <<= blkbits; } - /* 32-bit extent-start container, ee_block */ - res = 1LL << 32; + /* + * 32-bit extent-start container, ee_block. We lower the maxbytes + * by one fs block, so ee_len can cover the extent of maximum file + * size + */ + res = (1LL << 32) - 1; res <<= blkbits; - res -= 1; /* Sanity check against vm- & vfs- imposed limits */ if (res > upper_limit) @@ -2425,6 +2452,18 @@ static ssize_t lifetime_write_kbytes_show(struct ext4_attr *a, EXT4_SB(sb)->s_sectors_written_start) >> 1))); } +static ssize_t extent_cache_hits_show(struct ext4_attr *a, + struct ext4_sb_info *sbi, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%lu\n", sbi->extent_cache_hits); +} + +static ssize_t extent_cache_misses_show(struct ext4_attr *a, + struct ext4_sb_info *sbi, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%lu\n", sbi->extent_cache_misses); +} + static ssize_t inode_readahead_blks_store(struct ext4_attr *a, struct ext4_sb_info *sbi, const char *buf, size_t count) @@ -2482,6 +2521,8 @@ static struct ext4_attr ext4_attr_##name = __ATTR(name, mode, show, store) EXT4_RO_ATTR(delayed_allocation_blocks); EXT4_RO_ATTR(session_write_kbytes); EXT4_RO_ATTR(lifetime_write_kbytes); +EXT4_RO_ATTR(extent_cache_hits); +EXT4_RO_ATTR(extent_cache_misses); EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, sbi_ui_show, inode_readahead_blks_store, s_inode_readahead_blks); EXT4_RW_ATTR_SBI_UI(inode_goal, s_inode_goal); @@ -2497,6 +2538,8 @@ static struct attribute *ext4_attrs[] = { ATTR_LIST(delayed_allocation_blocks), ATTR_LIST(session_write_kbytes), ATTR_LIST(lifetime_write_kbytes), + ATTR_LIST(extent_cache_hits), + ATTR_LIST(extent_cache_misses), ATTR_LIST(inode_readahead_blks), ATTR_LIST(inode_goal), ATTR_LIST(mb_stats), @@ -2659,12 +2702,6 @@ static void print_daily_error_info(unsigned long arg) mod_timer(&sbi->s_err_report, jiffies + 24*60*60*HZ); /* Once a day */ } -static void ext4_lazyinode_timeout(unsigned long data) -{ - struct task_struct *p = (struct task_struct *)data; - wake_up_process(p); -} - /* Find next suitable group and run ext4_init_inode_table */ static int ext4_run_li_request(struct ext4_li_request *elr) { @@ -2696,11 +2733,8 @@ static int ext4_run_li_request(struct ext4_li_request *elr) ret = ext4_init_inode_table(sb, group, elr->lr_timeout ? 0 : 1); if (elr->lr_timeout == 0) { - timeout = jiffies - timeout; - if (elr->lr_sbi->s_li_wait_mult) - timeout *= elr->lr_sbi->s_li_wait_mult; - else - timeout *= 20; + timeout = (jiffies - timeout) * + elr->lr_sbi->s_li_wait_mult; elr->lr_timeout = timeout; } elr->lr_next_sched = jiffies + elr->lr_timeout; @@ -2712,7 +2746,7 @@ static int ext4_run_li_request(struct ext4_li_request *elr) /* * Remove lr_request from the list_request and free the - * request tructure. Should be called with li_list_mtx held + * request structure. Should be called with li_list_mtx held */ static void ext4_remove_li_request(struct ext4_li_request *elr) { @@ -2730,14 +2764,16 @@ static void ext4_remove_li_request(struct ext4_li_request *elr) static void ext4_unregister_li_request(struct super_block *sb) { - struct ext4_li_request *elr = EXT4_SB(sb)->s_li_request; - - if (!ext4_li_info) + mutex_lock(&ext4_li_mtx); + if (!ext4_li_info) { + mutex_unlock(&ext4_li_mtx); return; + } mutex_lock(&ext4_li_info->li_list_mtx); - ext4_remove_li_request(elr); + ext4_remove_li_request(EXT4_SB(sb)->s_li_request); mutex_unlock(&ext4_li_info->li_list_mtx); + mutex_unlock(&ext4_li_mtx); } static struct task_struct *ext4_lazyinit_task; @@ -2756,17 +2792,10 @@ static int ext4_lazyinit_thread(void *arg) struct ext4_lazy_init *eli = (struct ext4_lazy_init *)arg; struct list_head *pos, *n; struct ext4_li_request *elr; - unsigned long next_wakeup; - DEFINE_WAIT(wait); + unsigned long next_wakeup, cur; BUG_ON(NULL == eli); - eli->li_timer.data = (unsigned long)current; - eli->li_timer.function = ext4_lazyinode_timeout; - - eli->li_task = current; - wake_up(&eli->li_wait_task); - cont_thread: while (true) { next_wakeup = MAX_JIFFY_OFFSET; @@ -2797,19 +2826,15 @@ cont_thread: if (freezing(current)) refrigerator(); - if ((time_after_eq(jiffies, next_wakeup)) || + cur = jiffies; + if ((time_after_eq(cur, next_wakeup)) || (MAX_JIFFY_OFFSET == next_wakeup)) { cond_resched(); continue; } - eli->li_timer.expires = next_wakeup; - add_timer(&eli->li_timer); - prepare_to_wait(&eli->li_wait_daemon, &wait, - TASK_INTERRUPTIBLE); - if (time_before(jiffies, next_wakeup)) - schedule(); - finish_wait(&eli->li_wait_daemon, &wait); + schedule_timeout_interruptible(next_wakeup - cur); + if (kthread_should_stop()) { ext4_clear_request_list(); goto exit_thread; @@ -2833,12 +2858,7 @@ exit_thread: goto cont_thread; } mutex_unlock(&eli->li_list_mtx); - del_timer_sync(&ext4_li_info->li_timer); - eli->li_task = NULL; - wake_up(&eli->li_wait_task); - kfree(ext4_li_info); - ext4_lazyinit_task = NULL; ext4_li_info = NULL; mutex_unlock(&ext4_li_mtx); @@ -2866,7 +2886,6 @@ static int ext4_run_lazyinit_thread(void) if (IS_ERR(ext4_lazyinit_task)) { int err = PTR_ERR(ext4_lazyinit_task); ext4_clear_request_list(); - del_timer_sync(&ext4_li_info->li_timer); kfree(ext4_li_info); ext4_li_info = NULL; printk(KERN_CRIT "EXT4: error %d creating inode table " @@ -2875,8 +2894,6 @@ static int ext4_run_lazyinit_thread(void) return err; } ext4_li_info->li_state |= EXT4_LAZYINIT_RUNNING; - - wait_event(ext4_li_info->li_wait_task, ext4_li_info->li_task != NULL); return 0; } @@ -2911,13 +2928,9 @@ static int ext4_li_info_new(void) if (!eli) return -ENOMEM; - eli->li_task = NULL; INIT_LIST_HEAD(&eli->li_request_list); mutex_init(&eli->li_list_mtx); - init_waitqueue_head(&eli->li_wait_daemon); - init_waitqueue_head(&eli->li_wait_task); - init_timer(&eli->li_timer); eli->li_state |= EXT4_LAZYINIT_QUIT; ext4_li_info = eli; @@ -2960,20 +2973,19 @@ static int ext4_register_li_request(struct super_block *sb, ext4_group_t ngroups = EXT4_SB(sb)->s_groups_count; int ret = 0; - if (sbi->s_li_request != NULL) + if (sbi->s_li_request != NULL) { + /* + * Reset timeout so it can be computed again, because + * s_li_wait_mult might have changed. + */ + sbi->s_li_request->lr_timeout = 0; return 0; + } if (first_not_zeroed == ngroups || (sb->s_flags & MS_RDONLY) || - !test_opt(sb, INIT_INODE_TABLE)) { - sbi->s_li_request = NULL; + !test_opt(sb, INIT_INODE_TABLE)) return 0; - } - - if (first_not_zeroed == ngroups) { - sbi->s_li_request = NULL; - return 0; - } elr = ext4_li_request_new(sb, first_not_zeroed); if (!elr) @@ -3166,6 +3178,12 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) ((def_mount_opts & EXT4_DEFM_NODELALLOC) == 0)) set_opt(sb, DELALLOC); + /* + * set default s_li_wait_mult for lazyinit, for the case there is + * no mount option specified. + */ + sbi->s_li_wait_mult = EXT4_DEF_LI_WAIT_MULT; + if (!parse_options((char *) sbi->s_es->s_mount_opts, sb, &journal_devnum, &journal_ioprio, NULL, 0)) { ext4_msg(sb, KERN_WARNING, @@ -3187,6 +3205,28 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) "feature flags set on rev 0 fs, " "running e2fsck is recommended"); + if (IS_EXT2_SB(sb)) { + if (ext2_feature_set_ok(sb)) + ext4_msg(sb, KERN_INFO, "mounting ext2 file system " + "using the ext4 subsystem"); + else { + ext4_msg(sb, KERN_ERR, "couldn't mount as ext2 due " + "to feature incompatibilities"); + goto failed_mount; + } + } + + if (IS_EXT3_SB(sb)) { + if (ext3_feature_set_ok(sb)) + ext4_msg(sb, KERN_INFO, "mounting ext3 file system " + "using the ext4 subsystem"); + else { + ext4_msg(sb, KERN_ERR, "couldn't mount as ext3 due " + "to feature incompatibilities"); + goto failed_mount; + } + } + /* * Check feature flags regardless of the revision level, since we * previously didn't change the revision level when setting the flags, @@ -3459,6 +3499,11 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_MMP) && + !(sb->s_flags & MS_RDONLY)) + if (ext4_multi_mount_protect(sb, le64_to_cpu(es->s_mmp_block))) + goto failed_mount3; + /* * The first inode we look at is the journal inode. Don't try * root first: it may be modified in the journal! @@ -3474,7 +3519,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount_wq; } else { clear_opt(sb, DATA_FLAGS); - set_opt(sb, WRITEBACK_DATA); sbi->s_journal = NULL; needs_recovery = 0; goto no_journal; @@ -3707,6 +3751,8 @@ failed_mount3: percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyblocks_counter); + if (sbi->s_mmp_tsk) + kthread_stop(sbi->s_mmp_tsk); failed_mount2: for (i = 0; i < db_count; i++) brelse(sbi->s_group_desc[i]); @@ -4242,7 +4288,7 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) int enable_quota = 0; ext4_group_t g; unsigned int journal_ioprio = DEFAULT_JOURNAL_IOPRIO; - int err; + int err = 0; #ifdef CONFIG_QUOTA int i; #endif @@ -4368,6 +4414,13 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) goto restore_opts; if (!ext4_setup_super(sb, es, 0)) sb->s_flags &= ~MS_RDONLY; + if (EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_MMP)) + if (ext4_multi_mount_protect(sb, + le64_to_cpu(es->s_mmp_block))) { + err = -EROFS; + goto restore_opts; + } enable_quota = 1; } } @@ -4432,6 +4485,7 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf) struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; u64 fsid; + s64 bfree; if (test_opt(sb, MINIX_DF)) { sbi->s_overhead_last = 0; @@ -4475,8 +4529,10 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_type = EXT4_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; buf->f_blocks = ext4_blocks_count(es) - sbi->s_overhead_last; - buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) - + bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) - percpu_counter_sum_positive(&sbi->s_dirtyblocks_counter); + /* prevent underflow in case that few free space is available */ + buf->f_bfree = max_t(s64, bfree, 0); buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es); if (buf->f_bfree < ext4_r_blocks_count(es)) buf->f_bavail = 0; @@ -4652,6 +4708,9 @@ static int ext4_quota_off(struct super_block *sb, int type) if (test_opt(sb, DELALLOC)) sync_filesystem(sb); + if (!inode) + goto out; + /* Update modification times of quota files when userspace can * start looking at them */ handle = ext4_journal_start(inode, 1); @@ -4772,14 +4831,6 @@ static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags, } #if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23) -static struct file_system_type ext2_fs_type = { - .owner = THIS_MODULE, - .name = "ext2", - .mount = ext4_mount, - .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, -}; - static inline void register_as_ext2(void) { int err = register_filesystem(&ext2_fs_type); @@ -4792,10 +4843,22 @@ static inline void unregister_as_ext2(void) { unregister_filesystem(&ext2_fs_type); } + +static inline int ext2_feature_set_ok(struct super_block *sb) +{ + if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT2_FEATURE_INCOMPAT_SUPP)) + return 0; + if (sb->s_flags & MS_RDONLY) + return 1; + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT2_FEATURE_RO_COMPAT_SUPP)) + return 0; + return 1; +} MODULE_ALIAS("ext2"); #else static inline void register_as_ext2(void) { } static inline void unregister_as_ext2(void) { } +static inline int ext2_feature_set_ok(struct super_block *sb) { return 0; } #endif #if !defined(CONFIG_EXT3_FS) && !defined(CONFIG_EXT3_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23) @@ -4811,10 +4874,24 @@ static inline void unregister_as_ext3(void) { unregister_filesystem(&ext3_fs_type); } + +static inline int ext3_feature_set_ok(struct super_block *sb) +{ + if (EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP)) + return 0; + if (!EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) + return 0; + if (sb->s_flags & MS_RDONLY) + return 1; + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP)) + return 0; + return 1; +} MODULE_ALIAS("ext3"); #else static inline void register_as_ext3(void) { } static inline void unregister_as_ext3(void) { } +static inline int ext3_feature_set_ok(struct super_block *sb) { return 0; } #endif static struct file_system_type ext4_fs_type = { @@ -4898,8 +4975,8 @@ static int __init ext4_init_fs(void) err = init_inodecache(); if (err) goto out1; - register_as_ext2(); register_as_ext3(); + register_as_ext2(); err = register_filesystem(&ext4_fs_type); if (err) goto out; diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index b545ca1..c757adc 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -820,8 +820,8 @@ inserted: if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) goal = goal & EXT4_MAX_BLOCK_FILE_PHYS; - block = ext4_new_meta_blocks(handle, inode, - goal, NULL, &error); + block = ext4_new_meta_blocks(handle, inode, goal, 0, + NULL, &error); if (error) goto cleanup; diff --git a/fs/fat/cache.c b/fs/fat/cache.c index ae8200f..1cc7038e 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -151,6 +151,13 @@ static void fat_cache_add(struct inode *inode, struct fat_cache_id *new) spin_unlock(&MSDOS_I(inode)->cache_lru_lock); tmp = fat_cache_alloc(inode); + if (!tmp) { + spin_lock(&MSDOS_I(inode)->cache_lru_lock); + MSDOS_I(inode)->nr_caches--; + spin_unlock(&MSDOS_I(inode)->cache_lru_lock); + return; + } + spin_lock(&MSDOS_I(inode)->cache_lru_lock); cache = fat_cache_merge(inode, new); if (cache != NULL) { diff --git a/fs/fat/dir.c b/fs/fat/dir.c index ee42b9e..4ad6473 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -98,7 +98,7 @@ next: *bh = sb_bread(sb, phys); if (*bh == NULL) { - printk(KERN_ERR "FAT: Directory bread(block %llu) failed\n", + fat_msg(sb, KERN_ERR, "Directory bread(block %llu) failed", (llu)phys); /* skip this block */ *pos = (iblock + 1) << sb->s_blocksize_bits; @@ -136,9 +136,10 @@ static inline int fat_get_entry(struct inode *dir, loff_t *pos, * but ignore that right now. * Ahem... Stack smashing in ring 0 isn't fun. Fixed. */ -static int uni16_to_x8(unsigned char *ascii, const wchar_t *uni, int len, - int uni_xlate, struct nls_table *nls) +static int uni16_to_x8(struct super_block *sb, unsigned char *ascii, + const wchar_t *uni, int len, struct nls_table *nls) { + int uni_xlate = MSDOS_SB(sb)->options.unicode_xlate; const wchar_t *ip; wchar_t ec; unsigned char *op; @@ -166,23 +167,23 @@ static int uni16_to_x8(unsigned char *ascii, const wchar_t *uni, int len, } if (unlikely(*ip)) { - printk(KERN_WARNING "FAT: filename was truncated while " - "converting."); + fat_msg(sb, KERN_WARNING, "filename was truncated while " + "converting."); } *op = 0; return (op - ascii); } -static inline int fat_uni_to_x8(struct msdos_sb_info *sbi, const wchar_t *uni, +static inline int fat_uni_to_x8(struct super_block *sb, const wchar_t *uni, unsigned char *buf, int size) { + struct msdos_sb_info *sbi = MSDOS_SB(sb); if (sbi->options.utf8) return utf16s_to_utf8s(uni, FAT_MAX_UNI_CHARS, UTF16_HOST_ENDIAN, buf, size); else - return uni16_to_x8(buf, uni, size, sbi->options.unicode_xlate, - sbi->nls_io); + return uni16_to_x8(sb, buf, uni, size, sbi->nls_io); } static inline int @@ -419,7 +420,7 @@ parse_record: /* Compare shortname */ bufuname[last_u] = 0x0000; - len = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname)); + len = fat_uni_to_x8(sb, bufuname, bufname, sizeof(bufname)); if (fat_name_match(sbi, name, name_len, bufname, len)) goto found; @@ -428,7 +429,7 @@ parse_record: int size = PATH_MAX - FAT_MAX_UNI_SIZE; /* Compare longname */ - len = fat_uni_to_x8(sbi, unicode, longname, size); + len = fat_uni_to_x8(sb, unicode, longname, size); if (fat_name_match(sbi, name, name_len, longname, len)) goto found; } @@ -545,7 +546,7 @@ parse_record: if (nr_slots) { void *longname = unicode + FAT_MAX_UNI_CHARS; int size = PATH_MAX - FAT_MAX_UNI_SIZE; - int len = fat_uni_to_x8(sbi, unicode, longname, size); + int len = fat_uni_to_x8(sb, unicode, longname, size); fill_name = longname; fill_len = len; @@ -621,7 +622,7 @@ parse_record: if (isvfat) { bufuname[j] = 0x0000; - i = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname)); + i = fat_uni_to_x8(sb, bufuname, bufname, sizeof(bufname)); } if (nr_slots) { /* hack for fat_ioctl_filldir() */ @@ -979,6 +980,7 @@ static int __fat_remove_entries(struct inode *dir, loff_t pos, int nr_slots) int fat_remove_entries(struct inode *dir, struct fat_slot_info *sinfo) { + struct super_block *sb = dir->i_sb; struct msdos_dir_entry *de; struct buffer_head *bh; int err = 0, nr_slots; @@ -1013,8 +1015,8 @@ int fat_remove_entries(struct inode *dir, struct fat_slot_info *sinfo) */ err = __fat_remove_entries(dir, sinfo->slot_off, nr_slots); if (err) { - printk(KERN_WARNING - "FAT: Couldn't remove the long name slots\n"); + fat_msg(sb, KERN_WARNING, + "Couldn't remove the long name slots"); } } @@ -1265,7 +1267,7 @@ int fat_add_entries(struct inode *dir, void *slots, int nr_slots, if (sbi->fat_bits != 32) goto error; } else if (MSDOS_I(dir)->i_start == 0) { - printk(KERN_ERR "FAT: Corrupted directory (i_pos %lld)\n", + fat_msg(sb, KERN_ERR, "Corrupted directory (i_pos %lld)", MSDOS_I(dir)->i_pos); err = -EIO; goto error; diff --git a/fs/fat/fat.h b/fs/fat/fat.h index f504089..8276cc2 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -319,19 +319,20 @@ extern struct inode *fat_build_inode(struct super_block *sb, struct msdos_dir_entry *de, loff_t i_pos); extern int fat_sync_inode(struct inode *inode); extern int fat_fill_super(struct super_block *sb, void *data, int silent, - const struct inode_operations *fs_dir_inode_ops, - int isvfat, void (*setup)(struct super_block *)); + int isvfat, void (*setup)(struct super_block *)); extern int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2); /* fat/misc.c */ extern void -__fat_fs_error(struct super_block *s, int report, const char *fmt, ...) +__fat_fs_error(struct super_block *sb, int report, const char *fmt, ...) + __attribute__ ((format (printf, 3, 4))) __cold; +#define fat_fs_error(sb, fmt, args...) \ + __fat_fs_error(sb, 1, fmt , ## args) +#define fat_fs_error_ratelimit(sb, fmt, args...) \ + __fat_fs_error(sb, __ratelimit(&MSDOS_SB(sb)->ratelimit), fmt , ## args) +void fat_msg(struct super_block *sb, const char *level, const char *fmt, ...) __attribute__ ((format (printf, 3, 4))) __cold; -#define fat_fs_error(s, fmt, args...) \ - __fat_fs_error(s, 1, fmt , ## args) -#define fat_fs_error_ratelimit(s, fmt, args...) \ - __fat_fs_error(s, __ratelimit(&MSDOS_SB(s)->ratelimit), fmt , ## args) extern int fat_clusters_flush(struct super_block *sb); extern int fat_chain_add(struct inode *inode, int new_dclus, int nr_cluster); extern void fat_time_fat2unix(struct msdos_sb_info *sbi, struct timespec *ts, diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index b47d2c9..2e81ac0 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -95,7 +95,7 @@ static int fat12_ent_bread(struct super_block *sb, struct fat_entry *fatent, err_brelse: brelse(bhs[0]); err: - printk(KERN_ERR "FAT: FAT read failed (blocknr %llu)\n", (llu)blocknr); + fat_msg(sb, KERN_ERR, "FAT read failed (blocknr %llu)", (llu)blocknr); return -EIO; } @@ -108,7 +108,7 @@ static int fat_ent_bread(struct super_block *sb, struct fat_entry *fatent, fatent->fat_inode = MSDOS_SB(sb)->fat_inode; fatent->bhs[0] = sb_bread(sb, blocknr); if (!fatent->bhs[0]) { - printk(KERN_ERR "FAT: FAT read failed (blocknr %llu)\n", + fat_msg(sb, KERN_ERR, "FAT read failed (blocknr %llu)", (llu)blocknr); return -EIO; } diff --git a/fs/fat/file.c b/fs/fat/file.c index 7257752..7018e1d 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -102,7 +102,7 @@ static int fat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) if (attr & ATTR_SYS) inode->i_flags |= S_IMMUTABLE; else - inode->i_flags &= S_IMMUTABLE; + inode->i_flags &= ~S_IMMUTABLE; } fat_save_attrs(inode, attr); diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 8d68690..cb8d839 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -581,7 +581,8 @@ static int fat_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_bavail = sbi->free_clusters; buf->f_fsid.val[0] = (u32)id; buf->f_fsid.val[1] = (u32)(id >> 32); - buf->f_namelen = sbi->options.isvfat ? FAT_LFN_LEN : 12; + buf->f_namelen = + (sbi->options.isvfat ? FAT_LFN_LEN : 12) * NLS_MAX_CHARSET_SIZE; return 0; } @@ -619,8 +620,8 @@ retry: bh = sb_bread(sb, i_pos >> sbi->dir_per_block_bits); if (!bh) { - printk(KERN_ERR "FAT: unable to read inode block " - "for updating (i_pos %lld)\n", i_pos); + fat_msg(sb, KERN_ERR, "unable to read inode block " + "for updating (i_pos %lld)", i_pos); return -EIO; } spin_lock(&sbi->inode_hash_lock); @@ -976,8 +977,8 @@ static const match_table_t vfat_tokens = { {Opt_err, NULL} }; -static int parse_options(char *options, int is_vfat, int silent, int *debug, - struct fat_mount_options *opts) +static int parse_options(struct super_block *sb, char *options, int is_vfat, + int silent, int *debug, struct fat_mount_options *opts) { char *p; substring_t args[MAX_OPT_ARGS]; @@ -1168,15 +1169,15 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug, /* obsolete mount options */ case Opt_obsolate: - printk(KERN_INFO "FAT: \"%s\" option is obsolete, " - "not supported now\n", p); + fat_msg(sb, KERN_INFO, "\"%s\" option is obsolete, " + "not supported now", p); break; /* unknown option */ default: if (!silent) { - printk(KERN_ERR - "FAT: Unrecognized mount option \"%s\" " - "or missing value\n", p); + fat_msg(sb, KERN_ERR, + "Unrecognized mount option \"%s\" " + "or missing value", p); } return -EINVAL; } @@ -1185,7 +1186,7 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug, out: /* UTF-8 doesn't provide FAT semantics */ if (!strcmp(opts->iocharset, "utf8")) { - printk(KERN_ERR "FAT: utf8 is not a recommended IO charset" + fat_msg(sb, KERN_ERR, "utf8 is not a recommended IO charset" " for FAT filesystems, filesystem will be " "case sensitive!\n"); } @@ -1238,8 +1239,7 @@ static int fat_read_root(struct inode *inode) /* * Read the super block of an MS-DOS FS. */ -int fat_fill_super(struct super_block *sb, void *data, int silent, - const struct inode_operations *fs_dir_inode_ops, int isvfat, +int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, void (*setup)(struct super_block *)) { struct inode *root_inode = NULL, *fat_inode = NULL; @@ -1268,11 +1268,10 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sb->s_magic = MSDOS_SUPER_MAGIC; sb->s_op = &fat_sops; sb->s_export_op = &fat_export_ops; - sbi->dir_ops = fs_dir_inode_ops; ratelimit_state_init(&sbi->ratelimit, DEFAULT_RATELIMIT_INTERVAL, DEFAULT_RATELIMIT_BURST); - error = parse_options(data, isvfat, silent, &debug, &sbi->options); + error = parse_options(sb, data, isvfat, silent, &debug, &sbi->options); if (error) goto out_fail; @@ -1282,20 +1281,20 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sb_min_blocksize(sb, 512); bh = sb_bread(sb, 0); if (bh == NULL) { - printk(KERN_ERR "FAT: unable to read boot sector\n"); + fat_msg(sb, KERN_ERR, "unable to read boot sector"); goto out_fail; } b = (struct fat_boot_sector *) bh->b_data; if (!b->reserved) { if (!silent) - printk(KERN_ERR "FAT: bogus number of reserved sectors\n"); + fat_msg(sb, KERN_ERR, "bogus number of reserved sectors"); brelse(bh); goto out_invalid; } if (!b->fats) { if (!silent) - printk(KERN_ERR "FAT: bogus number of FAT structure\n"); + fat_msg(sb, KERN_ERR, "bogus number of FAT structure"); brelse(bh); goto out_invalid; } @@ -1308,7 +1307,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, media = b->media; if (!fat_valid_media(media)) { if (!silent) - printk(KERN_ERR "FAT: invalid media value (0x%02x)\n", + fat_msg(sb, KERN_ERR, "invalid media value (0x%02x)", media); brelse(bh); goto out_invalid; @@ -1318,7 +1317,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, || (logical_sector_size < 512) || (logical_sector_size > 4096)) { if (!silent) - printk(KERN_ERR "FAT: bogus logical sector size %u\n", + fat_msg(sb, KERN_ERR, "bogus logical sector size %u", logical_sector_size); brelse(bh); goto out_invalid; @@ -1326,15 +1325,15 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sbi->sec_per_clus = b->sec_per_clus; if (!is_power_of_2(sbi->sec_per_clus)) { if (!silent) - printk(KERN_ERR "FAT: bogus sectors per cluster %u\n", + fat_msg(sb, KERN_ERR, "bogus sectors per cluster %u", sbi->sec_per_clus); brelse(bh); goto out_invalid; } if (logical_sector_size < sb->s_blocksize) { - printk(KERN_ERR "FAT: logical sector size too small for device" - " (logical sector size = %u)\n", logical_sector_size); + fat_msg(sb, KERN_ERR, "logical sector size too small for device" + " (logical sector size = %u)", logical_sector_size); brelse(bh); goto out_fail; } @@ -1342,14 +1341,14 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, brelse(bh); if (!sb_set_blocksize(sb, logical_sector_size)) { - printk(KERN_ERR "FAT: unable to set blocksize %u\n", + fat_msg(sb, KERN_ERR, "unable to set blocksize %u", logical_sector_size); goto out_fail; } bh = sb_bread(sb, 0); if (bh == NULL) { - printk(KERN_ERR "FAT: unable to read boot sector" - " (logical sector size = %lu)\n", + fat_msg(sb, KERN_ERR, "unable to read boot sector" + " (logical sector size = %lu)", sb->s_blocksize); goto out_fail; } @@ -1385,16 +1384,16 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, fsinfo_bh = sb_bread(sb, sbi->fsinfo_sector); if (fsinfo_bh == NULL) { - printk(KERN_ERR "FAT: bread failed, FSINFO block" - " (sector = %lu)\n", sbi->fsinfo_sector); + fat_msg(sb, KERN_ERR, "bread failed, FSINFO block" + " (sector = %lu)", sbi->fsinfo_sector); brelse(bh); goto out_fail; } fsinfo = (struct fat_boot_fsinfo *)fsinfo_bh->b_data; if (!IS_FSINFO(fsinfo)) { - printk(KERN_WARNING "FAT: Invalid FSINFO signature: " - "0x%08x, 0x%08x (sector = %lu)\n", + fat_msg(sb, KERN_WARNING, "Invalid FSINFO signature: " + "0x%08x, 0x%08x (sector = %lu)", le32_to_cpu(fsinfo->signature1), le32_to_cpu(fsinfo->signature2), sbi->fsinfo_sector); @@ -1415,8 +1414,8 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sbi->dir_entries = get_unaligned_le16(&b->dir_entries); if (sbi->dir_entries & (sbi->dir_per_block - 1)) { if (!silent) - printk(KERN_ERR "FAT: bogus directroy-entries per block" - " (%u)\n", sbi->dir_entries); + fat_msg(sb, KERN_ERR, "bogus directroy-entries per block" + " (%u)", sbi->dir_entries); brelse(bh); goto out_invalid; } @@ -1438,7 +1437,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, total_clusters = min(total_clusters, fat_clusters - FAT_START_ENT); if (total_clusters > MAX_FAT(sb)) { if (!silent) - printk(KERN_ERR "FAT: count of clusters too big (%u)\n", + fat_msg(sb, KERN_ERR, "count of clusters too big (%u)", total_clusters); brelse(bh); goto out_invalid; @@ -1471,7 +1470,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sprintf(buf, "cp%d", sbi->options.codepage); sbi->nls_disk = load_nls(buf); if (!sbi->nls_disk) { - printk(KERN_ERR "FAT: codepage %s not found\n", buf); + fat_msg(sb, KERN_ERR, "codepage %s not found", buf); goto out_fail; } @@ -1479,7 +1478,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, if (sbi->options.isvfat) { sbi->nls_io = load_nls(sbi->options.iocharset); if (!sbi->nls_io) { - printk(KERN_ERR "FAT: IO charset %s not found\n", + fat_msg(sb, KERN_ERR, "IO charset %s not found", sbi->options.iocharset); goto out_fail; } @@ -1503,7 +1502,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, insert_inode_hash(root_inode); sb->s_root = d_alloc_root(root_inode); if (!sb->s_root) { - printk(KERN_ERR "FAT: get root inode failed\n"); + fat_msg(sb, KERN_ERR, "get root inode failed"); goto out_fail; } @@ -1512,8 +1511,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, out_invalid: error = -EINVAL; if (!silent) - printk(KERN_INFO "VFS: Can't find a valid FAT filesystem" - " on dev %s.\n", sb->s_id); + fat_msg(sb, KERN_INFO, "Can't find a valid FAT filesystem"); out_fail: if (fat_inode) diff --git a/fs/fat/misc.c b/fs/fat/misc.c index 970e682..6d93360 100644 --- a/fs/fat/misc.c +++ b/fs/fat/misc.c @@ -20,30 +20,46 @@ * In case the file system is remounted read-only, it can be made writable * again by remounting it. */ -void __fat_fs_error(struct super_block *s, int report, const char *fmt, ...) +void __fat_fs_error(struct super_block *sb, int report, const char *fmt, ...) { - struct fat_mount_options *opts = &MSDOS_SB(s)->options; + struct fat_mount_options *opts = &MSDOS_SB(sb)->options; va_list args; + struct va_format vaf; if (report) { - printk(KERN_ERR "FAT: Filesystem error (dev %s)\n", s->s_id); - - printk(KERN_ERR " "); va_start(args, fmt); - vprintk(fmt, args); + vaf.fmt = fmt; + vaf.va = &args; + printk(KERN_ERR "FAT-fs (%s): error, %pV\n", sb->s_id, &vaf); va_end(args); - printk("\n"); } if (opts->errors == FAT_ERRORS_PANIC) - panic("FAT: fs panic from previous error\n"); - else if (opts->errors == FAT_ERRORS_RO && !(s->s_flags & MS_RDONLY)) { - s->s_flags |= MS_RDONLY; - printk(KERN_ERR "FAT: Filesystem has been set read-only\n"); + panic("FAT-fs (%s): fs panic from previous error\n", sb->s_id); + else if (opts->errors == FAT_ERRORS_RO && !(sb->s_flags & MS_RDONLY)) { + sb->s_flags |= MS_RDONLY; + printk(KERN_ERR "FAT-fs (%s): Filesystem has been " + "set read-only\n", sb->s_id); } } EXPORT_SYMBOL_GPL(__fat_fs_error); +/** + * fat_msg() - print preformated FAT specific messages. Every thing what is + * not fat_fs_error() should be fat_msg(). + */ +void fat_msg(struct super_block *sb, const char *level, const char *fmt, ...) +{ + struct va_format vaf; + va_list args; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + printk("%sFAT-fs (%s): %pV\n", level, sb->s_id, &vaf); + va_end(args); +} + /* Flushes the number of free clusters on FAT32 */ /* XXX: Need to write one per FSINFO block. Currently only writes 1 */ int fat_clusters_flush(struct super_block *sb) @@ -57,15 +73,15 @@ int fat_clusters_flush(struct super_block *sb) bh = sb_bread(sb, sbi->fsinfo_sector); if (bh == NULL) { - printk(KERN_ERR "FAT: bread failed in fat_clusters_flush\n"); + fat_msg(sb, KERN_ERR, "bread failed in fat_clusters_flush"); return -EIO; } fsinfo = (struct fat_boot_fsinfo *)bh->b_data; /* Sanity check */ if (!IS_FSINFO(fsinfo)) { - printk(KERN_ERR "FAT: Invalid FSINFO signature: " - "0x%08x, 0x%08x (sector = %lu)\n", + fat_msg(sb, KERN_ERR, "Invalid FSINFO signature: " + "0x%08x, 0x%08x (sector = %lu)", le32_to_cpu(fsinfo->signature1), le32_to_cpu(fsinfo->signature2), sbi->fsinfo_sector); diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 7114990..3b222da 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -659,14 +659,14 @@ static const struct inode_operations msdos_dir_inode_operations = { static void setup(struct super_block *sb) { + MSDOS_SB(sb)->dir_ops = &msdos_dir_inode_operations; sb->s_d_op = &msdos_dentry_operations; sb->s_flags |= MS_NOATIME; } static int msdos_fill_super(struct super_block *sb, void *data, int silent) { - return fat_fill_super(sb, data, silent, &msdos_dir_inode_operations, - 0, setup); + return fat_fill_super(sb, data, silent, 0, setup); } static struct dentry *msdos_mount(struct file_system_type *fs_type, diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index adae3fb..20b4ea5 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -1065,6 +1065,7 @@ static const struct inode_operations vfat_dir_inode_operations = { static void setup(struct super_block *sb) { + MSDOS_SB(sb)->dir_ops = &vfat_dir_inode_operations; if (MSDOS_SB(sb)->options.name_check != 's') sb->s_d_op = &vfat_ci_dentry_ops; else @@ -1073,8 +1074,7 @@ static void setup(struct super_block *sb) static int vfat_fill_super(struct super_block *sb, void *data, int silent) { - return fat_fill_super(sb, data, silent, &vfat_dir_inode_operations, - 1, setup); + return fat_fill_super(sb, data, silent, 1, setup); } static struct dentry *vfat_mount(struct file_system_type *fs_type, diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 2ba6719..1a43114 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -272,7 +272,7 @@ vxfs_get_fake_inode(struct super_block *sbp, struct vxfs_inode_info *vip) * *ip: VFS inode * * Description: - * vxfs_put_fake_inode frees all data asssociated with @ip. + * vxfs_put_fake_inode frees all data associated with @ip. */ void vxfs_put_fake_inode(struct inode *ip) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 34591ee8..0f015a0 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1007,9 +1007,6 @@ static noinline void block_dump___mark_inode_dirty(struct inode *inode) * In short, make sure you hash any inodes _before_ you start marking * them dirty. * - * This function *must* be atomic for the I_DIRTY_PAGES case - - * set_page_dirty() is called under spinlock in several places. - * * Note that for blockdevs, inode->dirtied_when represents the dirtying time of * the block-special inode (/dev/hda1) itself. And the ->dirtied_when field of * the kernel-internal blockdev inode represents the dirtying time of the @@ -1028,7 +1025,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) */ if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) { if (sb->s_op->dirty_inode) - sb->s_op->dirty_inode(inode); + sb->s_op->dirty_inode(inode, flags); } /* diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c index 48a18f1..30afdfa 100644 --- a/fs/fscache/operation.c +++ b/fs/fscache/operation.c @@ -33,8 +33,6 @@ void fscache_enqueue_operation(struct fscache_operation *op) _enter("{OBJ%x OP%x,%u}", op->object->debug_id, op->debug_id, atomic_read(&op->usage)); - fscache_set_op_state(op, "EnQ"); - ASSERT(list_empty(&op->pend_link)); ASSERT(op->processor != NULL); ASSERTCMP(op->object->state, >=, FSCACHE_OBJECT_AVAILABLE); @@ -66,8 +64,6 @@ EXPORT_SYMBOL(fscache_enqueue_operation); static void fscache_run_op(struct fscache_object *object, struct fscache_operation *op) { - fscache_set_op_state(op, "Run"); - object->n_in_progress++; if (test_and_clear_bit(FSCACHE_OP_WAITING, &op->flags)) wake_up_bit(&op->flags, FSCACHE_OP_WAITING); @@ -88,8 +84,6 @@ int fscache_submit_exclusive_op(struct fscache_object *object, _enter("{OBJ%x OP%x},", object->debug_id, op->debug_id); - fscache_set_op_state(op, "SubmitX"); - spin_lock(&object->lock); ASSERTCMP(object->n_ops, >=, object->n_in_progress); ASSERTCMP(object->n_ops, >=, object->n_exclusive); @@ -194,8 +188,6 @@ int fscache_submit_op(struct fscache_object *object, ASSERTCMP(atomic_read(&op->usage), >, 0); - fscache_set_op_state(op, "Submit"); - spin_lock(&object->lock); ASSERTCMP(object->n_ops, >=, object->n_in_progress); ASSERTCMP(object->n_ops, >=, object->n_exclusive); @@ -335,8 +327,6 @@ void fscache_put_operation(struct fscache_operation *op) if (!atomic_dec_and_test(&op->usage)) return; - fscache_set_op_state(op, "Put"); - _debug("PUT OP"); if (test_and_set_bit(FSCACHE_OP_DEAD, &op->flags)) BUG(); diff --git a/fs/fscache/page.c b/fs/fscache/page.c index 41c441c..a2a5d19 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -155,11 +155,9 @@ static void fscache_attr_changed_op(struct fscache_operation *op) fscache_stat(&fscache_n_attr_changed_calls); if (fscache_object_is_active(object)) { - fscache_set_op_state(op, "CallFS"); fscache_stat(&fscache_n_cop_attr_changed); ret = object->cache->ops->attr_changed(object); fscache_stat_d(&fscache_n_cop_attr_changed); - fscache_set_op_state(op, "Done"); if (ret < 0) fscache_abort_object(object); } @@ -190,7 +188,6 @@ int __fscache_attr_changed(struct fscache_cookie *cookie) fscache_operation_init(op, fscache_attr_changed_op, NULL); op->flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_EXCLUSIVE); - fscache_set_op_name(op, "Attr"); spin_lock(&cookie->lock); @@ -257,7 +254,6 @@ static struct fscache_retrieval *fscache_alloc_retrieval( op->context = context; op->start_time = jiffies; INIT_LIST_HEAD(&op->to_do); - fscache_set_op_name(&op->op, "Retr"); return op; } @@ -368,7 +364,6 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie, _leave(" = -ENOMEM"); return -ENOMEM; } - fscache_set_op_name(&op->op, "RetrRA1"); spin_lock(&cookie->lock); @@ -487,7 +482,6 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie, op = fscache_alloc_retrieval(mapping, end_io_func, context); if (!op) return -ENOMEM; - fscache_set_op_name(&op->op, "RetrRAN"); spin_lock(&cookie->lock); @@ -589,7 +583,6 @@ int __fscache_alloc_page(struct fscache_cookie *cookie, op = fscache_alloc_retrieval(page->mapping, NULL, NULL); if (!op) return -ENOMEM; - fscache_set_op_name(&op->op, "RetrAL1"); spin_lock(&cookie->lock); @@ -662,8 +655,6 @@ static void fscache_write_op(struct fscache_operation *_op) _enter("{OP%x,%d}", op->op.debug_id, atomic_read(&op->op.usage)); - fscache_set_op_state(&op->op, "GetPage"); - spin_lock(&object->lock); cookie = object->cookie; @@ -698,15 +689,12 @@ static void fscache_write_op(struct fscache_operation *_op) spin_unlock(&cookie->stores_lock); spin_unlock(&object->lock); - fscache_set_op_state(&op->op, "Store"); fscache_stat(&fscache_n_store_pages); fscache_stat(&fscache_n_cop_write_page); ret = object->cache->ops->write_page(op, page); fscache_stat_d(&fscache_n_cop_write_page); - fscache_set_op_state(&op->op, "EndWrite"); fscache_end_page_write(object, page); if (ret < 0) { - fscache_set_op_state(&op->op, "Abort"); fscache_abort_object(object); } else { fscache_enqueue_operation(&op->op); @@ -778,7 +766,6 @@ int __fscache_write_page(struct fscache_cookie *cookie, fscache_operation_init(&op->op, fscache_write_op, fscache_release_write_op); op->op.flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_WAITING); - fscache_set_op_name(&op->op, "Write1"); ret = radix_tree_preload(gfp & ~__GFP_HIGHMEM); if (ret < 0) diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index b32eb29..d5016071 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -691,6 +691,7 @@ static int fuse_rename(struct inode *olddir, struct dentry *oldent, struct fuse_rename_in inarg; struct fuse_conn *fc = get_fuse_conn(olddir); struct fuse_req *req = fuse_get_req(fc); + if (IS_ERR(req)) return PTR_ERR(req); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index cc6ec4b..38f84cd 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -921,6 +921,8 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (sb->s_flags & MS_MANDLOCK) goto err; + sb->s_flags &= ~MS_NOSEC; + if (!parse_fuse_opt((char *) data, &d, is_bdev)) goto err; diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile index f3d23ef..8612820 100644 --- a/fs/gfs2/Makefile +++ b/fs/gfs2/Makefile @@ -1,9 +1,9 @@ ccflags-y := -I$(src) obj-$(CONFIG_GFS2_FS) += gfs2.o gfs2-y := acl.o bmap.o dir.o xattr.o glock.o \ - glops.o inode.o log.o lops.o main.o meta_io.o \ + glops.o log.o lops.o main.o meta_io.o \ aops.o dentry.o export.o file.o \ - ops_fstype.o ops_inode.o quota.o \ + ops_fstype.o inode.o quota.o \ recovery.o rgrp.o super.o sys.o trans.o util.o gfs2-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 0f5c4f9..802ac5e 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -1076,8 +1076,8 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) bd = bh->b_private; if (bd && bd->bd_ail) goto cannot_release; - gfs2_assert_warn(sdp, !buffer_pinned(bh)); - gfs2_assert_warn(sdp, !buffer_dirty(bh)); + if (buffer_pinned(bh) || buffer_dirty(bh)) + goto not_possible; bh = bh->b_this_page; } while(bh != head); gfs2_log_unlock(sdp); @@ -1107,6 +1107,10 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) } while (bh != head); return try_to_free_buffers(page); + +not_possible: /* Should never happen */ + WARN_ON(buffer_dirty(bh)); + WARN_ON(buffer_pinned(bh)); cannot_release: gfs2_log_unlock(sdp); return 0; diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 74add2d..e65493a 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -780,6 +780,8 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, metadata = (height != ip->i_height - 1); if (metadata) revokes = (height) ? sdp->sd_inptrs : sdp->sd_diptrs; + else if (ip->i_depth) + revokes = sdp->sd_inptrs; if (ip != GFS2_I(sdp->sd_rindex)) error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index f789c57..091ee47 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -82,12 +82,9 @@ struct qstr gfs2_qdot __read_mostly; struct qstr gfs2_qdotdot __read_mostly; -typedef int (*leaf_call_t) (struct gfs2_inode *dip, u32 index, u32 len, - u64 leaf_no, void *data); typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent, const struct qstr *name, void *opaque); - int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block, struct buffer_head **bhp) { @@ -1600,7 +1597,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) */ int gfs2_dir_add(struct inode *inode, const struct qstr *name, - const struct gfs2_inode *nip, unsigned type) + const struct gfs2_inode *nip) { struct gfs2_inode *ip = GFS2_I(inode); struct buffer_head *bh; @@ -1616,7 +1613,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, return PTR_ERR(dent); dent = gfs2_init_dirent(inode, dent, name, bh); gfs2_inum_out(nip, dent); - dent->de_type = cpu_to_be16(type); + dent->de_type = cpu_to_be16(IF2DT(nip->i_inode.i_mode)); if (ip->i_diskflags & GFS2_DIF_EXHASH) { leaf = (struct gfs2_leaf *)bh->b_data; be16_add_cpu(&leaf->lf_entries, 1); @@ -1628,6 +1625,8 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_entries++; ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; + if (S_ISDIR(nip->i_inode.i_mode)) + inc_nlink(&ip->i_inode); gfs2_dinode_out(ip, bh->b_data); brelse(bh); error = 0; @@ -1672,8 +1671,9 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, * Returns: 0 on success, error code on failure */ -int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) +int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry) { + const struct qstr *name = &dentry->d_name; struct gfs2_dirent *dent, *prev = NULL; struct buffer_head *bh; int error; @@ -1714,6 +1714,8 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) gfs2_trans_add_bh(dip->i_gl, bh, 1); dip->i_entries--; dip->i_inode.i_mtime = dip->i_inode.i_ctime = CURRENT_TIME; + if (S_ISDIR(dentry->d_inode->i_mode)) + drop_nlink(&dip->i_inode); gfs2_dinode_out(dip, bh->b_data); brelse(bh); mark_inode_dirty(&dip->i_inode); @@ -1768,94 +1770,20 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, } /** - * foreach_leaf - call a function for each leaf in a directory - * @dip: the directory - * @lc: the function to call for each each - * @data: private data to pass to it - * - * Returns: errno - */ - -static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data) -{ - struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct buffer_head *bh; - struct gfs2_leaf *leaf; - u32 hsize, len; - u32 ht_offset, lp_offset, ht_offset_cur = -1; - u32 index = 0; - __be64 *lp; - u64 leaf_no; - int error = 0; - - hsize = 1 << dip->i_depth; - if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) { - gfs2_consist_inode(dip); - return -EIO; - } - - lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS); - if (!lp) - return -ENOMEM; - - while (index < hsize) { - lp_offset = index & (sdp->sd_hash_ptrs - 1); - ht_offset = index - lp_offset; - - if (ht_offset_cur != ht_offset) { - error = gfs2_dir_read_data(dip, (char *)lp, - ht_offset * sizeof(__be64), - sdp->sd_hash_bsize, 1); - if (error != sdp->sd_hash_bsize) { - if (error >= 0) - error = -EIO; - goto out; - } - ht_offset_cur = ht_offset; - } - - leaf_no = be64_to_cpu(lp[lp_offset]); - if (leaf_no) { - error = get_leaf(dip, leaf_no, &bh); - if (error) - goto out; - leaf = (struct gfs2_leaf *)bh->b_data; - len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth)); - brelse(bh); - - error = lc(dip, index, len, leaf_no, data); - if (error) - goto out; - - index = (index & ~(len - 1)) + len; - } else - index++; - } - - if (index != hsize) { - gfs2_consist_inode(dip); - error = -EIO; - } - -out: - kfree(lp); - - return error; -} - -/** * leaf_dealloc - Deallocate a directory leaf * @dip: the directory * @index: the hash table offset in the directory * @len: the number of pointers to this leaf * @leaf_no: the leaf number - * @data: not used + * @leaf_bh: buffer_head for the starting leaf + * last_dealloc: 1 if this is the final dealloc for the leaf, else 0 * * Returns: errno */ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, - u64 leaf_no, void *data) + u64 leaf_no, struct buffer_head *leaf_bh, + int last_dealloc) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_leaf *tmp_leaf; @@ -1887,14 +1815,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_qs; /* Count the number of leaves */ + bh = leaf_bh; for (blk = leaf_no; blk; blk = nblk) { - error = get_leaf(dip, blk, &bh); - if (error) - goto out_rlist; + if (blk != leaf_no) { + error = get_leaf(dip, blk, &bh); + if (error) + goto out_rlist; + } tmp_leaf = (struct gfs2_leaf *)bh->b_data; nblk = be64_to_cpu(tmp_leaf->lf_next); - brelse(bh); + if (blk != leaf_no) + brelse(bh); gfs2_rlist_add(sdp, &rlist, blk); l_blocks++; @@ -1918,13 +1850,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, if (error) goto out_rg_gunlock; + bh = leaf_bh; + for (blk = leaf_no; blk; blk = nblk) { - error = get_leaf(dip, blk, &bh); - if (error) - goto out_end_trans; + if (blk != leaf_no) { + error = get_leaf(dip, blk, &bh); + if (error) + goto out_end_trans; + } tmp_leaf = (struct gfs2_leaf *)bh->b_data; nblk = be64_to_cpu(tmp_leaf->lf_next); - brelse(bh); + if (blk != leaf_no) + brelse(bh); gfs2_free_meta(dip, blk, 1); gfs2_add_inode_blocks(&dip->i_inode, -1); @@ -1942,6 +1879,10 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_end_trans; gfs2_trans_add_bh(dip->i_gl, dibh, 1); + /* On the last dealloc, make this a regular file in case we crash. + (We don't want to free these blocks a second time.) */ + if (last_dealloc) + dip->i_inode.i_mode = S_IFREG; gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); @@ -1975,29 +1916,67 @@ int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct buffer_head *bh; - int error; + struct gfs2_leaf *leaf; + u32 hsize, len; + u32 ht_offset, lp_offset, ht_offset_cur = -1; + u32 index = 0, next_index; + __be64 *lp; + u64 leaf_no; + int error = 0, last; - /* Dealloc on-disk leaves to FREEMETA state */ - error = foreach_leaf(dip, leaf_dealloc, NULL); - if (error) - return error; + hsize = 1 << dip->i_depth; + if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) { + gfs2_consist_inode(dip); + return -EIO; + } - /* Make this a regular file in case we crash. - (We don't want to free these blocks a second time.) */ + lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS); + if (!lp) + return -ENOMEM; - error = gfs2_trans_begin(sdp, RES_DINODE, 0); - if (error) - return error; + while (index < hsize) { + lp_offset = index & (sdp->sd_hash_ptrs - 1); + ht_offset = index - lp_offset; - error = gfs2_meta_inode_buffer(dip, &bh); - if (!error) { - gfs2_trans_add_bh(dip->i_gl, bh, 1); - ((struct gfs2_dinode *)bh->b_data)->di_mode = - cpu_to_be32(S_IFREG); - brelse(bh); + if (ht_offset_cur != ht_offset) { + error = gfs2_dir_read_data(dip, (char *)lp, + ht_offset * sizeof(__be64), + sdp->sd_hash_bsize, 1); + if (error != sdp->sd_hash_bsize) { + if (error >= 0) + error = -EIO; + goto out; + } + ht_offset_cur = ht_offset; + } + + leaf_no = be64_to_cpu(lp[lp_offset]); + if (leaf_no) { + error = get_leaf(dip, leaf_no, &bh); + if (error) + goto out; + leaf = (struct gfs2_leaf *)bh->b_data; + len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth)); + + next_index = (index & ~(len - 1)) + len; + last = ((next_index >= hsize) ? 1 : 0); + error = leaf_dealloc(dip, index, len, leaf_no, bh, + last); + brelse(bh); + if (error) + goto out; + index = next_index; + } else + index++; } - gfs2_trans_end(sdp); + if (index != hsize) { + gfs2_consist_inode(dip); + error = -EIO; + } + +out: + kfree(lp); return error; } diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index a98f644..e686af1 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -22,8 +22,8 @@ extern struct inode *gfs2_dir_search(struct inode *dir, extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename, const struct gfs2_inode *ip); extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename, - const struct gfs2_inode *ip, unsigned int type); -extern int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename); + const struct gfs2_inode *ip); +extern int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry); extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, filldir_t filldir); extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index b5a5e60..fe9945f 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -139,7 +139,7 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, struct gfs2_sbd *sdp = sb->s_fs_info; struct inode *inode; - inode = gfs2_ilookup(sb, inum->no_addr); + inode = gfs2_ilookup(sb, inum->no_addr, 0); if (inode) { if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) { iput(inode); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index e483108..a9f5cbe 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -545,18 +545,10 @@ static int gfs2_close(struct inode *inode, struct file *file) /** * gfs2_fsync - sync the dirty data for a file (across the cluster) * @file: the file that points to the dentry (we ignore this) - * @dentry: the dentry that points to the inode to sync + * @datasync: set if we can ignore timestamp changes * - * The VFS will flush "normal" data for us. We only need to worry - * about metadata here. For journaled data, we just do a log flush - * as we can't avoid it. Otherwise we can just bale out if datasync - * is set. For stuffed inodes we must flush the log in order to - * ensure that all data is on disk. - * - * The call to write_inode_now() is there to write back metadata and - * the inode itself. It does also try and write the data, but thats - * (hopefully) a no-op due to the VFS having already called filemap_fdatawrite() - * for us. + * The VFS will flush data for us. We only need to worry + * about metadata here. * * Returns: errno */ @@ -565,22 +557,20 @@ static int gfs2_fsync(struct file *file, int datasync) { struct inode *inode = file->f_mapping->host; int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC); - int ret = 0; - - if (gfs2_is_jdata(GFS2_I(inode))) { - gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); - return 0; - } + struct gfs2_inode *ip = GFS2_I(inode); + int ret; - if (sync_state != 0) { - if (!datasync) - ret = write_inode_now(inode, 0); + if (datasync) + sync_state &= ~I_DIRTY_SYNC; - if (gfs2_is_stuffed(GFS2_I(inode))) - gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); + if (sync_state) { + ret = sync_inode_metadata(inode, 1); + if (ret) + return ret; + gfs2_ail_flush(ip->i_gl); } - return ret; + return 0; } /** @@ -826,6 +816,7 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t bytes, max_bytes; struct gfs2_alloc *al; int error; + loff_t bsize_mask = ~((loff_t)sdp->sd_sb.sb_bsize - 1); loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift; next = (next + 1) << sdp->sd_sb.sb_bsize_shift; @@ -833,13 +824,15 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, if (mode & ~FALLOC_FL_KEEP_SIZE) return -EOPNOTSUPP; - offset = (offset >> sdp->sd_sb.sb_bsize_shift) << - sdp->sd_sb.sb_bsize_shift; + offset &= bsize_mask; len = next - offset; bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2; if (!bytes) bytes = UINT_MAX; + bytes &= bsize_mask; + if (bytes == 0) + bytes = sdp->sd_sb.sb_bsize; gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); error = gfs2_glock_nq(&ip->i_gh); @@ -870,6 +863,9 @@ retry: if (error) { if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) { bytes >>= 1; + bytes &= bsize_mask; + if (bytes == 0) + bytes = sdp->sd_sb.sb_bsize; goto retry; } goto out_qunlock; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 7a4fb63..1c1336e 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -143,14 +143,9 @@ static int demote_ok(const struct gfs2_glock *gl) { const struct gfs2_glock_operations *glops = gl->gl_ops; - /* assert_spin_locked(&gl->gl_spin); */ - if (gl->gl_state == LM_ST_UNLOCKED) return 0; - if (test_bit(GLF_LFLUSH, &gl->gl_flags)) - return 0; - if ((gl->gl_name.ln_type != LM_TYPE_INODE) && - !list_empty(&gl->gl_holders)) + if (!list_empty(&gl->gl_holders)) return 0; if (glops->go_demote_ok) return glops->go_demote_ok(gl); @@ -158,6 +153,31 @@ static int demote_ok(const struct gfs2_glock *gl) } +void gfs2_glock_add_to_lru(struct gfs2_glock *gl) +{ + spin_lock(&lru_lock); + + if (!list_empty(&gl->gl_lru)) + list_del_init(&gl->gl_lru); + else + atomic_inc(&lru_count); + + list_add_tail(&gl->gl_lru, &lru_list); + set_bit(GLF_LRU, &gl->gl_flags); + spin_unlock(&lru_lock); +} + +static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl) +{ + spin_lock(&lru_lock); + if (!list_empty(&gl->gl_lru)) { + list_del_init(&gl->gl_lru); + atomic_dec(&lru_count); + clear_bit(GLF_LRU, &gl->gl_flags); + } + spin_unlock(&lru_lock); +} + /** * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list * @gl: the glock @@ -168,24 +188,8 @@ static int demote_ok(const struct gfs2_glock *gl) static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl) { - if (demote_ok(gl)) { - spin_lock(&lru_lock); - - if (!list_empty(&gl->gl_lru)) - list_del_init(&gl->gl_lru); - else - atomic_inc(&lru_count); - - list_add_tail(&gl->gl_lru, &lru_list); - spin_unlock(&lru_lock); - } -} - -void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl) -{ - spin_lock(&gl->gl_spin); - __gfs2_glock_schedule_for_reclaim(gl); - spin_unlock(&gl->gl_spin); + if (demote_ok(gl)) + gfs2_glock_add_to_lru(gl); } /** @@ -217,12 +221,7 @@ void gfs2_glock_put(struct gfs2_glock *gl) spin_lock_bucket(gl->gl_hash); hlist_bl_del_rcu(&gl->gl_list); spin_unlock_bucket(gl->gl_hash); - spin_lock(&lru_lock); - if (!list_empty(&gl->gl_lru)) { - list_del_init(&gl->gl_lru); - atomic_dec(&lru_count); - } - spin_unlock(&lru_lock); + gfs2_glock_remove_from_lru(gl); GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders)); GLOCK_BUG_ON(gl, mapping && mapping->nrpages); trace_gfs2_glock_put(gl); @@ -542,11 +541,6 @@ __acquires(&gl->gl_spin) clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); gfs2_glock_hold(gl); - if (target != LM_ST_UNLOCKED && (gl->gl_state == LM_ST_SHARED || - gl->gl_state == LM_ST_DEFERRED) && - !(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) - lck_flags |= LM_FLAG_TRY_1CB; - if (sdp->sd_lockstruct.ls_ops->lm_lock) { /* lock_dlm */ ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); @@ -648,7 +642,7 @@ static void delete_work_func(struct work_struct *work) /* Note: Unsafe to dereference ip as we don't hold right refs/locks */ if (ip) - inode = gfs2_ilookup(sdp->sd_vfs, no_addr); + inode = gfs2_ilookup(sdp->sd_vfs, no_addr, 1); else inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED); if (inode && !IS_ERR(inode)) { @@ -669,14 +663,19 @@ static void glock_work_func(struct work_struct *work) drop_ref = 1; } spin_lock(&gl->gl_spin); - if (test_and_clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && + if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && gl->gl_state != LM_ST_UNLOCKED && gl->gl_demote_state != LM_ST_EXCLUSIVE) { unsigned long holdtime, now = jiffies; + holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time; if (time_before(now, holdtime)) delay = holdtime - now; - set_bit(delay ? GLF_PENDING_DEMOTE : GLF_DEMOTE, &gl->gl_flags); + + if (!delay) { + clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags); + set_bit(GLF_DEMOTE, &gl->gl_flags); + } } run_queue(gl, 0); spin_unlock(&gl->gl_spin); @@ -1025,6 +1024,9 @@ int gfs2_glock_nq(struct gfs2_holder *gh) if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) return -EIO; + if (test_bit(GLF_LRU, &gl->gl_flags)) + gfs2_glock_remove_from_lru(gl); + spin_lock(&gl->gl_spin); add_to_queue(gh); if ((LM_FLAG_NOEXP & gh->gh_flags) && @@ -1082,7 +1084,8 @@ void gfs2_glock_dq(struct gfs2_holder *gh) !test_bit(GLF_DEMOTE, &gl->gl_flags)) fast_path = 1; } - __gfs2_glock_schedule_for_reclaim(gl); + if (!test_bit(GLF_LFLUSH, &gl->gl_flags)) + __gfs2_glock_schedule_for_reclaim(gl); trace_gfs2_glock_queue(gh, 0); spin_unlock(&gl->gl_spin); if (likely(fast_path)) @@ -1348,11 +1351,14 @@ void gfs2_glock_complete(struct gfs2_glock *gl, int ret) } -static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) +static int gfs2_shrink_glock_memory(struct shrinker *shrink, + struct shrink_control *sc) { struct gfs2_glock *gl; int may_demote; int nr_skipped = 0; + int nr = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; LIST_HEAD(skipped); if (nr == 0) @@ -1365,6 +1371,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m while(nr && !list_empty(&lru_list)) { gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru); list_del_init(&gl->gl_lru); + clear_bit(GLF_LRU, &gl->gl_flags); atomic_dec(&lru_count); /* Test for being demotable */ @@ -1387,6 +1394,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m } nr_skipped++; list_add(&gl->gl_lru, &skipped); + set_bit(GLF_LRU, &gl->gl_flags); } list_splice(&skipped, &lru_list); atomic_add(nr_skipped, &lru_count); @@ -1459,12 +1467,7 @@ static void thaw_glock(struct gfs2_glock *gl) static void clear_glock(struct gfs2_glock *gl) { - spin_lock(&lru_lock); - if (!list_empty(&gl->gl_lru)) { - list_del_init(&gl->gl_lru); - atomic_dec(&lru_count); - } - spin_unlock(&lru_lock); + gfs2_glock_remove_from_lru(gl); spin_lock(&gl->gl_spin); if (gl->gl_state != LM_ST_UNLOCKED) @@ -1599,9 +1602,11 @@ static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh) return 0; } -static const char *gflags2str(char *buf, const unsigned long *gflags) +static const char *gflags2str(char *buf, const struct gfs2_glock *gl) { + const unsigned long *gflags = &gl->gl_flags; char *p = buf; + if (test_bit(GLF_LOCK, gflags)) *p++ = 'l'; if (test_bit(GLF_DEMOTE, gflags)) @@ -1624,6 +1629,10 @@ static const char *gflags2str(char *buf, const unsigned long *gflags) *p++ = 'F'; if (test_bit(GLF_QUEUED, gflags)) *p++ = 'q'; + if (test_bit(GLF_LRU, gflags)) + *p++ = 'L'; + if (gl->gl_object) + *p++ = 'o'; *p = 0; return buf; } @@ -1658,14 +1667,15 @@ static int __dump_glock(struct seq_file *seq, const struct gfs2_glock *gl) dtime *= 1000000/HZ; /* demote time in uSec */ if (!test_bit(GLF_DEMOTE, &gl->gl_flags)) dtime = 0; - gfs2_print_dbg(seq, "G: s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d r:%d\n", + gfs2_print_dbg(seq, "G: s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d v:%d r:%d\n", state2str(gl->gl_state), gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number, - gflags2str(gflags_buf, &gl->gl_flags), + gflags2str(gflags_buf, gl), state2str(gl->gl_target), state2str(gl->gl_demote_state), dtime, atomic_read(&gl->gl_ail_count), + atomic_read(&gl->gl_revokes), atomic_read(&gl->gl_ref)); list_for_each_entry(gh, &gl->gl_holders, gh_list) { diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index aea1606..6b2f757 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -225,11 +225,10 @@ static inline int gfs2_glock_nq_init(struct gfs2_glock *gl, extern void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state); extern void gfs2_glock_complete(struct gfs2_glock *gl, int ret); -extern void gfs2_reclaim_glock(struct gfs2_sbd *sdp); extern void gfs2_gl_hash_clear(struct gfs2_sbd *sdp); extern void gfs2_glock_finish_truncate(struct gfs2_inode *ip); extern void gfs2_glock_thaw(struct gfs2_sbd *sdp); -extern void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl); +extern void gfs2_glock_add_to_lru(struct gfs2_glock *gl); extern void gfs2_glock_free(struct gfs2_glock *gl); extern int __init gfs2_glock_init(void); diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 25eeb2b..8ef70f4 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -28,33 +28,18 @@ #include "trans.h" /** - * ail_empty_gl - remove all buffers for a given lock from the AIL + * __gfs2_ail_flush - remove all buffers for a given lock from the AIL * @gl: the glock * * None of the buffers should be dirty, locked, or pinned. */ -static void gfs2_ail_empty_gl(struct gfs2_glock *gl) +static void __gfs2_ail_flush(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_sbd; struct list_head *head = &gl->gl_ail_list; struct gfs2_bufdata *bd; struct buffer_head *bh; - struct gfs2_trans tr; - - memset(&tr, 0, sizeof(tr)); - tr.tr_revokes = atomic_read(&gl->gl_ail_count); - - if (!tr.tr_revokes) - return; - - /* A shortened, inline version of gfs2_trans_begin() */ - tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64)); - tr.tr_ip = (unsigned long)__builtin_return_address(0); - INIT_LIST_HEAD(&tr.tr_list_buf); - gfs2_log_reserve(sdp, tr.tr_reserved); - BUG_ON(current->journal_info); - current->journal_info = &tr; spin_lock(&sdp->sd_ail_lock); while (!list_empty(head)) { @@ -76,7 +61,47 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl) } gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count)); spin_unlock(&sdp->sd_ail_lock); +} + + +static void gfs2_ail_empty_gl(struct gfs2_glock *gl) +{ + struct gfs2_sbd *sdp = gl->gl_sbd; + struct gfs2_trans tr; + + memset(&tr, 0, sizeof(tr)); + tr.tr_revokes = atomic_read(&gl->gl_ail_count); + + if (!tr.tr_revokes) + return; + + /* A shortened, inline version of gfs2_trans_begin() */ + tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64)); + tr.tr_ip = (unsigned long)__builtin_return_address(0); + INIT_LIST_HEAD(&tr.tr_list_buf); + gfs2_log_reserve(sdp, tr.tr_reserved); + BUG_ON(current->journal_info); + current->journal_info = &tr; + + __gfs2_ail_flush(gl); + + gfs2_trans_end(sdp); + gfs2_log_flush(sdp, NULL); +} + +void gfs2_ail_flush(struct gfs2_glock *gl) +{ + struct gfs2_sbd *sdp = gl->gl_sbd; + unsigned int revokes = atomic_read(&gl->gl_ail_count); + int ret; + + if (!revokes) + return; + ret = gfs2_trans_begin(sdp, 0, revokes); + if (ret) + return; + __gfs2_ail_flush(gl); gfs2_trans_end(sdp); gfs2_log_flush(sdp, NULL); } @@ -227,6 +252,119 @@ static int inode_go_demote_ok(const struct gfs2_glock *gl) } /** + * gfs2_set_nlink - Set the inode's link count based on on-disk info + * @inode: The inode in question + * @nlink: The link count + * + * If the link count has hit zero, it must never be raised, whatever the + * on-disk inode might say. When new struct inodes are created the link + * count is set to 1, so that we can safely use this test even when reading + * in on disk information for the first time. + */ + +static void gfs2_set_nlink(struct inode *inode, u32 nlink) +{ + /* + * We will need to review setting the nlink count here in the + * light of the forthcoming ro bind mount work. This is a reminder + * to do that. + */ + if ((inode->i_nlink != nlink) && (inode->i_nlink != 0)) { + if (nlink == 0) + clear_nlink(inode); + else + inode->i_nlink = nlink; + } +} + +static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) +{ + const struct gfs2_dinode *str = buf; + struct timespec atime; + u16 height, depth; + + if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) + goto corrupt; + ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino); + ip->i_inode.i_mode = be32_to_cpu(str->di_mode); + ip->i_inode.i_rdev = 0; + switch (ip->i_inode.i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), + be32_to_cpu(str->di_minor)); + break; + }; + + ip->i_inode.i_uid = be32_to_cpu(str->di_uid); + ip->i_inode.i_gid = be32_to_cpu(str->di_gid); + gfs2_set_nlink(&ip->i_inode, be32_to_cpu(str->di_nlink)); + i_size_write(&ip->i_inode, be64_to_cpu(str->di_size)); + gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); + atime.tv_sec = be64_to_cpu(str->di_atime); + atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); + if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0) + ip->i_inode.i_atime = atime; + ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); + ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec); + ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); + ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec); + + ip->i_goal = be64_to_cpu(str->di_goal_meta); + ip->i_generation = be64_to_cpu(str->di_generation); + + ip->i_diskflags = be32_to_cpu(str->di_flags); + gfs2_set_inode_flags(&ip->i_inode); + height = be16_to_cpu(str->di_height); + if (unlikely(height > GFS2_MAX_META_HEIGHT)) + goto corrupt; + ip->i_height = (u8)height; + + depth = be16_to_cpu(str->di_depth); + if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) + goto corrupt; + ip->i_depth = (u8)depth; + ip->i_entries = be32_to_cpu(str->di_entries); + + ip->i_eattr = be64_to_cpu(str->di_eattr); + if (S_ISREG(ip->i_inode.i_mode)) + gfs2_set_aops(&ip->i_inode); + + return 0; +corrupt: + gfs2_consist_inode(ip); + return -EIO; +} + +/** + * gfs2_inode_refresh - Refresh the incore copy of the dinode + * @ip: The GFS2 inode + * + * Returns: errno + */ + +int gfs2_inode_refresh(struct gfs2_inode *ip) +{ + struct buffer_head *dibh; + int error; + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + return error; + + if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) { + brelse(dibh); + return -EIO; + } + + error = gfs2_dinode_in(ip, dibh->b_data); + brelse(dibh); + clear_bit(GIF_INVALID, &ip->i_flags); + + return error; +} + +/** * inode_go_lock - operation done after an inode lock is locked by a process * @gl: the glock * @flags: diff --git a/fs/gfs2/glops.h b/fs/gfs2/glops.h index b3aa2e3..6fce409 100644 --- a/fs/gfs2/glops.h +++ b/fs/gfs2/glops.h @@ -23,4 +23,6 @@ extern const struct gfs2_glock_operations gfs2_quota_glops; extern const struct gfs2_glock_operations gfs2_journal_glops; extern const struct gfs2_glock_operations *gfs2_glops_list[]; +extern void gfs2_ail_flush(struct gfs2_glock *gl); + #endif /* __GLOPS_DOT_H__ */ diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 870a89d..0a064e9 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -20,7 +20,6 @@ #define DIO_WAIT 0x00000010 #define DIO_METADATA 0x00000020 -#define DIO_ALL 0x00000100 struct gfs2_log_operations; struct gfs2_log_element; @@ -200,6 +199,8 @@ enum { GLF_INITIAL = 10, GLF_FROZEN = 11, GLF_QUEUED = 12, + GLF_LRU = 13, + GLF_OBJECT = 14, /* Used only for tracing */ }; struct gfs2_glock { @@ -234,6 +235,7 @@ struct gfs2_glock { struct list_head gl_ail_list; atomic_t gl_ail_count; + atomic_t gl_revokes; struct delayed_work gl_work; struct work_struct gl_delete; struct rcu_head gl_rcu; @@ -374,8 +376,6 @@ struct gfs2_ail { unsigned int ai_first; struct list_head ai_ail1_list; struct list_head ai_ail2_list; - - u64 ai_sync_gen; }; struct gfs2_journal_extent { @@ -488,7 +488,6 @@ struct gfs2_sb_host { char sb_lockproto[GFS2_LOCKNAME_LEN]; char sb_locktable[GFS2_LOCKNAME_LEN]; - u8 sb_uuid[16]; }; /* @@ -654,7 +653,6 @@ struct gfs2_sbd { spinlock_t sd_ail_lock; struct list_head sd_ail1_list; struct list_head sd_ail2_list; - u64 sd_ail_sync_gen; /* Replay stuff */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 9134dcb..03e0c52 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -1,23 +1,25 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions * of the GNU General Public License version 2. */ -#include <linux/sched.h> #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/buffer_head.h> +#include <linux/namei.h> +#include <linux/mm.h> +#include <linux/xattr.h> #include <linux/posix_acl.h> -#include <linux/sort.h> #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> +#include <linux/fiemap.h> #include <linux/security.h> -#include <linux/time.h> +#include <asm/uaccess.h> #include "gfs2.h" #include "incore.h" @@ -26,19 +28,14 @@ #include "dir.h" #include "xattr.h" #include "glock.h" -#include "glops.h" #include "inode.h" -#include "log.h" #include "meta_io.h" #include "quota.h" #include "rgrp.h" #include "trans.h" #include "util.h" - -struct gfs2_inum_range_host { - u64 ir_start; - u64 ir_length; -}; +#include "super.h" +#include "glops.h" struct gfs2_skip_data { u64 no_addr; @@ -74,14 +71,14 @@ static int iget_set(struct inode *inode, void *opaque) return 0; } -struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr) +struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int non_block) { unsigned long hash = (unsigned long)no_addr; struct gfs2_skip_data data; data.no_addr = no_addr; data.skipped = 0; - data.non_block = 0; + data.non_block = non_block; return ilookup5(sb, hash, iget_test, &data); } @@ -248,203 +245,6 @@ fail_iput: goto fail; } -static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) -{ - const struct gfs2_dinode *str = buf; - struct timespec atime; - u16 height, depth; - - if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) - goto corrupt; - ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino); - ip->i_inode.i_mode = be32_to_cpu(str->di_mode); - ip->i_inode.i_rdev = 0; - switch (ip->i_inode.i_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), - be32_to_cpu(str->di_minor)); - break; - }; - - ip->i_inode.i_uid = be32_to_cpu(str->di_uid); - ip->i_inode.i_gid = be32_to_cpu(str->di_gid); - /* - * We will need to review setting the nlink count here in the - * light of the forthcoming ro bind mount work. This is a reminder - * to do that. - */ - ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink); - i_size_write(&ip->i_inode, be64_to_cpu(str->di_size)); - gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); - atime.tv_sec = be64_to_cpu(str->di_atime); - atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); - if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0) - ip->i_inode.i_atime = atime; - ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); - ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec); - ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); - ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec); - - ip->i_goal = be64_to_cpu(str->di_goal_meta); - ip->i_generation = be64_to_cpu(str->di_generation); - - ip->i_diskflags = be32_to_cpu(str->di_flags); - gfs2_set_inode_flags(&ip->i_inode); - height = be16_to_cpu(str->di_height); - if (unlikely(height > GFS2_MAX_META_HEIGHT)) - goto corrupt; - ip->i_height = (u8)height; - - depth = be16_to_cpu(str->di_depth); - if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) - goto corrupt; - ip->i_depth = (u8)depth; - ip->i_entries = be32_to_cpu(str->di_entries); - - ip->i_eattr = be64_to_cpu(str->di_eattr); - if (S_ISREG(ip->i_inode.i_mode)) - gfs2_set_aops(&ip->i_inode); - - return 0; -corrupt: - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; -} - -/** - * gfs2_inode_refresh - Refresh the incore copy of the dinode - * @ip: The GFS2 inode - * - * Returns: errno - */ - -int gfs2_inode_refresh(struct gfs2_inode *ip) -{ - struct buffer_head *dibh; - int error; - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - return error; - - if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) { - brelse(dibh); - return -EIO; - } - - error = gfs2_dinode_in(ip, dibh->b_data); - brelse(dibh); - clear_bit(GIF_INVALID, &ip->i_flags); - - return error; -} - -int gfs2_dinode_dealloc(struct gfs2_inode *ip) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al; - struct gfs2_rgrpd *rgd; - int error; - - if (gfs2_get_inode_blocks(&ip->i_inode) != 1) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - al = gfs2_alloc_get(ip); - if (!al) - return -ENOMEM; - - error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); - if (error) - goto out; - - error = gfs2_rindex_hold(sdp, &al->al_ri_gh); - if (error) - goto out_qs; - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - if (!rgd) { - gfs2_consist_inode(ip); - error = -EIO; - goto out_rindex_relse; - } - - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, - &al->al_rgd_gh); - if (error) - goto out_rindex_relse; - - error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA, 1); - if (error) - goto out_rg_gunlock; - - set_bit(GLF_DIRTY, &ip->i_gl->gl_flags); - set_bit(GLF_LFLUSH, &ip->i_gl->gl_flags); - - gfs2_free_di(rgd, ip); - - gfs2_trans_end(sdp); - -out_rg_gunlock: - gfs2_glock_dq_uninit(&al->al_rgd_gh); -out_rindex_relse: - gfs2_glock_dq_uninit(&al->al_ri_gh); -out_qs: - gfs2_quota_unhold(ip); -out: - gfs2_alloc_put(ip); - return error; -} - -/** - * gfs2_change_nlink - Change nlink count on inode - * @ip: The GFS2 inode - * @diff: The change in the nlink count required - * - * Returns: errno - */ -int gfs2_change_nlink(struct gfs2_inode *ip, int diff) -{ - struct buffer_head *dibh; - u32 nlink; - int error; - - BUG_ON(diff != 1 && diff != -1); - nlink = ip->i_inode.i_nlink + diff; - - /* If we are reducing the nlink count, but the new value ends up being - bigger than the old one, we must have underflowed. */ - if (diff < 0 && nlink > ip->i_inode.i_nlink) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - return error; - - if (diff > 0) - inc_nlink(&ip->i_inode); - else - drop_nlink(&ip->i_inode); - - ip->i_inode.i_ctime = CURRENT_TIME; - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - mark_inode_dirty(&ip->i_inode); - - if (ip->i_inode.i_nlink == 0) - gfs2_unlink_di(&ip->i_inode); /* mark inode unlinked */ - - return error; -} struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) { @@ -543,7 +343,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, /* Don't create entries in an unlinked directory */ if (!dip->i_inode.i_nlink) - return -EPERM; + return -ENOENT; error = gfs2_dir_check(&dip->i_inode, name, NULL); switch (error) { @@ -613,21 +413,44 @@ out: return error; } +static void gfs2_init_dir(struct buffer_head *dibh, + const struct gfs2_inode *parent) +{ + struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; + struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1); + + gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent); + dent->de_inum = di->di_num; /* already GFS2 endian */ + dent->de_type = cpu_to_be16(DT_DIR); + + dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1)); + gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent); + gfs2_inum_out(parent, dent); + dent->de_type = cpu_to_be16(DT_DIR); + +} + /** * init_dinode - Fill in a new dinode structure - * @dip: the directory this inode is being created in + * @dip: The directory this inode is being created in * @gl: The glock covering the new inode - * @inum: the inode number - * @mode: the file permissions - * @uid: - * @gid: + * @inum: The inode number + * @mode: The file permissions + * @uid: The uid of the new inode + * @gid: The gid of the new inode + * @generation: The generation number of the new inode + * @dev: The device number (if a device node) + * @symname: The symlink destination (if a symlink) + * @size: The inode size (ignored for directories) + * @bhp: The buffer head (returned to caller) * */ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, const struct gfs2_inum_host *inum, unsigned int mode, unsigned int uid, unsigned int gid, - const u64 *generation, dev_t dev, struct buffer_head **bhp) + const u64 *generation, dev_t dev, const char *symname, + unsigned size, struct buffer_head **bhp) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_dinode *di; @@ -646,7 +469,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_uid = cpu_to_be32(uid); di->di_gid = cpu_to_be32(gid); di->di_nlink = 0; - di->di_size = 0; + di->di_size = cpu_to_be64(size); di->di_blocks = cpu_to_be64(1); di->di_atime = di->di_mtime = di->di_ctime = cpu_to_be64(tv.tv_sec); di->di_major = cpu_to_be32(MAJOR(dev)); @@ -654,16 +477,6 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_goal_meta = di->di_goal_data = cpu_to_be64(inum->no_addr); di->di_generation = cpu_to_be64(*generation); di->di_flags = 0; - - if (S_ISREG(mode)) { - if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) || - gfs2_tune_get(sdp, gt_new_files_jdata)) - di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); - } else if (S_ISDIR(mode)) { - di->di_flags |= cpu_to_be32(dip->i_diskflags & - GFS2_DIF_INHERIT_JDATA); - } - di->__pad1 = 0; di->di_payload_format = cpu_to_be32(S_ISDIR(mode) ? GFS2_FORMAT_DE : 0); di->di_height = 0; @@ -677,7 +490,26 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_mtime_nsec = cpu_to_be32(tv.tv_nsec); di->di_ctime_nsec = cpu_to_be32(tv.tv_nsec); memset(&di->di_reserved, 0, sizeof(di->di_reserved)); - + + switch(mode & S_IFMT) { + case S_IFREG: + if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) || + gfs2_tune_get(sdp, gt_new_files_jdata)) + di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); + break; + case S_IFDIR: + di->di_flags |= cpu_to_be32(dip->i_diskflags & + GFS2_DIF_INHERIT_JDATA); + di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); + di->di_size = cpu_to_be64(sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)); + di->di_entries = cpu_to_be32(2); + gfs2_init_dir(dibh, dip); + break; + case S_IFLNK: + memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, size); + break; + } + set_buffer_uptodate(dibh); *bhp = dibh; @@ -685,7 +517,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, unsigned int mode, const struct gfs2_inum_host *inum, - const u64 *generation, dev_t dev, struct buffer_head **bhp) + const u64 *generation, dev_t dev, const char *symname, + unsigned int size, struct buffer_head **bhp) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); unsigned int uid, gid; @@ -707,7 +540,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, if (error) goto out_quota; - init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, bhp); + init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, symname, size, bhp); gfs2_quota_change(dip, +1, uid, gid); gfs2_trans_end(sdp); @@ -761,14 +594,16 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, goto fail_quota_locks; } - error = gfs2_dir_add(&dip->i_inode, name, ip, IF2DT(ip->i_inode.i_mode)); + error = gfs2_dir_add(&dip->i_inode, name, ip); if (error) goto fail_end_trans; error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto fail_end_trans; - ip->i_inode.i_nlink = 1; + inc_nlink(&ip->i_inode); + if (S_ISDIR(ip->i_inode.i_mode)) + inc_nlink(&ip->i_inode); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -815,27 +650,25 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip, } /** - * gfs2_createi - Create a new inode - * @ghs: An array of two holders - * @name: The name of the new file - * @mode: the permissions on the new inode - * - * @ghs[0] is an initialized holder for the directory - * @ghs[1] is the holder for the inode lock + * gfs2_create_inode - Create a new inode + * @dir: The parent directory + * @dentry: The new dentry + * @mode: The permissions on the new inode + * @dev: For device nodes, this is the device number + * @symname: For symlinks, this is the link destination + * @size: The initial size of the inode (ignored for directories) * - * If the return value is not NULL, the glocks on both the directory and the new - * file are held. A transaction has been started and an inplace reservation - * is held, as well. - * - * Returns: An inode + * Returns: 0 on success, or error code */ -struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode, dev_t dev) +static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, + unsigned int mode, dev_t dev, const char *symname, + unsigned int size) { + const struct qstr *name = &dentry->d_name; + struct gfs2_holder ghs[2]; struct inode *inode = NULL; - struct gfs2_inode *dip = ghs->gh_gl->gl_object; - struct inode *dir = &dip->i_inode; + struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_inum_host inum = { .no_addr = 0, .no_formal_ino = 0 }; int error; @@ -843,10 +676,9 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, struct buffer_head *bh = NULL; if (!name->len || name->len > GFS2_FNAMESIZE) - return ERR_PTR(-ENAMETOOLONG); + return -ENAMETOOLONG; - gfs2_holder_reinit(LM_ST_EXCLUSIVE, 0, ghs); - error = gfs2_glock_nq(ghs); + error = gfs2_glock_nq_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); if (error) goto fail; @@ -864,7 +696,7 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (error) goto fail_gunlock; - error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, &bh); + error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, symname, size, &bh); if (error) goto fail_gunlock2; @@ -891,18 +723,852 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (bh) brelse(bh); - return inode; + + gfs2_trans_end(sdp); + if (dip->i_alloc->al_rgd) + gfs2_inplace_release(dip); + gfs2_quota_unlock(dip); + gfs2_alloc_put(dip); + gfs2_glock_dq_uninit_m(2, ghs); + mark_inode_dirty(inode); + d_instantiate(dentry, inode); + return 0; fail_gunlock2: gfs2_glock_dq_uninit(ghs + 1); if (inode && !IS_ERR(inode)) iput(inode); fail_gunlock: - gfs2_glock_dq(ghs); + gfs2_glock_dq_uninit(ghs); fail: if (bh) brelse(bh); - return ERR_PTR(error); + return error; +} + +/** + * gfs2_create - Create a file + * @dir: The directory in which to create the file + * @dentry: The dentry of the new file + * @mode: The mode of the new file + * + * Returns: errno + */ + +static int gfs2_create(struct inode *dir, struct dentry *dentry, + int mode, struct nameidata *nd) +{ + struct inode *inode; + int ret; + + for (;;) { + ret = gfs2_create_inode(dir, dentry, S_IFREG | mode, 0, NULL, 0); + if (ret != -EEXIST || (nd && (nd->flags & LOOKUP_EXCL))) + return ret; + + inode = gfs2_lookupi(dir, &dentry->d_name, 0); + if (inode) { + if (!IS_ERR(inode)) + break; + return PTR_ERR(inode); + } + } + + d_instantiate(dentry, inode); + return 0; +} + +/** + * gfs2_lookup - Look up a filename in a directory and return its inode + * @dir: The directory inode + * @dentry: The dentry of the new inode + * @nd: passed from Linux VFS, ignored by us + * + * Called by the VFS layer. Lock dir and call gfs2_lookupi() + * + * Returns: errno + */ + +static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct inode *inode = NULL; + + inode = gfs2_lookupi(dir, &dentry->d_name, 0); + if (inode && IS_ERR(inode)) + return ERR_CAST(inode); + + if (inode) { + struct gfs2_glock *gl = GFS2_I(inode)->i_gl; + struct gfs2_holder gh; + int error; + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) { + iput(inode); + return ERR_PTR(error); + } + gfs2_glock_dq_uninit(&gh); + return d_splice_alias(inode, dentry); + } + d_add(dentry, inode); + + return NULL; +} + +/** + * gfs2_link - Link to a file + * @old_dentry: The inode to link + * @dir: Add link to this directory + * @dentry: The name of the link + * + * Link the inode in "old_dentry" into the directory "dir" with the + * name in "dentry". + * + * Returns: errno + */ + +static int gfs2_link(struct dentry *old_dentry, struct inode *dir, + struct dentry *dentry) +{ + struct gfs2_inode *dip = GFS2_I(dir); + struct gfs2_sbd *sdp = GFS2_SB(dir); + struct inode *inode = old_dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder ghs[2]; + struct buffer_head *dibh; + int alloc_required; + int error; + + if (S_ISDIR(inode->i_mode)) + return -EPERM; + + gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); + + error = gfs2_glock_nq(ghs); /* parent */ + if (error) + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; + + error = -ENOENT; + if (inode->i_nlink == 0) + goto out_gunlock; + + error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0); + if (error) + goto out_gunlock; + + error = gfs2_dir_check(dir, &dentry->d_name, NULL); + switch (error) { + case -ENOENT: + break; + case 0: + error = -EEXIST; + default: + goto out_gunlock; + } + + error = -EINVAL; + if (!dip->i_inode.i_nlink) + goto out_gunlock; + error = -EFBIG; + if (dip->i_entries == (u32)-1) + goto out_gunlock; + error = -EPERM; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + goto out_gunlock; + error = -EINVAL; + if (!ip->i_inode.i_nlink) + goto out_gunlock; + error = -EMLINK; + if (ip->i_inode.i_nlink == (u32)-1) + goto out_gunlock; + + alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); + if (error < 0) + goto out_gunlock; + error = 0; + + if (alloc_required) { + struct gfs2_alloc *al = gfs2_alloc_get(dip); + if (!al) { + error = -ENOMEM; + goto out_gunlock; + } + + error = gfs2_quota_lock_check(dip); + if (error) + goto out_alloc; + + al->al_requested = sdp->sd_max_dirres; + + error = gfs2_inplace_reserve(dip); + if (error) + goto out_gunlock_q; + + error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + + gfs2_rg_blocks(al) + + 2 * RES_DINODE + RES_STATFS + + RES_QUOTA, 0); + if (error) + goto out_ipres; + } else { + error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0); + if (error) + goto out_ipres; + } + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + goto out_end_trans; + + error = gfs2_dir_add(dir, &dentry->d_name, ip); + if (error) + goto out_brelse; + + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + inc_nlink(&ip->i_inode); + ip->i_inode.i_ctime = CURRENT_TIME; + gfs2_dinode_out(ip, dibh->b_data); + mark_inode_dirty(&ip->i_inode); + +out_brelse: + brelse(dibh); +out_end_trans: + gfs2_trans_end(sdp); +out_ipres: + if (alloc_required) + gfs2_inplace_release(dip); +out_gunlock_q: + if (alloc_required) + gfs2_quota_unlock(dip); +out_alloc: + if (alloc_required) + gfs2_alloc_put(dip); +out_gunlock: + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_glock_dq(ghs); +out_parent: + gfs2_holder_uninit(ghs); + gfs2_holder_uninit(ghs + 1); + if (!error) { + ihold(inode); + d_instantiate(dentry, inode); + mark_inode_dirty(inode); + } + return error; +} + +/* + * gfs2_unlink_ok - check to see that a inode is still in a directory + * @dip: the directory + * @name: the name of the file + * @ip: the inode + * + * Assumes that the lock on (at least) @dip is held. + * + * Returns: 0 if the parent/child relationship is correct, errno if it isn't + */ + +static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, + const struct gfs2_inode *ip) +{ + int error; + + if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) + return -EPERM; + + if ((dip->i_inode.i_mode & S_ISVTX) && + dip->i_inode.i_uid != current_fsuid() && + ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER)) + return -EPERM; + + if (IS_APPEND(&dip->i_inode)) + return -EPERM; + + error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); + if (error) + return error; + + error = gfs2_dir_check(&dip->i_inode, name, ip); + if (error) + return error; + + return 0; +} + +/** + * gfs2_unlink_inode - Removes an inode from its parent dir and unlinks it + * @dip: The parent directory + * @name: The name of the entry in the parent directory + * @bh: The inode buffer for the inode to be removed + * @inode: The inode to be removed + * + * Called with all the locks and in a transaction. This will only be + * called for a directory after it has been checked to ensure it is empty. + * + * Returns: 0 on success, or an error + */ + +static int gfs2_unlink_inode(struct gfs2_inode *dip, + const struct dentry *dentry, + struct buffer_head *bh) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + int error; + + error = gfs2_dir_del(dip, dentry); + if (error) + return error; + + ip->i_entries = 0; + inode->i_ctime = CURRENT_TIME; + if (S_ISDIR(inode->i_mode)) + clear_nlink(inode); + else + drop_nlink(inode); + gfs2_trans_add_bh(ip->i_gl, bh, 1); + gfs2_dinode_out(ip, bh->b_data); + mark_inode_dirty(inode); + if (inode->i_nlink == 0) + gfs2_unlink_di(inode); + return 0; +} + + +/** + * gfs2_unlink - Unlink an inode (this does rmdir as well) + * @dir: The inode of the directory containing the inode to unlink + * @dentry: The file itself + * + * This routine uses the type of the inode as a flag to figure out + * whether this is an unlink or an rmdir. + * + * Returns: errno + */ + +static int gfs2_unlink(struct inode *dir, struct dentry *dentry) +{ + struct gfs2_inode *dip = GFS2_I(dir); + struct gfs2_sbd *sdp = GFS2_SB(dir); + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct buffer_head *bh; + struct gfs2_holder ghs[3]; + struct gfs2_rgrpd *rgd; + struct gfs2_holder ri_gh; + int error; + + error = gfs2_rindex_hold(sdp, &ri_gh); + if (error) + return error; + + gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); + + rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); + gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); + + + error = gfs2_glock_nq(ghs); /* parent */ + if (error) + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; + + error = -ENOENT; + if (inode->i_nlink == 0) + goto out_rgrp; + + if (S_ISDIR(inode->i_mode)) { + error = -ENOTEMPTY; + if (ip->i_entries > 2 || inode->i_nlink > 2) + goto out_rgrp; + } + + error = gfs2_glock_nq(ghs + 2); /* rgrp */ + if (error) + goto out_rgrp; + + error = gfs2_unlink_ok(dip, &dentry->d_name, ip); + if (error) + goto out_gunlock; + + error = gfs2_trans_begin(sdp, 2*RES_DINODE + 3*RES_LEAF + RES_RG_BIT, 0); + if (error) + goto out_gunlock; + + error = gfs2_meta_inode_buffer(ip, &bh); + if (error) + goto out_end_trans; + + error = gfs2_unlink_inode(dip, dentry, bh); + brelse(bh); + +out_end_trans: + gfs2_trans_end(sdp); +out_gunlock: + gfs2_glock_dq(ghs + 2); +out_rgrp: + gfs2_holder_uninit(ghs + 2); + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_holder_uninit(ghs + 1); + gfs2_glock_dq(ghs); +out_parent: + gfs2_holder_uninit(ghs); + gfs2_glock_dq_uninit(&ri_gh); + return error; +} + +/** + * gfs2_symlink - Create a symlink + * @dir: The directory to create the symlink in + * @dentry: The dentry to put the symlink in + * @symname: The thing which the link points to + * + * Returns: errno + */ + +static int gfs2_symlink(struct inode *dir, struct dentry *dentry, + const char *symname) +{ + struct gfs2_sbd *sdp = GFS2_SB(dir); + unsigned int size; + + size = strlen(symname); + if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1) + return -ENAMETOOLONG; + + return gfs2_create_inode(dir, dentry, S_IFLNK | S_IRWXUGO, 0, symname, size); +} + +/** + * gfs2_mkdir - Make a directory + * @dir: The parent directory of the new one + * @dentry: The dentry of the new directory + * @mode: The mode of the new directory + * + * Returns: errno + */ + +static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) +{ + return gfs2_create_inode(dir, dentry, S_IFDIR | mode, 0, NULL, 0); +} + +/** + * gfs2_mknod - Make a special file + * @dir: The directory in which the special file will reside + * @dentry: The dentry of the special file + * @mode: The mode of the special file + * @dev: The device specification of the special file + * + */ + +static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, + dev_t dev) +{ + return gfs2_create_inode(dir, dentry, mode, dev, NULL, 0); +} + +/* + * gfs2_ok_to_move - check if it's ok to move a directory to another directory + * @this: move this + * @to: to here + * + * Follow @to back to the root and make sure we don't encounter @this + * Assumes we already hold the rename lock. + * + * Returns: errno + */ + +static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) +{ + struct inode *dir = &to->i_inode; + struct super_block *sb = dir->i_sb; + struct inode *tmp; + int error = 0; + + igrab(dir); + + for (;;) { + if (dir == &this->i_inode) { + error = -EINVAL; + break; + } + if (dir == sb->s_root->d_inode) { + error = 0; + break; + } + + tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1); + if (IS_ERR(tmp)) { + error = PTR_ERR(tmp); + break; + } + + iput(dir); + dir = tmp; + } + + iput(dir); + + return error; +} + +/** + * gfs2_rename - Rename a file + * @odir: Parent directory of old file name + * @odentry: The old dentry of the file + * @ndir: Parent directory of new file name + * @ndentry: The new dentry of the file + * + * Returns: errno + */ + +static int gfs2_rename(struct inode *odir, struct dentry *odentry, + struct inode *ndir, struct dentry *ndentry) +{ + struct gfs2_inode *odip = GFS2_I(odir); + struct gfs2_inode *ndip = GFS2_I(ndir); + struct gfs2_inode *ip = GFS2_I(odentry->d_inode); + struct gfs2_inode *nip = NULL; + struct gfs2_sbd *sdp = GFS2_SB(odir); + struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh; + struct gfs2_rgrpd *nrgd; + unsigned int num_gh; + int dir_rename = 0; + int alloc_required = 0; + unsigned int x; + int error; + + if (ndentry->d_inode) { + nip = GFS2_I(ndentry->d_inode); + if (ip == nip) + return 0; + } + + error = gfs2_rindex_hold(sdp, &ri_gh); + if (error) + return error; + + if (odip != ndip) { + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, + 0, &r_gh); + if (error) + goto out; + + if (S_ISDIR(ip->i_inode.i_mode)) { + dir_rename = 1; + /* don't move a dirctory into it's subdir */ + error = gfs2_ok_to_move(ip, ndip); + if (error) + goto out_gunlock_r; + } + } + + num_gh = 1; + gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + if (odip != ndip) { + gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + } + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + + if (nip) { + gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + /* grab the resource lock for unlink flag twiddling + * this is the case of the target file already existing + * so we unlink before doing the rename + */ + nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr); + if (nrgd) + gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); + } + + for (x = 0; x < num_gh; x++) { + error = gfs2_glock_nq(ghs + x); + if (error) + goto out_gunlock; + } + + error = -ENOENT; + if (ip->i_inode.i_nlink == 0) + goto out_gunlock; + + /* Check out the old directory */ + + error = gfs2_unlink_ok(odip, &odentry->d_name, ip); + if (error) + goto out_gunlock; + + /* Check out the new directory */ + + if (nip) { + error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip); + if (error) + goto out_gunlock; + + if (nip->i_inode.i_nlink == 0) { + error = -EAGAIN; + goto out_gunlock; + } + + if (S_ISDIR(nip->i_inode.i_mode)) { + if (nip->i_entries < 2) { + gfs2_consist_inode(nip); + error = -EIO; + goto out_gunlock; + } + if (nip->i_entries > 2) { + error = -ENOTEMPTY; + goto out_gunlock; + } + } + } else { + error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0); + if (error) + goto out_gunlock; + + error = gfs2_dir_check(ndir, &ndentry->d_name, NULL); + switch (error) { + case -ENOENT: + error = 0; + break; + case 0: + error = -EEXIST; + default: + goto out_gunlock; + }; + + if (odip != ndip) { + if (!ndip->i_inode.i_nlink) { + error = -ENOENT; + goto out_gunlock; + } + if (ndip->i_entries == (u32)-1) { + error = -EFBIG; + goto out_gunlock; + } + if (S_ISDIR(ip->i_inode.i_mode) && + ndip->i_inode.i_nlink == (u32)-1) { + error = -EMLINK; + goto out_gunlock; + } + } + } + + /* Check out the dir to be renamed */ + + if (dir_rename) { + error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0); + if (error) + goto out_gunlock; + } + + if (nip == NULL) + alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name); + error = alloc_required; + if (error < 0) + goto out_gunlock; + error = 0; + + if (alloc_required) { + struct gfs2_alloc *al = gfs2_alloc_get(ndip); + if (!al) { + error = -ENOMEM; + goto out_gunlock; + } + + error = gfs2_quota_lock_check(ndip); + if (error) + goto out_alloc; + + al->al_requested = sdp->sd_max_dirres; + + error = gfs2_inplace_reserve_ri(ndip); + if (error) + goto out_gunlock_q; + + error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + + gfs2_rg_blocks(al) + + 4 * RES_DINODE + 4 * RES_LEAF + + RES_STATFS + RES_QUOTA + 4, 0); + if (error) + goto out_ipreserv; + } else { + error = gfs2_trans_begin(sdp, 4 * RES_DINODE + + 5 * RES_LEAF + 4, 0); + if (error) + goto out_gunlock; + } + + /* Remove the target file, if it exists */ + + if (nip) { + struct buffer_head *bh; + error = gfs2_meta_inode_buffer(nip, &bh); + if (error) + goto out_end_trans; + error = gfs2_unlink_inode(ndip, ndentry, bh); + brelse(bh); + } + + if (dir_rename) { + error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR); + if (error) + goto out_end_trans; + } else { + struct buffer_head *dibh; + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + goto out_end_trans; + ip->i_inode.i_ctime = CURRENT_TIME; + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_dinode_out(ip, dibh->b_data); + brelse(dibh); + } + + error = gfs2_dir_del(odip, odentry); + if (error) + goto out_end_trans; + + error = gfs2_dir_add(ndir, &ndentry->d_name, ip); + if (error) + goto out_end_trans; + +out_end_trans: + gfs2_trans_end(sdp); +out_ipreserv: + if (alloc_required) + gfs2_inplace_release(ndip); +out_gunlock_q: + if (alloc_required) + gfs2_quota_unlock(ndip); +out_alloc: + if (alloc_required) + gfs2_alloc_put(ndip); +out_gunlock: + while (x--) { + gfs2_glock_dq(ghs + x); + gfs2_holder_uninit(ghs + x); + } +out_gunlock_r: + if (r_gh.gh_gl) + gfs2_glock_dq_uninit(&r_gh); +out: + gfs2_glock_dq_uninit(&ri_gh); + return error; +} + +/** + * gfs2_follow_link - Follow a symbolic link + * @dentry: The dentry of the link + * @nd: Data that we pass to vfs_follow_link() + * + * This can handle symlinks of any size. + * + * Returns: 0 on success or error code + */ + +static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + struct gfs2_inode *ip = GFS2_I(dentry->d_inode); + struct gfs2_holder i_gh; + struct buffer_head *dibh; + unsigned int size; + char *buf; + int error; + + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); + error = gfs2_glock_nq(&i_gh); + if (error) { + gfs2_holder_uninit(&i_gh); + nd_set_link(nd, ERR_PTR(error)); + return NULL; + } + + size = (unsigned int)i_size_read(&ip->i_inode); + if (size == 0) { + gfs2_consist_inode(ip); + buf = ERR_PTR(-EIO); + goto out; + } + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) { + buf = ERR_PTR(error); + goto out; + } + + buf = kzalloc(size + 1, GFP_NOFS); + if (!buf) + buf = ERR_PTR(-ENOMEM); + else + memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), size); + brelse(dibh); +out: + gfs2_glock_dq_uninit(&i_gh); + nd_set_link(nd, buf); + return NULL; +} + +static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p) +{ + char *s = nd_get_link(nd); + if (!IS_ERR(s)) + kfree(s); +} + +/** + * gfs2_permission - + * @inode: The inode + * @mask: The mask to be tested + * @flags: Indicates whether this is an RCU path walk or not + * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. + * + * Returns: errno + */ + +int gfs2_permission(struct inode *inode, int mask, unsigned int flags) +{ + struct gfs2_inode *ip; + struct gfs2_holder i_gh; + int error; + int unlock = 0; + + + ip = GFS2_I(inode); + if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); + if (error) + return error; + unlock = 1; + } + + if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) + error = -EACCES; + else + error = generic_permission(inode, mask, flags, gfs2_check_acl); + if (unlock) + gfs2_glock_dq_uninit(&i_gh); + + return error; } static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) @@ -928,8 +1594,6 @@ static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) * @ip: * @attr: * - * Called with a reference on the vnode. - * * Returns: errno */ @@ -949,60 +1613,280 @@ int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) return error; } -void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) -{ - struct gfs2_dinode *str = buf; - - str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); - str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); - str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); - str->di_num.no_addr = cpu_to_be64(ip->i_no_addr); - str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino); - str->di_mode = cpu_to_be32(ip->i_inode.i_mode); - str->di_uid = cpu_to_be32(ip->i_inode.i_uid); - str->di_gid = cpu_to_be32(ip->i_inode.i_gid); - str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); - str->di_size = cpu_to_be64(i_size_read(&ip->i_inode)); - str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode)); - str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); - str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); - str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); - - str->di_goal_meta = cpu_to_be64(ip->i_goal); - str->di_goal_data = cpu_to_be64(ip->i_goal); - str->di_generation = cpu_to_be64(ip->i_generation); - - str->di_flags = cpu_to_be32(ip->i_diskflags); - str->di_height = cpu_to_be16(ip->i_height); - str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && - !(ip->i_diskflags & GFS2_DIF_EXHASH) ? - GFS2_FORMAT_DE : 0); - str->di_depth = cpu_to_be16(ip->i_depth); - str->di_entries = cpu_to_be32(ip->i_entries); - - str->di_eattr = cpu_to_be64(ip->i_eattr); - str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); - str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec); - str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec); -} - -void gfs2_dinode_print(const struct gfs2_inode *ip) -{ - printk(KERN_INFO " no_formal_ino = %llu\n", - (unsigned long long)ip->i_no_formal_ino); - printk(KERN_INFO " no_addr = %llu\n", - (unsigned long long)ip->i_no_addr); - printk(KERN_INFO " i_size = %llu\n", - (unsigned long long)i_size_read(&ip->i_inode)); - printk(KERN_INFO " blocks = %llu\n", - (unsigned long long)gfs2_get_inode_blocks(&ip->i_inode)); - printk(KERN_INFO " i_goal = %llu\n", - (unsigned long long)ip->i_goal); - printk(KERN_INFO " i_diskflags = 0x%.8X\n", ip->i_diskflags); - printk(KERN_INFO " i_height = %u\n", ip->i_height); - printk(KERN_INFO " i_depth = %u\n", ip->i_depth); - printk(KERN_INFO " i_entries = %u\n", ip->i_entries); - printk(KERN_INFO " i_eattr = %llu\n", - (unsigned long long)ip->i_eattr); +static int setattr_chown(struct inode *inode, struct iattr *attr) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + u32 ouid, ogid, nuid, ngid; + int error; + + ouid = inode->i_uid; + ogid = inode->i_gid; + nuid = attr->ia_uid; + ngid = attr->ia_gid; + + if (!(attr->ia_valid & ATTR_UID) || ouid == nuid) + ouid = nuid = NO_QUOTA_CHANGE; + if (!(attr->ia_valid & ATTR_GID) || ogid == ngid) + ogid = ngid = NO_QUOTA_CHANGE; + + if (!gfs2_alloc_get(ip)) + return -ENOMEM; + + error = gfs2_quota_lock(ip, nuid, ngid); + if (error) + goto out_alloc; + + if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { + error = gfs2_quota_check(ip, nuid, ngid); + if (error) + goto out_gunlock_q; + } + + error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0); + if (error) + goto out_gunlock_q; + + error = gfs2_setattr_simple(ip, attr); + if (error) + goto out_end_trans; + + if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { + u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); + gfs2_quota_change(ip, -blocks, ouid, ogid); + gfs2_quota_change(ip, blocks, nuid, ngid); + } + +out_end_trans: + gfs2_trans_end(sdp); +out_gunlock_q: + gfs2_quota_unlock(ip); +out_alloc: + gfs2_alloc_put(ip); + return error; +} + +/** + * gfs2_setattr - Change attributes on an inode + * @dentry: The dentry which is changing + * @attr: The structure describing the change + * + * The VFS layer wants to change one or more of an inodes attributes. Write + * that change out to disk. + * + * Returns: errno + */ + +static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder i_gh; + int error; + + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); + if (error) + return error; + + error = -EPERM; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + goto out; + + error = inode_change_ok(inode, attr); + if (error) + goto out; + + if (attr->ia_valid & ATTR_SIZE) + error = gfs2_setattr_size(inode, attr->ia_size); + else if (attr->ia_valid & (ATTR_UID | ATTR_GID)) + error = setattr_chown(inode, attr); + else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode)) + error = gfs2_acl_chmod(ip, attr); + else + error = gfs2_setattr_simple(ip, attr); + +out: + gfs2_glock_dq_uninit(&i_gh); + if (!error) + mark_inode_dirty(inode); + return error; +} + +/** + * gfs2_getattr - Read out an inode's attributes + * @mnt: The vfsmount the inode is being accessed from + * @dentry: The dentry to stat + * @stat: The inode's stats + * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. Note that its the NFS + * readdirplus operation which causes this to be called (from filldir) + * with the glock already held. + * + * Returns: errno + */ + +static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int error; + int unlock = 0; + + if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) + return error; + unlock = 1; + } + + generic_fillattr(inode, stat); + if (unlock) + gfs2_glock_dq_uninit(&gh); + + return 0; +} + +static int gfs2_setxattr(struct dentry *dentry, const char *name, + const void *data, size_t size, int flags) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_setxattr(dentry, name, data, size, flags); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name, + void *data, size_t size) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_getxattr(dentry, name, data, size); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static int gfs2_removexattr(struct dentry *dentry, const char *name) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_removexattr(dentry, name); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC); + if (ret) + return ret; + + mutex_lock(&inode->i_mutex); + + ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + if (ret) + goto out; + + if (gfs2_is_stuffed(ip)) { + u64 phys = ip->i_no_addr << inode->i_blkbits; + u64 size = i_size_read(inode); + u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED| + FIEMAP_EXTENT_DATA_INLINE; + phys += sizeof(struct gfs2_dinode); + phys += start; + if (start + len > size) + len = size - start; + if (start < size) + ret = fiemap_fill_next_extent(fieinfo, start, phys, + len, flags); + if (ret == 1) + ret = 0; + } else { + ret = __generic_block_fiemap(inode, fieinfo, start, len, + gfs2_block_map); + } + + gfs2_glock_dq_uninit(&gh); +out: + mutex_unlock(&inode->i_mutex); + return ret; } +const struct inode_operations gfs2_file_iops = { + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + +const struct inode_operations gfs2_dir_iops = { + .create = gfs2_create, + .lookup = gfs2_lookup, + .link = gfs2_link, + .unlink = gfs2_unlink, + .symlink = gfs2_symlink, + .mkdir = gfs2_mkdir, + .rmdir = gfs2_unlink, + .mknod = gfs2_mknod, + .rename = gfs2_rename, + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + +const struct inode_operations gfs2_symlink_iops = { + .readlink = generic_readlink, + .follow_link = gfs2_follow_link, + .put_link = gfs2_put_link, + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 099ca30..3160607 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -102,22 +102,16 @@ extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, u64 *no_formal_ino, unsigned int blktype); -extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr); +extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int nonblock); extern int gfs2_inode_refresh(struct gfs2_inode *ip); -extern int gfs2_dinode_dealloc(struct gfs2_inode *inode); -extern int gfs2_change_nlink(struct gfs2_inode *ip, int diff); extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, int is_root); -extern struct inode *gfs2_createi(struct gfs2_holder *ghs, - const struct qstr *name, - unsigned int mode, dev_t dev); extern int gfs2_permission(struct inode *inode, int mask, unsigned int flags); extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); -extern void gfs2_dinode_print(const struct gfs2_inode *ip); extern const struct inode_operations gfs2_file_iops; extern const struct inode_operations gfs2_dir_iops; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 5b102c1..903115f 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -18,6 +18,7 @@ #include <linux/kthread.h> #include <linux/freezer.h> #include <linux/bio.h> +#include <linux/writeback.h> #include "gfs2.h" #include "incore.h" @@ -83,55 +84,97 @@ void gfs2_remove_from_ail(struct gfs2_bufdata *bd) /** * gfs2_ail1_start_one - Start I/O on a part of the AIL * @sdp: the filesystem - * @tr: the part of the AIL + * @wbc: The writeback control structure + * @ai: The ail structure * */ -static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) +static int gfs2_ail1_start_one(struct gfs2_sbd *sdp, + struct writeback_control *wbc, + struct gfs2_ail *ai) __releases(&sdp->sd_ail_lock) __acquires(&sdp->sd_ail_lock) { + struct gfs2_glock *gl = NULL; + struct address_space *mapping; struct gfs2_bufdata *bd, *s; struct buffer_head *bh; - int retry; - do { - retry = 0; + list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, bd_ail_st_list) { + bh = bd->bd_bh; - list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, - bd_ail_st_list) { - bh = bd->bd_bh; + gfs2_assert(sdp, bd->bd_ail == ai); - gfs2_assert(sdp, bd->bd_ail == ai); + if (!buffer_busy(bh)) { + if (!buffer_uptodate(bh)) + gfs2_io_error_bh(sdp, bh); + list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); + continue; + } - if (!buffer_busy(bh)) { - if (!buffer_uptodate(bh)) - gfs2_io_error_bh(sdp, bh); - list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); - continue; - } + if (!buffer_dirty(bh)) + continue; + if (gl == bd->bd_gl) + continue; + gl = bd->bd_gl; + list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list); + mapping = bh->b_page->mapping; + if (!mapping) + continue; + spin_unlock(&sdp->sd_ail_lock); + generic_writepages(mapping, wbc); + spin_lock(&sdp->sd_ail_lock); + if (wbc->nr_to_write <= 0) + break; + return 1; + } - if (!buffer_dirty(bh)) - continue; + return 0; +} - list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list); - get_bh(bh); - spin_unlock(&sdp->sd_ail_lock); - lock_buffer(bh); - if (test_clear_buffer_dirty(bh)) { - bh->b_end_io = end_buffer_write_sync; - submit_bh(WRITE_SYNC, bh); - } else { - unlock_buffer(bh); - brelse(bh); - } - spin_lock(&sdp->sd_ail_lock); - - retry = 1; +/** + * gfs2_ail1_flush - start writeback of some ail1 entries + * @sdp: The super block + * @wbc: The writeback control structure + * + * Writes back some ail1 entries, according to the limits in the + * writeback control structure + */ + +void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc) +{ + struct list_head *head = &sdp->sd_ail1_list; + struct gfs2_ail *ai; + + trace_gfs2_ail_flush(sdp, wbc, 1); + spin_lock(&sdp->sd_ail_lock); +restart: + list_for_each_entry_reverse(ai, head, ai_list) { + if (wbc->nr_to_write <= 0) break; - } - } while (retry); + if (gfs2_ail1_start_one(sdp, wbc, ai)) + goto restart; + } + spin_unlock(&sdp->sd_ail_lock); + trace_gfs2_ail_flush(sdp, wbc, 0); +} + +/** + * gfs2_ail1_start - start writeback of all ail1 entries + * @sdp: The superblock + */ + +static void gfs2_ail1_start(struct gfs2_sbd *sdp) +{ + struct writeback_control wbc = { + .sync_mode = WB_SYNC_NONE, + .nr_to_write = LONG_MAX, + .range_start = 0, + .range_end = LLONG_MAX, + }; + + return gfs2_ail1_flush(sdp, &wbc); } /** @@ -141,7 +184,7 @@ __acquires(&sdp->sd_ail_lock) * */ -static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int flags) +static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) { struct gfs2_bufdata *bd, *s; struct buffer_head *bh; @@ -149,76 +192,63 @@ static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int fl list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, bd_ail_st_list) { bh = bd->bd_bh; - gfs2_assert(sdp, bd->bd_ail == ai); - - if (buffer_busy(bh)) { - if (flags & DIO_ALL) - continue; - else - break; - } - + if (buffer_busy(bh)) + continue; if (!buffer_uptodate(bh)) gfs2_io_error_bh(sdp, bh); - list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); } - return list_empty(&ai->ai_ail1_list); } -static void gfs2_ail1_start(struct gfs2_sbd *sdp) -{ - struct list_head *head; - u64 sync_gen; - struct gfs2_ail *ai; - int done = 0; - - spin_lock(&sdp->sd_ail_lock); - head = &sdp->sd_ail1_list; - if (list_empty(head)) { - spin_unlock(&sdp->sd_ail_lock); - return; - } - sync_gen = sdp->sd_ail_sync_gen++; - - while(!done) { - done = 1; - list_for_each_entry_reverse(ai, head, ai_list) { - if (ai->ai_sync_gen >= sync_gen) - continue; - ai->ai_sync_gen = sync_gen; - gfs2_ail1_start_one(sdp, ai); /* This may drop ail lock */ - done = 0; - break; - } - } - - spin_unlock(&sdp->sd_ail_lock); -} +/** + * gfs2_ail1_empty - Try to empty the ail1 lists + * @sdp: The superblock + * + * Tries to empty the ail1 lists, starting with the oldest first + */ -static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags) +static int gfs2_ail1_empty(struct gfs2_sbd *sdp) { struct gfs2_ail *ai, *s; int ret; spin_lock(&sdp->sd_ail_lock); - list_for_each_entry_safe_reverse(ai, s, &sdp->sd_ail1_list, ai_list) { - if (gfs2_ail1_empty_one(sdp, ai, flags)) + gfs2_ail1_empty_one(sdp, ai); + if (list_empty(&ai->ai_ail1_list)) list_move(&ai->ai_list, &sdp->sd_ail2_list); - else if (!(flags & DIO_ALL)) + else break; } - ret = list_empty(&sdp->sd_ail1_list); - spin_unlock(&sdp->sd_ail_lock); return ret; } +static void gfs2_ail1_wait(struct gfs2_sbd *sdp) +{ + struct gfs2_ail *ai; + struct gfs2_bufdata *bd; + struct buffer_head *bh; + + spin_lock(&sdp->sd_ail_lock); + list_for_each_entry_reverse(ai, &sdp->sd_ail1_list, ai_list) { + list_for_each_entry(bd, &ai->ai_ail1_list, bd_ail_st_list) { + bh = bd->bd_bh; + if (!buffer_locked(bh)) + continue; + get_bh(bh); + spin_unlock(&sdp->sd_ail_lock); + wait_on_buffer(bh); + brelse(bh); + return; + } + } + spin_unlock(&sdp->sd_ail_lock); +} /** * gfs2_ail2_empty_one - Check whether or not a trans in the AIL has been synced @@ -574,7 +604,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) set_buffer_uptodate(bh); clear_buffer_dirty(bh); - gfs2_ail1_empty(sdp, 0); + gfs2_ail1_empty(sdp); tail = current_tail(sdp); lh = (struct gfs2_log_header *)bh->b_data; @@ -869,9 +899,9 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp) gfs2_log_flush(sdp, NULL); for (;;) { gfs2_ail1_start(sdp); - if (gfs2_ail1_empty(sdp, DIO_ALL)) + gfs2_ail1_wait(sdp); + if (gfs2_ail1_empty(sdp)) break; - msleep(10); } } @@ -905,20 +935,20 @@ int gfs2_logd(void *data) preflush = atomic_read(&sdp->sd_log_pinned); if (gfs2_jrnl_flush_reqd(sdp) || t == 0) { - gfs2_ail1_empty(sdp, DIO_ALL); + gfs2_ail1_empty(sdp); gfs2_log_flush(sdp, NULL); - gfs2_ail1_empty(sdp, DIO_ALL); } if (gfs2_ail_flush_reqd(sdp)) { gfs2_ail1_start(sdp); - io_schedule(); - gfs2_ail1_empty(sdp, 0); + gfs2_ail1_wait(sdp); + gfs2_ail1_empty(sdp); gfs2_log_flush(sdp, NULL); - gfs2_ail1_empty(sdp, DIO_ALL); } - wake_up(&sdp->sd_log_waitq); + if (!gfs2_ail_flush_reqd(sdp)) + wake_up(&sdp->sd_log_waitq); + t = gfs2_tune_get(sdp, gt_logd_secs) * HZ; if (freezing(current)) refrigerator(); diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 0d007f9..ab06216 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -12,6 +12,7 @@ #include <linux/list.h> #include <linux/spinlock.h> +#include <linux/writeback.h> #include "incore.h" /** @@ -59,6 +60,7 @@ extern struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp, extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl); extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans); extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd); +extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc); extern void gfs2_log_shutdown(struct gfs2_sbd *sdp); extern void gfs2_meta_syncfs(struct gfs2_sbd *sdp); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 51d27f0..05bbb12 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -40,7 +40,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh) { struct gfs2_bufdata *bd; - gfs2_assert_withdraw(sdp, test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)); + BUG_ON(!current->journal_info); clear_buffer_dirty(bh); if (test_set_buffer_pinned(bh)) @@ -65,6 +65,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh) * @sdp: the filesystem the buffer belongs to * @bh: The buffer to unpin * @ai: + * @flags: The inode dirty flags * */ @@ -73,10 +74,8 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, { struct gfs2_bufdata *bd = bh->b_private; - gfs2_assert_withdraw(sdp, buffer_uptodate(bh)); - - if (!buffer_pinned(bh)) - gfs2_assert_withdraw(sdp, 0); + BUG_ON(!buffer_uptodate(bh)); + BUG_ON(!buffer_pinned(bh)); lock_buffer(bh); mark_buffer_dirty(bh); @@ -95,8 +94,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list); spin_unlock(&sdp->sd_ail_lock); - if (test_and_clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags)) - gfs2_glock_schedule_for_reclaim(bd->bd_gl); + clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags); trace_gfs2_pin(bd, 0); unlock_buffer(bh); atomic_dec(&sdp->sd_log_pinned); @@ -322,12 +320,16 @@ static void buf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass) static void revoke_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) { + struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le); + struct gfs2_glock *gl = bd->bd_gl; struct gfs2_trans *tr; tr = current->journal_info; tr->tr_touched = 1; tr->tr_num_revoke++; sdp->sd_log_num_revoke++; + atomic_inc(&gl->gl_revokes); + set_bit(GLF_LFLUSH, &gl->gl_flags); list_add(&le->le_list, &sdp->sd_log_le_revoke); } @@ -350,9 +352,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) ld->ld_data1 = cpu_to_be32(sdp->sd_log_num_revoke); offset = sizeof(struct gfs2_log_descriptor); - while (!list_empty(head)) { - bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list); - list_del_init(&bd->bd_le.le_list); + list_for_each_entry(bd, head, bd_le.le_list) { sdp->sd_log_num_revoke--; if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) { @@ -367,8 +367,6 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) } *(__be64 *)(bh->b_data + offset) = cpu_to_be64(bd->bd_blkno); - kmem_cache_free(gfs2_bufdata_cachep, bd); - offset += sizeof(u64); } gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); @@ -376,6 +374,22 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) submit_bh(WRITE_SYNC, bh); } +static void revoke_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) +{ + struct list_head *head = &sdp->sd_log_le_revoke; + struct gfs2_bufdata *bd; + struct gfs2_glock *gl; + + while (!list_empty(head)) { + bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list); + list_del_init(&bd->bd_le.le_list); + gl = bd->bd_gl; + atomic_dec(&gl->gl_revokes); + clear_bit(GLF_LFLUSH, &gl->gl_flags); + kmem_cache_free(gfs2_bufdata_cachep, bd); + } +} + static void revoke_lo_before_scan(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head, int pass) { @@ -749,6 +763,7 @@ const struct gfs2_log_operations gfs2_buf_lops = { const struct gfs2_log_operations gfs2_revoke_lops = { .lo_add = revoke_lo_add, .lo_before_commit = revoke_lo_before_commit, + .lo_after_commit = revoke_lo_after_commit, .lo_before_scan = revoke_lo_before_scan, .lo_scan_elements = revoke_lo_scan_elements, .lo_after_scan = revoke_lo_after_scan, diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 888a5f5..c2b34cd 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -53,6 +53,7 @@ static void gfs2_init_glock_once(void *foo) INIT_LIST_HEAD(&gl->gl_lru); INIT_LIST_HEAD(&gl->gl_ail_list); atomic_set(&gl->gl_ail_count, 0); + atomic_set(&gl->gl_revokes, 0); } static void gfs2_init_gl_aspace_once(void *foo) @@ -145,7 +146,7 @@ static int __init init_gfs2_fs(void) gfs2_register_debugfs(); - printk("GFS2 (built %s %s) installed\n", __DATE__, __TIME__); + printk("GFS2 installed\n"); return 0; diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 675349b..747238c 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -31,6 +31,7 @@ #include "rgrp.h" #include "trans.h" #include "util.h" +#include "trace_gfs2.h" static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc) { @@ -310,6 +311,7 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, int struct gfs2_bufdata *bd = bh->b_private; if (test_clear_buffer_pinned(bh)) { + trace_gfs2_pin(bd, 0); atomic_dec(&sdp->sd_log_pinned); list_del_init(&bd->bd_le.le_list); if (meta) { diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h index 6a1d9ba..22c5265 100644 --- a/fs/gfs2/meta_io.h +++ b/fs/gfs2/meta_io.h @@ -77,8 +77,6 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen); #define buffer_busy(bh) \ ((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock) | (1ul << BH_Pinned))) -#define buffer_in_io(bh) \ -((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock))) #endif /* __DIO_DOT_H__ */ diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index d3c69eb..8ac9ae1 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -126,8 +126,10 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) * changed. */ -static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) +static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent) { + struct gfs2_sb_host *sb = &sdp->sd_sb; + if (sb->sb_magic != GFS2_MAGIC || sb->sb_type != GFS2_METATYPE_SB) { if (!silent) @@ -157,8 +159,10 @@ static void end_bio_io_page(struct bio *bio, int error) unlock_page(page); } -static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) +static void gfs2_sb_in(struct gfs2_sbd *sdp, const void *buf) { + struct gfs2_sb_host *sb = &sdp->sd_sb; + struct super_block *s = sdp->sd_vfs; const struct gfs2_sb *str = buf; sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); @@ -175,7 +179,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); - memcpy(sb->sb_uuid, str->sb_uuid, 16); + memcpy(s->s_uuid, str->sb_uuid, 16); } /** @@ -197,7 +201,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) * Returns: 0 on success or error */ -static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) +static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent) { struct super_block *sb = sdp->sd_vfs; struct gfs2_sb *p; @@ -227,10 +231,10 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) return -EIO; } p = kmap(page); - gfs2_sb_in(&sdp->sd_sb, p); + gfs2_sb_in(sdp, p); kunmap(page); __free_page(page); - return 0; + return gfs2_check_sb(sdp, silent); } /** @@ -247,17 +251,13 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent) unsigned int x; int error; - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent); if (error) { if (!silent) fs_err(sdp, "can't read superblock\n"); return error; } - error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); - if (error) - return error; - sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT; sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift; @@ -340,14 +340,10 @@ static int init_names(struct gfs2_sbd *sdp, int silent) /* Try to autodetect */ if (!proto[0] || !table[0]) { - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent); if (error) return error; - error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); - if (error) - goto out; - if (!proto[0]) proto = sdp->sd_sb.sb_lockproto; if (!table[0]) @@ -364,7 +360,6 @@ static int init_names(struct gfs2_sbd *sdp, int silent) while ((table = strchr(table, '/'))) *table = '_'; -out: return error; } @@ -1119,8 +1114,7 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent if (sdp->sd_args.ar_statfs_quantum) { sdp->sd_tune.gt_statfs_slow = 0; sdp->sd_tune.gt_statfs_quantum = sdp->sd_args.ar_statfs_quantum; - } - else { + } else { sdp->sd_tune.gt_statfs_slow = 1; sdp->sd_tune.gt_statfs_quantum = 30; } diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c deleted file mode 100644 index 09e436a..0000000 --- a/fs/gfs2/ops_inode.c +++ /dev/null @@ -1,1344 +0,0 @@ -/* - * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License version 2. - */ - -#include <linux/slab.h> -#include <linux/spinlock.h> -#include <linux/completion.h> -#include <linux/buffer_head.h> -#include <linux/namei.h> -#include <linux/mm.h> -#include <linux/xattr.h> -#include <linux/posix_acl.h> -#include <linux/gfs2_ondisk.h> -#include <linux/crc32.h> -#include <linux/fiemap.h> -#include <asm/uaccess.h> - -#include "gfs2.h" -#include "incore.h" -#include "acl.h" -#include "bmap.h" -#include "dir.h" -#include "xattr.h" -#include "glock.h" -#include "inode.h" -#include "meta_io.h" -#include "quota.h" -#include "rgrp.h" -#include "trans.h" -#include "util.h" -#include "super.h" - -/** - * gfs2_create - Create a file - * @dir: The directory in which to create the file - * @dentry: The dentry of the new file - * @mode: The mode of the new file - * - * Returns: errno - */ - -static int gfs2_create(struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - for (;;) { - inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0); - if (!IS_ERR(inode)) { - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - gfs2_glock_dq_uninit_m(2, ghs); - mark_inode_dirty(inode); - break; - } else if (PTR_ERR(inode) != -EEXIST || - (nd && nd->flags & LOOKUP_EXCL)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - inode = gfs2_lookupi(dir, &dentry->d_name, 0); - if (inode) { - if (!IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - break; - } else { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - } - } - - d_instantiate(dentry, inode); - - return 0; -} - -/** - * gfs2_lookup - Look up a filename in a directory and return its inode - * @dir: The directory inode - * @dentry: The dentry of the new inode - * @nd: passed from Linux VFS, ignored by us - * - * Called by the VFS layer. Lock dir and call gfs2_lookupi() - * - * Returns: errno - */ - -static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) -{ - struct inode *inode = NULL; - - inode = gfs2_lookupi(dir, &dentry->d_name, 0); - if (inode && IS_ERR(inode)) - return ERR_CAST(inode); - - if (inode) { - struct gfs2_glock *gl = GFS2_I(inode)->i_gl; - struct gfs2_holder gh; - int error; - error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (error) { - iput(inode); - return ERR_PTR(error); - } - gfs2_glock_dq_uninit(&gh); - return d_splice_alias(inode, dentry); - } - d_add(dentry, inode); - - return NULL; -} - -/** - * gfs2_link - Link to a file - * @old_dentry: The inode to link - * @dir: Add link to this directory - * @dentry: The name of the link - * - * Link the inode in "old_dentry" into the directory "dir" with the - * name in "dentry". - * - * Returns: errno - */ - -static int gfs2_link(struct dentry *old_dentry, struct inode *dir, - struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct inode *inode = old_dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder ghs[2]; - int alloc_required; - int error; - - if (S_ISDIR(inode->i_mode)) - return -EPERM; - - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_check(dir, &dentry->d_name, NULL); - switch (error) { - case -ENOENT: - break; - case 0: - error = -EEXIST; - default: - goto out_gunlock; - } - - error = -EINVAL; - if (!dip->i_inode.i_nlink) - goto out_gunlock; - error = -EFBIG; - if (dip->i_entries == (u32)-1) - goto out_gunlock; - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_gunlock; - error = -EINVAL; - if (!ip->i_inode.i_nlink) - goto out_gunlock; - error = -EMLINK; - if (ip->i_inode.i_nlink == (u32)-1) - goto out_gunlock; - - alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); - if (error < 0) - goto out_gunlock; - error = 0; - - if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(dip); - if (!al) { - error = -ENOMEM; - goto out_gunlock; - } - - error = gfs2_quota_lock_check(dip); - if (error) - goto out_alloc; - - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve(dip); - if (error) - goto out_gunlock_q; - - error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + - gfs2_rg_blocks(al) + - 2 * RES_DINODE + RES_STATFS + - RES_QUOTA, 0); - if (error) - goto out_ipres; - } else { - error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0); - if (error) - goto out_ipres; - } - - error = gfs2_dir_add(dir, &dentry->d_name, ip, IF2DT(inode->i_mode)); - if (error) - goto out_end_trans; - - error = gfs2_change_nlink(ip, +1); - -out_end_trans: - gfs2_trans_end(sdp); -out_ipres: - if (alloc_required) - gfs2_inplace_release(dip); -out_gunlock_q: - if (alloc_required) - gfs2_quota_unlock(dip); -out_alloc: - if (alloc_required) - gfs2_alloc_put(dip); -out_gunlock: - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_holder_uninit(ghs + 1); - if (!error) { - ihold(inode); - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - } - return error; -} - -/* - * gfs2_unlink_ok - check to see that a inode is still in a directory - * @dip: the directory - * @name: the name of the file - * @ip: the inode - * - * Assumes that the lock on (at least) @dip is held. - * - * Returns: 0 if the parent/child relationship is correct, errno if it isn't - */ - -static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, - const struct gfs2_inode *ip) -{ - int error; - - if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) - return -EPERM; - - if ((dip->i_inode.i_mode & S_ISVTX) && - dip->i_inode.i_uid != current_fsuid() && - ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER)) - return -EPERM; - - if (IS_APPEND(&dip->i_inode)) - return -EPERM; - - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); - if (error) - return error; - - error = gfs2_dir_check(&dip->i_inode, name, ip); - if (error) - return error; - - return 0; -} - -/** - * gfs2_unlink - Unlink a file - * @dir: The inode of the directory containing the file to unlink - * @dentry: The file itself - * - * Unlink a file. Call gfs2_unlinki() - * - * Returns: errno - */ - -static int gfs2_unlink(struct inode *dir, struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder ghs[3]; - struct gfs2_rgrpd *rgd; - struct gfs2_holder ri_gh; - int error; - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_glock_nq(ghs + 2); /* rgrp */ - if (error) - goto out_rgrp; - - error = gfs2_unlink_ok(dip, &dentry->d_name, ip); - if (error) - goto out_gunlock; - - error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_del(dip, &dentry->d_name); - if (error) - goto out_end_trans; - - error = gfs2_change_nlink(ip, -1); - -out_end_trans: - gfs2_trans_end(sdp); -out_gunlock: - gfs2_glock_dq(ghs + 2); -out_rgrp: - gfs2_holder_uninit(ghs + 2); - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_holder_uninit(ghs + 1); - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_symlink - Create a symlink - * @dir: The directory to create the symlink in - * @dentry: The dentry to put the symlink in - * @symname: The thing which the link points to - * - * Returns: errno - */ - -static int gfs2_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) -{ - struct gfs2_inode *dip = GFS2_I(dir), *ip; - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - struct buffer_head *dibh; - int size; - int error; - - /* Must be stuffed with a null terminator for gfs2_follow_link() */ - size = strlen(symname); - if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1) - return -ENAMETOOLONG; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO, 0); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - ip = ghs[1].gh_gl->gl_object; - - i_size_write(inode, size); - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(ip, dibh->b_data); - memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, - size); - brelse(dibh); - } - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/** - * gfs2_mkdir - Make a directory - * @dir: The parent directory of the new one - * @dentry: The dentry of the new directory - * @mode: The mode of the new directory - * - * Returns: errno - */ - -static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) -{ - struct gfs2_inode *dip = GFS2_I(dir), *ip; - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - struct buffer_head *dibh; - int error; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode, 0); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - ip = ghs[1].gh_gl->gl_object; - - ip->i_inode.i_nlink = 2; - i_size_write(inode, sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)); - ip->i_diskflags |= GFS2_DIF_JDATA; - ip->i_entries = 2; - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; - struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent); - dent->de_inum = di->di_num; /* already GFS2 endian */ - dent->de_type = cpu_to_be16(DT_DIR); - di->di_entries = cpu_to_be32(1); - - dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1)); - gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent); - - gfs2_inum_out(dip, dent); - dent->de_type = cpu_to_be16(DT_DIR); - - gfs2_dinode_out(ip, di); - - brelse(dibh); - } - - error = gfs2_change_nlink(dip, +1); - gfs2_assert_withdraw(sdp, !error); /* dip already pinned */ - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/** - * gfs2_rmdiri - Remove a directory - * @dip: The parent directory of the directory to be removed - * @name: The name of the directory to be removed - * @ip: The GFS2 inode of the directory to be removed - * - * Assumes Glocks on dip and ip are held - * - * Returns: errno - */ - -static int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, - struct gfs2_inode *ip) -{ - int error; - - if (ip->i_entries != 2) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - error = gfs2_dir_del(dip, name); - if (error) - return error; - - error = gfs2_change_nlink(dip, -1); - if (error) - return error; - - error = gfs2_dir_del(ip, &gfs2_qdot); - if (error) - return error; - - error = gfs2_dir_del(ip, &gfs2_qdotdot); - if (error) - return error; - - /* It looks odd, but it really should be done twice */ - error = gfs2_change_nlink(ip, -1); - if (error) - return error; - - error = gfs2_change_nlink(ip, -1); - if (error) - return error; - - return error; -} - -/** - * gfs2_rmdir - Remove a directory - * @dir: The parent directory of the directory to be removed - * @dentry: The dentry of the directory to remove - * - * Remove a directory. Call gfs2_rmdiri() - * - * Returns: errno - */ - -static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder ghs[3]; - struct gfs2_rgrpd *rgd; - struct gfs2_holder ri_gh; - int error; - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_glock_nq(ghs + 2); /* rgrp */ - if (error) - goto out_rgrp; - - error = gfs2_unlink_ok(dip, &dentry->d_name, ip); - if (error) - goto out_gunlock; - - if (ip->i_entries < 2) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - error = -EIO; - goto out_gunlock; - } - if (ip->i_entries > 2) { - error = -ENOTEMPTY; - goto out_gunlock; - } - - error = gfs2_trans_begin(sdp, 2 * RES_DINODE + 3 * RES_LEAF + RES_RG_BIT, 0); - if (error) - goto out_gunlock; - - error = gfs2_rmdiri(dip, &dentry->d_name, ip); - - gfs2_trans_end(sdp); - -out_gunlock: - gfs2_glock_dq(ghs + 2); -out_rgrp: - gfs2_holder_uninit(ghs + 2); - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_holder_uninit(ghs + 1); - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_mknod - Make a special file - * @dir: The directory in which the special file will reside - * @dentry: The dentry of the special file - * @mode: The mode of the special file - * @rdev: The device specification of the special file - * - */ - -static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, - dev_t dev) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, mode, dev); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/* - * gfs2_ok_to_move - check if it's ok to move a directory to another directory - * @this: move this - * @to: to here - * - * Follow @to back to the root and make sure we don't encounter @this - * Assumes we already hold the rename lock. - * - * Returns: errno - */ - -static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) -{ - struct inode *dir = &to->i_inode; - struct super_block *sb = dir->i_sb; - struct inode *tmp; - int error = 0; - - igrab(dir); - - for (;;) { - if (dir == &this->i_inode) { - error = -EINVAL; - break; - } - if (dir == sb->s_root->d_inode) { - error = 0; - break; - } - - tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1); - if (IS_ERR(tmp)) { - error = PTR_ERR(tmp); - break; - } - - iput(dir); - dir = tmp; - } - - iput(dir); - - return error; -} - -/** - * gfs2_rename - Rename a file - * @odir: Parent directory of old file name - * @odentry: The old dentry of the file - * @ndir: Parent directory of new file name - * @ndentry: The new dentry of the file - * - * Returns: errno - */ - -static int gfs2_rename(struct inode *odir, struct dentry *odentry, - struct inode *ndir, struct dentry *ndentry) -{ - struct gfs2_inode *odip = GFS2_I(odir); - struct gfs2_inode *ndip = GFS2_I(ndir); - struct gfs2_inode *ip = GFS2_I(odentry->d_inode); - struct gfs2_inode *nip = NULL; - struct gfs2_sbd *sdp = GFS2_SB(odir); - struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh; - struct gfs2_rgrpd *nrgd; - unsigned int num_gh; - int dir_rename = 0; - int alloc_required = 0; - unsigned int x; - int error; - - if (ndentry->d_inode) { - nip = GFS2_I(ndentry->d_inode); - if (ip == nip) - return 0; - } - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - - if (odip != ndip) { - error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, - 0, &r_gh); - if (error) - goto out; - - if (S_ISDIR(ip->i_inode.i_mode)) { - dir_rename = 1; - /* don't move a dirctory into it's subdir */ - error = gfs2_ok_to_move(ip, ndip); - if (error) - goto out_gunlock_r; - } - } - - num_gh = 1; - gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - if (odip != ndip) { - gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - } - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - - if (nip) { - gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - /* grab the resource lock for unlink flag twiddling - * this is the case of the target file already existing - * so we unlink before doing the rename - */ - nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr); - if (nrgd) - gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); - } - - for (x = 0; x < num_gh; x++) { - error = gfs2_glock_nq(ghs + x); - if (error) - goto out_gunlock; - } - - /* Check out the old directory */ - - error = gfs2_unlink_ok(odip, &odentry->d_name, ip); - if (error) - goto out_gunlock; - - /* Check out the new directory */ - - if (nip) { - error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip); - if (error) - goto out_gunlock; - - if (S_ISDIR(nip->i_inode.i_mode)) { - if (nip->i_entries < 2) { - if (gfs2_consist_inode(nip)) - gfs2_dinode_print(nip); - error = -EIO; - goto out_gunlock; - } - if (nip->i_entries > 2) { - error = -ENOTEMPTY; - goto out_gunlock; - } - } - } else { - error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_check(ndir, &ndentry->d_name, NULL); - switch (error) { - case -ENOENT: - error = 0; - break; - case 0: - error = -EEXIST; - default: - goto out_gunlock; - }; - - if (odip != ndip) { - if (!ndip->i_inode.i_nlink) { - error = -EINVAL; - goto out_gunlock; - } - if (ndip->i_entries == (u32)-1) { - error = -EFBIG; - goto out_gunlock; - } - if (S_ISDIR(ip->i_inode.i_mode) && - ndip->i_inode.i_nlink == (u32)-1) { - error = -EMLINK; - goto out_gunlock; - } - } - } - - /* Check out the dir to be renamed */ - - if (dir_rename) { - error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0); - if (error) - goto out_gunlock; - } - - if (nip == NULL) - alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name); - error = alloc_required; - if (error < 0) - goto out_gunlock; - error = 0; - - if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(ndip); - if (!al) { - error = -ENOMEM; - goto out_gunlock; - } - - error = gfs2_quota_lock_check(ndip); - if (error) - goto out_alloc; - - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve_ri(ndip); - if (error) - goto out_gunlock_q; - - error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + - gfs2_rg_blocks(al) + - 4 * RES_DINODE + 4 * RES_LEAF + - RES_STATFS + RES_QUOTA + 4, 0); - if (error) - goto out_ipreserv; - } else { - error = gfs2_trans_begin(sdp, 4 * RES_DINODE + - 5 * RES_LEAF + 4, 0); - if (error) - goto out_gunlock; - } - - /* Remove the target file, if it exists */ - - if (nip) { - if (S_ISDIR(nip->i_inode.i_mode)) - error = gfs2_rmdiri(ndip, &ndentry->d_name, nip); - else { - error = gfs2_dir_del(ndip, &ndentry->d_name); - if (error) - goto out_end_trans; - error = gfs2_change_nlink(nip, -1); - } - if (error) - goto out_end_trans; - } - - if (dir_rename) { - error = gfs2_change_nlink(ndip, +1); - if (error) - goto out_end_trans; - error = gfs2_change_nlink(odip, -1); - if (error) - goto out_end_trans; - - error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR); - if (error) - goto out_end_trans; - } else { - struct buffer_head *dibh; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto out_end_trans; - ip->i_inode.i_ctime = CURRENT_TIME; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - } - - error = gfs2_dir_del(odip, &odentry->d_name); - if (error) - goto out_end_trans; - - error = gfs2_dir_add(ndir, &ndentry->d_name, ip, IF2DT(ip->i_inode.i_mode)); - if (error) - goto out_end_trans; - -out_end_trans: - gfs2_trans_end(sdp); -out_ipreserv: - if (alloc_required) - gfs2_inplace_release(ndip); -out_gunlock_q: - if (alloc_required) - gfs2_quota_unlock(ndip); -out_alloc: - if (alloc_required) - gfs2_alloc_put(ndip); -out_gunlock: - while (x--) { - gfs2_glock_dq(ghs + x); - gfs2_holder_uninit(ghs + x); - } -out_gunlock_r: - if (r_gh.gh_gl) - gfs2_glock_dq_uninit(&r_gh); -out: - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_follow_link - Follow a symbolic link - * @dentry: The dentry of the link - * @nd: Data that we pass to vfs_follow_link() - * - * This can handle symlinks of any size. - * - * Returns: 0 on success or error code - */ - -static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd) -{ - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder i_gh; - struct buffer_head *dibh; - unsigned int x, size; - char *buf; - int error; - - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); - error = gfs2_glock_nq(&i_gh); - if (error) { - gfs2_holder_uninit(&i_gh); - nd_set_link(nd, ERR_PTR(error)); - return NULL; - } - - size = (unsigned int)i_size_read(&ip->i_inode); - if (size == 0) { - gfs2_consist_inode(ip); - buf = ERR_PTR(-EIO); - goto out; - } - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) { - buf = ERR_PTR(error); - goto out; - } - - x = size + 1; - buf = kmalloc(x, GFP_NOFS); - if (!buf) - buf = ERR_PTR(-ENOMEM); - else - memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), x); - brelse(dibh); -out: - gfs2_glock_dq_uninit(&i_gh); - nd_set_link(nd, buf); - return NULL; -} - -static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p) -{ - char *s = nd_get_link(nd); - if (!IS_ERR(s)) - kfree(s); -} - -/** - * gfs2_permission - - * @inode: The inode - * @mask: The mask to be tested - * @flags: Indicates whether this is an RCU path walk or not - * - * This may be called from the VFS directly, or from within GFS2 with the - * inode locked, so we look to see if the glock is already locked and only - * lock the glock if its not already been done. - * - * Returns: errno - */ - -int gfs2_permission(struct inode *inode, int mask, unsigned int flags) -{ - struct gfs2_inode *ip; - struct gfs2_holder i_gh; - int error; - int unlock = 0; - - - ip = GFS2_I(inode); - if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); - if (error) - return error; - unlock = 1; - } - - if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) - error = -EACCES; - else - error = generic_permission(inode, mask, flags, gfs2_check_acl); - if (unlock) - gfs2_glock_dq_uninit(&i_gh); - - return error; -} - -static int setattr_chown(struct inode *inode, struct iattr *attr) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); - u32 ouid, ogid, nuid, ngid; - int error; - - ouid = inode->i_uid; - ogid = inode->i_gid; - nuid = attr->ia_uid; - ngid = attr->ia_gid; - - if (!(attr->ia_valid & ATTR_UID) || ouid == nuid) - ouid = nuid = NO_QUOTA_CHANGE; - if (!(attr->ia_valid & ATTR_GID) || ogid == ngid) - ogid = ngid = NO_QUOTA_CHANGE; - - if (!gfs2_alloc_get(ip)) - return -ENOMEM; - - error = gfs2_quota_lock(ip, nuid, ngid); - if (error) - goto out_alloc; - - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { - error = gfs2_quota_check(ip, nuid, ngid); - if (error) - goto out_gunlock_q; - } - - error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0); - if (error) - goto out_gunlock_q; - - error = gfs2_setattr_simple(ip, attr); - if (error) - goto out_end_trans; - - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { - u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); - gfs2_quota_change(ip, -blocks, ouid, ogid); - gfs2_quota_change(ip, blocks, nuid, ngid); - } - -out_end_trans: - gfs2_trans_end(sdp); -out_gunlock_q: - gfs2_quota_unlock(ip); -out_alloc: - gfs2_alloc_put(ip); - return error; -} - -/** - * gfs2_setattr - Change attributes on an inode - * @dentry: The dentry which is changing - * @attr: The structure describing the change - * - * The VFS layer wants to change one or more of an inodes attributes. Write - * that change out to disk. - * - * Returns: errno - */ - -static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder i_gh; - int error; - - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); - if (error) - return error; - - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out; - - error = inode_change_ok(inode, attr); - if (error) - goto out; - - if (attr->ia_valid & ATTR_SIZE) - error = gfs2_setattr_size(inode, attr->ia_size); - else if (attr->ia_valid & (ATTR_UID | ATTR_GID)) - error = setattr_chown(inode, attr); - else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode)) - error = gfs2_acl_chmod(ip, attr); - else - error = gfs2_setattr_simple(ip, attr); - -out: - gfs2_glock_dq_uninit(&i_gh); - if (!error) - mark_inode_dirty(inode); - return error; -} - -/** - * gfs2_getattr - Read out an inode's attributes - * @mnt: The vfsmount the inode is being accessed from - * @dentry: The dentry to stat - * @stat: The inode's stats - * - * This may be called from the VFS directly, or from within GFS2 with the - * inode locked, so we look to see if the glock is already locked and only - * lock the glock if its not already been done. Note that its the NFS - * readdirplus operation which causes this to be called (from filldir) - * with the glock already held. - * - * Returns: errno - */ - -static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, - struct kstat *stat) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int error; - int unlock = 0; - - if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (error) - return error; - unlock = 1; - } - - generic_fillattr(inode, stat); - if (unlock) - gfs2_glock_dq_uninit(&gh); - - return 0; -} - -static int gfs2_setxattr(struct dentry *dentry, const char *name, - const void *data, size_t size, int flags) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_setxattr(dentry, name, data, size, flags); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name, - void *data, size_t size) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_getxattr(dentry, name, data, size); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static int gfs2_removexattr(struct dentry *dentry, const char *name) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_removexattr(dentry, name); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, - u64 start, u64 len) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC); - if (ret) - return ret; - - mutex_lock(&inode->i_mutex); - - ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh); - if (ret) - goto out; - - if (gfs2_is_stuffed(ip)) { - u64 phys = ip->i_no_addr << inode->i_blkbits; - u64 size = i_size_read(inode); - u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED| - FIEMAP_EXTENT_DATA_INLINE; - phys += sizeof(struct gfs2_dinode); - phys += start; - if (start + len > size) - len = size - start; - if (start < size) - ret = fiemap_fill_next_extent(fieinfo, start, phys, - len, flags); - if (ret == 1) - ret = 0; - } else { - ret = __generic_block_fiemap(inode, fieinfo, start, len, - gfs2_block_map); - } - - gfs2_glock_dq_uninit(&gh); -out: - mutex_unlock(&inode->i_mutex); - return ret; -} - -const struct inode_operations gfs2_file_iops = { - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - -const struct inode_operations gfs2_dir_iops = { - .create = gfs2_create, - .lookup = gfs2_lookup, - .link = gfs2_link, - .unlink = gfs2_unlink, - .symlink = gfs2_symlink, - .mkdir = gfs2_mkdir, - .rmdir = gfs2_rmdir, - .mknod = gfs2_mknod, - .rename = gfs2_rename, - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - -const struct inode_operations gfs2_symlink_iops = { - .readlink = generic_readlink, - .follow_link = gfs2_follow_link, - .put_link = gfs2_put_link, - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index e23d986..42e8d23 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -38,6 +38,7 @@ #include <linux/sched.h> #include <linux/slab.h> +#include <linux/mm.h> #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/buffer_head.h> @@ -77,19 +78,20 @@ static LIST_HEAD(qd_lru_list); static atomic_t qd_lru_count = ATOMIC_INIT(0); static DEFINE_SPINLOCK(qd_lru_lock); -int gfs2_shrink_qd_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) +int gfs2_shrink_qd_memory(struct shrinker *shrink, struct shrink_control *sc) { struct gfs2_quota_data *qd; struct gfs2_sbd *sdp; + int nr_to_scan = sc->nr_to_scan; - if (nr == 0) + if (nr_to_scan == 0) goto out; - if (!(gfp_mask & __GFP_FS)) + if (!(sc->gfp_mask & __GFP_FS)) return -1; spin_lock(&qd_lru_lock); - while (nr && !list_empty(&qd_lru_list)) { + while (nr_to_scan && !list_empty(&qd_lru_list)) { qd = list_entry(qd_lru_list.next, struct gfs2_quota_data, qd_reclaim); sdp = qd->qd_gl->gl_sbd; @@ -110,7 +112,7 @@ int gfs2_shrink_qd_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) spin_unlock(&qd_lru_lock); kmem_cache_free(gfs2_quotad_cachep, qd); spin_lock(&qd_lru_lock); - nr--; + nr_to_scan--; } spin_unlock(&qd_lru_lock); diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h index e7d236c..90bf1c3 100644 --- a/fs/gfs2/quota.h +++ b/fs/gfs2/quota.h @@ -12,6 +12,7 @@ struct gfs2_inode; struct gfs2_sbd; +struct shrink_control; #define NO_QUOTA_CHANGE ((u32)-1) @@ -51,7 +52,8 @@ static inline int gfs2_quota_lock_check(struct gfs2_inode *ip) return ret; } -extern int gfs2_shrink_qd_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask); +extern int gfs2_shrink_qd_memory(struct shrinker *shrink, + struct shrink_control *sc); extern const struct quotactl_ops gfs2_quotactl_ops; #endif /* __QUOTA_DOT_H__ */ diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 6fcae84..9b780df 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -78,10 +78,11 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, unsigned char *buf2, unsigned int offset, - unsigned int buflen, u32 block, + struct gfs2_bitmap *bi, u32 block, unsigned char new_state) { unsigned char *byte1, *byte2, *end, cur_state; + unsigned int buflen = bi->bi_len; const unsigned int bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE; byte1 = buf1 + offset + (block / GFS2_NBBY); @@ -92,6 +93,16 @@ static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, cur_state = (*byte1 >> bit) & GFS2_BIT_MASK; if (unlikely(!valid_change[new_state * 4 + cur_state])) { + printk(KERN_WARNING "GFS2: buf_blk = 0x%llx old_state=%d, " + "new_state=%d\n", + (unsigned long long)block, cur_state, new_state); + printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%lx\n", + (unsigned long long)rgd->rd_addr, + (unsigned long)bi->bi_start); + printk(KERN_WARNING "GFS2: bi_offset=0x%lx bi_len=0x%lx\n", + (unsigned long)bi->bi_offset, + (unsigned long)bi->bi_len); + dump_stack(); gfs2_consist_rgrpd(rgd); return; } @@ -381,6 +392,7 @@ static void clear_rgrpdi(struct gfs2_sbd *sdp) if (gl) { gl->gl_object = NULL; + gfs2_glock_add_to_lru(gl); gfs2_glock_put(gl); } @@ -1365,7 +1377,7 @@ skip: gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi->bi_len, blk, new_state); + bi, blk, new_state); goal = blk; while (*n < elen) { goal++; @@ -1375,7 +1387,7 @@ skip: GFS2_BLKST_FREE) break; gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi->bi_len, goal, new_state); + bi, goal, new_state); (*n)++; } out: @@ -1432,7 +1444,7 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, } gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); gfs2_setbit(rgd, bi->bi_bh->b_data, NULL, bi->bi_offset, - bi->bi_len, buf_blk, new_state); + bi, buf_blk, new_state); } return rgd; @@ -1617,6 +1629,10 @@ void __gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen) gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); gfs2_trans_add_rg(rgd); + + /* Directories keep their data in the metadata address space */ + if (ip->i_depth) + gfs2_meta_wipe(ip, bstart, blen); } /** diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index b9f28e6..ed540e7 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -23,6 +23,7 @@ #include <linux/time.h> #include <linux/wait.h> #include <linux/writeback.h> +#include <linux/backing-dev.h> #include "gfs2.h" #include "incore.h" @@ -700,11 +701,47 @@ void gfs2_unfreeze_fs(struct gfs2_sbd *sdp) mutex_unlock(&sdp->sd_freeze_lock); } +void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) +{ + struct gfs2_dinode *str = buf; + + str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); + str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); + str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); + str->di_num.no_addr = cpu_to_be64(ip->i_no_addr); + str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino); + str->di_mode = cpu_to_be32(ip->i_inode.i_mode); + str->di_uid = cpu_to_be32(ip->i_inode.i_uid); + str->di_gid = cpu_to_be32(ip->i_inode.i_gid); + str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); + str->di_size = cpu_to_be64(i_size_read(&ip->i_inode)); + str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode)); + str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); + str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); + str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); + + str->di_goal_meta = cpu_to_be64(ip->i_goal); + str->di_goal_data = cpu_to_be64(ip->i_goal); + str->di_generation = cpu_to_be64(ip->i_generation); + + str->di_flags = cpu_to_be32(ip->i_diskflags); + str->di_height = cpu_to_be16(ip->i_height); + str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && + !(ip->i_diskflags & GFS2_DIF_EXHASH) ? + GFS2_FORMAT_DE : 0); + str->di_depth = cpu_to_be16(ip->i_depth); + str->di_entries = cpu_to_be32(ip->i_entries); + + str->di_eattr = cpu_to_be64(ip->i_eattr); + str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); + str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec); + str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec); +} /** * gfs2_write_inode - Make sure the inode is stable on the disk * @inode: The inode - * @sync: synchronous write flag + * @wbc: The writeback control structure * * Returns: errno */ @@ -713,15 +750,17 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); + struct address_space *metamapping = gfs2_glock2aspace(ip->i_gl); + struct backing_dev_info *bdi = metamapping->backing_dev_info; struct gfs2_holder gh; struct buffer_head *bh; struct timespec atime; struct gfs2_dinode *di; - int ret = 0; + int ret = -EAGAIN; - /* Check this is a "normal" inode, etc */ + /* Skip timestamp update, if this is from a memalloc */ if (current->flags & PF_MEMALLOC) - return 0; + goto do_flush; ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); if (ret) goto do_flush; @@ -745,6 +784,13 @@ do_unlock: do_flush: if (wbc->sync_mode == WB_SYNC_ALL) gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + filemap_fdatawrite(metamapping); + if (bdi->dirty_exceeded) + gfs2_ail1_flush(sdp, wbc); + if (!ret && (wbc->sync_mode == WB_SYNC_ALL)) + ret = filemap_fdatawait(metamapping); + if (ret) + mark_inode_dirty_sync(inode); return ret; } @@ -874,8 +920,9 @@ restart: static int gfs2_sync_fs(struct super_block *sb, int wait) { - if (wait && sb->s_fs_info) - gfs2_log_flush(sb->s_fs_info, NULL); + struct gfs2_sbd *sdp = sb->s_fs_info; + if (wait && sdp) + gfs2_log_flush(sdp, NULL); return 0; } @@ -1308,6 +1355,78 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) return 0; } +static void gfs2_final_release_pages(struct gfs2_inode *ip) +{ + struct inode *inode = &ip->i_inode; + struct gfs2_glock *gl = ip->i_gl; + + truncate_inode_pages(gfs2_glock2aspace(ip->i_gl), 0); + truncate_inode_pages(&inode->i_data, 0); + + if (atomic_read(&gl->gl_revokes) == 0) { + clear_bit(GLF_LFLUSH, &gl->gl_flags); + clear_bit(GLF_DIRTY, &gl->gl_flags); + } +} + +static int gfs2_dinode_dealloc(struct gfs2_inode *ip) +{ + struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + struct gfs2_alloc *al; + struct gfs2_rgrpd *rgd; + int error; + + if (gfs2_get_inode_blocks(&ip->i_inode) != 1) { + gfs2_consist_inode(ip); + return -EIO; + } + + al = gfs2_alloc_get(ip); + if (!al) + return -ENOMEM; + + error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); + if (error) + goto out; + + error = gfs2_rindex_hold(sdp, &al->al_ri_gh); + if (error) + goto out_qs; + + rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); + if (!rgd) { + gfs2_consist_inode(ip); + error = -EIO; + goto out_rindex_relse; + } + + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, + &al->al_rgd_gh); + if (error) + goto out_rindex_relse; + + error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA, + sdp->sd_jdesc->jd_blocks); + if (error) + goto out_rg_gunlock; + + gfs2_free_di(rgd, ip); + + gfs2_final_release_pages(ip); + + gfs2_trans_end(sdp); + +out_rg_gunlock: + gfs2_glock_dq_uninit(&al->al_rgd_gh); +out_rindex_relse: + gfs2_glock_dq_uninit(&al->al_ri_gh); +out_qs: + gfs2_quota_unhold(ip); +out: + gfs2_alloc_put(ip); + return error; +} + /* * We have to (at the moment) hold the inodes main lock to cover * the gap between unlocking the shared lock on the iopen lock and @@ -1371,15 +1490,13 @@ static void gfs2_evict_inode(struct inode *inode) } error = gfs2_dinode_dealloc(ip); - if (error) - goto out_unlock; + goto out_unlock; out_truncate: error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks); if (error) goto out_unlock; - /* Needs to be done before glock release & also in a transaction */ - truncate_inode_pages(&inode->i_data, 0); + gfs2_final_release_pages(ip); gfs2_trans_end(sdp); out_unlock: @@ -1394,6 +1511,7 @@ out: end_writeback(inode); ip->i_gl->gl_object = NULL; + gfs2_glock_add_to_lru(ip->i_gl); gfs2_glock_put(ip->i_gl); ip->i_gl = NULL; if (ip->i_iopen_gh.gh_gl) { diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 748ccb5..e20eab3 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -81,7 +81,8 @@ static int gfs2_uuid_valid(const u8 *uuid) static ssize_t uuid_show(struct gfs2_sbd *sdp, char *buf) { - const u8 *uuid = sdp->sd_sb.sb_uuid; + struct super_block *s = sdp->sd_vfs; + const u8 *uuid = s->s_uuid; buf[0] = '\0'; if (!gfs2_uuid_valid(uuid)) return 0; @@ -616,7 +617,8 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj, struct kobj_uevent_env *env) { struct gfs2_sbd *sdp = container_of(kobj, struct gfs2_sbd, sd_kobj); - const u8 *uuid = sdp->sd_sb.sb_uuid; + struct super_block *s = sdp->sd_vfs; + const u8 *uuid = s->s_uuid; add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name); add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name); diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h index cedb0bb..5d07609 100644 --- a/fs/gfs2/trace_gfs2.h +++ b/fs/gfs2/trace_gfs2.h @@ -10,6 +10,7 @@ #include <linux/buffer_head.h> #include <linux/dlmconstants.h> #include <linux/gfs2_ondisk.h> +#include <linux/writeback.h> #include "incore.h" #include "glock.h" @@ -40,7 +41,9 @@ {(1UL << GLF_REPLY_PENDING), "r" }, \ {(1UL << GLF_INITIAL), "I" }, \ {(1UL << GLF_FROZEN), "F" }, \ - {(1UL << GLF_QUEUED), "q" }) + {(1UL << GLF_QUEUED), "q" }, \ + {(1UL << GLF_LRU), "L" }, \ + {(1UL << GLF_OBJECT), "o" }) #ifndef NUMPTY #define NUMPTY @@ -94,7 +97,7 @@ TRACE_EVENT(gfs2_glock_state_change, __entry->new_state = glock_trace_state(new_state); __entry->tgt_state = glock_trace_state(gl->gl_target); __entry->dmt_state = glock_trace_state(gl->gl_demote_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld state %s to %s tgt:%s dmt:%s flags:%s", @@ -127,7 +130,7 @@ TRACE_EVENT(gfs2_glock_put, __entry->gltype = gl->gl_name.ln_type; __entry->glnum = gl->gl_name.ln_number; __entry->cur_state = glock_trace_state(gl->gl_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld state %s => %s flags:%s", @@ -161,7 +164,7 @@ TRACE_EVENT(gfs2_demote_rq, __entry->glnum = gl->gl_name.ln_number; __entry->cur_state = glock_trace_state(gl->gl_state); __entry->dmt_state = glock_trace_state(gl->gl_demote_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld demote %s to %s flags:%s", @@ -318,6 +321,33 @@ TRACE_EVENT(gfs2_log_blocks, MINOR(__entry->dev), __entry->blocks) ); +/* Writing back the AIL */ +TRACE_EVENT(gfs2_ail_flush, + + TP_PROTO(const struct gfs2_sbd *sdp, const struct writeback_control *wbc, int start), + + TP_ARGS(sdp, wbc, start), + + TP_STRUCT__entry( + __field( dev_t, dev ) + __field( int, start ) + __field( int, sync_mode ) + __field( long, nr_to_write ) + ), + + TP_fast_assign( + __entry->dev = sdp->sd_vfs->s_dev; + __entry->start = start; + __entry->sync_mode = wbc->sync_mode; + __entry->nr_to_write = wbc->nr_to_write; + ), + + TP_printk("%u,%u ail flush %s %s %ld", MAJOR(__entry->dev), + MINOR(__entry->dev), __entry->start ? "start" : "end", + __entry->sync_mode == WB_SYNC_ALL ? "all" : "none", + __entry->nr_to_write) +); + /* Section 3 - bmap * * Objectives: diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index 1f05839..acf95da 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -395,7 +395,6 @@ again: dentry_unhash(dentry); if (!d_unhashed(dentry)) { - dput(dentry); hpfs_unlock(dir->i_sb); return -ENOSPC; } @@ -403,7 +402,6 @@ again: !S_ISREG(inode->i_mode) || get_write_access(inode)) { d_rehash(dentry); - dput(dentry); } else { struct iattr newattrs; /*printk("HPFS: truncating file before delete.\n");*/ @@ -411,7 +409,6 @@ again: newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; err = notify_change(dentry, &newattrs); put_write_access(inode); - dput(dentry); if (!err) goto again; } @@ -535,6 +532,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct buffer_head *bh; struct fnode *fnode; int err; + if ((err = hpfs_chk_name(new_name, &new_len))) return err; err = 0; hpfs_adjust_length(old_name, &old_len); diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index b9eeb1c..7aafeb8 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -412,10 +412,10 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) pgoff = offset >> PAGE_SHIFT; i_size_write(inode, offset); - spin_lock(&mapping->i_mmap_lock); + mutex_lock(&mapping->i_mmap_mutex); if (!prio_tree_empty(&mapping->i_mmap)) hugetlb_vmtruncate_list(&mapping->i_mmap, pgoff); - spin_unlock(&mapping->i_mmap_lock); + mutex_unlock(&mapping->i_mmap_mutex); truncate_hugepages(inode, offset); return 0; } @@ -921,7 +921,8 @@ static int can_do_hugetlb_shm(void) return capable(CAP_IPC_LOCK) || in_group_p(sysctl_hugetlb_shm_group); } -struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag, +struct file *hugetlb_file_setup(const char *name, size_t size, + vm_flags_t acctflag, struct user_struct **user, int creat_flags) { int error = -ENOMEM; @@ -1,9 +1,7 @@ /* - * linux/fs/inode.c - * * (C) 1997 Linus Torvalds + * (C) 1999 Andrea Arcangeli <andrea@suse.de> (dynamic inode allocation) */ - #include <linux/fs.h> #include <linux/mm.h> #include <linux/dcache.h> @@ -24,12 +22,14 @@ #include <linux/mount.h> #include <linux/async.h> #include <linux/posix_acl.h> +#include <linux/prefetch.h> #include <linux/ima.h> #include <linux/cred.h> +#include <linux/buffer_head.h> /* for inode_has_buffers */ #include "internal.h" /* - * inode locking rules. + * Inode locking rules: * * inode->i_lock protects: * inode->i_state, inode->i_hash, __iget() @@ -59,54 +59,11 @@ * inode_hash_lock */ -/* - * This is needed for the following functions: - * - inode_has_buffers - * - invalidate_bdev - * - * FIXME: remove all knowledge of the buffer layer from this file - */ -#include <linux/buffer_head.h> - -/* - * New inode.c implementation. - * - * This implementation has the basic premise of trying - * to be extremely low-overhead and SMP-safe, yet be - * simple enough to be "obviously correct". - * - * Famous last words. - */ - -/* inode dynamic allocation 1999, Andrea Arcangeli <andrea@suse.de> */ - -/* #define INODE_PARANOIA 1 */ -/* #define INODE_DEBUG 1 */ - -/* - * Inode lookup is no longer as critical as it used to be: - * most of the lookups are going to be through the dcache. - */ -#define I_HASHBITS i_hash_shift -#define I_HASHMASK i_hash_mask - static unsigned int i_hash_mask __read_mostly; static unsigned int i_hash_shift __read_mostly; static struct hlist_head *inode_hashtable __read_mostly; static __cacheline_aligned_in_smp DEFINE_SPINLOCK(inode_hash_lock); -/* - * Each inode can be on two separate lists. One is - * the hash list of the inode, used for lookups. The - * other linked list is the "type" list: - * "in_use" - valid inode, i_count > 0, i_nlink > 0 - * "dirty" - as "in_use" but also dirty - * "unused" - valid inode, i_count = 0 - * - * A "dirty" list is maintained for each super block, - * allowing for low-overhead inode sync() operations. - */ - static LIST_HEAD(inode_lru); static DEFINE_SPINLOCK(inode_lru_lock); @@ -325,12 +282,11 @@ void address_space_init_once(struct address_space *mapping) memset(mapping, 0, sizeof(*mapping)); INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC); spin_lock_init(&mapping->tree_lock); - spin_lock_init(&mapping->i_mmap_lock); + mutex_init(&mapping->i_mmap_mutex); INIT_LIST_HEAD(&mapping->private_list); spin_lock_init(&mapping->private_lock); INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap); INIT_LIST_HEAD(&mapping->i_mmap_nonlinear); - mutex_init(&mapping->unmap_mutex); } EXPORT_SYMBOL(address_space_init_once); @@ -424,8 +380,8 @@ static unsigned long hash(struct super_block *sb, unsigned long hashval) tmp = (hashval * (unsigned long)sb) ^ (GOLDEN_RATIO_PRIME + hashval) / L1_CACHE_BYTES; - tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> I_HASHBITS); - return tmp & I_HASHMASK; + tmp = tmp ^ ((tmp ^ GOLDEN_RATIO_PRIME) >> i_hash_shift); + return tmp & i_hash_mask; } /** @@ -751,8 +707,12 @@ static void prune_icache(int nr_to_scan) * This function is passed the number of inodes to scan, and it returns the * total number of remaining possibly-reclaimable inodes. */ -static int shrink_icache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) +static int shrink_icache_memory(struct shrinker *shrink, + struct shrink_control *sc) { + int nr = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; + if (nr) { /* * Nasty deadlock avoidance. We may hold various FS locks, diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 3db5ba4..b3cc858 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -974,7 +974,7 @@ out_no_inode: out_no_read: printk(KERN_WARNING "%s: bread failed, dev=%s, iso_blknum=%d, block=%d\n", __func__, s->s_id, iso_blknum, block); - goto out_freesbi; + goto out_freebh; out_bad_zone_size: printk(KERN_WARNING "ISOFS: Bad logical zone size %ld\n", sbi->s_log_zone_size); @@ -989,6 +989,7 @@ out_unknown_format: out_freebh: brelse(bh); + brelse(pri_bh); out_freesbi: kfree(opt.iocharset); kfree(sbi); diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 69b1804..72ffa97 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -302,12 +302,6 @@ void journal_commit_transaction(journal_t *journal) * all outstanding updates to complete. */ -#ifdef COMMIT_STATS - spin_lock(&journal->j_list_lock); - summarise_journal_usage(journal); - spin_unlock(&journal->j_list_lock); -#endif - /* Do we need to erase the effects of a prior journal_flush? */ if (journal->j_flags & JFS_FLUSHED) { jbd_debug(3, "super block updated\n"); @@ -722,8 +716,13 @@ wait_for_iobuf: required. */ JBUFFER_TRACE(jh, "file as BJ_Forget"); journal_file_buffer(jh, commit_transaction, BJ_Forget); - /* Wake up any transactions which were waiting for this - IO to complete */ + /* + * Wake up any transactions which were waiting for this + * IO to complete. The barrier must be here so that changes + * by journal_file_buffer() take effect before wake_up_bit() + * does the waitqueue check. + */ + smp_mb(); wake_up_bit(&bh->b_state, BH_Unshadow); JBUFFER_TRACE(jh, "brelse shadowed buffer"); __brelse(bh); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index b3713af..e2d4285 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -437,9 +437,12 @@ int __log_space_left(journal_t *journal) int __log_start_commit(journal_t *journal, tid_t target) { /* - * Are we already doing a recent enough commit? + * The only transaction we can possibly wait upon is the + * currently running transaction (if it exists). Otherwise, + * the target tid must be an old one. */ - if (!tid_geq(journal->j_commit_request, target)) { + if (journal->j_running_transaction && + journal->j_running_transaction->t_tid == target) { /* * We want a new commit: OK, mark the request and wakeup the * commit thread. We do _not_ do the commit ourselves. @@ -451,7 +454,14 @@ int __log_start_commit(journal_t *journal, tid_t target) journal->j_commit_sequence); wake_up(&journal->j_wait_commit); return 1; - } + } else if (!tid_geq(journal->j_commit_request, target)) + /* This should never happen, but if it does, preserve + the evidence before kjournald goes into a loop and + increments j_commit_sequence beyond all recognition. */ + WARN_ONCE(1, "jbd: bad log_start_commit: %u %u %u %u\n", + journal->j_commit_request, journal->j_commit_sequence, + target, journal->j_running_transaction ? + journal->j_running_transaction->t_tid : 0); return 0; } diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 60d2319..f7ee81a 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -266,7 +266,8 @@ static handle_t *new_handle(int nblocks) * This function is visible to journal users (like ext3fs), so is not * called with the journal already locked. * - * Return a pointer to a newly allocated handle, or NULL on failure + * Return a pointer to a newly allocated handle, or an ERR_PTR() value + * on failure. */ handle_t *journal_start(journal_t *journal, int nblocks) { diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 6a79fd0..2c62c5a 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -97,10 +97,14 @@ static int __try_to_free_cp_buf(struct journal_head *jh) if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh) && !buffer_write_io_error(bh)) { + /* + * Get our reference so that bh cannot be freed before + * we unlock it + */ + get_bh(bh); JBUFFER_TRACE(jh, "remove from checkpoint list"); ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); BUFFER_TRACE(bh, "release"); __brelse(bh); } else { @@ -223,8 +227,8 @@ restart: spin_lock(&journal->j_list_lock); goto restart; } + get_bh(bh); if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -243,7 +247,6 @@ restart: */ released = __jbd2_journal_remove_checkpoint(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } @@ -284,7 +287,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, int ret = 0; if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); + get_bh(bh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -316,12 +319,12 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, ret = 1; if (unlikely(buffer_write_io_error(bh))) ret = -EIO; + get_bh(bh); J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); __jbd2_journal_remove_checkpoint(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } else { /* @@ -554,7 +557,8 @@ int jbd2_cleanup_journal_tail(journal_t *journal) /* * journal_clean_one_cp_list * - * Find all the written-back checkpoint buffers in the given list and release them. + * Find all the written-back checkpoint buffers in the given list and + * release them. * * Called with the journal locked. * Called with j_list_lock held. @@ -663,8 +667,8 @@ out: * checkpoint lists. * * The function returns 1 if it frees the transaction, 0 otherwise. + * The function can free jh and bh. * - * This function is called with the journal locked. * This function is called with j_list_lock held. * This function is called with jbd_lock_bh_state(jh2bh(jh)) */ @@ -684,13 +688,14 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) } journal = transaction->t_journal; + JBUFFER_TRACE(jh, "removing from transaction"); __buffer_unlink(jh); jh->b_cp_transaction = NULL; + jbd2_journal_put_journal_head(jh); if (transaction->t_checkpoint_list != NULL || transaction->t_checkpoint_io_list != NULL) goto out; - JBUFFER_TRACE(jh, "transaction has no more buffers"); /* * There is one special case to worry about: if we have just pulled the @@ -701,10 +706,8 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) * The locking here around t_state is a bit sleazy. * See the comment at the end of jbd2_journal_commit_transaction(). */ - if (transaction->t_state != T_FINISHED) { - JBUFFER_TRACE(jh, "belongs to running/committing transaction"); + if (transaction->t_state != T_FINISHED) goto out; - } /* OK, that was the last buffer for the transaction: we can now safely remove this transaction from the log */ @@ -723,7 +726,6 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) wake_up(&journal->j_wait_logspace); ret = 1; out: - JBUFFER_TRACE(jh, "exit"); return ret; } @@ -742,6 +744,8 @@ void __jbd2_journal_insert_checkpoint(struct journal_head *jh, J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh))); J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + /* Get reference for checkpointing transaction */ + jbd2_journal_grab_journal_head(jh2bh(jh)); jh->b_cp_transaction = transaction; if (!transaction->t_checkpoint_list) { diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6e28000..eef6979 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -219,7 +219,6 @@ static int journal_submit_data_buffers(journal_t *journal, ret = err; spin_lock(&journal->j_list_lock); J_ASSERT(jinode->i_transaction == commit_transaction); - commit_transaction->t_flushed_data_blocks = 1; clear_bit(__JI_COMMIT_RUNNING, &jinode->i_flags); smp_mb__after_clear_bit(); wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); @@ -338,12 +337,6 @@ void jbd2_journal_commit_transaction(journal_t *journal) * all outstanding updates to complete. */ -#ifdef COMMIT_STATS - spin_lock(&journal->j_list_lock); - summarise_journal_usage(journal); - spin_unlock(&journal->j_list_lock); -#endif - /* Do we need to erase the effects of a prior jbd2_journal_flush? */ if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); @@ -678,12 +671,16 @@ start_journal_io: err = 0; } + write_lock(&journal->j_state_lock); + J_ASSERT(commit_transaction->t_state == T_COMMIT); + commit_transaction->t_state = T_COMMIT_DFLUSH; + write_unlock(&journal->j_state_lock); /* * If the journal is not located on the file system device, * then we must flush the file system device before we issue * the commit record */ - if (commit_transaction->t_flushed_data_blocks && + if (commit_transaction->t_need_data_flush && (journal->j_fs_dev != journal->j_dev) && (journal->j_flags & JBD2_BARRIER)) blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); @@ -760,8 +757,13 @@ wait_for_iobuf: required. */ JBUFFER_TRACE(jh, "file as BJ_Forget"); jbd2_journal_file_buffer(jh, commit_transaction, BJ_Forget); - /* Wake up any transactions which were waiting for this - IO to complete */ + /* + * Wake up any transactions which were waiting for this IO to + * complete. The barrier must be here so that changes by + * jbd2_journal_file_buffer() take effect before wake_up_bit() + * does the waitqueue check. + */ + smp_mb(); wake_up_bit(&bh->b_state, BH_Unshadow); JBUFFER_TRACE(jh, "brelse shadowed buffer"); __brelse(bh); @@ -800,6 +802,10 @@ wait_for_iobuf: jbd2_journal_abort(journal, err); jbd_debug(3, "JBD: commit phase 5\n"); + write_lock(&journal->j_state_lock); + J_ASSERT(commit_transaction->t_state == T_COMMIT_DFLUSH); + commit_transaction->t_state = T_COMMIT_JFLUSH; + write_unlock(&journal->j_state_lock); if (!JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { @@ -842,10 +848,16 @@ restart_loop: while (commit_transaction->t_forget) { transaction_t *cp_transaction; struct buffer_head *bh; + int try_to_free = 0; jh = commit_transaction->t_forget; spin_unlock(&journal->j_list_lock); bh = jh2bh(jh); + /* + * Get a reference so that bh cannot be freed before we are + * done with it. + */ + get_bh(bh); jbd_lock_bh_state(bh); J_ASSERT_JH(jh, jh->b_transaction == commit_transaction); @@ -908,28 +920,27 @@ restart_loop: __jbd2_journal_insert_checkpoint(jh, commit_transaction); if (is_journal_aborted(journal)) clear_buffer_jbddirty(bh); - JBUFFER_TRACE(jh, "refile for checkpoint writeback"); - __jbd2_journal_refile_buffer(jh); - jbd_unlock_bh_state(bh); } else { J_ASSERT_BH(bh, !buffer_dirty(bh)); - /* The buffer on BJ_Forget list and not jbddirty means + /* + * The buffer on BJ_Forget list and not jbddirty means * it has been freed by this transaction and hence it * could not have been reallocated until this * transaction has committed. *BUT* it could be * reallocated once we have written all the data to * disk and before we process the buffer on BJ_Forget - * list. */ - JBUFFER_TRACE(jh, "refile or unfile freed buffer"); - __jbd2_journal_refile_buffer(jh); - if (!jh->b_transaction) { - jbd_unlock_bh_state(bh); - /* needs a brelse */ - jbd2_journal_remove_journal_head(bh); - release_buffer_page(bh); - } else - jbd_unlock_bh_state(bh); + * list. + */ + if (!jh->b_next_transaction) + try_to_free = 1; } + JBUFFER_TRACE(jh, "refile or unfile buffer"); + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + if (try_to_free) + release_buffer_page(bh); /* Drops bh reference */ + else + __brelse(bh); cond_resched_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); @@ -955,7 +966,7 @@ restart_loop: jbd_debug(3, "JBD: commit phase 7\n"); - J_ASSERT(commit_transaction->t_state == T_COMMIT); + J_ASSERT(commit_transaction->t_state == T_COMMIT_JFLUSH); commit_transaction->t_start = jiffies; stats.run.rs_logging = jbd2_time_diff(stats.run.rs_logging, diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index e0ec3db..0dfa5b59 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -479,9 +479,12 @@ int __jbd2_log_space_left(journal_t *journal) int __jbd2_log_start_commit(journal_t *journal, tid_t target) { /* - * Are we already doing a recent enough commit? + * The only transaction we can possibly wait upon is the + * currently running transaction (if it exists). Otherwise, + * the target tid must be an old one. */ - if (!tid_geq(journal->j_commit_request, target)) { + if (journal->j_running_transaction && + journal->j_running_transaction->t_tid == target) { /* * We want a new commit: OK, mark the request and wakeup the * commit thread. We do _not_ do the commit ourselves. @@ -493,7 +496,15 @@ int __jbd2_log_start_commit(journal_t *journal, tid_t target) journal->j_commit_sequence); wake_up(&journal->j_wait_commit); return 1; - } + } else if (!tid_geq(journal->j_commit_request, target)) + /* This should never happen, but if it does, preserve + the evidence before kjournald goes into a loop and + increments j_commit_sequence beyond all recognition. */ + WARN_ONCE(1, "jbd: bad log_start_commit: %u %u %u %u\n", + journal->j_commit_request, + journal->j_commit_sequence, + target, journal->j_running_transaction ? + journal->j_running_transaction->t_tid : 0); return 0; } @@ -577,6 +588,47 @@ int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid) } /* + * Return 1 if a given transaction has not yet sent barrier request + * connected with a transaction commit. If 0 is returned, transaction + * may or may not have sent the barrier. Used to avoid sending barrier + * twice in common cases. + */ +int jbd2_trans_will_send_data_barrier(journal_t *journal, tid_t tid) +{ + int ret = 0; + transaction_t *commit_trans; + + if (!(journal->j_flags & JBD2_BARRIER)) + return 0; + read_lock(&journal->j_state_lock); + /* Transaction already committed? */ + if (tid_geq(journal->j_commit_sequence, tid)) + goto out; + commit_trans = journal->j_committing_transaction; + if (!commit_trans || commit_trans->t_tid != tid) { + ret = 1; + goto out; + } + /* + * Transaction is being committed and we already proceeded to + * submitting a flush to fs partition? + */ + if (journal->j_fs_dev != journal->j_dev) { + if (!commit_trans->t_need_data_flush || + commit_trans->t_state >= T_COMMIT_DFLUSH) + goto out; + } else { + if (commit_trans->t_state >= T_COMMIT_JFLUSH) + goto out; + } + ret = 1; +out: + read_unlock(&journal->j_state_lock); + return ret; +} +EXPORT_SYMBOL(jbd2_trans_will_send_data_barrier); + +/* * Wait for a specified commit to complete. * The caller may not hold the journal lock. */ @@ -2026,10 +2078,9 @@ static void journal_free_journal_head(struct journal_head *jh) * When a buffer has its BH_JBD bit set it is immune from being released by * core kernel code, mainly via ->b_count. * - * A journal_head may be detached from its buffer_head when the journal_head's - * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. - * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the - * journal_head can be dropped if needed. + * A journal_head is detached from its buffer_head when the journal_head's + * b_jcount reaches zero. Running transaction (b_transaction) and checkpoint + * transaction (b_cp_transaction) hold their references to b_jcount. * * Various places in the kernel want to attach a journal_head to a buffer_head * _before_ attaching the journal_head to a transaction. To protect the @@ -2042,17 +2093,16 @@ static void journal_free_journal_head(struct journal_head *jh) * (Attach a journal_head if needed. Increments b_jcount) * struct journal_head *jh = jbd2_journal_add_journal_head(bh); * ... + * (Get another reference for transaction) + * jbd2_journal_grab_journal_head(bh); * jh->b_transaction = xxx; + * (Put original reference) * jbd2_journal_put_journal_head(jh); - * - * Now, the journal_head's b_jcount is zero, but it is safe from being released - * because it has a non-zero b_transaction. */ /* * Give a buffer_head a journal_head. * - * Doesn't need the journal lock. * May sleep. */ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) @@ -2116,61 +2166,29 @@ static void __journal_remove_journal_head(struct buffer_head *bh) struct journal_head *jh = bh2jh(bh); J_ASSERT_JH(jh, jh->b_jcount >= 0); - - get_bh(bh); - if (jh->b_jcount == 0) { - if (jh->b_transaction == NULL && - jh->b_next_transaction == NULL && - jh->b_cp_transaction == NULL) { - J_ASSERT_JH(jh, jh->b_jlist == BJ_None); - J_ASSERT_BH(bh, buffer_jbd(bh)); - J_ASSERT_BH(bh, jh2bh(jh) == bh); - BUFFER_TRACE(bh, "remove journal_head"); - if (jh->b_frozen_data) { - printk(KERN_WARNING "%s: freeing " - "b_frozen_data\n", - __func__); - jbd2_free(jh->b_frozen_data, bh->b_size); - } - if (jh->b_committed_data) { - printk(KERN_WARNING "%s: freeing " - "b_committed_data\n", - __func__); - jbd2_free(jh->b_committed_data, bh->b_size); - } - bh->b_private = NULL; - jh->b_bh = NULL; /* debug, really */ - clear_buffer_jbd(bh); - __brelse(bh); - journal_free_journal_head(jh); - } else { - BUFFER_TRACE(bh, "journal_head was locked"); - } + J_ASSERT_JH(jh, jh->b_transaction == NULL); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + J_ASSERT_JH(jh, jh->b_jlist == BJ_None); + J_ASSERT_BH(bh, buffer_jbd(bh)); + J_ASSERT_BH(bh, jh2bh(jh) == bh); + BUFFER_TRACE(bh, "remove journal_head"); + if (jh->b_frozen_data) { + printk(KERN_WARNING "%s: freeing b_frozen_data\n", __func__); + jbd2_free(jh->b_frozen_data, bh->b_size); } + if (jh->b_committed_data) { + printk(KERN_WARNING "%s: freeing b_committed_data\n", __func__); + jbd2_free(jh->b_committed_data, bh->b_size); + } + bh->b_private = NULL; + jh->b_bh = NULL; /* debug, really */ + clear_buffer_jbd(bh); + journal_free_journal_head(jh); } /* - * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction - * and has a zero b_jcount then remove and release its journal_head. If we did - * see that the buffer is not used by any transaction we also "logically" - * decrement ->b_count. - * - * We in fact take an additional increment on ->b_count as a convenience, - * because the caller usually wants to do additional things with the bh - * after calling here. - * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some - * time. Once the caller has run __brelse(), the buffer is eligible for - * reaping by try_to_free_buffers(). - */ -void jbd2_journal_remove_journal_head(struct buffer_head *bh) -{ - jbd_lock_bh_journal_head(bh); - __journal_remove_journal_head(bh); - jbd_unlock_bh_journal_head(bh); -} - -/* - * Drop a reference on the passed journal_head. If it fell to zero then try to + * Drop a reference on the passed journal_head. If it fell to zero then * release the journal_head from the buffer_head. */ void jbd2_journal_put_journal_head(struct journal_head *jh) @@ -2180,11 +2198,12 @@ void jbd2_journal_put_journal_head(struct journal_head *jh) jbd_lock_bh_journal_head(bh); J_ASSERT_JH(jh, jh->b_jcount > 0); --jh->b_jcount; - if (!jh->b_jcount && !jh->b_transaction) { + if (!jh->b_jcount) { __journal_remove_journal_head(bh); + jbd_unlock_bh_journal_head(bh); __brelse(bh); - } - jbd_unlock_bh_journal_head(bh); + } else + jbd_unlock_bh_journal_head(bh); } /* diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 05fa77a..2d71094 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -30,6 +30,7 @@ #include <linux/module.h> static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); +static void __jbd2_journal_unfile_buffer(struct journal_head *jh); /* * jbd2_get_transaction: obtain a new transaction_t object. @@ -82,7 +83,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) */ /* - * Update transiaction's maximum wait time, if debugging is enabled. + * Update transaction's maximum wait time, if debugging is enabled. * * In order for t_max_wait to be reliable, it must be protected by a * lock. But doing so will mean that start_this_handle() can not be @@ -91,11 +92,10 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) * means that maximum wait time reported by the jbd2_run_stats * tracepoint will always be zero. */ -static inline void update_t_max_wait(transaction_t *transaction) +static inline void update_t_max_wait(transaction_t *transaction, + unsigned long ts) { #ifdef CONFIG_JBD2_DEBUG - unsigned long ts = jiffies; - if (jbd2_journal_enable_debug && time_after(transaction->t_start, ts)) { ts = jbd2_time_diff(ts, transaction->t_start); @@ -121,6 +121,7 @@ static int start_this_handle(journal_t *journal, handle_t *handle, tid_t tid; int needed, need_to_start; int nblocks = handle->h_buffer_credits; + unsigned long ts = jiffies; if (nblocks > journal->j_max_transaction_buffers) { printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n", @@ -271,7 +272,7 @@ repeat: /* OK, account for the buffers that this operation expects to * use and add the handle to the running transaction. */ - update_t_max_wait(transaction); + update_t_max_wait(transaction, ts); handle->h_transaction = transaction; atomic_inc(&transaction->t_updates); atomic_inc(&transaction->t_handle_count); @@ -316,7 +317,8 @@ static handle_t *new_handle(int nblocks) * This function is visible to journal users (like ext3fs), so is not * called with the journal already locked. * - * Return a pointer to a newly allocated handle, or NULL on failure + * Return a pointer to a newly allocated handle, or an ERR_PTR() value + * on failure. */ handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int gfp_mask) { @@ -763,7 +765,6 @@ repeat: if (!jh->b_transaction) { JBUFFER_TRACE(jh, "no transaction"); J_ASSERT_JH(jh, !jh->b_next_transaction); - jh->b_transaction = transaction; JBUFFER_TRACE(jh, "file as BJ_Reserved"); spin_lock(&journal->j_list_lock); __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); @@ -813,7 +814,6 @@ out: * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. * @handle: transaction to add buffer modifications to * @bh: bh to be used for metadata writes - * @credits: variable that will receive credits for the buffer * * Returns an error code or 0 on success. * @@ -895,8 +895,6 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) * committed and so it's safe to clear the dirty bit. */ clear_buffer_dirty(jh2bh(jh)); - jh->b_transaction = transaction; - /* first access by this transaction */ jh->b_modified = 0; @@ -921,8 +919,8 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) */ JBUFFER_TRACE(jh, "cancelling revoke"); jbd2_journal_cancel_revoke(handle, jh); - jbd2_journal_put_journal_head(jh); out: + jbd2_journal_put_journal_head(jh); return err; } @@ -931,7 +929,6 @@ out: * non-rewindable consequences * @handle: transaction * @bh: buffer to undo - * @credits: store the number of taken credits here (if not NULL) * * Sometimes there is a need to distinguish between metadata which has * been committed to disk and that which has not. The ext3fs code uses @@ -1231,8 +1228,6 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); } else { __jbd2_journal_unfile_buffer(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); if (!buffer_jbd(bh)) { spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); @@ -1555,19 +1550,32 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) mark_buffer_dirty(bh); /* Expose it to the VM */ } -void __jbd2_journal_unfile_buffer(struct journal_head *jh) +/* + * Remove buffer from all transactions. + * + * Called with bh_state lock and j_list_lock + * + * jh and bh may be already freed when this function returns. + */ +static void __jbd2_journal_unfile_buffer(struct journal_head *jh) { __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = NULL; + jbd2_journal_put_journal_head(jh); } void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh) { - jbd_lock_bh_state(jh2bh(jh)); + struct buffer_head *bh = jh2bh(jh); + + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); + jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); __jbd2_journal_unfile_buffer(jh); spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(jh2bh(jh)); + jbd_unlock_bh_state(bh); + __brelse(bh); } /* @@ -1594,8 +1602,6 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); __jbd2_journal_remove_checkpoint(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); } } spin_unlock(&journal->j_list_lock); @@ -1658,7 +1664,6 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal, /* * We take our own ref against the journal_head here to avoid * having to add tons of locking around each instance of - * jbd2_journal_remove_journal_head() and * jbd2_journal_put_journal_head(). */ jh = jbd2_journal_grab_journal_head(bh); @@ -1696,10 +1701,9 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) int may_free = 1; struct buffer_head *bh = jh2bh(jh); - __jbd2_journal_unfile_buffer(jh); - if (jh->b_cp_transaction) { JBUFFER_TRACE(jh, "on running+cp transaction"); + __jbd2_journal_temp_unlink_buffer(jh); /* * We don't want to write the buffer anymore, clear the * bit so that we don't confuse checks in @@ -1710,8 +1714,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) may_free = 0; } else { JBUFFER_TRACE(jh, "on running transaction"); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); + __jbd2_journal_unfile_buffer(jh); } return may_free; } @@ -1989,6 +1992,8 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, if (jh->b_transaction) __jbd2_journal_temp_unlink_buffer(jh); + else + jbd2_journal_grab_journal_head(bh); jh->b_transaction = transaction; switch (jlist) { @@ -2040,9 +2045,10 @@ void jbd2_journal_file_buffer(struct journal_head *jh, * already started to be used by a subsequent transaction, refile the * buffer on that transaction's metadata list. * - * Called under journal->j_list_lock - * + * Called under j_list_lock * Called under jbd_lock_bh_state(jh2bh(jh)) + * + * jh and bh may be already free when this function returns */ void __jbd2_journal_refile_buffer(struct journal_head *jh) { @@ -2066,6 +2072,11 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) was_dirty = test_clear_buffer_jbddirty(bh); __jbd2_journal_temp_unlink_buffer(jh); + /* + * We set b_transaction here because b_next_transaction will inherit + * our jh reference and thus __jbd2_journal_file_buffer() must not + * take a new one. + */ jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; if (buffer_freed(bh)) @@ -2082,30 +2093,21 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) } /* - * For the unlocked version of this call, also make sure that any - * hanging journal_head is cleaned up if necessary. - * - * __jbd2_journal_refile_buffer is usually called as part of a single locked - * operation on a buffer_head, in which the caller is probably going to - * be hooking the journal_head onto other lists. In that case it is up - * to the caller to remove the journal_head if necessary. For the - * unlocked jbd2_journal_refile_buffer call, the caller isn't going to be - * doing anything else to the buffer so we need to do the cleanup - * ourselves to avoid a jh leak. - * - * *** The journal_head may be freed by this call! *** + * __jbd2_journal_refile_buffer() with necessary locking added. We take our + * bh reference so that we can safely unlock bh. + * + * The jh and bh may be freed by this call. */ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); - __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - spin_unlock(&journal->j_list_lock); __brelse(bh); } @@ -2147,6 +2149,13 @@ int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode) jinode->i_next_transaction == transaction) goto done; + /* + * We only ever set this variable to 1 so the test is safe. Since + * t_need_data_flush is likely to be set, we do the test to save some + * cacheline bouncing + */ + if (!transaction->t_need_data_flush) + transaction->t_need_data_flush = 1; /* On some different transaction's list - should be * the committing one */ if (jinode->i_transaction) { diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 82faddd..4bca6a2 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -75,7 +75,6 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, struct nameidata *nd) { struct jffs2_inode_info *dir_f; - struct jffs2_sb_info *c; struct jffs2_full_dirent *fd = NULL, *fd_list; uint32_t ino = 0; struct inode *inode = NULL; @@ -86,7 +85,6 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, return ERR_PTR(-ENAMETOOLONG); dir_f = JFFS2_INODE_INFO(dir_i); - c = JFFS2_SB_INFO(dir_i->i_sb); mutex_lock(&dir_f->sem); @@ -119,7 +117,6 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, static int jffs2_readdir(struct file *filp, void *dirent, filldir_t filldir) { struct jffs2_inode_info *f; - struct jffs2_sb_info *c; struct inode *inode = filp->f_path.dentry->d_inode; struct jffs2_full_dirent *fd; unsigned long offset, curofs; @@ -127,7 +124,6 @@ static int jffs2_readdir(struct file *filp, void *dirent, filldir_t filldir) D1(printk(KERN_DEBUG "jffs2_readdir() for dir_i #%lu\n", filp->f_path.dentry->d_inode->i_ino)); f = JFFS2_INODE_INFO(inode); - c = JFFS2_SB_INFO(inode->i_sb); offset = filp->f_pos; diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index e896e67..46ad619 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -357,7 +357,7 @@ error: return ERR_PTR(ret); } -void jffs2_dirty_inode(struct inode *inode) +void jffs2_dirty_inode(struct inode *inode, int flags) { struct iattr iattr; diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 00bae7c..65c6c43 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -172,7 +172,7 @@ int jffs2_setattr (struct dentry *, struct iattr *); int jffs2_do_setattr (struct inode *, struct iattr *); struct inode *jffs2_iget(struct super_block *, unsigned long); void jffs2_evict_inode (struct inode *); -void jffs2_dirty_inode(struct inode *inode); +void jffs2_dirty_inode(struct inode *inode, int flags); struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_inode *ri); int jffs2_statfs (struct dentry *, struct kstatfs *); diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index b632ddd..8d8cd34 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -94,7 +94,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) uint32_t buf_size = 0; struct jffs2_summary *s = NULL; /* summary info collected by the scan process */ #ifndef __ECOS - size_t pointlen; + size_t pointlen, try_size; if (c->mtd->point) { ret = c->mtd->point(c->mtd, 0, c->mtd->size, &pointlen, @@ -113,18 +113,21 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) /* For NAND it's quicker to read a whole eraseblock at a time, apparently */ if (jffs2_cleanmarker_oob(c)) - buf_size = c->sector_size; + try_size = c->sector_size; else - buf_size = PAGE_SIZE; + try_size = PAGE_SIZE; - /* Respect kmalloc limitations */ - if (buf_size > 128*1024) - buf_size = 128*1024; + D1(printk(KERN_DEBUG "Trying to allocate readbuf of %zu " + "bytes\n", try_size)); - D1(printk(KERN_DEBUG "Allocating readbuf of %d bytes\n", buf_size)); - flashbuf = kmalloc(buf_size, GFP_KERNEL); + flashbuf = mtd_kmalloc_up_to(c->mtd, &try_size); if (!flashbuf) return -ENOMEM; + + D1(printk(KERN_DEBUG "Allocated readbuf of %zu bytes\n", + try_size)); + + buf_size = (uint32_t)try_size; } if (jffs2_sum_active()) { diff --git a/fs/jfs/file.c b/fs/jfs/file.c index c5ce6c1..2f3f531 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -66,9 +66,9 @@ static int jfs_open(struct inode *inode, struct file *file) struct jfs_inode_info *ji = JFS_IP(inode); spin_lock_irq(&ji->ag_lock); if (ji->active_ag == -1) { - ji->active_ag = ji->agno; - atomic_inc( - &JFS_SBI(inode->i_sb)->bmap->db_active[ji->agno]); + struct jfs_sb_info *jfs_sb = JFS_SBI(inode->i_sb); + ji->active_ag = BLKTOAG(addressPXD(&ji->ixpxd), jfs_sb); + atomic_inc( &jfs_sb->bmap->db_active[ji->active_ag]); } spin_unlock_irq(&ji->ag_lock); } diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index eddbb37..1096559 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -173,7 +173,7 @@ void jfs_evict_inode(struct inode *inode) dquot_drop(inode); } -void jfs_dirty_inode(struct inode *inode) +void jfs_dirty_inode(struct inode *inode, int flags) { static int noisy = 5; diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index ed53a47..b78b2f9 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -397,7 +397,7 @@ int diRead(struct inode *ip) release_metapage(mp); /* set the ag for the inode */ - JFS_IP(ip)->agno = BLKTOAG(agstart, sbi); + JFS_IP(ip)->agstart = agstart; JFS_IP(ip)->active_ag = -1; return (rc); @@ -901,7 +901,7 @@ int diFree(struct inode *ip) /* get the allocation group for this ino. */ - agno = JFS_IP(ip)->agno; + agno = BLKTOAG(JFS_IP(ip)->agstart, JFS_SBI(ip->i_sb)); /* Lock the AG specific inode map information */ @@ -1315,12 +1315,11 @@ int diFree(struct inode *ip) static inline void diInitInode(struct inode *ip, int iagno, int ino, int extno, struct iag * iagp) { - struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb); struct jfs_inode_info *jfs_ip = JFS_IP(ip); ip->i_ino = (iagno << L2INOSPERIAG) + ino; jfs_ip->ixpxd = iagp->inoext[extno]; - jfs_ip->agno = BLKTOAG(le64_to_cpu(iagp->agstart), sbi); + jfs_ip->agstart = le64_to_cpu(iagp->agstart); jfs_ip->active_ag = -1; } @@ -1379,7 +1378,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip) */ /* get the ag number of this iag */ - agno = JFS_IP(pip)->agno; + agno = BLKTOAG(JFS_IP(pip)->agstart, JFS_SBI(pip->i_sb)); if (atomic_read(&JFS_SBI(pip->i_sb)->bmap->db_active[agno])) { /* @@ -2921,10 +2920,9 @@ int diExtendFS(struct inode *ipimap, struct inode *ipbmap) continue; } - /* agstart that computes to the same ag is treated as same; */ agstart = le64_to_cpu(iagp->agstart); - /* iagp->agstart = agstart & ~(mp->db_agsize - 1); */ n = agstart >> mp->db_agl2size; + iagp->agstart = cpu_to_le64((s64)n << mp->db_agl2size); /* compute backed inodes */ numinos = (EXTSPERIAG - le32_to_cpu(iagp->nfreeexts)) diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h index 1439f11..584a4a1 100644 --- a/fs/jfs/jfs_incore.h +++ b/fs/jfs/jfs_incore.h @@ -50,8 +50,9 @@ struct jfs_inode_info { short btindex; /* btpage entry index*/ struct inode *ipimap; /* inode map */ unsigned long cflag; /* commit flags */ + u64 agstart; /* agstart of the containing IAG */ u16 bxflag; /* xflag of pseudo buffer? */ - unchar agno; /* ag number */ + unchar pad; signed char active_ag; /* ag currently allocating from */ lid_t blid; /* lid of pseudo buffer? */ lid_t atlhead; /* anonymous tlock list head */ diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h index 155e91e..ec2fb8b 100644 --- a/fs/jfs/jfs_inode.h +++ b/fs/jfs/jfs_inode.h @@ -28,7 +28,7 @@ extern struct inode *jfs_iget(struct super_block *, unsigned long); extern int jfs_commit_inode(struct inode *, int); extern int jfs_write_inode(struct inode *, struct writeback_control *); extern void jfs_evict_inode(struct inode *); -extern void jfs_dirty_inode(struct inode *); +extern void jfs_dirty_inode(struct inode *, int); extern void jfs_truncate(struct inode *); extern void jfs_truncate_nolock(struct inode *, loff_t); extern void jfs_free_zero_link(struct inode *); diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index 278e3fb..583636f 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -1123,7 +1123,7 @@ int lmLogOpen(struct super_block *sb) bdev = blkdev_get_by_dev(sbi->logdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, log); if (IS_ERR(bdev)) { - rc = -PTR_ERR(bdev); + rc = PTR_ERR(bdev); goto free; } diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c index 8ea5efb..8d0c1c7 100644 --- a/fs/jfs/resize.c +++ b/fs/jfs/resize.c @@ -80,7 +80,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) int log_formatted = 0; struct inode *iplist[1]; struct jfs_superblock *j_sb, *j_sb2; - uint old_agsize; + s64 old_agsize; int agsizechanged = 0; struct buffer_head *bh, *bh2; diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index adb45ec..e374050 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -708,7 +708,13 @@ static void nlmclnt_unlock_callback(struct rpc_task *task, void *data) if (task->tk_status < 0) { dprintk("lockd: unlock failed (err = %d)\n", -task->tk_status); - goto retry_rebind; + switch (task->tk_status) { + case -EACCES: + case -EIO: + goto die; + default: + goto retry_rebind; + } } if (status == NLM_LCK_DENIED_GRACE_PERIOD) { rpc_delay(task, NLMCLNT_GRACE_WAIT); diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c index 1adc8d4..df0de27 100644 --- a/fs/logfs/dev_bdev.c +++ b/fs/logfs/dev_bdev.c @@ -10,6 +10,7 @@ #include <linux/blkdev.h> #include <linux/buffer_head.h> #include <linux/gfp.h> +#include <linux/prefetch.h> #define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index 9ed89d1..1afae26 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -555,13 +555,6 @@ static int logfs_symlink(struct inode *dir, struct dentry *dentry, return __logfs_create(dir, dentry, inode, target, destlen); } -static int logfs_permission(struct inode *inode, int mask, unsigned int flags) -{ - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - return generic_permission(inode, mask, flags, NULL); -} - static int logfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) { @@ -820,7 +813,6 @@ const struct inode_operations logfs_dir_iops = { .mknod = logfs_mknod, .rename = logfs_rename, .rmdir = logfs_rmdir, - .permission = logfs_permission, .symlink = logfs_symlink, .unlink = logfs_unlink, }; diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c index 9e22085..d8d0938 100644 --- a/fs/logfs/readwrite.c +++ b/fs/logfs/readwrite.c @@ -481,7 +481,7 @@ static int inode_write_alias(struct super_block *sb, val = inode_val0(inode); break; case INODE_USED_OFS: - val = cpu_to_be64(li->li_used_bytes);; + val = cpu_to_be64(li->li_used_bytes); break; case INODE_SIZE_OFS: val = cpu_to_be64(i_size_read(inode)); diff --git a/fs/mbcache.c b/fs/mbcache.c index 2f174be..8c32ef3 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -90,7 +90,8 @@ static DEFINE_SPINLOCK(mb_cache_spinlock); * What the mbcache registers as to get shrunk dynamically. */ -static int mb_cache_shrink_fn(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask); +static int mb_cache_shrink_fn(struct shrinker *shrink, + struct shrink_control *sc); static struct shrinker mb_cache_shrinker = { .shrink = mb_cache_shrink_fn, @@ -156,18 +157,19 @@ forget: * gets low. * * @shrink: (ignored) - * @nr_to_scan: Number of objects to scan - * @gfp_mask: (ignored) + * @sc: shrink_control passed from reclaim * * Returns the number of objects which are present in the cache. */ static int -mb_cache_shrink_fn(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask) +mb_cache_shrink_fn(struct shrinker *shrink, struct shrink_control *sc) { LIST_HEAD(free_list); struct mb_cache *cache; struct mb_cache_entry *entry, *tmp; int count = 0; + int nr_to_scan = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; mb_debug("trying to free %d entries", nr_to_scan); spin_lock(&mb_cache_spinlock); @@ -27,6 +27,7 @@ #include <linux/writeback.h> #include <linux/backing-dev.h> #include <linux/pagevec.h> +#include <linux/cleancache.h> /* * I/O completion handler for multipage BIOs. @@ -271,6 +272,12 @@ do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages, SetPageMappedToDisk(page); } + if (fully_mapped && blocks_per_page == 1 && !PageUptodate(page) && + cleancache_get_page(page) == 0) { + SetPageUptodate(page); + goto confused; + } + /* * This page will go to BIO. Do we need to send this BIO off first? */ @@ -238,7 +238,8 @@ int generic_permission(struct inode *inode, int mask, unsigned int flags, /* * Read/write DACs are always overridable. - * Executable DACs are overridable if at least one exec bit is set. + * Executable DACs are overridable for all directories and + * for non-directories that have least one exec bit set. */ if (!(mask & MAY_EXEC) || execute_ok(inode)) if (ns_capable(inode_userns(inode), CAP_DAC_OVERRIDE)) @@ -391,79 +392,28 @@ void path_put(struct path *path) } EXPORT_SYMBOL(path_put); -/** - * nameidata_drop_rcu - drop this nameidata out of rcu-walk - * @nd: nameidata pathwalk data to drop - * Returns: 0 on success, -ECHILD on failure - * +/* * Path walking has 2 modes, rcu-walk and ref-walk (see - * Documentation/filesystems/path-lookup.txt). __drop_rcu* functions attempt - * to drop out of rcu-walk mode and take normal reference counts on dentries - * and vfsmounts to transition to rcu-walk mode. __drop_rcu* functions take - * refcounts at the last known good point before rcu-walk got stuck, so - * ref-walk may continue from there. If this is not successful (eg. a seqcount - * has changed), then failure is returned and path walk restarts from the - * beginning in ref-walk mode. - * - * nameidata_drop_rcu attempts to drop the current nd->path and nd->root into - * ref-walk. Must be called from rcu-walk context. + * Documentation/filesystems/path-lookup.txt). In situations when we can't + * continue in RCU mode, we attempt to drop out of rcu-walk mode and grab + * normal reference counts on dentries and vfsmounts to transition to rcu-walk + * mode. Refcounts are grabbed at the last known good point before rcu-walk + * got stuck, so ref-walk may continue from there. If this is not successful + * (eg. a seqcount has changed), then failure is returned and it's up to caller + * to restart the path walk from the beginning in ref-walk mode. */ -static int nameidata_drop_rcu(struct nameidata *nd) -{ - struct fs_struct *fs = current->fs; - struct dentry *dentry = nd->path.dentry; - int want_root = 0; - - BUG_ON(!(nd->flags & LOOKUP_RCU)); - if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) { - want_root = 1; - spin_lock(&fs->lock); - if (nd->root.mnt != fs->root.mnt || - nd->root.dentry != fs->root.dentry) - goto err_root; - } - spin_lock(&dentry->d_lock); - if (!__d_rcu_to_refcount(dentry, nd->seq)) - goto err; - BUG_ON(nd->inode != dentry->d_inode); - spin_unlock(&dentry->d_lock); - if (want_root) { - path_get(&nd->root); - spin_unlock(&fs->lock); - } - mntget(nd->path.mnt); - - rcu_read_unlock(); - br_read_unlock(vfsmount_lock); - nd->flags &= ~LOOKUP_RCU; - return 0; -err: - spin_unlock(&dentry->d_lock); -err_root: - if (want_root) - spin_unlock(&fs->lock); - return -ECHILD; -} - -/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing. */ -static inline int nameidata_drop_rcu_maybe(struct nameidata *nd) -{ - if (nd->flags & LOOKUP_RCU) - return nameidata_drop_rcu(nd); - return 0; -} /** - * nameidata_dentry_drop_rcu - drop nameidata and dentry out of rcu-walk - * @nd: nameidata pathwalk data to drop - * @dentry: dentry to drop + * unlazy_walk - try to switch to ref-walk mode. + * @nd: nameidata pathwalk data + * @dentry: child of nd->path.dentry or NULL * Returns: 0 on success, -ECHILD on failure * - * nameidata_dentry_drop_rcu attempts to drop the current nd->path and nd->root, - * and dentry into ref-walk. @dentry must be a path found by a do_lookup call on - * @nd. Must be called from rcu-walk context. + * unlazy_walk attempts to legitimize the current nd->path, nd->root and dentry + * for ref-walk mode. @dentry must be a path found by a do_lookup call on + * @nd or NULL. Must be called from rcu-walk context. */ -static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry) +static int unlazy_walk(struct nameidata *nd, struct dentry *dentry) { struct fs_struct *fs = current->fs; struct dentry *parent = nd->path.dentry; @@ -478,18 +428,25 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry goto err_root; } spin_lock(&parent->d_lock); - spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - if (!__d_rcu_to_refcount(dentry, nd->seq)) - goto err; - /* - * If the sequence check on the child dentry passed, then the child has - * not been removed from its parent. This means the parent dentry must - * be valid and able to take a reference at this point. - */ - BUG_ON(!IS_ROOT(dentry) && dentry->d_parent != parent); - BUG_ON(!parent->d_count); - parent->d_count++; - spin_unlock(&dentry->d_lock); + if (!dentry) { + if (!__d_rcu_to_refcount(parent, nd->seq)) + goto err_parent; + BUG_ON(nd->inode != parent->d_inode); + } else { + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + if (!__d_rcu_to_refcount(dentry, nd->seq)) + goto err_child; + /* + * If the sequence check on the child dentry passed, then + * the child has not been removed from its parent. This + * means the parent dentry must be valid and able to take + * a reference at this point. + */ + BUG_ON(!IS_ROOT(dentry) && dentry->d_parent != parent); + BUG_ON(!parent->d_count); + parent->d_count++; + spin_unlock(&dentry->d_lock); + } spin_unlock(&parent->d_lock); if (want_root) { path_get(&nd->root); @@ -501,8 +458,10 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry br_read_unlock(vfsmount_lock); nd->flags &= ~LOOKUP_RCU; return 0; -err: + +err_child: spin_unlock(&dentry->d_lock); +err_parent: spin_unlock(&parent->d_lock); err_root: if (want_root) @@ -510,59 +469,6 @@ err_root: return -ECHILD; } -/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing. */ -static inline int nameidata_dentry_drop_rcu_maybe(struct nameidata *nd, struct dentry *dentry) -{ - if (nd->flags & LOOKUP_RCU) { - if (unlikely(nameidata_dentry_drop_rcu(nd, dentry))) { - nd->flags &= ~LOOKUP_RCU; - if (!(nd->flags & LOOKUP_ROOT)) - nd->root.mnt = NULL; - rcu_read_unlock(); - br_read_unlock(vfsmount_lock); - return -ECHILD; - } - } - return 0; -} - -/** - * nameidata_drop_rcu_last - drop nameidata ending path walk out of rcu-walk - * @nd: nameidata pathwalk data to drop - * Returns: 0 on success, -ECHILD on failure - * - * nameidata_drop_rcu_last attempts to drop the current nd->path into ref-walk. - * nd->path should be the final element of the lookup, so nd->root is discarded. - * Must be called from rcu-walk context. - */ -static int nameidata_drop_rcu_last(struct nameidata *nd) -{ - struct dentry *dentry = nd->path.dentry; - - BUG_ON(!(nd->flags & LOOKUP_RCU)); - nd->flags &= ~LOOKUP_RCU; - if (!(nd->flags & LOOKUP_ROOT)) - nd->root.mnt = NULL; - spin_lock(&dentry->d_lock); - if (!__d_rcu_to_refcount(dentry, nd->seq)) - goto err_unlock; - BUG_ON(nd->inode != dentry->d_inode); - spin_unlock(&dentry->d_lock); - - mntget(nd->path.mnt); - - rcu_read_unlock(); - br_read_unlock(vfsmount_lock); - - return 0; - -err_unlock: - spin_unlock(&dentry->d_lock); - rcu_read_unlock(); - br_read_unlock(vfsmount_lock); - return -ECHILD; -} - /** * release_open_intent - free up open intent resources * @nd: pointer to nameidata @@ -606,26 +512,39 @@ do_revalidate(struct dentry *dentry, struct nameidata *nd) return dentry; } -/* - * handle_reval_path - force revalidation of a dentry - * - * In some situations the path walking code will trust dentries without - * revalidating them. This causes problems for filesystems that depend on - * d_revalidate to handle file opens (e.g. NFSv4). When FS_REVAL_DOT is set - * (which indicates that it's possible for the dentry to go stale), force - * a d_revalidate call before proceeding. +/** + * complete_walk - successful completion of path walk + * @nd: pointer nameidata * - * Returns 0 if the revalidation was successful. If the revalidation fails, - * either return the error returned by d_revalidate or -ESTALE if the - * revalidation it just returned 0. If d_revalidate returns 0, we attempt to - * invalidate the dentry. It's up to the caller to handle putting references - * to the path if necessary. + * If we had been in RCU mode, drop out of it and legitimize nd->path. + * Revalidate the final result, unless we'd already done that during + * the path walk or the filesystem doesn't ask for it. Return 0 on + * success, -error on failure. In case of failure caller does not + * need to drop nd->path. */ -static inline int handle_reval_path(struct nameidata *nd) +static int complete_walk(struct nameidata *nd) { struct dentry *dentry = nd->path.dentry; int status; + if (nd->flags & LOOKUP_RCU) { + nd->flags &= ~LOOKUP_RCU; + if (!(nd->flags & LOOKUP_ROOT)) + nd->root.mnt = NULL; + spin_lock(&dentry->d_lock); + if (unlikely(!__d_rcu_to_refcount(dentry, nd->seq))) { + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + return -ECHILD; + } + BUG_ON(nd->inode != dentry->d_inode); + spin_unlock(&dentry->d_lock); + mntget(nd->path.mnt); + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + } + if (likely(!(nd->flags & LOOKUP_JUMPED))) return 0; @@ -643,6 +562,7 @@ static inline int handle_reval_path(struct nameidata *nd) if (!status) status = -ESTALE; + path_put(&nd->path); return status; } @@ -893,6 +813,11 @@ static int follow_automount(struct path *path, unsigned flags, if (!mnt) /* mount collision */ return 0; + if (!*need_mntput) { + /* lock_mount() may release path->mnt on error */ + mntget(path->mnt); + *need_mntput = true; + } err = finish_automount(mnt, path); switch (err) { @@ -900,12 +825,9 @@ static int follow_automount(struct path *path, unsigned flags, /* Someone else made a mount here whilst we were busy */ return 0; case 0: - dput(path->dentry); - if (*need_mntput) - mntput(path->mnt); + path_put(path); path->mnt = mnt; path->dentry = dget(mnt->mnt_root); - *need_mntput = true; return 0; default: return err; @@ -925,9 +847,10 @@ static int follow_automount(struct path *path, unsigned flags, */ static int follow_managed(struct path *path, unsigned flags) { + struct vfsmount *mnt = path->mnt; /* held by caller, must be left alone */ unsigned managed; bool need_mntput = false; - int ret; + int ret = 0; /* Given that we're not holding a lock here, we retain the value in a * local variable for each dentry as we look at it so that we don't see @@ -942,7 +865,7 @@ static int follow_managed(struct path *path, unsigned flags) BUG_ON(!path->dentry->d_op->d_manage); ret = path->dentry->d_op->d_manage(path->dentry, false); if (ret < 0) - return ret == -EISDIR ? 0 : ret; + break; } /* Transit to a mounted filesystem. */ @@ -968,14 +891,19 @@ static int follow_managed(struct path *path, unsigned flags) if (managed & DCACHE_NEED_AUTOMOUNT) { ret = follow_automount(path, flags, &need_mntput); if (ret < 0) - return ret == -EISDIR ? 0 : ret; + break; continue; } /* We didn't change the current path point */ break; } - return 0; + + if (need_mntput && path->mnt == mnt) + mntput(path->mnt); + if (ret == -EISDIR) + ret = 0; + return ret; } int follow_down_one(struct path *path) @@ -1000,12 +928,11 @@ static inline bool managed_dentry_might_block(struct dentry *dentry) } /* - * Skip to top of mountpoint pile in rcuwalk mode. We abort the rcu-walk if we - * meet a managed dentry and we're not walking to "..". True is returned to - * continue, false to abort. + * Try to skip to top of mountpoint pile in rcuwalk mode. Fail if + * we meet a managed dentry that would need blocking. */ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, - struct inode **inode, bool reverse_transit) + struct inode **inode) { for (;;) { struct vfsmount *mounted; @@ -1014,8 +941,7 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, * that wants to block transit. */ *inode = path->dentry->d_inode; - if (!reverse_transit && - unlikely(managed_dentry_might_block(path->dentry))) + if (unlikely(managed_dentry_might_block(path->dentry))) return false; if (!d_mountpoint(path->dentry)) @@ -1028,16 +954,24 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, path->dentry = mounted->mnt_root; nd->seq = read_seqcount_begin(&path->dentry->d_seq); } - - if (unlikely(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT)) - return reverse_transit; return true; } -static int follow_dotdot_rcu(struct nameidata *nd) +static void follow_mount_rcu(struct nameidata *nd) { - struct inode *inode = nd->inode; + while (d_mountpoint(nd->path.dentry)) { + struct vfsmount *mounted; + mounted = __lookup_mnt(nd->path.mnt, nd->path.dentry, 1); + if (!mounted) + break; + nd->path.mnt = mounted; + nd->path.dentry = mounted->mnt_root; + nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq); + } +} +static int follow_dotdot_rcu(struct nameidata *nd) +{ set_root_rcu(nd); while (1) { @@ -1053,7 +987,6 @@ static int follow_dotdot_rcu(struct nameidata *nd) seq = read_seqcount_begin(&parent->d_seq); if (read_seqcount_retry(&old->d_seq, nd->seq)) goto failed; - inode = parent->d_inode; nd->path.dentry = parent; nd->seq = seq; break; @@ -1061,10 +994,9 @@ static int follow_dotdot_rcu(struct nameidata *nd) if (!follow_up_rcu(&nd->path)) break; nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq); - inode = nd->path.dentry->d_inode; } - __follow_mount_rcu(nd, &nd->path, &inode, true); - nd->inode = inode; + follow_mount_rcu(nd); + nd->inode = nd->path.dentry->d_inode; return 0; failed: @@ -1080,9 +1012,6 @@ failed: * Follow down to the covering mount currently visible to userspace. At each * point, the filesystem owning that dentry may be queried as to whether the * caller is permitted to proceed or not. - * - * Care must be taken as namespace_sem may be held (indicated by mounting_here - * being true). */ int follow_down(struct path *path) { @@ -1238,16 +1167,14 @@ static int do_lookup(struct nameidata *nd, struct qstr *name, } path->mnt = mnt; path->dentry = dentry; - if (likely(__follow_mount_rcu(nd, path, inode, false))) - return 0; + if (unlikely(!__follow_mount_rcu(nd, path, inode))) + goto unlazy; + if (unlikely(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT)) + goto unlazy; + return 0; unlazy: - if (dentry) { - if (nameidata_dentry_drop_rcu(nd, dentry)) - return -ECHILD; - } else { - if (nameidata_drop_rcu(nd)) - return -ECHILD; - } + if (unlazy_walk(nd, dentry)) + return -ECHILD; } else { dentry = __d_lookup(parent, name); } @@ -1303,7 +1230,7 @@ static inline int may_lookup(struct nameidata *nd) int err = exec_permission(nd->inode, IPERM_FLAG_RCU); if (err != -ECHILD) return err; - if (nameidata_drop_rcu(nd)) + if (unlazy_walk(nd, NULL)) return -ECHILD; } return exec_permission(nd->inode, 0); @@ -1357,8 +1284,12 @@ static inline int walk_component(struct nameidata *nd, struct path *path, return -ENOENT; } if (unlikely(inode->i_op->follow_link) && follow) { - if (nameidata_dentry_drop_rcu_maybe(nd, path->dentry)) - return -ECHILD; + if (nd->flags & LOOKUP_RCU) { + if (unlikely(unlazy_walk(nd, path->dentry))) { + terminate_walk(nd); + return -ECHILD; + } + } BUG_ON(inode != path->dentry->d_inode); return 1; } @@ -1378,12 +1309,12 @@ static inline int nested_symlink(struct path *path, struct nameidata *nd) { int res; - BUG_ON(nd->depth >= MAX_NESTED_LINKS); if (unlikely(current->link_count >= MAX_NESTED_LINKS)) { path_put_conditional(path, nd); path_put(&nd->path); return -ELOOP; } + BUG_ON(nd->depth >= MAX_NESTED_LINKS); nd->depth++; current->link_count++; @@ -1657,18 +1588,8 @@ static int path_lookupat(int dfd, const char *name, } } - if (nd->flags & LOOKUP_RCU) { - /* went all way through without dropping RCU */ - BUG_ON(err); - if (nameidata_drop_rcu_last(nd)) - err = -ECHILD; - } - - if (!err) { - err = handle_reval_path(nd); - if (err) - path_put(&nd->path); - } + if (!err) + err = complete_walk(nd); if (!err && nd->flags & LOOKUP_DIRECTORY) { if (!nd->inode->i_op->lookup) { @@ -2134,13 +2055,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, return ERR_PTR(error); /* fallthrough */ case LAST_ROOT: - if (nd->flags & LOOKUP_RCU) { - if (nameidata_drop_rcu_last(nd)) - return ERR_PTR(-ECHILD); - } - error = handle_reval_path(nd); + error = complete_walk(nd); if (error) - goto exit; + return ERR_PTR(error); audit_inode(pathname, nd->path.dentry); if (open_flag & O_CREAT) { error = -EISDIR; @@ -2148,10 +2065,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, } goto ok; case LAST_BIND: - /* can't be RCU mode here */ - error = handle_reval_path(nd); + error = complete_walk(nd); if (error) - goto exit; + return ERR_PTR(error); audit_inode(pathname, dir); goto ok; } @@ -2170,10 +2086,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, if (error) /* symlink */ return NULL; /* sayonara */ - if (nd->flags & LOOKUP_RCU) { - if (nameidata_drop_rcu_last(nd)) - return ERR_PTR(-ECHILD); - } + error = complete_walk(nd); + if (error) + return ERR_PTR(-ECHILD); error = -ENOTDIR; if (nd->flags & LOOKUP_DIRECTORY) { @@ -2185,11 +2100,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, } /* create side of things */ - - if (nd->flags & LOOKUP_RCU) { - if (nameidata_drop_rcu_last(nd)) - return ERR_PTR(-ECHILD); - } + error = complete_walk(nd); + if (error) + return ERR_PTR(error); audit_inode(pathname, dir); error = -EISDIR; @@ -2629,10 +2542,10 @@ SYSCALL_DEFINE2(mkdir, const char __user *, pathname, int, mode) } /* - * We try to drop the dentry early: we should have - * a usage count of 2 if we're the only user of this - * dentry, and if that is true (possibly after pruning - * the dcache), then we drop the dentry now. + * The dentry_unhash() helper will try to drop the dentry early: we + * should have a usage count of 2 if we're the only user of this + * dentry, and if that is true (possibly after pruning the dcache), + * then we drop the dentry now. * * A low-level filesystem can, if it choses, legally * do a @@ -2645,10 +2558,9 @@ SYSCALL_DEFINE2(mkdir, const char __user *, pathname, int, mode) */ void dentry_unhash(struct dentry *dentry) { - dget(dentry); shrink_dcache_parent(dentry); spin_lock(&dentry->d_lock); - if (dentry->d_count == 2) + if (dentry->d_count == 1) __d_drop(dentry); spin_unlock(&dentry->d_lock); } @@ -2664,25 +2576,27 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) return -EPERM; mutex_lock(&dentry->d_inode->i_mutex); - dentry_unhash(dentry); + + error = -EBUSY; if (d_mountpoint(dentry)) - error = -EBUSY; - else { - error = security_inode_rmdir(dir, dentry); - if (!error) { - error = dir->i_op->rmdir(dir, dentry); - if (!error) { - dentry->d_inode->i_flags |= S_DEAD; - dont_mount(dentry); - } - } - } + goto out; + + error = security_inode_rmdir(dir, dentry); + if (error) + goto out; + + shrink_dcache_parent(dentry); + error = dir->i_op->rmdir(dir, dentry); + if (error) + goto out; + + dentry->d_inode->i_flags |= S_DEAD; + dont_mount(dentry); + +out: mutex_unlock(&dentry->d_inode->i_mutex); - if (!error) { + if (!error) d_delete(dentry); - } - dput(dentry); - return error; } @@ -2716,6 +2630,10 @@ static long do_rmdir(int dfd, const char __user *pathname) error = PTR_ERR(dentry); if (IS_ERR(dentry)) goto exit2; + if (!dentry->d_inode) { + error = -ENOENT; + goto exit3; + } error = mnt_want_write(nd.path.mnt); if (error) goto exit3; @@ -2804,8 +2722,9 @@ static long do_unlinkat(int dfd, const char __user *pathname) if (nd.last.name[nd.last.len]) goto slashes; inode = dentry->d_inode; - if (inode) - ihold(inode); + if (!inode) + goto slashes; + ihold(inode); error = mnt_want_write(nd.path.mnt); if (error) goto exit2; @@ -3053,12 +2972,7 @@ SYSCALL_DEFINE2(link, const char __user *, oldname, const char __user *, newname * HOWEVER, it relies on the assumption that any object with ->lookup() * has no more than 1 dentry. If "hybrid" objects will ever appear, * we'd better make sure that there's no link(2) for them. - * d) some filesystems don't support opened-but-unlinked directories, - * either because of layout or because they are not ready to deal with - * all cases correctly. The latter will be fixed (taking this sort of - * stuff into VFS), but the former is not going away. Solution: the same - * trick as in rmdir(). - * e) conversion from fhandle to dentry may come in the wrong moment - when + * d) conversion from fhandle to dentry may come in the wrong moment - when * we are removing the target. Solution: we will have to grab ->i_mutex * in the fhandle_to_dentry code. [FIXME - current nfsfh.c relies on * ->i_mutex on parents, which works but leads to some truly excessive @@ -3068,7 +2982,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry) { int error = 0; - struct inode *target; + struct inode *target = new_dentry->d_inode; /* * If we are going to change the parent - check write permissions, @@ -3084,26 +2998,26 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, if (error) return error; - target = new_dentry->d_inode; if (target) mutex_lock(&target->i_mutex); - if (d_mountpoint(old_dentry)||d_mountpoint(new_dentry)) - error = -EBUSY; - else { - if (target) - dentry_unhash(new_dentry); - error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry); - } + + error = -EBUSY; + if (d_mountpoint(old_dentry) || d_mountpoint(new_dentry)) + goto out; + + if (target) + shrink_dcache_parent(new_dentry); + error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry); + if (error) + goto out; + if (target) { - if (!error) { - target->i_flags |= S_DEAD; - dont_mount(new_dentry); - } - mutex_unlock(&target->i_mutex); - if (d_unhashed(new_dentry)) - d_rehash(new_dentry); - dput(new_dentry); + target->i_flags |= S_DEAD; + dont_mount(new_dentry); } +out: + if (target) + mutex_unlock(&target->i_mutex); if (!error) if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE)) d_move(old_dentry,new_dentry); @@ -3113,7 +3027,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry) { - struct inode *target; + struct inode *target = new_dentry->d_inode; int error; error = security_inode_rename(old_dir, old_dentry, new_dir, new_dentry); @@ -3121,19 +3035,22 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, return error; dget(new_dentry); - target = new_dentry->d_inode; if (target) mutex_lock(&target->i_mutex); + + error = -EBUSY; if (d_mountpoint(old_dentry)||d_mountpoint(new_dentry)) - error = -EBUSY; - else - error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry); - if (!error) { - if (target) - dont_mount(new_dentry); - if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE)) - d_move(old_dentry, new_dentry); - } + goto out; + + error = old_dir->i_op->rename(old_dir, old_dentry, new_dir, new_dentry); + if (error) + goto out; + + if (target) + dont_mount(new_dentry); + if (!(old_dir->i_sb->s_type->fs_flags & FS_RENAME_DOES_D_MOVE)) + d_move(old_dentry, new_dentry); +out: if (target) mutex_unlock(&target->i_mutex); dput(new_dentry); diff --git a/fs/namespace.c b/fs/namespace.c index d99bcf5..fe59bd1 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1695,7 +1695,7 @@ static int graft_tree(struct vfsmount *mnt, struct path *path) static int flags_to_propagation_type(int flags) { - int type = flags & ~MS_REC; + int type = flags & ~(MS_REC | MS_SILENT); /* Fail if any non-propagation flags are set */ if (type & ~(MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index f6946bb..9c51f62 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -1033,6 +1033,11 @@ static int ncp_rmdir(struct inode *dir, struct dentry *dentry) DPRINTK("ncp_rmdir: removing %s/%s\n", dentry->d_parent->d_name.name, dentry->d_name.name); + /* + * fail with EBUSY if there are still references to this + * directory. + */ + dentry_unhash(dentry); error = -EBUSY; if (!d_unhashed(dentry)) goto out; @@ -1139,6 +1144,17 @@ static int ncp_rename(struct inode *old_dir, struct dentry *old_dentry, old_dentry->d_parent->d_name.name, old_dentry->d_name.name, new_dentry->d_parent->d_name.name, new_dentry->d_name.name); + if (new_dentry->d_inode && S_ISDIR(new_dentry->d_inode->i_mode)) { + /* + * fail with EBUSY if there are still references to this + * directory. + */ + dentry_unhash(new_dentry); + error = -EBUSY; + if (!d_unhashed(new_dentry)) + goto out; + } + ncp_age_dentry(server, old_dentry); ncp_age_dentry(server, new_dentry); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 0250e4c..202f370 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -461,7 +461,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) #endif struct ncp_entry_info finfo; - data.wdog_pid = NULL; + memset(&data, 0, sizeof(data)); server = kzalloc(sizeof(struct ncp_server), GFP_KERNEL); if (!server) return -ENOMEM; @@ -496,7 +496,6 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) struct ncp_mount_data_v4* md = (struct ncp_mount_data_v4*)raw_data; data.flags = md->flags; - data.int_flags = 0; data.mounted_uid = md->mounted_uid; data.wdog_pid = find_get_pid(md->wdog_pid); data.ncp_fd = md->ncp_fd; @@ -507,7 +506,6 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) data.file_mode = md->file_mode; data.dir_mode = md->dir_mode; data.info_fd = -1; - data.mounted_vol[0] = 0; } break; default: diff --git a/fs/ncpfs/mmap.c b/fs/ncpfs/mmap.c index a7c07b4..e5d71b2 100644 --- a/fs/ncpfs/mmap.c +++ b/fs/ncpfs/mmap.c @@ -16,6 +16,7 @@ #include <linux/mman.h> #include <linux/string.h> #include <linux/fcntl.h> +#include <linux/memcontrol.h> #include <asm/uaccess.h> #include <asm/system.h> @@ -92,6 +93,7 @@ static int ncp_file_mmap_fault(struct vm_area_struct *area, * -- wli */ count_vm_event(PGMAJFAULT); + mem_cgroup_count_vm_event(area->vm_mm, PGMAJFAULT); return VM_FAULT_MAJOR; } diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index ba30665..8151554 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -87,6 +87,16 @@ config NFS_V4_1 config PNFS_FILE_LAYOUT tristate +config PNFS_OBJLAYOUT + tristate "Provide support for the pNFS Objects Layout Driver for NFSv4.1 pNFS (EXPERIMENTAL)" + depends on NFS_FS && NFS_V4_1 && SCSI_OSD_ULD + help + Say M here if you want your pNFS client to support the Objects Layout Driver. + Requires the SCSI osd initiator library (SCSI_OSD_INITIATOR) and + upper level driver (SCSI_OSD_ULD). + + If unsure, say N. + config ROOT_NFS bool "Root file system on NFS" depends on NFS_FS=y && IP_PNP diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index 4776ff9..6a34f7d 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -15,9 +15,11 @@ nfs-$(CONFIG_NFS_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o \ delegation.o idmap.o \ callback.o callback_xdr.o callback_proc.o \ nfs4namespace.o -nfs-$(CONFIG_NFS_V4_1) += pnfs.o +nfs-$(CONFIG_NFS_V4_1) += pnfs.o pnfs_dev.o nfs-$(CONFIG_SYSCTL) += sysctl.o nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o nfs_layout_nfsv41_files-y := nfs4filelayout.o nfs4filelayoutdev.o + +obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayout/ diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 46d93ce..b257383 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -167,6 +167,23 @@ extern unsigned nfs4_callback_layoutrecall( extern void nfs4_check_drain_bc_complete(struct nfs4_session *ses); extern void nfs4_cb_take_slot(struct nfs_client *clp); + +struct cb_devicenotifyitem { + uint32_t cbd_notify_type; + uint32_t cbd_layout_type; + struct nfs4_deviceid cbd_dev_id; + uint32_t cbd_immediate; +}; + +struct cb_devicenotifyargs { + int ndevs; + struct cb_devicenotifyitem *devs; +}; + +extern __be32 nfs4_callback_devicenotify( + struct cb_devicenotifyargs *args, + void *dummy, struct cb_process_state *cps); + #endif /* CONFIG_NFS_V4_1 */ extern int check_gss_callback_principal(struct nfs_client *, struct svc_rqst *); extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 2f41dcce..d4d1954 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -139,7 +139,7 @@ static u32 initiate_file_draining(struct nfs_client *clp, spin_lock(&ino->i_lock); if (test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags) || mark_matching_lsegs_invalid(lo, &free_me_list, - args->cbl_range.iomode)) + &args->cbl_range)) rv = NFS4ERR_DELAY; else rv = NFS4ERR_NOMATCHING_LAYOUT; @@ -184,7 +184,7 @@ static u32 initiate_bulk_draining(struct nfs_client *clp, ino = lo->plh_inode; spin_lock(&ino->i_lock); set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); - if (mark_matching_lsegs_invalid(lo, &free_me_list, range.iomode)) + if (mark_matching_lsegs_invalid(lo, &free_me_list, &range)) rv = NFS4ERR_DELAY; list_del_init(&lo->plh_bulk_recall); spin_unlock(&ino->i_lock); @@ -241,6 +241,53 @@ static void pnfs_recall_all_layouts(struct nfs_client *clp) do_callback_layoutrecall(clp, &args); } +__be32 nfs4_callback_devicenotify(struct cb_devicenotifyargs *args, + void *dummy, struct cb_process_state *cps) +{ + int i; + __be32 res = 0; + struct nfs_client *clp = cps->clp; + struct nfs_server *server = NULL; + + dprintk("%s: -->\n", __func__); + + if (!clp) { + res = cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION); + goto out; + } + + for (i = 0; i < args->ndevs; i++) { + struct cb_devicenotifyitem *dev = &args->devs[i]; + + if (!server || + server->pnfs_curr_ld->id != dev->cbd_layout_type) { + rcu_read_lock(); + list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) + if (server->pnfs_curr_ld && + server->pnfs_curr_ld->id == dev->cbd_layout_type) { + rcu_read_unlock(); + goto found; + } + rcu_read_unlock(); + dprintk("%s: layout type %u not found\n", + __func__, dev->cbd_layout_type); + continue; + } + + found: + if (dev->cbd_notify_type == NOTIFY_DEVICEID4_CHANGE) + dprintk("%s: NOTIFY_DEVICEID4_CHANGE not supported, " + "deleting instead\n", __func__); + nfs4_delete_deviceid(server->pnfs_curr_ld, clp, &dev->cbd_dev_id); + } + +out: + kfree(args->devs); + dprintk("%s: exit with status = %u\n", + __func__, be32_to_cpu(res)); + return res; +} + int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid) { if (delegation == NULL) diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 00ecf62..c6c86a7 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -25,6 +25,7 @@ #if defined(CONFIG_NFS_V4_1) #define CB_OP_LAYOUTRECALL_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) +#define CB_OP_DEVICENOTIFY_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) #define CB_OP_SEQUENCE_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ + \ 4 + 1 + 3) #define CB_OP_RECALLANY_RES_MAXSZ (CB_OP_HDR_RES_MAXSZ) @@ -284,6 +285,93 @@ out: return status; } +static +__be32 decode_devicenotify_args(struct svc_rqst *rqstp, + struct xdr_stream *xdr, + struct cb_devicenotifyargs *args) +{ + __be32 *p; + __be32 status = 0; + u32 tmp; + int n, i; + args->ndevs = 0; + + /* Num of device notifications */ + p = read_buf(xdr, sizeof(uint32_t)); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto out; + } + n = ntohl(*p++); + if (n <= 0) + goto out; + + args->devs = kmalloc(n * sizeof(*args->devs), GFP_KERNEL); + if (!args->devs) { + status = htonl(NFS4ERR_DELAY); + goto out; + } + + /* Decode each dev notification */ + for (i = 0; i < n; i++) { + struct cb_devicenotifyitem *dev = &args->devs[i]; + + p = read_buf(xdr, (4 * sizeof(uint32_t)) + NFS4_DEVICEID4_SIZE); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto err; + } + + tmp = ntohl(*p++); /* bitmap size */ + if (tmp != 1) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + dev->cbd_notify_type = ntohl(*p++); + if (dev->cbd_notify_type != NOTIFY_DEVICEID4_CHANGE && + dev->cbd_notify_type != NOTIFY_DEVICEID4_DELETE) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + + tmp = ntohl(*p++); /* opaque size */ + if (((dev->cbd_notify_type == NOTIFY_DEVICEID4_CHANGE) && + (tmp != NFS4_DEVICEID4_SIZE + 8)) || + ((dev->cbd_notify_type == NOTIFY_DEVICEID4_DELETE) && + (tmp != NFS4_DEVICEID4_SIZE + 4))) { + status = htonl(NFS4ERR_INVAL); + goto err; + } + dev->cbd_layout_type = ntohl(*p++); + memcpy(dev->cbd_dev_id.data, p, NFS4_DEVICEID4_SIZE); + p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); + + if (dev->cbd_layout_type == NOTIFY_DEVICEID4_CHANGE) { + p = read_buf(xdr, sizeof(uint32_t)); + if (unlikely(p == NULL)) { + status = htonl(NFS4ERR_BADXDR); + goto err; + } + dev->cbd_immediate = ntohl(*p++); + } else { + dev->cbd_immediate = 0; + } + + args->ndevs++; + + dprintk("%s: type %d layout 0x%x immediate %d\n", + __func__, dev->cbd_notify_type, dev->cbd_layout_type, + dev->cbd_immediate); + } +out: + dprintk("%s: status %d ndevs %d\n", + __func__, ntohl(status), args->ndevs); + return status; +err: + kfree(args->devs); + goto out; +} + static __be32 decode_sessionid(struct xdr_stream *xdr, struct nfs4_sessionid *sid) { @@ -639,10 +727,10 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op) case OP_CB_RECALL_ANY: case OP_CB_RECALL_SLOT: case OP_CB_LAYOUTRECALL: + case OP_CB_NOTIFY_DEVICEID: *op = &callback_ops[op_nr]; break; - case OP_CB_NOTIFY_DEVICEID: case OP_CB_NOTIFY: case OP_CB_PUSH_DELEG: case OP_CB_RECALLABLE_OBJ_AVAIL: @@ -849,6 +937,12 @@ static struct callback_op callback_ops[] = { (callback_decode_arg_t)decode_layoutrecall_args, .res_maxsize = CB_OP_LAYOUTRECALL_RES_MAXSZ, }, + [OP_CB_NOTIFY_DEVICEID] = { + .process_op = (callback_process_op_t)nfs4_callback_devicenotify, + .decode_args = + (callback_decode_arg_t)decode_devicenotify_args, + .res_maxsize = CB_OP_DEVICENOTIFY_RES_MAXSZ, + }, [OP_CB_SEQUENCE] = { .process_op = (callback_process_op_t)nfs4_callback_sequence, .decode_args = (callback_decode_arg_t)decode_cb_sequence_args, diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 139be96..b3dc2b8 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -290,6 +290,8 @@ static void nfs_free_client(struct nfs_client *clp) if (clp->cl_machine_cred != NULL) put_rpccred(clp->cl_machine_cred); + nfs4_deviceid_purge_client(clp); + kfree(clp->cl_hostname); kfree(clp); diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index bbbc6bf..dd25c2a 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -21,25 +21,13 @@ #include "delegation.h" #include "internal.h" -static void nfs_do_free_delegation(struct nfs_delegation *delegation) -{ - kfree(delegation); -} - -static void nfs_free_delegation_callback(struct rcu_head *head) -{ - struct nfs_delegation *delegation = container_of(head, struct nfs_delegation, rcu); - - nfs_do_free_delegation(delegation); -} - static void nfs_free_delegation(struct nfs_delegation *delegation) { if (delegation->cred) { put_rpccred(delegation->cred); delegation->cred = NULL; } - call_rcu(&delegation->rcu, nfs_free_delegation_callback); + kfree_rcu(delegation, rcu); } /** diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 7237672..ededdbd 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -512,12 +512,7 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en struct page **xdr_pages, struct page *page, unsigned int buflen) { struct xdr_stream stream; - struct xdr_buf buf = { - .pages = xdr_pages, - .page_len = buflen, - .buflen = buflen, - .len = buflen, - }; + struct xdr_buf buf; struct page *scratch; struct nfs_cache_array *array; unsigned int count = 0; @@ -527,7 +522,7 @@ int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *en if (scratch == NULL) return -ENOMEM; - xdr_init_decode(&stream, &buf, NULL); + xdr_init_decode_pages(&stream, &buf, xdr_pages, buflen); xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); do { @@ -2042,11 +2037,14 @@ static void nfs_access_free_list(struct list_head *head) } } -int nfs_access_cache_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask) +int nfs_access_cache_shrinker(struct shrinker *shrink, + struct shrink_control *sc) { LIST_HEAD(head); struct nfs_inode *nfsi, *next; struct nfs_access_entry *cache; + int nr_to_scan = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; if ((gfp_mask & GFP_KERNEL) != GFP_KERNEL) return (nr_to_scan == 0) ? 0 : -1; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 57bb31a..6f4850d 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -256,7 +256,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) nfs_attr_check_mountpoint(sb, fattr); - if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0 && (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) + if (((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0) && + !nfs_attr_use_mounted_on_fileid(fattr)) goto out_no_inode; if ((fattr->valid & NFS_ATTR_FATTR_TYPE) == 0) goto out_no_inode; @@ -1294,12 +1295,17 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (new_isize != cur_isize) { /* Do we perhaps have any outstanding writes, or has * the file grown beyond our last write? */ - if (nfsi->npages == 0 || new_isize > cur_isize) { + if ((nfsi->npages == 0 && !test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) || + new_isize > cur_isize) { i_size_write(inode, new_isize); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; } - dprintk("NFS: isize change on server for file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); + dprintk("NFS: isize change on server for file %s/%ld " + "(%Ld to %Ld)\n", + inode->i_sb->s_id, + inode->i_ino, + (long long)cur_isize, + (long long)new_isize); } } else invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR @@ -1424,9 +1430,10 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) */ void nfs4_evict_inode(struct inode *inode) { - pnfs_destroy_layout(NFS_I(inode)); truncate_inode_pages(&inode->i_data, 0); end_writeback(inode); + pnfs_return_layout(inode); + pnfs_destroy_layout(NFS_I(inode)); /* If we are holding a delegation, return it! */ nfs_inode_return_delegation_noreclaim(inode); /* First call standard NFS clear_inode() code */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index ce118ce..2a55347 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -45,6 +45,17 @@ static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct fattr->valid |= NFS_ATTR_FATTR_MOUNTPOINT; } +static inline int nfs_attr_use_mounted_on_fileid(struct nfs_fattr *fattr) +{ + if (((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) == 0) || + (((fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) && + ((fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) == 0))) + return 0; + + fattr->fileid = fattr->mounted_on_fileid; + return 1; +} + struct nfs_clone_mount { const struct super_block *sb; const struct dentry *dentry; @@ -234,7 +245,7 @@ extern int nfs_init_client(struct nfs_client *clp, /* dir.c */ extern int nfs_access_cache_shrinker(struct shrinker *shrink, - int nr_to_scan, gfp_t gfp_mask); + struct shrink_control *sc); /* inode.c */ extern struct workqueue_struct *nfsiod_workqueue; @@ -310,6 +321,7 @@ extern int nfs_migrate_page(struct address_space *, #endif /* nfs4proc.c */ +extern void __nfs4_read_done_cb(struct nfs_read_data *); extern void nfs4_reset_read(struct rpc_task *task, struct nfs_read_data *data); extern int nfs4_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index be79dc9..0bafcc9 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -30,6 +30,7 @@ */ #include <linux/nfs_fs.h> +#include <linux/nfs_page.h> #include "internal.h" #include "nfs4filelayout.h" @@ -421,6 +422,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, struct nfs4_deviceid *id, gfp_t gfp_flags) { + struct nfs4_deviceid_node *d; struct nfs4_file_layout_dsaddr *dsaddr; int status = -EINVAL; struct nfs_server *nfss = NFS_SERVER(lo->plh_inode); @@ -428,7 +430,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, dprintk("--> %s\n", __func__); if (fl->pattern_offset > lgr->range.offset) { - dprintk("%s pattern_offset %lld to large\n", + dprintk("%s pattern_offset %lld too large\n", __func__, fl->pattern_offset); goto out; } @@ -440,12 +442,14 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, } /* find and reference the deviceid */ - dsaddr = nfs4_fl_find_get_deviceid(id); - if (dsaddr == NULL) { + d = nfs4_find_get_deviceid(NFS_SERVER(lo->plh_inode)->pnfs_curr_ld, + NFS_SERVER(lo->plh_inode)->nfs_client, id); + if (d == NULL) { dsaddr = get_device_info(lo->plh_inode, id, gfp_flags); if (dsaddr == NULL) goto out; - } + } else + dsaddr = container_of(d, struct nfs4_file_layout_dsaddr, id_node); fl->dsaddr = dsaddr; if (fl->first_stripe_index < 0 || @@ -507,12 +511,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, gfp_t gfp_flags) { struct xdr_stream stream; - struct xdr_buf buf = { - .pages = lgr->layoutp->pages, - .page_len = lgr->layoutp->len, - .buflen = lgr->layoutp->len, - .len = lgr->layoutp->len, - }; + struct xdr_buf buf; struct page *scratch; __be32 *p; uint32_t nfl_util; @@ -524,7 +523,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, if (!scratch) return -ENOMEM; - xdr_init_decode(&stream, &buf, NULL); + xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, lgr->layoutp->len); xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); /* 20 = ufl_util (4), first_stripe_index (4), pattern_offset (8), @@ -535,7 +534,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, memcpy(id, p, sizeof(*id)); p += XDR_QUADLEN(NFS4_DEVICEID4_SIZE); - print_deviceid(id); + nfs4_print_deviceid(id); nfl_util = be32_to_cpup(p++); if (nfl_util & NFL4_UFLG_COMMIT_THRU_MDS) @@ -554,13 +553,18 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, __func__, nfl_util, fl->num_fh, fl->first_stripe_index, fl->pattern_offset); - if (!fl->num_fh) + /* Note that a zero value for num_fh is legal for STRIPE_SPARSE. + * Futher checking is done in filelayout_check_layout */ + if (fl->num_fh < 0 || fl->num_fh > + max(NFS4_PNFS_MAX_STRIPE_CNT, NFS4_PNFS_MAX_MULTI_CNT)) goto out_err; - fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), - gfp_flags); - if (!fl->fh_array) - goto out_err; + if (fl->num_fh > 0) { + fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), + gfp_flags); + if (!fl->fh_array) + goto out_err; + } for (i = 0; i < fl->num_fh; i++) { /* Do we want to use a mempool here? */ @@ -653,16 +657,20 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, /* * filelayout_pg_test(). Called by nfs_can_coalesce_requests() * - * return 1 : coalesce page - * return 0 : don't coalesce page + * return true : coalesce page + * return false : don't coalesce page */ -int +bool filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { u64 p_stripe, r_stripe; u32 stripe_unit; + if (!pnfs_generic_pg_test(pgio, prev, req) || + !nfs_generic_pg_test(pgio, prev, req)) + return false; + if (!pgio->pg_lseg) return 1; p_stripe = (u64)prev->wb_index << PAGE_CACHE_SHIFT; @@ -860,6 +868,12 @@ filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages, return -ENOMEM; } +static void +filelayout_free_deveiceid_node(struct nfs4_deviceid_node *d) +{ + nfs4_fl_free_deviceid(container_of(d, struct nfs4_file_layout_dsaddr, id_node)); +} + static struct pnfs_layoutdriver_type filelayout_type = { .id = LAYOUT_NFSV4_1_FILES, .name = "LAYOUT_NFSV4_1_FILES", @@ -872,6 +886,7 @@ static struct pnfs_layoutdriver_type filelayout_type = { .commit_pagelist = filelayout_commit_pagelist, .read_pagelist = filelayout_read_pagelist, .write_pagelist = filelayout_write_pagelist, + .free_deviceid_node = filelayout_free_deveiceid_node, }; static int __init nfs4filelayout_init(void) diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h index 2b461d7..cebe01e 100644 --- a/fs/nfs/nfs4filelayout.h +++ b/fs/nfs/nfs4filelayout.h @@ -59,9 +59,7 @@ struct nfs4_pnfs_ds { #define NFS4_DEVICE_ID_NEG_ENTRY 0x00000001 struct nfs4_file_layout_dsaddr { - struct hlist_node node; - struct nfs4_deviceid deviceid; - atomic_t ref; + struct nfs4_deviceid_node id_node; unsigned long flags; u32 stripe_count; u8 *stripe_indices; @@ -95,14 +93,12 @@ extern struct nfs_fh * nfs4_fl_select_ds_fh(struct pnfs_layout_segment *lseg, u32 j); extern void print_ds(struct nfs4_pnfs_ds *ds); -extern void print_deviceid(struct nfs4_deviceid *dev_id); u32 nfs4_fl_calc_j_index(struct pnfs_layout_segment *lseg, loff_t offset); u32 nfs4_fl_calc_ds_index(struct pnfs_layout_segment *lseg, u32 j); struct nfs4_pnfs_ds *nfs4_fl_prepare_ds(struct pnfs_layout_segment *lseg, u32 ds_idx); -extern struct nfs4_file_layout_dsaddr * -nfs4_fl_find_get_deviceid(struct nfs4_deviceid *dev_id); extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); +extern void nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); struct nfs4_file_layout_dsaddr * get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags); diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c index db07c7a..3b7bf13 100644 --- a/fs/nfs/nfs4filelayoutdev.c +++ b/fs/nfs/nfs4filelayoutdev.c @@ -37,30 +37,6 @@ #define NFSDBG_FACILITY NFSDBG_PNFS_LD /* - * Device ID RCU cache. A device ID is unique per client ID and layout type. - */ -#define NFS4_FL_DEVICE_ID_HASH_BITS 5 -#define NFS4_FL_DEVICE_ID_HASH_SIZE (1 << NFS4_FL_DEVICE_ID_HASH_BITS) -#define NFS4_FL_DEVICE_ID_HASH_MASK (NFS4_FL_DEVICE_ID_HASH_SIZE - 1) - -static inline u32 -nfs4_fl_deviceid_hash(struct nfs4_deviceid *id) -{ - unsigned char *cptr = (unsigned char *)id->data; - unsigned int nbytes = NFS4_DEVICEID4_SIZE; - u32 x = 0; - - while (nbytes--) { - x *= 37; - x += *cptr++; - } - return x & NFS4_FL_DEVICE_ID_HASH_MASK; -} - -static struct hlist_head filelayout_deviceid_cache[NFS4_FL_DEVICE_ID_HASH_SIZE]; -static DEFINE_SPINLOCK(filelayout_deviceid_lock); - -/* * Data server cache * * Data servers can be mapped to different device ids. @@ -89,27 +65,6 @@ print_ds(struct nfs4_pnfs_ds *ds) ds->ds_clp ? ds->ds_clp->cl_exchange_flags : 0); } -void -print_ds_list(struct nfs4_file_layout_dsaddr *dsaddr) -{ - int i; - - ifdebug(FACILITY) { - printk("%s dsaddr->ds_num %d\n", __func__, - dsaddr->ds_num); - for (i = 0; i < dsaddr->ds_num; i++) - print_ds(dsaddr->ds_list[i]); - } -} - -void print_deviceid(struct nfs4_deviceid *id) -{ - u32 *p = (u32 *)id; - - dprintk("%s: device id= [%x%x%x%x]\n", __func__, - p[0], p[1], p[2], p[3]); -} - /* nfs4_ds_cache_lock is held */ static struct nfs4_pnfs_ds * _data_server_lookup_locked(u32 ip_addr, u32 port) @@ -201,13 +156,13 @@ destroy_ds(struct nfs4_pnfs_ds *ds) kfree(ds); } -static void +void nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr) { struct nfs4_pnfs_ds *ds; int i; - print_deviceid(&dsaddr->deviceid); + nfs4_print_deviceid(&dsaddr->id_node.deviceid); for (i = 0; i < dsaddr->ds_num; i++) { ds = dsaddr->ds_list[i]; @@ -353,12 +308,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags) u8 max_stripe_index; struct nfs4_file_layout_dsaddr *dsaddr = NULL; struct xdr_stream stream; - struct xdr_buf buf = { - .pages = pdev->pages, - .page_len = pdev->pglen, - .buflen = pdev->pglen, - .len = pdev->pglen, - }; + struct xdr_buf buf; struct page *scratch; /* set up xdr stream */ @@ -366,7 +316,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags) if (!scratch) goto out_err; - xdr_init_decode(&stream, &buf, NULL); + xdr_init_decode_pages(&stream, &buf, pdev->pages, pdev->pglen); xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); /* Get the stripe count (number of stripe index) */ @@ -431,8 +381,10 @@ decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags) dsaddr->stripe_indices = stripe_indices; stripe_indices = NULL; dsaddr->ds_num = num; - - memcpy(&dsaddr->deviceid, &pdev->dev_id, sizeof(pdev->dev_id)); + nfs4_init_deviceid_node(&dsaddr->id_node, + NFS_SERVER(ino)->pnfs_curr_ld, + NFS_SERVER(ino)->nfs_client, + &pdev->dev_id); for (i = 0; i < dsaddr->ds_num; i++) { int j; @@ -505,8 +457,8 @@ out_err: static struct nfs4_file_layout_dsaddr * decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_flags) { - struct nfs4_file_layout_dsaddr *d, *new; - long hash; + struct nfs4_deviceid_node *d; + struct nfs4_file_layout_dsaddr *n, *new; new = decode_device(inode, dev, gfp_flags); if (!new) { @@ -515,20 +467,13 @@ decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_fl return NULL; } - spin_lock(&filelayout_deviceid_lock); - d = nfs4_fl_find_get_deviceid(&new->deviceid); - if (d) { - spin_unlock(&filelayout_deviceid_lock); + d = nfs4_insert_deviceid_node(&new->id_node); + n = container_of(d, struct nfs4_file_layout_dsaddr, id_node); + if (n != new) { nfs4_fl_free_deviceid(new); - return d; + return n; } - INIT_HLIST_NODE(&new->node); - atomic_set(&new->ref, 1); - hash = nfs4_fl_deviceid_hash(&new->deviceid); - hlist_add_head_rcu(&new->node, &filelayout_deviceid_cache[hash]); - spin_unlock(&filelayout_deviceid_lock); - return new; } @@ -600,35 +545,7 @@ out_free: void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr) { - if (atomic_dec_and_lock(&dsaddr->ref, &filelayout_deviceid_lock)) { - hlist_del_rcu(&dsaddr->node); - spin_unlock(&filelayout_deviceid_lock); - - synchronize_rcu(); - nfs4_fl_free_deviceid(dsaddr); - } -} - -struct nfs4_file_layout_dsaddr * -nfs4_fl_find_get_deviceid(struct nfs4_deviceid *id) -{ - struct nfs4_file_layout_dsaddr *d; - struct hlist_node *n; - long hash = nfs4_fl_deviceid_hash(id); - - - rcu_read_lock(); - hlist_for_each_entry_rcu(d, n, &filelayout_deviceid_cache[hash], node) { - if (!memcmp(&d->deviceid, id, sizeof(*id))) { - if (!atomic_inc_not_zero(&d->ref)) - goto fail; - rcu_read_unlock(); - return d; - } - } -fail: - rcu_read_unlock(); - return NULL; + nfs4_put_deviceid_node(&dsaddr->id_node); } /* @@ -676,15 +593,15 @@ static void filelayout_mark_devid_negative(struct nfs4_file_layout_dsaddr *dsaddr, int err, u32 ds_addr) { - u32 *p = (u32 *)&dsaddr->deviceid; + u32 *p = (u32 *)&dsaddr->id_node.deviceid; printk(KERN_ERR "NFS: data server %x connection error %d." " Deviceid [%x%x%x%x] marked out of use.\n", ds_addr, err, p[0], p[1], p[2], p[3]); - spin_lock(&filelayout_deviceid_lock); + spin_lock(&nfs4_ds_cache_lock); dsaddr->flags |= NFS4_DEVICE_ID_NEG_ENTRY; - spin_unlock(&filelayout_deviceid_lock); + spin_unlock(&nfs4_ds_cache_lock); } struct nfs4_pnfs_ds * diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index cf1b339..5879b23 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -267,9 +267,11 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc break; nfs4_schedule_stateid_recovery(server, state); goto wait_on_recovery; + case -NFS4ERR_EXPIRED: + if (state != NULL) + nfs4_schedule_stateid_recovery(server, state); case -NFS4ERR_STALE_STATEID: case -NFS4ERR_STALE_CLIENTID: - case -NFS4ERR_EXPIRED: nfs4_schedule_lease_recovery(clp); goto wait_on_recovery; #if defined(CONFIG_NFS_V4_1) @@ -2263,12 +2265,14 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, return nfs4_map_errors(status); } +static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr); /* * Get locations and (maybe) other attributes of a referral. * Note that we'll actually follow the referral later when * we detect fsid mismatch in inode revalidation */ -static int nfs4_get_referral(struct inode *dir, const struct qstr *name, struct nfs_fattr *fattr, struct nfs_fh *fhandle) +static int nfs4_get_referral(struct inode *dir, const struct qstr *name, + struct nfs_fattr *fattr, struct nfs_fh *fhandle) { int status = -ENOMEM; struct page *page = NULL; @@ -2286,15 +2290,16 @@ static int nfs4_get_referral(struct inode *dir, const struct qstr *name, struct goto out; /* Make sure server returned a different fsid for the referral */ if (nfs_fsid_equal(&NFS_SERVER(dir)->fsid, &locations->fattr.fsid)) { - dprintk("%s: server did not return a different fsid for a referral at %s\n", __func__, name->name); + dprintk("%s: server did not return a different fsid for" + " a referral at %s\n", __func__, name->name); status = -EIO; goto out; } + /* Fixup attributes for the nfs_lookup() call to nfs_fhget() */ + nfs_fixup_referral_attributes(&locations->fattr); + /* replace the lookup nfs_fattr with the locations nfs_fattr */ memcpy(fattr, &locations->fattr, sizeof(struct nfs_fattr)); - fattr->valid |= NFS_ATTR_FATTR_V4_REFERRAL; - if (!fattr->mode) - fattr->mode = S_IFDIR; memset(fhandle, 0, sizeof(struct nfs_fh)); out: if (page) @@ -2361,6 +2366,9 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, struct nfs4_state *state = NULL; int status; + if (pnfs_ld_layoutret_on_setattr(inode)) + pnfs_return_layout(inode); + nfs_fattr_init(fattr); /* Search for an existing open(O_WRITE) file */ @@ -3175,6 +3183,11 @@ static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return err; } +void __nfs4_read_done_cb(struct nfs_read_data *data) +{ + nfs_invalidate_atime(data->inode); +} + static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) { struct nfs_server *server = NFS_SERVER(data->inode); @@ -3184,7 +3197,7 @@ static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) return -EAGAIN; } - nfs_invalidate_atime(data->inode); + __nfs4_read_done_cb(data); if (task->tk_status > 0) renew_lease(server, data->timestamp); return 0; @@ -3198,7 +3211,8 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) if (!nfs4_sequence_done(task, &data->res.seq_res)) return -EAGAIN; - return data->read_done_cb(task, data); + return data->read_done_cb ? data->read_done_cb(task, data) : + nfs4_read_done_cb(task, data); } static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) @@ -3243,7 +3257,8 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) { if (!nfs4_sequence_done(task, &data->res.seq_res)) return -EAGAIN; - return data->write_done_cb(task, data); + return data->write_done_cb ? data->write_done_cb(task, data) : + nfs4_write_done_cb(task, data); } /* Reset the the nfs_write_data to send the write to the MDS. */ @@ -3670,9 +3685,11 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, break; nfs4_schedule_stateid_recovery(server, state); goto wait_on_recovery; + case -NFS4ERR_EXPIRED: + if (state != NULL) + nfs4_schedule_stateid_recovery(server, state); case -NFS4ERR_STALE_STATEID: case -NFS4ERR_STALE_CLIENTID: - case -NFS4ERR_EXPIRED: nfs4_schedule_lease_recovery(clp); goto wait_on_recovery; #if defined(CONFIG_NFS_V4_1) @@ -4543,6 +4560,7 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl) case -ESTALE: goto out; case -NFS4ERR_EXPIRED: + nfs4_schedule_stateid_recovery(server, state); case -NFS4ERR_STALE_CLIENTID: case -NFS4ERR_STALE_STATEID: nfs4_schedule_lease_recovery(server->nfs_client); @@ -4652,11 +4670,15 @@ static size_t nfs4_xattr_list_nfs4_acl(struct dentry *dentry, char *list, return len; } +/* + * nfs_fhget will use either the mounted_on_fileid or the fileid + */ static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr) { - if (!((fattr->valid & NFS_ATTR_FATTR_FILEID) && - (fattr->valid & NFS_ATTR_FATTR_FSID) && - (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL))) + if (!(((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) || + (fattr->valid & NFS_ATTR_FATTR_FILEID)) && + (fattr->valid & NFS_ATTR_FATTR_FSID) && + (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL))) return; fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE | @@ -4671,7 +4693,6 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs_server *server = NFS_SERVER(dir); u32 bitmask[2] = { [0] = FATTR4_WORD0_FSID | FATTR4_WORD0_FS_LOCATIONS, - [1] = FATTR4_WORD1_MOUNTED_ON_FILEID, }; struct nfs4_fs_locations_arg args = { .dir_fh = NFS_FH(dir), @@ -4690,11 +4711,18 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, int status; dprintk("%s: start\n", __func__); + + /* Ask for the fileid of the absent filesystem if mounted_on_fileid + * is not supported */ + if (NFS_SERVER(dir)->attr_bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID) + bitmask[1] |= FATTR4_WORD1_MOUNTED_ON_FILEID; + else + bitmask[0] |= FATTR4_WORD0_FILEID; + nfs_fattr_init(&fs_locations->fattr); fs_locations->server = server; fs_locations->nlocations = 0; status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); - nfs_fixup_referral_attributes(&fs_locations->fattr); dprintk("%s: returned status = %d\n", __func__, status); return status; } @@ -5083,7 +5111,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) if (mxresp_sz == 0) mxresp_sz = NFS_MAX_FILE_IO_SIZE; /* Fore channel attributes */ - args->fc_attrs.headerpadsz = 0; args->fc_attrs.max_rqst_sz = mxrqst_sz; args->fc_attrs.max_resp_sz = mxresp_sz; args->fc_attrs.max_ops = NFS4_MAX_OPS; @@ -5096,7 +5123,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) args->fc_attrs.max_ops, args->fc_attrs.max_reqs); /* Back channel attributes */ - args->bc_attrs.headerpadsz = 0; args->bc_attrs.max_rqst_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz_cached = 0; @@ -5116,8 +5142,6 @@ static int nfs4_verify_fore_channel_attrs(struct nfs41_create_session_args *args struct nfs4_channel_attrs *sent = &args->fc_attrs; struct nfs4_channel_attrs *rcvd = &session->fc_attrs; - if (rcvd->headerpadsz > sent->headerpadsz) - return -EINVAL; if (rcvd->max_resp_sz > sent->max_resp_sz) return -EINVAL; /* @@ -5666,6 +5690,88 @@ int nfs4_proc_layoutget(struct nfs4_layoutget *lgp) return status; } +static void +nfs4_layoutreturn_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + + dprintk("--> %s\n", __func__); + if (nfs41_setup_sequence(lrp->clp->cl_session, &lrp->args.seq_args, + &lrp->res.seq_res, 0, task)) + return; + rpc_call_start(task); +} + +static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + struct nfs_server *server; + struct pnfs_layout_hdr *lo = NFS_I(lrp->args.inode)->layout; + + dprintk("--> %s\n", __func__); + + if (!nfs4_sequence_done(task, &lrp->res.seq_res)) + return; + + server = NFS_SERVER(lrp->args.inode); + if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) { + nfs_restart_rpc(task, lrp->clp); + return; + } + spin_lock(&lo->plh_inode->i_lock); + if (task->tk_status == 0) { + if (lrp->res.lrs_present) { + pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); + } else + BUG_ON(!list_empty(&lo->plh_segs)); + } + lo->plh_block_lgets--; + spin_unlock(&lo->plh_inode->i_lock); + dprintk("<-- %s\n", __func__); +} + +static void nfs4_layoutreturn_release(void *calldata) +{ + struct nfs4_layoutreturn *lrp = calldata; + + dprintk("--> %s\n", __func__); + put_layout_hdr(NFS_I(lrp->args.inode)->layout); + kfree(calldata); + dprintk("<-- %s\n", __func__); +} + +static const struct rpc_call_ops nfs4_layoutreturn_call_ops = { + .rpc_call_prepare = nfs4_layoutreturn_prepare, + .rpc_call_done = nfs4_layoutreturn_done, + .rpc_release = nfs4_layoutreturn_release, +}; + +int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp) +{ + struct rpc_task *task; + struct rpc_message msg = { + .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTRETURN], + .rpc_argp = &lrp->args, + .rpc_resp = &lrp->res, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = lrp->clp->cl_rpcclient, + .rpc_message = &msg, + .callback_ops = &nfs4_layoutreturn_call_ops, + .callback_data = lrp, + }; + int status; + + dprintk("--> %s\n", __func__); + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) + return PTR_ERR(task); + status = task->tk_status; + dprintk("<-- %s status=%d\n", __func__, status); + rpc_put_task(task); + return status; +} + static int _nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *pdev) { diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 036f5ad..e97dd21 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1466,7 +1466,10 @@ static int nfs4_reclaim_lease(struct nfs_client *clp) #ifdef CONFIG_NFS_V4_1 void nfs4_schedule_session_recovery(struct nfs4_session *session) { - nfs4_schedule_lease_recovery(session->clp); + struct nfs_client *clp = session->clp; + + set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); + nfs4_schedule_lease_recovery(clp); } EXPORT_SYMBOL_GPL(nfs4_schedule_session_recovery); @@ -1549,6 +1552,7 @@ static int nfs4_reset_session(struct nfs_client *clp) status = nfs4_recovery_handle_error(clp, status); goto out; } + clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state); /* create_session negotiated new slot table */ clear_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index c3ccd2c..6870bc6 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -255,7 +255,7 @@ static int nfs4_stat_to_errno(int); #define decode_fs_locations_maxsz \ (0) #define encode_secinfo_maxsz (op_encode_hdr_maxsz + nfs4_name_maxsz) -#define decode_secinfo_maxsz (op_decode_hdr_maxsz + 4 + (NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN))) +#define decode_secinfo_maxsz (op_decode_hdr_maxsz + 1 + ((NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN)) / 4)) #if defined(CONFIG_NFS_V4_1) #define NFS4_MAX_MACHINE_NAME_LEN (64) @@ -338,7 +338,11 @@ static int nfs4_stat_to_errno(int); 1 /* layoutupdate4 layout type */ + \ 1 /* NULL filelayout layoutupdate4 payload */) #define decode_layoutcommit_maxsz (op_decode_hdr_maxsz + 3) - +#define encode_layoutreturn_maxsz (8 + op_encode_hdr_maxsz + \ + encode_stateid_maxsz + \ + 1 /* FIXME: opaque lrf_body always empty at the moment */) +#define decode_layoutreturn_maxsz (op_decode_hdr_maxsz + \ + 1 + decode_stateid_maxsz) #else /* CONFIG_NFS_V4_1 */ #define encode_sequence_maxsz 0 #define decode_sequence_maxsz 0 @@ -760,7 +764,14 @@ static int nfs4_stat_to_errno(int); decode_putfh_maxsz + \ decode_layoutcommit_maxsz + \ decode_getattr_maxsz) - +#define NFS4_enc_layoutreturn_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_layoutreturn_maxsz) +#define NFS4_dec_layoutreturn_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_layoutreturn_maxsz) const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH + compound_encode_hdr_maxsz + @@ -1714,7 +1725,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(args->flags); /*flags */ /* Fore Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->fc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->fc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(max_resp_sz_cached); /* Max resp sz cached */ @@ -1723,7 +1734,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(0); /* rdmachannel_attrs */ /* Back Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->bc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz_cached); /* Max resp sz cached */ @@ -1864,6 +1875,7 @@ encode_layoutget(struct xdr_stream *xdr, static int encode_layoutcommit(struct xdr_stream *xdr, + struct inode *inode, const struct nfs4_layoutcommit_args *args, struct compound_hdr *hdr) { @@ -1872,7 +1884,7 @@ encode_layoutcommit(struct xdr_stream *xdr, dprintk("%s: lbw: %llu type: %d\n", __func__, args->lastbytewritten, NFS_SERVER(args->inode)->pnfs_curr_ld->id); - p = reserve_space(xdr, 48 + NFS4_STATEID_SIZE); + p = reserve_space(xdr, 44 + NFS4_STATEID_SIZE); *p++ = cpu_to_be32(OP_LAYOUTCOMMIT); /* Only whole file layouts */ p = xdr_encode_hyper(p, 0); /* offset */ @@ -1883,12 +1895,49 @@ encode_layoutcommit(struct xdr_stream *xdr, p = xdr_encode_hyper(p, args->lastbytewritten); *p++ = cpu_to_be32(0); /* Never send time_modify_changed */ *p++ = cpu_to_be32(NFS_SERVER(args->inode)->pnfs_curr_ld->id);/* type */ - *p++ = cpu_to_be32(0); /* no file layout payload */ + + if (NFS_SERVER(inode)->pnfs_curr_ld->encode_layoutcommit) + NFS_SERVER(inode)->pnfs_curr_ld->encode_layoutcommit( + NFS_I(inode)->layout, xdr, args); + else { + p = reserve_space(xdr, 4); + *p = cpu_to_be32(0); /* no layout-type payload */ + } hdr->nops++; hdr->replen += decode_layoutcommit_maxsz; return 0; } + +static void +encode_layoutreturn(struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args, + struct compound_hdr *hdr) +{ + __be32 *p; + + p = reserve_space(xdr, 20); + *p++ = cpu_to_be32(OP_LAYOUTRETURN); + *p++ = cpu_to_be32(0); /* reclaim. always 0 for now */ + *p++ = cpu_to_be32(args->layout_type); + *p++ = cpu_to_be32(IOMODE_ANY); + *p = cpu_to_be32(RETURN_FILE); + p = reserve_space(xdr, 16 + NFS4_STATEID_SIZE); + p = xdr_encode_hyper(p, 0); + p = xdr_encode_hyper(p, NFS4_MAX_UINT64); + spin_lock(&args->inode->i_lock); + xdr_encode_opaque_fixed(p, &args->stateid.data, NFS4_STATEID_SIZE); + spin_unlock(&args->inode->i_lock); + if (NFS_SERVER(args->inode)->pnfs_curr_ld->encode_layoutreturn) { + NFS_SERVER(args->inode)->pnfs_curr_ld->encode_layoutreturn( + NFS_I(args->inode)->layout, xdr, args); + } else { + p = reserve_space(xdr, 4); + *p = cpu_to_be32(0); + } + hdr->nops++; + hdr->replen += decode_layoutreturn_maxsz; +} #endif /* CONFIG_NFS_V4_1 */ /* @@ -2706,10 +2755,12 @@ static void nfs4_xdr_enc_layoutget(struct rpc_rqst *req, /* * Encode LAYOUTCOMMIT request */ -static int nfs4_xdr_enc_layoutcommit(struct rpc_rqst *req, - struct xdr_stream *xdr, - struct nfs4_layoutcommit_args *args) +static void nfs4_xdr_enc_layoutcommit(struct rpc_rqst *req, + struct xdr_stream *xdr, + struct nfs4_layoutcommit_args *args) { + struct nfs4_layoutcommit_data *data = + container_of(args, struct nfs4_layoutcommit_data, args); struct compound_hdr hdr = { .minorversion = nfs4_xdr_minorversion(&args->seq_args), }; @@ -2717,10 +2768,27 @@ static int nfs4_xdr_enc_layoutcommit(struct rpc_rqst *req, encode_compound_hdr(xdr, req, &hdr); encode_sequence(xdr, &args->seq_args, &hdr); encode_putfh(xdr, NFS_FH(args->inode), &hdr); - encode_layoutcommit(xdr, args, &hdr); + encode_layoutcommit(xdr, data->args.inode, args, &hdr); encode_getfattr(xdr, args->bitmask, &hdr); encode_nops(&hdr); - return 0; +} + +/* + * Encode LAYOUTRETURN request + */ +static void nfs4_xdr_enc_layoutreturn(struct rpc_rqst *req, + struct xdr_stream *xdr, + struct nfs4_layoutreturn_args *args) +{ + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, NFS_FH(args->inode), &hdr); + encode_layoutreturn(xdr, args, &hdr); + encode_nops(&hdr); } #endif /* CONFIG_NFS_V4_1 */ @@ -3030,7 +3098,7 @@ out_overflow: return -EIO; } -static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap) +static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap, int32_t *res) { __be32 *p; @@ -3041,7 +3109,7 @@ static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap) if (unlikely(!p)) goto out_overflow; bitmap[0] &= ~FATTR4_WORD0_RDATTR_ERROR; - return -be32_to_cpup(p); + *res = -be32_to_cpup(p); } return 0; out_overflow: @@ -4002,6 +4070,7 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, int status; umode_t fmode = 0; uint32_t type; + int32_t err; status = decode_attr_type(xdr, bitmap, &type); if (status < 0) @@ -4027,13 +4096,12 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, goto xdr_error; fattr->valid |= status; - status = decode_attr_error(xdr, bitmap); - if (status == -NFS4ERR_WRONGSEC) { - nfs_fixup_secinfo_attributes(fattr, fh); - status = 0; - } + err = 0; + status = decode_attr_error(xdr, bitmap, &err); if (status < 0) goto xdr_error; + if (err == -NFS4ERR_WRONGSEC) + nfs_fixup_secinfo_attributes(fattr, fh); status = decode_attr_filehandle(xdr, bitmap, fh); if (status < 0) @@ -4929,12 +4997,14 @@ static int decode_chan_attrs(struct xdr_stream *xdr, struct nfs4_channel_attrs *attrs) { __be32 *p; - u32 nr_attrs; + u32 nr_attrs, val; p = xdr_inline_decode(xdr, 28); if (unlikely(!p)) goto out_overflow; - attrs->headerpadsz = be32_to_cpup(p++); + val = be32_to_cpup(p++); /* headerpadsz */ + if (val) + return -EINVAL; /* no support for header padding yet */ attrs->max_rqst_sz = be32_to_cpup(p++); attrs->max_resp_sz = be32_to_cpup(p++); attrs->max_resp_sz_cached = be32_to_cpup(p++); @@ -5203,6 +5273,27 @@ out_overflow: return -EIO; } +static int decode_layoutreturn(struct xdr_stream *xdr, + struct nfs4_layoutreturn_res *res) +{ + __be32 *p; + int status; + + status = decode_op_hdr(xdr, OP_LAYOUTRETURN); + if (status) + return status; + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + goto out_overflow; + res->lrs_present = be32_to_cpup(p); + if (res->lrs_present) + status = decode_stateid(xdr, &res->stateid); + return status; +out_overflow: + print_overflow_msg(__func__, xdr); + return -EIO; +} + static int decode_layoutcommit(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs4_layoutcommit_res *res) @@ -6320,6 +6411,30 @@ out: } /* + * Decode LAYOUTRETURN response + */ +static int nfs4_xdr_dec_layoutreturn(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + struct nfs4_layoutreturn_res *res) +{ + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_layoutreturn(xdr, res); +out: + return status; +} + +/* * Decode LAYOUTCOMMIT response */ static int nfs4_xdr_dec_layoutcommit(struct rpc_rqst *rqstp, @@ -6547,6 +6662,7 @@ struct rpc_procinfo nfs4_procedures[] = { PROC(GETDEVICEINFO, enc_getdeviceinfo, dec_getdeviceinfo), PROC(LAYOUTGET, enc_layoutget, dec_layoutget), PROC(LAYOUTCOMMIT, enc_layoutcommit, dec_layoutcommit), + PROC(LAYOUTRETURN, enc_layoutreturn, dec_layoutreturn), #endif /* CONFIG_NFS_V4_1 */ }; diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index c541093..c4744e1 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -87,7 +87,7 @@ #define NFS_ROOT "/tftpboot/%s" /* Default NFSROOT mount options. */ -#define NFS_DEF_OPTIONS "udp" +#define NFS_DEF_OPTIONS "vers=2,udp,rsize=4096,wsize=4096" /* Parameters passed from the kernel command line */ static char nfs_root_parms[256] __initdata = ""; diff --git a/fs/nfs/objlayout/Kbuild b/fs/nfs/objlayout/Kbuild new file mode 100644 index 0000000..ed30ea0 --- /dev/null +++ b/fs/nfs/objlayout/Kbuild @@ -0,0 +1,5 @@ +# +# Makefile for the pNFS Objects Layout Driver kernel module +# +objlayoutdriver-y := objio_osd.o pnfs_osd_xdr_cli.o objlayout.o +obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayoutdriver.o diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c new file mode 100644 index 0000000..8ff2ea3 --- /dev/null +++ b/fs/nfs/objlayout/objio_osd.c @@ -0,0 +1,1059 @@ +/* + * pNFS Objects layout implementation over open-osd initiator library + * + * Copyright (C) 2009 Panasas Inc. [year of first publication] + * All rights reserved. + * + * Benny Halevy <bhalevy@panasas.com> + * Boaz Harrosh <bharrosh@panasas.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * See the file COPYING included with this distribution for more details. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the Panasas company nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/module.h> +#include <scsi/osd_initiator.h> + +#include "objlayout.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +#define _LLU(x) ((unsigned long long)x) + +enum { BIO_MAX_PAGES_KMALLOC = + (PAGE_SIZE - sizeof(struct bio)) / sizeof(struct bio_vec), +}; + +struct objio_dev_ent { + struct nfs4_deviceid_node id_node; + struct osd_dev *od; +}; + +static void +objio_free_deviceid_node(struct nfs4_deviceid_node *d) +{ + struct objio_dev_ent *de = container_of(d, struct objio_dev_ent, id_node); + + dprintk("%s: free od=%p\n", __func__, de->od); + osduld_put_device(de->od); + kfree(de); +} + +static struct objio_dev_ent *_dev_list_find(const struct nfs_server *nfss, + const struct nfs4_deviceid *d_id) +{ + struct nfs4_deviceid_node *d; + struct objio_dev_ent *de; + + d = nfs4_find_get_deviceid(nfss->pnfs_curr_ld, nfss->nfs_client, d_id); + if (!d) + return NULL; + + de = container_of(d, struct objio_dev_ent, id_node); + return de; +} + +static struct objio_dev_ent * +_dev_list_add(const struct nfs_server *nfss, + const struct nfs4_deviceid *d_id, struct osd_dev *od, + gfp_t gfp_flags) +{ + struct nfs4_deviceid_node *d; + struct objio_dev_ent *de = kzalloc(sizeof(*de), gfp_flags); + struct objio_dev_ent *n; + + if (!de) { + dprintk("%s: -ENOMEM od=%p\n", __func__, od); + return NULL; + } + + dprintk("%s: Adding od=%p\n", __func__, od); + nfs4_init_deviceid_node(&de->id_node, + nfss->pnfs_curr_ld, + nfss->nfs_client, + d_id); + de->od = od; + + d = nfs4_insert_deviceid_node(&de->id_node); + n = container_of(d, struct objio_dev_ent, id_node); + if (n != de) { + dprintk("%s: Race with other n->od=%p\n", __func__, n->od); + objio_free_deviceid_node(&de->id_node); + de = n; + } + + return de; +} + +struct caps_buffers { + u8 caps_key[OSD_CRYPTO_KEYID_SIZE]; + u8 creds[OSD_CAP_LEN]; +}; + +struct objio_segment { + struct pnfs_layout_segment lseg; + + struct pnfs_osd_object_cred *comps; + + unsigned mirrors_p1; + unsigned stripe_unit; + unsigned group_width; /* Data stripe_units without integrity comps */ + u64 group_depth; + unsigned group_count; + + unsigned max_io_size; + + unsigned comps_index; + unsigned num_comps; + /* variable length */ + struct objio_dev_ent *ods[]; +}; + +static inline struct objio_segment * +OBJIO_LSEG(struct pnfs_layout_segment *lseg) +{ + return container_of(lseg, struct objio_segment, lseg); +} + +struct objio_state; +typedef ssize_t (*objio_done_fn)(struct objio_state *ios); + +struct objio_state { + /* Generic layer */ + struct objlayout_io_state ol_state; + + struct objio_segment *layout; + + struct kref kref; + objio_done_fn done; + void *private; + + unsigned long length; + unsigned numdevs; /* Actually used devs in this IO */ + /* A per-device variable array of size numdevs */ + struct _objio_per_comp { + struct bio *bio; + struct osd_request *or; + unsigned long length; + u64 offset; + unsigned dev; + } per_dev[]; +}; + +/* Send and wait for a get_device_info of devices in the layout, + then look them up with the osd_initiator library */ +static struct objio_dev_ent *_device_lookup(struct pnfs_layout_hdr *pnfslay, + struct objio_segment *objio_seg, unsigned comp, + gfp_t gfp_flags) +{ + struct pnfs_osd_deviceaddr *deviceaddr; + struct nfs4_deviceid *d_id; + struct objio_dev_ent *ode; + struct osd_dev *od; + struct osd_dev_info odi; + int err; + + d_id = &objio_seg->comps[comp].oc_object_id.oid_device_id; + + ode = _dev_list_find(NFS_SERVER(pnfslay->plh_inode), d_id); + if (ode) + return ode; + + err = objlayout_get_deviceinfo(pnfslay, d_id, &deviceaddr, gfp_flags); + if (unlikely(err)) { + dprintk("%s: objlayout_get_deviceinfo dev(%llx:%llx) =>%d\n", + __func__, _DEVID_LO(d_id), _DEVID_HI(d_id), err); + return ERR_PTR(err); + } + + odi.systemid_len = deviceaddr->oda_systemid.len; + if (odi.systemid_len > sizeof(odi.systemid)) { + err = -EINVAL; + goto out; + } else if (odi.systemid_len) + memcpy(odi.systemid, deviceaddr->oda_systemid.data, + odi.systemid_len); + odi.osdname_len = deviceaddr->oda_osdname.len; + odi.osdname = (u8 *)deviceaddr->oda_osdname.data; + + if (!odi.osdname_len && !odi.systemid_len) { + dprintk("%s: !odi.osdname_len && !odi.systemid_len\n", + __func__); + err = -ENODEV; + goto out; + } + + od = osduld_info_lookup(&odi); + if (unlikely(IS_ERR(od))) { + err = PTR_ERR(od); + dprintk("%s: osduld_info_lookup => %d\n", __func__, err); + goto out; + } + + ode = _dev_list_add(NFS_SERVER(pnfslay->plh_inode), d_id, od, + gfp_flags); + +out: + dprintk("%s: return=%d\n", __func__, err); + objlayout_put_deviceinfo(deviceaddr); + return err ? ERR_PTR(err) : ode; +} + +static int objio_devices_lookup(struct pnfs_layout_hdr *pnfslay, + struct objio_segment *objio_seg, + gfp_t gfp_flags) +{ + unsigned i; + int err; + + /* lookup all devices */ + for (i = 0; i < objio_seg->num_comps; i++) { + struct objio_dev_ent *ode; + + ode = _device_lookup(pnfslay, objio_seg, i, gfp_flags); + if (unlikely(IS_ERR(ode))) { + err = PTR_ERR(ode); + goto out; + } + objio_seg->ods[i] = ode; + } + err = 0; + +out: + dprintk("%s: return=%d\n", __func__, err); + return err; +} + +static int _verify_data_map(struct pnfs_osd_layout *layout) +{ + struct pnfs_osd_data_map *data_map = &layout->olo_map; + u64 stripe_length; + u32 group_width; + +/* FIXME: Only raid0 for now. if not go through MDS */ + if (data_map->odm_raid_algorithm != PNFS_OSD_RAID_0) { + printk(KERN_ERR "Only RAID_0 for now\n"); + return -ENOTSUPP; + } + if (0 != (data_map->odm_num_comps % (data_map->odm_mirror_cnt + 1))) { + printk(KERN_ERR "Data Map wrong, num_comps=%u mirrors=%u\n", + data_map->odm_num_comps, data_map->odm_mirror_cnt); + return -EINVAL; + } + + if (data_map->odm_group_width) + group_width = data_map->odm_group_width; + else + group_width = data_map->odm_num_comps / + (data_map->odm_mirror_cnt + 1); + + stripe_length = (u64)data_map->odm_stripe_unit * group_width; + if (stripe_length >= (1ULL << 32)) { + printk(KERN_ERR "Total Stripe length(0x%llx)" + " >= 32bit is not supported\n", _LLU(stripe_length)); + return -ENOTSUPP; + } + + if (0 != (data_map->odm_stripe_unit & ~PAGE_MASK)) { + printk(KERN_ERR "Stripe Unit(0x%llx)" + " must be Multples of PAGE_SIZE(0x%lx)\n", + _LLU(data_map->odm_stripe_unit), PAGE_SIZE); + return -ENOTSUPP; + } + + return 0; +} + +static void copy_single_comp(struct pnfs_osd_object_cred *cur_comp, + struct pnfs_osd_object_cred *src_comp, + struct caps_buffers *caps_p) +{ + WARN_ON(src_comp->oc_cap_key.cred_len > sizeof(caps_p->caps_key)); + WARN_ON(src_comp->oc_cap.cred_len > sizeof(caps_p->creds)); + + *cur_comp = *src_comp; + + memcpy(caps_p->caps_key, src_comp->oc_cap_key.cred, + sizeof(caps_p->caps_key)); + cur_comp->oc_cap_key.cred = caps_p->caps_key; + + memcpy(caps_p->creds, src_comp->oc_cap.cred, + sizeof(caps_p->creds)); + cur_comp->oc_cap.cred = caps_p->creds; +} + +int objio_alloc_lseg(struct pnfs_layout_segment **outp, + struct pnfs_layout_hdr *pnfslay, + struct pnfs_layout_range *range, + struct xdr_stream *xdr, + gfp_t gfp_flags) +{ + struct objio_segment *objio_seg; + struct pnfs_osd_xdr_decode_layout_iter iter; + struct pnfs_osd_layout layout; + struct pnfs_osd_object_cred *cur_comp, src_comp; + struct caps_buffers *caps_p; + int err; + + err = pnfs_osd_xdr_decode_layout_map(&layout, &iter, xdr); + if (unlikely(err)) + return err; + + err = _verify_data_map(&layout); + if (unlikely(err)) + return err; + + objio_seg = kzalloc(sizeof(*objio_seg) + + sizeof(objio_seg->ods[0]) * layout.olo_num_comps + + sizeof(*objio_seg->comps) * layout.olo_num_comps + + sizeof(struct caps_buffers) * layout.olo_num_comps, + gfp_flags); + if (!objio_seg) + return -ENOMEM; + + objio_seg->comps = (void *)(objio_seg->ods + layout.olo_num_comps); + cur_comp = objio_seg->comps; + caps_p = (void *)(cur_comp + layout.olo_num_comps); + while (pnfs_osd_xdr_decode_layout_comp(&src_comp, &iter, xdr, &err)) + copy_single_comp(cur_comp++, &src_comp, caps_p++); + if (unlikely(err)) + goto err; + + objio_seg->num_comps = layout.olo_num_comps; + objio_seg->comps_index = layout.olo_comps_index; + err = objio_devices_lookup(pnfslay, objio_seg, gfp_flags); + if (err) + goto err; + + objio_seg->mirrors_p1 = layout.olo_map.odm_mirror_cnt + 1; + objio_seg->stripe_unit = layout.olo_map.odm_stripe_unit; + if (layout.olo_map.odm_group_width) { + objio_seg->group_width = layout.olo_map.odm_group_width; + objio_seg->group_depth = layout.olo_map.odm_group_depth; + objio_seg->group_count = layout.olo_map.odm_num_comps / + objio_seg->mirrors_p1 / + objio_seg->group_width; + } else { + objio_seg->group_width = layout.olo_map.odm_num_comps / + objio_seg->mirrors_p1; + objio_seg->group_depth = -1; + objio_seg->group_count = 1; + } + + /* Cache this calculation it will hit for every page */ + objio_seg->max_io_size = (BIO_MAX_PAGES_KMALLOC * PAGE_SIZE - + objio_seg->stripe_unit) * + objio_seg->group_width; + + *outp = &objio_seg->lseg; + return 0; + +err: + kfree(objio_seg); + dprintk("%s: Error: return %d\n", __func__, err); + *outp = NULL; + return err; +} + +void objio_free_lseg(struct pnfs_layout_segment *lseg) +{ + int i; + struct objio_segment *objio_seg = OBJIO_LSEG(lseg); + + for (i = 0; i < objio_seg->num_comps; i++) { + if (!objio_seg->ods[i]) + break; + nfs4_put_deviceid_node(&objio_seg->ods[i]->id_node); + } + kfree(objio_seg); +} + +int objio_alloc_io_state(struct pnfs_layout_segment *lseg, + struct objlayout_io_state **outp, + gfp_t gfp_flags) +{ + struct objio_segment *objio_seg = OBJIO_LSEG(lseg); + struct objio_state *ios; + const unsigned first_size = sizeof(*ios) + + objio_seg->num_comps * sizeof(ios->per_dev[0]); + const unsigned sec_size = objio_seg->num_comps * + sizeof(ios->ol_state.ioerrs[0]); + + ios = kzalloc(first_size + sec_size, gfp_flags); + if (unlikely(!ios)) + return -ENOMEM; + + ios->layout = objio_seg; + ios->ol_state.ioerrs = ((void *)ios) + first_size; + ios->ol_state.num_comps = objio_seg->num_comps; + + *outp = &ios->ol_state; + return 0; +} + +void objio_free_io_state(struct objlayout_io_state *ol_state) +{ + struct objio_state *ios = container_of(ol_state, struct objio_state, + ol_state); + + kfree(ios); +} + +enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep) +{ + switch (oep) { + case OSD_ERR_PRI_NO_ERROR: + return (enum pnfs_osd_errno)0; + + case OSD_ERR_PRI_CLEAR_PAGES: + BUG_ON(1); + return 0; + + case OSD_ERR_PRI_RESOURCE: + return PNFS_OSD_ERR_RESOURCE; + case OSD_ERR_PRI_BAD_CRED: + return PNFS_OSD_ERR_BAD_CRED; + case OSD_ERR_PRI_NO_ACCESS: + return PNFS_OSD_ERR_NO_ACCESS; + case OSD_ERR_PRI_UNREACHABLE: + return PNFS_OSD_ERR_UNREACHABLE; + case OSD_ERR_PRI_NOT_FOUND: + return PNFS_OSD_ERR_NOT_FOUND; + case OSD_ERR_PRI_NO_SPACE: + return PNFS_OSD_ERR_NO_SPACE; + default: + WARN_ON(1); + /* fallthrough */ + case OSD_ERR_PRI_EIO: + return PNFS_OSD_ERR_EIO; + } +} + +static void _clear_bio(struct bio *bio) +{ + struct bio_vec *bv; + unsigned i; + + __bio_for_each_segment(bv, bio, i, 0) { + unsigned this_count = bv->bv_len; + + if (likely(PAGE_SIZE == this_count)) + clear_highpage(bv->bv_page); + else + zero_user(bv->bv_page, bv->bv_offset, this_count); + } +} + +static int _io_check(struct objio_state *ios, bool is_write) +{ + enum osd_err_priority oep = OSD_ERR_PRI_NO_ERROR; + int lin_ret = 0; + int i; + + for (i = 0; i < ios->numdevs; i++) { + struct osd_sense_info osi; + struct osd_request *or = ios->per_dev[i].or; + unsigned dev; + int ret; + + if (!or) + continue; + + ret = osd_req_decode_sense(or, &osi); + if (likely(!ret)) + continue; + + if (OSD_ERR_PRI_CLEAR_PAGES == osi.osd_err_pri) { + /* start read offset passed endof file */ + BUG_ON(is_write); + _clear_bio(ios->per_dev[i].bio); + dprintk("%s: start read offset passed end of file " + "offset=0x%llx, length=0x%lx\n", __func__, + _LLU(ios->per_dev[i].offset), + ios->per_dev[i].length); + + continue; /* we recovered */ + } + dev = ios->per_dev[i].dev; + objlayout_io_set_result(&ios->ol_state, dev, + &ios->layout->comps[dev].oc_object_id, + osd_pri_2_pnfs_err(osi.osd_err_pri), + ios->per_dev[i].offset, + ios->per_dev[i].length, + is_write); + + if (osi.osd_err_pri >= oep) { + oep = osi.osd_err_pri; + lin_ret = ret; + } + } + + return lin_ret; +} + +/* + * Common IO state helpers. + */ +static void _io_free(struct objio_state *ios) +{ + unsigned i; + + for (i = 0; i < ios->numdevs; i++) { + struct _objio_per_comp *per_dev = &ios->per_dev[i]; + + if (per_dev->or) { + osd_end_request(per_dev->or); + per_dev->or = NULL; + } + + if (per_dev->bio) { + bio_put(per_dev->bio); + per_dev->bio = NULL; + } + } +} + +struct osd_dev *_io_od(struct objio_state *ios, unsigned dev) +{ + unsigned min_dev = ios->layout->comps_index; + unsigned max_dev = min_dev + ios->layout->num_comps; + + BUG_ON(dev < min_dev || max_dev <= dev); + return ios->layout->ods[dev - min_dev]->od; +} + +struct _striping_info { + u64 obj_offset; + u64 group_length; + unsigned dev; + unsigned unit_off; +}; + +static void _calc_stripe_info(struct objio_state *ios, u64 file_offset, + struct _striping_info *si) +{ + u32 stripe_unit = ios->layout->stripe_unit; + u32 group_width = ios->layout->group_width; + u64 group_depth = ios->layout->group_depth; + u32 U = stripe_unit * group_width; + + u64 T = U * group_depth; + u64 S = T * ios->layout->group_count; + u64 M = div64_u64(file_offset, S); + + /* + G = (L - (M * S)) / T + H = (L - (M * S)) % T + */ + u64 LmodU = file_offset - M * S; + u32 G = div64_u64(LmodU, T); + u64 H = LmodU - G * T; + + u32 N = div_u64(H, U); + + div_u64_rem(file_offset, stripe_unit, &si->unit_off); + si->obj_offset = si->unit_off + (N * stripe_unit) + + (M * group_depth * stripe_unit); + + /* "H - (N * U)" is just "H % U" so it's bound to u32 */ + si->dev = (u32)(H - (N * U)) / stripe_unit + G * group_width; + si->dev *= ios->layout->mirrors_p1; + + si->group_length = T - H; +} + +static int _add_stripe_unit(struct objio_state *ios, unsigned *cur_pg, + unsigned pgbase, struct _objio_per_comp *per_dev, int cur_len, + gfp_t gfp_flags) +{ + unsigned pg = *cur_pg; + struct request_queue *q = + osd_request_queue(_io_od(ios, per_dev->dev)); + + per_dev->length += cur_len; + + if (per_dev->bio == NULL) { + unsigned stripes = ios->layout->num_comps / + ios->layout->mirrors_p1; + unsigned pages_in_stripe = stripes * + (ios->layout->stripe_unit / PAGE_SIZE); + unsigned bio_size = (ios->ol_state.nr_pages + pages_in_stripe) / + stripes; + + if (BIO_MAX_PAGES_KMALLOC < bio_size) + bio_size = BIO_MAX_PAGES_KMALLOC; + + per_dev->bio = bio_kmalloc(gfp_flags, bio_size); + if (unlikely(!per_dev->bio)) { + dprintk("Faild to allocate BIO size=%u\n", bio_size); + return -ENOMEM; + } + } + + while (cur_len > 0) { + unsigned pglen = min_t(unsigned, PAGE_SIZE - pgbase, cur_len); + unsigned added_len; + + BUG_ON(ios->ol_state.nr_pages <= pg); + cur_len -= pglen; + + added_len = bio_add_pc_page(q, per_dev->bio, + ios->ol_state.pages[pg], pglen, pgbase); + if (unlikely(pglen != added_len)) + return -ENOMEM; + pgbase = 0; + ++pg; + } + BUG_ON(cur_len); + + *cur_pg = pg; + return 0; +} + +static int _prepare_one_group(struct objio_state *ios, u64 length, + struct _striping_info *si, unsigned *last_pg, + gfp_t gfp_flags) +{ + unsigned stripe_unit = ios->layout->stripe_unit; + unsigned mirrors_p1 = ios->layout->mirrors_p1; + unsigned devs_in_group = ios->layout->group_width * mirrors_p1; + unsigned dev = si->dev; + unsigned first_dev = dev - (dev % devs_in_group); + unsigned max_comp = ios->numdevs ? ios->numdevs - mirrors_p1 : 0; + unsigned cur_pg = *last_pg; + int ret = 0; + + while (length) { + struct _objio_per_comp *per_dev = &ios->per_dev[dev]; + unsigned cur_len, page_off = 0; + + if (!per_dev->length) { + per_dev->dev = dev; + if (dev < si->dev) { + per_dev->offset = si->obj_offset + stripe_unit - + si->unit_off; + cur_len = stripe_unit; + } else if (dev == si->dev) { + per_dev->offset = si->obj_offset; + cur_len = stripe_unit - si->unit_off; + page_off = si->unit_off & ~PAGE_MASK; + BUG_ON(page_off && + (page_off != ios->ol_state.pgbase)); + } else { /* dev > si->dev */ + per_dev->offset = si->obj_offset - si->unit_off; + cur_len = stripe_unit; + } + + if (max_comp < dev) + max_comp = dev; + } else { + cur_len = stripe_unit; + } + if (cur_len >= length) + cur_len = length; + + ret = _add_stripe_unit(ios, &cur_pg, page_off , per_dev, + cur_len, gfp_flags); + if (unlikely(ret)) + goto out; + + dev += mirrors_p1; + dev = (dev % devs_in_group) + first_dev; + + length -= cur_len; + ios->length += cur_len; + } +out: + ios->numdevs = max_comp + mirrors_p1; + *last_pg = cur_pg; + return ret; +} + +static int _io_rw_pagelist(struct objio_state *ios, gfp_t gfp_flags) +{ + u64 length = ios->ol_state.count; + u64 offset = ios->ol_state.offset; + struct _striping_info si; + unsigned last_pg = 0; + int ret = 0; + + while (length) { + _calc_stripe_info(ios, offset, &si); + + if (length < si.group_length) + si.group_length = length; + + ret = _prepare_one_group(ios, si.group_length, &si, &last_pg, gfp_flags); + if (unlikely(ret)) + goto out; + + offset += si.group_length; + length -= si.group_length; + } + +out: + if (!ios->length) + return ret; + + return 0; +} + +static ssize_t _sync_done(struct objio_state *ios) +{ + struct completion *waiting = ios->private; + + complete(waiting); + return 0; +} + +static void _last_io(struct kref *kref) +{ + struct objio_state *ios = container_of(kref, struct objio_state, kref); + + ios->done(ios); +} + +static void _done_io(struct osd_request *or, void *p) +{ + struct objio_state *ios = p; + + kref_put(&ios->kref, _last_io); +} + +static ssize_t _io_exec(struct objio_state *ios) +{ + DECLARE_COMPLETION_ONSTACK(wait); + ssize_t status = 0; /* sync status */ + unsigned i; + objio_done_fn saved_done_fn = ios->done; + bool sync = ios->ol_state.sync; + + if (sync) { + ios->done = _sync_done; + ios->private = &wait; + } + + kref_init(&ios->kref); + + for (i = 0; i < ios->numdevs; i++) { + struct osd_request *or = ios->per_dev[i].or; + + if (!or) + continue; + + kref_get(&ios->kref); + osd_execute_request_async(or, _done_io, ios); + } + + kref_put(&ios->kref, _last_io); + + if (sync) { + wait_for_completion(&wait); + status = saved_done_fn(ios); + } + + return status; +} + +/* + * read + */ +static ssize_t _read_done(struct objio_state *ios) +{ + ssize_t status; + int ret = _io_check(ios, false); + + _io_free(ios); + + if (likely(!ret)) + status = ios->length; + else + status = ret; + + objlayout_read_done(&ios->ol_state, status, ios->ol_state.sync); + return status; +} + +static int _read_mirrors(struct objio_state *ios, unsigned cur_comp) +{ + struct osd_request *or = NULL; + struct _objio_per_comp *per_dev = &ios->per_dev[cur_comp]; + unsigned dev = per_dev->dev; + struct pnfs_osd_object_cred *cred = + &ios->layout->comps[dev]; + struct osd_obj_id obj = { + .partition = cred->oc_object_id.oid_partition_id, + .id = cred->oc_object_id.oid_object_id, + }; + int ret; + + or = osd_start_request(_io_od(ios, dev), GFP_KERNEL); + if (unlikely(!or)) { + ret = -ENOMEM; + goto err; + } + per_dev->or = or; + + osd_req_read(or, &obj, per_dev->offset, per_dev->bio, per_dev->length); + + ret = osd_finalize_request(or, 0, cred->oc_cap.cred, NULL); + if (ret) { + dprintk("%s: Faild to osd_finalize_request() => %d\n", + __func__, ret); + goto err; + } + + dprintk("%s:[%d] dev=%d obj=0x%llx start=0x%llx length=0x%lx\n", + __func__, cur_comp, dev, obj.id, _LLU(per_dev->offset), + per_dev->length); + +err: + return ret; +} + +static ssize_t _read_exec(struct objio_state *ios) +{ + unsigned i; + int ret; + + for (i = 0; i < ios->numdevs; i += ios->layout->mirrors_p1) { + if (!ios->per_dev[i].length) + continue; + ret = _read_mirrors(ios, i); + if (unlikely(ret)) + goto err; + } + + ios->done = _read_done; + return _io_exec(ios); /* In sync mode exec returns the io status */ + +err: + _io_free(ios); + return ret; +} + +ssize_t objio_read_pagelist(struct objlayout_io_state *ol_state) +{ + struct objio_state *ios = container_of(ol_state, struct objio_state, + ol_state); + int ret; + + ret = _io_rw_pagelist(ios, GFP_KERNEL); + if (unlikely(ret)) + return ret; + + return _read_exec(ios); +} + +/* + * write + */ +static ssize_t _write_done(struct objio_state *ios) +{ + ssize_t status; + int ret = _io_check(ios, true); + + _io_free(ios); + + if (likely(!ret)) { + /* FIXME: should be based on the OSD's persistence model + * See OSD2r05 Section 4.13 Data persistence model */ + ios->ol_state.committed = NFS_FILE_SYNC; + status = ios->length; + } else { + status = ret; + } + + objlayout_write_done(&ios->ol_state, status, ios->ol_state.sync); + return status; +} + +static int _write_mirrors(struct objio_state *ios, unsigned cur_comp) +{ + struct _objio_per_comp *master_dev = &ios->per_dev[cur_comp]; + unsigned dev = ios->per_dev[cur_comp].dev; + unsigned last_comp = cur_comp + ios->layout->mirrors_p1; + int ret; + + for (; cur_comp < last_comp; ++cur_comp, ++dev) { + struct osd_request *or = NULL; + struct pnfs_osd_object_cred *cred = + &ios->layout->comps[dev]; + struct osd_obj_id obj = { + .partition = cred->oc_object_id.oid_partition_id, + .id = cred->oc_object_id.oid_object_id, + }; + struct _objio_per_comp *per_dev = &ios->per_dev[cur_comp]; + struct bio *bio; + + or = osd_start_request(_io_od(ios, dev), GFP_NOFS); + if (unlikely(!or)) { + ret = -ENOMEM; + goto err; + } + per_dev->or = or; + + if (per_dev != master_dev) { + bio = bio_kmalloc(GFP_NOFS, + master_dev->bio->bi_max_vecs); + if (unlikely(!bio)) { + dprintk("Faild to allocate BIO size=%u\n", + master_dev->bio->bi_max_vecs); + ret = -ENOMEM; + goto err; + } + + __bio_clone(bio, master_dev->bio); + bio->bi_bdev = NULL; + bio->bi_next = NULL; + per_dev->bio = bio; + per_dev->dev = dev; + per_dev->length = master_dev->length; + per_dev->offset = master_dev->offset; + } else { + bio = master_dev->bio; + bio->bi_rw |= REQ_WRITE; + } + + osd_req_write(or, &obj, per_dev->offset, bio, per_dev->length); + + ret = osd_finalize_request(or, 0, cred->oc_cap.cred, NULL); + if (ret) { + dprintk("%s: Faild to osd_finalize_request() => %d\n", + __func__, ret); + goto err; + } + + dprintk("%s:[%d] dev=%d obj=0x%llx start=0x%llx length=0x%lx\n", + __func__, cur_comp, dev, obj.id, _LLU(per_dev->offset), + per_dev->length); + } + +err: + return ret; +} + +static ssize_t _write_exec(struct objio_state *ios) +{ + unsigned i; + int ret; + + for (i = 0; i < ios->numdevs; i += ios->layout->mirrors_p1) { + if (!ios->per_dev[i].length) + continue; + ret = _write_mirrors(ios, i); + if (unlikely(ret)) + goto err; + } + + ios->done = _write_done; + return _io_exec(ios); /* In sync mode exec returns the io->status */ + +err: + _io_free(ios); + return ret; +} + +ssize_t objio_write_pagelist(struct objlayout_io_state *ol_state, bool stable) +{ + struct objio_state *ios = container_of(ol_state, struct objio_state, + ol_state); + int ret; + + /* TODO: ios->stable = stable; */ + ret = _io_rw_pagelist(ios, GFP_NOFS); + if (unlikely(ret)) + return ret; + + return _write_exec(ios); +} + +static bool objio_pg_test(struct nfs_pageio_descriptor *pgio, + struct nfs_page *prev, struct nfs_page *req) +{ + if (!pnfs_generic_pg_test(pgio, prev, req)) + return false; + + if (pgio->pg_lseg == NULL) + return true; + + return pgio->pg_count + req->wb_bytes <= + OBJIO_LSEG(pgio->pg_lseg)->max_io_size; +} + +static struct pnfs_layoutdriver_type objlayout_type = { + .id = LAYOUT_OSD2_OBJECTS, + .name = "LAYOUT_OSD2_OBJECTS", + .flags = PNFS_LAYOUTRET_ON_SETATTR, + + .alloc_layout_hdr = objlayout_alloc_layout_hdr, + .free_layout_hdr = objlayout_free_layout_hdr, + + .alloc_lseg = objlayout_alloc_lseg, + .free_lseg = objlayout_free_lseg, + + .read_pagelist = objlayout_read_pagelist, + .write_pagelist = objlayout_write_pagelist, + .pg_test = objio_pg_test, + + .free_deviceid_node = objio_free_deviceid_node, + + .encode_layoutcommit = objlayout_encode_layoutcommit, + .encode_layoutreturn = objlayout_encode_layoutreturn, +}; + +MODULE_DESCRIPTION("pNFS Layout Driver for OSD2 objects"); +MODULE_AUTHOR("Benny Halevy <bhalevy@panasas.com>"); +MODULE_LICENSE("GPL"); + +static int __init +objlayout_init(void) +{ + int ret = pnfs_register_layoutdriver(&objlayout_type); + + if (ret) + printk(KERN_INFO + "%s: Registering OSD pNFS Layout Driver failed: error=%d\n", + __func__, ret); + else + printk(KERN_INFO "%s: Registered OSD pNFS Layout Driver\n", + __func__); + return ret; +} + +static void __exit +objlayout_exit(void) +{ + pnfs_unregister_layoutdriver(&objlayout_type); + printk(KERN_INFO "%s: Unregistered OSD pNFS Layout Driver\n", + __func__); +} + +module_init(objlayout_init); +module_exit(objlayout_exit); diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c new file mode 100644 index 0000000..1d06f8e --- /dev/null +++ b/fs/nfs/objlayout/objlayout.c @@ -0,0 +1,712 @@ +/* + * pNFS Objects layout driver high level definitions + * + * Copyright (C) 2007 Panasas Inc. [year of first publication] + * All rights reserved. + * + * Benny Halevy <bhalevy@panasas.com> + * Boaz Harrosh <bharrosh@panasas.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * See the file COPYING included with this distribution for more details. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the Panasas company nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <scsi/osd_initiator.h> +#include "objlayout.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD +/* + * Create a objlayout layout structure for the given inode and return it. + */ +struct pnfs_layout_hdr * +objlayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags) +{ + struct objlayout *objlay; + + objlay = kzalloc(sizeof(struct objlayout), gfp_flags); + if (objlay) { + spin_lock_init(&objlay->lock); + INIT_LIST_HEAD(&objlay->err_list); + } + dprintk("%s: Return %p\n", __func__, objlay); + return &objlay->pnfs_layout; +} + +/* + * Free an objlayout layout structure + */ +void +objlayout_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct objlayout *objlay = OBJLAYOUT(lo); + + dprintk("%s: objlay %p\n", __func__, objlay); + + WARN_ON(!list_empty(&objlay->err_list)); + kfree(objlay); +} + +/* + * Unmarshall layout and store it in pnfslay. + */ +struct pnfs_layout_segment * +objlayout_alloc_lseg(struct pnfs_layout_hdr *pnfslay, + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) +{ + int status = -ENOMEM; + struct xdr_stream stream; + struct xdr_buf buf = { + .pages = lgr->layoutp->pages, + .page_len = lgr->layoutp->len, + .buflen = lgr->layoutp->len, + .len = lgr->layoutp->len, + }; + struct page *scratch; + struct pnfs_layout_segment *lseg; + + dprintk("%s: Begin pnfslay %p\n", __func__, pnfslay); + + scratch = alloc_page(gfp_flags); + if (!scratch) + goto err_nofree; + + xdr_init_decode(&stream, &buf, NULL); + xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); + + status = objio_alloc_lseg(&lseg, pnfslay, &lgr->range, &stream, gfp_flags); + if (unlikely(status)) { + dprintk("%s: objio_alloc_lseg Return err %d\n", __func__, + status); + goto err; + } + + __free_page(scratch); + + dprintk("%s: Return %p\n", __func__, lseg); + return lseg; + +err: + __free_page(scratch); +err_nofree: + dprintk("%s: Err Return=>%d\n", __func__, status); + return ERR_PTR(status); +} + +/* + * Free a layout segement + */ +void +objlayout_free_lseg(struct pnfs_layout_segment *lseg) +{ + dprintk("%s: freeing layout segment %p\n", __func__, lseg); + + if (unlikely(!lseg)) + return; + + objio_free_lseg(lseg); +} + +/* + * I/O Operations + */ +static inline u64 +end_offset(u64 start, u64 len) +{ + u64 end; + + end = start + len; + return end >= start ? end : NFS4_MAX_UINT64; +} + +/* last octet in a range */ +static inline u64 +last_byte_offset(u64 start, u64 len) +{ + u64 end; + + BUG_ON(!len); + end = start + len; + return end > start ? end - 1 : NFS4_MAX_UINT64; +} + +static struct objlayout_io_state * +objlayout_alloc_io_state(struct pnfs_layout_hdr *pnfs_layout_type, + struct page **pages, + unsigned pgbase, + loff_t offset, + size_t count, + struct pnfs_layout_segment *lseg, + void *rpcdata, + gfp_t gfp_flags) +{ + struct objlayout_io_state *state; + u64 lseg_end_offset; + + dprintk("%s: allocating io_state\n", __func__); + if (objio_alloc_io_state(lseg, &state, gfp_flags)) + return NULL; + + BUG_ON(offset < lseg->pls_range.offset); + lseg_end_offset = end_offset(lseg->pls_range.offset, + lseg->pls_range.length); + BUG_ON(offset >= lseg_end_offset); + if (offset + count > lseg_end_offset) { + count = lseg->pls_range.length - + (offset - lseg->pls_range.offset); + dprintk("%s: truncated count %Zd\n", __func__, count); + } + + if (pgbase > PAGE_SIZE) { + pages += pgbase >> PAGE_SHIFT; + pgbase &= ~PAGE_MASK; + } + + INIT_LIST_HEAD(&state->err_list); + state->lseg = lseg; + state->rpcdata = rpcdata; + state->pages = pages; + state->pgbase = pgbase; + state->nr_pages = (pgbase + count + PAGE_SIZE - 1) >> PAGE_SHIFT; + state->offset = offset; + state->count = count; + state->sync = 0; + + return state; +} + +static void +objlayout_free_io_state(struct objlayout_io_state *state) +{ + dprintk("%s: freeing io_state\n", __func__); + if (unlikely(!state)) + return; + + objio_free_io_state(state); +} + +/* + * I/O done common code + */ +static void +objlayout_iodone(struct objlayout_io_state *state) +{ + dprintk("%s: state %p status\n", __func__, state); + + if (likely(state->status >= 0)) { + objlayout_free_io_state(state); + } else { + struct objlayout *objlay = OBJLAYOUT(state->lseg->pls_layout); + + spin_lock(&objlay->lock); + objlay->delta_space_valid = OBJ_DSU_INVALID; + list_add(&objlay->err_list, &state->err_list); + spin_unlock(&objlay->lock); + } +} + +/* + * objlayout_io_set_result - Set an osd_error code on a specific osd comp. + * + * The @index component IO failed (error returned from target). Register + * the error for later reporting at layout-return. + */ +void +objlayout_io_set_result(struct objlayout_io_state *state, unsigned index, + struct pnfs_osd_objid *pooid, int osd_error, + u64 offset, u64 length, bool is_write) +{ + struct pnfs_osd_ioerr *ioerr = &state->ioerrs[index]; + + BUG_ON(index >= state->num_comps); + if (osd_error) { + ioerr->oer_component = *pooid; + ioerr->oer_comp_offset = offset; + ioerr->oer_comp_length = length; + ioerr->oer_iswrite = is_write; + ioerr->oer_errno = osd_error; + + dprintk("%s: err[%d]: errno=%d is_write=%d dev(%llx:%llx) " + "par=0x%llx obj=0x%llx offset=0x%llx length=0x%llx\n", + __func__, index, ioerr->oer_errno, + ioerr->oer_iswrite, + _DEVID_LO(&ioerr->oer_component.oid_device_id), + _DEVID_HI(&ioerr->oer_component.oid_device_id), + ioerr->oer_component.oid_partition_id, + ioerr->oer_component.oid_object_id, + ioerr->oer_comp_offset, + ioerr->oer_comp_length); + } else { + /* User need not call if no error is reported */ + ioerr->oer_errno = 0; + } +} + +/* Function scheduled on rpc workqueue to call ->nfs_readlist_complete(). + * This is because the osd completion is called with ints-off from + * the block layer + */ +static void _rpc_read_complete(struct work_struct *work) +{ + struct rpc_task *task; + struct nfs_read_data *rdata; + + dprintk("%s enter\n", __func__); + task = container_of(work, struct rpc_task, u.tk_work); + rdata = container_of(task, struct nfs_read_data, task); + + pnfs_ld_read_done(rdata); +} + +void +objlayout_read_done(struct objlayout_io_state *state, ssize_t status, bool sync) +{ + int eof = state->eof; + struct nfs_read_data *rdata; + + state->status = status; + dprintk("%s: Begin status=%zd eof=%d\n", __func__, status, eof); + rdata = state->rpcdata; + rdata->task.tk_status = status; + if (status >= 0) { + rdata->res.count = status; + rdata->res.eof = eof; + } + objlayout_iodone(state); + /* must not use state after this point */ + + if (sync) + pnfs_ld_read_done(rdata); + else { + INIT_WORK(&rdata->task.u.tk_work, _rpc_read_complete); + schedule_work(&rdata->task.u.tk_work); + } +} + +/* + * Perform sync or async reads. + */ +enum pnfs_try_status +objlayout_read_pagelist(struct nfs_read_data *rdata) +{ + loff_t offset = rdata->args.offset; + size_t count = rdata->args.count; + struct objlayout_io_state *state; + ssize_t status = 0; + loff_t eof; + + dprintk("%s: Begin inode %p offset %llu count %d\n", + __func__, rdata->inode, offset, (int)count); + + eof = i_size_read(rdata->inode); + if (unlikely(offset + count > eof)) { + if (offset >= eof) { + status = 0; + rdata->res.count = 0; + rdata->res.eof = 1; + goto out; + } + count = eof - offset; + } + + state = objlayout_alloc_io_state(NFS_I(rdata->inode)->layout, + rdata->args.pages, rdata->args.pgbase, + offset, count, + rdata->lseg, rdata, + GFP_KERNEL); + if (unlikely(!state)) { + status = -ENOMEM; + goto out; + } + + state->eof = state->offset + state->count >= eof; + + status = objio_read_pagelist(state); + out: + dprintk("%s: Return status %Zd\n", __func__, status); + rdata->pnfs_error = status; + return PNFS_ATTEMPTED; +} + +/* Function scheduled on rpc workqueue to call ->nfs_writelist_complete(). + * This is because the osd completion is called with ints-off from + * the block layer + */ +static void _rpc_write_complete(struct work_struct *work) +{ + struct rpc_task *task; + struct nfs_write_data *wdata; + + dprintk("%s enter\n", __func__); + task = container_of(work, struct rpc_task, u.tk_work); + wdata = container_of(task, struct nfs_write_data, task); + + pnfs_ld_write_done(wdata); +} + +void +objlayout_write_done(struct objlayout_io_state *state, ssize_t status, + bool sync) +{ + struct nfs_write_data *wdata; + + dprintk("%s: Begin\n", __func__); + wdata = state->rpcdata; + state->status = status; + wdata->task.tk_status = status; + if (status >= 0) { + wdata->res.count = status; + wdata->verf.committed = state->committed; + dprintk("%s: Return status %d committed %d\n", + __func__, wdata->task.tk_status, + wdata->verf.committed); + } else + dprintk("%s: Return status %d\n", + __func__, wdata->task.tk_status); + objlayout_iodone(state); + /* must not use state after this point */ + + if (sync) + pnfs_ld_write_done(wdata); + else { + INIT_WORK(&wdata->task.u.tk_work, _rpc_write_complete); + schedule_work(&wdata->task.u.tk_work); + } +} + +/* + * Perform sync or async writes. + */ +enum pnfs_try_status +objlayout_write_pagelist(struct nfs_write_data *wdata, + int how) +{ + struct objlayout_io_state *state; + ssize_t status; + + dprintk("%s: Begin inode %p offset %llu count %u\n", + __func__, wdata->inode, wdata->args.offset, wdata->args.count); + + state = objlayout_alloc_io_state(NFS_I(wdata->inode)->layout, + wdata->args.pages, + wdata->args.pgbase, + wdata->args.offset, + wdata->args.count, + wdata->lseg, wdata, + GFP_NOFS); + if (unlikely(!state)) { + status = -ENOMEM; + goto out; + } + + state->sync = how & FLUSH_SYNC; + + status = objio_write_pagelist(state, how & FLUSH_STABLE); + out: + dprintk("%s: Return status %Zd\n", __func__, status); + wdata->pnfs_error = status; + return PNFS_ATTEMPTED; +} + +void +objlayout_encode_layoutcommit(struct pnfs_layout_hdr *pnfslay, + struct xdr_stream *xdr, + const struct nfs4_layoutcommit_args *args) +{ + struct objlayout *objlay = OBJLAYOUT(pnfslay); + struct pnfs_osd_layoutupdate lou; + __be32 *start; + + dprintk("%s: Begin\n", __func__); + + spin_lock(&objlay->lock); + lou.dsu_valid = (objlay->delta_space_valid == OBJ_DSU_VALID); + lou.dsu_delta = objlay->delta_space_used; + objlay->delta_space_used = 0; + objlay->delta_space_valid = OBJ_DSU_INIT; + lou.olu_ioerr_flag = !list_empty(&objlay->err_list); + spin_unlock(&objlay->lock); + + start = xdr_reserve_space(xdr, 4); + + BUG_ON(pnfs_osd_xdr_encode_layoutupdate(xdr, &lou)); + + *start = cpu_to_be32((xdr->p - start - 1) * 4); + + dprintk("%s: Return delta_space_used %lld err %d\n", __func__, + lou.dsu_delta, lou.olu_ioerr_flag); +} + +static int +err_prio(u32 oer_errno) +{ + switch (oer_errno) { + case 0: + return 0; + + case PNFS_OSD_ERR_RESOURCE: + return OSD_ERR_PRI_RESOURCE; + case PNFS_OSD_ERR_BAD_CRED: + return OSD_ERR_PRI_BAD_CRED; + case PNFS_OSD_ERR_NO_ACCESS: + return OSD_ERR_PRI_NO_ACCESS; + case PNFS_OSD_ERR_UNREACHABLE: + return OSD_ERR_PRI_UNREACHABLE; + case PNFS_OSD_ERR_NOT_FOUND: + return OSD_ERR_PRI_NOT_FOUND; + case PNFS_OSD_ERR_NO_SPACE: + return OSD_ERR_PRI_NO_SPACE; + default: + WARN_ON(1); + /* fallthrough */ + case PNFS_OSD_ERR_EIO: + return OSD_ERR_PRI_EIO; + } +} + +static void +merge_ioerr(struct pnfs_osd_ioerr *dest_err, + const struct pnfs_osd_ioerr *src_err) +{ + u64 dest_end, src_end; + + if (!dest_err->oer_errno) { + *dest_err = *src_err; + /* accumulated device must be blank */ + memset(&dest_err->oer_component.oid_device_id, 0, + sizeof(dest_err->oer_component.oid_device_id)); + + return; + } + + if (dest_err->oer_component.oid_partition_id != + src_err->oer_component.oid_partition_id) + dest_err->oer_component.oid_partition_id = 0; + + if (dest_err->oer_component.oid_object_id != + src_err->oer_component.oid_object_id) + dest_err->oer_component.oid_object_id = 0; + + if (dest_err->oer_comp_offset > src_err->oer_comp_offset) + dest_err->oer_comp_offset = src_err->oer_comp_offset; + + dest_end = end_offset(dest_err->oer_comp_offset, + dest_err->oer_comp_length); + src_end = end_offset(src_err->oer_comp_offset, + src_err->oer_comp_length); + if (dest_end < src_end) + dest_end = src_end; + + dest_err->oer_comp_length = dest_end - dest_err->oer_comp_offset; + + if ((src_err->oer_iswrite == dest_err->oer_iswrite) && + (err_prio(src_err->oer_errno) > err_prio(dest_err->oer_errno))) { + dest_err->oer_errno = src_err->oer_errno; + } else if (src_err->oer_iswrite) { + dest_err->oer_iswrite = true; + dest_err->oer_errno = src_err->oer_errno; + } +} + +static void +encode_accumulated_error(struct objlayout *objlay, __be32 *p) +{ + struct objlayout_io_state *state, *tmp; + struct pnfs_osd_ioerr accumulated_err = {.oer_errno = 0}; + + list_for_each_entry_safe(state, tmp, &objlay->err_list, err_list) { + unsigned i; + + for (i = 0; i < state->num_comps; i++) { + struct pnfs_osd_ioerr *ioerr = &state->ioerrs[i]; + + if (!ioerr->oer_errno) + continue; + + printk(KERN_ERR "%s: err[%d]: errno=%d is_write=%d " + "dev(%llx:%llx) par=0x%llx obj=0x%llx " + "offset=0x%llx length=0x%llx\n", + __func__, i, ioerr->oer_errno, + ioerr->oer_iswrite, + _DEVID_LO(&ioerr->oer_component.oid_device_id), + _DEVID_HI(&ioerr->oer_component.oid_device_id), + ioerr->oer_component.oid_partition_id, + ioerr->oer_component.oid_object_id, + ioerr->oer_comp_offset, + ioerr->oer_comp_length); + + merge_ioerr(&accumulated_err, ioerr); + } + list_del(&state->err_list); + objlayout_free_io_state(state); + } + + pnfs_osd_xdr_encode_ioerr(p, &accumulated_err); +} + +void +objlayout_encode_layoutreturn(struct pnfs_layout_hdr *pnfslay, + struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args) +{ + struct objlayout *objlay = OBJLAYOUT(pnfslay); + struct objlayout_io_state *state, *tmp; + __be32 *start; + + dprintk("%s: Begin\n", __func__); + start = xdr_reserve_space(xdr, 4); + BUG_ON(!start); + + spin_lock(&objlay->lock); + + list_for_each_entry_safe(state, tmp, &objlay->err_list, err_list) { + __be32 *last_xdr = NULL, *p; + unsigned i; + int res = 0; + + for (i = 0; i < state->num_comps; i++) { + struct pnfs_osd_ioerr *ioerr = &state->ioerrs[i]; + + if (!ioerr->oer_errno) + continue; + + dprintk("%s: err[%d]: errno=%d is_write=%d " + "dev(%llx:%llx) par=0x%llx obj=0x%llx " + "offset=0x%llx length=0x%llx\n", + __func__, i, ioerr->oer_errno, + ioerr->oer_iswrite, + _DEVID_LO(&ioerr->oer_component.oid_device_id), + _DEVID_HI(&ioerr->oer_component.oid_device_id), + ioerr->oer_component.oid_partition_id, + ioerr->oer_component.oid_object_id, + ioerr->oer_comp_offset, + ioerr->oer_comp_length); + + p = pnfs_osd_xdr_ioerr_reserve_space(xdr); + if (unlikely(!p)) { + res = -E2BIG; + break; /* accumulated_error */ + } + + last_xdr = p; + pnfs_osd_xdr_encode_ioerr(p, &state->ioerrs[i]); + } + + /* TODO: use xdr_write_pages */ + if (unlikely(res)) { + /* no space for even one error descriptor */ + BUG_ON(!last_xdr); + + /* we've encountered a situation with lots and lots of + * errors and no space to encode them all. Use the last + * available slot to report the union of all the + * remaining errors. + */ + encode_accumulated_error(objlay, last_xdr); + goto loop_done; + } + list_del(&state->err_list); + objlayout_free_io_state(state); + } +loop_done: + spin_unlock(&objlay->lock); + + *start = cpu_to_be32((xdr->p - start - 1) * 4); + dprintk("%s: Return\n", __func__); +} + + +/* + * Get Device Info API for io engines + */ +struct objlayout_deviceinfo { + struct page *page; + struct pnfs_osd_deviceaddr da; /* This must be last */ +}; + +/* Initialize and call nfs_getdeviceinfo, then decode and return a + * "struct pnfs_osd_deviceaddr *" Eventually objlayout_put_deviceinfo() + * should be called. + */ +int objlayout_get_deviceinfo(struct pnfs_layout_hdr *pnfslay, + struct nfs4_deviceid *d_id, struct pnfs_osd_deviceaddr **deviceaddr, + gfp_t gfp_flags) +{ + struct objlayout_deviceinfo *odi; + struct pnfs_device pd; + struct super_block *sb; + struct page *page, **pages; + u32 *p; + int err; + + page = alloc_page(gfp_flags); + if (!page) + return -ENOMEM; + + pages = &page; + pd.pages = pages; + + memcpy(&pd.dev_id, d_id, sizeof(*d_id)); + pd.layout_type = LAYOUT_OSD2_OBJECTS; + pd.pages = &page; + pd.pgbase = 0; + pd.pglen = PAGE_SIZE; + pd.mincount = 0; + + sb = pnfslay->plh_inode->i_sb; + err = nfs4_proc_getdeviceinfo(NFS_SERVER(pnfslay->plh_inode), &pd); + dprintk("%s nfs_getdeviceinfo returned %d\n", __func__, err); + if (err) + goto err_out; + + p = page_address(page); + odi = kzalloc(sizeof(*odi), gfp_flags); + if (!odi) { + err = -ENOMEM; + goto err_out; + } + pnfs_osd_xdr_decode_deviceaddr(&odi->da, p); + odi->page = page; + *deviceaddr = &odi->da; + return 0; + +err_out: + __free_page(page); + return err; +} + +void objlayout_put_deviceinfo(struct pnfs_osd_deviceaddr *deviceaddr) +{ + struct objlayout_deviceinfo *odi = container_of(deviceaddr, + struct objlayout_deviceinfo, + da); + + __free_page(odi->page); + kfree(odi); +} diff --git a/fs/nfs/objlayout/objlayout.h b/fs/nfs/objlayout/objlayout.h new file mode 100644 index 0000000..a8244c8 --- /dev/null +++ b/fs/nfs/objlayout/objlayout.h @@ -0,0 +1,187 @@ +/* + * Data types and function declerations for interfacing with the + * pNFS standard object layout driver. + * + * Copyright (C) 2007 Panasas Inc. [year of first publication] + * All rights reserved. + * + * Benny Halevy <bhalevy@panasas.com> + * Boaz Harrosh <bharrosh@panasas.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * See the file COPYING included with this distribution for more details. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the Panasas company nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _OBJLAYOUT_H +#define _OBJLAYOUT_H + +#include <linux/nfs_fs.h> +#include <linux/pnfs_osd_xdr.h> +#include "../pnfs.h" + +/* + * per-inode layout + */ +struct objlayout { + struct pnfs_layout_hdr pnfs_layout; + + /* for layout_commit */ + enum osd_delta_space_valid_enum { + OBJ_DSU_INIT = 0, + OBJ_DSU_VALID, + OBJ_DSU_INVALID, + } delta_space_valid; + s64 delta_space_used; /* consumed by write ops */ + + /* for layout_return */ + spinlock_t lock; + struct list_head err_list; +}; + +static inline struct objlayout * +OBJLAYOUT(struct pnfs_layout_hdr *lo) +{ + return container_of(lo, struct objlayout, pnfs_layout); +} + +/* + * per-I/O operation state + * embedded in objects provider io_state data structure + */ +struct objlayout_io_state { + struct pnfs_layout_segment *lseg; + + struct page **pages; + unsigned pgbase; + unsigned nr_pages; + unsigned long count; + loff_t offset; + bool sync; + + void *rpcdata; + int status; /* res */ + int eof; /* res */ + int committed; /* res */ + + /* Error reporting (layout_return) */ + struct list_head err_list; + unsigned num_comps; + /* Pointer to array of error descriptors of size num_comps. + * It should contain as many entries as devices in the osd_layout + * that participate in the I/O. It is up to the io_engine to allocate + * needed space and set num_comps. + */ + struct pnfs_osd_ioerr *ioerrs; +}; + +/* + * Raid engine I/O API + */ +extern int objio_alloc_lseg(struct pnfs_layout_segment **outp, + struct pnfs_layout_hdr *pnfslay, + struct pnfs_layout_range *range, + struct xdr_stream *xdr, + gfp_t gfp_flags); +extern void objio_free_lseg(struct pnfs_layout_segment *lseg); + +extern int objio_alloc_io_state( + struct pnfs_layout_segment *lseg, + struct objlayout_io_state **outp, + gfp_t gfp_flags); +extern void objio_free_io_state(struct objlayout_io_state *state); + +extern ssize_t objio_read_pagelist(struct objlayout_io_state *ol_state); +extern ssize_t objio_write_pagelist(struct objlayout_io_state *ol_state, + bool stable); + +/* + * callback API + */ +extern void objlayout_io_set_result(struct objlayout_io_state *state, + unsigned index, struct pnfs_osd_objid *pooid, + int osd_error, u64 offset, u64 length, bool is_write); + +static inline void +objlayout_add_delta_space_used(struct objlayout_io_state *state, s64 space_used) +{ + struct objlayout *objlay = OBJLAYOUT(state->lseg->pls_layout); + + /* If one of the I/Os errored out and the delta_space_used was + * invalid we render the complete report as invalid. Protocol mandate + * the DSU be accurate or not reported. + */ + spin_lock(&objlay->lock); + if (objlay->delta_space_valid != OBJ_DSU_INVALID) { + objlay->delta_space_valid = OBJ_DSU_VALID; + objlay->delta_space_used += space_used; + } + spin_unlock(&objlay->lock); +} + +extern void objlayout_read_done(struct objlayout_io_state *state, + ssize_t status, bool sync); +extern void objlayout_write_done(struct objlayout_io_state *state, + ssize_t status, bool sync); + +extern int objlayout_get_deviceinfo(struct pnfs_layout_hdr *pnfslay, + struct nfs4_deviceid *d_id, struct pnfs_osd_deviceaddr **deviceaddr, + gfp_t gfp_flags); +extern void objlayout_put_deviceinfo(struct pnfs_osd_deviceaddr *deviceaddr); + +/* + * exported generic objects function vectors + */ + +extern struct pnfs_layout_hdr *objlayout_alloc_layout_hdr(struct inode *, gfp_t gfp_flags); +extern void objlayout_free_layout_hdr(struct pnfs_layout_hdr *); + +extern struct pnfs_layout_segment *objlayout_alloc_lseg( + struct pnfs_layout_hdr *, + struct nfs4_layoutget_res *, + gfp_t gfp_flags); +extern void objlayout_free_lseg(struct pnfs_layout_segment *); + +extern enum pnfs_try_status objlayout_read_pagelist( + struct nfs_read_data *); + +extern enum pnfs_try_status objlayout_write_pagelist( + struct nfs_write_data *, + int how); + +extern void objlayout_encode_layoutcommit( + struct pnfs_layout_hdr *, + struct xdr_stream *, + const struct nfs4_layoutcommit_args *); + +extern void objlayout_encode_layoutreturn( + struct pnfs_layout_hdr *, + struct xdr_stream *, + const struct nfs4_layoutreturn_args *); + +#endif /* _OBJLAYOUT_H */ diff --git a/fs/nfs/objlayout/pnfs_osd_xdr_cli.c b/fs/nfs/objlayout/pnfs_osd_xdr_cli.c new file mode 100644 index 0000000..16fc758 --- /dev/null +++ b/fs/nfs/objlayout/pnfs_osd_xdr_cli.c @@ -0,0 +1,412 @@ +/* + * Object-Based pNFS Layout XDR layer + * + * Copyright (C) 2007 Panasas Inc. [year of first publication] + * All rights reserved. + * + * Benny Halevy <bhalevy@panasas.com> + * Boaz Harrosh <bharrosh@panasas.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * See the file COPYING included with this distribution for more details. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the name of the Panasas company nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR + * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/pnfs_osd_xdr.h> + +#define NFSDBG_FACILITY NFSDBG_PNFS_LD + +/* + * The following implementation is based on RFC5664 + */ + +/* + * struct pnfs_osd_objid { + * struct nfs4_deviceid oid_device_id; + * u64 oid_partition_id; + * u64 oid_object_id; + * }; // xdr size 32 bytes + */ +static __be32 * +_osd_xdr_decode_objid(__be32 *p, struct pnfs_osd_objid *objid) +{ + p = xdr_decode_opaque_fixed(p, objid->oid_device_id.data, + sizeof(objid->oid_device_id.data)); + + p = xdr_decode_hyper(p, &objid->oid_partition_id); + p = xdr_decode_hyper(p, &objid->oid_object_id); + return p; +} +/* + * struct pnfs_osd_opaque_cred { + * u32 cred_len; + * void *cred; + * }; // xdr size [variable] + * The return pointers are from the xdr buffer + */ +static int +_osd_xdr_decode_opaque_cred(struct pnfs_osd_opaque_cred *opaque_cred, + struct xdr_stream *xdr) +{ + __be32 *p = xdr_inline_decode(xdr, 1); + + if (!p) + return -EINVAL; + + opaque_cred->cred_len = be32_to_cpu(*p++); + + p = xdr_inline_decode(xdr, opaque_cred->cred_len); + if (!p) + return -EINVAL; + + opaque_cred->cred = p; + return 0; +} + +/* + * struct pnfs_osd_object_cred { + * struct pnfs_osd_objid oc_object_id; + * u32 oc_osd_version; + * u32 oc_cap_key_sec; + * struct pnfs_osd_opaque_cred oc_cap_key + * struct pnfs_osd_opaque_cred oc_cap; + * }; // xdr size 32 + 4 + 4 + [variable] + [variable] + */ +static int +_osd_xdr_decode_object_cred(struct pnfs_osd_object_cred *comp, + struct xdr_stream *xdr) +{ + __be32 *p = xdr_inline_decode(xdr, 32 + 4 + 4); + int ret; + + if (!p) + return -EIO; + + p = _osd_xdr_decode_objid(p, &comp->oc_object_id); + comp->oc_osd_version = be32_to_cpup(p++); + comp->oc_cap_key_sec = be32_to_cpup(p); + + ret = _osd_xdr_decode_opaque_cred(&comp->oc_cap_key, xdr); + if (unlikely(ret)) + return ret; + + ret = _osd_xdr_decode_opaque_cred(&comp->oc_cap, xdr); + return ret; +} + +/* + * struct pnfs_osd_data_map { + * u32 odm_num_comps; + * u64 odm_stripe_unit; + * u32 odm_group_width; + * u32 odm_group_depth; + * u32 odm_mirror_cnt; + * u32 odm_raid_algorithm; + * }; // xdr size 4 + 8 + 4 + 4 + 4 + 4 + */ +static inline int +_osd_data_map_xdr_sz(void) +{ + return 4 + 8 + 4 + 4 + 4 + 4; +} + +static __be32 * +_osd_xdr_decode_data_map(__be32 *p, struct pnfs_osd_data_map *data_map) +{ + data_map->odm_num_comps = be32_to_cpup(p++); + p = xdr_decode_hyper(p, &data_map->odm_stripe_unit); + data_map->odm_group_width = be32_to_cpup(p++); + data_map->odm_group_depth = be32_to_cpup(p++); + data_map->odm_mirror_cnt = be32_to_cpup(p++); + data_map->odm_raid_algorithm = be32_to_cpup(p++); + dprintk("%s: odm_num_comps=%u odm_stripe_unit=%llu odm_group_width=%u " + "odm_group_depth=%u odm_mirror_cnt=%u odm_raid_algorithm=%u\n", + __func__, + data_map->odm_num_comps, + (unsigned long long)data_map->odm_stripe_unit, + data_map->odm_group_width, + data_map->odm_group_depth, + data_map->odm_mirror_cnt, + data_map->odm_raid_algorithm); + return p; +} + +int pnfs_osd_xdr_decode_layout_map(struct pnfs_osd_layout *layout, + struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr) +{ + __be32 *p; + + memset(iter, 0, sizeof(*iter)); + + p = xdr_inline_decode(xdr, _osd_data_map_xdr_sz() + 4 + 4); + if (unlikely(!p)) + return -EINVAL; + + p = _osd_xdr_decode_data_map(p, &layout->olo_map); + layout->olo_comps_index = be32_to_cpup(p++); + layout->olo_num_comps = be32_to_cpup(p++); + iter->total_comps = layout->olo_num_comps; + return 0; +} + +bool pnfs_osd_xdr_decode_layout_comp(struct pnfs_osd_object_cred *comp, + struct pnfs_osd_xdr_decode_layout_iter *iter, struct xdr_stream *xdr, + int *err) +{ + BUG_ON(iter->decoded_comps > iter->total_comps); + if (iter->decoded_comps == iter->total_comps) + return false; + + *err = _osd_xdr_decode_object_cred(comp, xdr); + if (unlikely(*err)) { + dprintk("%s: _osd_xdr_decode_object_cred=>%d decoded_comps=%d " + "total_comps=%d\n", __func__, *err, + iter->decoded_comps, iter->total_comps); + return false; /* stop the loop */ + } + dprintk("%s: dev(%llx:%llx) par=0x%llx obj=0x%llx " + "key_len=%u cap_len=%u\n", + __func__, + _DEVID_LO(&comp->oc_object_id.oid_device_id), + _DEVID_HI(&comp->oc_object_id.oid_device_id), + comp->oc_object_id.oid_partition_id, + comp->oc_object_id.oid_object_id, + comp->oc_cap_key.cred_len, comp->oc_cap.cred_len); + + iter->decoded_comps++; + return true; +} + +/* + * Get Device Information Decoding + * + * Note: since Device Information is currently done synchronously, all + * variable strings fields are left inside the rpc buffer and are only + * pointed to by the pnfs_osd_deviceaddr members. So the read buffer + * should not be freed while the returned information is in use. + */ +/* + *struct nfs4_string { + * unsigned int len; + * char *data; + *}; // size [variable] + * NOTE: Returned string points to inside the XDR buffer + */ +static __be32 * +__read_u8_opaque(__be32 *p, struct nfs4_string *str) +{ + str->len = be32_to_cpup(p++); + str->data = (char *)p; + + p += XDR_QUADLEN(str->len); + return p; +} + +/* + * struct pnfs_osd_targetid { + * u32 oti_type; + * struct nfs4_string oti_scsi_device_id; + * };// size 4 + [variable] + */ +static __be32 * +__read_targetid(__be32 *p, struct pnfs_osd_targetid* targetid) +{ + u32 oti_type; + + oti_type = be32_to_cpup(p++); + targetid->oti_type = oti_type; + + switch (oti_type) { + case OBJ_TARGET_SCSI_NAME: + case OBJ_TARGET_SCSI_DEVICE_ID: + p = __read_u8_opaque(p, &targetid->oti_scsi_device_id); + } + + return p; +} + +/* + * struct pnfs_osd_net_addr { + * struct nfs4_string r_netid; + * struct nfs4_string r_addr; + * }; + */ +static __be32 * +__read_net_addr(__be32 *p, struct pnfs_osd_net_addr* netaddr) +{ + p = __read_u8_opaque(p, &netaddr->r_netid); + p = __read_u8_opaque(p, &netaddr->r_addr); + + return p; +} + +/* + * struct pnfs_osd_targetaddr { + * u32 ota_available; + * struct pnfs_osd_net_addr ota_netaddr; + * }; + */ +static __be32 * +__read_targetaddr(__be32 *p, struct pnfs_osd_targetaddr *targetaddr) +{ + u32 ota_available; + + ota_available = be32_to_cpup(p++); + targetaddr->ota_available = ota_available; + + if (ota_available) + p = __read_net_addr(p, &targetaddr->ota_netaddr); + + + return p; +} + +/* + * struct pnfs_osd_deviceaddr { + * struct pnfs_osd_targetid oda_targetid; + * struct pnfs_osd_targetaddr oda_targetaddr; + * u8 oda_lun[8]; + * struct nfs4_string oda_systemid; + * struct pnfs_osd_object_cred oda_root_obj_cred; + * struct nfs4_string oda_osdname; + * }; + */ + +/* We need this version for the pnfs_osd_xdr_decode_deviceaddr which does + * not have an xdr_stream + */ +static __be32 * +__read_opaque_cred(__be32 *p, + struct pnfs_osd_opaque_cred *opaque_cred) +{ + opaque_cred->cred_len = be32_to_cpu(*p++); + opaque_cred->cred = p; + return p + XDR_QUADLEN(opaque_cred->cred_len); +} + +static __be32 * +__read_object_cred(__be32 *p, struct pnfs_osd_object_cred *comp) +{ + p = _osd_xdr_decode_objid(p, &comp->oc_object_id); + comp->oc_osd_version = be32_to_cpup(p++); + comp->oc_cap_key_sec = be32_to_cpup(p++); + + p = __read_opaque_cred(p, &comp->oc_cap_key); + p = __read_opaque_cred(p, &comp->oc_cap); + return p; +} + +void pnfs_osd_xdr_decode_deviceaddr( + struct pnfs_osd_deviceaddr *deviceaddr, __be32 *p) +{ + p = __read_targetid(p, &deviceaddr->oda_targetid); + + p = __read_targetaddr(p, &deviceaddr->oda_targetaddr); + + p = xdr_decode_opaque_fixed(p, deviceaddr->oda_lun, + sizeof(deviceaddr->oda_lun)); + + p = __read_u8_opaque(p, &deviceaddr->oda_systemid); + + p = __read_object_cred(p, &deviceaddr->oda_root_obj_cred); + + p = __read_u8_opaque(p, &deviceaddr->oda_osdname); + + /* libosd likes this terminated in dbg. It's last, so no problems */ + deviceaddr->oda_osdname.data[deviceaddr->oda_osdname.len] = 0; +} + +/* + * struct pnfs_osd_layoutupdate { + * u32 dsu_valid; + * s64 dsu_delta; + * u32 olu_ioerr_flag; + * }; xdr size 4 + 8 + 4 + */ +int +pnfs_osd_xdr_encode_layoutupdate(struct xdr_stream *xdr, + struct pnfs_osd_layoutupdate *lou) +{ + __be32 *p = xdr_reserve_space(xdr, 4 + 8 + 4); + + if (!p) + return -E2BIG; + + *p++ = cpu_to_be32(lou->dsu_valid); + if (lou->dsu_valid) + p = xdr_encode_hyper(p, lou->dsu_delta); + *p++ = cpu_to_be32(lou->olu_ioerr_flag); + return 0; +} + +/* + * struct pnfs_osd_objid { + * struct nfs4_deviceid oid_device_id; + * u64 oid_partition_id; + * u64 oid_object_id; + * }; // xdr size 32 bytes + */ +static inline __be32 * +pnfs_osd_xdr_encode_objid(__be32 *p, struct pnfs_osd_objid *object_id) +{ + p = xdr_encode_opaque_fixed(p, &object_id->oid_device_id.data, + sizeof(object_id->oid_device_id.data)); + p = xdr_encode_hyper(p, object_id->oid_partition_id); + p = xdr_encode_hyper(p, object_id->oid_object_id); + + return p; +} + +/* + * struct pnfs_osd_ioerr { + * struct pnfs_osd_objid oer_component; + * u64 oer_comp_offset; + * u64 oer_comp_length; + * u32 oer_iswrite; + * u32 oer_errno; + * }; // xdr size 32 + 24 bytes + */ +void pnfs_osd_xdr_encode_ioerr(__be32 *p, struct pnfs_osd_ioerr *ioerr) +{ + p = pnfs_osd_xdr_encode_objid(p, &ioerr->oer_component); + p = xdr_encode_hyper(p, ioerr->oer_comp_offset); + p = xdr_encode_hyper(p, ioerr->oer_comp_length); + *p++ = cpu_to_be32(ioerr->oer_iswrite); + *p = cpu_to_be32(ioerr->oer_errno); +} + +__be32 *pnfs_osd_xdr_ioerr_reserve_space(struct xdr_stream *xdr) +{ + __be32 *p; + + p = xdr_reserve_space(xdr, 32 + 24); + if (unlikely(!p)) + dprintk("%s: out of xdr space\n", __func__); + + return p; +} diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index c80add6..0098557 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -204,6 +204,22 @@ nfs_wait_on_request(struct nfs_page *req) TASK_UNINTERRUPTIBLE); } +bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) +{ + /* + * FIXME: ideally we should be able to coalesce all requests + * that are not block boundary aligned, but currently this + * is problematic for the case of bsize < PAGE_CACHE_SIZE, + * since nfs_flush_multi and nfs_pagein_multi assume you + * can have only one struct nfs_page. + */ + if (desc->pg_bsize < PAGE_SIZE) + return 0; + + return desc->pg_count + req->wb_bytes <= desc->pg_bsize; +} +EXPORT_SYMBOL_GPL(nfs_generic_pg_test); + /** * nfs_pageio_init - initialise a page io descriptor * @desc: pointer to descriptor @@ -229,6 +245,8 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, desc->pg_ioflags = io_flags; desc->pg_error = 0; desc->pg_lseg = NULL; + desc->pg_test = nfs_generic_pg_test; + pnfs_pageio_init(desc, inode); } /** @@ -242,29 +260,23 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, * * Return 'true' if this is the case, else return 'false'. */ -static int nfs_can_coalesce_requests(struct nfs_page *prev, - struct nfs_page *req, - struct nfs_pageio_descriptor *pgio) +static bool nfs_can_coalesce_requests(struct nfs_page *prev, + struct nfs_page *req, + struct nfs_pageio_descriptor *pgio) { if (req->wb_context->cred != prev->wb_context->cred) - return 0; + return false; if (req->wb_lock_context->lockowner != prev->wb_lock_context->lockowner) - return 0; + return false; if (req->wb_context->state != prev->wb_context->state) - return 0; + return false; if (req->wb_index != (prev->wb_index + 1)) - return 0; + return false; if (req->wb_pgbase != 0) - return 0; + return false; if (prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE) - return 0; - /* - * Non-whole file layouts need to check that req is inside of - * pgio->pg_lseg. - */ - if (pgio->pg_test && !pgio->pg_test(pgio, prev, req)) - return 0; - return 1; + return false; + return pgio->pg_test(pgio, prev, req); } /** @@ -278,31 +290,18 @@ static int nfs_can_coalesce_requests(struct nfs_page *prev, static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc, struct nfs_page *req) { - size_t newlen = req->wb_bytes; - if (desc->pg_count != 0) { struct nfs_page *prev; - /* - * FIXME: ideally we should be able to coalesce all requests - * that are not block boundary aligned, but currently this - * is problematic for the case of bsize < PAGE_CACHE_SIZE, - * since nfs_flush_multi and nfs_pagein_multi assume you - * can have only one struct nfs_page. - */ - if (desc->pg_bsize < PAGE_SIZE) - return 0; - newlen += desc->pg_count; - if (newlen > desc->pg_bsize) - return 0; prev = nfs_list_entry(desc->pg_list.prev); if (!nfs_can_coalesce_requests(prev, req, desc)) return 0; - } else + } else { desc->pg_base = req->wb_pgbase; + } nfs_list_remove_request(req); nfs_list_add_request(req, &desc->pg_list); - desc->pg_count = newlen; + desc->pg_count += req->wb_bytes; return 1; } diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index f57f528..29c0ca7f 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -177,13 +177,28 @@ get_layout_hdr(struct pnfs_layout_hdr *lo) atomic_inc(&lo->plh_refcount); } +static struct pnfs_layout_hdr * +pnfs_alloc_layout_hdr(struct inode *ino, gfp_t gfp_flags) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(ino)->pnfs_curr_ld; + return ld->alloc_layout_hdr ? ld->alloc_layout_hdr(ino, gfp_flags) : + kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags); +} + +static void +pnfs_free_layout_hdr(struct pnfs_layout_hdr *lo) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(lo->plh_inode)->pnfs_curr_ld; + return ld->alloc_layout_hdr ? ld->free_layout_hdr(lo) : kfree(lo); +} + static void destroy_layout_hdr(struct pnfs_layout_hdr *lo) { dprintk("%s: freeing layout cache %p\n", __func__, lo); BUG_ON(!list_empty(&lo->plh_layouts)); NFS_I(lo->plh_inode)->layout = NULL; - kfree(lo); + pnfs_free_layout_hdr(lo); } static void @@ -228,7 +243,7 @@ put_lseg_common(struct pnfs_layout_segment *lseg) { struct inode *inode = lseg->pls_layout->plh_inode; - BUG_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); + WARN_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); list_del_init(&lseg->pls_list); if (list_empty(&lseg->pls_layout->plh_segs)) { set_bit(NFS_LAYOUT_DESTROYED, &lseg->pls_layout->plh_flags); @@ -261,11 +276,72 @@ put_lseg(struct pnfs_layout_segment *lseg) } EXPORT_SYMBOL_GPL(put_lseg); +static inline u64 +end_offset(u64 start, u64 len) +{ + u64 end; + + end = start + len; + return end >= start ? end : NFS4_MAX_UINT64; +} + +/* last octet in a range */ +static inline u64 +last_byte_offset(u64 start, u64 len) +{ + u64 end; + + BUG_ON(!len); + end = start + len; + return end > start ? end - 1 : NFS4_MAX_UINT64; +} + +/* + * is l2 fully contained in l1? + * start1 end1 + * [----------------------------------) + * start2 end2 + * [----------------) + */ +static inline int +lo_seg_contained(struct pnfs_layout_range *l1, + struct pnfs_layout_range *l2) +{ + u64 start1 = l1->offset; + u64 end1 = end_offset(start1, l1->length); + u64 start2 = l2->offset; + u64 end2 = end_offset(start2, l2->length); + + return (start1 <= start2) && (end1 >= end2); +} + +/* + * is l1 and l2 intersecting? + * start1 end1 + * [----------------------------------) + * start2 end2 + * [----------------) + */ +static inline int +lo_seg_intersecting(struct pnfs_layout_range *l1, + struct pnfs_layout_range *l2) +{ + u64 start1 = l1->offset; + u64 end1 = end_offset(start1, l1->length); + u64 start2 = l2->offset; + u64 end2 = end_offset(start2, l2->length); + + return (end1 == NFS4_MAX_UINT64 || end1 > start2) && + (end2 == NFS4_MAX_UINT64 || end2 > start1); +} + static bool -should_free_lseg(u32 lseg_iomode, u32 recall_iomode) +should_free_lseg(struct pnfs_layout_range *lseg_range, + struct pnfs_layout_range *recall_range) { - return (recall_iomode == IOMODE_ANY || - lseg_iomode == recall_iomode); + return (recall_range->iomode == IOMODE_ANY || + lseg_range->iomode == recall_range->iomode) && + lo_seg_intersecting(lseg_range, recall_range); } /* Returns 1 if lseg is removed from list, 0 otherwise */ @@ -296,7 +372,7 @@ static int mark_lseg_invalid(struct pnfs_layout_segment *lseg, int mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, struct list_head *tmp_list, - u32 iomode) + struct pnfs_layout_range *recall_range) { struct pnfs_layout_segment *lseg, *next; int invalid = 0, removed = 0; @@ -309,7 +385,8 @@ mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, return 0; } list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list) - if (should_free_lseg(lseg->pls_range.iomode, iomode)) { + if (!recall_range || + should_free_lseg(&lseg->pls_range, recall_range)) { dprintk("%s: freeing lseg %p iomode %d " "offset %llu length %llu\n", __func__, lseg, lseg->pls_range.iomode, lseg->pls_range.offset, @@ -358,7 +435,7 @@ pnfs_destroy_layout(struct nfs_inode *nfsi) lo = nfsi->layout; if (lo) { lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */ - mark_matching_lsegs_invalid(lo, &tmp_list, IOMODE_ANY); + mark_matching_lsegs_invalid(lo, &tmp_list, NULL); } spin_unlock(&nfsi->vfs_inode.i_lock); pnfs_free_lseg_list(&tmp_list); @@ -467,7 +544,7 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo, static struct pnfs_layout_segment * send_layoutget(struct pnfs_layout_hdr *lo, struct nfs_open_context *ctx, - u32 iomode, + struct pnfs_layout_range *range, gfp_t gfp_flags) { struct inode *ino = lo->plh_inode; @@ -499,11 +576,11 @@ send_layoutget(struct pnfs_layout_hdr *lo, goto out_err_free; } - lgp->args.minlength = NFS4_MAX_UINT64; + lgp->args.minlength = PAGE_CACHE_SIZE; + if (lgp->args.minlength > range->length) + lgp->args.minlength = range->length; lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE; - lgp->args.range.iomode = iomode; - lgp->args.range.offset = 0; - lgp->args.range.length = NFS4_MAX_UINT64; + lgp->args.range = *range; lgp->args.type = server->pnfs_curr_ld->id; lgp->args.inode = ino; lgp->args.ctx = get_nfs_open_context(ctx); @@ -518,7 +595,7 @@ send_layoutget(struct pnfs_layout_hdr *lo, nfs4_proc_layoutget(lgp); if (!lseg) { /* remember that LAYOUTGET failed and suspend trying */ - set_bit(lo_fail_bit(iomode), &lo->plh_flags); + set_bit(lo_fail_bit(range->iomode), &lo->plh_flags); } /* free xdr pages */ @@ -542,6 +619,56 @@ out_err_free: return NULL; } +/* Initiates a LAYOUTRETURN(FILE) */ +int +_pnfs_return_layout(struct inode *ino) +{ + struct pnfs_layout_hdr *lo = NULL; + struct nfs_inode *nfsi = NFS_I(ino); + LIST_HEAD(tmp_list); + struct nfs4_layoutreturn *lrp; + nfs4_stateid stateid; + int status = 0; + + dprintk("--> %s\n", __func__); + + spin_lock(&ino->i_lock); + lo = nfsi->layout; + if (!lo) { + spin_unlock(&ino->i_lock); + dprintk("%s: no layout to return\n", __func__); + return status; + } + stateid = nfsi->layout->plh_stateid; + /* Reference matched in nfs4_layoutreturn_release */ + get_layout_hdr(lo); + mark_matching_lsegs_invalid(lo, &tmp_list, NULL); + lo->plh_block_lgets++; + spin_unlock(&ino->i_lock); + pnfs_free_lseg_list(&tmp_list); + + WARN_ON(test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)); + + lrp = kzalloc(sizeof(*lrp), GFP_KERNEL); + if (unlikely(lrp == NULL)) { + status = -ENOMEM; + set_bit(NFS_LAYOUT_RW_FAILED, &lo->plh_flags); + set_bit(NFS_LAYOUT_RO_FAILED, &lo->plh_flags); + put_layout_hdr(lo); + goto out; + } + + lrp->args.stateid = stateid; + lrp->args.layout_type = NFS_SERVER(ino)->pnfs_curr_ld->id; + lrp->args.inode = ino; + lrp->clp = NFS_SERVER(ino)->nfs_client; + + status = nfs4_proc_layoutreturn(lrp); +out: + dprintk("<-- %s status: %d\n", __func__, status); + return status; +} + bool pnfs_roc(struct inode *ino) { struct pnfs_layout_hdr *lo; @@ -625,10 +752,23 @@ bool pnfs_roc_drain(struct inode *ino, u32 *barrier) * are seen first. */ static s64 -cmp_layout(u32 iomode1, u32 iomode2) +cmp_layout(struct pnfs_layout_range *l1, + struct pnfs_layout_range *l2) { + s64 d; + + /* high offset > low offset */ + d = l1->offset - l2->offset; + if (d) + return d; + + /* short length > long length */ + d = l2->length - l1->length; + if (d) + return d; + /* read > read/write */ - return (int)(iomode2 == IOMODE_READ) - (int)(iomode1 == IOMODE_READ); + return (int)(l1->iomode == IOMODE_READ) - (int)(l2->iomode == IOMODE_READ); } static void @@ -636,13 +776,12 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo, struct pnfs_layout_segment *lseg) { struct pnfs_layout_segment *lp; - int found = 0; dprintk("%s:Begin\n", __func__); assert_spin_locked(&lo->plh_inode->i_lock); list_for_each_entry(lp, &lo->plh_segs, pls_list) { - if (cmp_layout(lp->pls_range.iomode, lseg->pls_range.iomode) > 0) + if (cmp_layout(&lseg->pls_range, &lp->pls_range) > 0) continue; list_add_tail(&lseg->pls_list, &lp->pls_list); dprintk("%s: inserted lseg %p " @@ -652,16 +791,14 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo, lseg->pls_range.offset, lseg->pls_range.length, lp, lp->pls_range.iomode, lp->pls_range.offset, lp->pls_range.length); - found = 1; - break; - } - if (!found) { - list_add_tail(&lseg->pls_list, &lo->plh_segs); - dprintk("%s: inserted lseg %p " - "iomode %d offset %llu length %llu at tail\n", - __func__, lseg, lseg->pls_range.iomode, - lseg->pls_range.offset, lseg->pls_range.length); + goto out; } + list_add_tail(&lseg->pls_list, &lo->plh_segs); + dprintk("%s: inserted lseg %p " + "iomode %d offset %llu length %llu at tail\n", + __func__, lseg, lseg->pls_range.iomode, + lseg->pls_range.offset, lseg->pls_range.length); +out: get_layout_hdr(lo); dprintk("%s:Return\n", __func__); @@ -672,7 +809,7 @@ alloc_init_layout_hdr(struct inode *ino, gfp_t gfp_flags) { struct pnfs_layout_hdr *lo; - lo = kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags); + lo = pnfs_alloc_layout_hdr(ino, gfp_flags); if (!lo) return NULL; atomic_set(&lo->plh_refcount, 1); @@ -705,7 +842,7 @@ pnfs_find_alloc_layout(struct inode *ino, gfp_t gfp_flags) if (likely(nfsi->layout == NULL)) /* Won the race? */ nfsi->layout = new; else - kfree(new); + pnfs_free_layout_hdr(new); return nfsi->layout; } @@ -721,16 +858,28 @@ pnfs_find_alloc_layout(struct inode *ino, gfp_t gfp_flags) * READ RW true */ static int -is_matching_lseg(struct pnfs_layout_segment *lseg, u32 iomode) +is_matching_lseg(struct pnfs_layout_range *ls_range, + struct pnfs_layout_range *range) { - return (iomode != IOMODE_RW || lseg->pls_range.iomode == IOMODE_RW); + struct pnfs_layout_range range1; + + if ((range->iomode == IOMODE_RW && + ls_range->iomode != IOMODE_RW) || + !lo_seg_intersecting(ls_range, range)) + return 0; + + /* range1 covers only the first byte in the range */ + range1 = *range; + range1.length = 1; + return lo_seg_contained(ls_range, &range1); } /* * lookup range in layout */ static struct pnfs_layout_segment * -pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode) +pnfs_find_lseg(struct pnfs_layout_hdr *lo, + struct pnfs_layout_range *range) { struct pnfs_layout_segment *lseg, *ret = NULL; @@ -739,11 +888,11 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode) assert_spin_locked(&lo->plh_inode->i_lock); list_for_each_entry(lseg, &lo->plh_segs, pls_list) { if (test_bit(NFS_LSEG_VALID, &lseg->pls_flags) && - is_matching_lseg(lseg, iomode)) { + is_matching_lseg(&lseg->pls_range, range)) { ret = get_lseg(lseg); break; } - if (cmp_layout(iomode, lseg->pls_range.iomode) > 0) + if (lseg->pls_range.offset > range->offset) break; } @@ -759,9 +908,17 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode) struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, + loff_t pos, + u64 count, enum pnfs_iomode iomode, gfp_t gfp_flags) { + struct pnfs_layout_range arg = { + .iomode = iomode, + .offset = pos, + .length = count, + }; + unsigned pg_offset; struct nfs_inode *nfsi = NFS_I(ino); struct nfs_client *clp = NFS_SERVER(ino)->nfs_client; struct pnfs_layout_hdr *lo; @@ -789,7 +946,7 @@ pnfs_update_layout(struct inode *ino, goto out_unlock; /* Check to see if the layout for the given range already exists */ - lseg = pnfs_find_lseg(lo, iomode); + lseg = pnfs_find_lseg(lo, &arg); if (lseg) goto out_unlock; @@ -811,7 +968,14 @@ pnfs_update_layout(struct inode *ino, spin_unlock(&clp->cl_lock); } - lseg = send_layoutget(lo, ctx, iomode, gfp_flags); + pg_offset = arg.offset & ~PAGE_CACHE_MASK; + if (pg_offset) { + arg.offset -= pg_offset; + arg.length += pg_offset; + } + arg.length = PAGE_CACHE_ALIGN(arg.length); + + lseg = send_layoutget(lo, ctx, &arg, gfp_flags); if (!lseg && first) { spin_lock(&clp->cl_lock); list_del_init(&lo->plh_layouts); @@ -838,17 +1002,6 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) struct nfs_client *clp = NFS_SERVER(ino)->nfs_client; int status = 0; - /* Verify we got what we asked for. - * Note that because the xdr parsing only accepts a single - * element array, this can fail even if the server is behaving - * correctly. - */ - if (lgp->args.range.iomode > res->range.iomode || - res->range.offset != 0 || - res->range.length != NFS4_MAX_UINT64) { - status = -EINVAL; - goto out; - } /* Inject layout blob into I/O device driver */ lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res, lgp->gfp_flags); if (!lseg || IS_ERR(lseg)) { @@ -895,51 +1048,77 @@ out_forget_reply: goto out; } -static int pnfs_read_pg_test(struct nfs_pageio_descriptor *pgio, - struct nfs_page *prev, - struct nfs_page *req) -{ - if (pgio->pg_count == prev->wb_bytes) { - /* This is first coelesce call for a series of nfs_pages */ - pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, - prev->wb_context, - IOMODE_READ, - GFP_KERNEL); - } - return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req); -} - -void -pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode) +bool +pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, + struct nfs_page *req) { - struct pnfs_layoutdriver_type *ld; + enum pnfs_iomode access_type; + gfp_t gfp_flags; - ld = NFS_SERVER(inode)->pnfs_curr_ld; - pgio->pg_test = (ld && ld->pg_test) ? pnfs_read_pg_test : NULL; -} + /* We assume that pg_ioflags == 0 iff we're reading a page */ + if (pgio->pg_ioflags == 0) { + access_type = IOMODE_READ; + gfp_flags = GFP_KERNEL; + } else { + access_type = IOMODE_RW; + gfp_flags = GFP_NOFS; + } -static int pnfs_write_pg_test(struct nfs_pageio_descriptor *pgio, - struct nfs_page *prev, - struct nfs_page *req) -{ - if (pgio->pg_count == prev->wb_bytes) { + if (pgio->pg_lseg == NULL) { + if (pgio->pg_count != prev->wb_bytes) + return true; /* This is first coelesce call for a series of nfs_pages */ pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, prev->wb_context, - IOMODE_RW, - GFP_NOFS); + req_offset(prev), + pgio->pg_count, + access_type, + gfp_flags); + if (pgio->pg_lseg == NULL) + return true; } - return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req); + + /* + * Test if a nfs_page is fully contained in the pnfs_layout_range. + * Note that this test makes several assumptions: + * - that the previous nfs_page in the struct nfs_pageio_descriptor + * is known to lie within the range. + * - that the nfs_page being tested is known to be contiguous with the + * previous nfs_page. + * - Layout ranges are page aligned, so we only have to test the + * start offset of the request. + * + * Please also note that 'end_offset' is actually the offset of the + * first byte that lies outside the pnfs_layout_range. FIXME? + * + */ + return req_offset(req) < end_offset(pgio->pg_lseg->pls_range.offset, + pgio->pg_lseg->pls_range.length); } +EXPORT_SYMBOL_GPL(pnfs_generic_pg_test); -void -pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode) +/* + * Called by non rpc-based layout drivers + */ +int +pnfs_ld_write_done(struct nfs_write_data *data) { - struct pnfs_layoutdriver_type *ld; + int status; - ld = NFS_SERVER(inode)->pnfs_curr_ld; - pgio->pg_test = (ld && ld->pg_test) ? pnfs_write_pg_test : NULL; + if (!data->pnfs_error) { + pnfs_set_layoutcommit(data); + data->mds_ops->rpc_call_done(&data->task, data); + data->mds_ops->rpc_release(data); + return 0; + } + + dprintk("%s: pnfs_error=%d, retry via MDS\n", __func__, + data->pnfs_error); + status = nfs_initiate_write(data, NFS_CLIENT(data->inode), + data->mds_ops, NFS_FILE_SYNC); + return status ? : -EAGAIN; } +EXPORT_SYMBOL_GPL(pnfs_ld_write_done); enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *wdata, @@ -966,6 +1145,29 @@ pnfs_try_to_write_data(struct nfs_write_data *wdata, } /* + * Called by non rpc-based layout drivers + */ +int +pnfs_ld_read_done(struct nfs_read_data *data) +{ + int status; + + if (!data->pnfs_error) { + __nfs4_read_done_cb(data); + data->mds_ops->rpc_call_done(&data->task, data); + data->mds_ops->rpc_release(data); + return 0; + } + + dprintk("%s: pnfs_error=%d, retry via MDS\n", __func__, + data->pnfs_error); + status = nfs_initiate_read(data, NFS_CLIENT(data->inode), + data->mds_ops); + return status ? : -EAGAIN; +} +EXPORT_SYMBOL_GPL(pnfs_ld_read_done); + +/* * Call the appropriate parallel I/O subsystem read function. */ enum pnfs_try_status @@ -1009,7 +1211,7 @@ void pnfs_set_layoutcommit(struct nfs_write_data *wdata) { struct nfs_inode *nfsi = NFS_I(wdata->inode); - loff_t end_pos = wdata->args.offset + wdata->res.count; + loff_t end_pos = wdata->mds_offset + wdata->res.count; bool mark_as_dirty = false; spin_lock(&nfsi->vfs_inode.i_lock); diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 0c015ba..96bf4e6 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -30,6 +30,7 @@ #ifndef FS_NFS_PNFS_H #define FS_NFS_PNFS_H +#include <linux/nfs_fs.h> #include <linux/nfs_page.h> enum { @@ -64,17 +65,29 @@ enum { NFS_LAYOUT_DESTROYED, /* no new use of layout allowed */ }; +enum layoutdriver_policy_flags { + /* Should the pNFS client commit and return the layout upon a setattr */ + PNFS_LAYOUTRET_ON_SETATTR = 1 << 0, +}; + +struct nfs4_deviceid_node; + /* Per-layout driver specific registration structure */ struct pnfs_layoutdriver_type { struct list_head pnfs_tblid; const u32 id; const char *name; struct module *owner; + unsigned flags; + + struct pnfs_layout_hdr * (*alloc_layout_hdr) (struct inode *inode, gfp_t gfp_flags); + void (*free_layout_hdr) (struct pnfs_layout_hdr *); + struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags); void (*free_lseg) (struct pnfs_layout_segment *lseg); /* test for nfs page cache coalescing */ - int (*pg_test)(struct nfs_pageio_descriptor *, struct nfs_page *, struct nfs_page *); + bool (*pg_test)(struct nfs_pageio_descriptor *, struct nfs_page *, struct nfs_page *); /* Returns true if layoutdriver wants to divert this request to * driver's commit routine. @@ -89,6 +102,16 @@ struct pnfs_layoutdriver_type { */ enum pnfs_try_status (*read_pagelist) (struct nfs_read_data *nfs_data); enum pnfs_try_status (*write_pagelist) (struct nfs_write_data *nfs_data, int how); + + void (*free_deviceid_node) (struct nfs4_deviceid_node *); + + void (*encode_layoutreturn) (struct pnfs_layout_hdr *layoutid, + struct xdr_stream *xdr, + const struct nfs4_layoutreturn_args *args); + + void (*encode_layoutcommit) (struct pnfs_layout_hdr *layoutid, + struct xdr_stream *xdr, + const struct nfs4_layoutcommit_args *args); }; struct pnfs_layout_hdr { @@ -120,21 +143,22 @@ extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *); extern int nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *dev); extern int nfs4_proc_layoutget(struct nfs4_layoutget *lgp); +extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp); /* pnfs.c */ void get_layout_hdr(struct pnfs_layout_hdr *lo); void put_lseg(struct pnfs_layout_segment *lseg); struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, - enum pnfs_iomode access_type, gfp_t gfp_flags); + loff_t pos, u64 count, enum pnfs_iomode access_type, + gfp_t gfp_flags); void set_pnfs_layoutdriver(struct nfs_server *, u32 id); void unset_pnfs_layoutdriver(struct nfs_server *); enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *, const struct rpc_call_ops *, int); enum pnfs_try_status pnfs_try_to_read_data(struct nfs_read_data *, const struct rpc_call_ops *); -void pnfs_pageio_init_read(struct nfs_pageio_descriptor *, struct inode *); -void pnfs_pageio_init_write(struct nfs_pageio_descriptor *, struct inode *); +bool pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req); int pnfs_layout_process(struct nfs4_layoutget *lgp); void pnfs_free_lseg_list(struct list_head *tmp_list); void pnfs_destroy_layout(struct nfs_inode *); @@ -148,13 +172,38 @@ int pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct nfs4_state *open_state); int mark_matching_lsegs_invalid(struct pnfs_layout_hdr *lo, struct list_head *tmp_list, - u32 iomode); + struct pnfs_layout_range *recall_range); bool pnfs_roc(struct inode *ino); void pnfs_roc_release(struct inode *ino); void pnfs_roc_set_barrier(struct inode *ino, u32 barrier); bool pnfs_roc_drain(struct inode *ino, u32 *barrier); void pnfs_set_layoutcommit(struct nfs_write_data *wdata); int pnfs_layoutcommit_inode(struct inode *inode, bool sync); +int _pnfs_return_layout(struct inode *); +int pnfs_ld_write_done(struct nfs_write_data *); +int pnfs_ld_read_done(struct nfs_read_data *); + +/* pnfs_dev.c */ +struct nfs4_deviceid_node { + struct hlist_node node; + struct hlist_node tmpnode; + const struct pnfs_layoutdriver_type *ld; + const struct nfs_client *nfs_client; + struct nfs4_deviceid deviceid; + atomic_t ref; +}; + +void nfs4_print_deviceid(const struct nfs4_deviceid *dev_id); +struct nfs4_deviceid_node *nfs4_find_get_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *); +struct nfs4_deviceid_node *nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *); +void nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *, const struct nfs_client *, const struct nfs4_deviceid *); +void nfs4_init_deviceid_node(struct nfs4_deviceid_node *, + const struct pnfs_layoutdriver_type *, + const struct nfs_client *, + const struct nfs4_deviceid *); +struct nfs4_deviceid_node *nfs4_insert_deviceid_node(struct nfs4_deviceid_node *); +bool nfs4_put_deviceid_node(struct nfs4_deviceid_node *); +void nfs4_deviceid_purge_client(const struct nfs_client *); static inline int lo_fail_bit(u32 iomode) { @@ -223,6 +272,36 @@ static inline void pnfs_clear_request_commit(struct nfs_page *req) put_lseg(req->wb_commit_lseg); } +/* Should the pNFS client commit and return the layout upon a setattr */ +static inline bool +pnfs_ld_layoutret_on_setattr(struct inode *inode) +{ + if (!pnfs_enabled_sb(NFS_SERVER(inode))) + return false; + return NFS_SERVER(inode)->pnfs_curr_ld->flags & + PNFS_LAYOUTRET_ON_SETATTR; +} + +static inline int pnfs_return_layout(struct inode *ino) +{ + struct nfs_inode *nfsi = NFS_I(ino); + struct nfs_server *nfss = NFS_SERVER(ino); + + if (pnfs_enabled_sb(nfss) && nfsi->layout) + return _pnfs_return_layout(ino); + + return 0; +} + +static inline void pnfs_pageio_init(struct nfs_pageio_descriptor *pgio, + struct inode *inode) +{ + struct pnfs_layoutdriver_type *ld = NFS_SERVER(inode)->pnfs_curr_ld; + + if (ld) + pgio->pg_test = ld->pg_test; +} + #else /* CONFIG_NFS_V4_1 */ static inline void pnfs_destroy_all_layouts(struct nfs_client *clp) @@ -245,7 +324,8 @@ static inline void put_lseg(struct pnfs_layout_segment *lseg) static inline struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, - enum pnfs_iomode access_type, gfp_t gfp_flags) + loff_t pos, u64 count, enum pnfs_iomode access_type, + gfp_t gfp_flags) { return NULL; } @@ -264,6 +344,17 @@ pnfs_try_to_write_data(struct nfs_write_data *data, return PNFS_NOT_ATTEMPTED; } +static inline int pnfs_return_layout(struct inode *ino) +{ + return 0; +} + +static inline bool +pnfs_ld_layoutret_on_setattr(struct inode *inode) +{ + return false; +} + static inline bool pnfs_roc(struct inode *ino) { @@ -294,16 +385,9 @@ static inline void unset_pnfs_layoutdriver(struct nfs_server *s) { } -static inline void -pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *ino) -{ - pgio->pg_test = NULL; -} - -static inline void -pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *ino) +static inline void pnfs_pageio_init(struct nfs_pageio_descriptor *pgio, + struct inode *inode) { - pgio->pg_test = NULL; } static inline void @@ -331,6 +415,10 @@ static inline int pnfs_layoutcommit_inode(struct inode *inode, bool sync) { return 0; } + +static inline void nfs4_deviceid_purge_client(struct nfs_client *ncl) +{ +} #endif /* CONFIG_NFS_V4_1 */ #endif /* FS_NFS_PNFS_H */ diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c new file mode 100644 index 0000000..f0f8e1e --- /dev/null +++ b/fs/nfs/pnfs_dev.c @@ -0,0 +1,277 @@ +/* + * Device operations for the pnfs client. + * + * Copyright (c) 2002 + * The Regents of the University of Michigan + * All Rights Reserved + * + * Dean Hildebrand <dhildebz@umich.edu> + * Garth Goodson <Garth.Goodson@netapp.com> + * + * Permission is granted to use, copy, create derivative works, and + * redistribute this software and such derivative works for any purpose, + * so long as the name of the University of Michigan is not used in + * any advertising or publicity pertaining to the use or distribution + * of this software without specific, written prior authorization. If + * the above copyright notice or any other identification of the + * University of Michigan is included in any copy of any portion of + * this software, then the disclaimer below must also be included. + * + * This software is provided as is, without representation or warranty + * of any kind either express or implied, including without limitation + * the implied warranties of merchantability, fitness for a particular + * purpose, or noninfringement. The Regents of the University of + * Michigan shall not be liable for any damages, including special, + * indirect, incidental, or consequential damages, with respect to any + * claim arising out of or in connection with the use of the software, + * even if it has been or is hereafter advised of the possibility of + * such damages. + */ + +#include "pnfs.h" + +#define NFSDBG_FACILITY NFSDBG_PNFS + +/* + * Device ID RCU cache. A device ID is unique per server and layout type. + */ +#define NFS4_DEVICE_ID_HASH_BITS 5 +#define NFS4_DEVICE_ID_HASH_SIZE (1 << NFS4_DEVICE_ID_HASH_BITS) +#define NFS4_DEVICE_ID_HASH_MASK (NFS4_DEVICE_ID_HASH_SIZE - 1) + +static struct hlist_head nfs4_deviceid_cache[NFS4_DEVICE_ID_HASH_SIZE]; +static DEFINE_SPINLOCK(nfs4_deviceid_lock); + +void +nfs4_print_deviceid(const struct nfs4_deviceid *id) +{ + u32 *p = (u32 *)id; + + dprintk("%s: device id= [%x%x%x%x]\n", __func__, + p[0], p[1], p[2], p[3]); +} +EXPORT_SYMBOL_GPL(nfs4_print_deviceid); + +static inline u32 +nfs4_deviceid_hash(const struct nfs4_deviceid *id) +{ + unsigned char *cptr = (unsigned char *)id->data; + unsigned int nbytes = NFS4_DEVICEID4_SIZE; + u32 x = 0; + + while (nbytes--) { + x *= 37; + x += *cptr++; + } + return x & NFS4_DEVICE_ID_HASH_MASK; +} + +static struct nfs4_deviceid_node * +_lookup_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id, + long hash) +{ + struct nfs4_deviceid_node *d; + struct hlist_node *n; + + hlist_for_each_entry_rcu(d, n, &nfs4_deviceid_cache[hash], node) + if (d->ld == ld && d->nfs_client == clp && + !memcmp(&d->deviceid, id, sizeof(*id))) { + if (atomic_read(&d->ref)) + return d; + else + continue; + } + return NULL; +} + +/* + * Lookup a deviceid in cache and get a reference count on it if found + * + * @clp nfs_client associated with deviceid + * @id deviceid to look up + */ +struct nfs4_deviceid_node * +_find_get_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id, + long hash) +{ + struct nfs4_deviceid_node *d; + + rcu_read_lock(); + d = _lookup_deviceid(ld, clp, id, hash); + if (d && !atomic_inc_not_zero(&d->ref)) + d = NULL; + rcu_read_unlock(); + return d; +} + +struct nfs4_deviceid_node * +nfs4_find_get_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id) +{ + return _find_get_deviceid(ld, clp, id, nfs4_deviceid_hash(id)); +} +EXPORT_SYMBOL_GPL(nfs4_find_get_deviceid); + +/* + * Unhash and put deviceid + * + * @clp nfs_client associated with deviceid + * @id the deviceid to unhash + * + * @ret the unhashed node, if found and dereferenced to zero, NULL otherwise. + */ +struct nfs4_deviceid_node * +nfs4_unhash_put_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id) +{ + struct nfs4_deviceid_node *d; + + spin_lock(&nfs4_deviceid_lock); + rcu_read_lock(); + d = _lookup_deviceid(ld, clp, id, nfs4_deviceid_hash(id)); + rcu_read_unlock(); + if (!d) { + spin_unlock(&nfs4_deviceid_lock); + return NULL; + } + hlist_del_init_rcu(&d->node); + spin_unlock(&nfs4_deviceid_lock); + synchronize_rcu(); + + /* balance the initial ref set in pnfs_insert_deviceid */ + if (atomic_dec_and_test(&d->ref)) + return d; + + return NULL; +} +EXPORT_SYMBOL_GPL(nfs4_unhash_put_deviceid); + +/* + * Delete a deviceid from cache + * + * @clp struct nfs_client qualifying the deviceid + * @id deviceid to delete + */ +void +nfs4_delete_deviceid(const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *clp, const struct nfs4_deviceid *id) +{ + struct nfs4_deviceid_node *d; + + d = nfs4_unhash_put_deviceid(ld, clp, id); + if (!d) + return; + d->ld->free_deviceid_node(d); +} +EXPORT_SYMBOL_GPL(nfs4_delete_deviceid); + +void +nfs4_init_deviceid_node(struct nfs4_deviceid_node *d, + const struct pnfs_layoutdriver_type *ld, + const struct nfs_client *nfs_client, + const struct nfs4_deviceid *id) +{ + INIT_HLIST_NODE(&d->node); + INIT_HLIST_NODE(&d->tmpnode); + d->ld = ld; + d->nfs_client = nfs_client; + d->deviceid = *id; + atomic_set(&d->ref, 1); +} +EXPORT_SYMBOL_GPL(nfs4_init_deviceid_node); + +/* + * Uniquely initialize and insert a deviceid node into cache + * + * @new new deviceid node + * Note that the caller must set up the following members: + * new->ld + * new->nfs_client + * new->deviceid + * + * @ret the inserted node, if none found, otherwise, the found entry. + */ +struct nfs4_deviceid_node * +nfs4_insert_deviceid_node(struct nfs4_deviceid_node *new) +{ + struct nfs4_deviceid_node *d; + long hash; + + spin_lock(&nfs4_deviceid_lock); + hash = nfs4_deviceid_hash(&new->deviceid); + d = _find_get_deviceid(new->ld, new->nfs_client, &new->deviceid, hash); + if (d) { + spin_unlock(&nfs4_deviceid_lock); + return d; + } + + hlist_add_head_rcu(&new->node, &nfs4_deviceid_cache[hash]); + spin_unlock(&nfs4_deviceid_lock); + atomic_inc(&new->ref); + + return new; +} +EXPORT_SYMBOL_GPL(nfs4_insert_deviceid_node); + +/* + * Dereference a deviceid node and delete it when its reference count drops + * to zero. + * + * @d deviceid node to put + * + * @ret true iff the node was deleted + */ +bool +nfs4_put_deviceid_node(struct nfs4_deviceid_node *d) +{ + if (!atomic_dec_and_lock(&d->ref, &nfs4_deviceid_lock)) + return false; + hlist_del_init_rcu(&d->node); + spin_unlock(&nfs4_deviceid_lock); + synchronize_rcu(); + d->ld->free_deviceid_node(d); + return true; +} +EXPORT_SYMBOL_GPL(nfs4_put_deviceid_node); + +static void +_deviceid_purge_client(const struct nfs_client *clp, long hash) +{ + struct nfs4_deviceid_node *d; + struct hlist_node *n; + HLIST_HEAD(tmp); + + spin_lock(&nfs4_deviceid_lock); + rcu_read_lock(); + hlist_for_each_entry_rcu(d, n, &nfs4_deviceid_cache[hash], node) + if (d->nfs_client == clp && atomic_read(&d->ref)) { + hlist_del_init_rcu(&d->node); + hlist_add_head(&d->tmpnode, &tmp); + } + rcu_read_unlock(); + spin_unlock(&nfs4_deviceid_lock); + + if (hlist_empty(&tmp)) + return; + + synchronize_rcu(); + while (!hlist_empty(&tmp)) { + d = hlist_entry(tmp.first, struct nfs4_deviceid_node, tmpnode); + hlist_del(&d->tmpnode); + if (atomic_dec_and_test(&d->ref)) + d->ld->free_deviceid_node(d); + } +} + +void +nfs4_deviceid_purge_client(const struct nfs_client *clp) +{ + long h; + + if (!(clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_MDS)) + return; + for (h = 0; h < NFS4_DEVICE_ID_HASH_SIZE; h++) + _deviceid_purge_client(clp, h); +} diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 2bcf0dc..20a7f95 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -288,7 +288,9 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc) atomic_set(&req->wb_complete, requests); BUG_ON(desc->pg_lseg != NULL); - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, + req_offset(req), desc->pg_count, + IOMODE_READ, GFP_KERNEL); ClearPageError(page); offset = 0; nbytes = desc->pg_count; @@ -351,7 +353,9 @@ static int nfs_pagein_one(struct nfs_pageio_descriptor *desc) } req = nfs_list_entry(data->pages.next); if ((!lseg) && list_is_singular(&data->pages)) - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, + req_offset(req), desc->pg_count, + IOMODE_READ, GFP_KERNEL); ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count, 0, lseg); @@ -660,7 +664,6 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, if (ret == 0) goto read_complete; /* all pages were read */ - pnfs_pageio_init_read(&pgio, inode); if (rsize < PAGE_CACHE_SIZE) nfs_pageio_init(&pgio, inode, nfs_pagein_multi, rsize, 0); else diff --git a/fs/nfs/super.c b/fs/nfs/super.c index e288f06..ce40e5c 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -63,6 +63,7 @@ #include "iostat.h" #include "internal.h" #include "fscache.h" +#include "pnfs.h" #define NFSDBG_FACILITY NFSDBG_VFS @@ -732,6 +733,28 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) return 0; } +#ifdef CONFIG_NFS_V4_1 +void show_sessions(struct seq_file *m, struct nfs_server *server) +{ + if (nfs4_has_session(server->nfs_client)) + seq_printf(m, ",sessions"); +} +#else +void show_sessions(struct seq_file *m, struct nfs_server *server) {} +#endif + +#ifdef CONFIG_NFS_V4_1 +void show_pnfs(struct seq_file *m, struct nfs_server *server) +{ + seq_printf(m, ",pnfs="); + if (server->pnfs_curr_ld) + seq_printf(m, "%s", server->pnfs_curr_ld->name); + else + seq_printf(m, "not configured"); +} +#else /* CONFIG_NFS_V4_1 */ +void show_pnfs(struct seq_file *m, struct nfs_server *server) {} +#endif /* CONFIG_NFS_V4_1 */ static int nfs_show_devname(struct seq_file *m, struct vfsmount *mnt) { @@ -792,6 +815,8 @@ static int nfs_show_stats(struct seq_file *m, struct vfsmount *mnt) seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]); seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]); seq_printf(m, ",acl=0x%x", nfss->acl_bitmask); + show_sessions(m, nfss); + show_pnfs(m, nfss); } #endif diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 49c715b..e268e3b 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -939,7 +939,9 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc) atomic_set(&req->wb_complete, requests); BUG_ON(desc->pg_lseg); - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, + req_offset(req), desc->pg_count, + IOMODE_RW, GFP_NOFS); ClearPageError(page); offset = 0; nbytes = desc->pg_count; @@ -1013,7 +1015,9 @@ static int nfs_flush_one(struct nfs_pageio_descriptor *desc) } req = nfs_list_entry(data->pages.next); if ((!lseg) && list_is_singular(&data->pages)) - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, + req_offset(req), desc->pg_count, + IOMODE_RW, GFP_NOFS); if ((desc->pg_ioflags & FLUSH_COND_STABLE) && (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit)) @@ -1032,8 +1036,6 @@ static void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, { size_t wsize = NFS_SERVER(inode)->wsize; - pnfs_pageio_init_write(pgio, inode); - if (wsize < PAGE_CACHE_SIZE) nfs_pageio_init(pgio, inode, nfs_flush_multi, wsize, ioflags); else diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig index 18b3e89..fbb2a5e 100644 --- a/fs/nfsd/Kconfig +++ b/fs/nfsd/Kconfig @@ -82,6 +82,7 @@ config NFSD_V4 select NFSD_V3 select FS_POSIX_ACL select SUNRPC_GSS + select CRYPTO help This option enables support in your system's NFS server for version 4 of the NFS protocol (RFC 3530). diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index ad000ae..b9566e4 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1354,12 +1354,6 @@ exp_pseudoroot(struct svc_rqst *rqstp, struct svc_fh *fhp) if (IS_ERR(exp)) return nfserrno(PTR_ERR(exp)); rv = fh_compose(fhp, exp, exp->ex_path.dentry, NULL); - if (rv) - goto out; - rv = check_nfsd_access(exp, rqstp); - if (rv) - fh_put(fhp); -out: exp_put(exp); return rv; } diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 2247fc9..9095f3c 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -245,7 +245,7 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, } /* Now create the file and set attributes */ - nfserr = nfsd_create_v3(rqstp, dirfhp, argp->name, argp->len, + nfserr = do_nfsd_create(rqstp, dirfhp, argp->name, argp->len, attr, newfhp, argp->createmode, argp->verf, NULL, NULL); diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index ad48fac..08c6e36 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -842,7 +842,7 @@ out: return rv; } -__be32 *encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen) +static __be32 *encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen) { struct svc_fh fh; int err; diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 5fcb139..3a6dbd7 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -196,9 +196,9 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o /* * Note: create modes (UNCHECKED,GUARDED...) are the same - * in NFSv4 as in v3. + * in NFSv4 as in v3 except EXCLUSIVE4_1. */ - status = nfsd_create_v3(rqstp, current_fh, open->op_fname.data, + status = do_nfsd_create(rqstp, current_fh, open->op_fname.data, open->op_fname.len, &open->op_iattr, &resfh, open->op_createmode, (u32 *)open->op_verf.data, @@ -403,7 +403,7 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, cstate->current_fh.fh_handle.fh_size = putfh->pf_fhlen; memcpy(&cstate->current_fh.fh_handle.fh_base, putfh->pf_fhval, putfh->pf_fhlen); - return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_NOP); + return fh_verify(rqstp, &cstate->current_fh, 0, NFSD_MAY_BYPASS_GSS); } static __be32 @@ -762,6 +762,9 @@ nfsd4_secinfo(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, __be32 err; fh_init(&resfh, NFS4_FHSIZE); + err = fh_verify(rqstp, &cstate->current_fh, S_IFDIR, NFSD_MAY_EXEC); + if (err) + return err; err = nfsd_lookup_dentry(rqstp, &cstate->current_fh, secinfo->si_name, secinfo->si_namelen, &exp, &dentry); @@ -986,6 +989,9 @@ enum nfsd4_op_flags { ALLOWED_WITHOUT_FH = 1 << 0, /* No current filehandle required */ ALLOWED_ON_ABSENT_FS = 1 << 1, /* ops processed on absent fs */ ALLOWED_AS_FIRST_OP = 1 << 2, /* ops reqired first in compound */ + /* For rfc 5661 section 2.6.3.1.1: */ + OP_HANDLES_WRONGSEC = 1 << 3, + OP_IS_PUTFH_LIKE = 1 << 4, }; struct nfsd4_operation { @@ -1031,6 +1037,44 @@ static __be32 nfs41_check_op_ordering(struct nfsd4_compoundargs *args) return nfs_ok; } +static inline struct nfsd4_operation *OPDESC(struct nfsd4_op *op) +{ + return &nfsd4_ops[op->opnum]; +} + +static bool need_wrongsec_check(struct svc_rqst *rqstp) +{ + struct nfsd4_compoundres *resp = rqstp->rq_resp; + struct nfsd4_compoundargs *argp = rqstp->rq_argp; + struct nfsd4_op *this = &argp->ops[resp->opcnt - 1]; + struct nfsd4_op *next = &argp->ops[resp->opcnt]; + struct nfsd4_operation *thisd; + struct nfsd4_operation *nextd; + + thisd = OPDESC(this); + /* + * Most ops check wronsec on our own; only the putfh-like ops + * have special rules. + */ + if (!(thisd->op_flags & OP_IS_PUTFH_LIKE)) + return false; + /* + * rfc 5661 2.6.3.1.1.6: don't bother erroring out a + * put-filehandle operation if we're not going to use the + * result: + */ + if (argp->opcnt == resp->opcnt) + return false; + + nextd = OPDESC(next); + /* + * Rest of 2.6.3.1.1: certain operations will return WRONGSEC + * errors themselves as necessary; others should check for them + * now: + */ + return !(nextd->op_flags & OP_HANDLES_WRONGSEC); +} + /* * COMPOUND call. */ @@ -1108,7 +1152,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, goto encode_op; } - opdesc = &nfsd4_ops[op->opnum]; + opdesc = OPDESC(op); if (!cstate->current_fh.fh_dentry) { if (!(opdesc->op_flags & ALLOWED_WITHOUT_FH)) { @@ -1126,6 +1170,9 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, else BUG_ON(op->status == nfs_ok); + if (!op->status && need_wrongsec_check(rqstp)) + op->status = check_nfsd_access(cstate->current_fh.fh_export, rqstp); + encode_op: /* Only from SEQUENCE */ if (resp->cstate.status == nfserr_replay_cache) { @@ -1217,10 +1264,12 @@ static struct nfsd4_operation nfsd4_ops[] = { }, [OP_LOOKUP] = { .op_func = (nfsd4op_func)nfsd4_lookup, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_LOOKUP", }, [OP_LOOKUPP] = { .op_func = (nfsd4op_func)nfsd4_lookupp, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_LOOKUPP", }, [OP_NVERIFY] = { @@ -1229,6 +1278,7 @@ static struct nfsd4_operation nfsd4_ops[] = { }, [OP_OPEN] = { .op_func = (nfsd4op_func)nfsd4_open, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_OPEN", }, [OP_OPEN_CONFIRM] = { @@ -1241,17 +1291,20 @@ static struct nfsd4_operation nfsd4_ops[] = { }, [OP_PUTFH] = { .op_func = (nfsd4op_func)nfsd4_putfh, - .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS, + .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS + | OP_IS_PUTFH_LIKE, .op_name = "OP_PUTFH", }, [OP_PUTPUBFH] = { .op_func = (nfsd4op_func)nfsd4_putrootfh, - .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS, + .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS + | OP_IS_PUTFH_LIKE, .op_name = "OP_PUTPUBFH", }, [OP_PUTROOTFH] = { .op_func = (nfsd4op_func)nfsd4_putrootfh, - .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS, + .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS + | OP_IS_PUTFH_LIKE, .op_name = "OP_PUTROOTFH", }, [OP_READ] = { @@ -1281,15 +1334,18 @@ static struct nfsd4_operation nfsd4_ops[] = { }, [OP_RESTOREFH] = { .op_func = (nfsd4op_func)nfsd4_restorefh, - .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS, + .op_flags = ALLOWED_WITHOUT_FH | ALLOWED_ON_ABSENT_FS + | OP_IS_PUTFH_LIKE, .op_name = "OP_RESTOREFH", }, [OP_SAVEFH] = { .op_func = (nfsd4op_func)nfsd4_savefh, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_SAVEFH", }, [OP_SECINFO] = { .op_func = (nfsd4op_func)nfsd4_secinfo, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_SECINFO", }, [OP_SETATTR] = { @@ -1353,6 +1409,7 @@ static struct nfsd4_operation nfsd4_ops[] = { }, [OP_SECINFO_NO_NAME] = { .op_func = (nfsd4op_func)nfsd4_secinfo_no_name, + .op_flags = OP_HANDLES_WRONGSEC, .op_name = "OP_SECINFO_NO_NAME", }, }; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 4cf04e1..e98f3c2 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1519,6 +1519,9 @@ nfsd4_create_session(struct svc_rqst *rqstp, bool confirm_me = false; int status = 0; + if (cr_ses->flags & ~SESSION4_FLAG_MASK_A) + return nfserr_inval; + nfs4_lock_state(); unconf = find_unconfirmed_client(&cr_ses->clientid); conf = find_confirmed_client(&cr_ses->clientid); @@ -1637,8 +1640,9 @@ __be32 nfsd4_bind_conn_to_session(struct svc_rqst *rqstp, return nfserr_badsession; status = nfsd4_map_bcts_dir(&bcts->dir); - nfsd4_new_conn(rqstp, cstate->session, bcts->dir); - return nfs_ok; + if (!status) + nfsd4_new_conn(rqstp, cstate->session, bcts->dir); + return status; } static bool nfsd4_compound_in_session(struct nfsd4_session *session, struct nfs4_sessionid *sid) @@ -1725,6 +1729,13 @@ static void nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_sessi return; } +static bool nfsd4_session_too_many_ops(struct svc_rqst *rqstp, struct nfsd4_session *session) +{ + struct nfsd4_compoundargs *args = rqstp->rq_argp; + + return args->opcnt > session->se_fchannel.maxops; +} + __be32 nfsd4_sequence(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, @@ -1753,6 +1764,10 @@ nfsd4_sequence(struct svc_rqst *rqstp, if (!session) goto out; + status = nfserr_too_many_ops; + if (nfsd4_session_too_many_ops(rqstp, session)) + goto out; + status = nfserr_badslot; if (seq->slotid >= session->se_fchannel.maxreqs) goto out; @@ -1808,6 +1823,8 @@ out: __be32 nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_reclaim_complete *rc) { + int status = 0; + if (rc->rca_one_fs) { if (!cstate->current_fh.fh_dentry) return nfserr_nofilehandle; @@ -1817,9 +1834,14 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta */ return nfs_ok; } + nfs4_lock_state(); - if (is_client_expired(cstate->session->se_client)) { - nfs4_unlock_state(); + status = nfserr_complete_already; + if (cstate->session->se_client->cl_firststate) + goto out; + + status = nfserr_stale_clientid; + if (is_client_expired(cstate->session->se_client)) /* * The following error isn't really legal. * But we only get here if the client just explicitly @@ -1827,11 +1849,13 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *csta * error it gets back on an operation for the dead * client. */ - return nfserr_stale_clientid; - } + goto out; + + status = nfs_ok; nfsd4_create_clid_dir(cstate->session->se_client); +out: nfs4_unlock_state(); - return nfs_ok; + return status; } __be32 @@ -2462,7 +2486,7 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) return NULL; } -int share_access_to_flags(u32 share_access) +static int share_access_to_flags(u32 share_access) { share_access &= ~NFS4_SHARE_WANT_MASK; @@ -2882,7 +2906,7 @@ out: return status; } -struct lock_manager nfsd4_manager = { +static struct lock_manager nfsd4_manager = { }; static void diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index c6766af..9901811 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -424,15 +424,12 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp, struct nfsd4_bind_conn_to_session *bcts) { DECODE_HEAD; - u32 dummy; READ_BUF(NFS4_MAX_SESSIONID_LEN + 8); COPYMEM(bcts->sessionid.data, NFS4_MAX_SESSIONID_LEN); READ32(bcts->dir); - /* XXX: Perhaps Tom Tucker could help us figure out how we - * should be using ctsa_use_conn_in_rdma_mode: */ - READ32(dummy); - + /* XXX: skipping ctsa_use_conn_in_rdma_mode. Perhaps Tom Tucker + * could help us figure out we should be using it. */ DECODE_TAIL; } @@ -588,8 +585,6 @@ nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) READ_BUF(lockt->lt_owner.len); READMEM(lockt->lt_owner.data, lockt->lt_owner.len); - if (argp->minorversion && !zero_clientid(&lockt->lt_clientid)) - return nfserr_inval; DECODE_TAIL; } @@ -3120,7 +3115,7 @@ nfsd4_encode_destroy_session(struct nfsd4_compoundres *resp, int nfserr, return nfserr; } -__be32 +static __be32 nfsd4_encode_sequence(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_sequence *seq) { diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 1f5eae4..2b1449d 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -13,6 +13,7 @@ #include <linux/lockd/lockd.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/gss_api.h> +#include <linux/sunrpc/gss_krb5_enctypes.h> #include "idmap.h" #include "nfsd.h" @@ -189,18 +190,10 @@ static struct file_operations export_features_operations = { .release = single_release, }; -#ifdef CONFIG_SUNRPC_GSS +#if defined(CONFIG_SUNRPC_GSS) || defined(CONFIG_SUNRPC_GSS_MODULE) static int supported_enctypes_show(struct seq_file *m, void *v) { - struct gss_api_mech *k5mech; - - k5mech = gss_mech_get_by_name("krb5"); - if (k5mech == NULL) - goto out; - if (k5mech->gm_upcall_enctypes != NULL) - seq_printf(m, k5mech->gm_upcall_enctypes); - gss_mech_put(k5mech); -out: + seq_printf(m, KRB5_SUPPORTED_ENCTYPES); return 0; } @@ -215,7 +208,7 @@ static struct file_operations supported_enctypes_ops = { .llseek = seq_lseek, .release = single_release, }; -#endif /* CONFIG_SUNRPC_GSS */ +#endif /* CONFIG_SUNRPC_GSS or CONFIG_SUNRPC_GSS_MODULE */ extern int nfsd_pool_stats_open(struct inode *inode, struct file *file); extern int nfsd_pool_stats_release(struct inode *inode, struct file *file); @@ -1427,9 +1420,9 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent) [NFSD_Versions] = {"versions", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Ports] = {"portlist", &transaction_ops, S_IWUSR|S_IRUGO}, [NFSD_MaxBlkSize] = {"max_block_size", &transaction_ops, S_IWUSR|S_IRUGO}, -#ifdef CONFIG_SUNRPC_GSS +#if defined(CONFIG_SUNRPC_GSS) || defined(CONFIG_SUNRPC_GSS_MODULE) [NFSD_SupportedEnctypes] = {"supported_krb5_enctypes", &supported_enctypes_ops, S_IRUGO}, -#endif /* CONFIG_SUNRPC_GSS */ +#endif /* CONFIG_SUNRPC_GSS or CONFIG_SUNRPC_GSS_MODULE */ #ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Gracetime] = {"nfsv4gracetime", &transaction_ops, S_IWUSR|S_IRUSR}, diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 55c8e63..90c6aa6 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -344,7 +344,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) * which clients virtually always use auth_sys for, * even while using RPCSEC_GSS for NFS. */ - if (access & NFSD_MAY_LOCK) + if (access & NFSD_MAY_LOCK || access & NFSD_MAY_BYPASS_GSS) goto skip_pseudoflavor_check; /* * Clients may expect to be able to use auth_sys during mount, diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c index 5232d3e..a2e2402 100644 --- a/fs/nfsd/stats.c +++ b/fs/nfsd/stats.c @@ -8,7 +8,7 @@ * Statistsics for the reply cache * fh <stale> <total-lookups> <anonlookups> <dir-not-in-dcache> <nondir-not-in-dcache> * statistics for filehandle lookup - * io <bytes-read> <bytes-writtten> + * io <bytes-read> <bytes-written> * statistics for IO throughput * th <threads> <fullcnt> <10%-20%> <20%-30%> ... <90%-100%> <100%> * time (seconds) when nfsd thread usage above thresholds diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 129f3c9..fd0acca 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -181,16 +181,10 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp, struct svc_export *exp; struct dentry *dparent; struct dentry *dentry; - __be32 err; int host_err; dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name); - /* Obtain dentry and export. */ - err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC); - if (err) - return err; - dparent = fhp->fh_dentry; exp = fhp->fh_export; exp_get(exp); @@ -254,6 +248,9 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, struct dentry *dentry; __be32 err; + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC); + if (err) + return err; err = nfsd_lookup_dentry(rqstp, fhp, name, len, &exp, &dentry); if (err) return err; @@ -699,7 +696,15 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor } #endif /* CONFIG_NFSD_V3 */ +static int nfsd_open_break_lease(struct inode *inode, int access) +{ + unsigned int mode; + if (access & NFSD_MAY_NOT_BREAK_LEASE) + return 0; + mode = (access & NFSD_MAY_WRITE) ? O_WRONLY : O_RDONLY; + return break_lease(inode, mode | O_NONBLOCK); +} /* * Open an existing file or directory. @@ -747,12 +752,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (!inode->i_fop) goto out; - /* - * Check to see if there are any leases on this file. - * This may block while leases are broken. - */ - if (!(access & NFSD_MAY_NOT_BREAK_LEASE)) - host_err = break_lease(inode, O_NONBLOCK | ((access & NFSD_MAY_WRITE) ? O_WRONLY : 0)); + host_err = nfsd_open_break_lease(inode, access); if (host_err) /* NOMEM or WOULDBLOCK */ goto out_nfserr; @@ -877,13 +877,11 @@ static __be32 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { - struct inode *inode; mm_segment_t oldfs; __be32 err; int host_err; err = nfserr_perm; - inode = file->f_path.dentry->d_inode; if (file->f_op->splice_read && rqstp->rq_splice_ok) { struct splice_desc sd = { @@ -1340,11 +1338,18 @@ out_nfserr: } #ifdef CONFIG_NFSD_V3 + +static inline int nfsd_create_is_exclusive(int createmode) +{ + return createmode == NFS3_CREATE_EXCLUSIVE + || createmode == NFS4_CREATE_EXCLUSIVE4_1; +} + /* - * NFSv3 version of nfsd_create + * NFSv3 and NFSv4 version of nfsd_create */ __be32 -nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, +do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, int *truncp, int *created) @@ -1396,7 +1401,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, if (err) goto out; - if (createmode == NFS3_CREATE_EXCLUSIVE) { + if (nfsd_create_is_exclusive(createmode)) { /* solaris7 gets confused (bugid 4218508) if these have * the high bit set, so just clear the high bits. If this is * ever changed to use different attrs for storing the @@ -1437,6 +1442,11 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, && dchild->d_inode->i_atime.tv_sec == v_atime && dchild->d_inode->i_size == 0 ) break; + case NFS4_CREATE_EXCLUSIVE4_1: + if ( dchild->d_inode->i_mtime.tv_sec == v_mtime + && dchild->d_inode->i_atime.tv_sec == v_atime + && dchild->d_inode->i_size == 0 ) + goto set_attr; /* fallthru */ case NFS3_CREATE_GUARDED: err = nfserr_exist; @@ -1455,7 +1465,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, nfsd_check_ignore_resizing(iap); - if (createmode == NFS3_CREATE_EXCLUSIVE) { + if (nfsd_create_is_exclusive(createmode)) { /* Cram the verifier into atime/mtime */ iap->ia_valid = ATTR_MTIME|ATTR_ATIME | ATTR_MTIME_SET|ATTR_ATIME_SET; @@ -1653,8 +1663,10 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, if (!dold->d_inode) goto out_drop_write; host_err = nfsd_break_lease(dold->d_inode); - if (host_err) + if (host_err) { + err = nfserrno(host_err); goto out_drop_write; + } host_err = vfs_link(dold, dirp, dnew); if (!host_err) { err = nfserrno(commit_metadata(ffhp)); @@ -2034,7 +2046,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, struct inode *inode = dentry->d_inode; int err; - if (acc == NFSD_MAY_NOP) + if ((acc & NFSD_MAY_MASK) == NFSD_MAY_NOP) return 0; #if 0 dprintk("nfsd: permission 0x%x%s%s%s%s%s%s%s mode 0%o%s%s%s\n", diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h index 9a370a5..e0bbac0 100644 --- a/fs/nfsd/vfs.h +++ b/fs/nfsd/vfs.h @@ -17,10 +17,14 @@ #define NFSD_MAY_SATTR 8 #define NFSD_MAY_TRUNC 16 #define NFSD_MAY_LOCK 32 +#define NFSD_MAY_MASK 63 + +/* extra hints to permission and open routines: */ #define NFSD_MAY_OWNER_OVERRIDE 64 #define NFSD_MAY_LOCAL_ACCESS 128 /* IRIX doing local access check on device special file*/ #define NFSD_MAY_BYPASS_GSS_ON_ROOT 256 #define NFSD_MAY_NOT_BREAK_LEASE 512 +#define NFSD_MAY_BYPASS_GSS 1024 #define NFSD_MAY_CREATE (NFSD_MAY_EXEC|NFSD_MAY_WRITE) #define NFSD_MAY_REMOVE (NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC) @@ -54,7 +58,7 @@ __be32 nfsd_create(struct svc_rqst *, struct svc_fh *, int type, dev_t rdev, struct svc_fh *res); #ifdef CONFIG_NFSD_V3 __be32 nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *); -__be32 nfsd_create_v3(struct svc_rqst *, struct svc_fh *, +__be32 do_nfsd_create(struct svc_rqst *, struct svc_fh *, char *name, int len, struct iattr *attrs, struct svc_fh *res, int createmode, u32 *verifier, int *truncp, int *created); diff --git a/fs/nilfs2/alloc.c b/fs/nilfs2/alloc.c index f768448..eed4d7b 100644 --- a/fs/nilfs2/alloc.c +++ b/fs/nilfs2/alloc.c @@ -489,8 +489,8 @@ int nilfs_palloc_prepare_alloc_entry(struct inode *inode, void nilfs_palloc_commit_alloc_entry(struct inode *inode, struct nilfs_palloc_req *req) { - nilfs_mdt_mark_buffer_dirty(req->pr_bitmap_bh); - nilfs_mdt_mark_buffer_dirty(req->pr_desc_bh); + mark_buffer_dirty(req->pr_bitmap_bh); + mark_buffer_dirty(req->pr_desc_bh); nilfs_mdt_mark_dirty(inode); brelse(req->pr_bitmap_bh); @@ -527,8 +527,8 @@ void nilfs_palloc_commit_free_entry(struct inode *inode, kunmap(req->pr_bitmap_bh->b_page); kunmap(req->pr_desc_bh->b_page); - nilfs_mdt_mark_buffer_dirty(req->pr_desc_bh); - nilfs_mdt_mark_buffer_dirty(req->pr_bitmap_bh); + mark_buffer_dirty(req->pr_desc_bh); + mark_buffer_dirty(req->pr_bitmap_bh); nilfs_mdt_mark_dirty(inode); brelse(req->pr_bitmap_bh); @@ -683,8 +683,8 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems) kunmap(bitmap_bh->b_page); kunmap(desc_bh->b_page); - nilfs_mdt_mark_buffer_dirty(desc_bh); - nilfs_mdt_mark_buffer_dirty(bitmap_bh); + mark_buffer_dirty(desc_bh); + mark_buffer_dirty(bitmap_bh); nilfs_mdt_mark_dirty(inode); brelse(bitmap_bh); diff --git a/fs/nilfs2/bmap.c b/fs/nilfs2/bmap.c index 4723f04..aadbd0b 100644 --- a/fs/nilfs2/bmap.c +++ b/fs/nilfs2/bmap.c @@ -34,7 +34,9 @@ struct inode *nilfs_bmap_get_dat(const struct nilfs_bmap *bmap) { - return NILFS_I_NILFS(bmap->b_inode)->ns_dat; + struct the_nilfs *nilfs = bmap->b_inode->i_sb->s_fs_info; + + return nilfs->ns_dat; } static int nilfs_bmap_convert_error(struct nilfs_bmap *bmap, diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c index 609cd22..a35ae35 100644 --- a/fs/nilfs2/btnode.c +++ b/fs/nilfs2/btnode.c @@ -34,12 +34,6 @@ #include "page.h" #include "btnode.h" -void nilfs_btnode_cache_init(struct address_space *btnc, - struct backing_dev_info *bdi) -{ - nilfs_mapping_init(btnc, bdi); -} - void nilfs_btnode_cache_clear(struct address_space *btnc) { invalidate_mapping_pages(btnc, 0, -1); @@ -62,7 +56,7 @@ nilfs_btnode_create_block(struct address_space *btnc, __u64 blocknr) BUG(); } memset(bh->b_data, 0, 1 << inode->i_blkbits); - bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev; + bh->b_bdev = inode->i_sb->s_bdev; bh->b_blocknr = blocknr; set_buffer_mapped(bh); set_buffer_uptodate(bh); @@ -94,10 +88,11 @@ int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr, if (pblocknr == 0) { pblocknr = blocknr; if (inode->i_ino != NILFS_DAT_INO) { - struct inode *dat = NILFS_I_NILFS(inode)->ns_dat; + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; /* blocknr is a virtual block number */ - err = nilfs_dat_translate(dat, blocknr, &pblocknr); + err = nilfs_dat_translate(nilfs->ns_dat, blocknr, + &pblocknr); if (unlikely(err)) { brelse(bh); goto out_locked; @@ -120,7 +115,7 @@ int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr, goto found; } set_buffer_mapped(bh); - bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev; + bh->b_bdev = inode->i_sb->s_bdev; bh->b_blocknr = pblocknr; /* set block address for read */ bh->b_end_io = end_buffer_read_sync; get_bh(bh); @@ -259,7 +254,7 @@ void nilfs_btnode_commit_change_key(struct address_space *btnc, "invalid oldkey %lld (newkey=%lld)", (unsigned long long)oldkey, (unsigned long long)newkey); - nilfs_btnode_mark_dirty(obh); + mark_buffer_dirty(obh); spin_lock_irq(&btnc->tree_lock); radix_tree_delete(&btnc->page_tree, oldkey); @@ -271,7 +266,7 @@ void nilfs_btnode_commit_change_key(struct address_space *btnc, unlock_page(opage); } else { nilfs_copy_buffer(nbh, obh); - nilfs_btnode_mark_dirty(nbh); + mark_buffer_dirty(nbh); nbh->b_blocknr = newkey; ctxt->bh = nbh; diff --git a/fs/nilfs2/btnode.h b/fs/nilfs2/btnode.h index 1b8ebd8..3a4dd2d 100644 --- a/fs/nilfs2/btnode.h +++ b/fs/nilfs2/btnode.h @@ -37,7 +37,6 @@ struct nilfs_btnode_chkey_ctxt { struct buffer_head *newbh; }; -void nilfs_btnode_cache_init(struct address_space *, struct backing_dev_info *); void nilfs_btnode_cache_clear(struct address_space *); struct buffer_head *nilfs_btnode_create_block(struct address_space *btnc, __u64 blocknr); @@ -51,7 +50,4 @@ void nilfs_btnode_commit_change_key(struct address_space *, void nilfs_btnode_abort_change_key(struct address_space *, struct nilfs_btnode_chkey_ctxt *); -#define nilfs_btnode_mark_dirty(bh) nilfs_mark_buffer_dirty(bh) - - #endif /* _NILFS_BTNODE_H */ diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c index d451ae0..b2e3ff3 100644 --- a/fs/nilfs2/btree.c +++ b/fs/nilfs2/btree.c @@ -714,7 +714,7 @@ static void nilfs_btree_promote_key(struct nilfs_bmap *btree, nilfs_btree_get_nonroot_node(path, level), path[level].bp_index, key); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); } while ((path[level].bp_index == 0) && (++level < nilfs_btree_height(btree) - 1)); } @@ -739,7 +739,7 @@ static void nilfs_btree_do_insert(struct nilfs_bmap *btree, nilfs_btree_node_insert(node, path[level].bp_index, *keyp, *ptrp, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (path[level].bp_index == 0) nilfs_btree_promote_key(btree, path, level + 1, @@ -777,9 +777,9 @@ static void nilfs_btree_carry_left(struct nilfs_bmap *btree, nilfs_btree_node_move_left(left, node, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); nilfs_btree_promote_key(btree, path, level + 1, nilfs_btree_node_get_key(node, 0)); @@ -823,9 +823,9 @@ static void nilfs_btree_carry_right(struct nilfs_bmap *btree, nilfs_btree_node_move_right(node, right, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); path[level + 1].bp_index++; nilfs_btree_promote_key(btree, path, level + 1, @@ -870,9 +870,9 @@ static void nilfs_btree_split(struct nilfs_bmap *btree, nilfs_btree_node_move_right(node, right, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); newkey = nilfs_btree_node_get_key(right, 0); newptr = path[level].bp_newreq.bpr_ptr; @@ -919,7 +919,7 @@ static void nilfs_btree_grow(struct nilfs_bmap *btree, nilfs_btree_node_set_level(root, level + 1); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); path[level].bp_bh = path[level].bp_sib_bh; path[level].bp_sib_bh = NULL; @@ -1194,7 +1194,7 @@ static void nilfs_btree_do_delete(struct nilfs_bmap *btree, nilfs_btree_node_delete(node, path[level].bp_index, keyp, ptrp, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (path[level].bp_index == 0) nilfs_btree_promote_key(btree, path, level + 1, nilfs_btree_node_get_key(node, 0)); @@ -1226,9 +1226,9 @@ static void nilfs_btree_borrow_left(struct nilfs_bmap *btree, nilfs_btree_node_move_right(left, node, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); nilfs_btree_promote_key(btree, path, level + 1, nilfs_btree_node_get_key(node, 0)); @@ -1258,9 +1258,9 @@ static void nilfs_btree_borrow_right(struct nilfs_bmap *btree, nilfs_btree_node_move_left(node, right, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); path[level + 1].bp_index++; nilfs_btree_promote_key(btree, path, level + 1, @@ -1289,7 +1289,7 @@ static void nilfs_btree_concat_left(struct nilfs_bmap *btree, nilfs_btree_node_move_left(left, node, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_sib_bh)) - nilfs_btnode_mark_dirty(path[level].bp_sib_bh); + mark_buffer_dirty(path[level].bp_sib_bh); nilfs_btnode_delete(path[level].bp_bh); path[level].bp_bh = path[level].bp_sib_bh; @@ -1315,7 +1315,7 @@ static void nilfs_btree_concat_right(struct nilfs_bmap *btree, nilfs_btree_node_move_left(node, right, n, ncblk, ncblk); if (!buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); nilfs_btnode_delete(path[level].bp_sib_bh); path[level].bp_sib_bh = NULL; @@ -1346,6 +1346,11 @@ static void nilfs_btree_shrink(struct nilfs_bmap *btree, path[level].bp_bh = NULL; } +static void nilfs_btree_nop(struct nilfs_bmap *btree, + struct nilfs_btree_path *path, + int level, __u64 *keyp, __u64 *ptrp) +{ +} static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree, struct nilfs_btree_path *path, @@ -1356,20 +1361,19 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree, struct buffer_head *bh; struct nilfs_btree_node *node, *parent, *sib; __u64 sibptr; - int pindex, level, ncmin, ncmax, ncblk, ret; + int pindex, dindex, level, ncmin, ncmax, ncblk, ret; ret = 0; stats->bs_nblocks = 0; ncmin = NILFS_BTREE_NODE_NCHILDREN_MIN(nilfs_btree_node_size(btree)); ncblk = nilfs_btree_nchildren_per_block(btree); - for (level = NILFS_BTREE_LEVEL_NODE_MIN; + for (level = NILFS_BTREE_LEVEL_NODE_MIN, dindex = path[level].bp_index; level < nilfs_btree_height(btree) - 1; level++) { node = nilfs_btree_get_nonroot_node(path, level); path[level].bp_oldreq.bpr_ptr = - nilfs_btree_node_get_ptr(node, path[level].bp_index, - ncblk); + nilfs_btree_node_get_ptr(node, dindex, ncblk); ret = nilfs_bmap_prepare_end_ptr(btree, &path[level].bp_oldreq, dat); if (ret < 0) @@ -1383,6 +1387,7 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree, parent = nilfs_btree_get_node(btree, path, level + 1, &ncmax); pindex = path[level + 1].bp_index; + dindex = pindex; if (pindex > 0) { /* left sibling */ @@ -1421,6 +1426,14 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree, path[level].bp_sib_bh = bh; path[level].bp_op = nilfs_btree_concat_right; stats->bs_nblocks++; + /* + * When merging right sibling node + * into the current node, pointer to + * the right sibling node must be + * terminated instead. The adjustment + * below is required for that. + */ + dindex = pindex + 1; /* continue; */ } } else { @@ -1431,29 +1444,31 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree, NILFS_BTREE_ROOT_NCHILDREN_MAX) { path[level].bp_op = nilfs_btree_shrink; stats->bs_nblocks += 2; + level++; + path[level].bp_op = nilfs_btree_nop; + goto shrink_root_child; } else { path[level].bp_op = nilfs_btree_do_delete; stats->bs_nblocks++; + goto out; } - - goto out; - } } + /* child of the root node is deleted */ + path[level].bp_op = nilfs_btree_do_delete; + stats->bs_nblocks++; + +shrink_root_child: node = nilfs_btree_get_root(btree); path[level].bp_oldreq.bpr_ptr = - nilfs_btree_node_get_ptr(node, path[level].bp_index, + nilfs_btree_node_get_ptr(node, dindex, NILFS_BTREE_ROOT_NCHILDREN_MAX); ret = nilfs_bmap_prepare_end_ptr(btree, &path[level].bp_oldreq, dat); if (ret < 0) goto err_out_child_node; - /* child of the root node is deleted */ - path[level].bp_op = nilfs_btree_do_delete; - stats->bs_nblocks++; - /* success */ out: *levelp = level; @@ -1709,7 +1724,7 @@ nilfs_btree_commit_convert_and_insert(struct nilfs_bmap *btree, nilfs_btree_node_init(node, 0, 1, n, ncblk, keys, ptrs); nilfs_btree_node_insert(node, n, key, dreq->bpr_ptr, ncblk); if (!buffer_dirty(bh)) - nilfs_btnode_mark_dirty(bh); + mark_buffer_dirty(bh); if (!nilfs_bmap_dirty(btree)) nilfs_bmap_set_dirty(btree); @@ -1787,7 +1802,7 @@ static int nilfs_btree_propagate_p(struct nilfs_bmap *btree, { while ((++level < nilfs_btree_height(btree) - 1) && !buffer_dirty(path[level].bp_bh)) - nilfs_btnode_mark_dirty(path[level].bp_bh); + mark_buffer_dirty(path[level].bp_bh); return 0; } @@ -2229,7 +2244,7 @@ static int nilfs_btree_mark(struct nilfs_bmap *btree, __u64 key, int level) } if (!buffer_dirty(bh)) - nilfs_btnode_mark_dirty(bh); + mark_buffer_dirty(bh); brelse(bh); if (!nilfs_bmap_dirty(btree)) nilfs_bmap_set_dirty(btree); diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c index 5ff15a8..c9b342c 100644 --- a/fs/nilfs2/cpfile.c +++ b/fs/nilfs2/cpfile.c @@ -216,14 +216,14 @@ int nilfs_cpfile_get_checkpoint(struct inode *cpfile, if (!nilfs_cpfile_is_in_first(cpfile, cno)) nilfs_cpfile_block_add_valid_checkpoints(cpfile, cp_bh, kaddr, 1); - nilfs_mdt_mark_buffer_dirty(cp_bh); + mark_buffer_dirty(cp_bh); kaddr = kmap_atomic(header_bh->b_page, KM_USER0); header = nilfs_cpfile_block_get_header(cpfile, header_bh, kaddr); le64_add_cpu(&header->ch_ncheckpoints, 1); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(header_bh); + mark_buffer_dirty(header_bh); nilfs_mdt_mark_dirty(cpfile); } @@ -326,7 +326,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, } if (nicps > 0) { tnicps += nicps; - nilfs_mdt_mark_buffer_dirty(cp_bh); + mark_buffer_dirty(cp_bh); nilfs_mdt_mark_dirty(cpfile); if (!nilfs_cpfile_is_in_first(cpfile, cno)) { count = @@ -358,7 +358,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, header = nilfs_cpfile_block_get_header(cpfile, header_bh, kaddr); le64_add_cpu(&header->ch_ncheckpoints, -(u64)tnicps); - nilfs_mdt_mark_buffer_dirty(header_bh); + mark_buffer_dirty(header_bh); nilfs_mdt_mark_dirty(cpfile); kunmap_atomic(kaddr, KM_USER0); } @@ -671,10 +671,10 @@ static int nilfs_cpfile_set_snapshot(struct inode *cpfile, __u64 cno) le64_add_cpu(&header->ch_nsnapshots, 1); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(prev_bh); - nilfs_mdt_mark_buffer_dirty(curr_bh); - nilfs_mdt_mark_buffer_dirty(cp_bh); - nilfs_mdt_mark_buffer_dirty(header_bh); + mark_buffer_dirty(prev_bh); + mark_buffer_dirty(curr_bh); + mark_buffer_dirty(cp_bh); + mark_buffer_dirty(header_bh); nilfs_mdt_mark_dirty(cpfile); brelse(prev_bh); @@ -774,10 +774,10 @@ static int nilfs_cpfile_clear_snapshot(struct inode *cpfile, __u64 cno) le64_add_cpu(&header->ch_nsnapshots, -1); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(next_bh); - nilfs_mdt_mark_buffer_dirty(prev_bh); - nilfs_mdt_mark_buffer_dirty(cp_bh); - nilfs_mdt_mark_buffer_dirty(header_bh); + mark_buffer_dirty(next_bh); + mark_buffer_dirty(prev_bh); + mark_buffer_dirty(cp_bh); + mark_buffer_dirty(header_bh); nilfs_mdt_mark_dirty(cpfile); brelse(prev_bh); diff --git a/fs/nilfs2/dat.c b/fs/nilfs2/dat.c index 59e5fe7..fcc2f86 100644 --- a/fs/nilfs2/dat.c +++ b/fs/nilfs2/dat.c @@ -54,7 +54,7 @@ static int nilfs_dat_prepare_entry(struct inode *dat, static void nilfs_dat_commit_entry(struct inode *dat, struct nilfs_palloc_req *req) { - nilfs_mdt_mark_buffer_dirty(req->pr_entry_bh); + mark_buffer_dirty(req->pr_entry_bh); nilfs_mdt_mark_dirty(dat); brelse(req->pr_entry_bh); } @@ -361,7 +361,7 @@ int nilfs_dat_move(struct inode *dat, __u64 vblocknr, sector_t blocknr) entry->de_blocknr = cpu_to_le64(blocknr); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(entry_bh); + mark_buffer_dirty(entry_bh); nilfs_mdt_mark_dirty(dat); brelse(entry_bh); diff --git a/fs/nilfs2/file.c b/fs/nilfs2/file.c index 397e732..d7eeca6 100644 --- a/fs/nilfs2/file.c +++ b/fs/nilfs2/file.c @@ -111,7 +111,6 @@ static int nilfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) nilfs_transaction_commit(inode->i_sb); mapped: - SetPageChecked(page); wait_on_page_writeback(page); return VM_FAULT_LOCKED; } diff --git a/fs/nilfs2/gcinode.c b/fs/nilfs2/gcinode.c index 1c2a3e2..08a07a2 100644 --- a/fs/nilfs2/gcinode.c +++ b/fs/nilfs2/gcinode.c @@ -48,9 +48,6 @@ #include "dat.h" #include "ifile.h" -static const struct address_space_operations def_gcinode_aops = { -}; - /* * nilfs_gccache_submit_read_data() - add data buffer and submit read request * @inode - gc inode @@ -87,9 +84,9 @@ int nilfs_gccache_submit_read_data(struct inode *inode, sector_t blkoff, goto out; if (pbn == 0) { - struct inode *dat_inode = NILFS_I_NILFS(inode)->ns_dat; - /* use original dat, not gc dat. */ - err = nilfs_dat_translate(dat_inode, vbn, &pbn); + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; + + err = nilfs_dat_translate(nilfs->ns_dat, vbn, &pbn); if (unlikely(err)) { /* -EIO, -ENOMEM, -ENOENT */ brelse(bh); goto failed; @@ -103,7 +100,7 @@ int nilfs_gccache_submit_read_data(struct inode *inode, sector_t blkoff, } if (!buffer_mapped(bh)) { - bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev; + bh->b_bdev = inode->i_sb->s_bdev; set_buffer_mapped(bh); } bh->b_blocknr = pbn; @@ -160,15 +157,11 @@ int nilfs_gccache_wait_and_mark_dirty(struct buffer_head *bh) if (buffer_dirty(bh)) return -EEXIST; - if (buffer_nilfs_node(bh)) { - if (nilfs_btree_broken_node_block(bh)) { - clear_buffer_uptodate(bh); - return -EIO; - } - nilfs_btnode_mark_dirty(bh); - } else { - nilfs_mark_buffer_dirty(bh); + if (buffer_nilfs_node(bh) && nilfs_btree_broken_node_block(bh)) { + clear_buffer_uptodate(bh); + return -EIO; } + mark_buffer_dirty(bh); return 0; } @@ -178,7 +171,7 @@ int nilfs_init_gcinode(struct inode *inode) inode->i_mode = S_IFREG; mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS); - inode->i_mapping->a_ops = &def_gcinode_aops; + inode->i_mapping->a_ops = &empty_aops; inode->i_mapping->backing_dev_info = inode->i_sb->s_bdi; ii->i_flags = 0; diff --git a/fs/nilfs2/ifile.c b/fs/nilfs2/ifile.c index bfc73d3..684d763 100644 --- a/fs/nilfs2/ifile.c +++ b/fs/nilfs2/ifile.c @@ -80,7 +80,7 @@ int nilfs_ifile_create_inode(struct inode *ifile, ino_t *out_ino, return ret; } nilfs_palloc_commit_alloc_entry(ifile, &req); - nilfs_mdt_mark_buffer_dirty(req.pr_entry_bh); + mark_buffer_dirty(req.pr_entry_bh); nilfs_mdt_mark_dirty(ifile); *out_ino = (ino_t)req.pr_entry_nr; *out_bh = req.pr_entry_bh; @@ -128,7 +128,7 @@ int nilfs_ifile_delete_inode(struct inode *ifile, ino_t ino) raw_inode->i_flags = 0; kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(req.pr_entry_bh); + mark_buffer_dirty(req.pr_entry_bh); brelse(req.pr_entry_bh); nilfs_palloc_commit_free_entry(ifile, &req); diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index c0aa274..b9b45fc 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -74,14 +74,14 @@ int nilfs_get_block(struct inode *inode, sector_t blkoff, struct buffer_head *bh_result, int create) { struct nilfs_inode_info *ii = NILFS_I(inode); + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; __u64 blknum = 0; int err = 0, ret; - struct inode *dat = NILFS_I_NILFS(inode)->ns_dat; unsigned maxblocks = bh_result->b_size >> inode->i_blkbits; - down_read(&NILFS_MDT(dat)->mi_sem); + down_read(&NILFS_MDT(nilfs->ns_dat)->mi_sem); ret = nilfs_bmap_lookup_contig(ii->i_bmap, blkoff, &blknum, maxblocks); - up_read(&NILFS_MDT(dat)->mi_sem); + up_read(&NILFS_MDT(nilfs->ns_dat)->mi_sem); if (ret >= 0) { /* found */ map_bh(bh_result, inode->i_sb, blknum); if (ret > 0) @@ -596,6 +596,16 @@ void nilfs_write_inode_common(struct inode *inode, raw_inode->i_flags = cpu_to_le32(ii->i_flags); raw_inode->i_generation = cpu_to_le32(inode->i_generation); + if (NILFS_ROOT_METADATA_FILE(inode->i_ino)) { + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; + + /* zero-fill unused portion in the case of super root block */ + raw_inode->i_xattr = 0; + raw_inode->i_pad = 0; + memset((void *)raw_inode + sizeof(*raw_inode), 0, + nilfs->ns_inode_size - sizeof(*raw_inode)); + } + if (has_bmap) nilfs_bmap_write(ii->i_bmap, raw_inode); else if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) @@ -791,12 +801,7 @@ out_err: int nilfs_permission(struct inode *inode, int mask, unsigned int flags) { - struct nilfs_root *root; - - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - - root = NILFS_I(inode)->i_root; + struct nilfs_root *root = NILFS_I(inode)->i_root; if ((mask & MAY_WRITE) && root && root->cno != NILFS_CPTREE_CURRENT_CNO) return -EROFS; /* snapshot is not writable */ @@ -872,8 +877,7 @@ int nilfs_set_file_dirty(struct inode *inode, unsigned nr_dirty) return -EINVAL; /* NILFS_I_DIRTY may remain for freeing inode */ } - list_del(&ii->i_dirty); - list_add_tail(&ii->i_dirty, &nilfs->ns_dirty_files); + list_move_tail(&ii->i_dirty, &nilfs->ns_dirty_files); set_bit(NILFS_I_QUEUED, &ii->i_state); } spin_unlock(&nilfs->ns_inode_lock); @@ -892,7 +896,7 @@ int nilfs_mark_inode_dirty(struct inode *inode) return err; } nilfs_update_inode(inode, ibh); - nilfs_mdt_mark_buffer_dirty(ibh); + mark_buffer_dirty(ibh); nilfs_mdt_mark_dirty(NILFS_I(inode)->i_root->ifile); brelse(ibh); return 0; @@ -908,7 +912,7 @@ int nilfs_mark_inode_dirty(struct inode *inode) * construction. This function can be called both as a single operation * and as a part of indivisible file operations. */ -void nilfs_dirty_inode(struct inode *inode) +void nilfs_dirty_inode(struct inode *inode, int flags) { struct nilfs_transaction_info ti; struct nilfs_mdt_info *mdi = NILFS_MDT(inode); @@ -931,7 +935,7 @@ void nilfs_dirty_inode(struct inode *inode) int nilfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len) { - struct the_nilfs *nilfs = NILFS_I_NILFS(inode); + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; __u64 logical = 0, phys = 0, size = 0; __u32 flags = 0; loff_t isize; diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index f2469ba..41d6743 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -698,6 +698,63 @@ static int nilfs_ioctl_sync(struct inode *inode, struct file *filp, return 0; } +static int nilfs_ioctl_resize(struct inode *inode, struct file *filp, + void __user *argp) +{ + __u64 newsize; + int ret = -EPERM; + + if (!capable(CAP_SYS_ADMIN)) + goto out; + + ret = mnt_want_write(filp->f_path.mnt); + if (ret) + goto out; + + ret = -EFAULT; + if (copy_from_user(&newsize, argp, sizeof(newsize))) + goto out_drop_write; + + ret = nilfs_resize_fs(inode->i_sb, newsize); + +out_drop_write: + mnt_drop_write(filp->f_path.mnt); +out: + return ret; +} + +static int nilfs_ioctl_set_alloc_range(struct inode *inode, void __user *argp) +{ + struct the_nilfs *nilfs = inode->i_sb->s_fs_info; + __u64 range[2]; + __u64 minseg, maxseg; + unsigned long segbytes; + int ret = -EPERM; + + if (!capable(CAP_SYS_ADMIN)) + goto out; + + ret = -EFAULT; + if (copy_from_user(range, argp, sizeof(__u64[2]))) + goto out; + + ret = -ERANGE; + if (range[1] > i_size_read(inode->i_sb->s_bdev->bd_inode)) + goto out; + + segbytes = nilfs->ns_blocks_per_segment * nilfs->ns_blocksize; + + minseg = range[0] + segbytes - 1; + do_div(minseg, segbytes); + maxseg = NILFS_SB2_OFFSET_BYTES(range[1]); + do_div(maxseg, segbytes); + maxseg--; + + ret = nilfs_sufile_set_alloc_range(nilfs->ns_sufile, minseg, maxseg); +out: + return ret; +} + static int nilfs_ioctl_get_info(struct inode *inode, struct file *filp, unsigned int cmd, void __user *argp, size_t membsz, @@ -763,6 +820,10 @@ long nilfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return nilfs_ioctl_clean_segments(inode, filp, cmd, argp); case NILFS_IOCTL_SYNC: return nilfs_ioctl_sync(inode, filp, cmd, argp); + case NILFS_IOCTL_RESIZE: + return nilfs_ioctl_resize(inode, filp, argp); + case NILFS_IOCTL_SET_ALLOC_RANGE: + return nilfs_ioctl_set_alloc_range(inode, argp); default: return -ENOTTY; } diff --git a/fs/nilfs2/mdt.c b/fs/nilfs2/mdt.c index a649b05..800e8d7 100644 --- a/fs/nilfs2/mdt.c +++ b/fs/nilfs2/mdt.c @@ -66,7 +66,7 @@ nilfs_mdt_insert_new_block(struct inode *inode, unsigned long block, kunmap_atomic(kaddr, KM_USER0); set_buffer_uptodate(bh); - nilfs_mark_buffer_dirty(bh); + mark_buffer_dirty(bh); nilfs_mdt_mark_dirty(inode); return 0; } @@ -355,7 +355,7 @@ int nilfs_mdt_mark_block_dirty(struct inode *inode, unsigned long block) err = nilfs_mdt_read_block(inode, block, 0, &bh); if (unlikely(err)) return err; - nilfs_mark_buffer_dirty(bh); + mark_buffer_dirty(bh); nilfs_mdt_mark_dirty(inode); brelse(bh); return 0; @@ -450,9 +450,9 @@ int nilfs_mdt_setup_shadow_map(struct inode *inode, INIT_LIST_HEAD(&shadow->frozen_buffers); address_space_init_once(&shadow->frozen_data); - nilfs_mapping_init(&shadow->frozen_data, bdi); + nilfs_mapping_init(&shadow->frozen_data, inode, bdi); address_space_init_once(&shadow->frozen_btnodes); - nilfs_mapping_init(&shadow->frozen_btnodes, bdi); + nilfs_mapping_init(&shadow->frozen_btnodes, inode, bdi); mi->mi_shadow = shadow; return 0; } diff --git a/fs/nilfs2/mdt.h b/fs/nilfs2/mdt.h index ed68563..ab20a4b 100644 --- a/fs/nilfs2/mdt.h +++ b/fs/nilfs2/mdt.h @@ -64,11 +64,6 @@ static inline struct nilfs_mdt_info *NILFS_MDT(const struct inode *inode) return inode->i_private; } -static inline struct the_nilfs *NILFS_I_NILFS(struct inode *inode) -{ - return inode->i_sb->s_fs_info; -} - /* Default GFP flags using highmem */ #define NILFS_MDT_GFP (__GFP_WAIT | __GFP_IO | __GFP_HIGHMEM) @@ -93,8 +88,6 @@ int nilfs_mdt_freeze_buffer(struct inode *inode, struct buffer_head *bh); struct buffer_head *nilfs_mdt_get_frozen_buffer(struct inode *inode, struct buffer_head *bh); -#define nilfs_mdt_mark_buffer_dirty(bh) nilfs_mark_buffer_dirty(bh) - static inline void nilfs_mdt_mark_dirty(struct inode *inode) { if (!test_bit(NILFS_I_DIRTY, &NILFS_I(inode)->i_state)) @@ -108,7 +101,7 @@ static inline void nilfs_mdt_clear_dirty(struct inode *inode) static inline __u64 nilfs_mdt_cno(struct inode *inode) { - return NILFS_I_NILFS(inode)->ns_cno; + return ((struct the_nilfs *)inode->i_sb->s_fs_info)->ns_cno; } #define nilfs_mdt_bgl_lock(inode, bg) \ diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index a8dd344..f02b9ad 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -80,12 +80,6 @@ static inline struct inode *NILFS_BTNC_I(struct address_space *btnc) return &ii->vfs_inode; } -static inline struct inode *NILFS_AS_I(struct address_space *mapping) -{ - return (mapping->host) ? : - container_of(mapping, struct inode, i_data); -} - /* * Dynamic state flags of NILFS on-memory inode (i_state) */ @@ -275,7 +269,7 @@ int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh); extern int nilfs_inode_dirty(struct inode *); int nilfs_set_file_dirty(struct inode *inode, unsigned nr_dirty); extern int nilfs_mark_inode_dirty(struct inode *); -extern void nilfs_dirty_inode(struct inode *); +extern void nilfs_dirty_inode(struct inode *, int flags); int nilfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len); @@ -298,6 +292,7 @@ struct nilfs_super_block **nilfs_prepare_super(struct super_block *sb, int flip); int nilfs_commit_super(struct super_block *sb, int flag); int nilfs_cleanup_super(struct super_block *sb); +int nilfs_resize_fs(struct super_block *sb, __u64 newsize); int nilfs_attach_checkpoint(struct super_block *sb, __u64 cno, int curr_mnt, struct nilfs_root **root); int nilfs_checkpoint_is_mounted(struct super_block *sb, __u64 cno); diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c index 1168059..65221a0 100644 --- a/fs/nilfs2/page.c +++ b/fs/nilfs2/page.c @@ -37,8 +37,7 @@ #define NILFS_BUFFER_INHERENT_BITS \ ((1UL << BH_Uptodate) | (1UL << BH_Mapped) | (1UL << BH_NILFS_Node) | \ - (1UL << BH_NILFS_Volatile) | (1UL << BH_NILFS_Allocated) | \ - (1UL << BH_NILFS_Checked)) + (1UL << BH_NILFS_Volatile) | (1UL << BH_NILFS_Checked)) static struct buffer_head * __nilfs_get_page_block(struct page *page, unsigned long block, pgoff_t index, @@ -59,19 +58,6 @@ __nilfs_get_page_block(struct page *page, unsigned long block, pgoff_t index, return bh; } -/* - * Since the page cache of B-tree node pages or data page cache of pseudo - * inodes does not have a valid mapping->host pointer, calling - * mark_buffer_dirty() for their buffers causes a NULL pointer dereference; - * it calls __mark_inode_dirty(NULL) through __set_page_dirty(). - * To avoid this problem, the old style mark_buffer_dirty() is used instead. - */ -void nilfs_mark_buffer_dirty(struct buffer_head *bh) -{ - if (!buffer_dirty(bh) && !test_set_buffer_dirty(bh)) - __set_page_dirty_nobuffers(bh->b_page); -} - struct buffer_head *nilfs_grab_buffer(struct inode *inode, struct address_space *mapping, unsigned long blkoff, @@ -183,7 +169,7 @@ int nilfs_page_buffers_clean(struct page *page) void nilfs_page_bug(struct page *page) { struct address_space *m; - unsigned long ino = 0; + unsigned long ino; if (unlikely(!page)) { printk(KERN_CRIT "NILFS_PAGE_BUG(NULL)\n"); @@ -191,11 +177,8 @@ void nilfs_page_bug(struct page *page) } m = page->mapping; - if (m) { - struct inode *inode = NILFS_AS_I(m); - if (inode != NULL) - ino = inode->i_ino; - } + ino = m ? m->host->i_ino : 0; + printk(KERN_CRIT "NILFS_PAGE_BUG(%p): cnt=%d index#=%llu flags=0x%lx " "mapping=%p ino=%lu\n", page, atomic_read(&page->_count), @@ -217,56 +200,6 @@ void nilfs_page_bug(struct page *page) } /** - * nilfs_alloc_private_page - allocate a private page with buffer heads - * - * Return Value: On success, a pointer to the allocated page is returned. - * On error, NULL is returned. - */ -struct page *nilfs_alloc_private_page(struct block_device *bdev, int size, - unsigned long state) -{ - struct buffer_head *bh, *head, *tail; - struct page *page; - - page = alloc_page(GFP_NOFS); /* page_count of the returned page is 1 */ - if (unlikely(!page)) - return NULL; - - lock_page(page); - head = alloc_page_buffers(page, size, 0); - if (unlikely(!head)) { - unlock_page(page); - __free_page(page); - return NULL; - } - - bh = head; - do { - bh->b_state = (1UL << BH_NILFS_Allocated) | state; - tail = bh; - bh->b_bdev = bdev; - bh = bh->b_this_page; - } while (bh); - - tail->b_this_page = head; - attach_page_buffers(page, head); - - return page; -} - -void nilfs_free_private_page(struct page *page) -{ - BUG_ON(!PageLocked(page)); - BUG_ON(page->mapping); - - if (page_has_buffers(page) && !try_to_free_buffers(page)) - NILFS_PAGE_BUG(page, "failed to free page"); - - unlock_page(page); - __free_page(page); -} - -/** * nilfs_copy_page -- copy the page with buffers * @dst: destination page * @src: source page @@ -492,10 +425,10 @@ unsigned nilfs_page_count_clean_buffers(struct page *page, return nc; } -void nilfs_mapping_init(struct address_space *mapping, +void nilfs_mapping_init(struct address_space *mapping, struct inode *inode, struct backing_dev_info *bdi) { - mapping->host = NULL; + mapping->host = inode; mapping->flags = 0; mapping_set_gfp_mask(mapping, GFP_NOFS); mapping->assoc_mapping = NULL; diff --git a/fs/nilfs2/page.h b/fs/nilfs2/page.h index f06b79a..fb7de71 100644 --- a/fs/nilfs2/page.h +++ b/fs/nilfs2/page.h @@ -38,14 +38,12 @@ enum { BH_NILFS_Redirected, }; -BUFFER_FNS(NILFS_Allocated, nilfs_allocated) /* nilfs private buffers */ BUFFER_FNS(NILFS_Node, nilfs_node) /* nilfs node buffers */ BUFFER_FNS(NILFS_Volatile, nilfs_volatile) BUFFER_FNS(NILFS_Checked, nilfs_checked) /* buffer is verified */ BUFFER_FNS(NILFS_Redirected, nilfs_redirected) /* redirected to a copy */ -void nilfs_mark_buffer_dirty(struct buffer_head *bh); int __nilfs_clear_page_dirty(struct page *); struct buffer_head *nilfs_grab_buffer(struct inode *, struct address_space *, @@ -54,14 +52,11 @@ void nilfs_forget_buffer(struct buffer_head *); void nilfs_copy_buffer(struct buffer_head *, struct buffer_head *); int nilfs_page_buffers_clean(struct page *); void nilfs_page_bug(struct page *); -struct page *nilfs_alloc_private_page(struct block_device *, int, - unsigned long); -void nilfs_free_private_page(struct page *); int nilfs_copy_dirty_pages(struct address_space *, struct address_space *); void nilfs_copy_back_pages(struct address_space *, struct address_space *); void nilfs_clear_dirty_pages(struct address_space *); -void nilfs_mapping_init(struct address_space *mapping, +void nilfs_mapping_init(struct address_space *mapping, struct inode *inode, struct backing_dev_info *bdi); unsigned nilfs_page_count_clean_buffers(struct page *, unsigned, unsigned); unsigned long nilfs_find_uncommitted_extent(struct inode *inode, diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c index ba4a645..a604ac0 100644 --- a/fs/nilfs2/recovery.c +++ b/fs/nilfs2/recovery.c @@ -387,9 +387,9 @@ static int nilfs_scan_dsync_log(struct the_nilfs *nilfs, sector_t start_blocknr, static void dispose_recovery_list(struct list_head *head) { while (!list_empty(head)) { - struct nilfs_recovery_block *rb - = list_entry(head->next, - struct nilfs_recovery_block, list); + struct nilfs_recovery_block *rb; + + rb = list_first_entry(head, struct nilfs_recovery_block, list); list_del(&rb->list); kfree(rb); } @@ -416,9 +416,9 @@ static int nilfs_segment_list_add(struct list_head *head, __u64 segnum) void nilfs_dispose_segment_list(struct list_head *head) { while (!list_empty(head)) { - struct nilfs_segment_entry *ent - = list_entry(head->next, - struct nilfs_segment_entry, list); + struct nilfs_segment_entry *ent; + + ent = list_first_entry(head, struct nilfs_segment_entry, list); list_del(&ent->list); kfree(ent); } diff --git a/fs/nilfs2/segbuf.c b/fs/nilfs2/segbuf.c index 2853ff2..850a7c0 100644 --- a/fs/nilfs2/segbuf.c +++ b/fs/nilfs2/segbuf.c @@ -239,12 +239,15 @@ nilfs_segbuf_fill_in_super_root_crc(struct nilfs_segment_buffer *segbuf, u32 seed) { struct nilfs_super_root *raw_sr; + struct the_nilfs *nilfs = segbuf->sb_super->s_fs_info; + unsigned srsize; u32 crc; raw_sr = (struct nilfs_super_root *)segbuf->sb_super_root->b_data; + srsize = NILFS_SR_BYTES(nilfs->ns_inode_size); crc = crc32_le(seed, (unsigned char *)raw_sr + sizeof(raw_sr->sr_sum), - NILFS_SR_BYTES - sizeof(raw_sr->sr_sum)); + srsize - sizeof(raw_sr->sr_sum)); raw_sr->sr_sum = cpu_to_le32(crc); } @@ -254,18 +257,6 @@ static void nilfs_release_buffers(struct list_head *list) list_for_each_entry_safe(bh, n, list, b_assoc_buffers) { list_del_init(&bh->b_assoc_buffers); - if (buffer_nilfs_allocated(bh)) { - struct page *clone_page = bh->b_page; - - /* remove clone page */ - brelse(bh); - page_cache_release(clone_page); /* for each bh */ - if (page_count(clone_page) <= 2) { - lock_page(clone_page); - nilfs_free_private_page(clone_page); - } - continue; - } brelse(bh); } } diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c index afe4f21..bb24ab6 100644 --- a/fs/nilfs2/segment.c +++ b/fs/nilfs2/segment.c @@ -655,13 +655,10 @@ static size_t nilfs_lookup_dirty_data_buffers(struct inode *inode, if (unlikely(page->index > last)) break; - if (mapping->host) { - lock_page(page); - if (!page_has_buffers(page)) - create_empty_buffers(page, - 1 << inode->i_blkbits, 0); - unlock_page(page); - } + lock_page(page); + if (!page_has_buffers(page)) + create_empty_buffers(page, 1 << inode->i_blkbits, 0); + unlock_page(page); bh = head = page_buffers(page); do { @@ -809,7 +806,7 @@ static int nilfs_segctor_create_checkpoint(struct nilfs_sc_info *sci) /* The following code is duplicated with cpfile. But, it is needed to collect the checkpoint even if it was not newly created */ - nilfs_mdt_mark_buffer_dirty(bh_cp); + mark_buffer_dirty(bh_cp); nilfs_mdt_mark_dirty(nilfs->ns_cpfile); nilfs_cpfile_put_checkpoint( nilfs->ns_cpfile, nilfs->ns_cno, bh_cp); @@ -889,12 +886,14 @@ static void nilfs_segctor_fill_in_super_root(struct nilfs_sc_info *sci, { struct buffer_head *bh_sr; struct nilfs_super_root *raw_sr; - unsigned isz = nilfs->ns_inode_size; + unsigned isz, srsz; bh_sr = NILFS_LAST_SEGBUF(&sci->sc_segbufs)->sb_super_root; raw_sr = (struct nilfs_super_root *)bh_sr->b_data; + isz = nilfs->ns_inode_size; + srsz = NILFS_SR_BYTES(isz); - raw_sr->sr_bytes = cpu_to_le16(NILFS_SR_BYTES); + raw_sr->sr_bytes = cpu_to_le16(srsz); raw_sr->sr_nongc_ctime = cpu_to_le64(nilfs_doing_gc() ? nilfs->ns_nongc_ctime : sci->sc_seg_ctime); @@ -906,6 +905,7 @@ static void nilfs_segctor_fill_in_super_root(struct nilfs_sc_info *sci, NILFS_SR_CPFILE_OFFSET(isz), 1); nilfs_write_inode_common(nilfs->ns_sufile, (void *)raw_sr + NILFS_SR_SUFILE_OFFSET(isz), 1); + memset((void *)raw_sr + srsz, 0, nilfs->ns_blocksize - srsz); } static void nilfs_redirty_inodes(struct list_head *head) @@ -954,8 +954,8 @@ static int nilfs_segctor_apply_buffers(struct nilfs_sc_info *sci, dispose_buffers: while (!list_empty(listp)) { - bh = list_entry(listp->next, struct buffer_head, - b_assoc_buffers); + bh = list_first_entry(listp, struct buffer_head, + b_assoc_buffers); list_del_init(&bh->b_assoc_buffers); brelse(bh); } @@ -1500,10 +1500,7 @@ nilfs_segctor_update_payload_blocknr(struct nilfs_sc_info *sci, nblocks = le32_to_cpu(finfo->fi_nblocks); ndatablk = le32_to_cpu(finfo->fi_ndatablk); - if (buffer_nilfs_node(bh)) - inode = NILFS_BTNC_I(bh->b_page->mapping); - else - inode = NILFS_AS_I(bh->b_page->mapping); + inode = bh->b_page->mapping->host; if (mode == SC_LSEG_DSYNC) sc_op = &nilfs_sc_dsync_ops; @@ -1556,83 +1553,24 @@ static int nilfs_segctor_assign(struct nilfs_sc_info *sci, int mode) return 0; } -static int -nilfs_copy_replace_page_buffers(struct page *page, struct list_head *out) -{ - struct page *clone_page; - struct buffer_head *bh, *head, *bh2; - void *kaddr; - - bh = head = page_buffers(page); - - clone_page = nilfs_alloc_private_page(bh->b_bdev, bh->b_size, 0); - if (unlikely(!clone_page)) - return -ENOMEM; - - bh2 = page_buffers(clone_page); - kaddr = kmap_atomic(page, KM_USER0); - do { - if (list_empty(&bh->b_assoc_buffers)) - continue; - get_bh(bh2); - page_cache_get(clone_page); /* for each bh */ - memcpy(bh2->b_data, kaddr + bh_offset(bh), bh2->b_size); - bh2->b_blocknr = bh->b_blocknr; - list_replace(&bh->b_assoc_buffers, &bh2->b_assoc_buffers); - list_add_tail(&bh->b_assoc_buffers, out); - } while (bh = bh->b_this_page, bh2 = bh2->b_this_page, bh != head); - kunmap_atomic(kaddr, KM_USER0); - - if (!TestSetPageWriteback(clone_page)) - account_page_writeback(clone_page); - unlock_page(clone_page); - - return 0; -} - -static int nilfs_test_page_to_be_frozen(struct page *page) -{ - struct address_space *mapping = page->mapping; - - if (!mapping || !mapping->host || S_ISDIR(mapping->host->i_mode)) - return 0; - - if (page_mapped(page)) { - ClearPageChecked(page); - return 1; - } - return PageChecked(page); -} - -static int nilfs_begin_page_io(struct page *page, struct list_head *out) +static void nilfs_begin_page_io(struct page *page) { if (!page || PageWriteback(page)) /* For split b-tree node pages, this function may be called twice. We ignore the 2nd or later calls by this check. */ - return 0; + return; lock_page(page); clear_page_dirty_for_io(page); set_page_writeback(page); unlock_page(page); - - if (nilfs_test_page_to_be_frozen(page)) { - int err = nilfs_copy_replace_page_buffers(page, out); - if (unlikely(err)) - return err; - } - return 0; } -static int nilfs_segctor_prepare_write(struct nilfs_sc_info *sci, - struct page **failed_page) +static void nilfs_segctor_prepare_write(struct nilfs_sc_info *sci) { struct nilfs_segment_buffer *segbuf; struct page *bd_page = NULL, *fs_page = NULL; - struct list_head *list = &sci->sc_copied_buffers; - int err; - *failed_page = NULL; list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list) { struct buffer_head *bh; @@ -1662,11 +1600,7 @@ static int nilfs_segctor_prepare_write(struct nilfs_sc_info *sci, break; } if (bh->b_page != fs_page) { - err = nilfs_begin_page_io(fs_page, list); - if (unlikely(err)) { - *failed_page = fs_page; - goto out; - } + nilfs_begin_page_io(fs_page); fs_page = bh->b_page; } } @@ -1677,11 +1611,7 @@ static int nilfs_segctor_prepare_write(struct nilfs_sc_info *sci, set_page_writeback(bd_page); unlock_page(bd_page); } - err = nilfs_begin_page_io(fs_page, list); - if (unlikely(err)) - *failed_page = fs_page; - out: - return err; + nilfs_begin_page_io(fs_page); } static int nilfs_segctor_write(struct nilfs_sc_info *sci, @@ -1694,24 +1624,6 @@ static int nilfs_segctor_write(struct nilfs_sc_info *sci, return ret; } -static void __nilfs_end_page_io(struct page *page, int err) -{ - if (!err) { - if (!nilfs_page_buffers_clean(page)) - __set_page_dirty_nobuffers(page); - ClearPageError(page); - } else { - __set_page_dirty_nobuffers(page); - SetPageError(page); - } - - if (buffer_nilfs_allocated(page_buffers(page))) { - if (TestClearPageWriteback(page)) - dec_zone_page_state(page, NR_WRITEBACK); - } else - end_page_writeback(page); -} - static void nilfs_end_page_io(struct page *page, int err) { if (!page) @@ -1738,40 +1650,19 @@ static void nilfs_end_page_io(struct page *page, int err) return; } - __nilfs_end_page_io(page, err); -} - -static void nilfs_clear_copied_buffers(struct list_head *list, int err) -{ - struct buffer_head *bh, *head; - struct page *page; - - while (!list_empty(list)) { - bh = list_entry(list->next, struct buffer_head, - b_assoc_buffers); - page = bh->b_page; - page_cache_get(page); - head = bh = page_buffers(page); - do { - if (!list_empty(&bh->b_assoc_buffers)) { - list_del_init(&bh->b_assoc_buffers); - if (!err) { - set_buffer_uptodate(bh); - clear_buffer_dirty(bh); - clear_buffer_delay(bh); - clear_buffer_nilfs_volatile(bh); - } - brelse(bh); /* for b_assoc_buffers */ - } - } while ((bh = bh->b_this_page) != head); - - __nilfs_end_page_io(page, err); - page_cache_release(page); + if (!err) { + if (!nilfs_page_buffers_clean(page)) + __set_page_dirty_nobuffers(page); + ClearPageError(page); + } else { + __set_page_dirty_nobuffers(page); + SetPageError(page); } + + end_page_writeback(page); } -static void nilfs_abort_logs(struct list_head *logs, struct page *failed_page, - int err) +static void nilfs_abort_logs(struct list_head *logs, int err) { struct nilfs_segment_buffer *segbuf; struct page *bd_page = NULL, *fs_page = NULL; @@ -1801,8 +1692,6 @@ static void nilfs_abort_logs(struct list_head *logs, struct page *failed_page, } if (bh->b_page != fs_page) { nilfs_end_page_io(fs_page, err); - if (fs_page && fs_page == failed_page) - return; fs_page = bh->b_page; } } @@ -1821,12 +1710,11 @@ static void nilfs_segctor_abort_construction(struct nilfs_sc_info *sci, list_splice_tail_init(&sci->sc_write_logs, &logs); ret = nilfs_wait_on_logs(&logs); - nilfs_abort_logs(&logs, NULL, ret ? : err); + nilfs_abort_logs(&logs, ret ? : err); list_splice_tail_init(&sci->sc_segbufs, &logs); nilfs_cancel_segusage(&logs, nilfs->ns_sufile); nilfs_free_incomplete_logs(&logs, nilfs); - nilfs_clear_copied_buffers(&sci->sc_copied_buffers, err); if (sci->sc_stage.flags & NILFS_CF_SUFREED) { ret = nilfs_sufile_cancel_freev(nilfs->ns_sufile, @@ -1920,8 +1808,6 @@ static void nilfs_segctor_complete_write(struct nilfs_sc_info *sci) nilfs_end_page_io(fs_page, 0); - nilfs_clear_copied_buffers(&sci->sc_copied_buffers, 0); - nilfs_drop_collected_inodes(&sci->sc_dirty_files); if (nilfs_doing_gc()) @@ -1979,7 +1865,7 @@ static int nilfs_segctor_collect_dirty_files(struct nilfs_sc_info *sci, "failed to get inode block.\n"); return err; } - nilfs_mdt_mark_buffer_dirty(ibh); + mark_buffer_dirty(ibh); nilfs_mdt_mark_dirty(ifile); spin_lock(&nilfs->ns_inode_lock); if (likely(!ii->i_bh)) @@ -1991,8 +1877,7 @@ static int nilfs_segctor_collect_dirty_files(struct nilfs_sc_info *sci, clear_bit(NILFS_I_QUEUED, &ii->i_state); set_bit(NILFS_I_BUSY, &ii->i_state); - list_del(&ii->i_dirty); - list_add_tail(&ii->i_dirty, &sci->sc_dirty_files); + list_move_tail(&ii->i_dirty, &sci->sc_dirty_files); } spin_unlock(&nilfs->ns_inode_lock); @@ -2014,8 +1899,7 @@ static void nilfs_segctor_drop_written_files(struct nilfs_sc_info *sci, clear_bit(NILFS_I_BUSY, &ii->i_state); brelse(ii->i_bh); ii->i_bh = NULL; - list_del(&ii->i_dirty); - list_add_tail(&ii->i_dirty, &ti->ti_garbage); + list_move_tail(&ii->i_dirty, &ti->ti_garbage); } spin_unlock(&nilfs->ns_inode_lock); } @@ -2026,7 +1910,6 @@ static void nilfs_segctor_drop_written_files(struct nilfs_sc_info *sci, static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode) { struct the_nilfs *nilfs = sci->sc_super->s_fs_info; - struct page *failed_page; int err; sci->sc_stage.scnt = NILFS_ST_INIT; @@ -2081,11 +1964,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode) nilfs_segctor_update_segusage(sci, nilfs->ns_sufile); /* Write partial segments */ - err = nilfs_segctor_prepare_write(sci, &failed_page); - if (err) { - nilfs_abort_logs(&sci->sc_segbufs, failed_page, err); - goto failed_to_write; - } + nilfs_segctor_prepare_write(sci); nilfs_add_checksums_on_logs(&sci->sc_segbufs, nilfs->ns_crc_seed); @@ -2687,7 +2566,6 @@ static struct nilfs_sc_info *nilfs_segctor_new(struct super_block *sb, INIT_LIST_HEAD(&sci->sc_segbufs); INIT_LIST_HEAD(&sci->sc_write_logs); INIT_LIST_HEAD(&sci->sc_gc_inodes); - INIT_LIST_HEAD(&sci->sc_copied_buffers); init_timer(&sci->sc_timer); sci->sc_interval = HZ * NILFS_SC_DEFAULT_TIMEOUT; @@ -2695,7 +2573,7 @@ static struct nilfs_sc_info *nilfs_segctor_new(struct super_block *sb, sci->sc_watermark = NILFS_SC_DEFAULT_WATERMARK; if (nilfs->ns_interval) - sci->sc_interval = nilfs->ns_interval; + sci->sc_interval = HZ * nilfs->ns_interval; if (nilfs->ns_watermark) sci->sc_watermark = nilfs->ns_watermark; return sci; @@ -2741,8 +2619,6 @@ static void nilfs_segctor_destroy(struct nilfs_sc_info *sci) if (flag || !nilfs_segctor_confirm(sci)) nilfs_segctor_write_out(sci); - WARN_ON(!list_empty(&sci->sc_copied_buffers)); - if (!list_empty(&sci->sc_dirty_files)) { nilfs_warning(sci->sc_super, __func__, "dirty file(s) after the final construction\n"); diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h index 6c02a86..38a1d00 100644 --- a/fs/nilfs2/segment.h +++ b/fs/nilfs2/segment.h @@ -92,7 +92,6 @@ struct nilfs_segsum_pointer { * @sc_nblk_inc: Block count of current generation * @sc_dirty_files: List of files to be written * @sc_gc_inodes: List of GC inodes having blocks to be written - * @sc_copied_buffers: List of copied buffers (buffer heads) to freeze data * @sc_freesegs: array of segment numbers to be freed * @sc_nfreesegs: number of segments on @sc_freesegs * @sc_dsync_inode: inode whose data pages are written for a sync operation @@ -136,7 +135,6 @@ struct nilfs_sc_info { struct list_head sc_dirty_files; struct list_head sc_gc_inodes; - struct list_head sc_copied_buffers; __u64 *sc_freesegs; size_t sc_nfreesegs; diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c index 1d6f488..0a0aba6 100644 --- a/fs/nilfs2/sufile.c +++ b/fs/nilfs2/sufile.c @@ -33,7 +33,9 @@ struct nilfs_sufile_info { struct nilfs_mdt_info mi; - unsigned long ncleansegs; + unsigned long ncleansegs;/* number of clean segments */ + __u64 allocmin; /* lower limit of allocatable segment range */ + __u64 allocmax; /* upper limit of allocatable segment range */ }; static inline struct nilfs_sufile_info *NILFS_SUI(struct inode *sufile) @@ -96,6 +98,13 @@ nilfs_sufile_get_segment_usage_block(struct inode *sufile, __u64 segnum, create, NULL, bhp); } +static int nilfs_sufile_delete_segment_usage_block(struct inode *sufile, + __u64 segnum) +{ + return nilfs_mdt_delete_block(sufile, + nilfs_sufile_get_blkoff(sufile, segnum)); +} + static void nilfs_sufile_mod_counter(struct buffer_head *header_bh, u64 ncleanadd, u64 ndirtyadd) { @@ -108,7 +117,7 @@ static void nilfs_sufile_mod_counter(struct buffer_head *header_bh, le64_add_cpu(&header->sh_ndirtysegs, ndirtyadd); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(header_bh); + mark_buffer_dirty(header_bh); } /** @@ -248,6 +257,35 @@ int nilfs_sufile_update(struct inode *sufile, __u64 segnum, int create, } /** + * nilfs_sufile_set_alloc_range - limit range of segment to be allocated + * @sufile: inode of segment usage file + * @start: minimum segment number of allocatable region (inclusive) + * @end: maximum segment number of allocatable region (inclusive) + * + * Return Value: On success, 0 is returned. On error, one of the + * following negative error codes is returned. + * + * %-ERANGE - invalid segment region + */ +int nilfs_sufile_set_alloc_range(struct inode *sufile, __u64 start, __u64 end) +{ + struct nilfs_sufile_info *sui = NILFS_SUI(sufile); + __u64 nsegs; + int ret = -ERANGE; + + down_write(&NILFS_MDT(sufile)->mi_sem); + nsegs = nilfs_sufile_get_nsegments(sufile); + + if (start <= end && end < nsegs) { + sui->allocmin = start; + sui->allocmax = end; + ret = 0; + } + up_write(&NILFS_MDT(sufile)->mi_sem); + return ret; +} + +/** * nilfs_sufile_alloc - allocate a segment * @sufile: inode of segment usage file * @segnump: pointer to segment number @@ -269,11 +307,12 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) struct buffer_head *header_bh, *su_bh; struct nilfs_sufile_header *header; struct nilfs_segment_usage *su; + struct nilfs_sufile_info *sui = NILFS_SUI(sufile); size_t susz = NILFS_MDT(sufile)->mi_entry_size; __u64 segnum, maxsegnum, last_alloc; void *kaddr; - unsigned long nsegments, ncleansegs, nsus; - int ret, i, j; + unsigned long nsegments, ncleansegs, nsus, cnt; + int ret, j; down_write(&NILFS_MDT(sufile)->mi_sem); @@ -287,13 +326,31 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) kunmap_atomic(kaddr, KM_USER0); nsegments = nilfs_sufile_get_nsegments(sufile); + maxsegnum = sui->allocmax; segnum = last_alloc + 1; - maxsegnum = nsegments - 1; - for (i = 0; i < nsegments; i += nsus) { - if (segnum >= nsegments) { - /* wrap around */ - segnum = 0; - maxsegnum = last_alloc; + if (segnum < sui->allocmin || segnum > sui->allocmax) + segnum = sui->allocmin; + + for (cnt = 0; cnt < nsegments; cnt += nsus) { + if (segnum > maxsegnum) { + if (cnt < sui->allocmax - sui->allocmin + 1) { + /* + * wrap around in the limited region. + * if allocation started from + * sui->allocmin, this never happens. + */ + segnum = sui->allocmin; + maxsegnum = last_alloc; + } else if (segnum > sui->allocmin && + sui->allocmax + 1 < nsegments) { + segnum = sui->allocmax + 1; + maxsegnum = nsegments - 1; + } else if (sui->allocmin > 0) { + segnum = 0; + maxsegnum = sui->allocmin - 1; + } else { + break; /* never happens */ + } } ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 1, &su_bh); @@ -319,9 +376,9 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump) header->sh_last_alloc = cpu_to_le64(segnum); kunmap_atomic(kaddr, KM_USER0); - NILFS_SUI(sufile)->ncleansegs--; - nilfs_mdt_mark_buffer_dirty(header_bh); - nilfs_mdt_mark_buffer_dirty(su_bh); + sui->ncleansegs--; + mark_buffer_dirty(header_bh); + mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); brelse(su_bh); *segnump = segnum; @@ -364,7 +421,7 @@ void nilfs_sufile_do_cancel_free(struct inode *sufile, __u64 segnum, nilfs_sufile_mod_counter(header_bh, -1, 1); NILFS_SUI(sufile)->ncleansegs--; - nilfs_mdt_mark_buffer_dirty(su_bh); + mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); } @@ -395,7 +452,7 @@ void nilfs_sufile_do_scrap(struct inode *sufile, __u64 segnum, nilfs_sufile_mod_counter(header_bh, clean ? (u64)-1 : 0, dirty ? 0 : 1); NILFS_SUI(sufile)->ncleansegs -= clean; - nilfs_mdt_mark_buffer_dirty(su_bh); + mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); } @@ -421,7 +478,7 @@ void nilfs_sufile_do_free(struct inode *sufile, __u64 segnum, sudirty = nilfs_segment_usage_dirty(su); nilfs_segment_usage_set_clean(su); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(su_bh); + mark_buffer_dirty(su_bh); nilfs_sufile_mod_counter(header_bh, 1, sudirty ? (u64)-1 : 0); NILFS_SUI(sufile)->ncleansegs++; @@ -441,7 +498,7 @@ int nilfs_sufile_mark_dirty(struct inode *sufile, __u64 segnum) ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, &bh); if (!ret) { - nilfs_mdt_mark_buffer_dirty(bh); + mark_buffer_dirty(bh); nilfs_mdt_mark_dirty(sufile); brelse(bh); } @@ -476,7 +533,7 @@ int nilfs_sufile_set_segment_usage(struct inode *sufile, __u64 segnum, su->su_nblocks = cpu_to_le32(nblocks); kunmap_atomic(kaddr, KM_USER0); - nilfs_mdt_mark_buffer_dirty(bh); + mark_buffer_dirty(bh); nilfs_mdt_mark_dirty(sufile); brelse(bh); @@ -505,7 +562,7 @@ int nilfs_sufile_get_stat(struct inode *sufile, struct nilfs_sustat *sustat) { struct buffer_head *header_bh; struct nilfs_sufile_header *header; - struct the_nilfs *nilfs = NILFS_I_NILFS(sufile); + struct the_nilfs *nilfs = sufile->i_sb->s_fs_info; void *kaddr; int ret; @@ -555,11 +612,183 @@ void nilfs_sufile_do_set_error(struct inode *sufile, __u64 segnum, nilfs_sufile_mod_counter(header_bh, -1, 0); NILFS_SUI(sufile)->ncleansegs--; } - nilfs_mdt_mark_buffer_dirty(su_bh); + mark_buffer_dirty(su_bh); nilfs_mdt_mark_dirty(sufile); } /** + * nilfs_sufile_truncate_range - truncate range of segment array + * @sufile: inode of segment usage file + * @start: start segment number (inclusive) + * @end: end segment number (inclusive) + * + * Return Value: On success, 0 is returned. On error, one of the + * following negative error codes is returned. + * + * %-EIO - I/O error. + * + * %-ENOMEM - Insufficient amount of memory available. + * + * %-EINVAL - Invalid number of segments specified + * + * %-EBUSY - Dirty or active segments are present in the range + */ +static int nilfs_sufile_truncate_range(struct inode *sufile, + __u64 start, __u64 end) +{ + struct the_nilfs *nilfs = sufile->i_sb->s_fs_info; + struct buffer_head *header_bh; + struct buffer_head *su_bh; + struct nilfs_segment_usage *su, *su2; + size_t susz = NILFS_MDT(sufile)->mi_entry_size; + unsigned long segusages_per_block; + unsigned long nsegs, ncleaned; + __u64 segnum; + void *kaddr; + ssize_t n, nc; + int ret; + int j; + + nsegs = nilfs_sufile_get_nsegments(sufile); + + ret = -EINVAL; + if (start > end || start >= nsegs) + goto out; + + ret = nilfs_sufile_get_header_block(sufile, &header_bh); + if (ret < 0) + goto out; + + segusages_per_block = nilfs_sufile_segment_usages_per_block(sufile); + ncleaned = 0; + + for (segnum = start; segnum <= end; segnum += n) { + n = min_t(unsigned long, + segusages_per_block - + nilfs_sufile_get_offset(sufile, segnum), + end - segnum + 1); + ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, + &su_bh); + if (ret < 0) { + if (ret != -ENOENT) + goto out_header; + /* hole */ + continue; + } + kaddr = kmap_atomic(su_bh->b_page, KM_USER0); + su = nilfs_sufile_block_get_segment_usage( + sufile, segnum, su_bh, kaddr); + su2 = su; + for (j = 0; j < n; j++, su = (void *)su + susz) { + if ((le32_to_cpu(su->su_flags) & + ~(1UL << NILFS_SEGMENT_USAGE_ERROR)) || + nilfs_segment_is_active(nilfs, segnum + j)) { + ret = -EBUSY; + kunmap_atomic(kaddr, KM_USER0); + brelse(su_bh); + goto out_header; + } + } + nc = 0; + for (su = su2, j = 0; j < n; j++, su = (void *)su + susz) { + if (nilfs_segment_usage_error(su)) { + nilfs_segment_usage_set_clean(su); + nc++; + } + } + kunmap_atomic(kaddr, KM_USER0); + if (nc > 0) { + mark_buffer_dirty(su_bh); + ncleaned += nc; + } + brelse(su_bh); + + if (n == segusages_per_block) { + /* make hole */ + nilfs_sufile_delete_segment_usage_block(sufile, segnum); + } + } + ret = 0; + +out_header: + if (ncleaned > 0) { + NILFS_SUI(sufile)->ncleansegs += ncleaned; + nilfs_sufile_mod_counter(header_bh, ncleaned, 0); + nilfs_mdt_mark_dirty(sufile); + } + brelse(header_bh); +out: + return ret; +} + +/** + * nilfs_sufile_resize - resize segment array + * @sufile: inode of segment usage file + * @newnsegs: new number of segments + * + * Return Value: On success, 0 is returned. On error, one of the + * following negative error codes is returned. + * + * %-EIO - I/O error. + * + * %-ENOMEM - Insufficient amount of memory available. + * + * %-ENOSPC - Enough free space is not left for shrinking + * + * %-EBUSY - Dirty or active segments exist in the region to be truncated + */ +int nilfs_sufile_resize(struct inode *sufile, __u64 newnsegs) +{ + struct the_nilfs *nilfs = sufile->i_sb->s_fs_info; + struct buffer_head *header_bh; + struct nilfs_sufile_header *header; + struct nilfs_sufile_info *sui = NILFS_SUI(sufile); + void *kaddr; + unsigned long nsegs, nrsvsegs; + int ret = 0; + + down_write(&NILFS_MDT(sufile)->mi_sem); + + nsegs = nilfs_sufile_get_nsegments(sufile); + if (nsegs == newnsegs) + goto out; + + ret = -ENOSPC; + nrsvsegs = nilfs_nrsvsegs(nilfs, newnsegs); + if (newnsegs < nsegs && nsegs - newnsegs + nrsvsegs > sui->ncleansegs) + goto out; + + ret = nilfs_sufile_get_header_block(sufile, &header_bh); + if (ret < 0) + goto out; + + if (newnsegs > nsegs) { + sui->ncleansegs += newnsegs - nsegs; + } else /* newnsegs < nsegs */ { + ret = nilfs_sufile_truncate_range(sufile, newnsegs, nsegs - 1); + if (ret < 0) + goto out_header; + + sui->ncleansegs -= nsegs - newnsegs; + } + + kaddr = kmap_atomic(header_bh->b_page, KM_USER0); + header = kaddr + bh_offset(header_bh); + header->sh_ncleansegs = cpu_to_le64(sui->ncleansegs); + kunmap_atomic(kaddr, KM_USER0); + + mark_buffer_dirty(header_bh); + nilfs_mdt_mark_dirty(sufile); + nilfs_set_nsegments(nilfs, newnsegs); + +out_header: + brelse(header_bh); +out: + up_write(&NILFS_MDT(sufile)->mi_sem); + return ret; +} + +/** * nilfs_sufile_get_suinfo - * @sufile: inode of segment usage file * @segnum: segment number to start looking @@ -583,7 +812,7 @@ ssize_t nilfs_sufile_get_suinfo(struct inode *sufile, __u64 segnum, void *buf, struct nilfs_segment_usage *su; struct nilfs_suinfo *si = buf; size_t susz = NILFS_MDT(sufile)->mi_entry_size; - struct the_nilfs *nilfs = NILFS_I_NILFS(sufile); + struct the_nilfs *nilfs = sufile->i_sb->s_fs_info; void *kaddr; unsigned long nsegs, segusages_per_block; ssize_t n; @@ -679,6 +908,9 @@ int nilfs_sufile_read(struct super_block *sb, size_t susize, kunmap_atomic(kaddr, KM_USER0); brelse(header_bh); + sui->allocmax = nilfs_sufile_get_nsegments(sufile) - 1; + sui->allocmin = 0; + unlock_new_inode(sufile); out: *inodep = sufile; diff --git a/fs/nilfs2/sufile.h b/fs/nilfs2/sufile.h index a943fba..e84bc5b 100644 --- a/fs/nilfs2/sufile.h +++ b/fs/nilfs2/sufile.h @@ -31,11 +31,12 @@ static inline unsigned long nilfs_sufile_get_nsegments(struct inode *sufile) { - return NILFS_I_NILFS(sufile)->ns_nsegments; + return ((struct the_nilfs *)sufile->i_sb->s_fs_info)->ns_nsegments; } unsigned long nilfs_sufile_get_ncleansegs(struct inode *sufile); +int nilfs_sufile_set_alloc_range(struct inode *sufile, __u64 start, __u64 end); int nilfs_sufile_alloc(struct inode *, __u64 *); int nilfs_sufile_mark_dirty(struct inode *sufile, __u64 segnum); int nilfs_sufile_set_segment_usage(struct inode *sufile, __u64 segnum, @@ -61,6 +62,7 @@ void nilfs_sufile_do_cancel_free(struct inode *, __u64, struct buffer_head *, void nilfs_sufile_do_set_error(struct inode *, __u64, struct buffer_head *, struct buffer_head *); +int nilfs_sufile_resize(struct inode *sufile, __u64 newnsegs); int nilfs_sufile_read(struct super_block *sb, size_t susize, struct nilfs_inode *raw_inode, struct inode **inodep); diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 062cca0..8351c44 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -56,6 +56,7 @@ #include "btnode.h" #include "page.h" #include "cpfile.h" +#include "sufile.h" /* nilfs_sufile_resize(), nilfs_sufile_set_alloc_range() */ #include "ifile.h" #include "dat.h" #include "segment.h" @@ -165,7 +166,7 @@ struct inode *nilfs_alloc_inode(struct super_block *sb) ii->i_state = 0; ii->i_cno = 0; ii->vfs_inode.i_version = 1; - nilfs_btnode_cache_init(&ii->i_btnode_cache, sb->s_bdi); + nilfs_mapping_init(&ii->i_btnode_cache, &ii->vfs_inode, sb->s_bdi); return &ii->vfs_inode; } @@ -347,6 +348,134 @@ int nilfs_cleanup_super(struct super_block *sb) return ret; } +/** + * nilfs_move_2nd_super - relocate secondary super block + * @sb: super block instance + * @sb2off: new offset of the secondary super block (in bytes) + */ +static int nilfs_move_2nd_super(struct super_block *sb, loff_t sb2off) +{ + struct the_nilfs *nilfs = sb->s_fs_info; + struct buffer_head *nsbh; + struct nilfs_super_block *nsbp; + sector_t blocknr, newblocknr; + unsigned long offset; + int sb2i = -1; /* array index of the secondary superblock */ + int ret = 0; + + /* nilfs->ns_sem must be locked by the caller. */ + if (nilfs->ns_sbh[1] && + nilfs->ns_sbh[1]->b_blocknr > nilfs->ns_first_data_block) { + sb2i = 1; + blocknr = nilfs->ns_sbh[1]->b_blocknr; + } else if (nilfs->ns_sbh[0]->b_blocknr > nilfs->ns_first_data_block) { + sb2i = 0; + blocknr = nilfs->ns_sbh[0]->b_blocknr; + } + if (sb2i >= 0 && (u64)blocknr << nilfs->ns_blocksize_bits == sb2off) + goto out; /* super block location is unchanged */ + + /* Get new super block buffer */ + newblocknr = sb2off >> nilfs->ns_blocksize_bits; + offset = sb2off & (nilfs->ns_blocksize - 1); + nsbh = sb_getblk(sb, newblocknr); + if (!nsbh) { + printk(KERN_WARNING + "NILFS warning: unable to move secondary superblock " + "to block %llu\n", (unsigned long long)newblocknr); + ret = -EIO; + goto out; + } + nsbp = (void *)nsbh->b_data + offset; + memset(nsbp, 0, nilfs->ns_blocksize); + + if (sb2i >= 0) { + memcpy(nsbp, nilfs->ns_sbp[sb2i], nilfs->ns_sbsize); + brelse(nilfs->ns_sbh[sb2i]); + nilfs->ns_sbh[sb2i] = nsbh; + nilfs->ns_sbp[sb2i] = nsbp; + } else if (nilfs->ns_sbh[0]->b_blocknr < nilfs->ns_first_data_block) { + /* secondary super block will be restored to index 1 */ + nilfs->ns_sbh[1] = nsbh; + nilfs->ns_sbp[1] = nsbp; + } else { + brelse(nsbh); + } +out: + return ret; +} + +/** + * nilfs_resize_fs - resize the filesystem + * @sb: super block instance + * @newsize: new size of the filesystem (in bytes) + */ +int nilfs_resize_fs(struct super_block *sb, __u64 newsize) +{ + struct the_nilfs *nilfs = sb->s_fs_info; + struct nilfs_super_block **sbp; + __u64 devsize, newnsegs; + loff_t sb2off; + int ret; + + ret = -ERANGE; + devsize = i_size_read(sb->s_bdev->bd_inode); + if (newsize > devsize) + goto out; + + /* + * Write lock is required to protect some functions depending + * on the number of segments, the number of reserved segments, + * and so forth. + */ + down_write(&nilfs->ns_segctor_sem); + + sb2off = NILFS_SB2_OFFSET_BYTES(newsize); + newnsegs = sb2off >> nilfs->ns_blocksize_bits; + do_div(newnsegs, nilfs->ns_blocks_per_segment); + + ret = nilfs_sufile_resize(nilfs->ns_sufile, newnsegs); + up_write(&nilfs->ns_segctor_sem); + if (ret < 0) + goto out; + + ret = nilfs_construct_segment(sb); + if (ret < 0) + goto out; + + down_write(&nilfs->ns_sem); + nilfs_move_2nd_super(sb, sb2off); + ret = -EIO; + sbp = nilfs_prepare_super(sb, 0); + if (likely(sbp)) { + nilfs_set_log_cursor(sbp[0], nilfs); + /* + * Drop NILFS_RESIZE_FS flag for compatibility with + * mount-time resize which may be implemented in a + * future release. + */ + sbp[0]->s_state = cpu_to_le16(le16_to_cpu(sbp[0]->s_state) & + ~NILFS_RESIZE_FS); + sbp[0]->s_dev_size = cpu_to_le64(newsize); + sbp[0]->s_nsegments = cpu_to_le64(nilfs->ns_nsegments); + if (sbp[1]) + memcpy(sbp[1], sbp[0], nilfs->ns_sbsize); + ret = nilfs_commit_super(sb, NILFS_SB_COMMIT_ALL); + } + up_write(&nilfs->ns_sem); + + /* + * Reset the range of allocatable segments last. This order + * is important in the case of expansion because the secondary + * superblock must be protected from log write until migration + * completes. + */ + if (!ret) + nilfs_sufile_set_alloc_range(nilfs->ns_sufile, 0, newnsegs - 1); +out: + return ret; +} + static void nilfs_put_super(struct super_block *sb) { struct the_nilfs *nilfs = sb->s_fs_info; diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index d2acd1a..d327140 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -363,6 +363,24 @@ static unsigned long long nilfs_max_size(unsigned int blkbits) return res; } +/** + * nilfs_nrsvsegs - calculate the number of reserved segments + * @nilfs: nilfs object + * @nsegs: total number of segments + */ +unsigned long nilfs_nrsvsegs(struct the_nilfs *nilfs, unsigned long nsegs) +{ + return max_t(unsigned long, NILFS_MIN_NRSVSEGS, + DIV_ROUND_UP(nsegs * nilfs->ns_r_segments_percentage, + 100)); +} + +void nilfs_set_nsegments(struct the_nilfs *nilfs, unsigned long nsegs) +{ + nilfs->ns_nsegments = nsegs; + nilfs->ns_nrsvsegs = nilfs_nrsvsegs(nilfs, nsegs); +} + static int nilfs_store_disk_layout(struct the_nilfs *nilfs, struct nilfs_super_block *sbp) { @@ -389,13 +407,9 @@ static int nilfs_store_disk_layout(struct the_nilfs *nilfs, } nilfs->ns_first_data_block = le64_to_cpu(sbp->s_first_data_block); - nilfs->ns_nsegments = le64_to_cpu(sbp->s_nsegments); nilfs->ns_r_segments_percentage = le32_to_cpu(sbp->s_r_segments_percentage); - nilfs->ns_nrsvsegs = - max_t(unsigned long, NILFS_MIN_NRSVSEGS, - DIV_ROUND_UP(nilfs->ns_nsegments * - nilfs->ns_r_segments_percentage, 100)); + nilfs_set_nsegments(nilfs, le64_to_cpu(sbp->s_nsegments)); nilfs->ns_crc_seed = le32_to_cpu(sbp->s_crc_seed); return 0; } diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index f496814..9992b11 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -268,6 +268,8 @@ struct the_nilfs *alloc_nilfs(struct block_device *bdev); void destroy_nilfs(struct the_nilfs *nilfs); int init_nilfs(struct the_nilfs *nilfs, struct super_block *sb, char *data); int load_nilfs(struct the_nilfs *nilfs, struct super_block *sb); +unsigned long nilfs_nrsvsegs(struct the_nilfs *nilfs, unsigned long nsegs); +void nilfs_set_nsegments(struct the_nilfs *nilfs, unsigned long nsegs); int nilfs_discard_segments(struct the_nilfs *, __u64 *, size_t); int nilfs_count_free_blocks(struct the_nilfs *, sector_t *); struct nilfs_root *nilfs_lookup_root(struct the_nilfs *nilfs, __u64 cno); diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile index d8a0313..f17e58b 100644 --- a/fs/ocfs2/Makefile +++ b/fs/ocfs2/Makefile @@ -30,6 +30,7 @@ ocfs2-objs := \ namei.o \ refcounttree.o \ reservations.o \ + move_extents.o \ resize.o \ slot_map.o \ suballoc.o \ diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 48aa9c7..ed553c6 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -29,6 +29,7 @@ #include <linux/highmem.h> #include <linux/swap.h> #include <linux/quotaops.h> +#include <linux/blkdev.h> #include <cluster/masklog.h> @@ -7184,3 +7185,168 @@ out_commit: out: return ret; } + +static int ocfs2_trim_extent(struct super_block *sb, + struct ocfs2_group_desc *gd, + u32 start, u32 count) +{ + u64 discard, bcount; + + bcount = ocfs2_clusters_to_blocks(sb, count); + discard = le64_to_cpu(gd->bg_blkno) + + ocfs2_clusters_to_blocks(sb, start); + + trace_ocfs2_trim_extent(sb, (unsigned long long)discard, bcount); + + return sb_issue_discard(sb, discard, bcount, GFP_NOFS, 0); +} + +static int ocfs2_trim_group(struct super_block *sb, + struct ocfs2_group_desc *gd, + u32 start, u32 max, u32 minbits) +{ + int ret = 0, count = 0, next; + void *bitmap = gd->bg_bitmap; + + if (le16_to_cpu(gd->bg_free_bits_count) < minbits) + return 0; + + trace_ocfs2_trim_group((unsigned long long)le64_to_cpu(gd->bg_blkno), + start, max, minbits); + + while (start < max) { + start = ocfs2_find_next_zero_bit(bitmap, max, start); + if (start >= max) + break; + next = ocfs2_find_next_bit(bitmap, max, start); + + if ((next - start) >= minbits) { + ret = ocfs2_trim_extent(sb, gd, + start, next - start); + if (ret < 0) { + mlog_errno(ret); + break; + } + count += next - start; + } + start = next + 1; + + if (fatal_signal_pending(current)) { + count = -ERESTARTSYS; + break; + } + + if ((le16_to_cpu(gd->bg_free_bits_count) - count) < minbits) + break; + } + + if (ret < 0) + count = ret; + + return count; +} + +int ocfs2_trim_fs(struct super_block *sb, struct fstrim_range *range) +{ + struct ocfs2_super *osb = OCFS2_SB(sb); + u64 start, len, trimmed, first_group, last_group, group; + int ret, cnt; + u32 first_bit, last_bit, minlen; + struct buffer_head *main_bm_bh = NULL; + struct inode *main_bm_inode = NULL; + struct buffer_head *gd_bh = NULL; + struct ocfs2_dinode *main_bm; + struct ocfs2_group_desc *gd = NULL; + + start = range->start >> osb->s_clustersize_bits; + len = range->len >> osb->s_clustersize_bits; + minlen = range->minlen >> osb->s_clustersize_bits; + trimmed = 0; + + if (!len) { + range->len = 0; + return 0; + } + + if (minlen >= osb->bitmap_cpg) + return -EINVAL; + + main_bm_inode = ocfs2_get_system_file_inode(osb, + GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT); + if (!main_bm_inode) { + ret = -EIO; + mlog_errno(ret); + goto out; + } + + mutex_lock(&main_bm_inode->i_mutex); + + ret = ocfs2_inode_lock(main_bm_inode, &main_bm_bh, 0); + if (ret < 0) { + mlog_errno(ret); + goto out_mutex; + } + main_bm = (struct ocfs2_dinode *)main_bm_bh->b_data; + + if (start >= le32_to_cpu(main_bm->i_clusters)) { + ret = -EINVAL; + goto out_unlock; + } + + if (start + len > le32_to_cpu(main_bm->i_clusters)) + len = le32_to_cpu(main_bm->i_clusters) - start; + + trace_ocfs2_trim_fs(start, len, minlen); + + /* Determine first and last group to examine based on start and len */ + first_group = ocfs2_which_cluster_group(main_bm_inode, start); + if (first_group == osb->first_cluster_group_blkno) + first_bit = start; + else + first_bit = start - ocfs2_blocks_to_clusters(sb, first_group); + last_group = ocfs2_which_cluster_group(main_bm_inode, start + len - 1); + last_bit = osb->bitmap_cpg; + + for (group = first_group; group <= last_group;) { + if (first_bit + len >= osb->bitmap_cpg) + last_bit = osb->bitmap_cpg; + else + last_bit = first_bit + len; + + ret = ocfs2_read_group_descriptor(main_bm_inode, + main_bm, group, + &gd_bh); + if (ret < 0) { + mlog_errno(ret); + break; + } + + gd = (struct ocfs2_group_desc *)gd_bh->b_data; + cnt = ocfs2_trim_group(sb, gd, first_bit, last_bit, minlen); + brelse(gd_bh); + gd_bh = NULL; + if (cnt < 0) { + ret = cnt; + mlog_errno(ret); + break; + } + + trimmed += cnt; + len -= osb->bitmap_cpg - first_bit; + first_bit = 0; + if (group == osb->first_cluster_group_blkno) + group = ocfs2_clusters_to_blocks(sb, osb->bitmap_cpg); + else + group += ocfs2_clusters_to_blocks(sb, osb->bitmap_cpg); + } + range->len = trimmed * sb->s_blocksize; +out_unlock: + ocfs2_inode_unlock(main_bm_inode, 0); + brelse(main_bm_bh); +out_mutex: + mutex_unlock(&main_bm_inode->i_mutex); + iput(main_bm_inode); +out: + return ret; +} diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h index 3bd08a0..ca381c5 100644 --- a/fs/ocfs2/alloc.h +++ b/fs/ocfs2/alloc.h @@ -239,6 +239,7 @@ int ocfs2_find_leaf(struct ocfs2_caching_info *ci, struct buffer_head **leaf_bh); int ocfs2_search_extent_list(struct ocfs2_extent_list *el, u32 v_cluster); +int ocfs2_trim_fs(struct super_block *sb, struct fstrim_range *range); /* * Helper function to look at the # of clusters in an extent record. */ diff --git a/fs/ocfs2/cluster/sys.c b/fs/ocfs2/cluster/sys.c index bc702da..a4b0773 100644 --- a/fs/ocfs2/cluster/sys.c +++ b/fs/ocfs2/cluster/sys.c @@ -57,7 +57,6 @@ static struct kset *o2cb_kset; void o2cb_sys_shutdown(void) { mlog_sys_shutdown(); - sysfs_remove_link(NULL, "o2cb"); kset_unregister(o2cb_kset); } @@ -69,14 +68,6 @@ int o2cb_sys_init(void) if (!o2cb_kset) return -ENOMEM; - /* - * Create this symlink for backwards compatibility with old - * versions of ocfs2-tools which look for things in /sys/o2cb. - */ - ret = sysfs_create_link(NULL, &o2cb_kset->kobj, "o2cb"); - if (ret) - goto error; - ret = sysfs_create_group(&o2cb_kset->kobj, &o2cb_attr_group); if (ret) goto error; diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index 4bdf7ba..d602abb 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -144,6 +144,7 @@ struct dlm_ctxt wait_queue_head_t dlm_join_events; unsigned long live_nodes_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long domain_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; + unsigned long exit_domain_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long recovery_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; struct dlm_recovery_ctxt reco; spinlock_t master_lock; @@ -401,6 +402,18 @@ static inline int dlm_lvb_is_empty(char *lvb) return 1; } +static inline char *dlm_list_in_text(enum dlm_lockres_list idx) +{ + if (idx == DLM_GRANTED_LIST) + return "granted"; + else if (idx == DLM_CONVERTING_LIST) + return "converting"; + else if (idx == DLM_BLOCKED_LIST) + return "blocked"; + else + return "unknown"; +} + static inline struct list_head * dlm_list_idx_to_ptr(struct dlm_lock_resource *res, enum dlm_lockres_list idx) { @@ -448,6 +461,7 @@ enum { DLM_FINALIZE_RECO_MSG = 518, DLM_QUERY_REGION = 519, DLM_QUERY_NODEINFO = 520, + DLM_BEGIN_EXIT_DOMAIN_MSG = 521, }; struct dlm_reco_node_data diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c index 04a32be..56f82cb 100644 --- a/fs/ocfs2/dlm/dlmdebug.c +++ b/fs/ocfs2/dlm/dlmdebug.c @@ -756,6 +756,12 @@ static int debug_state_print(struct dlm_ctxt *dlm, char *buf, int len) buf + out, len - out); out += snprintf(buf + out, len - out, "\n"); + /* Exit Domain Map: xx xx xx */ + out += snprintf(buf + out, len - out, "Exit Domain Map: "); + out += stringify_nodemap(dlm->exit_domain_map, O2NM_MAX_NODES, + buf + out, len - out); + out += snprintf(buf + out, len - out, "\n"); + /* Live Map: xx xx xx */ out += snprintf(buf + out, len - out, "Live Map: "); out += stringify_nodemap(dlm->live_nodes_map, O2NM_MAX_NODES, diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 3b179d6..6ed6b95 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -132,10 +132,12 @@ static DECLARE_WAIT_QUEUE_HEAD(dlm_domain_events); * New in version 1.1: * - Message DLM_QUERY_REGION added to support global heartbeat * - Message DLM_QUERY_NODEINFO added to allow online node removes + * New in version 1.2: + * - Message DLM_BEGIN_EXIT_DOMAIN_MSG added to mark start of exit domain */ static const struct dlm_protocol_version dlm_protocol = { .pv_major = 1, - .pv_minor = 1, + .pv_minor = 2, }; #define DLM_DOMAIN_BACKOFF_MS 200 @@ -449,14 +451,18 @@ redo_bucket: dropped = dlm_empty_lockres(dlm, res); spin_lock(&res->spinlock); - __dlm_lockres_calc_usage(dlm, res); - iter = res->hash_node.next; + if (dropped) + __dlm_lockres_calc_usage(dlm, res); + else + iter = res->hash_node.next; spin_unlock(&res->spinlock); dlm_lockres_put(res); - if (dropped) + if (dropped) { + cond_resched_lock(&dlm->spinlock); goto redo_bucket; + } } cond_resched_lock(&dlm->spinlock); num += n; @@ -486,6 +492,28 @@ static int dlm_no_joining_node(struct dlm_ctxt *dlm) return ret; } +static int dlm_begin_exit_domain_handler(struct o2net_msg *msg, u32 len, + void *data, void **ret_data) +{ + struct dlm_ctxt *dlm = data; + unsigned int node; + struct dlm_exit_domain *exit_msg = (struct dlm_exit_domain *) msg->buf; + + if (!dlm_grab(dlm)) + return 0; + + node = exit_msg->node_idx; + mlog(0, "%s: Node %u sent a begin exit domain message\n", dlm->name, node); + + spin_lock(&dlm->spinlock); + set_bit(node, dlm->exit_domain_map); + spin_unlock(&dlm->spinlock); + + dlm_put(dlm); + + return 0; +} + static void dlm_mark_domain_leaving(struct dlm_ctxt *dlm) { /* Yikes, a double spinlock! I need domain_lock for the dlm @@ -542,6 +570,7 @@ static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, spin_lock(&dlm->spinlock); clear_bit(node, dlm->domain_map); + clear_bit(node, dlm->exit_domain_map); __dlm_print_nodes(dlm); /* notify anything attached to the heartbeat events */ @@ -554,29 +583,56 @@ static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, return 0; } -static int dlm_send_one_domain_exit(struct dlm_ctxt *dlm, +static int dlm_send_one_domain_exit(struct dlm_ctxt *dlm, u32 msg_type, unsigned int node) { int status; struct dlm_exit_domain leave_msg; - mlog(0, "Asking node %u if we can leave the domain %s me = %u\n", - node, dlm->name, dlm->node_num); + mlog(0, "%s: Sending domain exit message %u to node %u\n", dlm->name, + msg_type, node); memset(&leave_msg, 0, sizeof(leave_msg)); leave_msg.node_idx = dlm->node_num; - status = o2net_send_message(DLM_EXIT_DOMAIN_MSG, dlm->key, - &leave_msg, sizeof(leave_msg), node, - NULL); + status = o2net_send_message(msg_type, dlm->key, &leave_msg, + sizeof(leave_msg), node, NULL); if (status < 0) - mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to " - "node %u\n", status, DLM_EXIT_DOMAIN_MSG, dlm->key, node); - mlog(0, "status return %d from o2net_send_message\n", status); + mlog(ML_ERROR, "Error %d sending domain exit message %u " + "to node %u on domain %s\n", status, msg_type, node, + dlm->name); return status; } +static void dlm_begin_exit_domain(struct dlm_ctxt *dlm) +{ + int node = -1; + + /* Support for begin exit domain was added in 1.2 */ + if (dlm->dlm_locking_proto.pv_major == 1 && + dlm->dlm_locking_proto.pv_minor < 2) + return; + + /* + * Unlike DLM_EXIT_DOMAIN_MSG, DLM_BEGIN_EXIT_DOMAIN_MSG is purely + * informational. Meaning if a node does not receive the message, + * so be it. + */ + spin_lock(&dlm->spinlock); + while (1) { + node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, node + 1); + if (node >= O2NM_MAX_NODES) + break; + if (node == dlm->node_num) + continue; + + spin_unlock(&dlm->spinlock); + dlm_send_one_domain_exit(dlm, DLM_BEGIN_EXIT_DOMAIN_MSG, node); + spin_lock(&dlm->spinlock); + } + spin_unlock(&dlm->spinlock); +} static void dlm_leave_domain(struct dlm_ctxt *dlm) { @@ -602,7 +658,8 @@ static void dlm_leave_domain(struct dlm_ctxt *dlm) clear_node = 1; - status = dlm_send_one_domain_exit(dlm, node); + status = dlm_send_one_domain_exit(dlm, DLM_EXIT_DOMAIN_MSG, + node); if (status < 0 && status != -ENOPROTOOPT && status != -ENOTCONN) { @@ -677,6 +734,7 @@ void dlm_unregister_domain(struct dlm_ctxt *dlm) if (leave) { mlog(0, "shutting down domain %s\n", dlm->name); + dlm_begin_exit_domain(dlm); /* We changed dlm state, notify the thread */ dlm_kick_thread(dlm, NULL); @@ -909,6 +967,7 @@ static int dlm_assert_joined_handler(struct o2net_msg *msg, u32 len, void *data, * leftover join state. */ BUG_ON(dlm->joining_node != assert->node_idx); set_bit(assert->node_idx, dlm->domain_map); + clear_bit(assert->node_idx, dlm->exit_domain_map); __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); printk(KERN_NOTICE "o2dlm: Node %u joins domain %s\n", @@ -1793,6 +1852,13 @@ static int dlm_register_domain_handlers(struct dlm_ctxt *dlm) if (status) goto bail; + status = o2net_register_handler(DLM_BEGIN_EXIT_DOMAIN_MSG, dlm->key, + sizeof(struct dlm_exit_domain), + dlm_begin_exit_domain_handler, + dlm, NULL, &dlm->dlm_domain_handlers); + if (status) + goto bail; + bail: if (status) dlm_unregister_domain_handlers(dlm); diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 84d1663..11eefb8 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -2339,65 +2339,55 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) dlm_lockres_put(res); } -/* Checks whether the lockres can be migrated. Returns 0 if yes, < 0 - * if not. If 0, numlocks is set to the number of locks in the lockres. +/* + * A migrateable resource is one that is : + * 1. locally mastered, and, + * 2. zero local locks, and, + * 3. one or more non-local locks, or, one or more references + * Returns 1 if yes, 0 if not. */ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - int *numlocks, - int *hasrefs) + struct dlm_lock_resource *res) { - int ret; - int i; - int count = 0; + enum dlm_lockres_list idx; + int nonlocal = 0, node_ref; struct list_head *queue; struct dlm_lock *lock; + u64 cookie; assert_spin_locked(&res->spinlock); - *numlocks = 0; - *hasrefs = 0; - - ret = -EINVAL; - if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { - mlog(0, "cannot migrate lockres with unknown owner!\n"); - goto leave; - } - - if (res->owner != dlm->node_num) { - mlog(0, "cannot migrate lockres this node doesn't own!\n"); - goto leave; - } + if (res->owner != dlm->node_num) + return 0; - ret = 0; - queue = &res->granted; - for (i = 0; i < 3; i++) { + for (idx = DLM_GRANTED_LIST; idx <= DLM_BLOCKED_LIST; idx++) { + queue = dlm_list_idx_to_ptr(res, idx); list_for_each_entry(lock, queue, list) { - ++count; - if (lock->ml.node == dlm->node_num) { - mlog(0, "found a lock owned by this node still " - "on the %s queue! will not migrate this " - "lockres\n", (i == 0 ? "granted" : - (i == 1 ? "converting" : - "blocked"))); - ret = -ENOTEMPTY; - goto leave; + if (lock->ml.node != dlm->node_num) { + nonlocal++; + continue; } + cookie = be64_to_cpu(lock->ml.cookie); + mlog(0, "%s: Not migrateable res %.*s, lock %u:%llu on " + "%s list\n", dlm->name, res->lockname.len, + res->lockname.name, + dlm_get_lock_cookie_node(cookie), + dlm_get_lock_cookie_seq(cookie), + dlm_list_in_text(idx)); + return 0; } - queue++; } - *numlocks = count; - - count = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); - if (count < O2NM_MAX_NODES) - *hasrefs = 1; + if (!nonlocal) { + node_ref = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); + if (node_ref >= O2NM_MAX_NODES) + return 0; + } - mlog(0, "%s: res %.*s, Migrateable, locks %d, refs %d\n", dlm->name, - res->lockname.len, res->lockname.name, *numlocks, *hasrefs); + mlog(0, "%s: res %.*s, Migrateable\n", dlm->name, res->lockname.len, + res->lockname.name); -leave: - return ret; + return 1; } /* @@ -2406,8 +2396,7 @@ leave: static int dlm_migrate_lockres(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - u8 target) + struct dlm_lock_resource *res, u8 target) { struct dlm_master_list_entry *mle = NULL; struct dlm_master_list_entry *oldmle = NULL; @@ -2416,37 +2405,20 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, const char *name; unsigned int namelen; int mle_added = 0; - int numlocks, hasrefs; int wake = 0; if (!dlm_grab(dlm)) return -EINVAL; + BUG_ON(target == O2NM_MAX_NODES); + name = res->lockname.name; namelen = res->lockname.len; - mlog(0, "%s: Migrating %.*s to %u\n", dlm->name, namelen, name, target); - - /* - * ensure this lockres is a proper candidate for migration - */ - spin_lock(&res->spinlock); - ret = dlm_is_lockres_migrateable(dlm, res, &numlocks, &hasrefs); - if (ret < 0) { - spin_unlock(&res->spinlock); - goto leave; - } - spin_unlock(&res->spinlock); - - /* no work to do */ - if (numlocks == 0 && !hasrefs) - goto leave; - - /* - * preallocate up front - * if this fails, abort - */ + mlog(0, "%s: Migrating %.*s to node %u\n", dlm->name, namelen, name, + target); + /* preallocate up front. if this fails, abort */ ret = -ENOMEM; mres = (struct dlm_migratable_lockres *) __get_free_page(GFP_NOFS); if (!mres) { @@ -2462,35 +2434,10 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, ret = 0; /* - * find a node to migrate the lockres to - */ - - spin_lock(&dlm->spinlock); - /* pick a new node */ - if (!test_bit(target, dlm->domain_map) || - target >= O2NM_MAX_NODES) { - target = dlm_pick_migration_target(dlm, res); - } - mlog(0, "%s: res %.*s, Node %u chosen for migration\n", dlm->name, - namelen, name, target); - - if (target >= O2NM_MAX_NODES || - !test_bit(target, dlm->domain_map)) { - /* target chosen is not alive */ - ret = -EINVAL; - } - - if (ret) { - spin_unlock(&dlm->spinlock); - goto fail; - } - - mlog(0, "continuing with target = %u\n", target); - - /* * clear any existing master requests and * add the migration mle to the list */ + spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); ret = dlm_add_migration_mle(dlm, res, mle, &oldmle, name, namelen, target, dlm->node_num); @@ -2531,6 +2478,7 @@ fail: dlm_put_mle(mle); } else if (mle) { kmem_cache_free(dlm_mle_cache, mle); + mle = NULL; } goto leave; } @@ -2652,69 +2600,52 @@ leave: if (wake) wake_up(&res->wq); - /* TODO: cleanup */ if (mres) free_page((unsigned long)mres); dlm_put(dlm); - mlog(0, "returning %d\n", ret); + mlog(0, "%s: Migrating %.*s to %u, returns %d\n", dlm->name, namelen, + name, target, ret); return ret; } #define DLM_MIGRATION_RETRY_MS 100 -/* Should be called only after beginning the domain leave process. +/* + * Should be called only after beginning the domain leave process. * There should not be any remaining locks on nonlocal lock resources, * and there should be no local locks left on locally mastered resources. * * Called with the dlm spinlock held, may drop it to do migration, but * will re-acquire before exit. * - * Returns: 1 if dlm->spinlock was dropped/retaken, 0 if never dropped */ + * Returns: 1 if dlm->spinlock was dropped/retaken, 0 if never dropped + */ int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { int ret; int lock_dropped = 0; - int numlocks, hasrefs; + u8 target = O2NM_MAX_NODES; + + assert_spin_locked(&dlm->spinlock); spin_lock(&res->spinlock); - if (res->owner != dlm->node_num) { - if (!__dlm_lockres_unused(res)) { - mlog(ML_ERROR, "%s:%.*s: this node is not master, " - "trying to free this but locks remain\n", - dlm->name, res->lockname.len, res->lockname.name); - } - spin_unlock(&res->spinlock); - goto leave; - } + if (dlm_is_lockres_migrateable(dlm, res)) + target = dlm_pick_migration_target(dlm, res); + spin_unlock(&res->spinlock); - /* No need to migrate a lockres having no locks */ - ret = dlm_is_lockres_migrateable(dlm, res, &numlocks, &hasrefs); - if (ret >= 0 && numlocks == 0 && !hasrefs) { - spin_unlock(&res->spinlock); + if (target == O2NM_MAX_NODES) goto leave; - } - spin_unlock(&res->spinlock); /* Wheee! Migrate lockres here! Will sleep so drop spinlock. */ spin_unlock(&dlm->spinlock); lock_dropped = 1; - while (1) { - ret = dlm_migrate_lockres(dlm, res, O2NM_MAX_NODES); - if (ret >= 0) - break; - if (ret == -ENOTEMPTY) { - mlog(ML_ERROR, "lockres %.*s still has local locks!\n", - res->lockname.len, res->lockname.name); - BUG(); - } - - mlog(0, "lockres %.*s: migrate failed, " - "retrying\n", res->lockname.len, - res->lockname.name); - msleep(DLM_MIGRATION_RETRY_MS); - } + ret = dlm_migrate_lockres(dlm, res, target); + if (ret) + mlog(0, "%s: res %.*s, Migrate to node %u failed with %d\n", + dlm->name, res->lockname.len, res->lockname.name, + target, ret); spin_lock(&dlm->spinlock); leave: return lock_dropped; @@ -2898,61 +2829,55 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, } } -/* for now this is not too intelligent. we will - * need stats to make this do the right thing. - * this just finds the first lock on one of the - * queues and uses that node as the target. */ +/* + * Pick a node to migrate the lock resource to. This function selects a + * potential target based first on the locks and then on refmap. It skips + * nodes that are in the process of exiting the domain. + */ static u8 dlm_pick_migration_target(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { - int i; + enum dlm_lockres_list idx; struct list_head *queue = &res->granted; struct dlm_lock *lock; - int nodenum; + int noderef; + u8 nodenum = O2NM_MAX_NODES; assert_spin_locked(&dlm->spinlock); + assert_spin_locked(&res->spinlock); - spin_lock(&res->spinlock); - for (i=0; i<3; i++) { + /* Go through all the locks */ + for (idx = DLM_GRANTED_LIST; idx <= DLM_BLOCKED_LIST; idx++) { + queue = dlm_list_idx_to_ptr(res, idx); list_for_each_entry(lock, queue, list) { - /* up to the caller to make sure this node - * is alive */ - if (lock->ml.node != dlm->node_num) { - spin_unlock(&res->spinlock); - return lock->ml.node; - } + if (lock->ml.node == dlm->node_num) + continue; + if (test_bit(lock->ml.node, dlm->exit_domain_map)) + continue; + nodenum = lock->ml.node; + goto bail; } - queue++; - } - - nodenum = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); - if (nodenum < O2NM_MAX_NODES) { - spin_unlock(&res->spinlock); - return nodenum; } - spin_unlock(&res->spinlock); - mlog(0, "have not found a suitable target yet! checking domain map\n"); - /* ok now we're getting desperate. pick anyone alive. */ - nodenum = -1; + /* Go thru the refmap */ + noderef = -1; while (1) { - nodenum = find_next_bit(dlm->domain_map, - O2NM_MAX_NODES, nodenum+1); - mlog(0, "found %d in domain map\n", nodenum); - if (nodenum >= O2NM_MAX_NODES) + noderef = find_next_bit(res->refmap, O2NM_MAX_NODES, + noderef + 1); + if (noderef >= O2NM_MAX_NODES) break; - if (nodenum != dlm->node_num) { - mlog(0, "picking %d\n", nodenum); - return nodenum; - } + if (noderef == dlm->node_num) + continue; + if (test_bit(noderef, dlm->exit_domain_map)) + continue; + nodenum = noderef; + goto bail; } - mlog(0, "giving up. no master to migrate to\n"); - return DLM_LOCK_RES_OWNER_UNKNOWN; +bail: + return nodenum; } - - /* this is called by the new master once all lockres * data has been received */ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index f1beb6f..7efab6d 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -2393,6 +2393,7 @@ static void __dlm_hb_node_down(struct dlm_ctxt *dlm, int idx) mlog(0, "node %u being removed from domain map!\n", idx); clear_bit(idx, dlm->domain_map); + clear_bit(idx, dlm->exit_domain_map); /* wake up migration waiters if a node goes down. * perhaps later we can genericize this for other waiters. */ wake_up(&dlm->migration_wq); diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index 8c5c0ed..b420767 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -88,7 +88,7 @@ struct workqueue_struct *user_dlm_worker; * signifies a bast fired on the lock. */ #define DLMFS_CAPABILITIES "bast stackglue" -extern int param_set_dlmfs_capabilities(const char *val, +static int param_set_dlmfs_capabilities(const char *val, struct kernel_param *kp) { printk(KERN_ERR "%s: readonly parameter\n", kp->name); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 89659d6..b1e35a3 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2670,6 +2670,7 @@ const struct file_operations ocfs2_fops_no_plocks = { .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, .splice_write = ocfs2_file_splice_write, + .fallocate = ocfs2_fallocate, }; const struct file_operations ocfs2_dops_no_plocks = { diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index 8f13c59..bc91072 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -22,6 +22,11 @@ #include "ioctl.h" #include "resize.h" #include "refcounttree.h" +#include "sysfile.h" +#include "dir.h" +#include "buffer_head_io.h" +#include "suballoc.h" +#include "move_extents.h" #include <linux/ext2_fs.h> @@ -35,31 +40,27 @@ * be -EFAULT. The error will be returned from the ioctl(2) call. It's * just a best-effort to tell userspace that this request caused the error. */ -static inline void __o2info_set_request_error(struct ocfs2_info_request *kreq, +static inline void o2info_set_request_error(struct ocfs2_info_request *kreq, struct ocfs2_info_request __user *req) { kreq->ir_flags |= OCFS2_INFO_FL_ERROR; (void)put_user(kreq->ir_flags, (__u32 __user *)&(req->ir_flags)); } -#define o2info_set_request_error(a, b) \ - __o2info_set_request_error((struct ocfs2_info_request *)&(a), b) - -static inline void __o2info_set_request_filled(struct ocfs2_info_request *req) +static inline void o2info_set_request_filled(struct ocfs2_info_request *req) { req->ir_flags |= OCFS2_INFO_FL_FILLED; } -#define o2info_set_request_filled(a) \ - __o2info_set_request_filled((struct ocfs2_info_request *)&(a)) - -static inline void __o2info_clear_request_filled(struct ocfs2_info_request *req) +static inline void o2info_clear_request_filled(struct ocfs2_info_request *req) { req->ir_flags &= ~OCFS2_INFO_FL_FILLED; } -#define o2info_clear_request_filled(a) \ - __o2info_clear_request_filled((struct ocfs2_info_request *)&(a)) +static inline int o2info_coherent(struct ocfs2_info_request *req) +{ + return (!(req->ir_flags & OCFS2_INFO_FL_NON_COHERENT)); +} static int ocfs2_get_inode_attr(struct inode *inode, unsigned *flags) { @@ -153,7 +154,7 @@ int ocfs2_info_handle_blocksize(struct inode *inode, oib.ib_blocksize = inode->i_sb->s_blocksize; - o2info_set_request_filled(oib); + o2info_set_request_filled(&oib.ib_req); if (o2info_to_user(oib, req)) goto bail; @@ -161,7 +162,7 @@ int ocfs2_info_handle_blocksize(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oib, req); + o2info_set_request_error(&oib.ib_req, req); return status; } @@ -178,7 +179,7 @@ int ocfs2_info_handle_clustersize(struct inode *inode, oic.ic_clustersize = osb->s_clustersize; - o2info_set_request_filled(oic); + o2info_set_request_filled(&oic.ic_req); if (o2info_to_user(oic, req)) goto bail; @@ -186,7 +187,7 @@ int ocfs2_info_handle_clustersize(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oic, req); + o2info_set_request_error(&oic.ic_req, req); return status; } @@ -203,7 +204,7 @@ int ocfs2_info_handle_maxslots(struct inode *inode, oim.im_max_slots = osb->max_slots; - o2info_set_request_filled(oim); + o2info_set_request_filled(&oim.im_req); if (o2info_to_user(oim, req)) goto bail; @@ -211,7 +212,7 @@ int ocfs2_info_handle_maxslots(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oim, req); + o2info_set_request_error(&oim.im_req, req); return status; } @@ -228,7 +229,7 @@ int ocfs2_info_handle_label(struct inode *inode, memcpy(oil.il_label, osb->vol_label, OCFS2_MAX_VOL_LABEL_LEN); - o2info_set_request_filled(oil); + o2info_set_request_filled(&oil.il_req); if (o2info_to_user(oil, req)) goto bail; @@ -236,7 +237,7 @@ int ocfs2_info_handle_label(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oil, req); + o2info_set_request_error(&oil.il_req, req); return status; } @@ -253,7 +254,7 @@ int ocfs2_info_handle_uuid(struct inode *inode, memcpy(oiu.iu_uuid_str, osb->uuid_str, OCFS2_TEXT_UUID_LEN + 1); - o2info_set_request_filled(oiu); + o2info_set_request_filled(&oiu.iu_req); if (o2info_to_user(oiu, req)) goto bail; @@ -261,7 +262,7 @@ int ocfs2_info_handle_uuid(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oiu, req); + o2info_set_request_error(&oiu.iu_req, req); return status; } @@ -280,7 +281,7 @@ int ocfs2_info_handle_fs_features(struct inode *inode, oif.if_incompat_features = osb->s_feature_incompat; oif.if_ro_compat_features = osb->s_feature_ro_compat; - o2info_set_request_filled(oif); + o2info_set_request_filled(&oif.if_req); if (o2info_to_user(oif, req)) goto bail; @@ -288,7 +289,7 @@ int ocfs2_info_handle_fs_features(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oif, req); + o2info_set_request_error(&oif.if_req, req); return status; } @@ -305,7 +306,7 @@ int ocfs2_info_handle_journal_size(struct inode *inode, oij.ij_journal_size = osb->journal->j_inode->i_size; - o2info_set_request_filled(oij); + o2info_set_request_filled(&oij.ij_req); if (o2info_to_user(oij, req)) goto bail; @@ -313,7 +314,408 @@ int ocfs2_info_handle_journal_size(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oij, req); + o2info_set_request_error(&oij.ij_req, req); + + return status; +} + +int ocfs2_info_scan_inode_alloc(struct ocfs2_super *osb, + struct inode *inode_alloc, u64 blkno, + struct ocfs2_info_freeinode *fi, u32 slot) +{ + int status = 0, unlock = 0; + + struct buffer_head *bh = NULL; + struct ocfs2_dinode *dinode_alloc = NULL; + + if (inode_alloc) + mutex_lock(&inode_alloc->i_mutex); + + if (o2info_coherent(&fi->ifi_req)) { + status = ocfs2_inode_lock(inode_alloc, &bh, 0); + if (status < 0) { + mlog_errno(status); + goto bail; + } + unlock = 1; + } else { + status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh); + if (status < 0) { + mlog_errno(status); + goto bail; + } + } + + dinode_alloc = (struct ocfs2_dinode *)bh->b_data; + + fi->ifi_stat[slot].lfi_total = + le32_to_cpu(dinode_alloc->id1.bitmap1.i_total); + fi->ifi_stat[slot].lfi_free = + le32_to_cpu(dinode_alloc->id1.bitmap1.i_total) - + le32_to_cpu(dinode_alloc->id1.bitmap1.i_used); + +bail: + if (unlock) + ocfs2_inode_unlock(inode_alloc, 0); + + if (inode_alloc) + mutex_unlock(&inode_alloc->i_mutex); + + brelse(bh); + + return status; +} + +int ocfs2_info_handle_freeinode(struct inode *inode, + struct ocfs2_info_request __user *req) +{ + u32 i; + u64 blkno = -1; + char namebuf[40]; + int status = -EFAULT, type = INODE_ALLOC_SYSTEM_INODE; + struct ocfs2_info_freeinode *oifi = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct inode *inode_alloc = NULL; + + oifi = kzalloc(sizeof(struct ocfs2_info_freeinode), GFP_KERNEL); + if (!oifi) { + status = -ENOMEM; + mlog_errno(status); + goto bail; + } + + if (o2info_from_user(*oifi, req)) + goto bail; + + oifi->ifi_slotnum = osb->max_slots; + + for (i = 0; i < oifi->ifi_slotnum; i++) { + if (o2info_coherent(&oifi->ifi_req)) { + inode_alloc = ocfs2_get_system_file_inode(osb, type, i); + if (!inode_alloc) { + mlog(ML_ERROR, "unable to get alloc inode in " + "slot %u\n", i); + status = -EIO; + goto bail; + } + } else { + ocfs2_sprintf_system_inode_name(namebuf, + sizeof(namebuf), + type, i); + status = ocfs2_lookup_ino_from_name(osb->sys_root_inode, + namebuf, + strlen(namebuf), + &blkno); + if (status < 0) { + status = -ENOENT; + goto bail; + } + } + + status = ocfs2_info_scan_inode_alloc(osb, inode_alloc, blkno, oifi, i); + if (status < 0) + goto bail; + + iput(inode_alloc); + inode_alloc = NULL; + } + + o2info_set_request_filled(&oifi->ifi_req); + + if (o2info_to_user(*oifi, req)) + goto bail; + + status = 0; +bail: + if (status) + o2info_set_request_error(&oifi->ifi_req, req); + + kfree(oifi); + + return status; +} + +static void o2ffg_update_histogram(struct ocfs2_info_free_chunk_list *hist, + unsigned int chunksize) +{ + int index; + + index = __ilog2_u32(chunksize); + if (index >= OCFS2_INFO_MAX_HIST) + index = OCFS2_INFO_MAX_HIST - 1; + + hist->fc_chunks[index]++; + hist->fc_clusters[index] += chunksize; +} + +static void o2ffg_update_stats(struct ocfs2_info_freefrag_stats *stats, + unsigned int chunksize) +{ + if (chunksize > stats->ffs_max) + stats->ffs_max = chunksize; + + if (chunksize < stats->ffs_min) + stats->ffs_min = chunksize; + + stats->ffs_avg += chunksize; + stats->ffs_free_chunks_real++; +} + +void ocfs2_info_update_ffg(struct ocfs2_info_freefrag *ffg, + unsigned int chunksize) +{ + o2ffg_update_histogram(&(ffg->iff_ffs.ffs_fc_hist), chunksize); + o2ffg_update_stats(&(ffg->iff_ffs), chunksize); +} + +int ocfs2_info_freefrag_scan_chain(struct ocfs2_super *osb, + struct inode *gb_inode, + struct ocfs2_dinode *gb_dinode, + struct ocfs2_chain_rec *rec, + struct ocfs2_info_freefrag *ffg, + u32 chunks_in_group) +{ + int status = 0, used; + u64 blkno; + + struct buffer_head *bh = NULL; + struct ocfs2_group_desc *bg = NULL; + + unsigned int max_bits, num_clusters; + unsigned int offset = 0, cluster, chunk; + unsigned int chunk_free, last_chunksize = 0; + + if (!le32_to_cpu(rec->c_free)) + goto bail; + + do { + if (!bg) + blkno = le64_to_cpu(rec->c_blkno); + else + blkno = le64_to_cpu(bg->bg_next_group); + + if (bh) { + brelse(bh); + bh = NULL; + } + + if (o2info_coherent(&ffg->iff_req)) + status = ocfs2_read_group_descriptor(gb_inode, + gb_dinode, + blkno, &bh); + else + status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh); + + if (status < 0) { + mlog(ML_ERROR, "Can't read the group descriptor # " + "%llu from device.", (unsigned long long)blkno); + status = -EIO; + goto bail; + } + + bg = (struct ocfs2_group_desc *)bh->b_data; + + if (!le16_to_cpu(bg->bg_free_bits_count)) + continue; + + max_bits = le16_to_cpu(bg->bg_bits); + offset = 0; + + for (chunk = 0; chunk < chunks_in_group; chunk++) { + /* + * last chunk may be not an entire one. + */ + if ((offset + ffg->iff_chunksize) > max_bits) + num_clusters = max_bits - offset; + else + num_clusters = ffg->iff_chunksize; + + chunk_free = 0; + for (cluster = 0; cluster < num_clusters; cluster++) { + used = ocfs2_test_bit(offset, + (unsigned long *)bg->bg_bitmap); + /* + * - chunk_free counts free clusters in #N chunk. + * - last_chunksize records the size(in) clusters + * for the last real free chunk being counted. + */ + if (!used) { + last_chunksize++; + chunk_free++; + } + + if (used && last_chunksize) { + ocfs2_info_update_ffg(ffg, + last_chunksize); + last_chunksize = 0; + } + + offset++; + } + + if (chunk_free == ffg->iff_chunksize) + ffg->iff_ffs.ffs_free_chunks++; + } + + /* + * need to update the info for last free chunk. + */ + if (last_chunksize) + ocfs2_info_update_ffg(ffg, last_chunksize); + + } while (le64_to_cpu(bg->bg_next_group)); + +bail: + brelse(bh); + + return status; +} + +int ocfs2_info_freefrag_scan_bitmap(struct ocfs2_super *osb, + struct inode *gb_inode, u64 blkno, + struct ocfs2_info_freefrag *ffg) +{ + u32 chunks_in_group; + int status = 0, unlock = 0, i; + + struct buffer_head *bh = NULL; + struct ocfs2_chain_list *cl = NULL; + struct ocfs2_chain_rec *rec = NULL; + struct ocfs2_dinode *gb_dinode = NULL; + + if (gb_inode) + mutex_lock(&gb_inode->i_mutex); + + if (o2info_coherent(&ffg->iff_req)) { + status = ocfs2_inode_lock(gb_inode, &bh, 0); + if (status < 0) { + mlog_errno(status); + goto bail; + } + unlock = 1; + } else { + status = ocfs2_read_blocks_sync(osb, blkno, 1, &bh); + if (status < 0) { + mlog_errno(status); + goto bail; + } + } + + gb_dinode = (struct ocfs2_dinode *)bh->b_data; + cl = &(gb_dinode->id2.i_chain); + + /* + * Chunksize(in) clusters from userspace should be + * less than clusters in a group. + */ + if (ffg->iff_chunksize > le16_to_cpu(cl->cl_cpg)) { + status = -EINVAL; + goto bail; + } + + memset(&ffg->iff_ffs, 0, sizeof(struct ocfs2_info_freefrag_stats)); + + ffg->iff_ffs.ffs_min = ~0U; + ffg->iff_ffs.ffs_clusters = + le32_to_cpu(gb_dinode->id1.bitmap1.i_total); + ffg->iff_ffs.ffs_free_clusters = ffg->iff_ffs.ffs_clusters - + le32_to_cpu(gb_dinode->id1.bitmap1.i_used); + + chunks_in_group = le16_to_cpu(cl->cl_cpg) / ffg->iff_chunksize + 1; + + for (i = 0; i < le16_to_cpu(cl->cl_next_free_rec); i++) { + rec = &(cl->cl_recs[i]); + status = ocfs2_info_freefrag_scan_chain(osb, gb_inode, + gb_dinode, + rec, ffg, + chunks_in_group); + if (status) + goto bail; + } + + if (ffg->iff_ffs.ffs_free_chunks_real) + ffg->iff_ffs.ffs_avg = (ffg->iff_ffs.ffs_avg / + ffg->iff_ffs.ffs_free_chunks_real); +bail: + if (unlock) + ocfs2_inode_unlock(gb_inode, 0); + + if (gb_inode) + mutex_unlock(&gb_inode->i_mutex); + + if (gb_inode) + iput(gb_inode); + + brelse(bh); + + return status; +} + +int ocfs2_info_handle_freefrag(struct inode *inode, + struct ocfs2_info_request __user *req) +{ + u64 blkno = -1; + char namebuf[40]; + int status = -EFAULT, type = GLOBAL_BITMAP_SYSTEM_INODE; + + struct ocfs2_info_freefrag *oiff; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct inode *gb_inode = NULL; + + oiff = kzalloc(sizeof(struct ocfs2_info_freefrag), GFP_KERNEL); + if (!oiff) { + status = -ENOMEM; + mlog_errno(status); + goto bail; + } + + if (o2info_from_user(*oiff, req)) + goto bail; + /* + * chunksize from userspace should be power of 2. + */ + if ((oiff->iff_chunksize & (oiff->iff_chunksize - 1)) || + (!oiff->iff_chunksize)) { + status = -EINVAL; + goto bail; + } + + if (o2info_coherent(&oiff->iff_req)) { + gb_inode = ocfs2_get_system_file_inode(osb, type, + OCFS2_INVALID_SLOT); + if (!gb_inode) { + mlog(ML_ERROR, "unable to get global_bitmap inode\n"); + status = -EIO; + goto bail; + } + } else { + ocfs2_sprintf_system_inode_name(namebuf, sizeof(namebuf), type, + OCFS2_INVALID_SLOT); + status = ocfs2_lookup_ino_from_name(osb->sys_root_inode, + namebuf, + strlen(namebuf), + &blkno); + if (status < 0) { + status = -ENOENT; + goto bail; + } + } + + status = ocfs2_info_freefrag_scan_bitmap(osb, gb_inode, blkno, oiff); + if (status < 0) + goto bail; + + o2info_set_request_filled(&oiff->iff_req); + + if (o2info_to_user(*oiff, req)) + goto bail; + + status = 0; +bail: + if (status) + o2info_set_request_error(&oiff->iff_req, req); + + kfree(oiff); return status; } @@ -327,7 +729,7 @@ int ocfs2_info_handle_unknown(struct inode *inode, if (o2info_from_user(oir, req)) goto bail; - o2info_clear_request_filled(oir); + o2info_clear_request_filled(&oir); if (o2info_to_user(oir, req)) goto bail; @@ -335,7 +737,7 @@ int ocfs2_info_handle_unknown(struct inode *inode, status = 0; bail: if (status) - o2info_set_request_error(oir, req); + o2info_set_request_error(&oir, req); return status; } @@ -389,6 +791,14 @@ int ocfs2_info_handle_request(struct inode *inode, if (oir.ir_size == sizeof(struct ocfs2_info_journal_size)) status = ocfs2_info_handle_journal_size(inode, req); break; + case OCFS2_INFO_FREEINODE: + if (oir.ir_size == sizeof(struct ocfs2_info_freeinode)) + status = ocfs2_info_handle_freeinode(inode, req); + break; + case OCFS2_INFO_FREEFRAG: + if (oir.ir_size == sizeof(struct ocfs2_info_freefrag)) + status = ocfs2_info_handle_freefrag(inode, req); + break; default: status = ocfs2_info_handle_unknown(inode, req); break; @@ -542,6 +952,31 @@ long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return -EFAULT; return ocfs2_info_handle(inode, &info, 0); + case FITRIM: + { + struct super_block *sb = inode->i_sb; + struct fstrim_range range; + int ret = 0; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(&range, (struct fstrim_range *)arg, + sizeof(range))) + return -EFAULT; + + ret = ocfs2_trim_fs(sb, &range); + if (ret < 0) + return ret; + + if (copy_to_user((struct fstrim_range *)arg, &range, + sizeof(range))) + return -EFAULT; + + return 0; + } + case OCFS2_IOC_MOVE_EXT: + return ocfs2_ioctl_move_extents(filp, (void __user *)arg); default: return -ENOTTY; } @@ -569,6 +1004,7 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg) case OCFS2_IOC_GROUP_EXTEND: case OCFS2_IOC_GROUP_ADD: case OCFS2_IOC_GROUP_ADD64: + case FITRIM: break; case OCFS2_IOC_REFLINK: if (copy_from_user(&args, (struct reflink_arguments *)arg, @@ -584,6 +1020,8 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg) return -EFAULT; return ocfs2_info_handle(inode, &info, 1); + case OCFS2_IOC_MOVE_EXT: + break; default: return -ENOIOCTLCMD; } diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c new file mode 100644 index 0000000..cd94270 --- /dev/null +++ b/fs/ocfs2/move_extents.c @@ -0,0 +1,1152 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * move_extents.c + * + * Copyright (C) 2011 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + */ +#include <linux/fs.h> +#include <linux/types.h> +#include <linux/mount.h> +#include <linux/swap.h> + +#include <cluster/masklog.h> + +#include "ocfs2.h" +#include "ocfs2_ioctl.h" + +#include "alloc.h" +#include "aops.h" +#include "dlmglue.h" +#include "extent_map.h" +#include "inode.h" +#include "journal.h" +#include "suballoc.h" +#include "uptodate.h" +#include "super.h" +#include "dir.h" +#include "buffer_head_io.h" +#include "sysfile.h" +#include "suballoc.h" +#include "refcounttree.h" +#include "move_extents.h" + +struct ocfs2_move_extents_context { + struct inode *inode; + struct file *file; + int auto_defrag; + int partial; + int credits; + u32 new_phys_cpos; + u32 clusters_moved; + u64 refcount_loc; + struct ocfs2_move_extents *range; + struct ocfs2_extent_tree et; + struct ocfs2_alloc_context *meta_ac; + struct ocfs2_alloc_context *data_ac; + struct ocfs2_cached_dealloc_ctxt dealloc; +}; + +static int __ocfs2_move_extent(handle_t *handle, + struct ocfs2_move_extents_context *context, + u32 cpos, u32 len, u32 p_cpos, u32 new_p_cpos, + int ext_flags) +{ + int ret = 0, index; + struct inode *inode = context->inode; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct ocfs2_extent_rec *rec, replace_rec; + struct ocfs2_path *path = NULL; + struct ocfs2_extent_list *el; + u64 ino = ocfs2_metadata_cache_owner(context->et.et_ci); + u64 old_blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cpos); + + ret = ocfs2_duplicate_clusters_by_page(handle, context->file, cpos, + p_cpos, new_p_cpos, len); + if (ret) { + mlog_errno(ret); + goto out; + } + + memset(&replace_rec, 0, sizeof(replace_rec)); + replace_rec.e_cpos = cpu_to_le32(cpos); + replace_rec.e_leaf_clusters = cpu_to_le16(len); + replace_rec.e_blkno = cpu_to_le64(ocfs2_clusters_to_blocks(inode->i_sb, + new_p_cpos)); + + path = ocfs2_new_path_from_et(&context->et); + if (!path) { + ret = -ENOMEM; + mlog_errno(ret); + goto out; + } + + ret = ocfs2_find_path(INODE_CACHE(inode), path, cpos); + if (ret) { + mlog_errno(ret); + goto out; + } + + el = path_leaf_el(path); + + index = ocfs2_search_extent_list(el, cpos); + if (index == -1 || index >= le16_to_cpu(el->l_next_free_rec)) { + ocfs2_error(inode->i_sb, + "Inode %llu has an extent at cpos %u which can no " + "longer be found.\n", + (unsigned long long)ino, cpos); + ret = -EROFS; + goto out; + } + + rec = &el->l_recs[index]; + + BUG_ON(ext_flags != rec->e_flags); + /* + * after moving/defraging to new location, the extent is not going + * to be refcounted anymore. + */ + replace_rec.e_flags = ext_flags & ~OCFS2_EXT_REFCOUNTED; + + ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), + context->et.et_root_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_split_extent(handle, &context->et, path, index, + &replace_rec, context->meta_ac, + &context->dealloc); + if (ret) { + mlog_errno(ret); + goto out; + } + + ocfs2_journal_dirty(handle, context->et.et_root_bh); + + context->new_phys_cpos = new_p_cpos; + + /* + * need I to append truncate log for old clusters? + */ + if (old_blkno) { + if (ext_flags & OCFS2_EXT_REFCOUNTED) + ret = ocfs2_decrease_refcount(inode, handle, + ocfs2_blocks_to_clusters(osb->sb, + old_blkno), + len, context->meta_ac, + &context->dealloc, 1); + else + ret = ocfs2_truncate_log_append(osb, handle, + old_blkno, len); + } + +out: + return ret; +} + +/* + * lock allocators, and reserving appropriate number of bits for + * meta blocks and data clusters. + * + * in some cases, we don't need to reserve clusters, just let data_ac + * be NULL. + */ +static int ocfs2_lock_allocators_move_extents(struct inode *inode, + struct ocfs2_extent_tree *et, + u32 clusters_to_move, + u32 extents_to_split, + struct ocfs2_alloc_context **meta_ac, + struct ocfs2_alloc_context **data_ac, + int extra_blocks, + int *credits) +{ + int ret, num_free_extents; + unsigned int max_recs_needed = 2 * extents_to_split + clusters_to_move; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + num_free_extents = ocfs2_num_free_extents(osb, et); + if (num_free_extents < 0) { + ret = num_free_extents; + mlog_errno(ret); + goto out; + } + + if (!num_free_extents || + (ocfs2_sparse_alloc(osb) && num_free_extents < max_recs_needed)) + extra_blocks += ocfs2_extend_meta_needed(et->et_root_el); + + ret = ocfs2_reserve_new_metadata_blocks(osb, extra_blocks, meta_ac); + if (ret) { + mlog_errno(ret); + goto out; + } + + if (data_ac) { + ret = ocfs2_reserve_clusters(osb, clusters_to_move, data_ac); + if (ret) { + mlog_errno(ret); + goto out; + } + } + + *credits += ocfs2_calc_extend_credits(osb->sb, et->et_root_el, + clusters_to_move + 2); + + mlog(0, "reserve metadata_blocks: %d, data_clusters: %u, credits: %d\n", + extra_blocks, clusters_to_move, *credits); +out: + if (ret) { + if (*meta_ac) { + ocfs2_free_alloc_context(*meta_ac); + *meta_ac = NULL; + } + } + + return ret; +} + +/* + * Using one journal handle to guarantee the data consistency in case + * crash happens anywhere. + * + * XXX: defrag can end up with finishing partial extent as requested, + * due to not enough contiguous clusters can be found in allocator. + */ +static int ocfs2_defrag_extent(struct ocfs2_move_extents_context *context, + u32 cpos, u32 phys_cpos, u32 *len, int ext_flags) +{ + int ret, credits = 0, extra_blocks = 0, partial = context->partial; + handle_t *handle; + struct inode *inode = context->inode; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct inode *tl_inode = osb->osb_tl_inode; + struct ocfs2_refcount_tree *ref_tree = NULL; + u32 new_phys_cpos, new_len; + u64 phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos); + + if ((ext_flags & OCFS2_EXT_REFCOUNTED) && *len) { + + BUG_ON(!(OCFS2_I(inode)->ip_dyn_features & + OCFS2_HAS_REFCOUNT_FL)); + + BUG_ON(!context->refcount_loc); + + ret = ocfs2_lock_refcount_tree(osb, context->refcount_loc, 1, + &ref_tree, NULL); + if (ret) { + mlog_errno(ret); + return ret; + } + + ret = ocfs2_prepare_refcount_change_for_del(inode, + context->refcount_loc, + phys_blkno, + *len, + &credits, + &extra_blocks); + if (ret) { + mlog_errno(ret); + goto out; + } + } + + ret = ocfs2_lock_allocators_move_extents(inode, &context->et, *len, 1, + &context->meta_ac, + &context->data_ac, + extra_blocks, &credits); + if (ret) { + mlog_errno(ret); + goto out; + } + + /* + * should be using allocation reservation strategy there? + * + * if (context->data_ac) + * context->data_ac->ac_resv = &OCFS2_I(inode)->ip_la_data_resv; + */ + + mutex_lock(&tl_inode->i_mutex); + + if (ocfs2_truncate_log_needs_flush(osb)) { + ret = __ocfs2_flush_truncate_log(osb); + if (ret < 0) { + mlog_errno(ret); + goto out_unlock_mutex; + } + } + + handle = ocfs2_start_trans(osb, credits); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + mlog_errno(ret); + goto out_unlock_mutex; + } + + ret = __ocfs2_claim_clusters(handle, context->data_ac, 1, *len, + &new_phys_cpos, &new_len); + if (ret) { + mlog_errno(ret); + goto out_commit; + } + + /* + * allowing partial extent moving is kind of 'pros and cons', it makes + * whole defragmentation less likely to fail, on the contrary, the bad + * thing is it may make the fs even more fragmented after moving, let + * userspace make a good decision here. + */ + if (new_len != *len) { + mlog(0, "len_claimed: %u, len: %u\n", new_len, *len); + if (!partial) { + context->range->me_flags &= ~OCFS2_MOVE_EXT_FL_COMPLETE; + ret = -ENOSPC; + goto out_commit; + } + } + + mlog(0, "cpos: %u, phys_cpos: %u, new_phys_cpos: %u\n", cpos, + phys_cpos, new_phys_cpos); + + ret = __ocfs2_move_extent(handle, context, cpos, new_len, phys_cpos, + new_phys_cpos, ext_flags); + if (ret) + mlog_errno(ret); + + if (partial && (new_len != *len)) + *len = new_len; + + /* + * Here we should write the new page out first if we are + * in write-back mode. + */ + ret = ocfs2_cow_sync_writeback(inode->i_sb, context->inode, cpos, *len); + if (ret) + mlog_errno(ret); + +out_commit: + ocfs2_commit_trans(osb, handle); + +out_unlock_mutex: + mutex_unlock(&tl_inode->i_mutex); + + if (context->data_ac) { + ocfs2_free_alloc_context(context->data_ac); + context->data_ac = NULL; + } + + if (context->meta_ac) { + ocfs2_free_alloc_context(context->meta_ac); + context->meta_ac = NULL; + } + +out: + if (ref_tree) + ocfs2_unlock_refcount_tree(osb, ref_tree, 1); + + return ret; +} + +/* + * find the victim alloc group, where #blkno fits. + */ +static int ocfs2_find_victim_alloc_group(struct inode *inode, + u64 vict_blkno, + int type, int slot, + int *vict_bit, + struct buffer_head **ret_bh) +{ + int ret, i, bits_per_unit = 0; + u64 blkno; + char namebuf[40]; + + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct buffer_head *ac_bh = NULL, *gd_bh = NULL; + struct ocfs2_chain_list *cl; + struct ocfs2_chain_rec *rec; + struct ocfs2_dinode *ac_dinode; + struct ocfs2_group_desc *bg; + + ocfs2_sprintf_system_inode_name(namebuf, sizeof(namebuf), type, slot); + ret = ocfs2_lookup_ino_from_name(osb->sys_root_inode, namebuf, + strlen(namebuf), &blkno); + if (ret) { + ret = -ENOENT; + goto out; + } + + ret = ocfs2_read_blocks_sync(osb, blkno, 1, &ac_bh); + if (ret) { + mlog_errno(ret); + goto out; + } + + ac_dinode = (struct ocfs2_dinode *)ac_bh->b_data; + cl = &(ac_dinode->id2.i_chain); + rec = &(cl->cl_recs[0]); + + if (type == GLOBAL_BITMAP_SYSTEM_INODE) + bits_per_unit = osb->s_clustersize_bits - + inode->i_sb->s_blocksize_bits; + /* + * 'vict_blkno' was out of the valid range. + */ + if ((vict_blkno < le64_to_cpu(rec->c_blkno)) || + (vict_blkno >= (le32_to_cpu(ac_dinode->id1.bitmap1.i_total) << + bits_per_unit))) { + ret = -EINVAL; + goto out; + } + + for (i = 0; i < le16_to_cpu(cl->cl_next_free_rec); i++) { + + rec = &(cl->cl_recs[i]); + if (!rec) + continue; + + bg = NULL; + + do { + if (!bg) + blkno = le64_to_cpu(rec->c_blkno); + else + blkno = le64_to_cpu(bg->bg_next_group); + + if (gd_bh) { + brelse(gd_bh); + gd_bh = NULL; + } + + ret = ocfs2_read_blocks_sync(osb, blkno, 1, &gd_bh); + if (ret) { + mlog_errno(ret); + goto out; + } + + bg = (struct ocfs2_group_desc *)gd_bh->b_data; + + if (vict_blkno < (le64_to_cpu(bg->bg_blkno) + + le16_to_cpu(bg->bg_bits))) { + + *ret_bh = gd_bh; + *vict_bit = (vict_blkno - blkno) >> + bits_per_unit; + mlog(0, "find the victim group: #%llu, " + "total_bits: %u, vict_bit: %u\n", + blkno, le16_to_cpu(bg->bg_bits), + *vict_bit); + goto out; + } + + } while (le64_to_cpu(bg->bg_next_group)); + } + + ret = -EINVAL; +out: + brelse(ac_bh); + + /* + * caller has to release the gd_bh properly. + */ + return ret; +} + +/* + * XXX: helper to validate and adjust moving goal. + */ +static int ocfs2_validate_and_adjust_move_goal(struct inode *inode, + struct ocfs2_move_extents *range) +{ + int ret, goal_bit = 0; + + struct buffer_head *gd_bh = NULL; + struct ocfs2_group_desc *bg = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + int c_to_b = 1 << (osb->s_clustersize_bits - + inode->i_sb->s_blocksize_bits); + + /* + * make goal become cluster aligned. + */ + range->me_goal = ocfs2_block_to_cluster_start(inode->i_sb, + range->me_goal); + /* + * moving goal is not allowd to start with a group desc blok(#0 blk) + * let's compromise to the latter cluster. + */ + if (range->me_goal == le64_to_cpu(bg->bg_blkno)) + range->me_goal += c_to_b; + + /* + * validate goal sits within global_bitmap, and return the victim + * group desc + */ + ret = ocfs2_find_victim_alloc_group(inode, range->me_goal, + GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT, + &goal_bit, &gd_bh); + if (ret) + goto out; + + bg = (struct ocfs2_group_desc *)gd_bh->b_data; + + /* + * movement is not gonna cross two groups. + */ + if ((le16_to_cpu(bg->bg_bits) - goal_bit) * osb->s_clustersize < + range->me_len) { + ret = -EINVAL; + goto out; + } + /* + * more exact validations/adjustments will be performed later during + * moving operation for each extent range. + */ + mlog(0, "extents get ready to be moved to #%llu block\n", + range->me_goal); + +out: + brelse(gd_bh); + + return ret; +} + +static void ocfs2_probe_alloc_group(struct inode *inode, struct buffer_head *bh, + int *goal_bit, u32 move_len, u32 max_hop, + u32 *phys_cpos) +{ + int i, used, last_free_bits = 0, base_bit = *goal_bit; + struct ocfs2_group_desc *gd = (struct ocfs2_group_desc *)bh->b_data; + u32 base_cpos = ocfs2_blocks_to_clusters(inode->i_sb, + le64_to_cpu(gd->bg_blkno)); + + for (i = base_bit; i < le16_to_cpu(gd->bg_bits); i++) { + + used = ocfs2_test_bit(i, (unsigned long *)gd->bg_bitmap); + if (used) { + /* + * we even tried searching the free chunk by jumping + * a 'max_hop' distance, but still failed. + */ + if ((i - base_bit) > max_hop) { + *phys_cpos = 0; + break; + } + + if (last_free_bits) + last_free_bits = 0; + + continue; + } else + last_free_bits++; + + if (last_free_bits == move_len) { + *goal_bit = i; + *phys_cpos = base_cpos + i; + break; + } + } + + mlog(0, "found phys_cpos: %u to fit the wanted moving.\n", *phys_cpos); +} + +static int ocfs2_alloc_dinode_update_counts(struct inode *inode, + handle_t *handle, + struct buffer_head *di_bh, + u32 num_bits, + u16 chain) +{ + int ret; + u32 tmp_used; + struct ocfs2_dinode *di = (struct ocfs2_dinode *) di_bh->b_data; + struct ocfs2_chain_list *cl = + (struct ocfs2_chain_list *) &di->id2.i_chain; + + ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + tmp_used = le32_to_cpu(di->id1.bitmap1.i_used); + di->id1.bitmap1.i_used = cpu_to_le32(num_bits + tmp_used); + le32_add_cpu(&cl->cl_recs[chain].c_free, -num_bits); + ocfs2_journal_dirty(handle, di_bh); + +out: + return ret; +} + +static inline int ocfs2_block_group_set_bits(handle_t *handle, + struct inode *alloc_inode, + struct ocfs2_group_desc *bg, + struct buffer_head *group_bh, + unsigned int bit_off, + unsigned int num_bits) +{ + int status; + void *bitmap = bg->bg_bitmap; + int journal_type = OCFS2_JOURNAL_ACCESS_WRITE; + + /* All callers get the descriptor via + * ocfs2_read_group_descriptor(). Any corruption is a code bug. */ + BUG_ON(!OCFS2_IS_VALID_GROUP_DESC(bg)); + BUG_ON(le16_to_cpu(bg->bg_free_bits_count) < num_bits); + + mlog(0, "block_group_set_bits: off = %u, num = %u\n", bit_off, + num_bits); + + if (ocfs2_is_cluster_bitmap(alloc_inode)) + journal_type = OCFS2_JOURNAL_ACCESS_UNDO; + + status = ocfs2_journal_access_gd(handle, + INODE_CACHE(alloc_inode), + group_bh, + journal_type); + if (status < 0) { + mlog_errno(status); + goto bail; + } + + le16_add_cpu(&bg->bg_free_bits_count, -num_bits); + if (le16_to_cpu(bg->bg_free_bits_count) > le16_to_cpu(bg->bg_bits)) { + ocfs2_error(alloc_inode->i_sb, "Group descriptor # %llu has bit" + " count %u but claims %u are freed. num_bits %d", + (unsigned long long)le64_to_cpu(bg->bg_blkno), + le16_to_cpu(bg->bg_bits), + le16_to_cpu(bg->bg_free_bits_count), num_bits); + return -EROFS; + } + while (num_bits--) + ocfs2_set_bit(bit_off++, bitmap); + + ocfs2_journal_dirty(handle, group_bh); + +bail: + return status; +} + +static int ocfs2_move_extent(struct ocfs2_move_extents_context *context, + u32 cpos, u32 phys_cpos, u32 *new_phys_cpos, + u32 len, int ext_flags) +{ + int ret, credits = 0, extra_blocks = 0, goal_bit = 0; + handle_t *handle; + struct inode *inode = context->inode; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct inode *tl_inode = osb->osb_tl_inode; + struct inode *gb_inode = NULL; + struct buffer_head *gb_bh = NULL; + struct buffer_head *gd_bh = NULL; + struct ocfs2_group_desc *gd; + struct ocfs2_refcount_tree *ref_tree = NULL; + u32 move_max_hop = ocfs2_blocks_to_clusters(inode->i_sb, + context->range->me_threshold); + u64 phys_blkno, new_phys_blkno; + + phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, phys_cpos); + + if ((ext_flags & OCFS2_EXT_REFCOUNTED) && len) { + + BUG_ON(!(OCFS2_I(inode)->ip_dyn_features & + OCFS2_HAS_REFCOUNT_FL)); + + BUG_ON(!context->refcount_loc); + + ret = ocfs2_lock_refcount_tree(osb, context->refcount_loc, 1, + &ref_tree, NULL); + if (ret) { + mlog_errno(ret); + return ret; + } + + ret = ocfs2_prepare_refcount_change_for_del(inode, + context->refcount_loc, + phys_blkno, + len, + &credits, + &extra_blocks); + if (ret) { + mlog_errno(ret); + goto out; + } + } + + ret = ocfs2_lock_allocators_move_extents(inode, &context->et, len, 1, + &context->meta_ac, + NULL, extra_blocks, &credits); + if (ret) { + mlog_errno(ret); + goto out; + } + + /* + * need to count 2 extra credits for global_bitmap inode and + * group descriptor. + */ + credits += OCFS2_INODE_UPDATE_CREDITS + 1; + + /* + * ocfs2_move_extent() didn't reserve any clusters in lock_allocators() + * logic, while we still need to lock the global_bitmap. + */ + gb_inode = ocfs2_get_system_file_inode(osb, GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT); + if (!gb_inode) { + mlog(ML_ERROR, "unable to get global_bitmap inode\n"); + ret = -EIO; + goto out; + } + + mutex_lock(&gb_inode->i_mutex); + + ret = ocfs2_inode_lock(gb_inode, &gb_bh, 1); + if (ret) { + mlog_errno(ret); + goto out_unlock_gb_mutex; + } + + mutex_lock(&tl_inode->i_mutex); + + handle = ocfs2_start_trans(osb, credits); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + mlog_errno(ret); + goto out_unlock_tl_inode; + } + + new_phys_blkno = ocfs2_clusters_to_blocks(inode->i_sb, *new_phys_cpos); + ret = ocfs2_find_victim_alloc_group(inode, new_phys_blkno, + GLOBAL_BITMAP_SYSTEM_INODE, + OCFS2_INVALID_SLOT, + &goal_bit, &gd_bh); + if (ret) { + mlog_errno(ret); + goto out_commit; + } + + /* + * probe the victim cluster group to find a proper + * region to fit wanted movement, it even will perfrom + * a best-effort attempt by compromising to a threshold + * around the goal. + */ + ocfs2_probe_alloc_group(inode, gd_bh, &goal_bit, len, move_max_hop, + new_phys_cpos); + if (!new_phys_cpos) { + ret = -ENOSPC; + goto out_commit; + } + + ret = __ocfs2_move_extent(handle, context, cpos, len, phys_cpos, + *new_phys_cpos, ext_flags); + if (ret) { + mlog_errno(ret); + goto out_commit; + } + + gd = (struct ocfs2_group_desc *)gd_bh->b_data; + ret = ocfs2_alloc_dinode_update_counts(gb_inode, handle, gb_bh, len, + le16_to_cpu(gd->bg_chain)); + if (ret) { + mlog_errno(ret); + goto out_commit; + } + + ret = ocfs2_block_group_set_bits(handle, gb_inode, gd, gd_bh, + goal_bit, len); + if (ret) + mlog_errno(ret); + + /* + * Here we should write the new page out first if we are + * in write-back mode. + */ + ret = ocfs2_cow_sync_writeback(inode->i_sb, context->inode, cpos, len); + if (ret) + mlog_errno(ret); + +out_commit: + ocfs2_commit_trans(osb, handle); + brelse(gd_bh); + +out_unlock_tl_inode: + mutex_unlock(&tl_inode->i_mutex); + + ocfs2_inode_unlock(gb_inode, 1); +out_unlock_gb_mutex: + mutex_unlock(&gb_inode->i_mutex); + brelse(gb_bh); + iput(gb_inode); + +out: + if (context->meta_ac) { + ocfs2_free_alloc_context(context->meta_ac); + context->meta_ac = NULL; + } + + if (ref_tree) + ocfs2_unlock_refcount_tree(osb, ref_tree, 1); + + return ret; +} + +/* + * Helper to calculate the defraging length in one run according to threshold. + */ +static void ocfs2_calc_extent_defrag_len(u32 *alloc_size, u32 *len_defraged, + u32 threshold, int *skip) +{ + if ((*alloc_size + *len_defraged) < threshold) { + /* + * proceed defragmentation until we meet the thresh + */ + *len_defraged += *alloc_size; + } else if (*len_defraged == 0) { + /* + * XXX: skip a large extent. + */ + *skip = 1; + } else { + /* + * split this extent to coalesce with former pieces as + * to reach the threshold. + * + * we're done here with one cycle of defragmentation + * in a size of 'thresh', resetting 'len_defraged' + * forces a new defragmentation. + */ + *alloc_size = threshold - *len_defraged; + *len_defraged = 0; + } +} + +static int __ocfs2_move_extents_range(struct buffer_head *di_bh, + struct ocfs2_move_extents_context *context) +{ + int ret = 0, flags, do_defrag, skip = 0; + u32 cpos, phys_cpos, move_start, len_to_move, alloc_size; + u32 len_defraged = 0, defrag_thresh = 0, new_phys_cpos = 0; + + struct inode *inode = context->inode; + struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data; + struct ocfs2_move_extents *range = context->range; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if ((inode->i_size == 0) || (range->me_len == 0)) + return 0; + + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) + return 0; + + context->refcount_loc = le64_to_cpu(di->i_refcount_loc); + + ocfs2_init_dinode_extent_tree(&context->et, INODE_CACHE(inode), di_bh); + ocfs2_init_dealloc_ctxt(&context->dealloc); + + /* + * TO-DO XXX: + * + * - xattr extents. + */ + + do_defrag = context->auto_defrag; + + /* + * extents moving happens in unit of clusters, for the sake + * of simplicity, we may ignore two clusters where 'byte_start' + * and 'byte_start + len' were within. + */ + move_start = ocfs2_clusters_for_bytes(osb->sb, range->me_start); + len_to_move = (range->me_start + range->me_len) >> + osb->s_clustersize_bits; + if (len_to_move >= move_start) + len_to_move -= move_start; + else + len_to_move = 0; + + if (do_defrag) { + defrag_thresh = range->me_threshold >> osb->s_clustersize_bits; + if (defrag_thresh <= 1) + goto done; + } else + new_phys_cpos = ocfs2_blocks_to_clusters(inode->i_sb, + range->me_goal); + + mlog(0, "Inode: %llu, start: %llu, len: %llu, cstart: %u, clen: %u, " + "thresh: %u\n", + (unsigned long long)OCFS2_I(inode)->ip_blkno, + (unsigned long long)range->me_start, + (unsigned long long)range->me_len, + move_start, len_to_move, defrag_thresh); + + cpos = move_start; + while (len_to_move) { + ret = ocfs2_get_clusters(inode, cpos, &phys_cpos, &alloc_size, + &flags); + if (ret) { + mlog_errno(ret); + goto out; + } + + if (alloc_size > len_to_move) + alloc_size = len_to_move; + + /* + * XXX: how to deal with a hole: + * + * - skip the hole of course + * - force a new defragmentation + */ + if (!phys_cpos) { + if (do_defrag) + len_defraged = 0; + + goto next; + } + + if (do_defrag) { + ocfs2_calc_extent_defrag_len(&alloc_size, &len_defraged, + defrag_thresh, &skip); + /* + * skip large extents + */ + if (skip) { + skip = 0; + goto next; + } + + mlog(0, "#Defrag: cpos: %u, phys_cpos: %u, " + "alloc_size: %u, len_defraged: %u\n", + cpos, phys_cpos, alloc_size, len_defraged); + + ret = ocfs2_defrag_extent(context, cpos, phys_cpos, + &alloc_size, flags); + } else { + ret = ocfs2_move_extent(context, cpos, phys_cpos, + &new_phys_cpos, alloc_size, + flags); + + new_phys_cpos += alloc_size; + } + + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + context->clusters_moved += alloc_size; +next: + cpos += alloc_size; + len_to_move -= alloc_size; + } + +done: + range->me_flags |= OCFS2_MOVE_EXT_FL_COMPLETE; + +out: + range->me_moved_len = ocfs2_clusters_to_bytes(osb->sb, + context->clusters_moved); + range->me_new_offset = ocfs2_clusters_to_bytes(osb->sb, + context->new_phys_cpos); + + ocfs2_schedule_truncate_log_flush(osb, 1); + ocfs2_run_deallocs(osb, &context->dealloc); + + return ret; +} + +static int ocfs2_move_extents(struct ocfs2_move_extents_context *context) +{ + int status; + handle_t *handle; + struct inode *inode = context->inode; + struct ocfs2_dinode *di; + struct buffer_head *di_bh = NULL; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + + if (!inode) + return -ENOENT; + + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) + return -EROFS; + + mutex_lock(&inode->i_mutex); + + /* + * This prevents concurrent writes from other nodes + */ + status = ocfs2_rw_lock(inode, 1); + if (status) { + mlog_errno(status); + goto out; + } + + status = ocfs2_inode_lock(inode, &di_bh, 1); + if (status) { + mlog_errno(status); + goto out_rw_unlock; + } + + /* + * rememer ip_xattr_sem also needs to be held if necessary + */ + down_write(&OCFS2_I(inode)->ip_alloc_sem); + + status = __ocfs2_move_extents_range(di_bh, context); + + up_write(&OCFS2_I(inode)->ip_alloc_sem); + if (status) { + mlog_errno(status); + goto out_inode_unlock; + } + + /* + * We update ctime for these changes + */ + handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS); + if (IS_ERR(handle)) { + status = PTR_ERR(handle); + mlog_errno(status); + goto out_inode_unlock; + } + + status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (status) { + mlog_errno(status); + goto out_commit; + } + + di = (struct ocfs2_dinode *)di_bh->b_data; + inode->i_ctime = CURRENT_TIME; + di->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec); + di->i_ctime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec); + + ocfs2_journal_dirty(handle, di_bh); + +out_commit: + ocfs2_commit_trans(osb, handle); + +out_inode_unlock: + brelse(di_bh); + ocfs2_inode_unlock(inode, 1); +out_rw_unlock: + ocfs2_rw_unlock(inode, 1); +out: + mutex_unlock(&inode->i_mutex); + + return status; +} + +int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp) +{ + int status; + + struct inode *inode = filp->f_path.dentry->d_inode; + struct ocfs2_move_extents range; + struct ocfs2_move_extents_context *context = NULL; + + status = mnt_want_write(filp->f_path.mnt); + if (status) + return status; + + if ((!S_ISREG(inode->i_mode)) || !(filp->f_mode & FMODE_WRITE)) + goto out; + + if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) { + status = -EPERM; + goto out; + } + + context = kzalloc(sizeof(struct ocfs2_move_extents_context), GFP_NOFS); + if (!context) { + status = -ENOMEM; + mlog_errno(status); + goto out; + } + + context->inode = inode; + context->file = filp; + + if (argp) { + if (copy_from_user(&range, (struct ocfs2_move_extents *)argp, + sizeof(range))) { + status = -EFAULT; + goto out; + } + } else { + status = -EINVAL; + goto out; + } + + if (range.me_start > i_size_read(inode)) + goto out; + + if (range.me_start + range.me_len > i_size_read(inode)) + range.me_len = i_size_read(inode) - range.me_start; + + context->range = ⦥ + + if (range.me_flags & OCFS2_MOVE_EXT_FL_AUTO_DEFRAG) { + context->auto_defrag = 1; + /* + * ok, the default theshold for the defragmentation + * is 1M, since our maximum clustersize was 1M also. + * any thought? + */ + if (!range.me_threshold) + range.me_threshold = 1024 * 1024; + + if (range.me_threshold > i_size_read(inode)) + range.me_threshold = i_size_read(inode); + + if (range.me_flags & OCFS2_MOVE_EXT_FL_PART_DEFRAG) + context->partial = 1; + } else { + /* + * first best-effort attempt to validate and adjust the goal + * (physical address in block), while it can't guarantee later + * operation can succeed all the time since global_bitmap may + * change a bit over time. + */ + + status = ocfs2_validate_and_adjust_move_goal(inode, &range); + if (status) + goto out; + } + + status = ocfs2_move_extents(context); + if (status) + mlog_errno(status); +out: + /* + * movement/defragmentation may end up being partially completed, + * that's the reason why we need to return userspace the finished + * length and new_offset even if failure happens somewhere. + */ + if (argp) { + if (copy_to_user((struct ocfs2_move_extents *)argp, &range, + sizeof(range))) + status = -EFAULT; + } + + kfree(context); + + mnt_drop_write(filp->f_path.mnt); + + return status; +} diff --git a/fs/ocfs2/move_extents.h b/fs/ocfs2/move_extents.h new file mode 100644 index 0000000..4e143e8 --- /dev/null +++ b/fs/ocfs2/move_extents.h @@ -0,0 +1,22 @@ +/* -*- mode: c; c-basic-offset: 8; -*- + * vim: noexpandtab sw=8 ts=8 sts=0: + * + * move_extents.h + * + * Copyright (C) 2011 Oracle. All rights reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public + * License version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + */ +#ifndef OCFS2_MOVE_EXTENTS_H +#define OCFS2_MOVE_EXTENTS_H + +int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp); + +#endif /* OCFS2_MOVE_EXTENTS_H */ diff --git a/fs/ocfs2/ocfs2_ioctl.h b/fs/ocfs2/ocfs2_ioctl.h index b46f39b..5b27ff1 100644 --- a/fs/ocfs2/ocfs2_ioctl.h +++ b/fs/ocfs2/ocfs2_ioctl.h @@ -142,6 +142,38 @@ struct ocfs2_info_journal_size { __u64 ij_journal_size; }; +struct ocfs2_info_freeinode { + struct ocfs2_info_request ifi_req; + struct ocfs2_info_local_freeinode { + __u64 lfi_total; + __u64 lfi_free; + } ifi_stat[OCFS2_MAX_SLOTS]; + __u32 ifi_slotnum; /* out */ + __u32 ifi_pad; +}; + +#define OCFS2_INFO_MAX_HIST (32) + +struct ocfs2_info_freefrag { + struct ocfs2_info_request iff_req; + struct ocfs2_info_freefrag_stats { /* (out) */ + struct ocfs2_info_free_chunk_list { + __u32 fc_chunks[OCFS2_INFO_MAX_HIST]; + __u32 fc_clusters[OCFS2_INFO_MAX_HIST]; + } ffs_fc_hist; + __u32 ffs_clusters; + __u32 ffs_free_clusters; + __u32 ffs_free_chunks; + __u32 ffs_free_chunks_real; + __u32 ffs_min; /* Minimum free chunksize in clusters */ + __u32 ffs_max; + __u32 ffs_avg; + __u32 ffs_pad; + } iff_ffs; + __u32 iff_chunksize; /* chunksize in clusters(in) */ + __u32 iff_pad; +}; + /* Codes for ocfs2_info_request */ enum ocfs2_info_type { OCFS2_INFO_CLUSTERSIZE = 1, @@ -151,6 +183,8 @@ enum ocfs2_info_type { OCFS2_INFO_UUID, OCFS2_INFO_FS_FEATURES, OCFS2_INFO_JOURNAL_SIZE, + OCFS2_INFO_FREEINODE, + OCFS2_INFO_FREEFRAG, OCFS2_INFO_NUM_TYPES }; @@ -171,4 +205,38 @@ enum ocfs2_info_type { #define OCFS2_IOC_INFO _IOR('o', 5, struct ocfs2_info) +struct ocfs2_move_extents { +/* All values are in bytes */ + /* in */ + __u64 me_start; /* Virtual start in the file to move */ + __u64 me_len; /* Length of the extents to be moved */ + __u64 me_goal; /* Physical offset of the goal, + it's in block unit */ + __u64 me_threshold; /* Maximum distance from goal or threshold + for auto defragmentation */ + __u64 me_flags; /* Flags for the operation: + * - auto defragmentation. + * - refcount,xattr cases. + */ + /* out */ + __u64 me_moved_len; /* Moved/defraged length */ + __u64 me_new_offset; /* Resulting physical location */ + __u32 me_reserved[2]; /* Reserved for futhure */ +}; + +#define OCFS2_MOVE_EXT_FL_AUTO_DEFRAG (0x00000001) /* Kernel manages to + claim new clusters + as the goal place + for extents moving */ +#define OCFS2_MOVE_EXT_FL_PART_DEFRAG (0x00000002) /* Allow partial extent + moving, is to make + movement less likely + to fail, may make fs + even more fragmented */ +#define OCFS2_MOVE_EXT_FL_COMPLETE (0x00000004) /* Move or defragmenation + completely gets done. + */ + +#define OCFS2_IOC_MOVE_EXT _IOW('o', 6, struct ocfs2_move_extents) + #endif /* OCFS2_IOCTL_H */ diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h index a1dae5b..3b481f4 100644 --- a/fs/ocfs2/ocfs2_trace.h +++ b/fs/ocfs2/ocfs2_trace.h @@ -688,6 +688,31 @@ TRACE_EVENT(ocfs2_cache_block_dealloc, __entry->blkno, __entry->bit) ); +TRACE_EVENT(ocfs2_trim_extent, + TP_PROTO(struct super_block *sb, unsigned long long blk, + unsigned long long count), + TP_ARGS(sb, blk, count), + TP_STRUCT__entry( + __field(int, dev_major) + __field(int, dev_minor) + __field(unsigned long long, blk) + __field(__u64, count) + ), + TP_fast_assign( + __entry->dev_major = MAJOR(sb->s_dev); + __entry->dev_minor = MINOR(sb->s_dev); + __entry->blk = blk; + __entry->count = count; + ), + TP_printk("%d %d %llu %llu", + __entry->dev_major, __entry->dev_minor, + __entry->blk, __entry->count) +); + +DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_trim_group); + +DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_trim_fs); + /* End of trace events for fs/ocfs2/alloc.c. */ /* Trace events for fs/ocfs2/localalloc.c. */ diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 5d32749..ebfd382 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -66,7 +66,7 @@ struct ocfs2_cow_context { u32 *num_clusters, unsigned int *extent_flags); int (*cow_duplicate_clusters)(handle_t *handle, - struct ocfs2_cow_context *context, + struct file *file, u32 cpos, u32 old_cluster, u32 new_cluster, u32 new_len); }; @@ -2921,20 +2921,21 @@ static int ocfs2_clear_cow_buffer(handle_t *handle, struct buffer_head *bh) return 0; } -static int ocfs2_duplicate_clusters_by_page(handle_t *handle, - struct ocfs2_cow_context *context, - u32 cpos, u32 old_cluster, - u32 new_cluster, u32 new_len) +int ocfs2_duplicate_clusters_by_page(handle_t *handle, + struct file *file, + u32 cpos, u32 old_cluster, + u32 new_cluster, u32 new_len) { int ret = 0, partial; - struct ocfs2_caching_info *ci = context->data_et.et_ci; + struct inode *inode = file->f_path.dentry->d_inode; + struct ocfs2_caching_info *ci = INODE_CACHE(inode); struct super_block *sb = ocfs2_metadata_cache_get_super(ci); u64 new_block = ocfs2_clusters_to_blocks(sb, new_cluster); struct page *page; pgoff_t page_index; unsigned int from, to, readahead_pages; loff_t offset, end, map_end; - struct address_space *mapping = context->inode->i_mapping; + struct address_space *mapping = inode->i_mapping; trace_ocfs2_duplicate_clusters_by_page(cpos, old_cluster, new_cluster, new_len); @@ -2948,8 +2949,8 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle, * We only duplicate pages until we reach the page contains i_size - 1. * So trim 'end' to i_size. */ - if (end > i_size_read(context->inode)) - end = i_size_read(context->inode); + if (end > i_size_read(inode)) + end = i_size_read(inode); while (offset < end) { page_index = offset >> PAGE_CACHE_SHIFT; @@ -2972,10 +2973,9 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle, if (PAGE_CACHE_SIZE <= OCFS2_SB(sb)->s_clustersize) BUG_ON(PageDirty(page)); - if (PageReadahead(page) && context->file) { + if (PageReadahead(page)) { page_cache_async_readahead(mapping, - &context->file->f_ra, - context->file, + &file->f_ra, file, page, page_index, readahead_pages); } @@ -2999,8 +2999,7 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle, } } - ocfs2_map_and_dirty_page(context->inode, - handle, from, to, + ocfs2_map_and_dirty_page(inode, handle, from, to, page, 0, &new_block); mark_page_accessed(page); unlock: @@ -3015,14 +3014,15 @@ unlock: return ret; } -static int ocfs2_duplicate_clusters_by_jbd(handle_t *handle, - struct ocfs2_cow_context *context, - u32 cpos, u32 old_cluster, - u32 new_cluster, u32 new_len) +int ocfs2_duplicate_clusters_by_jbd(handle_t *handle, + struct file *file, + u32 cpos, u32 old_cluster, + u32 new_cluster, u32 new_len) { int ret = 0; - struct super_block *sb = context->inode->i_sb; - struct ocfs2_caching_info *ci = context->data_et.et_ci; + struct inode *inode = file->f_path.dentry->d_inode; + struct super_block *sb = inode->i_sb; + struct ocfs2_caching_info *ci = INODE_CACHE(inode); int i, blocks = ocfs2_clusters_to_blocks(sb, new_len); u64 old_block = ocfs2_clusters_to_blocks(sb, old_cluster); u64 new_block = ocfs2_clusters_to_blocks(sb, new_cluster); @@ -3145,8 +3145,8 @@ static int ocfs2_replace_clusters(handle_t *handle, /*If the old clusters is unwritten, no need to duplicate. */ if (!(ext_flags & OCFS2_EXT_UNWRITTEN)) { - ret = context->cow_duplicate_clusters(handle, context, cpos, - old, new, len); + ret = context->cow_duplicate_clusters(handle, context->file, + cpos, old, new, len); if (ret) { mlog_errno(ret); goto out; @@ -3162,22 +3162,22 @@ out: return ret; } -static int ocfs2_cow_sync_writeback(struct super_block *sb, - struct ocfs2_cow_context *context, - u32 cpos, u32 num_clusters) +int ocfs2_cow_sync_writeback(struct super_block *sb, + struct inode *inode, + u32 cpos, u32 num_clusters) { int ret = 0; loff_t offset, end, map_end; pgoff_t page_index; struct page *page; - if (ocfs2_should_order_data(context->inode)) + if (ocfs2_should_order_data(inode)) return 0; offset = ((loff_t)cpos) << OCFS2_SB(sb)->s_clustersize_bits; end = offset + (num_clusters << OCFS2_SB(sb)->s_clustersize_bits); - ret = filemap_fdatawrite_range(context->inode->i_mapping, + ret = filemap_fdatawrite_range(inode->i_mapping, offset, end - 1); if (ret < 0) { mlog_errno(ret); @@ -3190,7 +3190,7 @@ static int ocfs2_cow_sync_writeback(struct super_block *sb, if (map_end > end) map_end = end; - page = find_or_create_page(context->inode->i_mapping, + page = find_or_create_page(inode->i_mapping, page_index, GFP_NOFS); BUG_ON(!page); @@ -3349,7 +3349,7 @@ static int ocfs2_make_clusters_writable(struct super_block *sb, * in write-back mode. */ if (context->get_clusters == ocfs2_di_get_clusters) { - ret = ocfs2_cow_sync_writeback(sb, context, cpos, + ret = ocfs2_cow_sync_writeback(sb, context->inode, cpos, orig_num_clusters); if (ret) mlog_errno(ret); @@ -3706,7 +3706,7 @@ int ocfs2_refcount_cow_xattr(struct inode *inode, context->cow_start = cow_start; context->cow_len = cow_len; context->ref_tree = ref_tree; - context->ref_root_bh = ref_root_bh;; + context->ref_root_bh = ref_root_bh; context->cow_object = xv; context->cow_duplicate_clusters = ocfs2_duplicate_clusters_by_jbd; diff --git a/fs/ocfs2/refcounttree.h b/fs/ocfs2/refcounttree.h index c8ce46f7..7754608 100644 --- a/fs/ocfs2/refcounttree.h +++ b/fs/ocfs2/refcounttree.h @@ -84,6 +84,17 @@ int ocfs2_refcount_cow_xattr(struct inode *inode, struct buffer_head *ref_root_bh, u32 cpos, u32 write_len, struct ocfs2_post_refcount *post); +int ocfs2_duplicate_clusters_by_page(handle_t *handle, + struct file *file, + u32 cpos, u32 old_cluster, + u32 new_cluster, u32 new_len); +int ocfs2_duplicate_clusters_by_jbd(handle_t *handle, + struct file *file, + u32 cpos, u32 old_cluster, + u32 new_cluster, u32 new_len); +int ocfs2_cow_sync_writeback(struct super_block *sb, + struct inode *inode, + u32 cpos, u32 num_clusters); int ocfs2_add_refcount_flag(struct inode *inode, struct ocfs2_extent_tree *data_et, struct ocfs2_caching_info *ref_ci, diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 5a521c7..56f6102 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -41,6 +41,7 @@ #include <linux/mount.h> #include <linux/seq_file.h> #include <linux/quotaops.h> +#include <linux/cleancache.h> #define CREATE_TRACE_POINTS #include "ocfs2_trace.h" @@ -1071,7 +1072,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) sb->s_magic = OCFS2_SUPER_MAGIC; - sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + sb->s_flags = (sb->s_flags & ~(MS_POSIXACL | MS_NOSEC)) | ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); /* Hard readonly mode only if: bdev_read_only, MS_RDONLY, @@ -1566,7 +1567,7 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (osb->preferred_slot != OCFS2_INVALID_SLOT) seq_printf(s, ",preferred_slot=%d", osb->preferred_slot); - if (osb->s_atime_quantum != OCFS2_DEFAULT_ATIME_QUANTUM) + if (!(mnt->mnt_flags & MNT_NOATIME) && !(mnt->mnt_flags & MNT_RELATIME)) seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum); if (osb->osb_commit_interval) @@ -2352,6 +2353,7 @@ static int ocfs2_initialize_super(struct super_block *sb, mlog_errno(status); goto bail; } + cleancache_init_shared_fs((char *)&uuid_net_key, sb); bail: return status; diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index de4ff29..3b8d397 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -240,7 +240,9 @@ static int omfs_remove(struct inode *dir, struct dentry *dentry) struct inode *inode = dentry->d_inode; int ret; - if (S_ISDIR(inode->i_mode) && !omfs_dir_is_empty(inode)) + + if (S_ISDIR(inode->i_mode) && + !omfs_dir_is_empty(inode)) return -ENOTEMPTY; ret = omfs_delete_entry(dentry); diff --git a/fs/omfs/file.c b/fs/omfs/file.c index d738a7e..2c6d952 100644 --- a/fs/omfs/file.c +++ b/fs/omfs/file.c @@ -4,7 +4,6 @@ * Released under GPL v2. */ -#include <linux/version.h> #include <linux/module.h> #include <linux/fs.h> #include <linux/buffer_head.h> diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c index 19d6750..6296b40 100644 --- a/fs/partitions/efi.c +++ b/fs/partitions/efi.c @@ -310,6 +310,15 @@ static int is_gpt_valid(struct parsed_partitions *state, u64 lba, goto fail; } + /* Check the GUID Partition Table header size */ + if (le32_to_cpu((*gpt)->header_size) > + bdev_logical_block_size(state->bdev)) { + pr_debug("GUID Partition Table Header size is wrong: %u > %u\n", + le32_to_cpu((*gpt)->header_size), + bdev_logical_block_size(state->bdev)); + goto fail; + } + /* Check the GUID Partition Table CRC */ origcrc = le32_to_cpu((*gpt)->header_crc32); (*gpt)->header_crc32 = 0; diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c index ce4f624..af9fdf0 100644 --- a/fs/partitions/ldm.c +++ b/fs/partitions/ldm.c @@ -565,7 +565,7 @@ static bool ldm_validate_partition_table(struct parsed_partitions *state) data = read_part_sector(state, 0, §); if (!data) { - ldm_crit ("Disk read failed."); + ldm_info ("Disk read failed."); return false; } @@ -1335,6 +1335,11 @@ static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags) list_add_tail (&f->list, frags); found: + if (rec >= f->num) { + ldm_error("REC value (%d) exceeds NUM value (%d)", rec, f->num); + return false; + } + if (f->map & (1 << rec)) { ldm_error ("Duplicate VBLK, part %d.", rec); f->map &= 0x7F; /* Mark the group as broken */ diff --git a/fs/proc/Makefile b/fs/proc/Makefile index df434c5..c1c7293 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -20,6 +20,7 @@ proc-y += stat.o proc-y += uptime.o proc-y += version.o proc-y += softirqs.o +proc-y += namespaces.o proc-$(CONFIG_PROC_SYSCTL) += proc_sysctl.o proc-$(CONFIG_NET) += proc_net.o proc-$(CONFIG_PROC_KCORE) += kcore.o diff --git a/fs/proc/array.c b/fs/proc/array.c index 5e4f776..9b45ee8 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -131,7 +131,7 @@ static inline void task_name(struct seq_file *m, struct task_struct *p) * you can test for combinations of others with * simple bit tests. */ -static const char *task_state_array[] = { +static const char * const task_state_array[] = { "R (running)", /* 0 */ "S (sleeping)", /* 1 */ "D (disk sleep)", /* 2 */ @@ -147,7 +147,7 @@ static const char *task_state_array[] = { static inline const char *get_task_state(struct task_struct *tsk) { unsigned int state = (tsk->state & TASK_REPORT) | tsk->exit_state; - const char **p = &task_state_array[0]; + const char * const *p = &task_state_array[0]; BUILD_BUG_ON(1 + ilog2(TASK_STATE_MAX) != ARRAY_SIZE(task_state_array)); diff --git a/fs/proc/base.c b/fs/proc/base.c index dfa5327..8a84210 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -83,6 +83,9 @@ #include <linux/pid_namespace.h> #include <linux/fs_struct.h> #include <linux/slab.h> +#ifdef CONFIG_HARDWALL +#include <asm/hardwall.h> +#endif #include "internal.h" /* NOTE: @@ -600,7 +603,7 @@ static int proc_fd_access_allowed(struct inode *inode) return allowed; } -static int proc_setattr(struct dentry *dentry, struct iattr *attr) +int proc_setattr(struct dentry *dentry, struct iattr *attr) { int error; struct inode *inode = dentry->d_inode; @@ -894,20 +897,20 @@ static ssize_t mem_write(struct file * file, const char __user *buf, if (!task) goto out_no_task; + copied = -ENOMEM; + page = (char *)__get_free_page(GFP_TEMPORARY); + if (!page) + goto out_task; + mm = check_mem_permission(task); copied = PTR_ERR(mm); if (IS_ERR(mm)) - goto out_task; + goto out_free; copied = -EIO; if (file->private_data != (void *)((long)current->self_exec_id)) goto out_mm; - copied = -ENOMEM; - page = (char *)__get_free_page(GFP_TEMPORARY); - if (!page) - goto out_mm; - copied = 0; while (count > 0) { int this_len, retval; @@ -929,9 +932,11 @@ static ssize_t mem_write(struct file * file, const char __user *buf, count -= retval; } *ppos = dst; - free_page((unsigned long) page); + out_mm: mmput(mm); +out_free: + free_page((unsigned long) page); out_task: put_task_struct(task); out_no_task: @@ -1059,7 +1064,7 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, { struct task_struct *task; char buffer[PROC_NUMBUF]; - long oom_adjust; + int oom_adjust; unsigned long flags; int err; @@ -1071,7 +1076,7 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, goto out; } - err = strict_strtol(strstrip(buffer), 0, &oom_adjust); + err = kstrtoint(strstrip(buffer), 0, &oom_adjust); if (err) goto out; if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) && @@ -1168,7 +1173,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, struct task_struct *task; char buffer[PROC_NUMBUF]; unsigned long flags; - long oom_score_adj; + int oom_score_adj; int err; memset(buffer, 0, sizeof(buffer)); @@ -1179,7 +1184,7 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, goto out; } - err = strict_strtol(strstrip(buffer), 0, &oom_score_adj); + err = kstrtoint(strstrip(buffer), 0, &oom_score_adj); if (err) goto out; if (oom_score_adj < OOM_SCORE_ADJ_MIN || @@ -1468,7 +1473,7 @@ sched_autogroup_write(struct file *file, const char __user *buf, struct inode *inode = file->f_path.dentry->d_inode; struct task_struct *p; char buffer[PROC_NUMBUF]; - long nice; + int nice; int err; memset(buffer, 0, sizeof(buffer)); @@ -1477,9 +1482,9 @@ sched_autogroup_write(struct file *file, const char __user *buf, if (copy_from_user(buffer, buf, count)) return -EFAULT; - err = strict_strtol(strstrip(buffer), 0, &nice); - if (err) - return -EINVAL; + err = kstrtoint(strstrip(buffer), 0, &nice); + if (err < 0) + return err; p = get_proc_task(inode); if (!p) @@ -1576,57 +1581,6 @@ static const struct file_operations proc_pid_set_comm_operations = { .release = single_release, }; -/* - * We added or removed a vma mapping the executable. The vmas are only mapped - * during exec and are not mapped with the mmap system call. - * Callers must hold down_write() on the mm's mmap_sem for these - */ -void added_exe_file_vma(struct mm_struct *mm) -{ - mm->num_exe_file_vmas++; -} - -void removed_exe_file_vma(struct mm_struct *mm) -{ - mm->num_exe_file_vmas--; - if ((mm->num_exe_file_vmas == 0) && mm->exe_file){ - fput(mm->exe_file); - mm->exe_file = NULL; - } - -} - -void set_mm_exe_file(struct mm_struct *mm, struct file *new_exe_file) -{ - if (new_exe_file) - get_file(new_exe_file); - if (mm->exe_file) - fput(mm->exe_file); - mm->exe_file = new_exe_file; - mm->num_exe_file_vmas = 0; -} - -struct file *get_mm_exe_file(struct mm_struct *mm) -{ - struct file *exe_file; - - /* We need mmap_sem to protect against races with removal of - * VM_EXECUTABLE vmas */ - down_read(&mm->mmap_sem); - exe_file = mm->exe_file; - if (exe_file) - get_file(exe_file); - up_read(&mm->mmap_sem); - return exe_file; -} - -void dup_mm_exe_file(struct mm_struct *oldmm, struct mm_struct *newmm) -{ - /* It's safe to write the exe_file pointer without exe_file_lock because - * this is called during fork when the task is not yet in /proc */ - newmm->exe_file = get_mm_exe_file(oldmm); -} - static int proc_exe_link(struct inode *inode, struct path *exe_path) { struct task_struct *task; @@ -1736,8 +1690,7 @@ static int task_dumpable(struct task_struct *task) return 0; } - -static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task) +struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task) { struct inode * inode; struct proc_inode *ei; @@ -1779,7 +1732,7 @@ out_unlock: return NULL; } -static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode = dentry->d_inode; struct task_struct *task; @@ -1820,7 +1773,7 @@ static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat * made this apply to all per process world readable and executable * directories. */ -static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) +int pid_revalidate(struct dentry *dentry, struct nameidata *nd) { struct inode *inode; struct task_struct *task; @@ -1862,7 +1815,7 @@ static int pid_delete_dentry(const struct dentry * dentry) return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; } -static const struct dentry_operations pid_dentry_operations = +const struct dentry_operations pid_dentry_operations = { .d_revalidate = pid_revalidate, .d_delete = pid_delete_dentry, @@ -1870,9 +1823,6 @@ static const struct dentry_operations pid_dentry_operations = /* Lookups */ -typedef struct dentry *instantiate_t(struct inode *, struct dentry *, - struct task_struct *, const void *); - /* * Fill a directory entry. * @@ -1885,8 +1835,8 @@ typedef struct dentry *instantiate_t(struct inode *, struct dentry *, * reported by readdir in sync with the inode numbers reported * by stat. */ -static int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, - char *name, int len, +int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, + const char *name, int len, instantiate_t instantiate, struct task_struct *task, const void *ptr) { struct dentry *child, *dir = filp->f_path.dentry; @@ -2219,11 +2169,7 @@ static const struct file_operations proc_fd_operations = { */ static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags) { - int rv; - - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - rv = generic_permission(inode, mask, flags, NULL); + int rv = generic_permission(inode, mask, flags, NULL); if (rv == 0) return 0; if (task_pid(current) == proc_pid(inode)) @@ -2820,6 +2766,7 @@ static const struct pid_entry tgid_base_stuff[] = { DIR("task", S_IRUGO|S_IXUGO, proc_task_inode_operations, proc_task_operations), DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), + DIR("ns", S_IRUSR|S_IXUGO, proc_ns_dir_inode_operations, proc_ns_dir_operations), #ifdef CONFIG_NET DIR("net", S_IRUGO|S_IXUGO, proc_net_inode_operations, proc_net_operations), #endif @@ -2894,6 +2841,9 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_TASK_IO_ACCOUNTING INF("io", S_IRUGO, proc_tgid_io_accounting), #endif +#ifdef CONFIG_HARDWALL + INF("hardwall", S_IRUGO, proc_pid_hardwall), +#endif }; static int proc_tgid_base_readdir(struct file * filp, @@ -3168,6 +3118,7 @@ out_no_task: static const struct pid_entry tid_base_stuff[] = { DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), + DIR("ns", S_IRUSR|S_IXUGO, proc_ns_dir_inode_operations, proc_ns_dir_operations), REG("environ", S_IRUSR, proc_environ_operations), INF("auxv", S_IRUSR, proc_pid_auxv), ONE("status", S_IRUGO, proc_pid_status), @@ -3232,6 +3183,9 @@ static const struct pid_entry tid_base_stuff[] = { #ifdef CONFIG_TASK_IO_ACCOUNTING INF("io", S_IRUGO, proc_tid_io_accounting), #endif +#ifdef CONFIG_HARDWALL + INF("hardwall", S_IRUGO, proc_pid_hardwall), +#endif }; static int proc_tid_base_readdir(struct file * filp, diff --git a/fs/proc/generic.c b/fs/proc/generic.c index f1281339..f1637f1 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -674,6 +674,7 @@ struct proc_dir_entry *proc_mkdir_mode(const char *name, mode_t mode, } return ent; } +EXPORT_SYMBOL(proc_mkdir_mode); struct proc_dir_entry *proc_net_mkdir(struct net *net, const char *name, struct proc_dir_entry *parent) diff --git a/fs/proc/inode.c b/fs/proc/inode.c index d15aa1b..74b48cf 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -28,6 +28,7 @@ static void proc_evict_inode(struct inode *inode) { struct proc_dir_entry *de; struct ctl_table_header *head; + const struct proc_ns_operations *ns_ops; truncate_inode_pages(&inode->i_data, 0); end_writeback(inode); @@ -44,6 +45,10 @@ static void proc_evict_inode(struct inode *inode) rcu_assign_pointer(PROC_I(inode)->sysctl, NULL); sysctl_head_put(head); } + /* Release any associated namespace */ + ns_ops = PROC_I(inode)->ns_ops; + if (ns_ops && ns_ops->put) + ns_ops->put(PROC_I(inode)->ns); } static struct kmem_cache * proc_inode_cachep; @@ -62,6 +67,8 @@ static struct inode *proc_alloc_inode(struct super_block *sb) ei->pde = NULL; ei->sysctl = NULL; ei->sysctl_entry = NULL; + ei->ns = NULL; + ei->ns_ops = NULL; inode = &ei->vfs_inode; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; return inode; diff --git a/fs/proc/internal.h b/fs/proc/internal.h index c03e8d3..7838e5c 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -61,6 +61,14 @@ extern const struct file_operations proc_pagemap_operations; extern const struct file_operations proc_net_operations; extern const struct inode_operations proc_net_inode_operations; +struct proc_maps_private { + struct pid *pid; + struct task_struct *task; +#ifdef CONFIG_MMU + struct vm_area_struct *tail_vma; +#endif +}; + void proc_init_inodecache(void); static inline struct pid *proc_pid(struct inode *inode) @@ -119,3 +127,21 @@ struct inode *proc_get_inode(struct super_block *, struct proc_dir_entry *); */ int proc_readdir(struct file *, void *, filldir_t); struct dentry *proc_lookup(struct inode *, struct dentry *, struct nameidata *); + + + +/* Lookups */ +typedef struct dentry *instantiate_t(struct inode *, struct dentry *, + struct task_struct *, const void *); +int proc_fill_cache(struct file *filp, void *dirent, filldir_t filldir, + const char *name, int len, + instantiate_t instantiate, struct task_struct *task, const void *ptr); +int pid_revalidate(struct dentry *dentry, struct nameidata *nd); +struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task); +extern const struct dentry_operations pid_dentry_operations; +int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); +int proc_setattr(struct dentry *dentry, struct iattr *attr); + +extern const struct inode_operations proc_ns_dir_inode_operations; +extern const struct file_operations proc_ns_dir_operations; + diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c new file mode 100644 index 0000000..be177f7 --- /dev/null +++ b/fs/proc/namespaces.c @@ -0,0 +1,201 @@ +#include <linux/proc_fs.h> +#include <linux/nsproxy.h> +#include <linux/sched.h> +#include <linux/ptrace.h> +#include <linux/fs_struct.h> +#include <linux/mount.h> +#include <linux/path.h> +#include <linux/namei.h> +#include <linux/file.h> +#include <linux/utsname.h> +#include <net/net_namespace.h> +#include <linux/mnt_namespace.h> +#include <linux/ipc_namespace.h> +#include <linux/pid_namespace.h> +#include "internal.h" + + +static const struct proc_ns_operations *ns_entries[] = { +#ifdef CONFIG_NET_NS + &netns_operations, +#endif +#ifdef CONFIG_UTS_NS + &utsns_operations, +#endif +#ifdef CONFIG_IPC_NS + &ipcns_operations, +#endif +}; + +static const struct file_operations ns_file_operations = { + .llseek = no_llseek, +}; + +static struct dentry *proc_ns_instantiate(struct inode *dir, + struct dentry *dentry, struct task_struct *task, const void *ptr) +{ + const struct proc_ns_operations *ns_ops = ptr; + struct inode *inode; + struct proc_inode *ei; + struct dentry *error = ERR_PTR(-ENOENT); + void *ns; + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + goto out; + + ns = ns_ops->get(task); + if (!ns) + goto out_iput; + + ei = PROC_I(inode); + inode->i_mode = S_IFREG|S_IRUSR; + inode->i_fop = &ns_file_operations; + ei->ns_ops = ns_ops; + ei->ns = ns; + + dentry->d_op = &pid_dentry_operations; + d_add(dentry, inode); + /* Close the race of the process dying before we return the dentry */ + if (pid_revalidate(dentry, NULL)) + error = NULL; +out: + return error; +out_iput: + iput(inode); + goto out; +} + +static int proc_ns_fill_cache(struct file *filp, void *dirent, + filldir_t filldir, struct task_struct *task, + const struct proc_ns_operations *ops) +{ + return proc_fill_cache(filp, dirent, filldir, + ops->name, strlen(ops->name), + proc_ns_instantiate, task, ops); +} + +static int proc_ns_dir_readdir(struct file *filp, void *dirent, + filldir_t filldir) +{ + int i; + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct task_struct *task = get_proc_task(inode); + const struct proc_ns_operations **entry, **last; + ino_t ino; + int ret; + + ret = -ENOENT; + if (!task) + goto out_no_task; + + ret = -EPERM; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out; + + ret = 0; + i = filp->f_pos; + switch (i) { + case 0: + ino = inode->i_ino; + if (filldir(dirent, ".", 1, i, ino, DT_DIR) < 0) + goto out; + i++; + filp->f_pos++; + /* fall through */ + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, i, ino, DT_DIR) < 0) + goto out; + i++; + filp->f_pos++; + /* fall through */ + default: + i -= 2; + if (i >= ARRAY_SIZE(ns_entries)) { + ret = 1; + goto out; + } + entry = ns_entries + i; + last = &ns_entries[ARRAY_SIZE(ns_entries) - 1]; + while (entry <= last) { + if (proc_ns_fill_cache(filp, dirent, filldir, + task, *entry) < 0) + goto out; + filp->f_pos++; + entry++; + } + } + + ret = 1; +out: + put_task_struct(task); +out_no_task: + return ret; +} + +const struct file_operations proc_ns_dir_operations = { + .read = generic_read_dir, + .readdir = proc_ns_dir_readdir, +}; + +static struct dentry *proc_ns_dir_lookup(struct inode *dir, + struct dentry *dentry, struct nameidata *nd) +{ + struct dentry *error; + struct task_struct *task = get_proc_task(dir); + const struct proc_ns_operations **entry, **last; + unsigned int len = dentry->d_name.len; + + error = ERR_PTR(-ENOENT); + + if (!task) + goto out_no_task; + + error = ERR_PTR(-EPERM); + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + goto out; + + last = &ns_entries[ARRAY_SIZE(ns_entries) - 1]; + for (entry = ns_entries; entry <= last; entry++) { + if (strlen((*entry)->name) != len) + continue; + if (!memcmp(dentry->d_name.name, (*entry)->name, len)) + break; + } + error = ERR_PTR(-ENOENT); + if (entry > last) + goto out; + + error = proc_ns_instantiate(dir, dentry, task, *entry); +out: + put_task_struct(task); +out_no_task: + return error; +} + +const struct inode_operations proc_ns_dir_inode_operations = { + .lookup = proc_ns_dir_lookup, + .getattr = pid_getattr, + .setattr = proc_setattr, +}; + +struct file *proc_ns_fget(int fd) +{ + struct file *file; + + file = fget(fd); + if (!file) + return ERR_PTR(-EBADF); + + if (file->f_op != &ns_file_operations) + goto out_invalid; + + return file; + +out_invalid: + fput(file); + return ERR_PTR(-EINVAL); +} + diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index f50133c..d167de3 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -304,9 +304,6 @@ static int proc_sys_permission(struct inode *inode, int mask,unsigned int flags) struct ctl_table *table; int error; - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - /* Executable files are not allowed under /proc/sys/ */ if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) return -EACCES; diff --git a/fs/proc/root.c b/fs/proc/root.c index a9000e9..d6c3b41 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -28,11 +28,12 @@ static int proc_test_super(struct super_block *sb, void *data) static int proc_set_super(struct super_block *sb, void *data) { - struct pid_namespace *ns; - - ns = (struct pid_namespace *)data; - sb->s_fs_info = get_pid_ns(ns); - return set_anon_super(sb, NULL); + int err = set_anon_super(sb, NULL); + if (!err) { + struct pid_namespace *ns = (struct pid_namespace *)data; + sb->s_fs_info = get_pid_ns(ns); + } + return err; } static struct dentry *proc_mount(struct file_system_type *fs_type, diff --git a/fs/proc/stat.c b/fs/proc/stat.c index 1cffa2b..9758b65 100644 --- a/fs/proc/stat.c +++ b/fs/proc/stat.c @@ -138,9 +138,9 @@ static int stat_open(struct inode *inode, struct file *file) struct seq_file *m; int res; - /* don't ask for more than the kmalloc() max size, currently 128 KB */ - if (size > 128 * 1024) - size = 128 * 1024; + /* don't ask for more than the kmalloc() max size */ + if (size > KMALLOC_MAX_SIZE) + size = KMALLOC_MAX_SIZE; buf = kmalloc(size, GFP_KERNEL); if (!buf) return -ENOMEM; diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 318d865..25b6a88 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -211,7 +211,7 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) { struct mm_struct *mm = vma->vm_mm; struct file *file = vma->vm_file; - int flags = vma->vm_flags; + vm_flags_t flags = vma->vm_flags; unsigned long ino = 0; unsigned long long pgoff = 0; unsigned long start, end; @@ -536,15 +536,17 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, char buffer[PROC_NUMBUF]; struct mm_struct *mm; struct vm_area_struct *vma; - long type; + int type; + int rv; memset(buffer, 0, sizeof(buffer)); if (count > sizeof(buffer) - 1) count = sizeof(buffer) - 1; if (copy_from_user(buffer, buf, count)) return -EFAULT; - if (strict_strtol(strstrip(buffer), 10, &type)) - return -EINVAL; + rv = kstrtoint(strstrip(buffer), 10, &type); + if (rv < 0) + return rv; if (type < CLEAR_REFS_ALL || type > CLEAR_REFS_MAPPED) return -EINVAL; task = get_proc_task(file->f_path.dentry->d_inode); @@ -769,18 +771,12 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, if (!task) goto out; - mm = mm_for_maps(task); - ret = PTR_ERR(mm); - if (!mm || IS_ERR(mm)) - goto out_task; - ret = -EINVAL; /* file position must be aligned */ if ((*ppos % PM_ENTRY_BYTES) || (count % PM_ENTRY_BYTES)) goto out_task; ret = 0; - if (!count) goto out_task; @@ -788,7 +784,12 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, pm.buffer = kmalloc(pm.len, GFP_TEMPORARY); ret = -ENOMEM; if (!pm.buffer) - goto out_mm; + goto out_task; + + mm = mm_for_maps(task); + ret = PTR_ERR(mm); + if (!mm || IS_ERR(mm)) + goto out_free; pagemap_walk.pmd_entry = pagemap_pte_range; pagemap_walk.pte_hole = pagemap_pte_hole; @@ -831,7 +832,7 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, len = min(count, PM_ENTRY_BYTES * pm.pos); if (copy_to_user(buf, pm.buffer, len)) { ret = -EFAULT; - goto out_free; + goto out_mm; } copied += len; buf += len; @@ -841,10 +842,10 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, if (!ret || ret == PM_END_OF_BUFFER) ret = copied; -out_free: - kfree(pm.buffer); out_mm: mmput(mm); +out_free: + kfree(pm.buffer); out_task: put_task_struct(task); out: @@ -858,7 +859,192 @@ const struct file_operations proc_pagemap_operations = { #endif /* CONFIG_PROC_PAGE_MONITOR */ #ifdef CONFIG_NUMA -extern int show_numa_map(struct seq_file *m, void *v); + +struct numa_maps { + struct vm_area_struct *vma; + unsigned long pages; + unsigned long anon; + unsigned long active; + unsigned long writeback; + unsigned long mapcount_max; + unsigned long dirty; + unsigned long swapcache; + unsigned long node[MAX_NUMNODES]; +}; + +struct numa_maps_private { + struct proc_maps_private proc_maps; + struct numa_maps md; +}; + +static void gather_stats(struct page *page, struct numa_maps *md, int pte_dirty) +{ + int count = page_mapcount(page); + + md->pages++; + if (pte_dirty || PageDirty(page)) + md->dirty++; + + if (PageSwapCache(page)) + md->swapcache++; + + if (PageActive(page) || PageUnevictable(page)) + md->active++; + + if (PageWriteback(page)) + md->writeback++; + + if (PageAnon(page)) + md->anon++; + + if (count > md->mapcount_max) + md->mapcount_max = count; + + md->node[page_to_nid(page)]++; +} + +static int gather_pte_stats(pmd_t *pmd, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct numa_maps *md; + spinlock_t *ptl; + pte_t *orig_pte; + pte_t *pte; + + md = walk->private; + orig_pte = pte = pte_offset_map_lock(walk->mm, pmd, addr, &ptl); + do { + struct page *page; + int nid; + + if (!pte_present(*pte)) + continue; + + page = vm_normal_page(md->vma, addr, *pte); + if (!page) + continue; + + if (PageReserved(page)) + continue; + + nid = page_to_nid(page); + if (!node_isset(nid, node_states[N_HIGH_MEMORY])) + continue; + + gather_stats(page, md, pte_dirty(*pte)); + + } while (pte++, addr += PAGE_SIZE, addr != end); + pte_unmap_unlock(orig_pte, ptl); + return 0; +} +#ifdef CONFIG_HUGETLB_PAGE +static int gather_hugetbl_stats(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, struct mm_walk *walk) +{ + struct numa_maps *md; + struct page *page; + + if (pte_none(*pte)) + return 0; + + page = pte_page(*pte); + if (!page) + return 0; + + md = walk->private; + gather_stats(page, md, pte_dirty(*pte)); + return 0; +} + +#else +static int gather_hugetbl_stats(pte_t *pte, unsigned long hmask, + unsigned long addr, unsigned long end, struct mm_walk *walk) +{ + return 0; +} +#endif + +/* + * Display pages allocated per node and memory policy via /proc. + */ +static int show_numa_map(struct seq_file *m, void *v) +{ + struct numa_maps_private *numa_priv = m->private; + struct proc_maps_private *proc_priv = &numa_priv->proc_maps; + struct vm_area_struct *vma = v; + struct numa_maps *md = &numa_priv->md; + struct file *file = vma->vm_file; + struct mm_struct *mm = vma->vm_mm; + struct mm_walk walk = {}; + struct mempolicy *pol; + int n; + char buffer[50]; + + if (!mm) + return 0; + + /* Ensure we start with an empty set of numa_maps statistics. */ + memset(md, 0, sizeof(*md)); + + md->vma = vma; + + walk.hugetlb_entry = gather_hugetbl_stats; + walk.pmd_entry = gather_pte_stats; + walk.private = md; + walk.mm = mm; + + pol = get_vma_policy(proc_priv->task, vma, vma->vm_start); + mpol_to_str(buffer, sizeof(buffer), pol, 0); + mpol_cond_put(pol); + + seq_printf(m, "%08lx %s", vma->vm_start, buffer); + + if (file) { + seq_printf(m, " file="); + seq_path(m, &file->f_path, "\n\t= "); + } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { + seq_printf(m, " heap"); + } else if (vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack) { + seq_printf(m, " stack"); + } + + walk_page_range(vma->vm_start, vma->vm_end, &walk); + + if (!md->pages) + goto out; + + if (md->anon) + seq_printf(m, " anon=%lu", md->anon); + + if (md->dirty) + seq_printf(m, " dirty=%lu", md->dirty); + + if (md->pages != md->anon && md->pages != md->dirty) + seq_printf(m, " mapped=%lu", md->pages); + + if (md->mapcount_max > 1) + seq_printf(m, " mapmax=%lu", md->mapcount_max); + + if (md->swapcache) + seq_printf(m, " swapcache=%lu", md->swapcache); + + if (md->active < md->pages && !is_vm_hugetlb_page(vma)) + seq_printf(m, " active=%lu", md->active); + + if (md->writeback) + seq_printf(m, " writeback=%lu", md->writeback); + + for_each_node_state(n, N_HIGH_MEMORY) + if (md->node[n]) + seq_printf(m, " N%d=%lu", n, md->node[n]); +out: + seq_putc(m, '\n'); + + if (m->count < m->size) + m->version = (vma != proc_priv->tail_vma) ? vma->vm_start : 0; + return 0; +} static const struct seq_operations proc_pid_numa_maps_op = { .start = m_start, @@ -869,7 +1055,20 @@ static const struct seq_operations proc_pid_numa_maps_op = { static int numa_maps_open(struct inode *inode, struct file *file) { - return do_maps_open(inode, file, &proc_pid_numa_maps_op); + struct numa_maps_private *priv; + int ret = -ENOMEM; + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (priv) { + priv->proc_maps.pid = proc_pid(inode); + ret = seq_open(file, &proc_pid_numa_maps_op); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = priv; + } else { + kfree(priv); + } + } + return ret; } const struct file_operations proc_numa_maps_operations = { @@ -878,4 +1077,4 @@ const struct file_operations proc_numa_maps_operations = { .llseek = seq_lseek, .release = seq_release_private, }; -#endif +#endif /* CONFIG_NUMA */ diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 74802bc5..cd99bf5 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -35,6 +35,46 @@ static u64 vmcore_size; static struct proc_dir_entry *proc_vmcore = NULL; +/* + * Returns > 0 for RAM pages, 0 for non-RAM pages, < 0 on error + * The called function has to take care of module refcounting. + */ +static int (*oldmem_pfn_is_ram)(unsigned long pfn); + +int register_oldmem_pfn_is_ram(int (*fn)(unsigned long pfn)) +{ + if (oldmem_pfn_is_ram) + return -EBUSY; + oldmem_pfn_is_ram = fn; + return 0; +} +EXPORT_SYMBOL_GPL(register_oldmem_pfn_is_ram); + +void unregister_oldmem_pfn_is_ram(void) +{ + oldmem_pfn_is_ram = NULL; + wmb(); +} +EXPORT_SYMBOL_GPL(unregister_oldmem_pfn_is_ram); + +static int pfn_is_ram(unsigned long pfn) +{ + int (*fn)(unsigned long pfn); + /* pfn is ram unless fn() checks pagetype */ + int ret = 1; + + /* + * Ask hypervisor if the pfn is really ram. + * A ballooned page contains no data and reading from such a page + * will cause high load in the hypervisor. + */ + fn = oldmem_pfn_is_ram; + if (fn) + ret = fn(pfn); + + return ret; +} + /* Reads a page from the oldmem device from given offset. */ static ssize_t read_from_oldmem(char *buf, size_t count, u64 *ppos, int userbuf) @@ -55,9 +95,15 @@ static ssize_t read_from_oldmem(char *buf, size_t count, else nr_bytes = count; - tmp = copy_oldmem_page(pfn, buf, nr_bytes, offset, userbuf); - if (tmp < 0) - return tmp; + /* If pfn is not ram, return zeros for sparse dump files */ + if (pfn_is_ram(pfn) == 0) + memset(buf, 0, nr_bytes); + else { + tmp = copy_oldmem_page(pfn, buf, nr_bytes, + offset, userbuf); + if (tmp < 0) + return tmp; + } *ppos += nr_bytes; count -= nr_bytes; buf += nr_bytes; diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index f835a25..f2c3ff2 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -152,21 +152,27 @@ EXPORT_SYMBOL_GPL(pstore_register); void pstore_get_records(void) { struct pstore_info *psi = psinfo; - size_t size; + ssize_t size; u64 id; enum pstore_type_id type; struct timespec time; - int failed = 0; + int failed = 0, rc; if (!psi) return; mutex_lock(&psinfo->buf_mutex); + rc = psi->open(psi); + if (rc) + goto out; + while ((size = psi->read(&id, &type, &time)) > 0) { - if (pstore_mkfile(type, psi->name, id, psi->buf, size, + if (pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size, time, psi->erase)) failed++; } + psi->close(psi); +out: mutex_unlock(&psinfo->buf_mutex); if (failed) diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index d3c032f..5b572c8 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -691,8 +691,11 @@ static void prune_dqcache(int count) * This is called from kswapd when we think we need some * more memory */ -static int shrink_dqcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) +static int shrink_dqcache_memory(struct shrinker *shrink, + struct shrink_control *sc) { + int nr = sc->nr_to_scan; + if (nr) { spin_lock(&dq_list_lock); prune_dqcache(nr); diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index b216ff6..aa91089 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -568,7 +568,7 @@ static void destroy_inodecache(void) } /* we don't mark inodes dirty, we just log them */ -static void reiserfs_dirty_inode(struct inode *inode) +static void reiserfs_dirty_inode(struct inode *inode, int flags) { struct reiserfs_transaction_handle th; diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 47d2a44..d780896 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -98,14 +98,12 @@ static int xattr_rmdir(struct inode *dir, struct dentry *dentry) reiserfs_mutex_lock_nested_safe(&dentry->d_inode->i_mutex, I_MUTEX_CHILD, dir->i_sb); - dentry_unhash(dentry); error = dir->i_op->rmdir(dir, dentry); if (!error) dentry->d_inode->i_flags |= S_DEAD; mutex_unlock(&dentry->d_inode->i_mutex); if (!error) d_delete(dentry); - dput(dentry); return error; } @@ -956,8 +954,6 @@ static int xattr_mount_check(struct super_block *s) int reiserfs_permission(struct inode *inode, int mask, unsigned int flags) { - if (flags & IPERM_FLAG_RCU) - return -ECHILD; /* * We don't do permission checks on the internal objects. * Permissions are determined by the "owning" object. diff --git a/fs/splice.c b/fs/splice.c index 50a5d978..aa866d3 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -162,6 +162,14 @@ static const struct pipe_buf_operations user_page_pipe_buf_ops = { .get = generic_pipe_buf_get, }; +static void wakeup_pipe_readers(struct pipe_inode_info *pipe) +{ + smp_mb(); + if (waitqueue_active(&pipe->wait)) + wake_up_interruptible(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); +} + /** * splice_to_pipe - fill passed data into a pipe * @pipe: pipe to fill @@ -247,12 +255,8 @@ ssize_t splice_to_pipe(struct pipe_inode_info *pipe, pipe_unlock(pipe); - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&pipe->wait)) - wake_up_interruptible(&pipe->wait); - kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); - } + if (do_wakeup) + wakeup_pipe_readers(pipe); while (page_nr < spd_pages) spd->spd_release(spd, page_nr++); @@ -1892,12 +1896,9 @@ retry: /* * If we put data in the output pipe, wakeup any potential readers. */ - if (ret > 0) { - smp_mb(); - if (waitqueue_active(&opipe->wait)) - wake_up_interruptible(&opipe->wait); - kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN); - } + if (ret > 0) + wakeup_pipe_readers(opipe); + if (input_wakeup) wakeup_pipe_writers(ipipe); @@ -1976,12 +1977,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, /* * If we put data in the output pipe, wakeup any potential readers. */ - if (ret > 0) { - smp_mb(); - if (waitqueue_active(&opipe->wait)) - wake_up_interruptible(&opipe->wait); - kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN); - } + if (ret > 0) + wakeup_pipe_readers(opipe); return ret; } diff --git a/fs/squashfs/Kconfig b/fs/squashfs/Kconfig index efc309f..7797218 100644 --- a/fs/squashfs/Kconfig +++ b/fs/squashfs/Kconfig @@ -42,7 +42,7 @@ config SQUASHFS_LZO select LZO_DECOMPRESS help Saying Y here includes support for reading Squashfs file systems - compressed with LZO compresssion. LZO compression is mainly + compressed with LZO compression. LZO compression is mainly aimed at embedded systems with slower CPUs where the overheads of zlib are too high. @@ -57,7 +57,7 @@ config SQUASHFS_XZ select XZ_DEC help Saying Y here includes support for reading Squashfs file systems - compressed with XZ compresssion. XZ gives better compression than + compressed with XZ compression. XZ gives better compression than the default zlib compression, at the expense of greater CPU and memory overhead. diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c index 8ab48bc..ed0eb2a 100644 --- a/fs/squashfs/block.c +++ b/fs/squashfs/block.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index c37b520..f744be9 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -29,7 +29,7 @@ * plus functions layered ontop of the generic cache implementation to * access the metadata and fragment caches. * - * To avoid out of memory and fragmentation isssues with vmalloc the cache + * To avoid out of memory and fragmentation issues with vmalloc the cache * uses sequences of kmalloced PAGE_CACHE_SIZE buffers. * * It should be noted that the cache is not used for file datablocks, these @@ -393,19 +393,36 @@ struct squashfs_cache_entry *squashfs_get_datablock(struct super_block *sb, /* * Read a filesystem table (uncompressed sequence of bytes) from disk */ -int squashfs_read_table(struct super_block *sb, void *buffer, u64 block, - int length) +void *squashfs_read_table(struct super_block *sb, u64 block, int length) { int pages = (length + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; int i, res; - void **data = kcalloc(pages, sizeof(void *), GFP_KERNEL); - if (data == NULL) - return -ENOMEM; + void *table, *buffer, **data; + + table = buffer = kmalloc(length, GFP_KERNEL); + if (table == NULL) + return ERR_PTR(-ENOMEM); + + data = kcalloc(pages, sizeof(void *), GFP_KERNEL); + if (data == NULL) { + res = -ENOMEM; + goto failed; + } for (i = 0; i < pages; i++, buffer += PAGE_CACHE_SIZE) data[i] = buffer; + res = squashfs_read_data(sb, data, block, length | SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, length, pages); + kfree(data); - return res; + + if (res < 0) + goto failed; + + return table; + +failed: + kfree(table); + return ERR_PTR(res); } diff --git a/fs/squashfs/decompressor.c b/fs/squashfs/decompressor.c index e921bd2..9f1b0bb 100644 --- a/fs/squashfs/decompressor.c +++ b/fs/squashfs/decompressor.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/decompressor.h b/fs/squashfs/decompressor.h index 099745a..8ba70cf 100644 --- a/fs/squashfs/decompressor.h +++ b/fs/squashfs/decompressor.h @@ -4,7 +4,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/dir.c b/fs/squashfs/dir.c index 3f79cd1..9dfe2ce 100644 --- a/fs/squashfs/dir.c +++ b/fs/squashfs/dir.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/export.c b/fs/squashfs/export.c index 7f93d5a..5e1101f 100644 --- a/fs/squashfs/export.c +++ b/fs/squashfs/export.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -121,30 +121,38 @@ static struct dentry *squashfs_get_parent(struct dentry *child) * Read uncompressed inode lookup table indexes off disk into memory */ __le64 *squashfs_read_inode_lookup_table(struct super_block *sb, - u64 lookup_table_start, unsigned int inodes) + u64 lookup_table_start, u64 next_table, unsigned int inodes) { unsigned int length = SQUASHFS_LOOKUP_BLOCK_BYTES(inodes); - __le64 *inode_lookup_table; - int err; + __le64 *table; TRACE("In read_inode_lookup_table, length %d\n", length); - /* Allocate inode lookup table indexes */ - inode_lookup_table = kmalloc(length, GFP_KERNEL); - if (inode_lookup_table == NULL) { - ERROR("Failed to allocate inode lookup table\n"); - return ERR_PTR(-ENOMEM); - } + /* Sanity check values */ + + /* there should always be at least one inode */ + if (inodes == 0) + return ERR_PTR(-EINVAL); + + /* length bytes should not extend into the next table - this check + * also traps instances where lookup_table_start is incorrectly larger + * than the next table start + */ + if (lookup_table_start + length > next_table) + return ERR_PTR(-EINVAL); + + table = squashfs_read_table(sb, lookup_table_start, length); - err = squashfs_read_table(sb, inode_lookup_table, lookup_table_start, - length); - if (err < 0) { - ERROR("unable to read inode lookup table\n"); - kfree(inode_lookup_table); - return ERR_PTR(err); + /* + * table[0] points to the first inode lookup table metadata block, + * this should be less than lookup_table_start + */ + if (!IS_ERR(table) && le64_to_cpu(table[0]) >= lookup_table_start) { + kfree(table); + return ERR_PTR(-EINVAL); } - return inode_lookup_table; + return table; } diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c index a25c506..38bb1c6 100644 --- a/fs/squashfs/file.c +++ b/fs/squashfs/file.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/fragment.c b/fs/squashfs/fragment.c index 7eef571..0ed6edb 100644 --- a/fs/squashfs/fragment.c +++ b/fs/squashfs/fragment.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -71,26 +71,29 @@ int squashfs_frag_lookup(struct super_block *sb, unsigned int fragment, * Read the uncompressed fragment lookup table indexes off disk into memory */ __le64 *squashfs_read_fragment_index_table(struct super_block *sb, - u64 fragment_table_start, unsigned int fragments) + u64 fragment_table_start, u64 next_table, unsigned int fragments) { unsigned int length = SQUASHFS_FRAGMENT_INDEX_BYTES(fragments); - __le64 *fragment_index; - int err; + __le64 *table; - /* Allocate fragment lookup table indexes */ - fragment_index = kmalloc(length, GFP_KERNEL); - if (fragment_index == NULL) { - ERROR("Failed to allocate fragment index table\n"); - return ERR_PTR(-ENOMEM); - } + /* + * Sanity check, length bytes should not extend into the next table - + * this check also traps instances where fragment_table_start is + * incorrectly larger than the next table start + */ + if (fragment_table_start + length > next_table) + return ERR_PTR(-EINVAL); + + table = squashfs_read_table(sb, fragment_table_start, length); - err = squashfs_read_table(sb, fragment_index, fragment_table_start, - length); - if (err < 0) { - ERROR("unable to read fragment index table\n"); - kfree(fragment_index); - return ERR_PTR(err); + /* + * table[0] points to the first fragment table metadata block, this + * should be less than fragment_table_start + */ + if (!IS_ERR(table) && le64_to_cpu(table[0]) >= fragment_table_start) { + kfree(table); + return ERR_PTR(-EINVAL); } - return fragment_index; + return table; } diff --git a/fs/squashfs/id.c b/fs/squashfs/id.c index d8f3245..d38ea3d 100644 --- a/fs/squashfs/id.c +++ b/fs/squashfs/id.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -66,27 +66,37 @@ int squashfs_get_id(struct super_block *sb, unsigned int index, * Read uncompressed id lookup table indexes from disk into memory */ __le64 *squashfs_read_id_index_table(struct super_block *sb, - u64 id_table_start, unsigned short no_ids) + u64 id_table_start, u64 next_table, unsigned short no_ids) { unsigned int length = SQUASHFS_ID_BLOCK_BYTES(no_ids); - __le64 *id_table; - int err; + __le64 *table; TRACE("In read_id_index_table, length %d\n", length); - /* Allocate id lookup table indexes */ - id_table = kmalloc(length, GFP_KERNEL); - if (id_table == NULL) { - ERROR("Failed to allocate id index table\n"); - return ERR_PTR(-ENOMEM); - } + /* Sanity check values */ + + /* there should always be at least one id */ + if (no_ids == 0) + return ERR_PTR(-EINVAL); + + /* + * length bytes should not extend into the next table - this check + * also traps instances where id_table_start is incorrectly larger + * than the next table start + */ + if (id_table_start + length > next_table) + return ERR_PTR(-EINVAL); + + table = squashfs_read_table(sb, id_table_start, length); - err = squashfs_read_table(sb, id_table, id_table_start, length); - if (err < 0) { - ERROR("unable to read id index table\n"); - kfree(id_table); - return ERR_PTR(err); + /* + * table[0] points to the first id lookup table metadata block, this + * should be less than id_table_start + */ + if (!IS_ERR(table) && le64_to_cpu(table[0]) >= id_table_start) { + kfree(table); + return ERR_PTR(-EINVAL); } - return id_table; + return table; } diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c index 62e63ad..04bebca 100644 --- a/fs/squashfs/inode.c +++ b/fs/squashfs/inode.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/namei.c b/fs/squashfs/namei.c index 5d922a6..4bc63ac 100644 --- a/fs/squashfs/namei.c +++ b/fs/squashfs/namei.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/squashfs.h b/fs/squashfs/squashfs.h index 1f2e608..e3be6a7 100644 --- a/fs/squashfs/squashfs.h +++ b/fs/squashfs/squashfs.h @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -44,24 +44,24 @@ extern struct squashfs_cache_entry *squashfs_get_fragment(struct super_block *, u64, int); extern struct squashfs_cache_entry *squashfs_get_datablock(struct super_block *, u64, int); -extern int squashfs_read_table(struct super_block *, void *, u64, int); +extern void *squashfs_read_table(struct super_block *, u64, int); /* decompressor.c */ extern const struct squashfs_decompressor *squashfs_lookup_decompressor(int); extern void *squashfs_decompressor_init(struct super_block *, unsigned short); /* export.c */ -extern __le64 *squashfs_read_inode_lookup_table(struct super_block *, u64, +extern __le64 *squashfs_read_inode_lookup_table(struct super_block *, u64, u64, unsigned int); /* fragment.c */ extern int squashfs_frag_lookup(struct super_block *, unsigned int, u64 *); extern __le64 *squashfs_read_fragment_index_table(struct super_block *, - u64, unsigned int); + u64, u64, unsigned int); /* id.c */ extern int squashfs_get_id(struct super_block *, unsigned int, unsigned int *); -extern __le64 *squashfs_read_id_index_table(struct super_block *, u64, +extern __le64 *squashfs_read_id_index_table(struct super_block *, u64, u64, unsigned short); /* inode.c */ diff --git a/fs/squashfs/squashfs_fs.h b/fs/squashfs/squashfs_fs.h index 4582c56..b4a4e53 100644 --- a/fs/squashfs/squashfs_fs.h +++ b/fs/squashfs/squashfs_fs.h @@ -4,7 +4,7 @@ * Squashfs * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/squashfs_fs_i.h b/fs/squashfs/squashfs_fs_i.h index 359baef..73588e7 100644 --- a/fs/squashfs/squashfs_fs_i.h +++ b/fs/squashfs/squashfs_fs_i.h @@ -4,7 +4,7 @@ * Squashfs * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/squashfs_fs_sb.h b/fs/squashfs/squashfs_fs_sb.h index d9037a5..651f0b3 100644 --- a/fs/squashfs/squashfs_fs_sb.h +++ b/fs/squashfs/squashfs_fs_sb.h @@ -4,7 +4,7 @@ * Squashfs * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 5c8184c..7438850 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -83,7 +83,7 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) long long root_inode; unsigned short flags; unsigned int fragments; - u64 lookup_table_start, xattr_id_table_start; + u64 lookup_table_start, xattr_id_table_start, next_table; int err; TRACE("Entered squashfs_fill_superblock\n"); @@ -95,12 +95,6 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) } msblk = sb->s_fs_info; - sblk = kzalloc(sizeof(*sblk), GFP_KERNEL); - if (sblk == NULL) { - ERROR("Failed to allocate squashfs_super_block\n"); - goto failure; - } - msblk->devblksize = sb_min_blocksize(sb, BLOCK_SIZE); msblk->devblksize_log2 = ffz(~msblk->devblksize); @@ -114,10 +108,12 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) * of bytes_used) we need to set it to an initial sensible dummy value */ msblk->bytes_used = sizeof(*sblk); - err = squashfs_read_table(sb, sblk, SQUASHFS_START, sizeof(*sblk)); + sblk = squashfs_read_table(sb, SQUASHFS_START, sizeof(*sblk)); - if (err < 0) { + if (IS_ERR(sblk)) { ERROR("unable to read squashfs_super_block\n"); + err = PTR_ERR(sblk); + sblk = NULL; goto failed_mount; } @@ -218,18 +214,61 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; } + /* Handle xattrs */ + sb->s_xattr = squashfs_xattr_handlers; + xattr_id_table_start = le64_to_cpu(sblk->xattr_id_table_start); + if (xattr_id_table_start == SQUASHFS_INVALID_BLK) { + next_table = msblk->bytes_used; + goto allocate_id_index_table; + } + + /* Allocate and read xattr id lookup table */ + msblk->xattr_id_table = squashfs_read_xattr_id_table(sb, + xattr_id_table_start, &msblk->xattr_table, &msblk->xattr_ids); + if (IS_ERR(msblk->xattr_id_table)) { + ERROR("unable to read xattr id index table\n"); + err = PTR_ERR(msblk->xattr_id_table); + msblk->xattr_id_table = NULL; + if (err != -ENOTSUPP) + goto failed_mount; + } + next_table = msblk->xattr_table; + +allocate_id_index_table: /* Allocate and read id index table */ msblk->id_table = squashfs_read_id_index_table(sb, - le64_to_cpu(sblk->id_table_start), le16_to_cpu(sblk->no_ids)); + le64_to_cpu(sblk->id_table_start), next_table, + le16_to_cpu(sblk->no_ids)); if (IS_ERR(msblk->id_table)) { + ERROR("unable to read id index table\n"); err = PTR_ERR(msblk->id_table); msblk->id_table = NULL; goto failed_mount; } + next_table = le64_to_cpu(msblk->id_table[0]); + + /* Handle inode lookup table */ + lookup_table_start = le64_to_cpu(sblk->lookup_table_start); + if (lookup_table_start == SQUASHFS_INVALID_BLK) + goto handle_fragments; + + /* Allocate and read inode lookup table */ + msblk->inode_lookup_table = squashfs_read_inode_lookup_table(sb, + lookup_table_start, next_table, msblk->inodes); + if (IS_ERR(msblk->inode_lookup_table)) { + ERROR("unable to read inode lookup table\n"); + err = PTR_ERR(msblk->inode_lookup_table); + msblk->inode_lookup_table = NULL; + goto failed_mount; + } + next_table = le64_to_cpu(msblk->inode_lookup_table[0]); + sb->s_export_op = &squashfs_export_ops; + +handle_fragments: fragments = le32_to_cpu(sblk->fragments); if (fragments == 0) - goto allocate_lookup_table; + goto check_directory_table; msblk->fragment_cache = squashfs_cache_init("fragment", SQUASHFS_CACHED_FRAGMENTS, msblk->block_size); @@ -240,45 +279,29 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) /* Allocate and read fragment index table */ msblk->fragment_index = squashfs_read_fragment_index_table(sb, - le64_to_cpu(sblk->fragment_table_start), fragments); + le64_to_cpu(sblk->fragment_table_start), next_table, fragments); if (IS_ERR(msblk->fragment_index)) { + ERROR("unable to read fragment index table\n"); err = PTR_ERR(msblk->fragment_index); msblk->fragment_index = NULL; goto failed_mount; } + next_table = le64_to_cpu(msblk->fragment_index[0]); -allocate_lookup_table: - lookup_table_start = le64_to_cpu(sblk->lookup_table_start); - if (lookup_table_start == SQUASHFS_INVALID_BLK) - goto allocate_xattr_table; - - /* Allocate and read inode lookup table */ - msblk->inode_lookup_table = squashfs_read_inode_lookup_table(sb, - lookup_table_start, msblk->inodes); - if (IS_ERR(msblk->inode_lookup_table)) { - err = PTR_ERR(msblk->inode_lookup_table); - msblk->inode_lookup_table = NULL; +check_directory_table: + /* Sanity check directory_table */ + if (msblk->directory_table >= next_table) { + err = -EINVAL; goto failed_mount; } - sb->s_export_op = &squashfs_export_ops; - -allocate_xattr_table: - sb->s_xattr = squashfs_xattr_handlers; - xattr_id_table_start = le64_to_cpu(sblk->xattr_id_table_start); - if (xattr_id_table_start == SQUASHFS_INVALID_BLK) - goto allocate_root; - - /* Allocate and read xattr id lookup table */ - msblk->xattr_id_table = squashfs_read_xattr_id_table(sb, - xattr_id_table_start, &msblk->xattr_table, &msblk->xattr_ids); - if (IS_ERR(msblk->xattr_id_table)) { - err = PTR_ERR(msblk->xattr_id_table); - msblk->xattr_id_table = NULL; - if (err != -ENOTSUPP) - goto failed_mount; + /* Sanity check inode_table */ + if (msblk->inode_table >= msblk->directory_table) { + err = -EINVAL; + goto failed_mount; } -allocate_root: + + /* allocate root */ root = new_inode(sb); if (!root) { err = -ENOMEM; @@ -318,11 +341,6 @@ failed_mount: sb->s_fs_info = NULL; kfree(sblk); return err; - -failure: - kfree(sb->s_fs_info); - sb->s_fs_info = NULL; - return -ENOMEM; } @@ -475,5 +493,5 @@ static const struct super_operations squashfs_super_ops = { module_init(init_squashfs_fs); module_exit(exit_squashfs_fs); MODULE_DESCRIPTION("squashfs 4.0, a compressed read-only filesystem"); -MODULE_AUTHOR("Phillip Lougher <phillip@lougher.demon.co.uk>"); +MODULE_AUTHOR("Phillip Lougher <phillip@squashfs.org.uk>"); MODULE_LICENSE("GPL"); diff --git a/fs/squashfs/symlink.c b/fs/squashfs/symlink.c index ec86434..1191817 100644 --- a/fs/squashfs/symlink.c +++ b/fs/squashfs/symlink.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/xattr.c b/fs/squashfs/xattr.c index 3876c36..92fcde7 100644 --- a/fs/squashfs/xattr.c +++ b/fs/squashfs/xattr.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2010 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/xattr.h b/fs/squashfs/xattr.h index b634efc..c83f5d9 100644 --- a/fs/squashfs/xattr.h +++ b/fs/squashfs/xattr.h @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2010 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -31,6 +31,7 @@ static inline __le64 *squashfs_read_xattr_id_table(struct super_block *sb, u64 start, u64 *xattr_table_start, int *xattr_ids) { ERROR("Xattrs in filesystem, these will be ignored\n"); + *xattr_table_start = start; return ERR_PTR(-ENOTSUPP); } diff --git a/fs/squashfs/xattr_id.c b/fs/squashfs/xattr_id.c index 05385db..c89607d 100644 --- a/fs/squashfs/xattr_id.c +++ b/fs/squashfs/xattr_id.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2010 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -67,34 +67,29 @@ __le64 *squashfs_read_xattr_id_table(struct super_block *sb, u64 start, u64 *xattr_table_start, int *xattr_ids) { unsigned int len; - __le64 *xid_table; - struct squashfs_xattr_id_table id_table; - int err; + struct squashfs_xattr_id_table *id_table; + + id_table = squashfs_read_table(sb, start, sizeof(*id_table)); + if (IS_ERR(id_table)) + return (__le64 *) id_table; + + *xattr_table_start = le64_to_cpu(id_table->xattr_table_start); + *xattr_ids = le32_to_cpu(id_table->xattr_ids); + kfree(id_table); + + /* Sanity check values */ + + /* there is always at least one xattr id */ + if (*xattr_ids == 0) + return ERR_PTR(-EINVAL); + + /* xattr_table should be less than start */ + if (*xattr_table_start >= start) + return ERR_PTR(-EINVAL); - err = squashfs_read_table(sb, &id_table, start, sizeof(id_table)); - if (err < 0) { - ERROR("unable to read xattr id table\n"); - return ERR_PTR(err); - } - *xattr_table_start = le64_to_cpu(id_table.xattr_table_start); - *xattr_ids = le32_to_cpu(id_table.xattr_ids); len = SQUASHFS_XATTR_BLOCK_BYTES(*xattr_ids); TRACE("In read_xattr_index_table, length %d\n", len); - /* Allocate xattr id lookup table indexes */ - xid_table = kmalloc(len, GFP_KERNEL); - if (xid_table == NULL) { - ERROR("Failed to allocate xattr id index table\n"); - return ERR_PTR(-ENOMEM); - } - - err = squashfs_read_table(sb, xid_table, start + sizeof(id_table), len); - if (err < 0) { - ERROR("unable to read xattr id index table\n"); - kfree(xid_table); - return ERR_PTR(err); - } - - return xid_table; + return squashfs_read_table(sb, start + sizeof(*id_table), len); } diff --git a/fs/squashfs/xz_wrapper.c b/fs/squashfs/xz_wrapper.c index aa47a28..1760b7d1 100644 --- a/fs/squashfs/xz_wrapper.c +++ b/fs/squashfs/xz_wrapper.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c index 517688b3..55d918f 100644 --- a/fs/squashfs/zlib_wrapper.c +++ b/fs/squashfs/zlib_wrapper.c @@ -2,7 +2,7 @@ * Squashfs - a compressed read only filesystem for Linux * * Copyright (c) 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 - * Phillip Lougher <phillip@lougher.demon.co.uk> + * Phillip Lougher <phillip@squashfs.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -31,6 +31,7 @@ #include <linux/mutex.h> #include <linux/backing-dev.h> #include <linux/rculist_bl.h> +#include <linux/cleancache.h> #include "internal.h" @@ -112,6 +113,7 @@ static struct super_block *alloc_super(struct file_system_type *type) s->s_maxbytes = MAX_NON_LFS; s->s_op = &default_op; s->s_time_gran = 1000000000; + s->cleancache_poolid = -1; } out: return s; @@ -177,6 +179,7 @@ void deactivate_locked_super(struct super_block *s) { struct file_system_type *fs = s->s_type; if (atomic_dec_and_test(&s->s_active)) { + cleancache_flush_fs(s); fs->kill_sb(s); /* * We need to call rcu_barrier so all the delayed rcu free @@ -819,7 +822,7 @@ struct dentry *mount_bdev(struct file_system_type *fs_type, } else { char b[BDEVNAME_SIZE]; - s->s_flags = flags; + s->s_flags = flags | MS_NOSEC; s->s_mode = mode; strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); sb_set_blocksize(s, block_size(bdev)); @@ -948,8 +951,7 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data) * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE * but s_maxbytes was an unsigned long long for many releases. Throw * this warning for a little while to try and catch filesystems that - * violate this rule. This warning should be either removed or - * converted to a BUG() in 2.6.34. + * violate this rule. */ WARN((sb->s_maxbytes < 0), "%s set sb->s_maxbytes to " "negative value (%lld)\n", type->name, sb->s_maxbytes); diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index da3fefe..1ad8c93 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -24,13 +24,6 @@ #include "sysfs.h" -/* used in crash dumps to help with debugging */ -static char last_sysfs_file[PATH_MAX]; -void sysfs_printk_last_file(void) -{ - printk(KERN_EMERG "last sysfs file: %s\n", last_sysfs_file); -} - /* * There's one sysfs_buffer for each open file and one * sysfs_open_dirent for each sysfs_dirent with one or more open @@ -337,11 +330,6 @@ static int sysfs_open_file(struct inode *inode, struct file *file) struct sysfs_buffer *buffer; const struct sysfs_ops *ops; int error = -EACCES; - char *p; - - p = d_path(&file->f_path, last_sysfs_file, sizeof(last_sysfs_file)); - if (!IS_ERR(p)) - memmove(last_sysfs_file, p, strlen(p) + 1); /* need attr_sd for attr and ops, its parent for kobj */ if (!sysfs_get_active(attr_sd)) diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index c8769dc2..194414f 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -101,9 +101,9 @@ int sysfs_create_group(struct kobject *kobj, } /** - * sysfs_update_group - given a directory kobject, create an attribute group - * @kobj: The kobject to create the group on - * @grp: The attribute group to create + * sysfs_update_group - given a directory kobject, update an attribute group + * @kobj: The kobject to update the group on + * @grp: The attribute group to update * * This function updates an attribute group. Unlike * sysfs_create_group(), it will explicitly not warn or error if any diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 2668957..e34f0d9 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -95,6 +95,14 @@ static int sysfs_set_super(struct super_block *sb, void *data) return error; } +static void free_sysfs_super_info(struct sysfs_super_info *info) +{ + int type; + for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) + kobj_ns_drop(type, info->ns[type]); + kfree(info); +} + static struct dentry *sysfs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { @@ -108,11 +116,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, return ERR_PTR(-ENOMEM); for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) - info->ns[type] = kobj_ns_current(type); + info->ns[type] = kobj_ns_grab_current(type); sb = sget(fs_type, sysfs_test_super, sysfs_set_super, info); if (IS_ERR(sb) || sb->s_fs_info != info) - kfree(info); + free_sysfs_super_info(info); if (IS_ERR(sb)) return ERR_CAST(sb); if (!sb->s_root) { @@ -131,12 +139,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, static void sysfs_kill_sb(struct super_block *sb) { struct sysfs_super_info *info = sysfs_info(sb); - /* Remove the superblock from fs_supers/s_instances * so we can't find it, before freeing sysfs_super_info. */ kill_anon_super(sb); - kfree(info); + free_sysfs_super_info(info); } static struct file_system_type sysfs_fs_type = { @@ -145,28 +152,6 @@ static struct file_system_type sysfs_fs_type = { .kill_sb = sysfs_kill_sb, }; -void sysfs_exit_ns(enum kobj_ns_type type, const void *ns) -{ - struct super_block *sb; - - mutex_lock(&sysfs_mutex); - spin_lock(&sb_lock); - list_for_each_entry(sb, &sysfs_fs_type.fs_supers, s_instances) { - struct sysfs_super_info *info = sysfs_info(sb); - /* - * If we see a superblock on the fs_supers/s_instances - * list the unmount has not completed and sb->s_fs_info - * points to a valid struct sysfs_super_info. - */ - /* Ignore superblocks with the wrong ns */ - if (info->ns[type] != ns) - continue; - info->ns[type] = NULL; - } - spin_unlock(&sb_lock); - mutex_unlock(&sysfs_mutex); -} - int __init sysfs_init(void) { int err = -ENOMEM; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 3d28af31..2ed2404 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -136,7 +136,7 @@ struct sysfs_addrm_cxt { * instance). */ struct sysfs_super_info { - const void *ns[KOBJ_NS_TYPES]; + void *ns[KOBJ_NS_TYPES]; }; #define sysfs_info(SB) ((struct sysfs_super_info *)(SB->s_fs_info)) extern struct sysfs_dirent sysfs_root; diff --git a/fs/timerfd.c b/fs/timerfd.c index 8c4fc14..dffeb37 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -22,16 +22,24 @@ #include <linux/anon_inodes.h> #include <linux/timerfd.h> #include <linux/syscalls.h> +#include <linux/rcupdate.h> struct timerfd_ctx { struct hrtimer tmr; ktime_t tintv; + ktime_t moffs; wait_queue_head_t wqh; u64 ticks; int expired; int clockid; + struct rcu_head rcu; + struct list_head clist; + bool might_cancel; }; +static LIST_HEAD(cancel_list); +static DEFINE_SPINLOCK(cancel_lock); + /* * This gets called when the timer event triggers. We set the "expired" * flag, but we do not re-arm the timer (in case it's necessary, @@ -51,6 +59,66 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) return HRTIMER_NORESTART; } +/* + * Called when the clock was set to cancel the timers in the cancel + * list. This will wake up processes waiting on these timers. The + * wake-up requires ctx->ticks to be non zero, therefore we increment + * it before calling wake_up_locked(). + */ +void timerfd_clock_was_set(void) +{ + ktime_t moffs = ktime_get_monotonic_offset(); + struct timerfd_ctx *ctx; + unsigned long flags; + + rcu_read_lock(); + list_for_each_entry_rcu(ctx, &cancel_list, clist) { + if (!ctx->might_cancel) + continue; + spin_lock_irqsave(&ctx->wqh.lock, flags); + if (ctx->moffs.tv64 != moffs.tv64) { + ctx->moffs.tv64 = KTIME_MAX; + ctx->ticks++; + wake_up_locked(&ctx->wqh); + } + spin_unlock_irqrestore(&ctx->wqh.lock, flags); + } + rcu_read_unlock(); +} + +static void timerfd_remove_cancel(struct timerfd_ctx *ctx) +{ + if (ctx->might_cancel) { + ctx->might_cancel = false; + spin_lock(&cancel_lock); + list_del_rcu(&ctx->clist); + spin_unlock(&cancel_lock); + } +} + +static bool timerfd_canceled(struct timerfd_ctx *ctx) +{ + if (!ctx->might_cancel || ctx->moffs.tv64 != KTIME_MAX) + return false; + ctx->moffs = ktime_get_monotonic_offset(); + return true; +} + +static void timerfd_setup_cancel(struct timerfd_ctx *ctx, int flags) +{ + if (ctx->clockid == CLOCK_REALTIME && (flags & TFD_TIMER_ABSTIME) && + (flags & TFD_TIMER_CANCEL_ON_SET)) { + if (!ctx->might_cancel) { + ctx->might_cancel = true; + spin_lock(&cancel_lock); + list_add_rcu(&ctx->clist, &cancel_list); + spin_unlock(&cancel_lock); + } + } else if (ctx->might_cancel) { + timerfd_remove_cancel(ctx); + } +} + static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx) { ktime_t remaining; @@ -59,11 +127,12 @@ static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx) return remaining.tv64 < 0 ? ktime_set(0, 0): remaining; } -static void timerfd_setup(struct timerfd_ctx *ctx, int flags, - const struct itimerspec *ktmr) +static int timerfd_setup(struct timerfd_ctx *ctx, int flags, + const struct itimerspec *ktmr) { enum hrtimer_mode htmode; ktime_t texp; + int clockid = ctx->clockid; htmode = (flags & TFD_TIMER_ABSTIME) ? HRTIMER_MODE_ABS: HRTIMER_MODE_REL; @@ -72,19 +141,24 @@ static void timerfd_setup(struct timerfd_ctx *ctx, int flags, ctx->expired = 0; ctx->ticks = 0; ctx->tintv = timespec_to_ktime(ktmr->it_interval); - hrtimer_init(&ctx->tmr, ctx->clockid, htmode); + hrtimer_init(&ctx->tmr, clockid, htmode); hrtimer_set_expires(&ctx->tmr, texp); ctx->tmr.function = timerfd_tmrproc; - if (texp.tv64 != 0) + if (texp.tv64 != 0) { hrtimer_start(&ctx->tmr, texp, htmode); + if (timerfd_canceled(ctx)) + return -ECANCELED; + } + return 0; } static int timerfd_release(struct inode *inode, struct file *file) { struct timerfd_ctx *ctx = file->private_data; + timerfd_remove_cancel(ctx); hrtimer_cancel(&ctx->tmr); - kfree(ctx); + kfree_rcu(ctx, rcu); return 0; } @@ -118,8 +192,21 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, res = -EAGAIN; else res = wait_event_interruptible_locked_irq(ctx->wqh, ctx->ticks); + + /* + * If clock has changed, we do not care about the + * ticks and we do not rearm the timer. Userspace must + * reevaluate anyway. + */ + if (timerfd_canceled(ctx)) { + ctx->ticks = 0; + ctx->expired = 0; + res = -ECANCELED; + } + if (ctx->ticks) { ticks = ctx->ticks; + if (ctx->expired && ctx->tintv.tv64) { /* * If tintv.tv64 != 0, this is a periodic timer that @@ -183,6 +270,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) init_waitqueue_head(&ctx->wqh); ctx->clockid = clockid; hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); + ctx->moffs = ktime_get_monotonic_offset(); ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS)); @@ -199,6 +287,7 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, struct file *file; struct timerfd_ctx *ctx; struct itimerspec ktmr, kotmr; + int ret; if (copy_from_user(&ktmr, utmr, sizeof(ktmr))) return -EFAULT; @@ -213,6 +302,8 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, return PTR_ERR(file); ctx = file->private_data; + timerfd_setup_cancel(ctx, flags); + /* * We need to stop the existing timer before reprogramming * it to the new values. @@ -240,14 +331,14 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, /* * Re-program the timer to the new value ... */ - timerfd_setup(ctx, flags, &ktmr); + ret = timerfd_setup(ctx, flags, &ktmr); spin_unlock_irq(&ctx->wqh.lock); fput(file); if (otmr && copy_to_user(otmr, &kotmr, sizeof(kotmr))) return -EFAULT; - return 0; + return ret; } SYSCALL_DEFINE2(timerfd_gettime, int, ufd, struct itimerspec __user *, otmr) diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 8b3a7da..315de66 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -106,7 +106,7 @@ static long long get_liability(struct ubifs_info *c) long long liab; spin_lock(&c->space_lock); - liab = c->budg_idx_growth + c->budg_data_growth + c->budg_dd_growth; + liab = c->bi.idx_growth + c->bi.data_growth + c->bi.dd_growth; spin_unlock(&c->space_lock); return liab; } @@ -180,7 +180,7 @@ int ubifs_calc_min_idx_lebs(struct ubifs_info *c) int idx_lebs; long long idx_size; - idx_size = c->old_idx_sz + c->budg_idx_growth + c->budg_uncommitted_idx; + idx_size = c->bi.old_idx_sz + c->bi.idx_growth + c->bi.uncommitted_idx; /* And make sure we have thrice the index size of space reserved */ idx_size += idx_size << 1; /* @@ -292,13 +292,13 @@ static int can_use_rp(struct ubifs_info *c) * budgeted index space to the size of the current index, multiplies this by 3, * and makes sure this does not exceed the amount of free LEBs. * - * Notes about @c->min_idx_lebs and @c->lst.idx_lebs variables: + * Notes about @c->bi.min_idx_lebs and @c->lst.idx_lebs variables: * o @c->lst.idx_lebs is the number of LEBs the index currently uses. It might * be large, because UBIFS does not do any index consolidation as long as * there is free space. IOW, the index may take a lot of LEBs, but the LEBs * will contain a lot of dirt. - * o @c->min_idx_lebs is the number of LEBS the index presumably takes. IOW, - * the index may be consolidated to take up to @c->min_idx_lebs LEBs. + * o @c->bi.min_idx_lebs is the number of LEBS the index presumably takes. IOW, + * the index may be consolidated to take up to @c->bi.min_idx_lebs LEBs. * * This function returns zero in case of success, and %-ENOSPC in case of * failure. @@ -343,13 +343,13 @@ static int do_budget_space(struct ubifs_info *c) c->lst.taken_empty_lebs; if (unlikely(rsvd_idx_lebs > lebs)) { dbg_budg("out of indexing space: min_idx_lebs %d (old %d), " - "rsvd_idx_lebs %d", min_idx_lebs, c->min_idx_lebs, + "rsvd_idx_lebs %d", min_idx_lebs, c->bi.min_idx_lebs, rsvd_idx_lebs); return -ENOSPC; } available = ubifs_calc_available(c, min_idx_lebs); - outstanding = c->budg_data_growth + c->budg_dd_growth; + outstanding = c->bi.data_growth + c->bi.dd_growth; if (unlikely(available < outstanding)) { dbg_budg("out of data space: available %lld, outstanding %lld", @@ -360,7 +360,7 @@ static int do_budget_space(struct ubifs_info *c) if (available - outstanding <= c->rp_size && !can_use_rp(c)) return -ENOSPC; - c->min_idx_lebs = min_idx_lebs; + c->bi.min_idx_lebs = min_idx_lebs; return 0; } @@ -393,11 +393,11 @@ static int calc_data_growth(const struct ubifs_info *c, { int data_growth; - data_growth = req->new_ino ? c->inode_budget : 0; + data_growth = req->new_ino ? c->bi.inode_budget : 0; if (req->new_page) - data_growth += c->page_budget; + data_growth += c->bi.page_budget; if (req->new_dent) - data_growth += c->dent_budget; + data_growth += c->bi.dent_budget; data_growth += req->new_ino_d; return data_growth; } @@ -413,12 +413,12 @@ static int calc_dd_growth(const struct ubifs_info *c, { int dd_growth; - dd_growth = req->dirtied_page ? c->page_budget : 0; + dd_growth = req->dirtied_page ? c->bi.page_budget : 0; if (req->dirtied_ino) - dd_growth += c->inode_budget << (req->dirtied_ino - 1); + dd_growth += c->bi.inode_budget << (req->dirtied_ino - 1); if (req->mod_dent) - dd_growth += c->dent_budget; + dd_growth += c->bi.dent_budget; dd_growth += req->dirtied_ino_d; return dd_growth; } @@ -460,19 +460,19 @@ int ubifs_budget_space(struct ubifs_info *c, struct ubifs_budget_req *req) again: spin_lock(&c->space_lock); - ubifs_assert(c->budg_idx_growth >= 0); - ubifs_assert(c->budg_data_growth >= 0); - ubifs_assert(c->budg_dd_growth >= 0); + ubifs_assert(c->bi.idx_growth >= 0); + ubifs_assert(c->bi.data_growth >= 0); + ubifs_assert(c->bi.dd_growth >= 0); - if (unlikely(c->nospace) && (c->nospace_rp || !can_use_rp(c))) { + if (unlikely(c->bi.nospace) && (c->bi.nospace_rp || !can_use_rp(c))) { dbg_budg("no space"); spin_unlock(&c->space_lock); return -ENOSPC; } - c->budg_idx_growth += idx_growth; - c->budg_data_growth += data_growth; - c->budg_dd_growth += dd_growth; + c->bi.idx_growth += idx_growth; + c->bi.data_growth += data_growth; + c->bi.dd_growth += dd_growth; err = do_budget_space(c); if (likely(!err)) { @@ -484,9 +484,9 @@ again: } /* Restore the old values */ - c->budg_idx_growth -= idx_growth; - c->budg_data_growth -= data_growth; - c->budg_dd_growth -= dd_growth; + c->bi.idx_growth -= idx_growth; + c->bi.data_growth -= data_growth; + c->bi.dd_growth -= dd_growth; spin_unlock(&c->space_lock); if (req->fast) { @@ -506,9 +506,9 @@ again: goto again; } dbg_budg("FS is full, -ENOSPC"); - c->nospace = 1; + c->bi.nospace = 1; if (can_use_rp(c) || c->rp_size == 0) - c->nospace_rp = 1; + c->bi.nospace_rp = 1; smp_wmb(); } else ubifs_err("cannot budget space, error %d", err); @@ -523,8 +523,8 @@ again: * This function releases the space budgeted by 'ubifs_budget_space()'. Note, * since the index changes (which were budgeted for in @req->idx_growth) will * only be written to the media on commit, this function moves the index budget - * from @c->budg_idx_growth to @c->budg_uncommitted_idx. The latter will be - * zeroed by the commit operation. + * from @c->bi.idx_growth to @c->bi.uncommitted_idx. The latter will be zeroed + * by the commit operation. */ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) { @@ -553,23 +553,23 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) if (!req->data_growth && !req->dd_growth) return; - c->nospace = c->nospace_rp = 0; + c->bi.nospace = c->bi.nospace_rp = 0; smp_wmb(); spin_lock(&c->space_lock); - c->budg_idx_growth -= req->idx_growth; - c->budg_uncommitted_idx += req->idx_growth; - c->budg_data_growth -= req->data_growth; - c->budg_dd_growth -= req->dd_growth; - c->min_idx_lebs = ubifs_calc_min_idx_lebs(c); - - ubifs_assert(c->budg_idx_growth >= 0); - ubifs_assert(c->budg_data_growth >= 0); - ubifs_assert(c->budg_dd_growth >= 0); - ubifs_assert(c->min_idx_lebs < c->main_lebs); - ubifs_assert(!(c->budg_idx_growth & 7)); - ubifs_assert(!(c->budg_data_growth & 7)); - ubifs_assert(!(c->budg_dd_growth & 7)); + c->bi.idx_growth -= req->idx_growth; + c->bi.uncommitted_idx += req->idx_growth; + c->bi.data_growth -= req->data_growth; + c->bi.dd_growth -= req->dd_growth; + c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c); + + ubifs_assert(c->bi.idx_growth >= 0); + ubifs_assert(c->bi.data_growth >= 0); + ubifs_assert(c->bi.dd_growth >= 0); + ubifs_assert(c->bi.min_idx_lebs < c->main_lebs); + ubifs_assert(!(c->bi.idx_growth & 7)); + ubifs_assert(!(c->bi.data_growth & 7)); + ubifs_assert(!(c->bi.dd_growth & 7)); spin_unlock(&c->space_lock); } @@ -586,13 +586,13 @@ void ubifs_convert_page_budget(struct ubifs_info *c) { spin_lock(&c->space_lock); /* Release the index growth reservation */ - c->budg_idx_growth -= c->max_idx_node_sz << UBIFS_BLOCKS_PER_PAGE_SHIFT; + c->bi.idx_growth -= c->max_idx_node_sz << UBIFS_BLOCKS_PER_PAGE_SHIFT; /* Release the data growth reservation */ - c->budg_data_growth -= c->page_budget; + c->bi.data_growth -= c->bi.page_budget; /* Increase the dirty data growth reservation instead */ - c->budg_dd_growth += c->page_budget; + c->bi.dd_growth += c->bi.page_budget; /* And re-calculate the indexing space reservation */ - c->min_idx_lebs = ubifs_calc_min_idx_lebs(c); + c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c); spin_unlock(&c->space_lock); } @@ -612,7 +612,7 @@ void ubifs_release_dirty_inode_budget(struct ubifs_info *c, memset(&req, 0, sizeof(struct ubifs_budget_req)); /* The "no space" flags will be cleared because dd_growth is > 0 */ - req.dd_growth = c->inode_budget + ALIGN(ui->data_len, 8); + req.dd_growth = c->bi.inode_budget + ALIGN(ui->data_len, 8); ubifs_release_budget(c, &req); } @@ -682,9 +682,9 @@ long long ubifs_get_free_space_nolock(struct ubifs_info *c) int rsvd_idx_lebs, lebs; long long available, outstanding, free; - ubifs_assert(c->min_idx_lebs == ubifs_calc_min_idx_lebs(c)); - outstanding = c->budg_data_growth + c->budg_dd_growth; - available = ubifs_calc_available(c, c->min_idx_lebs); + ubifs_assert(c->bi.min_idx_lebs == ubifs_calc_min_idx_lebs(c)); + outstanding = c->bi.data_growth + c->bi.dd_growth; + available = ubifs_calc_available(c, c->bi.min_idx_lebs); /* * When reporting free space to user-space, UBIFS guarantees that it is @@ -697,8 +697,8 @@ long long ubifs_get_free_space_nolock(struct ubifs_info *c) * Note, the calculations below are similar to what we have in * 'do_budget_space()', so refer there for comments. */ - if (c->min_idx_lebs > c->lst.idx_lebs) - rsvd_idx_lebs = c->min_idx_lebs - c->lst.idx_lebs; + if (c->bi.min_idx_lebs > c->lst.idx_lebs) + rsvd_idx_lebs = c->bi.min_idx_lebs - c->lst.idx_lebs; else rsvd_idx_lebs = 0; lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c index 1bd01de..87cd0ea 100644 --- a/fs/ubifs/commit.c +++ b/fs/ubifs/commit.c @@ -182,7 +182,7 @@ static int do_commit(struct ubifs_info *c) c->mst_node->root_len = cpu_to_le32(zroot.len); c->mst_node->ihead_lnum = cpu_to_le32(c->ihead_lnum); c->mst_node->ihead_offs = cpu_to_le32(c->ihead_offs); - c->mst_node->index_size = cpu_to_le64(c->old_idx_sz); + c->mst_node->index_size = cpu_to_le64(c->bi.old_idx_sz); c->mst_node->lpt_lnum = cpu_to_le32(c->lpt_lnum); c->mst_node->lpt_offs = cpu_to_le32(c->lpt_offs); c->mst_node->nhead_lnum = cpu_to_le32(c->nhead_lnum); diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index 004d374..0bb2bce 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -34,7 +34,6 @@ #include <linux/moduleparam.h> #include <linux/debugfs.h> #include <linux/math64.h> -#include <linux/slab.h> #ifdef CONFIG_UBIFS_FS_DEBUG @@ -43,15 +42,12 @@ DEFINE_SPINLOCK(dbg_lock); static char dbg_key_buf0[128]; static char dbg_key_buf1[128]; -unsigned int ubifs_msg_flags; unsigned int ubifs_chk_flags; unsigned int ubifs_tst_flags; -module_param_named(debug_msgs, ubifs_msg_flags, uint, S_IRUGO | S_IWUSR); module_param_named(debug_chks, ubifs_chk_flags, uint, S_IRUGO | S_IWUSR); module_param_named(debug_tsts, ubifs_tst_flags, uint, S_IRUGO | S_IWUSR); -MODULE_PARM_DESC(debug_msgs, "Debug message type flags"); MODULE_PARM_DESC(debug_chks, "Debug check flags"); MODULE_PARM_DESC(debug_tsts, "Debug special test flags"); @@ -317,6 +313,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) printk(KERN_DEBUG "\tflags %#x\n", sup_flags); printk(KERN_DEBUG "\t big_lpt %u\n", !!(sup_flags & UBIFS_FLG_BIGLPT)); + printk(KERN_DEBUG "\t space_fixup %u\n", + !!(sup_flags & UBIFS_FLG_SPACE_FIXUP)); printk(KERN_DEBUG "\tmin_io_size %u\n", le32_to_cpu(sup->min_io_size)); printk(KERN_DEBUG "\tleb_size %u\n", @@ -602,7 +600,7 @@ void dbg_dump_lstats(const struct ubifs_lp_stats *lst) spin_unlock(&dbg_lock); } -void dbg_dump_budg(struct ubifs_info *c) +void dbg_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi) { int i; struct rb_node *rb; @@ -610,26 +608,42 @@ void dbg_dump_budg(struct ubifs_info *c) struct ubifs_gced_idx_leb *idx_gc; long long available, outstanding, free; - ubifs_assert(spin_is_locked(&c->space_lock)); + spin_lock(&c->space_lock); spin_lock(&dbg_lock); - printk(KERN_DEBUG "(pid %d) Budgeting info: budg_data_growth %lld, " - "budg_dd_growth %lld, budg_idx_growth %lld\n", current->pid, - c->budg_data_growth, c->budg_dd_growth, c->budg_idx_growth); - printk(KERN_DEBUG "\tdata budget sum %lld, total budget sum %lld, " - "freeable_cnt %d\n", c->budg_data_growth + c->budg_dd_growth, - c->budg_data_growth + c->budg_dd_growth + c->budg_idx_growth, - c->freeable_cnt); - printk(KERN_DEBUG "\tmin_idx_lebs %d, old_idx_sz %lld, " - "calc_idx_sz %lld, idx_gc_cnt %d\n", c->min_idx_lebs, - c->old_idx_sz, c->calc_idx_sz, c->idx_gc_cnt); + printk(KERN_DEBUG "(pid %d) Budgeting info: data budget sum %lld, " + "total budget sum %lld\n", current->pid, + bi->data_growth + bi->dd_growth, + bi->data_growth + bi->dd_growth + bi->idx_growth); + printk(KERN_DEBUG "\tbudg_data_growth %lld, budg_dd_growth %lld, " + "budg_idx_growth %lld\n", bi->data_growth, bi->dd_growth, + bi->idx_growth); + printk(KERN_DEBUG "\tmin_idx_lebs %d, old_idx_sz %llu, " + "uncommitted_idx %lld\n", bi->min_idx_lebs, bi->old_idx_sz, + bi->uncommitted_idx); + printk(KERN_DEBUG "\tpage_budget %d, inode_budget %d, dent_budget %d\n", + bi->page_budget, bi->inode_budget, bi->dent_budget); + printk(KERN_DEBUG "\tnospace %u, nospace_rp %u\n", + bi->nospace, bi->nospace_rp); + printk(KERN_DEBUG "\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n", + c->dark_wm, c->dead_wm, c->max_idx_node_sz); + + if (bi != &c->bi) + /* + * If we are dumping saved budgeting data, do not print + * additional information which is about the current state, not + * the old one which corresponded to the saved budgeting data. + */ + goto out_unlock; + + printk(KERN_DEBUG "\tfreeable_cnt %d, calc_idx_sz %lld, idx_gc_cnt %d\n", + c->freeable_cnt, c->calc_idx_sz, c->idx_gc_cnt); printk(KERN_DEBUG "\tdirty_pg_cnt %ld, dirty_zn_cnt %ld, " "clean_zn_cnt %ld\n", atomic_long_read(&c->dirty_pg_cnt), atomic_long_read(&c->dirty_zn_cnt), atomic_long_read(&c->clean_zn_cnt)); - printk(KERN_DEBUG "\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n", - c->dark_wm, c->dead_wm, c->max_idx_node_sz); printk(KERN_DEBUG "\tgc_lnum %d, ihead_lnum %d\n", c->gc_lnum, c->ihead_lnum); + /* If we are in R/O mode, journal heads do not exist */ if (c->jheads) for (i = 0; i < c->jhead_cnt; i++) @@ -648,13 +662,15 @@ void dbg_dump_budg(struct ubifs_info *c) printk(KERN_DEBUG "\tcommit state %d\n", c->cmt_state); /* Print budgeting predictions */ - available = ubifs_calc_available(c, c->min_idx_lebs); - outstanding = c->budg_data_growth + c->budg_dd_growth; + available = ubifs_calc_available(c, c->bi.min_idx_lebs); + outstanding = c->bi.data_growth + c->bi.dd_growth; free = ubifs_get_free_space_nolock(c); printk(KERN_DEBUG "Budgeting predictions:\n"); printk(KERN_DEBUG "\tavailable: %lld, outstanding %lld, free %lld\n", available, outstanding, free); +out_unlock: spin_unlock(&dbg_lock); + spin_unlock(&c->space_lock); } void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) @@ -729,7 +745,13 @@ void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) if (bud->lnum == lp->lnum) { int head = 0; for (i = 0; i < c->jhead_cnt; i++) { - if (lp->lnum == c->jheads[i].wbuf.lnum) { + /* + * Note, if we are in R/O mode or in the middle + * of mounting/re-mounting, the write-buffers do + * not exist. + */ + if (c->jheads && + lp->lnum == c->jheads[i].wbuf.lnum) { printk(KERN_CONT ", jhead %s", dbg_jhead(i)); head = 1; @@ -976,6 +998,8 @@ void dbg_save_space_info(struct ubifs_info *c) spin_lock(&c->space_lock); memcpy(&d->saved_lst, &c->lst, sizeof(struct ubifs_lp_stats)); + memcpy(&d->saved_bi, &c->bi, sizeof(struct ubifs_budg_info)); + d->saved_idx_gc_cnt = c->idx_gc_cnt; /* * We use a dirty hack here and zero out @c->freeable_cnt, because it @@ -1042,14 +1066,14 @@ int dbg_check_space_info(struct ubifs_info *c) out: ubifs_msg("saved lprops statistics dump"); dbg_dump_lstats(&d->saved_lst); - ubifs_get_lp_stats(c, &lst); - + ubifs_msg("saved budgeting info dump"); + dbg_dump_budg(c, &d->saved_bi); + ubifs_msg("saved idx_gc_cnt %d", d->saved_idx_gc_cnt); ubifs_msg("current lprops statistics dump"); + ubifs_get_lp_stats(c, &lst); dbg_dump_lstats(&lst); - - spin_lock(&c->space_lock); - dbg_dump_budg(c); - spin_unlock(&c->space_lock); + ubifs_msg("current budgeting info dump"); + dbg_dump_budg(c, &c->bi); dump_stack(); return -EINVAL; } @@ -1793,6 +1817,8 @@ static struct fsck_inode *add_inode(struct ubifs_info *c, struct rb_node **p, *parent = NULL; struct fsck_inode *fscki; ino_t inum = key_inum_flash(c, &ino->key); + struct inode *inode; + struct ubifs_inode *ui; p = &fsckd->inodes.rb_node; while (*p) { @@ -1816,19 +1842,46 @@ static struct fsck_inode *add_inode(struct ubifs_info *c, if (!fscki) return ERR_PTR(-ENOMEM); + inode = ilookup(c->vfs_sb, inum); + fscki->inum = inum; - fscki->nlink = le32_to_cpu(ino->nlink); - fscki->size = le64_to_cpu(ino->size); - fscki->xattr_cnt = le32_to_cpu(ino->xattr_cnt); - fscki->xattr_sz = le32_to_cpu(ino->xattr_size); - fscki->xattr_nms = le32_to_cpu(ino->xattr_names); - fscki->mode = le32_to_cpu(ino->mode); + /* + * If the inode is present in the VFS inode cache, use it instead of + * the on-flash inode which might be out-of-date. E.g., the size might + * be out-of-date. If we do not do this, the following may happen, for + * example: + * 1. A power cut happens + * 2. We mount the file-system R/O, the replay process fixes up the + * inode size in the VFS cache, but on on-flash. + * 3. 'check_leaf()' fails because it hits a data node beyond inode + * size. + */ + if (!inode) { + fscki->nlink = le32_to_cpu(ino->nlink); + fscki->size = le64_to_cpu(ino->size); + fscki->xattr_cnt = le32_to_cpu(ino->xattr_cnt); + fscki->xattr_sz = le32_to_cpu(ino->xattr_size); + fscki->xattr_nms = le32_to_cpu(ino->xattr_names); + fscki->mode = le32_to_cpu(ino->mode); + } else { + ui = ubifs_inode(inode); + fscki->nlink = inode->i_nlink; + fscki->size = inode->i_size; + fscki->xattr_cnt = ui->xattr_cnt; + fscki->xattr_sz = ui->xattr_size; + fscki->xattr_nms = ui->xattr_names; + fscki->mode = inode->i_mode; + iput(inode); + } + if (S_ISDIR(fscki->mode)) { fscki->calc_sz = UBIFS_INO_NODE_SZ; fscki->calc_cnt = 2; } + rb_link_node(&fscki->rb, parent, p); rb_insert_color(&fscki->rb, &fsckd->inodes); + return fscki; } @@ -2421,7 +2474,8 @@ int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head) hashb = key_block(c, &sb->key); if (hasha > hashb) { - ubifs_err("larger hash %u goes before %u", hasha, hashb); + ubifs_err("larger hash %u goes before %u", + hasha, hashb); goto error_dump; } } @@ -2437,14 +2491,12 @@ error_dump: return 0; } -static int invocation_cnt; - int dbg_force_in_the_gaps(void) { - if (!dbg_force_in_the_gaps_enabled) + if (!(ubifs_chk_flags & UBIFS_CHK_GEN)) return 0; - /* Force in-the-gaps every 8th commit */ - return !((invocation_cnt++) & 0x7); + + return !(random32() & 7); } /* Failure mode for recovery testing */ @@ -2632,7 +2684,7 @@ int dbg_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset, int len, int check) { if (in_failure_mode(desc)) - return -EIO; + return -EROFS; return ubi_leb_read(desc, lnum, buf, offset, len, check); } @@ -2642,7 +2694,7 @@ int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf, int err, failing; if (in_failure_mode(desc)) - return -EIO; + return -EROFS; failing = do_fail(desc, lnum, 1); if (failing) cut_data(buf, len); @@ -2650,7 +2702,7 @@ int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf, if (err) return err; if (failing) - return -EIO; + return -EROFS; return 0; } @@ -2660,12 +2712,12 @@ int dbg_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf, int err; if (do_fail(desc, lnum, 1)) - return -EIO; + return -EROFS; err = ubi_leb_change(desc, lnum, buf, len, dtype); if (err) return err; if (do_fail(desc, lnum, 1)) - return -EIO; + return -EROFS; return 0; } @@ -2674,12 +2726,12 @@ int dbg_leb_erase(struct ubi_volume_desc *desc, int lnum) int err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; err = ubi_leb_erase(desc, lnum); if (err) return err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; return 0; } @@ -2688,19 +2740,19 @@ int dbg_leb_unmap(struct ubi_volume_desc *desc, int lnum) int err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; err = ubi_leb_unmap(desc, lnum); if (err) return err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; return 0; } int dbg_is_mapped(struct ubi_volume_desc *desc, int lnum) { if (in_failure_mode(desc)) - return -EIO; + return -EROFS; return ubi_is_mapped(desc, lnum); } @@ -2709,12 +2761,12 @@ int dbg_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype) int err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; err = ubi_leb_map(desc, lnum, dtype); if (err) return err; if (do_fail(desc, lnum, 0)) - return -EIO; + return -EROFS; return 0; } @@ -2784,7 +2836,7 @@ void dbg_debugfs_exit(void) static int open_debugfs_file(struct inode *inode, struct file *file) { file->private_data = inode->i_private; - return 0; + return nonseekable_open(inode, file); } static ssize_t write_debugfs_file(struct file *file, const char __user *buf, @@ -2795,18 +2847,15 @@ static ssize_t write_debugfs_file(struct file *file, const char __user *buf, if (file->f_path.dentry == d->dfs_dump_lprops) dbg_dump_lprops(c); - else if (file->f_path.dentry == d->dfs_dump_budg) { - spin_lock(&c->space_lock); - dbg_dump_budg(c); - spin_unlock(&c->space_lock); - } else if (file->f_path.dentry == d->dfs_dump_tnc) { + else if (file->f_path.dentry == d->dfs_dump_budg) + dbg_dump_budg(c, &c->bi); + else if (file->f_path.dentry == d->dfs_dump_tnc) { mutex_lock(&c->tnc_mutex); dbg_dump_tnc(c); mutex_unlock(&c->tnc_mutex); } else return -EINVAL; - *ppos += count; return count; } @@ -2814,7 +2863,7 @@ static const struct file_operations dfs_fops = { .open = open_debugfs_file, .write = write_debugfs_file, .owner = THIS_MODULE, - .llseek = default_llseek, + .llseek = no_llseek, }; /** diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index e6493ca..a811ac4 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -31,6 +31,8 @@ typedef int (*dbg_znode_callback)(struct ubifs_info *c, #ifdef CONFIG_UBIFS_FS_DEBUG +#include <linux/random.h> + /** * ubifs_debug_info - per-FS debugging information. * @old_zroot: old index root - used by 'dbg_check_old_index()' @@ -50,13 +52,15 @@ typedef int (*dbg_znode_callback)(struct ubifs_info *c, * @new_ihead_offs: used by debugging to check @c->ihead_offs * * @saved_lst: saved lprops statistics (used by 'dbg_save_space_info()') - * @saved_free: saved free space (used by 'dbg_save_space_info()') + * @saved_bi: saved budgeting information + * @saved_free: saved amount of free space + * @saved_idx_gc_cnt: saved value of @c->idx_gc_cnt * - * dfs_dir_name: name of debugfs directory containing this file-system's files - * dfs_dir: direntry object of the file-system debugfs directory - * dfs_dump_lprops: "dump lprops" debugfs knob - * dfs_dump_budg: "dump budgeting information" debugfs knob - * dfs_dump_tnc: "dump TNC" debugfs knob + * @dfs_dir_name: name of debugfs directory containing this file-system's files + * @dfs_dir: direntry object of the file-system debugfs directory + * @dfs_dump_lprops: "dump lprops" debugfs knob + * @dfs_dump_budg: "dump budgeting information" debugfs knob + * @dfs_dump_tnc: "dump TNC" debugfs knob */ struct ubifs_debug_info { struct ubifs_zbranch old_zroot; @@ -76,7 +80,9 @@ struct ubifs_debug_info { int new_ihead_offs; struct ubifs_lp_stats saved_lst; + struct ubifs_budg_info saved_bi; long long saved_free; + int saved_idx_gc_cnt; char dfs_dir_name[100]; struct dentry *dfs_dir; @@ -101,23 +107,7 @@ struct ubifs_debug_info { } \ } while (0) -#define dbg_dump_stack() do { \ - if (!dbg_failure_mode) \ - dump_stack(); \ -} while (0) - -/* Generic debugging messages */ -#define dbg_msg(fmt, ...) do { \ - spin_lock(&dbg_lock); \ - printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n", current->pid, \ - __func__, ##__VA_ARGS__); \ - spin_unlock(&dbg_lock); \ -} while (0) - -#define dbg_do_msg(typ, fmt, ...) do { \ - if (ubifs_msg_flags & typ) \ - dbg_msg(fmt, ##__VA_ARGS__); \ -} while (0) +#define dbg_dump_stack() dump_stack() #define dbg_err(fmt, ...) do { \ spin_lock(&dbg_lock); \ @@ -137,77 +127,40 @@ const char *dbg_key_str1(const struct ubifs_info *c, #define DBGKEY(key) dbg_key_str0(c, (key)) #define DBGKEY1(key) dbg_key_str1(c, (key)) -/* General messages */ -#define dbg_gen(fmt, ...) dbg_do_msg(UBIFS_MSG_GEN, fmt, ##__VA_ARGS__) +#define ubifs_dbg_msg(type, fmt, ...) do { \ + spin_lock(&dbg_lock); \ + pr_debug("UBIFS DBG " type ": " fmt "\n", ##__VA_ARGS__); \ + spin_unlock(&dbg_lock); \ +} while (0) +/* Just a debugging messages not related to any specific UBIFS subsystem */ +#define dbg_msg(fmt, ...) ubifs_dbg_msg("msg", fmt, ##__VA_ARGS__) +/* General messages */ +#define dbg_gen(fmt, ...) ubifs_dbg_msg("gen", fmt, ##__VA_ARGS__) /* Additional journal messages */ -#define dbg_jnl(fmt, ...) dbg_do_msg(UBIFS_MSG_JNL, fmt, ##__VA_ARGS__) - +#define dbg_jnl(fmt, ...) ubifs_dbg_msg("jnl", fmt, ##__VA_ARGS__) /* Additional TNC messages */ -#define dbg_tnc(fmt, ...) dbg_do_msg(UBIFS_MSG_TNC, fmt, ##__VA_ARGS__) - +#define dbg_tnc(fmt, ...) ubifs_dbg_msg("tnc", fmt, ##__VA_ARGS__) /* Additional lprops messages */ -#define dbg_lp(fmt, ...) dbg_do_msg(UBIFS_MSG_LP, fmt, ##__VA_ARGS__) - +#define dbg_lp(fmt, ...) ubifs_dbg_msg("lp", fmt, ##__VA_ARGS__) /* Additional LEB find messages */ -#define dbg_find(fmt, ...) dbg_do_msg(UBIFS_MSG_FIND, fmt, ##__VA_ARGS__) - +#define dbg_find(fmt, ...) ubifs_dbg_msg("find", fmt, ##__VA_ARGS__) /* Additional mount messages */ -#define dbg_mnt(fmt, ...) dbg_do_msg(UBIFS_MSG_MNT, fmt, ##__VA_ARGS__) - +#define dbg_mnt(fmt, ...) ubifs_dbg_msg("mnt", fmt, ##__VA_ARGS__) /* Additional I/O messages */ -#define dbg_io(fmt, ...) dbg_do_msg(UBIFS_MSG_IO, fmt, ##__VA_ARGS__) - +#define dbg_io(fmt, ...) ubifs_dbg_msg("io", fmt, ##__VA_ARGS__) /* Additional commit messages */ -#define dbg_cmt(fmt, ...) dbg_do_msg(UBIFS_MSG_CMT, fmt, ##__VA_ARGS__) - +#define dbg_cmt(fmt, ...) ubifs_dbg_msg("cmt", fmt, ##__VA_ARGS__) /* Additional budgeting messages */ -#define dbg_budg(fmt, ...) dbg_do_msg(UBIFS_MSG_BUDG, fmt, ##__VA_ARGS__) - +#define dbg_budg(fmt, ...) ubifs_dbg_msg("budg", fmt, ##__VA_ARGS__) /* Additional log messages */ -#define dbg_log(fmt, ...) dbg_do_msg(UBIFS_MSG_LOG, fmt, ##__VA_ARGS__) - +#define dbg_log(fmt, ...) ubifs_dbg_msg("log", fmt, ##__VA_ARGS__) /* Additional gc messages */ -#define dbg_gc(fmt, ...) dbg_do_msg(UBIFS_MSG_GC, fmt, ##__VA_ARGS__) - +#define dbg_gc(fmt, ...) ubifs_dbg_msg("gc", fmt, ##__VA_ARGS__) /* Additional scan messages */ -#define dbg_scan(fmt, ...) dbg_do_msg(UBIFS_MSG_SCAN, fmt, ##__VA_ARGS__) - +#define dbg_scan(fmt, ...) ubifs_dbg_msg("scan", fmt, ##__VA_ARGS__) /* Additional recovery messages */ -#define dbg_rcvry(fmt, ...) dbg_do_msg(UBIFS_MSG_RCVRY, fmt, ##__VA_ARGS__) - -/* - * Debugging message type flags. - * - * UBIFS_MSG_GEN: general messages - * UBIFS_MSG_JNL: journal messages - * UBIFS_MSG_MNT: mount messages - * UBIFS_MSG_CMT: commit messages - * UBIFS_MSG_FIND: LEB find messages - * UBIFS_MSG_BUDG: budgeting messages - * UBIFS_MSG_GC: garbage collection messages - * UBIFS_MSG_TNC: TNC messages - * UBIFS_MSG_LP: lprops messages - * UBIFS_MSG_IO: I/O messages - * UBIFS_MSG_LOG: log messages - * UBIFS_MSG_SCAN: scan messages - * UBIFS_MSG_RCVRY: recovery messages - */ -enum { - UBIFS_MSG_GEN = 0x1, - UBIFS_MSG_JNL = 0x2, - UBIFS_MSG_MNT = 0x4, - UBIFS_MSG_CMT = 0x8, - UBIFS_MSG_FIND = 0x10, - UBIFS_MSG_BUDG = 0x20, - UBIFS_MSG_GC = 0x40, - UBIFS_MSG_TNC = 0x80, - UBIFS_MSG_LP = 0x100, - UBIFS_MSG_IO = 0x200, - UBIFS_MSG_LOG = 0x400, - UBIFS_MSG_SCAN = 0x800, - UBIFS_MSG_RCVRY = 0x1000, -}; +#define dbg_rcvry(fmt, ...) ubifs_dbg_msg("rcvry", fmt, ##__VA_ARGS__) /* * Debugging check flags. @@ -233,11 +186,9 @@ enum { /* * Special testing flags. * - * UBIFS_TST_FORCE_IN_THE_GAPS: force the use of in-the-gaps method * UBIFS_TST_RCVRY: failure mode for recovery testing */ enum { - UBIFS_TST_FORCE_IN_THE_GAPS = 0x2, UBIFS_TST_RCVRY = 0x4, }; @@ -262,7 +213,7 @@ void dbg_dump_lpt_node(const struct ubifs_info *c, void *node, int lnum, int offs); void dbg_dump_budget_req(const struct ubifs_budget_req *req); void dbg_dump_lstats(const struct ubifs_lp_stats *lst); -void dbg_dump_budg(struct ubifs_info *c); +void dbg_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi); void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp); void dbg_dump_lprops(struct ubifs_info *c); void dbg_dump_lpt_info(struct ubifs_info *c); @@ -304,18 +255,16 @@ int dbg_check_data_nodes_order(struct ubifs_info *c, struct list_head *head); int dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head); /* Force the use of in-the-gaps method for testing */ - -#define dbg_force_in_the_gaps_enabled \ - (ubifs_tst_flags & UBIFS_TST_FORCE_IN_THE_GAPS) - +static inline int dbg_force_in_the_gaps_enabled(void) +{ + return ubifs_chk_flags & UBIFS_CHK_GEN; +} int dbg_force_in_the_gaps(void); /* Failure mode for recovery testing */ - #define dbg_failure_mode (ubifs_tst_flags & UBIFS_TST_RCVRY) #ifndef UBIFS_DBG_PRESERVE_UBI - #define ubi_leb_read dbg_leb_read #define ubi_leb_write dbg_leb_write #define ubi_leb_change dbg_leb_change @@ -323,7 +272,6 @@ int dbg_force_in_the_gaps(void); #define ubi_leb_unmap dbg_leb_unmap #define ubi_is_mapped dbg_is_mapped #define ubi_leb_map dbg_leb_map - #endif int dbg_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset, @@ -370,33 +318,33 @@ void dbg_debugfs_exit_fs(struct ubifs_info *c); __func__, __LINE__, current->pid); \ } while (0) -#define dbg_err(fmt, ...) do { \ - if (0) \ - ubifs_err(fmt, ##__VA_ARGS__); \ +#define dbg_err(fmt, ...) do { \ + if (0) \ + ubifs_err(fmt, ##__VA_ARGS__); \ } while (0) -#define dbg_msg(fmt, ...) do { \ - if (0) \ - printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n", \ - current->pid, __func__, ##__VA_ARGS__); \ +#define ubifs_dbg_msg(fmt, ...) do { \ + if (0) \ + pr_debug(fmt "\n", ##__VA_ARGS__); \ } while (0) #define dbg_dump_stack() #define ubifs_assert_cmt_locked(c) -#define dbg_gen(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_jnl(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_tnc(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_lp(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_find(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_mnt(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_io(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_cmt(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_budg(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_log(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_gc(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_scan(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) -#define dbg_rcvry(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_msg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_gen(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_jnl(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_tnc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_lp(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_find(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_mnt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_io(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_cmt(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_budg(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_log(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_gc(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_scan(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) +#define dbg_rcvry(fmt, ...) ubifs_dbg_msg(fmt, ##__VA_ARGS__) #define DBGKEY(key) ((char *)(key)) #define DBGKEY1(key) ((char *)(key)) @@ -420,7 +368,9 @@ static inline void dbg_dump_budget_req(const struct ubifs_budget_req *req) { return; } static inline void dbg_dump_lstats(const struct ubifs_lp_stats *lst) { return; } -static inline void dbg_dump_budg(struct ubifs_info *c) { return; } +static inline void +dbg_dump_budg(struct ubifs_info *c, + const struct ubifs_budg_info *bi) { return; } static inline void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) { return; } static inline void dbg_dump_lprops(struct ubifs_info *c) { return; } @@ -482,8 +432,8 @@ dbg_check_nondata_nodes_order(struct ubifs_info *c, struct list_head *head) { return 0; } static inline int dbg_force_in_the_gaps(void) { return 0; } -#define dbg_force_in_the_gaps_enabled 0 -#define dbg_failure_mode 0 +#define dbg_force_in_the_gaps_enabled() 0 +#define dbg_failure_mode 0 static inline int dbg_debugfs_init(void) { return 0; } static inline void dbg_debugfs_exit(void) { return; } diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 7217d67..ef5abd3 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -603,7 +603,7 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry) ubifs_release_budget(c, &req); else { /* We've deleted something - clean the "no space" flags */ - c->nospace = c->nospace_rp = 0; + c->bi.nospace = c->bi.nospace_rp = 0; smp_wmb(); } return 0; @@ -693,7 +693,7 @@ static int ubifs_rmdir(struct inode *dir, struct dentry *dentry) ubifs_release_budget(c, &req); else { /* We've deleted something - clean the "no space" flags */ - c->nospace = c->nospace_rp = 0; + c->bi.nospace = c->bi.nospace_rp = 0; smp_wmb(); } return 0; diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index b286db7..5e7fccf 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -212,7 +212,7 @@ static void release_new_page_budget(struct ubifs_info *c) */ static void release_existing_page_budget(struct ubifs_info *c) { - struct ubifs_budget_req req = { .dd_growth = c->page_budget}; + struct ubifs_budget_req req = { .dd_growth = c->bi.page_budget}; ubifs_release_budget(c, &req); } @@ -971,11 +971,11 @@ static int do_writepage(struct page *page, int len) * the page locked, and it locks @ui_mutex. However, write-back does take inode * @i_mutex, which means other VFS operations may be run on this inode at the * same time. And the problematic one is truncation to smaller size, from where - * we have to call 'truncate_setsize()', which first changes @inode->i_size, then - * drops the truncated pages. And while dropping the pages, it takes the page - * lock. This means that 'do_truncation()' cannot call 'truncate_setsize()' with - * @ui_mutex locked, because it would deadlock with 'ubifs_writepage()'. This - * means that @inode->i_size is changed while @ui_mutex is unlocked. + * we have to call 'truncate_setsize()', which first changes @inode->i_size, + * then drops the truncated pages. And while dropping the pages, it takes the + * page lock. This means that 'do_truncation()' cannot call 'truncate_setsize()' + * with @ui_mutex locked, because it would deadlock with 'ubifs_writepage()'. + * This means that @inode->i_size is changed while @ui_mutex is unlocked. * * XXX(truncate): with the new truncate sequence this is not true anymore, * and the calls to truncate_setsize can be move around freely. They should @@ -1189,7 +1189,7 @@ out_budg: if (budgeted) ubifs_release_budget(c, &req); else { - c->nospace = c->nospace_rp = 0; + c->bi.nospace = c->bi.nospace_rp = 0; smp_wmb(); } return err; @@ -1312,7 +1312,11 @@ int ubifs_fsync(struct file *file, int datasync) dbg_gen("syncing inode %lu", inode->i_ino); - if (inode->i_sb->s_flags & MS_RDONLY) + if (c->ro_mount) + /* + * For some really strange reasons VFS does not filter out + * 'fsync()' for R/O mounted file-systems as per 2.6.39. + */ return 0; /* @@ -1432,10 +1436,11 @@ static int ubifs_releasepage(struct page *page, gfp_t unused_gfp_flags) } /* - * mmap()d file has taken write protection fault and is being made - * writable. UBIFS must ensure page is budgeted for. + * mmap()d file has taken write protection fault and is being made writable. + * UBIFS must ensure page is budgeted for. */ -static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) +static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, + struct vm_fault *vmf) { struct page *page = vmf->page; struct inode *inode = vma->vm_file->f_path.dentry->d_inode; @@ -1536,7 +1541,6 @@ static int ubifs_file_mmap(struct file *file, struct vm_area_struct *vma) { int err; - /* 'generic_file_mmap()' takes care of NOMMU case */ err = generic_file_mmap(file, vma); if (err) return err; diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c index 1d54383..2559d17 100644 --- a/fs/ubifs/find.c +++ b/fs/ubifs/find.c @@ -252,8 +252,8 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, * But if the index takes fewer LEBs than it is reserved for it, * this function must avoid picking those reserved LEBs. */ - if (c->min_idx_lebs >= c->lst.idx_lebs) { - rsvd_idx_lebs = c->min_idx_lebs - c->lst.idx_lebs; + if (c->bi.min_idx_lebs >= c->lst.idx_lebs) { + rsvd_idx_lebs = c->bi.min_idx_lebs - c->lst.idx_lebs; exclude_index = 1; } spin_unlock(&c->space_lock); @@ -276,7 +276,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, pick_free = 0; } else { spin_lock(&c->space_lock); - exclude_index = (c->min_idx_lebs >= c->lst.idx_lebs); + exclude_index = (c->bi.min_idx_lebs >= c->lst.idx_lebs); spin_unlock(&c->space_lock); } @@ -501,8 +501,8 @@ int ubifs_find_free_space(struct ubifs_info *c, int min_space, int *offs, /* Check if there are enough empty LEBs for commit */ spin_lock(&c->space_lock); - if (c->min_idx_lebs > c->lst.idx_lebs) - rsvd_idx_lebs = c->min_idx_lebs - c->lst.idx_lebs; + if (c->bi.min_idx_lebs > c->lst.idx_lebs) + rsvd_idx_lebs = c->bi.min_idx_lebs - c->lst.idx_lebs; else rsvd_idx_lebs = 0; lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index 151f108..ded29f6 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -100,6 +100,10 @@ static int switch_gc_head(struct ubifs_info *c) if (err) return err; + err = ubifs_wbuf_sync_nolock(wbuf); + if (err) + return err; + err = ubifs_add_bud_to_log(c, GCHD, gc_lnum, 0); if (err) return err; @@ -118,7 +122,7 @@ static int switch_gc_head(struct ubifs_info *c) * This function compares data nodes @a and @b. Returns %1 if @a has greater * inode or block number, and %-1 otherwise. */ -int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b) +static int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b) { ino_t inuma, inumb; struct ubifs_info *c = priv; @@ -161,7 +165,8 @@ int data_nodes_cmp(void *priv, struct list_head *a, struct list_head *b) * first and sorted by length in descending order. Directory entry nodes go * after inode nodes and are sorted in ascending hash valuer order. */ -int nondata_nodes_cmp(void *priv, struct list_head *a, struct list_head *b) +static int nondata_nodes_cmp(void *priv, struct list_head *a, + struct list_head *b) { ino_t inuma, inumb; struct ubifs_info *c = priv; @@ -473,6 +478,37 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp) ubifs_assert(c->gc_lnum != lnum); ubifs_assert(wbuf->lnum != lnum); + if (lp->free + lp->dirty == c->leb_size) { + /* Special case - a free LEB */ + dbg_gc("LEB %d is free, return it", lp->lnum); + ubifs_assert(!(lp->flags & LPROPS_INDEX)); + + if (lp->free != c->leb_size) { + /* + * Write buffers must be sync'd before unmapping + * freeable LEBs, because one of them may contain data + * which obsoletes something in 'lp->pnum'. + */ + err = gc_sync_wbufs(c); + if (err) + return err; + err = ubifs_change_one_lp(c, lp->lnum, c->leb_size, + 0, 0, 0, 0); + if (err) + return err; + } + err = ubifs_leb_unmap(c, lp->lnum); + if (err) + return err; + + if (c->gc_lnum == -1) { + c->gc_lnum = lnum; + return LEB_RETAINED; + } + + return LEB_FREED; + } + /* * We scan the entire LEB even though we only really need to scan up to * (c->leb_size - lp->free). @@ -682,37 +718,6 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway) "(min. space %d)", lp.lnum, lp.free, lp.dirty, lp.free + lp.dirty, min_space); - if (lp.free + lp.dirty == c->leb_size) { - /* An empty LEB was returned */ - dbg_gc("LEB %d is free, return it", lp.lnum); - /* - * ubifs_find_dirty_leb() doesn't return freeable index - * LEBs. - */ - ubifs_assert(!(lp.flags & LPROPS_INDEX)); - if (lp.free != c->leb_size) { - /* - * Write buffers must be sync'd before - * unmapping freeable LEBs, because one of them - * may contain data which obsoletes something - * in 'lp.pnum'. - */ - ret = gc_sync_wbufs(c); - if (ret) - goto out; - ret = ubifs_change_one_lp(c, lp.lnum, - c->leb_size, 0, 0, 0, - 0); - if (ret) - goto out; - } - ret = ubifs_leb_unmap(c, lp.lnum); - if (ret) - goto out; - ret = lp.lnum; - break; - } - space_before = c->leb_size - wbuf->offs - wbuf->used; if (wbuf->lnum == -1) space_before = 0; diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c index dfd168b..3be645e 100644 --- a/fs/ubifs/io.c +++ b/fs/ubifs/io.c @@ -393,7 +393,7 @@ int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf) ubifs_assert(wbuf->size % c->min_io_size == 0); ubifs_assert(!c->ro_media && !c->ro_mount); if (c->leb_size - wbuf->offs >= c->max_write_size) - ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size )); + ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size)); if (c->ro_error) return -EROFS; @@ -452,8 +452,8 @@ int ubifs_wbuf_sync_nolock(struct ubifs_wbuf *wbuf) * @dtype: data type * * This function targets the write-buffer to logical eraseblock @lnum:@offs. - * The write-buffer is synchronized if it is not empty. Returns zero in case of - * success and a negative error code in case of failure. + * The write-buffer has to be empty. Returns zero in case of success and a + * negative error code in case of failure. */ int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs, int dtype) @@ -465,13 +465,7 @@ int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs, ubifs_assert(offs >= 0 && offs <= c->leb_size); ubifs_assert(offs % c->min_io_size == 0 && !(offs & 7)); ubifs_assert(lnum != wbuf->lnum); - - if (wbuf->used > 0) { - int err = ubifs_wbuf_sync_nolock(wbuf); - - if (err) - return err; - } + ubifs_assert(wbuf->used == 0); spin_lock(&wbuf->lock); wbuf->lnum = lnum; @@ -573,7 +567,7 @@ out_timers: int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) { struct ubifs_info *c = wbuf->c; - int err, written, n, aligned_len = ALIGN(len, 8), offs; + int err, written, n, aligned_len = ALIGN(len, 8); dbg_io("%d bytes (%s) to jhead %s wbuf at LEB %d:%d", len, dbg_ntype(((struct ubifs_ch *)buf)->node_type), @@ -587,8 +581,9 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) ubifs_assert(wbuf->size % c->min_io_size == 0); ubifs_assert(mutex_is_locked(&wbuf->io_mutex)); ubifs_assert(!c->ro_media && !c->ro_mount); + ubifs_assert(!c->space_fixup); if (c->leb_size - wbuf->offs >= c->max_write_size) - ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size )); + ubifs_assert(!((wbuf->offs + wbuf->size) % c->max_write_size)); if (c->leb_size - wbuf->offs - wbuf->used < aligned_len) { err = -ENOSPC; @@ -636,7 +631,6 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) goto exit; } - offs = wbuf->offs; written = 0; if (wbuf->used) { @@ -653,7 +647,7 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) if (err) goto out; - offs += wbuf->size; + wbuf->offs += wbuf->size; len -= wbuf->avail; aligned_len -= wbuf->avail; written += wbuf->avail; @@ -672,7 +666,7 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) if (err) goto out; - offs += wbuf->size; + wbuf->offs += wbuf->size; len -= wbuf->size; aligned_len -= wbuf->size; written += wbuf->size; @@ -687,12 +681,13 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) n = aligned_len >> c->max_write_shift; if (n) { n <<= c->max_write_shift; - dbg_io("write %d bytes to LEB %d:%d", n, wbuf->lnum, offs); - err = ubi_leb_write(c->ubi, wbuf->lnum, buf + written, offs, n, - wbuf->dtype); + dbg_io("write %d bytes to LEB %d:%d", n, wbuf->lnum, + wbuf->offs); + err = ubi_leb_write(c->ubi, wbuf->lnum, buf + written, + wbuf->offs, n, wbuf->dtype); if (err) goto out; - offs += n; + wbuf->offs += n; aligned_len -= n; len -= n; written += n; @@ -707,7 +702,6 @@ int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len) */ memcpy(wbuf->buf, buf + written, len); - wbuf->offs = offs; if (c->leb_size - wbuf->offs >= c->max_write_size) wbuf->size = c->max_write_size; else @@ -766,6 +760,7 @@ int ubifs_write_node(struct ubifs_info *c, void *buf, int len, int lnum, ubifs_assert(lnum >= 0 && lnum < c->leb_cnt && offs >= 0); ubifs_assert(offs % c->min_io_size == 0 && offs < c->leb_size); ubifs_assert(!c->ro_media && !c->ro_mount); + ubifs_assert(!c->space_fixup); if (c->ro_error) return -EROFS; diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index aed25e8..cef0460 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -141,14 +141,8 @@ again: * LEB with some empty space. */ lnum = ubifs_find_free_space(c, len, &offs, squeeze); - if (lnum >= 0) { - /* Found an LEB, add it to the journal head */ - err = ubifs_add_bud_to_log(c, jhead, lnum, offs); - if (err) - goto out_return; - /* A new bud was successfully allocated and added to the log */ + if (lnum >= 0) goto out; - } err = lnum; if (err != -ENOSPC) @@ -203,12 +197,23 @@ again: return 0; } - err = ubifs_add_bud_to_log(c, jhead, lnum, 0); - if (err) - goto out_return; offs = 0; out: + /* + * Make sure we synchronize the write-buffer before we add the new bud + * to the log. Otherwise we may have a power cut after the log + * reference node for the last bud (@lnum) is written but before the + * write-buffer data are written to the next-to-last bud + * (@wbuf->lnum). And the effect would be that the recovery would see + * that there is corruption in the next-to-last bud. + */ + err = ubifs_wbuf_sync_nolock(wbuf); + if (err) + goto out_return; + err = ubifs_add_bud_to_log(c, jhead, lnum, offs); + if (err) + goto out_return; err = ubifs_wbuf_seek_nolock(wbuf, lnum, offs, wbuf->dtype); if (err) goto out_unlock; @@ -380,10 +385,8 @@ out: if (err == -ENOSPC) { /* This are some budgeting problems, print useful information */ down_write(&c->commit_sem); - spin_lock(&c->space_lock); dbg_dump_stack(); - dbg_dump_budg(c); - spin_unlock(&c->space_lock); + dbg_dump_budg(c, &c->bi); dbg_dump_lprops(c); cmt_retries = dbg_check_lprops(c); up_write(&c->commit_sem); @@ -666,6 +669,7 @@ out_free: out_release: release_head(c, BASEHD); + kfree(dent); out_ro: ubifs_ro_mode(c, err); if (last_reference) diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c index 40fa780..affea94 100644 --- a/fs/ubifs/log.c +++ b/fs/ubifs/log.c @@ -100,20 +100,6 @@ struct ubifs_wbuf *ubifs_get_wbuf(struct ubifs_info *c, int lnum) } /** - * next_log_lnum - switch to the next log LEB. - * @c: UBIFS file-system description object - * @lnum: current log LEB - */ -static inline int next_log_lnum(const struct ubifs_info *c, int lnum) -{ - lnum += 1; - if (lnum > c->log_last) - lnum = UBIFS_LOG_LNUM; - - return lnum; -} - -/** * empty_log_bytes - calculate amount of empty space in the log. * @c: UBIFS file-system description object */ @@ -257,7 +243,7 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs) ref->jhead = cpu_to_le32(jhead); if (c->lhead_offs > c->leb_size - c->ref_node_alsz) { - c->lhead_lnum = next_log_lnum(c, c->lhead_lnum); + c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum); c->lhead_offs = 0; } @@ -425,7 +411,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum) /* Switch to the next log LEB */ if (c->lhead_offs) { - c->lhead_lnum = next_log_lnum(c, c->lhead_lnum); + c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum); c->lhead_offs = 0; } @@ -446,7 +432,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum) c->lhead_offs += len; if (c->lhead_offs == c->leb_size) { - c->lhead_lnum = next_log_lnum(c, c->lhead_lnum); + c->lhead_lnum = ubifs_next_log_lnum(c, c->lhead_lnum); c->lhead_offs = 0; } @@ -533,7 +519,7 @@ int ubifs_log_post_commit(struct ubifs_info *c, int old_ltail_lnum) } mutex_lock(&c->log_mutex); for (lnum = old_ltail_lnum; lnum != c->ltail_lnum; - lnum = next_log_lnum(c, lnum)) { + lnum = ubifs_next_log_lnum(c, lnum)) { dbg_log("unmap log LEB %d", lnum); err = ubifs_leb_unmap(c, lnum); if (err) @@ -642,7 +628,7 @@ static int add_node(struct ubifs_info *c, void *buf, int *lnum, int *offs, err = ubifs_leb_change(c, *lnum, buf, sz, UBI_SHORTTERM); if (err) return err; - *lnum = next_log_lnum(c, *lnum); + *lnum = ubifs_next_log_lnum(c, *lnum); *offs = 0; } memcpy(buf + *offs, node, len); @@ -712,7 +698,7 @@ int ubifs_consolidate_log(struct ubifs_info *c) ubifs_scan_destroy(sleb); if (lnum == c->lhead_lnum) break; - lnum = next_log_lnum(c, lnum); + lnum = ubifs_next_log_lnum(c, lnum); } if (offs) { int sz = ALIGN(offs, c->min_io_size); @@ -732,7 +718,7 @@ int ubifs_consolidate_log(struct ubifs_info *c) /* Unmap remaining LEBs */ lnum = write_lnum; do { - lnum = next_log_lnum(c, lnum); + lnum = ubifs_next_log_lnum(c, lnum); err = ubifs_leb_unmap(c, lnum); if (err) return err; diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c index 0ee0847..667884f 100644 --- a/fs/ubifs/lprops.c +++ b/fs/ubifs/lprops.c @@ -1007,21 +1007,11 @@ out: } /** - * struct scan_check_data - data provided to scan callback function. - * @lst: LEB properties statistics - * @err: error code - */ -struct scan_check_data { - struct ubifs_lp_stats lst; - int err; -}; - -/** * scan_check_cb - scan callback. * @c: the UBIFS file-system description object * @lp: LEB properties to scan * @in_tree: whether the LEB properties are in main memory - * @data: information passed to and from the caller of the scan + * @lst: lprops statistics to update * * This function returns a code that indicates whether the scan should continue * (%LPT_SCAN_CONTINUE), whether the LEB properties should be added to the tree @@ -1030,11 +1020,10 @@ struct scan_check_data { */ static int scan_check_cb(struct ubifs_info *c, const struct ubifs_lprops *lp, int in_tree, - struct scan_check_data *data) + struct ubifs_lp_stats *lst) { struct ubifs_scan_leb *sleb; struct ubifs_scan_node *snod; - struct ubifs_lp_stats *lst = &data->lst; int cat, lnum = lp->lnum, is_idx = 0, used = 0, free, dirty, ret; void *buf = NULL; @@ -1044,7 +1033,7 @@ static int scan_check_cb(struct ubifs_info *c, if (cat != (lp->flags & LPROPS_CAT_MASK)) { ubifs_err("bad LEB category %d expected %d", (lp->flags & LPROPS_CAT_MASK), cat); - goto out; + return -EINVAL; } } @@ -1078,7 +1067,7 @@ static int scan_check_cb(struct ubifs_info *c, } if (!found) { ubifs_err("bad LPT list (category %d)", cat); - goto out; + return -EINVAL; } } } @@ -1090,45 +1079,40 @@ static int scan_check_cb(struct ubifs_info *c, if ((lp->hpos != -1 && heap->arr[lp->hpos]->lnum != lnum) || lp != heap->arr[lp->hpos]) { ubifs_err("bad LPT heap (category %d)", cat); - goto out; + return -EINVAL; } } buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL); - if (!buf) { - ubifs_err("cannot allocate memory to scan LEB %d", lnum); - goto out; + if (!buf) + return -ENOMEM; + + /* + * After an unclean unmount, empty and freeable LEBs + * may contain garbage - do not scan them. + */ + if (lp->free == c->leb_size) { + lst->empty_lebs += 1; + lst->total_free += c->leb_size; + lst->total_dark += ubifs_calc_dark(c, c->leb_size); + return LPT_SCAN_CONTINUE; + } + if (lp->free + lp->dirty == c->leb_size && + !(lp->flags & LPROPS_INDEX)) { + lst->total_free += lp->free; + lst->total_dirty += lp->dirty; + lst->total_dark += ubifs_calc_dark(c, c->leb_size); + return LPT_SCAN_CONTINUE; } sleb = ubifs_scan(c, lnum, 0, buf, 0); if (IS_ERR(sleb)) { - /* - * After an unclean unmount, empty and freeable LEBs - * may contain garbage. - */ - if (lp->free == c->leb_size) { - ubifs_err("scan errors were in empty LEB " - "- continuing checking"); - lst->empty_lebs += 1; - lst->total_free += c->leb_size; - lst->total_dark += ubifs_calc_dark(c, c->leb_size); - ret = LPT_SCAN_CONTINUE; - goto exit; - } - - if (lp->free + lp->dirty == c->leb_size && - !(lp->flags & LPROPS_INDEX)) { - ubifs_err("scan errors were in freeable LEB " - "- continuing checking"); - lst->total_free += lp->free; - lst->total_dirty += lp->dirty; - lst->total_dark += ubifs_calc_dark(c, c->leb_size); - ret = LPT_SCAN_CONTINUE; - goto exit; + ret = PTR_ERR(sleb); + if (ret == -EUCLEAN) { + dbg_dump_lprops(c); + dbg_dump_budg(c, &c->bi); } - data->err = PTR_ERR(sleb); - ret = LPT_SCAN_STOP; - goto exit; + goto out; } is_idx = -1; @@ -1246,10 +1230,8 @@ static int scan_check_cb(struct ubifs_info *c, } ubifs_scan_destroy(sleb); - ret = LPT_SCAN_CONTINUE; -exit: vfree(buf); - return ret; + return LPT_SCAN_CONTINUE; out_print: ubifs_err("bad accounting of LEB %d: free %d, dirty %d flags %#x, " @@ -1258,10 +1240,10 @@ out_print: dbg_dump_leb(c, lnum); out_destroy: ubifs_scan_destroy(sleb); + ret = -EINVAL; out: vfree(buf); - data->err = -EINVAL; - return LPT_SCAN_STOP; + return ret; } /** @@ -1278,8 +1260,7 @@ out: int dbg_check_lprops(struct ubifs_info *c) { int i, err; - struct scan_check_data data; - struct ubifs_lp_stats *lst = &data.lst; + struct ubifs_lp_stats lst; if (!(ubifs_chk_flags & UBIFS_CHK_LPROPS)) return 0; @@ -1294,29 +1275,23 @@ int dbg_check_lprops(struct ubifs_info *c) return err; } - memset(lst, 0, sizeof(struct ubifs_lp_stats)); - - data.err = 0; + memset(&lst, 0, sizeof(struct ubifs_lp_stats)); err = ubifs_lpt_scan_nolock(c, c->main_first, c->leb_cnt - 1, (ubifs_lpt_scan_callback)scan_check_cb, - &data); + &lst); if (err && err != -ENOSPC) goto out; - if (data.err) { - err = data.err; - goto out; - } - if (lst->empty_lebs != c->lst.empty_lebs || - lst->idx_lebs != c->lst.idx_lebs || - lst->total_free != c->lst.total_free || - lst->total_dirty != c->lst.total_dirty || - lst->total_used != c->lst.total_used) { + if (lst.empty_lebs != c->lst.empty_lebs || + lst.idx_lebs != c->lst.idx_lebs || + lst.total_free != c->lst.total_free || + lst.total_dirty != c->lst.total_dirty || + lst.total_used != c->lst.total_used) { ubifs_err("bad overall accounting"); ubifs_err("calculated: empty_lebs %d, idx_lebs %d, " "total_free %lld, total_dirty %lld, total_used %lld", - lst->empty_lebs, lst->idx_lebs, lst->total_free, - lst->total_dirty, lst->total_used); + lst.empty_lebs, lst.idx_lebs, lst.total_free, + lst.total_dirty, lst.total_used); ubifs_err("read from lprops: empty_lebs %d, idx_lebs %d, " "total_free %lld, total_dirty %lld, total_used %lld", c->lst.empty_lebs, c->lst.idx_lebs, c->lst.total_free, @@ -1325,11 +1300,11 @@ int dbg_check_lprops(struct ubifs_info *c) goto out; } - if (lst->total_dead != c->lst.total_dead || - lst->total_dark != c->lst.total_dark) { + if (lst.total_dead != c->lst.total_dead || + lst.total_dark != c->lst.total_dark) { ubifs_err("bad dead/dark space accounting"); ubifs_err("calculated: total_dead %lld, total_dark %lld", - lst->total_dead, lst->total_dark); + lst.total_dead, lst.total_dark); ubifs_err("read from lprops: total_dead %lld, total_dark %lld", c->lst.total_dead, c->lst.total_dark); err = -EINVAL; diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c index 0c9c69b..dfcb574 100644 --- a/fs/ubifs/lpt_commit.c +++ b/fs/ubifs/lpt_commit.c @@ -29,6 +29,12 @@ #include <linux/slab.h> #include "ubifs.h" +#ifdef CONFIG_UBIFS_FS_DEBUG +static int dbg_populate_lsave(struct ubifs_info *c); +#else +#define dbg_populate_lsave(c) 0 +#endif + /** * first_dirty_cnode - find first dirty cnode. * @c: UBIFS file-system description object @@ -586,7 +592,7 @@ static struct ubifs_pnode *next_pnode_to_dirty(struct ubifs_info *c, if (nnode->nbranch[iip].lnum) break; } - } while (iip >= UBIFS_LPT_FANOUT); + } while (iip >= UBIFS_LPT_FANOUT); /* Go right */ nnode = ubifs_get_nnode(c, nnode, iip); @@ -815,6 +821,10 @@ static void populate_lsave(struct ubifs_info *c) c->lpt_drty_flgs |= LSAVE_DIRTY; ubifs_add_lpt_dirt(c, c->lsave_lnum, c->lsave_sz); } + + if (dbg_populate_lsave(c)) + return; + list_for_each_entry(lprops, &c->empty_list, list) { c->lsave[cnt++] = lprops->lnum; if (cnt >= c->lsave_cnt) @@ -1994,4 +2004,47 @@ void dbg_dump_lpt_lebs(const struct ubifs_info *c) current->pid); } +/** + * dbg_populate_lsave - debugging version of 'populate_lsave()' + * @c: UBIFS file-system description object + * + * This is a debugging version for 'populate_lsave()' which populates lsave + * with random LEBs instead of useful LEBs, which is good for test coverage. + * Returns zero if lsave has not been populated (this debugging feature is + * disabled) an non-zero if lsave has been populated. + */ +static int dbg_populate_lsave(struct ubifs_info *c) +{ + struct ubifs_lprops *lprops; + struct ubifs_lpt_heap *heap; + int i; + + if (!(ubifs_chk_flags & UBIFS_CHK_GEN)) + return 0; + if (random32() & 3) + return 0; + + for (i = 0; i < c->lsave_cnt; i++) + c->lsave[i] = c->main_first; + + list_for_each_entry(lprops, &c->empty_list, list) + c->lsave[random32() % c->lsave_cnt] = lprops->lnum; + list_for_each_entry(lprops, &c->freeable_list, list) + c->lsave[random32() % c->lsave_cnt] = lprops->lnum; + list_for_each_entry(lprops, &c->frdi_idx_list, list) + c->lsave[random32() % c->lsave_cnt] = lprops->lnum; + + heap = &c->lpt_heap[LPROPS_DIRTY_IDX - 1]; + for (i = 0; i < heap->cnt; i++) + c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum; + heap = &c->lpt_heap[LPROPS_DIRTY - 1]; + for (i = 0; i < heap->cnt; i++) + c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum; + heap = &c->lpt_heap[LPROPS_FREE - 1]; + for (i = 0; i < heap->cnt; i++) + c->lsave[random32() % c->lsave_cnt] = heap->arr[i]->lnum; + + return 1; +} + #endif /* CONFIG_UBIFS_FS_DEBUG */ diff --git a/fs/ubifs/master.c b/fs/ubifs/master.c index 21f47af..278c238 100644 --- a/fs/ubifs/master.c +++ b/fs/ubifs/master.c @@ -148,7 +148,7 @@ static int validate_master(const struct ubifs_info *c) } main_sz = (long long)c->main_lebs * c->leb_size; - if (c->old_idx_sz & 7 || c->old_idx_sz >= main_sz) { + if (c->bi.old_idx_sz & 7 || c->bi.old_idx_sz >= main_sz) { err = 9; goto out; } @@ -218,7 +218,7 @@ static int validate_master(const struct ubifs_info *c) } if (c->lst.total_dead + c->lst.total_dark + - c->lst.total_used + c->old_idx_sz > main_sz) { + c->lst.total_used + c->bi.old_idx_sz > main_sz) { err = 21; goto out; } @@ -286,7 +286,7 @@ int ubifs_read_master(struct ubifs_info *c) c->gc_lnum = le32_to_cpu(c->mst_node->gc_lnum); c->ihead_lnum = le32_to_cpu(c->mst_node->ihead_lnum); c->ihead_offs = le32_to_cpu(c->mst_node->ihead_offs); - c->old_idx_sz = le64_to_cpu(c->mst_node->index_size); + c->bi.old_idx_sz = le64_to_cpu(c->mst_node->index_size); c->lpt_lnum = le32_to_cpu(c->mst_node->lpt_lnum); c->lpt_offs = le32_to_cpu(c->mst_node->lpt_offs); c->nhead_lnum = le32_to_cpu(c->mst_node->nhead_lnum); @@ -305,7 +305,7 @@ int ubifs_read_master(struct ubifs_info *c) c->lst.total_dead = le64_to_cpu(c->mst_node->total_dead); c->lst.total_dark = le64_to_cpu(c->mst_node->total_dark); - c->calc_idx_sz = c->old_idx_sz; + c->calc_idx_sz = c->bi.old_idx_sz; if (c->mst_node->flags & cpu_to_le32(UBIFS_MST_NO_ORPHS)) c->no_orphs = 1; diff --git a/fs/ubifs/misc.h b/fs/ubifs/misc.h index c3de04d..0b5296a 100644 --- a/fs/ubifs/misc.h +++ b/fs/ubifs/misc.h @@ -340,4 +340,21 @@ static inline void ubifs_release_lprops(struct ubifs_info *c) mutex_unlock(&c->lp_mutex); } +/** + * ubifs_next_log_lnum - switch to the next log LEB. + * @c: UBIFS file-system description object + * @lnum: current log LEB + * + * This helper function returns the log LEB number which goes next after LEB + * 'lnum'. + */ +static inline int ubifs_next_log_lnum(const struct ubifs_info *c, int lnum) +{ + lnum += 1; + if (lnum > c->log_last) + lnum = UBIFS_LOG_LNUM; + + return lnum; +} + #endif /* __UBIFS_MISC_H__ */ diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c index 09df318..a5422ff 100644 --- a/fs/ubifs/orphan.c +++ b/fs/ubifs/orphan.c @@ -673,7 +673,8 @@ static int kill_orphans(struct ubifs_info *c) sleb = ubifs_scan(c, lnum, 0, c->sbuf, 1); if (IS_ERR(sleb)) { if (PTR_ERR(sleb) == -EUCLEAN) - sleb = ubifs_recover_leb(c, lnum, 0, c->sbuf, 0); + sleb = ubifs_recover_leb(c, lnum, 0, + c->sbuf, -1); if (IS_ERR(sleb)) { err = PTR_ERR(sleb); break; diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c index 3dbad6f..783d8e0 100644 --- a/fs/ubifs/recovery.c +++ b/fs/ubifs/recovery.c @@ -564,16 +564,15 @@ static int fix_unclean_leb(struct ubifs_info *c, struct ubifs_scan_leb *sleb, } /** - * drop_incomplete_group - drop nodes from an incomplete group. + * drop_last_group - drop the last group of nodes. * @sleb: scanned LEB information * @offs: offset of dropped nodes is returned here * - * This function returns %1 if nodes are dropped and %0 otherwise. + * This is a helper function for 'ubifs_recover_leb()' which drops the last + * group of nodes of the scanned LEB. */ -static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs) +static void drop_last_group(struct ubifs_scan_leb *sleb, int *offs) { - int dropped = 0; - while (!list_empty(&sleb->nodes)) { struct ubifs_scan_node *snod; struct ubifs_ch *ch; @@ -582,15 +581,40 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs) list); ch = snod->node; if (ch->group_type != UBIFS_IN_NODE_GROUP) - return dropped; - dbg_rcvry("dropping node at %d:%d", sleb->lnum, snod->offs); + break; + + dbg_rcvry("dropping grouped node at %d:%d", + sleb->lnum, snod->offs); + *offs = snod->offs; + list_del(&snod->list); + kfree(snod); + sleb->nodes_cnt -= 1; + } +} + +/** + * drop_last_node - drop the last node. + * @sleb: scanned LEB information + * @offs: offset of dropped nodes is returned here + * @grouped: non-zero if whole group of nodes have to be dropped + * + * This is a helper function for 'ubifs_recover_leb()' which drops the last + * node of the scanned LEB. + */ +static void drop_last_node(struct ubifs_scan_leb *sleb, int *offs) +{ + struct ubifs_scan_node *snod; + + if (!list_empty(&sleb->nodes)) { + snod = list_entry(sleb->nodes.prev, struct ubifs_scan_node, + list); + + dbg_rcvry("dropping last node at %d:%d", sleb->lnum, snod->offs); *offs = snod->offs; list_del(&snod->list); kfree(snod); sleb->nodes_cnt -= 1; - dropped = 1; } - return dropped; } /** @@ -599,7 +623,8 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs) * @lnum: LEB number * @offs: offset * @sbuf: LEB-sized buffer to use - * @grouped: nodes may be grouped for recovery + * @jhead: journal head number this LEB belongs to (%-1 if the LEB does not + * belong to any journal head) * * This function does a scan of a LEB, but caters for errors that might have * been caused by the unclean unmount from which we are attempting to recover. @@ -607,25 +632,21 @@ static int drop_incomplete_group(struct ubifs_scan_leb *sleb, int *offs) * found, and a negative error code in case of failure. */ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, - int offs, void *sbuf, int grouped) + int offs, void *sbuf, int jhead) { - int err, len = c->leb_size - offs, need_clean = 0, quiet = 1; - int empty_chkd = 0, start = offs; + int ret = 0, err, len = c->leb_size - offs, start = offs, min_io_unit; + int grouped = jhead == -1 ? 0 : c->jheads[jhead].grouped; struct ubifs_scan_leb *sleb; void *buf = sbuf + offs; - dbg_rcvry("%d:%d", lnum, offs); + dbg_rcvry("%d:%d, jhead %d, grouped %d", lnum, offs, jhead, grouped); sleb = ubifs_start_scan(c, lnum, offs, sbuf); if (IS_ERR(sleb)) return sleb; - if (sleb->ecc) - need_clean = 1; - + ubifs_assert(len >= 8); while (len >= 8) { - int ret; - dbg_scan("look at LEB %d:%d (%d bytes left)", lnum, offs, len); @@ -635,8 +656,7 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, * Scan quietly until there is an error from which we cannot * recover */ - ret = ubifs_scan_a_node(c, buf, len, lnum, offs, quiet); - + ret = ubifs_scan_a_node(c, buf, len, lnum, offs, 1); if (ret == SCANNED_A_NODE) { /* A valid node, and not a padding node */ struct ubifs_ch *ch = buf; @@ -649,70 +669,32 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, offs += node_len; buf += node_len; len -= node_len; - continue; - } - - if (ret > 0) { + } else if (ret > 0) { /* Padding bytes or a valid padding node */ offs += ret; buf += ret; len -= ret; - continue; - } - - if (ret == SCANNED_EMPTY_SPACE) { - if (!is_empty(buf, len)) { - if (!is_last_write(c, buf, offs)) - break; - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; - } - empty_chkd = 1; + } else if (ret == SCANNED_EMPTY_SPACE || + ret == SCANNED_GARBAGE || + ret == SCANNED_A_BAD_PAD_NODE || + ret == SCANNED_A_CORRUPT_NODE) { + dbg_rcvry("found corruption - %d", ret); break; - } - - if (ret == SCANNED_GARBAGE || ret == SCANNED_A_BAD_PAD_NODE) - if (is_last_write(c, buf, offs)) { - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; - empty_chkd = 1; - break; - } - - if (ret == SCANNED_A_CORRUPT_NODE) - if (no_more_nodes(c, buf, len, lnum, offs)) { - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; - empty_chkd = 1; - break; - } - - if (quiet) { - /* Redo the last scan but noisily */ - quiet = 0; - continue; - } - - switch (ret) { - case SCANNED_GARBAGE: - dbg_err("garbage"); - goto corrupted; - case SCANNED_A_CORRUPT_NODE: - case SCANNED_A_BAD_PAD_NODE: - dbg_err("bad node"); - goto corrupted; - default: - dbg_err("unknown"); + } else { + dbg_err("unexpected return value %d", ret); err = -EINVAL; goto error; } } - if (!empty_chkd && !is_empty(buf, len)) { - if (is_last_write(c, buf, offs)) { - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; - } else { + if (ret == SCANNED_GARBAGE || ret == SCANNED_A_BAD_PAD_NODE) { + if (!is_last_write(c, buf, offs)) + goto corrupted_rescan; + } else if (ret == SCANNED_A_CORRUPT_NODE) { + if (!no_more_nodes(c, buf, len, lnum, offs)) + goto corrupted_rescan; + } else if (!is_empty(buf, len)) { + if (!is_last_write(c, buf, offs)) { int corruption = first_non_ff(buf, len); /* @@ -728,29 +710,85 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, } } - /* Drop nodes from incomplete group */ - if (grouped && drop_incomplete_group(sleb, &offs)) { - buf = sbuf + offs; - len = c->leb_size - offs; - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; - } + min_io_unit = round_down(offs, c->min_io_size); + if (grouped) + /* + * If nodes are grouped, always drop the incomplete group at + * the end. + */ + drop_last_group(sleb, &offs); - if (offs % c->min_io_size) { - clean_buf(c, &buf, lnum, &offs, &len); - need_clean = 1; + if (jhead == GCHD) { + /* + * If this LEB belongs to the GC head then while we are in the + * middle of the same min. I/O unit keep dropping nodes. So + * basically, what we want is to make sure that the last min. + * I/O unit where we saw the corruption is dropped completely + * with all the uncorrupted nodes which may possibly sit there. + * + * In other words, let's name the min. I/O unit where the + * corruption starts B, and the previous min. I/O unit A. The + * below code tries to deal with a situation when half of B + * contains valid nodes or the end of a valid node, and the + * second half of B contains corrupted data or garbage. This + * means that UBIFS had been writing to B just before the power + * cut happened. I do not know how realistic is this scenario + * that half of the min. I/O unit had been written successfully + * and the other half not, but this is possible in our 'failure + * mode emulation' infrastructure at least. + * + * So what is the problem, why we need to drop those nodes? Why + * can't we just clean-up the second half of B by putting a + * padding node there? We can, and this works fine with one + * exception which was reproduced with power cut emulation + * testing and happens extremely rarely. + * + * Imagine the file-system is full, we run GC which starts + * moving valid nodes from LEB X to LEB Y (obviously, LEB Y is + * the current GC head LEB). The @c->gc_lnum is -1, which means + * that GC will retain LEB X and will try to continue. Imagine + * that LEB X is currently the dirtiest LEB, and the amount of + * used space in LEB Y is exactly the same as amount of free + * space in LEB X. + * + * And a power cut happens when nodes are moved from LEB X to + * LEB Y. We are here trying to recover LEB Y which is the GC + * head LEB. We find the min. I/O unit B as described above. + * Then we clean-up LEB Y by padding min. I/O unit. And later + * 'ubifs_rcvry_gc_commit()' function fails, because it cannot + * find a dirty LEB which could be GC'd into LEB Y! Even LEB X + * does not match because the amount of valid nodes there does + * not fit the free space in LEB Y any more! And this is + * because of the padding node which we added to LEB Y. The + * user-visible effect of this which I once observed and + * analysed is that we cannot mount the file-system with + * -ENOSPC error. + * + * So obviously, to make sure that situation does not happen we + * should free min. I/O unit B in LEB Y completely and the last + * used min. I/O unit in LEB Y should be A. This is basically + * what the below code tries to do. + */ + while (offs > min_io_unit) + drop_last_node(sleb, &offs); } + buf = sbuf + offs; + len = c->leb_size - offs; + + clean_buf(c, &buf, lnum, &offs, &len); ubifs_end_scan(c, sleb, lnum, offs); - if (need_clean) { - err = fix_unclean_leb(c, sleb, start); - if (err) - goto error; - } + err = fix_unclean_leb(c, sleb, start); + if (err) + goto error; return sleb; +corrupted_rescan: + /* Re-scan the corrupted data with verbose messages */ + dbg_err("corruptio %d", ret); + ubifs_scan_a_node(c, buf, len, lnum, offs, 1); corrupted: ubifs_scanned_corruption(c, lnum, offs, buf); err = -EUCLEAN; @@ -867,7 +905,7 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum, } ubifs_scan_destroy(sleb); } - return ubifs_recover_leb(c, lnum, offs, sbuf, 0); + return ubifs_recover_leb(c, lnum, offs, sbuf, -1); } /** @@ -1070,6 +1108,53 @@ int ubifs_clean_lebs(const struct ubifs_info *c, void *sbuf) } /** + * grab_empty_leb - grab an empty LEB to use as GC LEB and run commit. + * @c: UBIFS file-system description object + * + * This is a helper function for 'ubifs_rcvry_gc_commit()' which grabs an empty + * LEB to be used as GC LEB (@c->gc_lnum), and then runs the commit. Returns + * zero in case of success and a negative error code in case of failure. + */ +static int grab_empty_leb(struct ubifs_info *c) +{ + int lnum, err; + + /* + * Note, it is very important to first search for an empty LEB and then + * run the commit, not vice-versa. The reason is that there might be + * only one empty LEB at the moment, the one which has been the + * @c->gc_lnum just before the power cut happened. During the regular + * UBIFS operation (not now) @c->gc_lnum is marked as "taken", so no + * one but GC can grab it. But at this moment this single empty LEB is + * not marked as taken, so if we run commit - what happens? Right, the + * commit will grab it and write the index there. Remember that the + * index always expands as long as there is free space, and it only + * starts consolidating when we run out of space. + * + * IOW, if we run commit now, we might not be able to find a free LEB + * after this. + */ + lnum = ubifs_find_free_leb_for_idx(c); + if (lnum < 0) { + dbg_err("could not find an empty LEB"); + dbg_dump_lprops(c); + dbg_dump_budg(c, &c->bi); + return lnum; + } + + /* Reset the index flag */ + err = ubifs_change_one_lp(c, lnum, LPROPS_NC, LPROPS_NC, 0, + LPROPS_INDEX, 0); + if (err) + return err; + + c->gc_lnum = lnum; + dbg_rcvry("found empty LEB %d, run commit", lnum); + + return ubifs_run_commit(c); +} + +/** * ubifs_rcvry_gc_commit - recover the GC LEB number and run the commit. * @c: UBIFS file-system description object * @@ -1091,71 +1176,26 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c) { struct ubifs_wbuf *wbuf = &c->jheads[GCHD].wbuf; struct ubifs_lprops lp; - int lnum, err; + int err; + + dbg_rcvry("GC head LEB %d, offs %d", wbuf->lnum, wbuf->offs); c->gc_lnum = -1; - if (wbuf->lnum == -1) { - dbg_rcvry("no GC head LEB"); - goto find_free; - } - /* - * See whether the used space in the dirtiest LEB fits in the GC head - * LEB. - */ - if (wbuf->offs == c->leb_size) { - dbg_rcvry("no room in GC head LEB"); - goto find_free; - } + if (wbuf->lnum == -1 || wbuf->offs == c->leb_size) + return grab_empty_leb(c); + err = ubifs_find_dirty_leb(c, &lp, wbuf->offs, 2); if (err) { - /* - * There are no dirty or empty LEBs subject to here being - * enough for the index. Try to use - * 'ubifs_find_free_leb_for_idx()', which will return any empty - * LEBs (ignoring index requirements). If the index then - * doesn't have enough LEBs the recovery commit will fail - - * which is the same result anyway i.e. recovery fails. So - * there is no problem ignoring index requirements and just - * grabbing a free LEB since we have already established there - * is not a dirty LEB we could have used instead. - */ - if (err == -ENOSPC) { - dbg_rcvry("could not find a dirty LEB"); - goto find_free; - } - return err; - } - ubifs_assert(!(lp.flags & LPROPS_INDEX)); - lnum = lp.lnum; - if (lp.free + lp.dirty == c->leb_size) { - /* An empty LEB was returned */ - if (lp.free != c->leb_size) { - err = ubifs_change_one_lp(c, lnum, c->leb_size, - 0, 0, 0, 0); - if (err) - return err; - } - err = ubifs_leb_unmap(c, lnum); - if (err) - return err; - c->gc_lnum = lnum; - dbg_rcvry("allocated LEB %d for GC", lnum); - /* Run the commit */ - dbg_rcvry("committing"); - return ubifs_run_commit(c); - } - /* - * There was no empty LEB so the used space in the dirtiest LEB must fit - * in the GC head LEB. - */ - if (lp.free + lp.dirty < wbuf->offs) { - dbg_rcvry("LEB %d doesn't fit in GC head LEB %d:%d", - lnum, wbuf->lnum, wbuf->offs); - err = ubifs_return_leb(c, lnum); - if (err) + if (err != -ENOSPC) return err; - goto find_free; + + dbg_rcvry("could not find a dirty LEB"); + return grab_empty_leb(c); } + + ubifs_assert(!(lp.flags & LPROPS_INDEX)); + ubifs_assert(lp.free + lp.dirty >= wbuf->offs); + /* * We run the commit before garbage collection otherwise subsequent * mounts will see the GC and orphan deletion in a different order. @@ -1164,11 +1204,8 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c) err = ubifs_run_commit(c); if (err) return err; - /* - * The data in the dirtiest LEB fits in the GC head LEB, so do the GC - * - use locking to keep 'ubifs_assert()' happy. - */ - dbg_rcvry("GC'ing LEB %d", lnum); + + dbg_rcvry("GC'ing LEB %d", lp.lnum); mutex_lock_nested(&wbuf->io_mutex, wbuf->jhead); err = ubifs_garbage_collect_leb(c, &lp); if (err >= 0) { @@ -1184,37 +1221,17 @@ int ubifs_rcvry_gc_commit(struct ubifs_info *c) err = -EINVAL; return err; } - if (err != LEB_RETAINED) { - dbg_err("GC returned %d", err); + + ubifs_assert(err == LEB_RETAINED); + if (err != LEB_RETAINED) return -EINVAL; - } + err = ubifs_leb_unmap(c, c->gc_lnum); if (err) return err; - dbg_rcvry("allocated LEB %d for GC", lnum); - return 0; -find_free: - /* - * There is no GC head LEB or the free space in the GC head LEB is too - * small, or there are not dirty LEBs. Allocate gc_lnum by calling - * 'ubifs_find_free_leb_for_idx()' so GC is not run. - */ - lnum = ubifs_find_free_leb_for_idx(c); - if (lnum < 0) { - dbg_err("could not find an empty LEB"); - return lnum; - } - /* And reset the index flag */ - err = ubifs_change_one_lp(c, lnum, LPROPS_NC, LPROPS_NC, 0, - LPROPS_INDEX, 0); - if (err) - return err; - c->gc_lnum = lnum; - dbg_rcvry("allocated LEB %d for GC", lnum); - /* Run the commit */ - dbg_rcvry("committing"); - return ubifs_run_commit(c); + dbg_rcvry("allocated LEB %d for GC", lp.lnum); + return 0; } /** @@ -1456,7 +1473,7 @@ static int fix_size_in_place(struct ubifs_info *c, struct size_entry *e) err = ubi_leb_change(c->ubi, lnum, c->sbuf, len, UBI_UNKNOWN); if (err) goto out; - dbg_rcvry("inode %lu at %d:%d size %lld -> %lld ", + dbg_rcvry("inode %lu at %d:%d size %lld -> %lld", (unsigned long)e->inum, lnum, offs, i_size, e->d_size); return 0; @@ -1505,20 +1522,27 @@ int ubifs_recover_size(struct ubifs_info *c) e->i_size = le64_to_cpu(ino->size); } } + if (e->exists && e->i_size < e->d_size) { - if (!e->inode && c->ro_mount) { + if (c->ro_mount) { /* Fix the inode size and pin it in memory */ struct inode *inode; + struct ubifs_inode *ui; + + ubifs_assert(!e->inode); inode = ubifs_iget(c->vfs_sb, e->inum); if (IS_ERR(inode)) return PTR_ERR(inode); + + ui = ubifs_inode(inode); if (inode->i_size < e->d_size) { dbg_rcvry("ino %lu size %lld -> %lld", (unsigned long)e->inum, - e->d_size, inode->i_size); + inode->i_size, e->d_size); inode->i_size = e->d_size; - ubifs_inode(inode)->ui_size = e->d_size; + ui->ui_size = e->d_size; + ui->synced_i_size = e->d_size; e->inode = inode; this = rb_next(this); continue; @@ -1533,9 +1557,11 @@ int ubifs_recover_size(struct ubifs_info *c) iput(e->inode); } } + this = rb_next(this); rb_erase(&e->rb, &c->size_tree); kfree(e); } + return 0; } diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c index d3d6d36..5e97161 100644 --- a/fs/ubifs/replay.c +++ b/fs/ubifs/replay.c @@ -33,44 +33,32 @@ */ #include "ubifs.h" - -/* - * Replay flags. - * - * REPLAY_DELETION: node was deleted - * REPLAY_REF: node is a reference node - */ -enum { - REPLAY_DELETION = 1, - REPLAY_REF = 2, -}; +#include <linux/list_sort.h> /** - * struct replay_entry - replay tree entry. + * struct replay_entry - replay list entry. * @lnum: logical eraseblock number of the node * @offs: node offset * @len: node length + * @deletion: non-zero if this entry corresponds to a node deletion * @sqnum: node sequence number - * @flags: replay flags - * @rb: links the replay tree + * @list: links the replay list * @key: node key * @nm: directory entry name * @old_size: truncation old size * @new_size: truncation new size - * @free: amount of free space in a bud - * @dirty: amount of dirty space in a bud from padding and deletion nodes - * @jhead: journal head number of the bud * - * UBIFS journal replay must compare node sequence numbers, which means it must - * build a tree of node information to insert into the TNC. + * The replay process first scans all buds and builds the replay list, then + * sorts the replay list in nodes sequence number order, and then inserts all + * the replay entries to the TNC. */ struct replay_entry { int lnum; int offs; int len; + unsigned int deletion:1; unsigned long long sqnum; - int flags; - struct rb_node rb; + struct list_head list; union ubifs_key key; union { struct qstr nm; @@ -78,11 +66,6 @@ struct replay_entry { loff_t old_size; loff_t new_size; }; - struct { - int free; - int dirty; - int jhead; - }; }; }; @@ -90,57 +73,64 @@ struct replay_entry { * struct bud_entry - entry in the list of buds to replay. * @list: next bud in the list * @bud: bud description object - * @free: free bytes in the bud * @sqnum: reference node sequence number + * @free: free bytes in the bud + * @dirty: dirty bytes in the bud */ struct bud_entry { struct list_head list; struct ubifs_bud *bud; - int free; unsigned long long sqnum; + int free; + int dirty; }; /** * set_bud_lprops - set free and dirty space used by a bud. * @c: UBIFS file-system description object - * @r: replay entry of bud + * @b: bud entry which describes the bud + * + * This function makes sure the LEB properties of bud @b are set correctly + * after the replay. Returns zero in case of success and a negative error code + * in case of failure. */ -static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r) +static int set_bud_lprops(struct ubifs_info *c, struct bud_entry *b) { const struct ubifs_lprops *lp; int err = 0, dirty; ubifs_get_lprops(c); - lp = ubifs_lpt_lookup_dirty(c, r->lnum); + lp = ubifs_lpt_lookup_dirty(c, b->bud->lnum); if (IS_ERR(lp)) { err = PTR_ERR(lp); goto out; } dirty = lp->dirty; - if (r->offs == 0 && (lp->free != c->leb_size || lp->dirty != 0)) { + if (b->bud->start == 0 && (lp->free != c->leb_size || lp->dirty != 0)) { /* * The LEB was added to the journal with a starting offset of * zero which means the LEB must have been empty. The LEB - * property values should be lp->free == c->leb_size and - * lp->dirty == 0, but that is not the case. The reason is that - * the LEB was garbage collected. The garbage collector resets - * the free and dirty space without recording it anywhere except - * lprops, so if there is not a commit then lprops does not have - * that information next time the file system is mounted. + * property values should be @lp->free == @c->leb_size and + * @lp->dirty == 0, but that is not the case. The reason is that + * the LEB had been garbage collected before it became the bud, + * and there was not commit inbetween. The garbage collector + * resets the free and dirty space without recording it + * anywhere except lprops, so if there was no commit then + * lprops does not have that information. * * We do not need to adjust free space because the scan has told * us the exact value which is recorded in the replay entry as - * r->free. + * @b->free. * * However we do need to subtract from the dirty space the * amount of space that the garbage collector reclaimed, which * is the whole LEB minus the amount of space that was free. */ - dbg_mnt("bud LEB %d was GC'd (%d free, %d dirty)", r->lnum, + dbg_mnt("bud LEB %d was GC'd (%d free, %d dirty)", b->bud->lnum, lp->free, lp->dirty); - dbg_gc("bud LEB %d was GC'd (%d free, %d dirty)", r->lnum, + dbg_gc("bud LEB %d was GC'd (%d free, %d dirty)", b->bud->lnum, lp->free, lp->dirty); dirty -= c->leb_size - lp->free; /* @@ -152,10 +142,10 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r) */ if (dirty != 0) dbg_msg("LEB %d lp: %d free %d dirty " - "replay: %d free %d dirty", r->lnum, lp->free, - lp->dirty, r->free, r->dirty); + "replay: %d free %d dirty", b->bud->lnum, + lp->free, lp->dirty, b->free, b->dirty); } - lp = ubifs_change_lp(c, lp, r->free, dirty + r->dirty, + lp = ubifs_change_lp(c, lp, b->free, dirty + b->dirty, lp->flags | LPROPS_TAKEN, 0); if (IS_ERR(lp)) { err = PTR_ERR(lp); @@ -163,8 +153,9 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r) } /* Make sure the journal head points to the latest bud */ - err = ubifs_wbuf_seek_nolock(&c->jheads[r->jhead].wbuf, r->lnum, - c->leb_size - r->free, UBI_SHORTTERM); + err = ubifs_wbuf_seek_nolock(&c->jheads[b->bud->jhead].wbuf, + b->bud->lnum, c->leb_size - b->free, + UBI_SHORTTERM); out: ubifs_release_lprops(c); @@ -172,6 +163,27 @@ out: } /** + * set_buds_lprops - set free and dirty space for all replayed buds. + * @c: UBIFS file-system description object + * + * This function sets LEB properties for all replayed buds. Returns zero in + * case of success and a negative error code in case of failure. + */ +static int set_buds_lprops(struct ubifs_info *c) +{ + struct bud_entry *b; + int err; + + list_for_each_entry(b, &c->replay_buds, list) { + err = set_bud_lprops(c, b); + if (err) + return err; + } + + return 0; +} + +/** * trun_remove_range - apply a replay entry for a truncation to the TNC. * @c: UBIFS file-system description object * @r: replay entry of truncation @@ -207,24 +219,22 @@ static int trun_remove_range(struct ubifs_info *c, struct replay_entry *r) */ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r) { - int err, deletion = ((r->flags & REPLAY_DELETION) != 0); + int err; - dbg_mnt("LEB %d:%d len %d flgs %d sqnum %llu %s", r->lnum, - r->offs, r->len, r->flags, r->sqnum, DBGKEY(&r->key)); + dbg_mnt("LEB %d:%d len %d deletion %d sqnum %llu %s", r->lnum, + r->offs, r->len, r->deletion, r->sqnum, DBGKEY(&r->key)); /* Set c->replay_sqnum to help deal with dangling branches. */ c->replay_sqnum = r->sqnum; - if (r->flags & REPLAY_REF) - err = set_bud_lprops(c, r); - else if (is_hash_key(c, &r->key)) { - if (deletion) + if (is_hash_key(c, &r->key)) { + if (r->deletion) err = ubifs_tnc_remove_nm(c, &r->key, &r->nm); else err = ubifs_tnc_add_nm(c, &r->key, r->lnum, r->offs, r->len, &r->nm); } else { - if (deletion) + if (r->deletion) switch (key_type(c, &r->key)) { case UBIFS_INO_KEY: { @@ -247,7 +257,7 @@ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r) return err; if (c->need_recovery) - err = ubifs_recover_size_accum(c, &r->key, deletion, + err = ubifs_recover_size_accum(c, &r->key, r->deletion, r->new_size); } @@ -255,68 +265,77 @@ static int apply_replay_entry(struct ubifs_info *c, struct replay_entry *r) } /** - * destroy_replay_tree - destroy the replay. - * @c: UBIFS file-system description object + * replay_entries_cmp - compare 2 replay entries. + * @priv: UBIFS file-system description object + * @a: first replay entry + * @a: second replay entry * - * Destroy the replay tree. + * This is a comparios function for 'list_sort()' which compares 2 replay + * entries @a and @b by comparing their sequence numer. Returns %1 if @a has + * greater sequence number and %-1 otherwise. */ -static void destroy_replay_tree(struct ubifs_info *c) +static int replay_entries_cmp(void *priv, struct list_head *a, + struct list_head *b) { - struct rb_node *this = c->replay_tree.rb_node; - struct replay_entry *r; - - while (this) { - if (this->rb_left) { - this = this->rb_left; - continue; - } else if (this->rb_right) { - this = this->rb_right; - continue; - } - r = rb_entry(this, struct replay_entry, rb); - this = rb_parent(this); - if (this) { - if (this->rb_left == &r->rb) - this->rb_left = NULL; - else - this->rb_right = NULL; - } - if (is_hash_key(c, &r->key)) - kfree(r->nm.name); - kfree(r); - } - c->replay_tree = RB_ROOT; + struct replay_entry *ra, *rb; + + cond_resched(); + if (a == b) + return 0; + + ra = list_entry(a, struct replay_entry, list); + rb = list_entry(b, struct replay_entry, list); + ubifs_assert(ra->sqnum != rb->sqnum); + if (ra->sqnum > rb->sqnum) + return 1; + return -1; } /** - * apply_replay_tree - apply the replay tree to the TNC. + * apply_replay_list - apply the replay list to the TNC. * @c: UBIFS file-system description object * - * Apply the replay tree. - * Returns zero in case of success and a negative error code in case of - * failure. + * Apply all entries in the replay list to the TNC. Returns zero in case of + * success and a negative error code in case of failure. */ -static int apply_replay_tree(struct ubifs_info *c) +static int apply_replay_list(struct ubifs_info *c) { - struct rb_node *this = rb_first(&c->replay_tree); + struct replay_entry *r; + int err; - while (this) { - struct replay_entry *r; - int err; + list_sort(c, &c->replay_list, &replay_entries_cmp); + list_for_each_entry(r, &c->replay_list, list) { cond_resched(); - r = rb_entry(this, struct replay_entry, rb); err = apply_replay_entry(c, r); if (err) return err; - this = rb_next(this); } + return 0; } /** - * insert_node - insert a node to the replay tree. + * destroy_replay_list - destroy the replay. + * @c: UBIFS file-system description object + * + * Destroy the replay list. + */ +static void destroy_replay_list(struct ubifs_info *c) +{ + struct replay_entry *r, *tmp; + + list_for_each_entry_safe(r, tmp, &c->replay_list, list) { + if (is_hash_key(c, &r->key)) + kfree(r->nm.name); + list_del(&r->list); + kfree(r); + } +} + +/** + * insert_node - insert a node to the replay list * @c: UBIFS file-system description object * @lnum: node logical eraseblock number * @offs: node offset @@ -328,39 +347,25 @@ static int apply_replay_tree(struct ubifs_info *c) * @old_size: truncation old size * @new_size: truncation new size * - * This function inserts a scanned non-direntry node to the replay tree. The - * replay tree is an RB-tree containing @struct replay_entry elements which are - * indexed by the sequence number. The replay tree is applied at the very end - * of the replay process. Since the tree is sorted in sequence number order, - * the older modifications are applied first. This function returns zero in - * case of success and a negative error code in case of failure. + * This function inserts a scanned non-direntry node to the replay list. The + * replay list contains @struct replay_entry elements, and we sort this list in + * sequence number order before applying it. The replay list is applied at the + * very end of the replay process. Since the list is sorted in sequence number + * order, the older modifications are applied first. This function returns zero + * in case of success and a negative error code in case of failure. */ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len, union ubifs_key *key, unsigned long long sqnum, int deletion, int *used, loff_t old_size, loff_t new_size) { - struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL; struct replay_entry *r; + dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); + if (key_inum(c, key) >= c->highest_inum) c->highest_inum = key_inum(c, key); - dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); - while (*p) { - parent = *p; - r = rb_entry(parent, struct replay_entry, rb); - if (sqnum < r->sqnum) { - p = &(*p)->rb_left; - continue; - } else if (sqnum > r->sqnum) { - p = &(*p)->rb_right; - continue; - } - ubifs_err("duplicate sqnum in replay"); - return -EINVAL; - } - r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL); if (!r) return -ENOMEM; @@ -370,19 +375,18 @@ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len, r->lnum = lnum; r->offs = offs; r->len = len; + r->deletion = !!deletion; r->sqnum = sqnum; - r->flags = (deletion ? REPLAY_DELETION : 0); + key_copy(c, key, &r->key); r->old_size = old_size; r->new_size = new_size; - key_copy(c, key, &r->key); - rb_link_node(&r->rb, parent, p); - rb_insert_color(&r->rb, &c->replay_tree); + list_add_tail(&r->list, &c->replay_list); return 0; } /** - * insert_dent - insert a directory entry node into the replay tree. + * insert_dent - insert a directory entry node into the replay list. * @c: UBIFS file-system description object * @lnum: node logical eraseblock number * @offs: node offset @@ -394,43 +398,25 @@ static int insert_node(struct ubifs_info *c, int lnum, int offs, int len, * @deletion: non-zero if this is a deletion * @used: number of bytes in use in a LEB * - * This function inserts a scanned directory entry node to the replay tree. - * Returns zero in case of success and a negative error code in case of - * failure. - * - * This function is also used for extended attribute entries because they are - * implemented as directory entry nodes. + * This function inserts a scanned directory entry node or an extended + * attribute entry to the replay list. Returns zero in case of success and a + * negative error code in case of failure. */ static int insert_dent(struct ubifs_info *c, int lnum, int offs, int len, union ubifs_key *key, const char *name, int nlen, unsigned long long sqnum, int deletion, int *used) { - struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL; struct replay_entry *r; char *nbuf; + dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); if (key_inum(c, key) >= c->highest_inum) c->highest_inum = key_inum(c, key); - dbg_mnt("add LEB %d:%d, key %s", lnum, offs, DBGKEY(key)); - while (*p) { - parent = *p; - r = rb_entry(parent, struct replay_entry, rb); - if (sqnum < r->sqnum) { - p = &(*p)->rb_left; - continue; - } - if (sqnum > r->sqnum) { - p = &(*p)->rb_right; - continue; - } - ubifs_err("duplicate sqnum in replay"); - return -EINVAL; - } - r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL); if (!r) return -ENOMEM; + nbuf = kmalloc(nlen + 1, GFP_KERNEL); if (!nbuf) { kfree(r); @@ -442,17 +428,15 @@ static int insert_dent(struct ubifs_info *c, int lnum, int offs, int len, r->lnum = lnum; r->offs = offs; r->len = len; + r->deletion = !!deletion; r->sqnum = sqnum; + key_copy(c, key, &r->key); r->nm.len = nlen; memcpy(nbuf, name, nlen); nbuf[nlen] = '\0'; r->nm.name = nbuf; - r->flags = (deletion ? REPLAY_DELETION : 0); - key_copy(c, key, &r->key); - ubifs_assert(!*p); - rb_link_node(&r->rb, parent, p); - rb_insert_color(&r->rb, &c->replay_tree); + list_add_tail(&r->list, &c->replay_list); return 0; } @@ -489,29 +473,91 @@ int ubifs_validate_entry(struct ubifs_info *c, } /** + * is_last_bud - check if the bud is the last in the journal head. + * @c: UBIFS file-system description object + * @bud: bud description object + * + * This function checks if bud @bud is the last bud in its journal head. This + * information is then used by 'replay_bud()' to decide whether the bud can + * have corruptions or not. Indeed, only last buds can be corrupted by power + * cuts. Returns %1 if this is the last bud, and %0 if not. + */ +static int is_last_bud(struct ubifs_info *c, struct ubifs_bud *bud) +{ + struct ubifs_jhead *jh = &c->jheads[bud->jhead]; + struct ubifs_bud *next; + uint32_t data; + int err; + + if (list_is_last(&bud->list, &jh->buds_list)) + return 1; + + /* + * The following is a quirk to make sure we work correctly with UBIFS + * images used with older UBIFS. + * + * Normally, the last bud will be the last in the journal head's list + * of bud. However, there is one exception if the UBIFS image belongs + * to older UBIFS. This is fairly unlikely: one would need to use old + * UBIFS, then have a power cut exactly at the right point, and then + * try to mount this image with new UBIFS. + * + * The exception is: it is possible to have 2 buds A and B, A goes + * before B, and B is the last, bud B is contains no data, and bud A is + * corrupted at the end. The reason is that in older versions when the + * journal code switched the next bud (from A to B), it first added a + * log reference node for the new bud (B), and only after this it + * synchronized the write-buffer of current bud (A). But later this was + * changed and UBIFS started to always synchronize the write-buffer of + * the bud (A) before writing the log reference for the new bud (B). + * + * But because older UBIFS always synchronized A's write-buffer before + * writing to B, we can recognize this exceptional situation but + * checking the contents of bud B - if it is empty, then A can be + * treated as the last and we can recover it. + * + * TODO: remove this piece of code in a couple of years (today it is + * 16.05.2011). + */ + next = list_entry(bud->list.next, struct ubifs_bud, list); + if (!list_is_last(&next->list, &jh->buds_list)) + return 0; + + err = ubi_read(c->ubi, next->lnum, (char *)&data, + next->start, 4); + if (err) + return 0; + + return data == 0xFFFFFFFF; +} + +/** * replay_bud - replay a bud logical eraseblock. * @c: UBIFS file-system description object - * @lnum: bud logical eraseblock number to replay - * @offs: bud start offset - * @jhead: journal head to which this bud belongs - * @free: amount of free space in the bud is returned here - * @dirty: amount of dirty space from padding and deletion nodes is returned - * here + * @b: bud entry which describes the bud * - * This function returns zero in case of success and a negative error code in - * case of failure. + * This function replays bud @bud, recovers it if needed, and adds all nodes + * from this bud to the replay list. Returns zero in case of success and a + * negative error code in case of failure. */ -static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead, - int *free, int *dirty) +static int replay_bud(struct ubifs_info *c, struct bud_entry *b) { - int err = 0, used = 0; + int is_last = is_last_bud(c, b->bud); + int err = 0, used = 0, lnum = b->bud->lnum, offs = b->bud->start; struct ubifs_scan_leb *sleb; struct ubifs_scan_node *snod; - struct ubifs_bud *bud; - dbg_mnt("replay bud LEB %d, head %d", lnum, jhead); - if (c->need_recovery) - sleb = ubifs_recover_leb(c, lnum, offs, c->sbuf, jhead != GCHD); + dbg_mnt("replay bud LEB %d, head %d, offs %d, is_last %d", + lnum, b->bud->jhead, offs, is_last); + + if (c->need_recovery && is_last) + /* + * Recover only last LEBs in the journal heads, because power + * cuts may cause corruptions only in these LEBs, because only + * these LEBs could possibly be written to at the power cut + * time. + */ + sleb = ubifs_recover_leb(c, lnum, offs, c->sbuf, b->bud->jhead); else sleb = ubifs_scan(c, lnum, offs, c->sbuf, 0); if (IS_ERR(sleb)) @@ -627,15 +673,13 @@ static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead, goto out; } - bud = ubifs_search_bud(c, lnum); - if (!bud) - BUG(); - + ubifs_assert(ubifs_search_bud(c, lnum)); ubifs_assert(sleb->endpt - offs >= used); ubifs_assert(sleb->endpt % c->min_io_size == 0); - *dirty = sleb->endpt - offs - used; - *free = c->leb_size - sleb->endpt; + b->dirty = sleb->endpt - offs - used; + b->free = c->leb_size - sleb->endpt; + dbg_mnt("bud LEB %d replied: dirty %d, free %d", lnum, b->dirty, b->free); out: ubifs_scan_destroy(sleb); @@ -649,58 +693,6 @@ out_dump: } /** - * insert_ref_node - insert a reference node to the replay tree. - * @c: UBIFS file-system description object - * @lnum: node logical eraseblock number - * @offs: node offset - * @sqnum: sequence number - * @free: amount of free space in bud - * @dirty: amount of dirty space from padding and deletion nodes - * @jhead: journal head number for the bud - * - * This function inserts a reference node to the replay tree and returns zero - * in case of success or a negative error code in case of failure. - */ -static int insert_ref_node(struct ubifs_info *c, int lnum, int offs, - unsigned long long sqnum, int free, int dirty, - int jhead) -{ - struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL; - struct replay_entry *r; - - dbg_mnt("add ref LEB %d:%d", lnum, offs); - while (*p) { - parent = *p; - r = rb_entry(parent, struct replay_entry, rb); - if (sqnum < r->sqnum) { - p = &(*p)->rb_left; - continue; - } else if (sqnum > r->sqnum) { - p = &(*p)->rb_right; - continue; - } - ubifs_err("duplicate sqnum in replay tree"); - return -EINVAL; - } - - r = kzalloc(sizeof(struct replay_entry), GFP_KERNEL); - if (!r) - return -ENOMEM; - - r->lnum = lnum; - r->offs = offs; - r->sqnum = sqnum; - r->flags = REPLAY_REF; - r->free = free; - r->dirty = dirty; - r->jhead = jhead; - - rb_link_node(&r->rb, parent, p); - rb_insert_color(&r->rb, &c->replay_tree); - return 0; -} - -/** * replay_buds - replay all buds. * @c: UBIFS file-system description object * @@ -710,17 +702,16 @@ static int insert_ref_node(struct ubifs_info *c, int lnum, int offs, static int replay_buds(struct ubifs_info *c) { struct bud_entry *b; - int err, uninitialized_var(free), uninitialized_var(dirty); + int err; + unsigned long long prev_sqnum = 0; list_for_each_entry(b, &c->replay_buds, list) { - err = replay_bud(c, b->bud->lnum, b->bud->start, b->bud->jhead, - &free, &dirty); - if (err) - return err; - err = insert_ref_node(c, b->bud->lnum, b->bud->start, b->sqnum, - free, dirty, b->bud->jhead); + err = replay_bud(c, b); if (err) return err; + + ubifs_assert(b->sqnum > prev_sqnum); + prev_sqnum = b->sqnum; } return 0; @@ -1060,25 +1051,29 @@ int ubifs_replay_journal(struct ubifs_info *c) if (err) goto out; - err = apply_replay_tree(c); + err = apply_replay_list(c); + if (err) + goto out; + + err = set_buds_lprops(c); if (err) goto out; /* - * UBIFS budgeting calculations use @c->budg_uncommitted_idx variable - * to roughly estimate index growth. Things like @c->min_idx_lebs + * UBIFS budgeting calculations use @c->bi.uncommitted_idx variable + * to roughly estimate index growth. Things like @c->bi.min_idx_lebs * depend on it. This means we have to initialize it to make sure * budgeting works properly. */ - c->budg_uncommitted_idx = atomic_long_read(&c->dirty_zn_cnt); - c->budg_uncommitted_idx *= c->max_idx_node_sz; + c->bi.uncommitted_idx = atomic_long_read(&c->dirty_zn_cnt); + c->bi.uncommitted_idx *= c->max_idx_node_sz; ubifs_assert(c->bud_bytes <= c->max_bud_bytes || c->need_recovery); dbg_mnt("finished, log head LEB %d:%d, max_sqnum %llu, " "highest_inum %lu", c->lhead_lnum, c->lhead_offs, c->max_sqnum, (unsigned long)c->highest_inum); out: - destroy_replay_tree(c); + destroy_replay_list(c); destroy_bud_list(c); c->replaying = 0; return err; diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c index bf31b47..c606f01 100644 --- a/fs/ubifs/sb.c +++ b/fs/ubifs/sb.c @@ -475,7 +475,8 @@ failed: * @c: UBIFS file-system description object * * This function returns a pointer to the superblock node or a negative error - * code. + * code. Note, the user of this function is responsible of kfree()'ing the + * returned superblock buffer. */ struct ubifs_sb_node *ubifs_read_sb_node(struct ubifs_info *c) { @@ -616,6 +617,7 @@ int ubifs_read_superblock(struct ubifs_info *c) c->vfs_sb->s_time_gran = le32_to_cpu(sup->time_gran); memcpy(&c->uuid, &sup->uuid, 16); c->big_lpt = !!(sup_flags & UBIFS_FLG_BIGLPT); + c->space_fixup = !!(sup_flags & UBIFS_FLG_SPACE_FIXUP); /* Automatically increase file system size to the maximum size */ c->old_leb_cnt = c->leb_cnt; @@ -650,3 +652,152 @@ out: kfree(sup); return err; } + +/** + * fixup_leb - fixup/unmap an LEB containing free space. + * @c: UBIFS file-system description object + * @lnum: the LEB number to fix up + * @len: number of used bytes in LEB (starting at offset 0) + * + * This function reads the contents of the given LEB number @lnum, then fixes + * it up, so that empty min. I/O units in the end of LEB are actually erased on + * flash (rather than being just all-0xff real data). If the LEB is completely + * empty, it is simply unmapped. + */ +static int fixup_leb(struct ubifs_info *c, int lnum, int len) +{ + int err; + + ubifs_assert(len >= 0); + ubifs_assert(len % c->min_io_size == 0); + ubifs_assert(len < c->leb_size); + + if (len == 0) { + dbg_mnt("unmap empty LEB %d", lnum); + return ubi_leb_unmap(c->ubi, lnum); + } + + dbg_mnt("fixup LEB %d, data len %d", lnum, len); + err = ubi_read(c->ubi, lnum, c->sbuf, 0, len); + if (err) + return err; + + return ubi_leb_change(c->ubi, lnum, c->sbuf, len, UBI_UNKNOWN); +} + +/** + * fixup_free_space - find & remap all LEBs containing free space. + * @c: UBIFS file-system description object + * + * This function walks through all LEBs in the filesystem and fiexes up those + * containing free/empty space. + */ +static int fixup_free_space(struct ubifs_info *c) +{ + int lnum, err = 0; + struct ubifs_lprops *lprops; + + ubifs_get_lprops(c); + + /* Fixup LEBs in the master area */ + for (lnum = UBIFS_MST_LNUM; lnum < UBIFS_LOG_LNUM; lnum++) { + err = fixup_leb(c, lnum, c->mst_offs + c->mst_node_alsz); + if (err) + goto out; + } + + /* Unmap unused log LEBs */ + lnum = ubifs_next_log_lnum(c, c->lhead_lnum); + while (lnum != c->ltail_lnum) { + err = fixup_leb(c, lnum, 0); + if (err) + goto out; + lnum = ubifs_next_log_lnum(c, lnum); + } + + /* Fixup the current log head */ + err = fixup_leb(c, c->lhead_lnum, c->lhead_offs); + if (err) + goto out; + + /* Fixup LEBs in the LPT area */ + for (lnum = c->lpt_first; lnum <= c->lpt_last; lnum++) { + int free = c->ltab[lnum - c->lpt_first].free; + + if (free > 0) { + err = fixup_leb(c, lnum, c->leb_size - free); + if (err) + goto out; + } + } + + /* Unmap LEBs in the orphans area */ + for (lnum = c->orph_first; lnum <= c->orph_last; lnum++) { + err = fixup_leb(c, lnum, 0); + if (err) + goto out; + } + + /* Fixup LEBs in the main area */ + for (lnum = c->main_first; lnum < c->leb_cnt; lnum++) { + lprops = ubifs_lpt_lookup(c, lnum); + if (IS_ERR(lprops)) { + err = PTR_ERR(lprops); + goto out; + } + + if (lprops->free > 0) { + err = fixup_leb(c, lnum, c->leb_size - lprops->free); + if (err) + goto out; + } + } + +out: + ubifs_release_lprops(c); + return err; +} + +/** + * ubifs_fixup_free_space - find & fix all LEBs with free space. + * @c: UBIFS file-system description object + * + * This function fixes up LEBs containing free space on first mount, if the + * appropriate flag was set when the FS was created. Each LEB with one or more + * empty min. I/O unit (i.e. free-space-count > 0) is re-written, to make sure + * the free space is actually erased. E.g., this is necessary for some NAND + * chips, since the free space may have been programmed like real "0xff" data + * (generating a non-0xff ECC), causing future writes to the not-really-erased + * NAND pages to behave badly. After the space is fixed up, the superblock flag + * is cleared, so that this is skipped for all future mounts. + */ +int ubifs_fixup_free_space(struct ubifs_info *c) +{ + int err; + struct ubifs_sb_node *sup; + + ubifs_assert(c->space_fixup); + ubifs_assert(!c->ro_mount); + + ubifs_msg("start fixing up free space"); + + err = fixup_free_space(c); + if (err) + return err; + + sup = ubifs_read_sb_node(c); + if (IS_ERR(sup)) + return PTR_ERR(sup); + + /* Free-space fixup is no longer required */ + c->space_fixup = 0; + sup->flags &= cpu_to_le32(~UBIFS_FLG_SPACE_FIXUP); + + err = ubifs_write_sb_node(c, sup); + kfree(sup); + if (err) + return err; + + ubifs_msg("free space fixup complete"); + return err; +} diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c index 46961c0..9e1d056 100644 --- a/fs/ubifs/shrinker.c +++ b/fs/ubifs/shrinker.c @@ -277,13 +277,18 @@ static int kick_a_thread(void) return 0; } -int ubifs_shrinker(struct shrinker *shrink, int nr, gfp_t gfp_mask) +int ubifs_shrinker(struct shrinker *shrink, struct shrink_control *sc) { + int nr = sc->nr_to_scan; int freed, contention = 0; long clean_zn_cnt = atomic_long_read(&ubifs_clean_zn_cnt); if (nr == 0) - return clean_zn_cnt; + /* + * Due to the way UBIFS updates the clean znode counter it may + * temporarily be negative. + */ + return clean_zn_cnt >= 0 ? clean_zn_cnt : 1; if (!clean_zn_cnt) { /* diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 04ad07f..529be05 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -375,14 +375,14 @@ out: ubifs_release_dirty_inode_budget(c, ui); else { /* We've deleted something - clean the "no space" flags */ - c->nospace = c->nospace_rp = 0; + c->bi.nospace = c->bi.nospace_rp = 0; smp_wmb(); } done: end_writeback(inode); } -static void ubifs_dirty_inode(struct inode *inode) +static void ubifs_dirty_inode(struct inode *inode, int flags) { struct ubifs_inode *ui = ubifs_inode(inode); @@ -694,11 +694,11 @@ static int init_constants_sb(struct ubifs_info *c) * be compressed and direntries are of the maximum size. * * Note, data, which may be stored in inodes is budgeted separately, so - * it is not included into 'c->inode_budget'. + * it is not included into 'c->bi.inode_budget'. */ - c->page_budget = UBIFS_MAX_DATA_NODE_SZ * UBIFS_BLOCKS_PER_PAGE; - c->inode_budget = UBIFS_INO_NODE_SZ; - c->dent_budget = UBIFS_MAX_DENT_NODE_SZ; + c->bi.page_budget = UBIFS_MAX_DATA_NODE_SZ * UBIFS_BLOCKS_PER_PAGE; + c->bi.inode_budget = UBIFS_INO_NODE_SZ; + c->bi.dent_budget = UBIFS_MAX_DENT_NODE_SZ; /* * When the amount of flash space used by buds becomes @@ -742,7 +742,7 @@ static void init_constants_master(struct ubifs_info *c) { long long tmp64; - c->min_idx_lebs = ubifs_calc_min_idx_lebs(c); + c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c); c->report_rp_size = ubifs_reported_space(c, c->rp_size); /* @@ -811,15 +811,18 @@ static int alloc_wbufs(struct ubifs_info *c) c->jheads[i].wbuf.sync_callback = &bud_wbuf_callback; c->jheads[i].wbuf.jhead = i; + c->jheads[i].grouped = 1; } c->jheads[BASEHD].wbuf.dtype = UBI_SHORTTERM; /* * Garbage Collector head likely contains long-term data and - * does not need to be synchronized by timer. + * does not need to be synchronized by timer. Also GC head nodes are + * not grouped. */ c->jheads[GCHD].wbuf.dtype = UBI_LONGTERM; c->jheads[GCHD].wbuf.no_timer = 1; + c->jheads[GCHD].grouped = 0; return 0; } @@ -1144,8 +1147,8 @@ static int check_free_space(struct ubifs_info *c) { ubifs_assert(c->dark_wm > 0); if (c->lst.total_free + c->lst.total_dirty < c->dark_wm) { - ubifs_err("insufficient free space to mount in read/write mode"); - dbg_dump_budg(c); + ubifs_err("insufficient free space to mount in R/W mode"); + dbg_dump_budg(c, &c->bi); dbg_dump_lprops(c); return -ENOSPC; } @@ -1284,12 +1287,25 @@ static int mount_ubifs(struct ubifs_info *c) if ((c->mst_node->flags & cpu_to_le32(UBIFS_MST_DIRTY)) != 0) { ubifs_msg("recovery needed"); c->need_recovery = 1; - if (!c->ro_mount) { - err = ubifs_recover_inl_heads(c, c->sbuf); - if (err) - goto out_master; - } - } else if (!c->ro_mount) { + } + + if (c->need_recovery && !c->ro_mount) { + err = ubifs_recover_inl_heads(c, c->sbuf); + if (err) + goto out_master; + } + + err = ubifs_lpt_init(c, 1, !c->ro_mount); + if (err) + goto out_master; + + if (!c->ro_mount && c->space_fixup) { + err = ubifs_fixup_free_space(c); + if (err) + goto out_master; + } + + if (!c->ro_mount) { /* * Set the "dirty" flag so that if we reboot uncleanly we * will notice this immediately on the next mount. @@ -1297,14 +1313,10 @@ static int mount_ubifs(struct ubifs_info *c) c->mst_node->flags |= cpu_to_le32(UBIFS_MST_DIRTY); err = ubifs_write_master(c); if (err) - goto out_master; + goto out_lpt; } - err = ubifs_lpt_init(c, 1, !c->ro_mount); - if (err) - goto out_lpt; - - err = dbg_check_idx_size(c, c->old_idx_sz); + err = dbg_check_idx_size(c, c->bi.old_idx_sz); if (err) goto out_lpt; @@ -1313,7 +1325,7 @@ static int mount_ubifs(struct ubifs_info *c) goto out_journal; /* Calculate 'min_idx_lebs' after journal replay */ - c->min_idx_lebs = ubifs_calc_min_idx_lebs(c); + c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c); err = ubifs_mount_orphans(c, c->need_recovery, c->ro_mount); if (err) @@ -1442,7 +1454,8 @@ static int mount_ubifs(struct ubifs_info *c) c->main_lebs, c->main_first, c->leb_cnt - 1); dbg_msg("index LEBs: %d", c->lst.idx_lebs); dbg_msg("total index bytes: %lld (%lld KiB, %lld MiB)", - c->old_idx_sz, c->old_idx_sz >> 10, c->old_idx_sz >> 20); + c->bi.old_idx_sz, c->bi.old_idx_sz >> 10, + c->bi.old_idx_sz >> 20); dbg_msg("key hash type: %d", c->key_hash_type); dbg_msg("tree fanout: %d", c->fanout); dbg_msg("reserved GC LEB: %d", c->gc_lnum); @@ -1456,7 +1469,7 @@ static int mount_ubifs(struct ubifs_info *c) dbg_msg("node sizes: ref %zu, cmt. start %zu, orph %zu", UBIFS_REF_NODE_SZ, UBIFS_CS_NODE_SZ, UBIFS_ORPH_NODE_SZ); dbg_msg("max. node sizes: data %zu, inode %zu dentry %zu, idx %d", - UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ, + UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ, UBIFS_MAX_DENT_NODE_SZ, ubifs_idx_node_sz(c, c->fanout)); dbg_msg("dead watermark: %d", c->dead_wm); dbg_msg("dark watermark: %d", c->dark_wm); @@ -1584,6 +1597,7 @@ static int ubifs_remount_rw(struct ubifs_info *c) } sup->leb_cnt = cpu_to_le32(c->leb_cnt); err = ubifs_write_sb_node(c, sup); + kfree(sup); if (err) goto out; } @@ -1684,6 +1698,13 @@ static int ubifs_remount_rw(struct ubifs_info *c) */ err = dbg_check_space_info(c); } + + if (c->space_fixup) { + err = ubifs_fixup_free_space(c); + if (err) + goto out; + } + mutex_unlock(&c->umount_mutex); return err; @@ -1766,10 +1787,9 @@ static void ubifs_put_super(struct super_block *sb) * to write them back because of I/O errors. */ if (!c->ro_error) { - ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0); - ubifs_assert(c->budg_idx_growth == 0); - ubifs_assert(c->budg_dd_growth == 0); - ubifs_assert(c->budg_data_growth == 0); + ubifs_assert(c->bi.idx_growth == 0); + ubifs_assert(c->bi.dd_growth == 0); + ubifs_assert(c->bi.data_growth == 0); } /* @@ -1828,7 +1848,6 @@ static void ubifs_put_super(struct super_block *sb) bdi_destroy(&c->bdi); ubi_close_volume(c->ubi); mutex_unlock(&c->umount_mutex); - kfree(c); } static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data) @@ -1951,61 +1970,65 @@ static struct ubi_volume_desc *open_ubi(const char *name, int mode) return ERR_PTR(-EINVAL); } -static int ubifs_fill_super(struct super_block *sb, void *data, int silent) +static struct ubifs_info *alloc_ubifs_info(struct ubi_volume_desc *ubi) { - struct ubi_volume_desc *ubi = sb->s_fs_info; struct ubifs_info *c; - struct inode *root; - int err; c = kzalloc(sizeof(struct ubifs_info), GFP_KERNEL); - if (!c) - return -ENOMEM; + if (c) { + spin_lock_init(&c->cnt_lock); + spin_lock_init(&c->cs_lock); + spin_lock_init(&c->buds_lock); + spin_lock_init(&c->space_lock); + spin_lock_init(&c->orphan_lock); + init_rwsem(&c->commit_sem); + mutex_init(&c->lp_mutex); + mutex_init(&c->tnc_mutex); + mutex_init(&c->log_mutex); + mutex_init(&c->mst_mutex); + mutex_init(&c->umount_mutex); + mutex_init(&c->bu_mutex); + mutex_init(&c->write_reserve_mutex); + init_waitqueue_head(&c->cmt_wq); + c->buds = RB_ROOT; + c->old_idx = RB_ROOT; + c->size_tree = RB_ROOT; + c->orph_tree = RB_ROOT; + INIT_LIST_HEAD(&c->infos_list); + INIT_LIST_HEAD(&c->idx_gc); + INIT_LIST_HEAD(&c->replay_list); + INIT_LIST_HEAD(&c->replay_buds); + INIT_LIST_HEAD(&c->uncat_list); + INIT_LIST_HEAD(&c->empty_list); + INIT_LIST_HEAD(&c->freeable_list); + INIT_LIST_HEAD(&c->frdi_idx_list); + INIT_LIST_HEAD(&c->unclean_leb_list); + INIT_LIST_HEAD(&c->old_buds); + INIT_LIST_HEAD(&c->orph_list); + INIT_LIST_HEAD(&c->orph_new); + c->no_chk_data_crc = 1; + + c->highest_inum = UBIFS_FIRST_INO; + c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM; + + ubi_get_volume_info(ubi, &c->vi); + ubi_get_device_info(c->vi.ubi_num, &c->di); + } + return c; +} - spin_lock_init(&c->cnt_lock); - spin_lock_init(&c->cs_lock); - spin_lock_init(&c->buds_lock); - spin_lock_init(&c->space_lock); - spin_lock_init(&c->orphan_lock); - init_rwsem(&c->commit_sem); - mutex_init(&c->lp_mutex); - mutex_init(&c->tnc_mutex); - mutex_init(&c->log_mutex); - mutex_init(&c->mst_mutex); - mutex_init(&c->umount_mutex); - mutex_init(&c->bu_mutex); - mutex_init(&c->write_reserve_mutex); - init_waitqueue_head(&c->cmt_wq); - c->buds = RB_ROOT; - c->old_idx = RB_ROOT; - c->size_tree = RB_ROOT; - c->orph_tree = RB_ROOT; - INIT_LIST_HEAD(&c->infos_list); - INIT_LIST_HEAD(&c->idx_gc); - INIT_LIST_HEAD(&c->replay_list); - INIT_LIST_HEAD(&c->replay_buds); - INIT_LIST_HEAD(&c->uncat_list); - INIT_LIST_HEAD(&c->empty_list); - INIT_LIST_HEAD(&c->freeable_list); - INIT_LIST_HEAD(&c->frdi_idx_list); - INIT_LIST_HEAD(&c->unclean_leb_list); - INIT_LIST_HEAD(&c->old_buds); - INIT_LIST_HEAD(&c->orph_list); - INIT_LIST_HEAD(&c->orph_new); - c->no_chk_data_crc = 1; +static int ubifs_fill_super(struct super_block *sb, void *data, int silent) +{ + struct ubifs_info *c = sb->s_fs_info; + struct inode *root; + int err; c->vfs_sb = sb; - c->highest_inum = UBIFS_FIRST_INO; - c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM; - - ubi_get_volume_info(ubi, &c->vi); - ubi_get_device_info(c->vi.ubi_num, &c->di); - /* Re-open the UBI device in read-write mode */ c->ubi = ubi_open_volume(c->vi.ubi_num, c->vi.vol_id, UBI_READWRITE); if (IS_ERR(c->ubi)) { err = PTR_ERR(c->ubi); - goto out_free; + goto out; } /* @@ -2071,24 +2094,29 @@ out_bdi: bdi_destroy(&c->bdi); out_close: ubi_close_volume(c->ubi); -out_free: - kfree(c); +out: return err; } static int sb_test(struct super_block *sb, void *data) { - dev_t *dev = data; + struct ubifs_info *c1 = data; struct ubifs_info *c = sb->s_fs_info; - return c->vi.cdev == *dev; + return c->vi.cdev == c1->vi.cdev; +} + +static int sb_set(struct super_block *sb, void *data) +{ + sb->s_fs_info = data; + return set_anon_super(sb, NULL); } static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, const char *name, void *data) { struct ubi_volume_desc *ubi; - struct ubi_volume_info vi; + struct ubifs_info *c; struct super_block *sb; int err; @@ -2105,19 +2133,25 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, name, (int)PTR_ERR(ubi)); return ERR_CAST(ubi); } - ubi_get_volume_info(ubi, &vi); - dbg_gen("opened ubi%d_%d", vi.ubi_num, vi.vol_id); + c = alloc_ubifs_info(ubi); + if (!c) { + err = -ENOMEM; + goto out_close; + } + + dbg_gen("opened ubi%d_%d", c->vi.ubi_num, c->vi.vol_id); - sb = sget(fs_type, &sb_test, &set_anon_super, &vi.cdev); + sb = sget(fs_type, sb_test, sb_set, c); if (IS_ERR(sb)) { err = PTR_ERR(sb); + kfree(c); goto out_close; } if (sb->s_root) { struct ubifs_info *c1 = sb->s_fs_info; - + kfree(c); /* A new mount point for already mounted UBIFS */ dbg_gen("this ubi volume is already mounted"); if (!!(flags & MS_RDONLY) != c1->ro_mount) { @@ -2126,11 +2160,6 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, } } else { sb->s_flags = flags; - /* - * Pass 'ubi' to 'fill_super()' in sb->s_fs_info where it is - * replaced by 'c'. - */ - sb->s_fs_info = ubi; err = ubifs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) goto out_deact; @@ -2150,11 +2179,18 @@ out_close: return ERR_PTR(err); } +static void kill_ubifs_super(struct super_block *s) +{ + struct ubifs_info *c = s->s_fs_info; + kill_anon_super(s); + kfree(c); +} + static struct file_system_type ubifs_fs_type = { .name = "ubifs", .owner = THIS_MODULE, .mount = ubifs_mount, - .kill_sb = kill_anon_super, + .kill_sb = kill_ubifs_super, }; /* diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index de48597..91b4213 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -2557,11 +2557,11 @@ int ubifs_tnc_remove_nm(struct ubifs_info *c, const union ubifs_key *key, if (err) { /* Ensure the znode is dirtied */ if (znode->cnext || !ubifs_zn_dirty(znode)) { - znode = dirty_cow_bottom_up(c, znode); - if (IS_ERR(znode)) { - err = PTR_ERR(znode); - goto out_unlock; - } + znode = dirty_cow_bottom_up(c, znode); + if (IS_ERR(znode)) { + err = PTR_ERR(znode); + goto out_unlock; + } } err = tnc_delete(c, znode, n); } @@ -2876,12 +2876,13 @@ static void tnc_destroy_cnext(struct ubifs_info *c) */ void ubifs_tnc_close(struct ubifs_info *c) { - long clean_freed; - tnc_destroy_cnext(c); if (c->zroot.znode) { - clean_freed = ubifs_destroy_tnc_subtree(c->zroot.znode); - atomic_long_sub(clean_freed, &ubifs_clean_zn_cnt); + long n; + + ubifs_destroy_tnc_subtree(c->zroot.znode); + n = atomic_long_read(&c->clean_zn_cnt); + atomic_long_sub(n, &ubifs_clean_zn_cnt); } kfree(c->gap_lebs); kfree(c->ilebs); diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c index 53288e5..41920f3 100644 --- a/fs/ubifs/tnc_commit.c +++ b/fs/ubifs/tnc_commit.c @@ -377,15 +377,13 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt) c->gap_lebs = NULL; return err; } - if (!dbg_force_in_the_gaps_enabled) { + if (dbg_force_in_the_gaps_enabled()) { /* * Do not print scary warnings if the debugging * option which forces in-the-gaps is enabled. */ - ubifs_err("out of space"); - spin_lock(&c->space_lock); - dbg_dump_budg(c); - spin_unlock(&c->space_lock); + ubifs_warn("out of space"); + dbg_dump_budg(c, &c->bi); dbg_dump_lprops(c); } /* Try to commit anyway */ @@ -796,16 +794,16 @@ int ubifs_tnc_start_commit(struct ubifs_info *c, struct ubifs_zbranch *zroot) spin_lock(&c->space_lock); /* * Although we have not finished committing yet, update size of the - * committed index ('c->old_idx_sz') and zero out the index growth + * committed index ('c->bi.old_idx_sz') and zero out the index growth * budget. It is OK to do this now, because we've reserved all the * space which is needed to commit the index, and it is save for the * budgeting subsystem to assume the index is already committed, * even though it is not. */ - ubifs_assert(c->min_idx_lebs == ubifs_calc_min_idx_lebs(c)); - c->old_idx_sz = c->calc_idx_sz; - c->budg_uncommitted_idx = 0; - c->min_idx_lebs = ubifs_calc_min_idx_lebs(c); + ubifs_assert(c->bi.min_idx_lebs == ubifs_calc_min_idx_lebs(c)); + c->bi.old_idx_sz = c->calc_idx_sz; + c->bi.uncommitted_idx = 0; + c->bi.min_idx_lebs = ubifs_calc_min_idx_lebs(c); spin_unlock(&c->space_lock); mutex_unlock(&c->tnc_mutex); diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h index 191ca78..e24380c 100644 --- a/fs/ubifs/ubifs-media.h +++ b/fs/ubifs/ubifs-media.h @@ -408,9 +408,11 @@ enum { * Superblock flags. * * UBIFS_FLG_BIGLPT: if "big" LPT model is used if set + * UBIFS_FLG_SPACE_FIXUP: first-mount "fixup" of free space within LEBs needed */ enum { UBIFS_FLG_BIGLPT = 0x02, + UBIFS_FLG_SPACE_FIXUP = 0x04, }; /** @@ -434,7 +436,7 @@ struct ubifs_ch { __u8 node_type; __u8 group_type; __u8 padding[2]; -} __attribute__ ((packed)); +} __packed; /** * union ubifs_dev_desc - device node descriptor. @@ -448,7 +450,7 @@ struct ubifs_ch { union ubifs_dev_desc { __le32 new; __le64 huge; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_ino_node - inode node. @@ -509,7 +511,7 @@ struct ubifs_ino_node { __le16 compr_type; __u8 padding2[26]; /* Watch 'zero_ino_node_unused()' if changing! */ __u8 data[]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_dent_node - directory entry node. @@ -534,7 +536,7 @@ struct ubifs_dent_node { __le16 nlen; __u8 padding2[4]; /* Watch 'zero_dent_node_unused()' if changing! */ __u8 name[]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_data_node - data node. @@ -555,7 +557,7 @@ struct ubifs_data_node { __le16 compr_type; __u8 padding[2]; /* Watch 'zero_data_node_unused()' if changing! */ __u8 data[]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_trun_node - truncation node. @@ -575,7 +577,7 @@ struct ubifs_trun_node { __u8 padding[12]; /* Watch 'zero_trun_node_unused()' if changing! */ __le64 old_size; __le64 new_size; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_pad_node - padding node. @@ -586,7 +588,7 @@ struct ubifs_trun_node { struct ubifs_pad_node { struct ubifs_ch ch; __le32 pad_len; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_sb_node - superblock node. @@ -644,7 +646,7 @@ struct ubifs_sb_node { __u8 uuid[16]; __le32 ro_compat_version; __u8 padding2[3968]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_mst_node - master node. @@ -711,7 +713,7 @@ struct ubifs_mst_node { __le32 idx_lebs; __le32 leb_cnt; __u8 padding[344]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_ref_node - logical eraseblock reference node. @@ -727,7 +729,7 @@ struct ubifs_ref_node { __le32 offs; __le32 jhead; __u8 padding[28]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_branch - key/reference/length branch @@ -741,7 +743,7 @@ struct ubifs_branch { __le32 offs; __le32 len; __u8 key[]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_idx_node - indexing node. @@ -755,7 +757,7 @@ struct ubifs_idx_node { __le16 child_cnt; __le16 level; __u8 branches[]; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_cs_node - commit start node. @@ -765,7 +767,7 @@ struct ubifs_idx_node { struct ubifs_cs_node { struct ubifs_ch ch; __le64 cmt_no; -} __attribute__ ((packed)); +} __packed; /** * struct ubifs_orph_node - orphan node. @@ -777,6 +779,6 @@ struct ubifs_orph_node { struct ubifs_ch ch; __le64 cmt_no; __le64 inos[]; -} __attribute__ ((packed)); +} __packed; #endif /* __UBIFS_MEDIA_H__ */ diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index 8c40ad3..f79983d 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -389,9 +389,9 @@ struct ubifs_gced_idx_leb { * The @ui_size is a "shadow" variable for @inode->i_size and UBIFS uses * @ui_size instead of @inode->i_size. The reason for this is that UBIFS cannot * make sure @inode->i_size is always changed under @ui_mutex, because it - * cannot call 'truncate_setsize()' with @ui_mutex locked, because it would deadlock - * with 'ubifs_writepage()' (see file.c). All the other inode fields are - * changed under @ui_mutex, so they do not need "shadow" fields. Note, one + * cannot call 'truncate_setsize()' with @ui_mutex locked, because it would + * deadlock with 'ubifs_writepage()' (see file.c). All the other inode fields + * are changed under @ui_mutex, so they do not need "shadow" fields. Note, one * could consider to rework locking and base it on "shadow" fields. */ struct ubifs_inode { @@ -722,12 +722,14 @@ struct ubifs_bud { * struct ubifs_jhead - journal head. * @wbuf: head's write-buffer * @buds_list: list of bud LEBs belonging to this journal head + * @grouped: non-zero if UBIFS groups nodes when writing to this journal head * * Note, the @buds list is protected by the @c->buds_lock. */ struct ubifs_jhead { struct ubifs_wbuf wbuf; struct list_head buds_list; + unsigned int grouped:1; }; /** @@ -937,6 +939,40 @@ struct ubifs_mount_opts { unsigned int compr_type:2; }; +/** + * struct ubifs_budg_info - UBIFS budgeting information. + * @idx_growth: amount of bytes budgeted for index growth + * @data_growth: amount of bytes budgeted for cached data + * @dd_growth: amount of bytes budgeted for cached data that will make + * other data dirty + * @uncommitted_idx: amount of bytes were budgeted for growth of the index, but + * which still have to be taken into account because the index + * has not been committed so far + * @old_idx_sz: size of index on flash + * @min_idx_lebs: minimum number of LEBs required for the index + * @nospace: non-zero if the file-system does not have flash space (used as + * optimization) + * @nospace_rp: the same as @nospace, but additionally means that even reserved + * pool is full + * @page_budget: budget for a page (constant, nenver changed after mount) + * @inode_budget: budget for an inode (constant, nenver changed after mount) + * @dent_budget: budget for a directory entry (constant, nenver changed after + * mount) + */ +struct ubifs_budg_info { + long long idx_growth; + long long data_growth; + long long dd_growth; + long long uncommitted_idx; + unsigned long long old_idx_sz; + int min_idx_lebs; + unsigned int nospace:1; + unsigned int nospace_rp:1; + int page_budget; + int inode_budget; + int dent_budget; +}; + struct ubifs_debug_info; /** @@ -980,6 +1016,7 @@ struct ubifs_debug_info; * @cmt_wq: wait queue to sleep on if the log is full and a commit is running * * @big_lpt: flag that LPT is too big to write whole during commit + * @space_fixup: flag indicating that free space in LEBs needs to be cleaned up * @no_chk_data_crc: do not check CRCs when reading data nodes (except during * recovery) * @bulk_read: enable bulk-reads @@ -1057,32 +1094,14 @@ struct ubifs_debug_info; * @dirty_zn_cnt: number of dirty znodes * @clean_zn_cnt: number of clean znodes * - * @budg_idx_growth: amount of bytes budgeted for index growth - * @budg_data_growth: amount of bytes budgeted for cached data - * @budg_dd_growth: amount of bytes budgeted for cached data that will make - * other data dirty - * @budg_uncommitted_idx: amount of bytes were budgeted for growth of the index, - * but which still have to be taken into account because - * the index has not been committed so far - * @space_lock: protects @budg_idx_growth, @budg_data_growth, @budg_dd_growth, - * @budg_uncommited_idx, @min_idx_lebs, @old_idx_sz, @lst, - * @nospace, and @nospace_rp; - * @min_idx_lebs: minimum number of LEBs required for the index - * @old_idx_sz: size of index on flash + * @space_lock: protects @bi and @lst + * @lst: lprops statistics + * @bi: budgeting information * @calc_idx_sz: temporary variable which is used to calculate new index size * (contains accurate new index size at end of TNC commit start) - * @lst: lprops statistics - * @nospace: non-zero if the file-system does not have flash space (used as - * optimization) - * @nospace_rp: the same as @nospace, but additionally means that even reserved - * pool is full - * - * @page_budget: budget for a page - * @inode_budget: budget for an inode - * @dent_budget: budget for a directory entry * * @ref_node_alsz: size of the LEB reference node aligned to the min. flash - * I/O unit + * I/O unit * @mst_node_alsz: master node aligned size * @min_idx_node_sz: minimum indexing node aligned on 8-bytes boundary * @max_idx_node_sz: maximum indexing node aligned on 8-bytes boundary @@ -1189,7 +1208,6 @@ struct ubifs_debug_info; * @replaying: %1 during journal replay * @mounting: %1 while mounting * @remounting_rw: %1 while re-mounting from R/O mode to R/W mode - * @replay_tree: temporary tree used during journal replay * @replay_list: temporary list used during journal replay * @replay_buds: list of buds to replay * @cs_sqnum: sequence number of first node in the log (commit start node) @@ -1238,6 +1256,7 @@ struct ubifs_info { wait_queue_head_t cmt_wq; unsigned int big_lpt:1; + unsigned int space_fixup:1; unsigned int no_chk_data_crc:1; unsigned int bulk_read:1; unsigned int default_compr:2; @@ -1308,21 +1327,10 @@ struct ubifs_info { atomic_long_t dirty_zn_cnt; atomic_long_t clean_zn_cnt; - long long budg_idx_growth; - long long budg_data_growth; - long long budg_dd_growth; - long long budg_uncommitted_idx; spinlock_t space_lock; - int min_idx_lebs; - unsigned long long old_idx_sz; - unsigned long long calc_idx_sz; struct ubifs_lp_stats lst; - unsigned int nospace:1; - unsigned int nospace_rp:1; - - int page_budget; - int inode_budget; - int dent_budget; + struct ubifs_budg_info bi; + unsigned long long calc_idx_sz; int ref_node_alsz; int mst_node_alsz; @@ -1430,7 +1438,6 @@ struct ubifs_info { unsigned int replaying:1; unsigned int mounting:1; unsigned int remounting_rw:1; - struct rb_root replay_tree; struct list_head replay_list; struct list_head replay_buds; unsigned long long cs_sqnum; @@ -1609,7 +1616,7 @@ int ubifs_tnc_start_commit(struct ubifs_info *c, struct ubifs_zbranch *zroot); int ubifs_tnc_end_commit(struct ubifs_info *c); /* shrinker.c */ -int ubifs_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask); +int ubifs_shrinker(struct shrinker *shrink, struct shrink_control *sc); /* commit.c */ int ubifs_bg_thread(void *info); @@ -1628,6 +1635,7 @@ int ubifs_write_master(struct ubifs_info *c); int ubifs_read_superblock(struct ubifs_info *c); struct ubifs_sb_node *ubifs_read_sb_node(struct ubifs_info *c); int ubifs_write_sb_node(struct ubifs_info *c, struct ubifs_sb_node *sup); +int ubifs_fixup_free_space(struct ubifs_info *c); /* replay.c */ int ubifs_validate_entry(struct ubifs_info *c, @@ -1736,7 +1744,7 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum); int ubifs_recover_master_node(struct ubifs_info *c); int ubifs_write_rcvrd_mst_node(struct ubifs_info *c); struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, - int offs, void *sbuf, int grouped); + int offs, void *sbuf, int jhead); struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf); int ubifs_recover_inl_heads(const struct ubifs_info *c, void *sbuf); diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index 3299f46..16f19f5 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -80,8 +80,8 @@ enum { SECURITY_XATTR, }; -static const struct inode_operations none_inode_operations; -static const struct file_operations none_file_operations; +static const struct inode_operations empty_iops; +static const struct file_operations empty_fops; /** * create_xattr - create an extended attribute. @@ -131,8 +131,8 @@ static int create_xattr(struct ubifs_info *c, struct inode *host, /* Re-define all operations to be "nothing" */ inode->i_mapping->a_ops = &empty_aops; - inode->i_op = &none_inode_operations; - inode->i_fop = &none_file_operations; + inode->i_op = &empty_iops; + inode->i_fop = &empty_fops; inode->i_flags |= S_SYNC | S_NOATIME | S_NOCMTIME | S_NOQUOTA; ui = ubifs_inode(inode); diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index 46f7a80..42694e1 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -424,8 +424,7 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, ufs_cpu_to_data_ptr(sb, p, result); *err = 0; UFS_I(inode)->i_lastfrag = - max_t(u32, UFS_I(inode)->i_lastfrag, - fragment + count); + max(UFS_I(inode)->i_lastfrag, fragment + count); ufs_clear_frags(inode, result + oldcount, newcount - oldcount, locked_page != NULL); } @@ -440,7 +439,8 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, result = ufs_add_fragments (inode, tmp, oldcount, newcount, err); if (result) { *err = 0; - UFS_I(inode)->i_lastfrag = max_t(u32, UFS_I(inode)->i_lastfrag, fragment + count); + UFS_I(inode)->i_lastfrag = max(UFS_I(inode)->i_lastfrag, + fragment + count); ufs_clear_frags(inode, result + oldcount, newcount - oldcount, locked_page != NULL); unlock_super(sb); @@ -479,7 +479,8 @@ u64 ufs_new_fragments(struct inode *inode, void *p, u64 fragment, uspi->s_sbbase + result, locked_page); ufs_cpu_to_data_ptr(sb, p, result); *err = 0; - UFS_I(inode)->i_lastfrag = max_t(u32, UFS_I(inode)->i_lastfrag, fragment + count); + UFS_I(inode)->i_lastfrag = max(UFS_I(inode)->i_lastfrag, + fragment + count); unlock_super(sb); if (newcount < request) ufs_free_fragments (inode, result + newcount, request - newcount); diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index e765743..b4d791a 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -409,7 +409,7 @@ out: } /** - * ufs_getfrag_bloc() - `get_block_t' function, interface between UFS and + * ufs_getfrag_block() - `get_block_t' function, interface between UFS and * readpage, writepage and so on */ diff --git a/fs/ufs/truncate.c b/fs/ufs/truncate.c index 5f821db..f04f89f 100644 --- a/fs/ufs/truncate.c +++ b/fs/ufs/truncate.c @@ -84,7 +84,7 @@ static int ufs_trunc_direct(struct inode *inode) retry = 0; frag1 = DIRECT_FRAGMENT; - frag4 = min_t(u32, UFS_NDIR_FRAGMENT, ufsi->i_lastfrag); + frag4 = min_t(u64, UFS_NDIR_FRAGMENT, ufsi->i_lastfrag); frag2 = ((frag1 & uspi->s_fpbmask) ? ((frag1 | uspi->s_fpbmask) + 1) : frag1); frag3 = frag4 & ~uspi->s_fpbmask; block1 = block2 = 0; @@ -46,18 +46,22 @@ xattr_permission(struct inode *inode, const char *name, int mask) return 0; /* - * The trusted.* namespace can only be accessed by a privileged user. + * The trusted.* namespace can only be accessed by privileged users. */ - if (!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN)) - return (capable(CAP_SYS_ADMIN) ? 0 : -EPERM); + if (!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN)) { + if (!capable(CAP_SYS_ADMIN)) + return (mask & MAY_WRITE) ? -EPERM : -ENODATA; + return 0; + } - /* In user.* namespace, only regular files and directories can have + /* + * In the user.* namespace, only regular files and directories can have * extended attributes. For sticky directories, only the owner and - * privileged user can write attributes. + * privileged users can write attributes. */ if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN)) { if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode)) - return -EPERM; + return (mask & MAY_WRITE) ? -EPERM : -ENODATA; if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) && (mask & MAY_WRITE) && !inode_owner_or_capable(inode)) return -EPERM; @@ -87,7 +91,11 @@ int __vfs_setxattr_noperm(struct dentry *dentry, const char *name, { struct inode *inode = dentry->d_inode; int error = -EOPNOTSUPP; + int issec = !strncmp(name, XATTR_SECURITY_PREFIX, + XATTR_SECURITY_PREFIX_LEN); + if (issec) + inode->i_flags &= ~S_NOSEC; if (inode->i_op->setxattr) { error = inode->i_op->setxattr(dentry, name, value, size, flags); if (!error) { @@ -95,8 +103,7 @@ int __vfs_setxattr_noperm(struct dentry *dentry, const char *name, security_inode_post_setxattr(dentry, name, value, size, flags); } - } else if (!strncmp(name, XATTR_SECURITY_PREFIX, - XATTR_SECURITY_PREFIX_LEN)) { + } else if (issec) { const char *suffix = name + XATTR_SECURITY_PREFIX_LEN; error = security_inode_setsecurity(inode, suffix, value, size, flags); diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 9ef9ed2..5e68099 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -33,7 +33,6 @@ #include <linux/migrate.h> #include <linux/backing-dev.h> #include <linux/freezer.h> -#include <linux/list_sort.h> #include "xfs_sb.h" #include "xfs_inum.h" @@ -709,6 +708,27 @@ xfs_buf_get_empty( return bp; } +/* + * Return a buffer allocated as an empty buffer and associated to external + * memory via xfs_buf_associate_memory() back to it's empty state. + */ +void +xfs_buf_set_empty( + struct xfs_buf *bp, + size_t len) +{ + if (bp->b_pages) + _xfs_buf_free_pages(bp); + + bp->b_pages = NULL; + bp->b_page_count = 0; + bp->b_addr = NULL; + bp->b_file_offset = 0; + bp->b_buffer_length = bp->b_count_desired = len; + bp->b_bn = XFS_BUF_DADDR_NULL; + bp->b_flags &= ~XBF_MAPPED; +} + static inline struct page * mem_to_page( void *addr) @@ -1402,12 +1422,12 @@ restart: int xfs_buftarg_shrink( struct shrinker *shrink, - int nr_to_scan, - gfp_t mask) + struct shrink_control *sc) { struct xfs_buftarg *btp = container_of(shrink, struct xfs_buftarg, bt_shrinker); struct xfs_buf *bp; + int nr_to_scan = sc->nr_to_scan; LIST_HEAD(dispose); if (!nr_to_scan) diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index a9a1c45..50a7d5f 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -178,6 +178,7 @@ extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t, xfs_buf_flags_t); extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *); +extern void xfs_buf_set_empty(struct xfs_buf *bp, size_t len); extern xfs_buf_t *xfs_buf_get_uncached(struct xfs_buftarg *, size_t, int); extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t); extern void xfs_buf_hold(xfs_buf_t *); diff --git a/fs/xfs/linux-2.6/xfs_discard.c b/fs/xfs/linux-2.6/xfs_discard.c index d61611c..244e797 100644 --- a/fs/xfs/linux-2.6/xfs_discard.c +++ b/fs/xfs/linux-2.6/xfs_discard.c @@ -191,3 +191,32 @@ xfs_ioc_trim( return -XFS_ERROR(EFAULT); return 0; } + +int +xfs_discard_extents( + struct xfs_mount *mp, + struct list_head *list) +{ + struct xfs_busy_extent *busyp; + int error = 0; + + list_for_each_entry(busyp, list, list) { + trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, + busyp->length); + + error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, + XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), + XFS_FSB_TO_BB(mp, busyp->length), + GFP_NOFS, 0); + if (error && error != EOPNOTSUPP) { + xfs_info(mp, + "discard failed for extent [0x%llu,%u], error %d", + (unsigned long long)busyp->bno, + busyp->length, + error); + return error; + } + } + + return 0; +} diff --git a/fs/xfs/linux-2.6/xfs_discard.h b/fs/xfs/linux-2.6/xfs_discard.h index e82b6dd..344879a 100644 --- a/fs/xfs/linux-2.6/xfs_discard.h +++ b/fs/xfs/linux-2.6/xfs_discard.h @@ -2,7 +2,9 @@ #define XFS_DISCARD_H 1 struct fstrim_range; +struct list_head; extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *); +extern int xfs_discard_extents(struct xfs_mount *, struct list_head *); #endif /* XFS_DISCARD_H */ diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index f4213ba..7f782af 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -131,19 +131,34 @@ xfs_file_fsync( { struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; int error = 0; int log_flushed = 0; trace_xfs_file_fsync(ip); - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + if (XFS_FORCED_SHUTDOWN(mp)) return -XFS_ERROR(EIO); xfs_iflags_clear(ip, XFS_ITRUNCATED); xfs_ioend_wait(ip); + if (mp->m_flags & XFS_MOUNT_BARRIER) { + /* + * If we have an RT and/or log subvolume we need to make sure + * to flush the write cache the device used for file data + * first. This is to ensure newly written file data make + * it to disk before logging the new inode size in case of + * an extending write. + */ + if (XFS_IS_REALTIME_INODE(ip)) + xfs_blkdev_issue_flush(mp->m_rtdev_targp); + else if (mp->m_logdev_targp != mp->m_ddev_targp) + xfs_blkdev_issue_flush(mp->m_ddev_targp); + } + /* * We always need to make sure that the required inode state is safe on * disk. The inode might be clean but we still might need to force the @@ -175,9 +190,9 @@ xfs_file_fsync( * updates. The sync transaction will also force the log. */ xfs_iunlock(ip, XFS_ILOCK_SHARED); - tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS); + tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); error = xfs_trans_reserve(tp, 0, - XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0); + XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); if (error) { xfs_trans_cancel(tp, 0); return -error; @@ -209,28 +224,25 @@ xfs_file_fsync( * force the log. */ if (xfs_ipincount(ip)) { - error = _xfs_log_force_lsn(ip->i_mount, + error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn, XFS_LOG_SYNC, &log_flushed); } xfs_iunlock(ip, XFS_ILOCK_SHARED); } - if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) { - /* - * If the log write didn't issue an ordered tag we need - * to flush the disk cache for the data device now. - */ - if (!log_flushed) - xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp); - - /* - * If this inode is on the RT dev we need to flush that - * cache as well. - */ - if (XFS_IS_REALTIME_INODE(ip)) - xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp); - } + /* + * If we only have a single device, and the log force about was + * a no-op we might have to flush the data device cache here. + * This can only happen for fdatasync/O_DSYNC if we were overwriting + * an already allocated file and thus do not have any metadata to + * commit. + */ + if ((mp->m_flags & XFS_MOUNT_BARRIER) && + mp->m_logdev_targp == mp->m_ddev_targp && + !XFS_IS_REALTIME_INODE(ip) && + !log_flushed) + xfs_blkdev_issue_flush(mp->m_ddev_targp); return -error; } diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c index b3486df..54e623b 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ b/fs/xfs/linux-2.6/xfs_ioctl32.c @@ -586,7 +586,8 @@ xfs_file_compat_ioctl( case XFS_IOC_RESVSP_32: case XFS_IOC_UNRESVSP_32: case XFS_IOC_RESVSP64_32: - case XFS_IOC_UNRESVSP64_32: { + case XFS_IOC_UNRESVSP64_32: + case XFS_IOC_ZERO_RANGE_32: { struct xfs_flock64 bf; if (xfs_compat_flock64_copyin(&bf, arg)) diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.h b/fs/xfs/linux-2.6/xfs_ioctl32.h index 08b6057..80f4060 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.h +++ b/fs/xfs/linux-2.6/xfs_ioctl32.h @@ -184,6 +184,7 @@ typedef struct compat_xfs_flock64 { #define XFS_IOC_UNRESVSP_32 _IOW('X', 41, struct compat_xfs_flock64) #define XFS_IOC_RESVSP64_32 _IOW('X', 42, struct compat_xfs_flock64) #define XFS_IOC_UNRESVSP64_32 _IOW('X', 43, struct compat_xfs_flock64) +#define XFS_IOC_ZERO_RANGE_32 _IOW('X', 57, struct compat_xfs_flock64) typedef struct compat_xfs_fsop_geom_v1 { __u32 blocksize; /* filesystem (data) block size */ diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index dd21784..d44d92c 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -182,7 +182,7 @@ xfs_vn_mknod( if (IS_POSIXACL(dir)) { default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT); if (IS_ERR(default_acl)) - return -PTR_ERR(default_acl); + return PTR_ERR(default_acl); if (!default_acl) mode &= ~current_umask(); diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 244be9c..8633521 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -70,6 +70,7 @@ #include <linux/ctype.h> #include <linux/writeback.h> #include <linux/capability.h> +#include <linux/list_sort.h> #include <asm/page.h> #include <asm/div64.h> diff --git a/fs/xfs/linux-2.6/xfs_message.c b/fs/xfs/linux-2.6/xfs_message.c index 9f76cce..bd672de 100644 --- a/fs/xfs/linux-2.6/xfs_message.c +++ b/fs/xfs/linux-2.6/xfs_message.c @@ -41,23 +41,6 @@ __xfs_printk( printk("%sXFS: %pV\n", level, vaf); } -void xfs_printk( - const char *level, - const struct xfs_mount *mp, - const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - - va_start(args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - __xfs_printk(level, mp, &vaf); - va_end(args); -} - #define define_xfs_printk_level(func, kern_level) \ void func(const struct xfs_mount *mp, const char *fmt, ...) \ { \ @@ -95,8 +78,7 @@ xfs_alert_tag( int do_panic = 0; if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) { - xfs_printk(KERN_ALERT, mp, - "XFS: Transforming an alert into a BUG."); + xfs_alert(mp, "Transforming an alert into a BUG."); do_panic = 1; } diff --git a/fs/xfs/linux-2.6/xfs_message.h b/fs/xfs/linux-2.6/xfs_message.h index f1b3fc1..7fb7ea0 100644 --- a/fs/xfs/linux-2.6/xfs_message.h +++ b/fs/xfs/linux-2.6/xfs_message.h @@ -3,9 +3,6 @@ struct xfs_mount; -extern void xfs_printk(const char *level, const struct xfs_mount *mp, - const char *fmt, ...) - __attribute__ ((format (printf, 3, 4))); extern void xfs_emerg(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); extern void xfs_alert(const struct xfs_mount *mp, const char *fmt, ...) @@ -28,7 +25,9 @@ extern void xfs_info(const struct xfs_mount *mp, const char *fmt, ...) extern void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); #else -static inline void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) +static inline void +__attribute__ ((format (printf, 2, 3))) +xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) { } #endif diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index b38e58d..a1a881e 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -110,8 +110,10 @@ mempool_t *xfs_ioend_pool; #define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */ #define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */ #define MNTOPT_QUOTANOENF "qnoenforce" /* same as uqnoenforce */ -#define MNTOPT_DELAYLOG "delaylog" /* Delayed loging enabled */ -#define MNTOPT_NODELAYLOG "nodelaylog" /* Delayed loging disabled */ +#define MNTOPT_DELAYLOG "delaylog" /* Delayed logging enabled */ +#define MNTOPT_NODELAYLOG "nodelaylog" /* Delayed logging disabled */ +#define MNTOPT_DISCARD "discard" /* Discard unused blocks */ +#define MNTOPT_NODISCARD "nodiscard" /* Do not discard unused blocks */ /* * Table driven mount option parser. @@ -355,6 +357,10 @@ xfs_parseargs( mp->m_flags |= XFS_MOUNT_DELAYLOG; } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { mp->m_flags &= ~XFS_MOUNT_DELAYLOG; + } else if (!strcmp(this_char, MNTOPT_DISCARD)) { + mp->m_flags |= XFS_MOUNT_DISCARD; + } else if (!strcmp(this_char, MNTOPT_NODISCARD)) { + mp->m_flags &= ~XFS_MOUNT_DISCARD; } else if (!strcmp(this_char, "ihashsize")) { xfs_warn(mp, "ihashsize no longer used, option is deprecated."); @@ -388,6 +394,13 @@ xfs_parseargs( return EINVAL; } + if ((mp->m_flags & XFS_MOUNT_DISCARD) && + !(mp->m_flags & XFS_MOUNT_DELAYLOG)) { + xfs_warn(mp, + "the discard option is incompatible with the nodelaylog option"); + return EINVAL; + } + #ifndef CONFIG_XFS_QUOTA if (XFS_IS_QUOTA_RUNNING(mp)) { xfs_warn(mp, "quota support not available in this kernel."); @@ -488,6 +501,7 @@ xfs_showargs( { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, { XFS_MOUNT_DELAYLOG, "," MNTOPT_DELAYLOG }, + { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, { 0, NULL } }; static struct proc_xfs_info xfs_info_unset[] = { @@ -613,68 +627,6 @@ xfs_blkdev_put( blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); } -/* - * Try to write out the superblock using barriers. - */ -STATIC int -xfs_barrier_test( - xfs_mount_t *mp) -{ - xfs_buf_t *sbp = xfs_getsb(mp, 0); - int error; - - XFS_BUF_UNDONE(sbp); - XFS_BUF_UNREAD(sbp); - XFS_BUF_UNDELAYWRITE(sbp); - XFS_BUF_WRITE(sbp); - XFS_BUF_UNASYNC(sbp); - XFS_BUF_ORDERED(sbp); - - xfsbdstrat(mp, sbp); - error = xfs_buf_iowait(sbp); - - /* - * Clear all the flags we set and possible error state in the - * buffer. We only did the write to try out whether barriers - * worked and shouldn't leave any traces in the superblock - * buffer. - */ - XFS_BUF_DONE(sbp); - XFS_BUF_ERROR(sbp, 0); - XFS_BUF_UNORDERED(sbp); - - xfs_buf_relse(sbp); - return error; -} - -STATIC void -xfs_mountfs_check_barriers(xfs_mount_t *mp) -{ - int error; - - if (mp->m_logdev_targp != mp->m_ddev_targp) { - xfs_notice(mp, - "Disabling barriers, not supported with external log device"); - mp->m_flags &= ~XFS_MOUNT_BARRIER; - return; - } - - if (xfs_readonly_buftarg(mp->m_ddev_targp)) { - xfs_notice(mp, - "Disabling barriers, underlying device is readonly"); - mp->m_flags &= ~XFS_MOUNT_BARRIER; - return; - } - - error = xfs_barrier_test(mp); - if (error) { - xfs_notice(mp, - "Disabling barriers, trial barrier write failed"); - mp->m_flags &= ~XFS_MOUNT_BARRIER; - return; - } -} - void xfs_blkdev_issue_flush( xfs_buftarg_t *buftarg) @@ -911,7 +863,8 @@ xfs_fs_inode_init_once( */ STATIC void xfs_fs_dirty_inode( - struct inode *inode) + struct inode *inode, + int flags) { barrier(); XFS_I(inode)->i_update_core = 1; @@ -1225,14 +1178,6 @@ xfs_fs_remount( switch (token) { case Opt_barrier: mp->m_flags |= XFS_MOUNT_BARRIER; - - /* - * Test if barriers are actually working if we can, - * else delay this check until the filesystem is - * marked writeable. - */ - if (!(mp->m_flags & XFS_MOUNT_RDONLY)) - xfs_mountfs_check_barriers(mp); break; case Opt_nobarrier: mp->m_flags &= ~XFS_MOUNT_BARRIER; @@ -1267,8 +1212,6 @@ xfs_fs_remount( /* ro -> rw */ if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) { mp->m_flags &= ~XFS_MOUNT_RDONLY; - if (mp->m_flags & XFS_MOUNT_BARRIER) - xfs_mountfs_check_barriers(mp); /* * If this is the first remount to writeable state we @@ -1450,9 +1393,6 @@ xfs_fs_fill_super( if (error) goto out_free_sb; - if (mp->m_flags & XFS_MOUNT_BARRIER) - xfs_mountfs_check_barriers(mp); - error = xfs_filestream_mount(mp); if (error) goto out_free_sb; @@ -1787,10 +1727,6 @@ init_xfs_fs(void) if (error) goto out_cleanup_procfs; - error = xfs_init_workqueues(); - if (error) - goto out_sysctl_unregister; - vfs_initquota(); error = register_filesystem(&xfs_fs_type); diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 3e898a4..8ecad5f 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -267,6 +267,16 @@ xfs_sync_inode_attr( error = xfs_iflush(ip, flags); + /* + * We don't want to try again on non-blocking flushes that can't run + * again immediately. If an inode really must be written, then that's + * what the SYNC_WAIT flag is for. + */ + if (error == EAGAIN) { + ASSERT(!(flags & SYNC_WAIT)); + error = 0; + } + out_unlock: xfs_iunlock(ip, XFS_ILOCK_SHARED); return error; @@ -1022,13 +1032,14 @@ xfs_reclaim_inodes( static int xfs_reclaim_inode_shrink( struct shrinker *shrink, - int nr_to_scan, - gfp_t gfp_mask) + struct shrink_control *sc) { struct xfs_mount *mp; struct xfs_perag *pag; xfs_agnumber_t ag; int reclaimable; + int nr_to_scan = sc->nr_to_scan; + gfp_t gfp_mask = sc->gfp_mask; mp = container_of(shrink, struct xfs_mount, m_inode_shrink); if (nr_to_scan) { diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h index 2d0bcb4..d48b7a5 100644 --- a/fs/xfs/linux-2.6/xfs_trace.h +++ b/fs/xfs/linux-2.6/xfs_trace.h @@ -1151,44 +1151,7 @@ TRACE_EVENT(xfs_bunmap, ); -#define XFS_BUSY_SYNC \ - { 0, "async" }, \ - { 1, "sync" } - -TRACE_EVENT(xfs_alloc_busy, - TP_PROTO(struct xfs_trans *trans, xfs_agnumber_t agno, - xfs_agblock_t agbno, xfs_extlen_t len, int sync), - TP_ARGS(trans, agno, agbno, len, sync), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(struct xfs_trans *, tp) - __field(int, tid) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, len) - __field(int, sync) - ), - TP_fast_assign( - __entry->dev = trans->t_mountp->m_super->s_dev; - __entry->tp = trans; - __entry->tid = trans->t_ticket->t_tid; - __entry->agno = agno; - __entry->agbno = agbno; - __entry->len = len; - __entry->sync = sync; - ), - TP_printk("dev %d:%d trans 0x%p tid 0x%x agno %u agbno %u len %u %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->tp, - __entry->tid, - __entry->agno, - __entry->agbno, - __entry->len, - __print_symbolic(__entry->sync, XFS_BUSY_SYNC)) - -); - -TRACE_EVENT(xfs_alloc_unbusy, +DECLARE_EVENT_CLASS(xfs_busy_class, TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, xfs_extlen_t len), TP_ARGS(mp, agno, agbno, len), @@ -1210,35 +1173,45 @@ TRACE_EVENT(xfs_alloc_unbusy, __entry->agbno, __entry->len) ); +#define DEFINE_BUSY_EVENT(name) \ +DEFINE_EVENT(xfs_busy_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ + xfs_agblock_t agbno, xfs_extlen_t len), \ + TP_ARGS(mp, agno, agbno, len)) +DEFINE_BUSY_EVENT(xfs_alloc_busy); +DEFINE_BUSY_EVENT(xfs_alloc_busy_enomem); +DEFINE_BUSY_EVENT(xfs_alloc_busy_force); +DEFINE_BUSY_EVENT(xfs_alloc_busy_reuse); +DEFINE_BUSY_EVENT(xfs_alloc_busy_clear); -#define XFS_BUSY_STATES \ - { 0, "missing" }, \ - { 1, "found" } - -TRACE_EVENT(xfs_alloc_busysearch, +TRACE_EVENT(xfs_alloc_busy_trim, TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno, xfs_extlen_t len, int found), - TP_ARGS(mp, agno, agbno, len, found), + xfs_agblock_t agbno, xfs_extlen_t len, + xfs_agblock_t tbno, xfs_extlen_t tlen), + TP_ARGS(mp, agno, agbno, len, tbno, tlen), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_agnumber_t, agno) __field(xfs_agblock_t, agbno) __field(xfs_extlen_t, len) - __field(int, found) + __field(xfs_agblock_t, tbno) + __field(xfs_extlen_t, tlen) ), TP_fast_assign( __entry->dev = mp->m_super->s_dev; __entry->agno = agno; __entry->agbno = agbno; __entry->len = len; - __entry->found = found; + __entry->tbno = tbno; + __entry->tlen = tlen; ), - TP_printk("dev %d:%d agno %u agbno %u len %u %s", + TP_printk("dev %d:%d agno %u agbno %u len %u tbno %u tlen %u", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->agno, __entry->agbno, __entry->len, - __print_symbolic(__entry->found, XFS_BUSY_STATES)) + __entry->tbno, + __entry->tlen) ); TRACE_EVENT(xfs_trans_commit_lsn, @@ -1418,7 +1391,7 @@ DECLARE_EVENT_CLASS(xfs_alloc_class, __entry->wasfromfl, __entry->isfl, __entry->userdata, - __entry->firstblock) + (unsigned long long)__entry->firstblock) ) #define DEFINE_ALLOC_EVENT(name) \ @@ -1433,11 +1406,14 @@ DEFINE_ALLOC_EVENT(xfs_alloc_near_first); DEFINE_ALLOC_EVENT(xfs_alloc_near_greater); DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser); DEFINE_ALLOC_EVENT(xfs_alloc_near_error); +DEFINE_ALLOC_EVENT(xfs_alloc_near_noentry); +DEFINE_ALLOC_EVENT(xfs_alloc_near_busy); DEFINE_ALLOC_EVENT(xfs_alloc_size_neither); DEFINE_ALLOC_EVENT(xfs_alloc_size_noentry); DEFINE_ALLOC_EVENT(xfs_alloc_size_nominleft); DEFINE_ALLOC_EVENT(xfs_alloc_size_done); DEFINE_ALLOC_EVENT(xfs_alloc_size_error); +DEFINE_ALLOC_EVENT(xfs_alloc_size_busy); DEFINE_ALLOC_EVENT(xfs_alloc_small_freelist); DEFINE_ALLOC_EVENT(xfs_alloc_small_notenough); DEFINE_ALLOC_EVENT(xfs_alloc_small_done); diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 69228aa..b94dace 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -60,7 +60,7 @@ STATIC void xfs_qm_list_destroy(xfs_dqlist_t *); STATIC int xfs_qm_init_quotainos(xfs_mount_t *); STATIC int xfs_qm_init_quotainfo(xfs_mount_t *); -STATIC int xfs_qm_shake(struct shrinker *, int, gfp_t); +STATIC int xfs_qm_shake(struct shrinker *, struct shrink_control *); static struct shrinker xfs_qm_shaker = { .shrink = xfs_qm_shake, @@ -2009,10 +2009,10 @@ xfs_qm_shake_freelist( STATIC int xfs_qm_shake( struct shrinker *shrink, - int nr_to_scan, - gfp_t gfp_mask) + struct shrink_control *sc) { int ndqused, nfree, n; + gfp_t gfp_mask = sc->gfp_mask; if (!kmem_shake_allow(gfp_mask)) return 0; diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index 58632cc..6530769 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -187,7 +187,9 @@ struct xfs_busy_extent { xfs_agnumber_t agno; xfs_agblock_t bno; xfs_extlen_t length; - xlog_tid_t tid; /* transaction that created this */ + unsigned int flags; +#define XFS_ALLOC_BUSY_DISCARDED 0x01 /* undergoing a discard op. */ +#define XFS_ALLOC_BUSY_SKIP_DISCARD 0x02 /* do not discard */ }; /* diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 27d64d7..95862bb 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -41,19 +41,13 @@ #define XFSA_FIXUP_BNO_OK 1 #define XFSA_FIXUP_CNT_OK 2 -/* - * Prototypes for per-ag allocation routines - */ - STATIC int xfs_alloc_ag_vextent_exact(xfs_alloc_arg_t *); STATIC int xfs_alloc_ag_vextent_near(xfs_alloc_arg_t *); STATIC int xfs_alloc_ag_vextent_size(xfs_alloc_arg_t *); STATIC int xfs_alloc_ag_vextent_small(xfs_alloc_arg_t *, - xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *); - -/* - * Internal functions. - */ + xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *); +STATIC void xfs_alloc_busy_trim(struct xfs_alloc_arg *, + xfs_agblock_t, xfs_extlen_t, xfs_agblock_t *, xfs_extlen_t *); /* * Lookup the record equal to [bno, len] in the btree given by cur. @@ -154,19 +148,21 @@ xfs_alloc_compute_aligned( xfs_extlen_t *reslen) /* result length */ { xfs_agblock_t bno; - xfs_extlen_t diff; xfs_extlen_t len; - if (args->alignment > 1 && foundlen >= args->minlen) { - bno = roundup(foundbno, args->alignment); - diff = bno - foundbno; - len = diff >= foundlen ? 0 : foundlen - diff; + /* Trim busy sections out of found extent */ + xfs_alloc_busy_trim(args, foundbno, foundlen, &bno, &len); + + if (args->alignment > 1 && len >= args->minlen) { + xfs_agblock_t aligned_bno = roundup(bno, args->alignment); + xfs_extlen_t diff = aligned_bno - bno; + + *resbno = aligned_bno; + *reslen = diff >= len ? 0 : len - diff; } else { - bno = foundbno; - len = foundlen; + *resbno = bno; + *reslen = len; } - *resbno = bno; - *reslen = len; } /* @@ -280,7 +276,6 @@ xfs_alloc_fix_minleft( return 1; agf = XFS_BUF_TO_AGF(args->agbp); diff = be32_to_cpu(agf->agf_freeblks) - + be32_to_cpu(agf->agf_flcount) - args->len - args->minleft; if (diff >= 0) return 1; @@ -541,16 +536,8 @@ xfs_alloc_ag_vextent( if (error) return error; - /* - * Search the busylist for these blocks and mark the - * transaction as synchronous if blocks are found. This - * avoids the need to block due to a synchronous log - * force to ensure correct ordering as the synchronous - * transaction will guarantee that for us. - */ - if (xfs_alloc_busy_search(args->mp, args->agno, - args->agbno, args->len)) - xfs_trans_set_sync(args->tp); + ASSERT(!xfs_alloc_busy_search(args->mp, args->agno, + args->agbno, args->len)); } if (!args->isfl) { @@ -577,14 +564,14 @@ xfs_alloc_ag_vextent_exact( { xfs_btree_cur_t *bno_cur;/* by block-number btree cursor */ xfs_btree_cur_t *cnt_cur;/* by count btree cursor */ - xfs_agblock_t end; /* end of allocated extent */ int error; xfs_agblock_t fbno; /* start block of found extent */ - xfs_agblock_t fend; /* end block of found extent */ xfs_extlen_t flen; /* length of found extent */ + xfs_agblock_t tbno; /* start block of trimmed extent */ + xfs_extlen_t tlen; /* length of trimmed extent */ + xfs_agblock_t tend; /* end block of trimmed extent */ + xfs_agblock_t end; /* end of allocated extent */ int i; /* success/failure of operation */ - xfs_agblock_t maxend; /* end of maximal extent */ - xfs_agblock_t minend; /* end of minimal extent */ xfs_extlen_t rlen; /* length of returned extent */ ASSERT(args->alignment == 1); @@ -614,14 +601,22 @@ xfs_alloc_ag_vextent_exact( goto error0; XFS_WANT_CORRUPTED_GOTO(i == 1, error0); ASSERT(fbno <= args->agbno); - minend = args->agbno + args->minlen; - maxend = args->agbno + args->maxlen; - fend = fbno + flen; /* - * Give up if the freespace isn't long enough for the minimum request. + * Check for overlapping busy extents. */ - if (fend < minend) + xfs_alloc_busy_trim(args, fbno, flen, &tbno, &tlen); + + /* + * Give up if the start of the extent is busy, or the freespace isn't + * long enough for the minimum request. + */ + if (tbno > args->agbno) + goto not_found; + if (tlen < args->minlen) + goto not_found; + tend = tbno + tlen; + if (tend < args->agbno + args->minlen) goto not_found; /* @@ -630,14 +625,14 @@ xfs_alloc_ag_vextent_exact( * * Fix the length according to mod and prod if given. */ - end = XFS_AGBLOCK_MIN(fend, maxend); + end = XFS_AGBLOCK_MIN(tend, args->agbno + args->maxlen); args->len = end - args->agbno; xfs_alloc_fix_len(args); if (!xfs_alloc_fix_minleft(args)) goto not_found; rlen = args->len; - ASSERT(args->agbno + rlen <= fend); + ASSERT(args->agbno + rlen <= tend); end = args->agbno + rlen; /* @@ -686,11 +681,11 @@ xfs_alloc_find_best_extent( struct xfs_btree_cur **scur, /* searching cursor */ xfs_agblock_t gdiff, /* difference for search comparison */ xfs_agblock_t *sbno, /* extent found by search */ - xfs_extlen_t *slen, - xfs_extlen_t *slena, /* aligned length */ + xfs_extlen_t *slen, /* extent length */ + xfs_agblock_t *sbnoa, /* aligned extent found by search */ + xfs_extlen_t *slena, /* aligned extent length */ int dir) /* 0 = search right, 1 = search left */ { - xfs_agblock_t bno; xfs_agblock_t new; xfs_agblock_t sdiff; int error; @@ -708,16 +703,16 @@ xfs_alloc_find_best_extent( if (error) goto error0; XFS_WANT_CORRUPTED_GOTO(i == 1, error0); - xfs_alloc_compute_aligned(args, *sbno, *slen, &bno, slena); + xfs_alloc_compute_aligned(args, *sbno, *slen, sbnoa, slena); /* * The good extent is closer than this one. */ if (!dir) { - if (bno >= args->agbno + gdiff) + if (*sbnoa >= args->agbno + gdiff) goto out_use_good; } else { - if (bno <= args->agbno - gdiff) + if (*sbnoa <= args->agbno - gdiff) goto out_use_good; } @@ -729,8 +724,8 @@ xfs_alloc_find_best_extent( xfs_alloc_fix_len(args); sdiff = xfs_alloc_compute_diff(args->agbno, args->len, - args->alignment, *sbno, - *slen, &new); + args->alignment, *sbnoa, + *slena, &new); /* * Choose closer size and invalidate other cursor. @@ -780,7 +775,7 @@ xfs_alloc_ag_vextent_near( xfs_agblock_t gtbnoa; /* aligned ... */ xfs_extlen_t gtdiff; /* difference to right side entry */ xfs_extlen_t gtlen; /* length of right side entry */ - xfs_extlen_t gtlena = 0; /* aligned ... */ + xfs_extlen_t gtlena; /* aligned ... */ xfs_agblock_t gtnew; /* useful start bno of right side */ int error; /* error code */ int i; /* result code, temporary */ @@ -789,9 +784,10 @@ xfs_alloc_ag_vextent_near( xfs_agblock_t ltbnoa; /* aligned ... */ xfs_extlen_t ltdiff; /* difference to left side entry */ xfs_extlen_t ltlen; /* length of left side entry */ - xfs_extlen_t ltlena = 0; /* aligned ... */ + xfs_extlen_t ltlena; /* aligned ... */ xfs_agblock_t ltnew; /* useful start bno of left side */ xfs_extlen_t rlen; /* length of returned extent */ + int forced = 0; #if defined(DEBUG) && defined(__KERNEL__) /* * Randomly don't execute the first algorithm. @@ -800,13 +796,20 @@ xfs_alloc_ag_vextent_near( dofirst = random32() & 1; #endif + +restart: + bno_cur_lt = NULL; + bno_cur_gt = NULL; + ltlen = 0; + gtlena = 0; + ltlena = 0; + /* * Get a cursor for the by-size btree. */ cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp, args->agno, XFS_BTNUM_CNT); - ltlen = 0; - bno_cur_lt = bno_cur_gt = NULL; + /* * See if there are any free extents as big as maxlen. */ @@ -822,11 +825,13 @@ xfs_alloc_ag_vextent_near( goto error0; if (i == 0 || ltlen == 0) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); + trace_xfs_alloc_near_noentry(args); return 0; } ASSERT(i == 1); } args->wasfromfl = 0; + /* * First algorithm. * If the requested extent is large wrt the freespaces available @@ -890,7 +895,7 @@ xfs_alloc_ag_vextent_near( if (args->len < blen) continue; ltdiff = xfs_alloc_compute_diff(args->agbno, args->len, - args->alignment, ltbno, ltlen, <new); + args->alignment, ltbnoa, ltlena, <new); if (ltnew != NULLAGBLOCK && (args->len > blen || ltdiff < bdiff)) { bdiff = ltdiff; @@ -1042,11 +1047,12 @@ xfs_alloc_ag_vextent_near( args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen); xfs_alloc_fix_len(args); ltdiff = xfs_alloc_compute_diff(args->agbno, args->len, - args->alignment, ltbno, ltlen, <new); + args->alignment, ltbnoa, ltlena, <new); error = xfs_alloc_find_best_extent(args, &bno_cur_lt, &bno_cur_gt, - ltdiff, >bno, >len, >lena, + ltdiff, >bno, >len, + >bnoa, >lena, 0 /* search right */); } else { ASSERT(gtlena >= args->minlen); @@ -1057,11 +1063,12 @@ xfs_alloc_ag_vextent_near( args->len = XFS_EXTLEN_MIN(gtlena, args->maxlen); xfs_alloc_fix_len(args); gtdiff = xfs_alloc_compute_diff(args->agbno, args->len, - args->alignment, gtbno, gtlen, >new); + args->alignment, gtbnoa, gtlena, >new); error = xfs_alloc_find_best_extent(args, &bno_cur_gt, &bno_cur_lt, - gtdiff, <bno, <len, <lena, + gtdiff, <bno, <len, + <bnoa, <lena, 1 /* search left */); } @@ -1073,6 +1080,12 @@ xfs_alloc_ag_vextent_near( * If we couldn't get anything, give up. */ if (bno_cur_lt == NULL && bno_cur_gt == NULL) { + if (!forced++) { + trace_xfs_alloc_near_busy(args); + xfs_log_force(args->mp, XFS_LOG_SYNC); + goto restart; + } + trace_xfs_alloc_size_neither(args); args->agbno = NULLAGBLOCK; return 0; @@ -1107,12 +1120,13 @@ xfs_alloc_ag_vextent_near( return 0; } rlen = args->len; - (void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment, ltbno, - ltlen, <new); + (void)xfs_alloc_compute_diff(args->agbno, rlen, args->alignment, + ltbnoa, ltlena, <new); ASSERT(ltnew >= ltbno); - ASSERT(ltnew + rlen <= ltbno + ltlen); + ASSERT(ltnew + rlen <= ltbnoa + ltlena); ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length)); args->agbno = ltnew; + if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen, ltnew, rlen, XFSA_FIXUP_BNO_OK))) goto error0; @@ -1155,26 +1169,35 @@ xfs_alloc_ag_vextent_size( int i; /* temp status variable */ xfs_agblock_t rbno; /* returned block number */ xfs_extlen_t rlen; /* length of returned extent */ + int forced = 0; +restart: /* * Allocate and initialize a cursor for the by-size btree. */ cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp, args->agno, XFS_BTNUM_CNT); bno_cur = NULL; + /* * Look for an entry >= maxlen+alignment-1 blocks. */ if ((error = xfs_alloc_lookup_ge(cnt_cur, 0, args->maxlen + args->alignment - 1, &i))) goto error0; + /* - * If none, then pick up the last entry in the tree unless the - * tree is empty. + * If none or we have busy extents that we cannot allocate from, then + * we have to settle for a smaller extent. In the case that there are + * no large extents, this will return the last entry in the tree unless + * the tree is empty. In the case that there are only busy large + * extents, this will return the largest small extent unless there + * are no smaller extents available. */ - if (!i) { - if ((error = xfs_alloc_ag_vextent_small(args, cnt_cur, &fbno, - &flen, &i))) + if (!i || forced > 1) { + error = xfs_alloc_ag_vextent_small(args, cnt_cur, + &fbno, &flen, &i); + if (error) goto error0; if (i == 0 || flen == 0) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); @@ -1182,22 +1205,56 @@ xfs_alloc_ag_vextent_size( return 0; } ASSERT(i == 1); + xfs_alloc_compute_aligned(args, fbno, flen, &rbno, &rlen); + } else { + /* + * Search for a non-busy extent that is large enough. + * If we are at low space, don't check, or if we fall of + * the end of the btree, turn off the busy check and + * restart. + */ + for (;;) { + error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i); + if (error) + goto error0; + XFS_WANT_CORRUPTED_GOTO(i == 1, error0); + + xfs_alloc_compute_aligned(args, fbno, flen, + &rbno, &rlen); + + if (rlen >= args->maxlen) + break; + + error = xfs_btree_increment(cnt_cur, 0, &i); + if (error) + goto error0; + if (i == 0) { + /* + * Our only valid extents must have been busy. + * Make it unbusy by forcing the log out and + * retrying. If we've been here before, forcing + * the log isn't making the extents available, + * which means they have probably been freed in + * this transaction. In that case, we have to + * give up on them and we'll attempt a minlen + * allocation the next time around. + */ + xfs_btree_del_cursor(cnt_cur, + XFS_BTREE_NOERROR); + trace_xfs_alloc_size_busy(args); + if (!forced++) + xfs_log_force(args->mp, XFS_LOG_SYNC); + goto restart; + } + } } - /* - * There's a freespace as big as maxlen+alignment-1, get it. - */ - else { - if ((error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i))) - goto error0; - XFS_WANT_CORRUPTED_GOTO(i == 1, error0); - } + /* * In the first case above, we got the last entry in the * by-size btree. Now we check to see if the space hits maxlen * once aligned; if not, we search left for something better. * This can't happen in the second case above. */ - xfs_alloc_compute_aligned(args, fbno, flen, &rbno, &rlen); rlen = XFS_EXTLEN_MIN(args->maxlen, rlen); XFS_WANT_CORRUPTED_GOTO(rlen == 0 || (rlen <= flen && rbno + rlen <= fbno + flen), error0); @@ -1251,13 +1308,19 @@ xfs_alloc_ag_vextent_size( * Fix up the length. */ args->len = rlen; - xfs_alloc_fix_len(args); - if (rlen < args->minlen || !xfs_alloc_fix_minleft(args)) { - xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); - trace_xfs_alloc_size_nominleft(args); - args->agbno = NULLAGBLOCK; - return 0; + if (rlen < args->minlen) { + if (!forced++) { + xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); + trace_xfs_alloc_size_busy(args); + xfs_log_force(args->mp, XFS_LOG_SYNC); + goto restart; + } + goto out_nominleft; } + xfs_alloc_fix_len(args); + + if (!xfs_alloc_fix_minleft(args)) + goto out_nominleft; rlen = args->len; XFS_WANT_CORRUPTED_GOTO(rlen <= flen, error0); /* @@ -1287,6 +1350,12 @@ error0: if (bno_cur) xfs_btree_del_cursor(bno_cur, XFS_BTREE_ERROR); return error; + +out_nominleft: + xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); + trace_xfs_alloc_size_nominleft(args); + args->agbno = NULLAGBLOCK; + return 0; } /* @@ -1326,6 +1395,9 @@ xfs_alloc_ag_vextent_small( if (error) goto error0; if (fbno != NULLAGBLOCK) { + xfs_alloc_busy_reuse(args->mp, args->agno, fbno, 1, + args->userdata); + if (args->userdata) { xfs_buf_t *bp; @@ -1617,18 +1689,6 @@ xfs_free_ag_extent( trace_xfs_free_extent(mp, agno, bno, len, isfl, haveleft, haveright); - /* - * Since blocks move to the free list without the coordination - * used in xfs_bmap_finish, we can't allow block to be available - * for reallocation and non-transaction writing (user data) - * until we know that the transaction that moved it to the free - * list is permanently on disk. We track the blocks by declaring - * these blocks as "busy"; the busy list is maintained on a per-ag - * basis and each transaction records which entries should be removed - * when the iclog commits to disk. If a busy block is allocated, - * the iclog is pushed up to the LSN that freed the block. - */ - xfs_alloc_busy_insert(tp, agno, bno, len); return 0; error0: @@ -1923,21 +1983,6 @@ xfs_alloc_get_freelist( xfs_alloc_log_agf(tp, agbp, logflags); *bnop = bno; - /* - * As blocks are freed, they are added to the per-ag busy list and - * remain there until the freeing transaction is committed to disk. - * Now that we have allocated blocks, this list must be searched to see - * if a block is being reused. If one is, then the freeing transaction - * must be pushed to disk before this transaction. - * - * We do this by setting the current transaction to a sync transaction - * which guarantees that the freeing transaction is on disk before this - * transaction. This is done instead of a synchronous log force here so - * that we don't sit and wait with the AGF locked in the transaction - * during the log force. - */ - if (xfs_alloc_busy_search(mp, be32_to_cpu(agf->agf_seqno), bno, 1)) - xfs_trans_set_sync(tp); return 0; } @@ -2423,119 +2468,26 @@ xfs_free_extent( } error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0); + if (!error) + xfs_alloc_busy_insert(tp, args.agno, args.agbno, len, 0); error0: xfs_perag_put(args.pag); return error; } - -/* - * AG Busy list management - * The busy list contains block ranges that have been freed but whose - * transactions have not yet hit disk. If any block listed in a busy - * list is reused, the transaction that freed it must be forced to disk - * before continuing to use the block. - * - * xfs_alloc_busy_insert - add to the per-ag busy list - * xfs_alloc_busy_clear - remove an item from the per-ag busy list - * xfs_alloc_busy_search - search for a busy extent - */ - -/* - * Insert a new extent into the busy tree. - * - * The busy extent tree is indexed by the start block of the busy extent. - * there can be multiple overlapping ranges in the busy extent tree but only - * ever one entry at a given start block. The reason for this is that - * multi-block extents can be freed, then smaller chunks of that extent - * allocated and freed again before the first transaction commit is on disk. - * If the exact same start block is freed a second time, we have to wait for - * that busy extent to pass out of the tree before the new extent is inserted. - * There are two main cases we have to handle here. - * - * The first case is a transaction that triggers a "free - allocate - free" - * cycle. This can occur during btree manipulations as a btree block is freed - * to the freelist, then allocated from the free list, then freed again. In - * this case, the second extxpnet free is what triggers the duplicate and as - * such the transaction IDs should match. Because the extent was allocated in - * this transaction, the transaction must be marked as synchronous. This is - * true for all cases where the free/alloc/free occurs in the one transaction, - * hence the addition of the ASSERT(tp->t_flags & XFS_TRANS_SYNC) to this case. - * This serves to catch violations of the second case quite effectively. - * - * The second case is where the free/alloc/free occur in different - * transactions. In this case, the thread freeing the extent the second time - * can't mark the extent busy immediately because it is already tracked in a - * transaction that may be committing. When the log commit for the existing - * busy extent completes, the busy extent will be removed from the tree. If we - * allow the second busy insert to continue using that busy extent structure, - * it can be freed before this transaction is safely in the log. Hence our - * only option in this case is to force the log to remove the existing busy - * extent from the list before we insert the new one with the current - * transaction ID. - * - * The problem we are trying to avoid in the free-alloc-free in separate - * transactions is most easily described with a timeline: - * - * Thread 1 Thread 2 Thread 3 xfslogd - * xact alloc - * free X - * mark busy - * commit xact - * free xact - * xact alloc - * alloc X - * busy search - * mark xact sync - * commit xact - * free xact - * force log - * checkpoint starts - * .... - * xact alloc - * free X - * mark busy - * finds match - * *** KABOOM! *** - * .... - * log IO completes - * unbusy X - * checkpoint completes - * - * By issuing a log force in thread 3 @ "KABOOM", the thread will block until - * the checkpoint completes, and the busy extent it matched will have been - * removed from the tree when it is woken. Hence it can then continue safely. - * - * However, to ensure this matching process is robust, we need to use the - * transaction ID for identifying transaction, as delayed logging results in - * the busy extent and transaction lifecycles being different. i.e. the busy - * extent is active for a lot longer than the transaction. Hence the - * transaction structure can be freed and reallocated, then mark the same - * extent busy again in the new transaction. In this case the new transaction - * will have a different tid but can have the same address, and hence we need - * to check against the tid. - * - * Future: for delayed logging, we could avoid the log force if the extent was - * first freed in the current checkpoint sequence. This, however, requires the - * ability to pin the current checkpoint in memory until this transaction - * commits to ensure that both the original free and the current one combine - * logically into the one checkpoint. If the checkpoint sequences are - * different, however, we still need to wait on a log force. - */ void xfs_alloc_busy_insert( struct xfs_trans *tp, xfs_agnumber_t agno, xfs_agblock_t bno, - xfs_extlen_t len) + xfs_extlen_t len, + unsigned int flags) { struct xfs_busy_extent *new; struct xfs_busy_extent *busyp; struct xfs_perag *pag; struct rb_node **rbp; - struct rb_node *parent; - int match; - + struct rb_node *parent = NULL; new = kmem_zalloc(sizeof(struct xfs_busy_extent), KM_MAYFAIL); if (!new) { @@ -2544,7 +2496,7 @@ xfs_alloc_busy_insert( * block, make this a synchronous transaction to insure that * the block is not reused before this transaction commits. */ - trace_xfs_alloc_busy(tp, agno, bno, len, 1); + trace_xfs_alloc_busy_enomem(tp->t_mountp, agno, bno, len); xfs_trans_set_sync(tp); return; } @@ -2552,66 +2504,29 @@ xfs_alloc_busy_insert( new->agno = agno; new->bno = bno; new->length = len; - new->tid = xfs_log_get_trans_ident(tp); - INIT_LIST_HEAD(&new->list); + new->flags = flags; /* trace before insert to be able to see failed inserts */ - trace_xfs_alloc_busy(tp, agno, bno, len, 0); + trace_xfs_alloc_busy(tp->t_mountp, agno, bno, len); pag = xfs_perag_get(tp->t_mountp, new->agno); -restart: spin_lock(&pag->pagb_lock); rbp = &pag->pagb_tree.rb_node; - parent = NULL; - busyp = NULL; - match = 0; - while (*rbp && match >= 0) { + while (*rbp) { parent = *rbp; busyp = rb_entry(parent, struct xfs_busy_extent, rb_node); if (new->bno < busyp->bno) { - /* may overlap, but exact start block is lower */ rbp = &(*rbp)->rb_left; - if (new->bno + new->length > busyp->bno) - match = busyp->tid == new->tid ? 1 : -1; + ASSERT(new->bno + new->length <= busyp->bno); } else if (new->bno > busyp->bno) { - /* may overlap, but exact start block is higher */ rbp = &(*rbp)->rb_right; - if (bno < busyp->bno + busyp->length) - match = busyp->tid == new->tid ? 1 : -1; + ASSERT(bno >= busyp->bno + busyp->length); } else { - match = busyp->tid == new->tid ? 1 : -1; - break; + ASSERT(0); } } - if (match < 0) { - /* overlap marked busy in different transaction */ - spin_unlock(&pag->pagb_lock); - xfs_log_force(tp->t_mountp, XFS_LOG_SYNC); - goto restart; - } - if (match > 0) { - /* - * overlap marked busy in same transaction. Update if exact - * start block match, otherwise combine the busy extents into - * a single range. - */ - if (busyp->bno == new->bno) { - busyp->length = max(busyp->length, new->length); - spin_unlock(&pag->pagb_lock); - ASSERT(tp->t_flags & XFS_TRANS_SYNC); - xfs_perag_put(pag); - kmem_free(new); - return; - } - rb_erase(&busyp->rb_node, &pag->pagb_tree); - new->length = max(busyp->bno + busyp->length, - new->bno + new->length) - - min(busyp->bno, new->bno); - new->bno = min(busyp->bno, new->bno); - } else - busyp = NULL; rb_link_node(&new->rb_node, parent, rbp); rb_insert_color(&new->rb_node, &pag->pagb_tree); @@ -2619,7 +2534,6 @@ restart: list_add(&new->list, &tp->t_busy); spin_unlock(&pag->pagb_lock); xfs_perag_put(pag); - kmem_free(busyp); } /* @@ -2668,31 +2582,466 @@ xfs_alloc_busy_search( } } spin_unlock(&pag->pagb_lock); - trace_xfs_alloc_busysearch(mp, agno, bno, len, !!match); xfs_perag_put(pag); return match; } +/* + * The found free extent [fbno, fend] overlaps part or all of the given busy + * extent. If the overlap covers the beginning, the end, or all of the busy + * extent, the overlapping portion can be made unbusy and used for the + * allocation. We can't split a busy extent because we can't modify a + * transaction/CIL context busy list, but we can update an entries block + * number or length. + * + * Returns true if the extent can safely be reused, or false if the search + * needs to be restarted. + */ +STATIC bool +xfs_alloc_busy_update_extent( + struct xfs_mount *mp, + struct xfs_perag *pag, + struct xfs_busy_extent *busyp, + xfs_agblock_t fbno, + xfs_extlen_t flen, + bool userdata) +{ + xfs_agblock_t fend = fbno + flen; + xfs_agblock_t bbno = busyp->bno; + xfs_agblock_t bend = bbno + busyp->length; + + /* + * This extent is currently being discarded. Give the thread + * performing the discard a chance to mark the extent unbusy + * and retry. + */ + if (busyp->flags & XFS_ALLOC_BUSY_DISCARDED) { + spin_unlock(&pag->pagb_lock); + delay(1); + spin_lock(&pag->pagb_lock); + return false; + } + + /* + * If there is a busy extent overlapping a user allocation, we have + * no choice but to force the log and retry the search. + * + * Fortunately this does not happen during normal operation, but + * only if the filesystem is very low on space and has to dip into + * the AGFL for normal allocations. + */ + if (userdata) + goto out_force_log; + + if (bbno < fbno && bend > fend) { + /* + * Case 1: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +---------+ + * fbno fend + */ + + /* + * We would have to split the busy extent to be able to track + * it correct, which we cannot do because we would have to + * modify the list of busy extents attached to the transaction + * or CIL context, which is immutable. + * + * Force out the log to clear the busy extent and retry the + * search. + */ + goto out_force_log; + } else if (bbno >= fbno && bend <= fend) { + /* + * Case 2: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-----------------+ + * fbno fend + * + * Case 3: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +--------------------------+ + * fbno fend + * + * Case 4: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +--------------------------+ + * fbno fend + * + * Case 5: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-----------------------------------+ + * fbno fend + * + */ + + /* + * The busy extent is fully covered by the extent we are + * allocating, and can simply be removed from the rbtree. + * However we cannot remove it from the immutable list + * tracking busy extents in the transaction or CIL context, + * so set the length to zero to mark it invalid. + * + * We also need to restart the busy extent search from the + * tree root, because erasing the node can rearrange the + * tree topology. + */ + rb_erase(&busyp->rb_node, &pag->pagb_tree); + busyp->length = 0; + return false; + } else if (fend < bend) { + /* + * Case 6: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +---------+ + * fbno fend + * + * Case 7: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +------------------+ + * fbno fend + * + */ + busyp->bno = fend; + } else if (bbno < fbno) { + /* + * Case 8: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-------------+ + * fbno fend + * + * Case 9: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +----------------------+ + * fbno fend + */ + busyp->length = fbno - busyp->bno; + } else { + ASSERT(0); + } + + trace_xfs_alloc_busy_reuse(mp, pag->pag_agno, fbno, flen); + return true; + +out_force_log: + spin_unlock(&pag->pagb_lock); + xfs_log_force(mp, XFS_LOG_SYNC); + trace_xfs_alloc_busy_force(mp, pag->pag_agno, fbno, flen); + spin_lock(&pag->pagb_lock); + return false; +} + + +/* + * For a given extent [fbno, flen], make sure we can reuse it safely. + */ void -xfs_alloc_busy_clear( +xfs_alloc_busy_reuse( struct xfs_mount *mp, - struct xfs_busy_extent *busyp) + xfs_agnumber_t agno, + xfs_agblock_t fbno, + xfs_extlen_t flen, + bool userdata) { struct xfs_perag *pag; + struct rb_node *rbp; - trace_xfs_alloc_unbusy(mp, busyp->agno, busyp->bno, - busyp->length); + ASSERT(flen > 0); - ASSERT(xfs_alloc_busy_search(mp, busyp->agno, busyp->bno, - busyp->length) == 1); + pag = xfs_perag_get(mp, agno); + spin_lock(&pag->pagb_lock); +restart: + rbp = pag->pagb_tree.rb_node; + while (rbp) { + struct xfs_busy_extent *busyp = + rb_entry(rbp, struct xfs_busy_extent, rb_node); + xfs_agblock_t bbno = busyp->bno; + xfs_agblock_t bend = bbno + busyp->length; - list_del_init(&busyp->list); + if (fbno + flen <= bbno) { + rbp = rbp->rb_left; + continue; + } else if (fbno >= bend) { + rbp = rbp->rb_right; + continue; + } - pag = xfs_perag_get(mp, busyp->agno); - spin_lock(&pag->pagb_lock); - rb_erase(&busyp->rb_node, &pag->pagb_tree); + if (!xfs_alloc_busy_update_extent(mp, pag, busyp, fbno, flen, + userdata)) + goto restart; + } spin_unlock(&pag->pagb_lock); xfs_perag_put(pag); +} + +/* + * For a given extent [fbno, flen], search the busy extent list to find a + * subset of the extent that is not busy. If *rlen is smaller than + * args->minlen no suitable extent could be found, and the higher level + * code needs to force out the log and retry the allocation. + */ +STATIC void +xfs_alloc_busy_trim( + struct xfs_alloc_arg *args, + xfs_agblock_t bno, + xfs_extlen_t len, + xfs_agblock_t *rbno, + xfs_extlen_t *rlen) +{ + xfs_agblock_t fbno; + xfs_extlen_t flen; + struct rb_node *rbp; + + ASSERT(len > 0); + + spin_lock(&args->pag->pagb_lock); +restart: + fbno = bno; + flen = len; + rbp = args->pag->pagb_tree.rb_node; + while (rbp && flen >= args->minlen) { + struct xfs_busy_extent *busyp = + rb_entry(rbp, struct xfs_busy_extent, rb_node); + xfs_agblock_t fend = fbno + flen; + xfs_agblock_t bbno = busyp->bno; + xfs_agblock_t bend = bbno + busyp->length; + + if (fend <= bbno) { + rbp = rbp->rb_left; + continue; + } else if (fbno >= bend) { + rbp = rbp->rb_right; + continue; + } + + /* + * If this is a metadata allocation, try to reuse the busy + * extent instead of trimming the allocation. + */ + if (!args->userdata && + !(busyp->flags & XFS_ALLOC_BUSY_DISCARDED)) { + if (!xfs_alloc_busy_update_extent(args->mp, args->pag, + busyp, fbno, flen, + false)) + goto restart; + continue; + } + + if (bbno <= fbno) { + /* start overlap */ + + /* + * Case 1: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +---------+ + * fbno fend + * + * Case 2: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-------------+ + * fbno fend + * + * Case 3: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-------------+ + * fbno fend + * + * Case 4: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-----------------+ + * fbno fend + * + * No unbusy region in extent, return failure. + */ + if (fend <= bend) + goto fail; + + /* + * Case 5: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +----------------------+ + * fbno fend + * + * Case 6: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +--------------------------+ + * fbno fend + * + * Needs to be trimmed to: + * +-------+ + * fbno fend + */ + fbno = bend; + } else if (bend >= fend) { + /* end overlap */ + + /* + * Case 7: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +------------------+ + * fbno fend + * + * Case 8: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +--------------------------+ + * fbno fend + * + * Needs to be trimmed to: + * +-------+ + * fbno fend + */ + fend = bbno; + } else { + /* middle overlap */ + + /* + * Case 9: + * bbno bend + * +BBBBBBBBBBBBBBBBB+ + * +-----------------------------------+ + * fbno fend + * + * Can be trimmed to: + * +-------+ OR +-------+ + * fbno fend fbno fend + * + * Backward allocation leads to significant + * fragmentation of directories, which degrades + * directory performance, therefore we always want to + * choose the option that produces forward allocation + * patterns. + * Preferring the lower bno extent will make the next + * request use "fend" as the start of the next + * allocation; if the segment is no longer busy at + * that point, we'll get a contiguous allocation, but + * even if it is still busy, we will get a forward + * allocation. + * We try to avoid choosing the segment at "bend", + * because that can lead to the next allocation + * taking the segment at "fbno", which would be a + * backward allocation. We only use the segment at + * "fbno" if it is much larger than the current + * requested size, because in that case there's a + * good chance subsequent allocations will be + * contiguous. + */ + if (bbno - fbno >= args->maxlen) { + /* left candidate fits perfect */ + fend = bbno; + } else if (fend - bend >= args->maxlen * 4) { + /* right candidate has enough free space */ + fbno = bend; + } else if (bbno - fbno >= args->minlen) { + /* left candidate fits minimum requirement */ + fend = bbno; + } else { + goto fail; + } + } + + flen = fend - fbno; + } + spin_unlock(&args->pag->pagb_lock); + + if (fbno != bno || flen != len) { + trace_xfs_alloc_busy_trim(args->mp, args->agno, bno, len, + fbno, flen); + } + *rbno = fbno; + *rlen = flen; + return; +fail: + /* + * Return a zero extent length as failure indications. All callers + * re-check if the trimmed extent satisfies the minlen requirement. + */ + spin_unlock(&args->pag->pagb_lock); + trace_xfs_alloc_busy_trim(args->mp, args->agno, bno, len, fbno, 0); + *rbno = fbno; + *rlen = 0; +} + +static void +xfs_alloc_busy_clear_one( + struct xfs_mount *mp, + struct xfs_perag *pag, + struct xfs_busy_extent *busyp) +{ + if (busyp->length) { + trace_xfs_alloc_busy_clear(mp, busyp->agno, busyp->bno, + busyp->length); + rb_erase(&busyp->rb_node, &pag->pagb_tree); + } + list_del_init(&busyp->list); kmem_free(busyp); } + +/* + * Remove all extents on the passed in list from the busy extents tree. + * If do_discard is set skip extents that need to be discarded, and mark + * these as undergoing a discard operation instead. + */ +void +xfs_alloc_busy_clear( + struct xfs_mount *mp, + struct list_head *list, + bool do_discard) +{ + struct xfs_busy_extent *busyp, *n; + struct xfs_perag *pag = NULL; + xfs_agnumber_t agno = NULLAGNUMBER; + + list_for_each_entry_safe(busyp, n, list, list) { + if (busyp->agno != agno) { + if (pag) { + spin_unlock(&pag->pagb_lock); + xfs_perag_put(pag); + } + pag = xfs_perag_get(mp, busyp->agno); + spin_lock(&pag->pagb_lock); + agno = busyp->agno; + } + + if (do_discard && busyp->length && + !(busyp->flags & XFS_ALLOC_BUSY_SKIP_DISCARD)) + busyp->flags = XFS_ALLOC_BUSY_DISCARDED; + else + xfs_alloc_busy_clear_one(mp, pag, busyp); + } + + if (pag) { + spin_unlock(&pag->pagb_lock); + xfs_perag_put(pag); + } +} + +/* + * Callback for list_sort to sort busy extents by the AG they reside in. + */ +int +xfs_busy_extent_ag_cmp( + void *priv, + struct list_head *a, + struct list_head *b) +{ + return container_of(a, struct xfs_busy_extent, list)->agno - + container_of(b, struct xfs_busy_extent, list)->agno; +} diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h index d0b3bc7..2f52b92 100644 --- a/fs/xfs/xfs_alloc.h +++ b/fs/xfs/xfs_alloc.h @@ -137,14 +137,28 @@ xfs_alloc_longest_free_extent(struct xfs_mount *mp, #ifdef __KERNEL__ void xfs_alloc_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno, - xfs_agblock_t bno, xfs_extlen_t len); + xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags); void -xfs_alloc_busy_clear(struct xfs_mount *mp, struct xfs_busy_extent *busyp); +xfs_alloc_busy_clear(struct xfs_mount *mp, struct list_head *list, + bool do_discard); int xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t bno, xfs_extlen_t len); + +void +xfs_alloc_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata); + +int +xfs_busy_extent_ag_cmp(void *priv, struct list_head *a, struct list_head *b); + +static inline void xfs_alloc_busy_sort(struct list_head *list) +{ + list_sort(NULL, list, xfs_busy_extent_ag_cmp); +} + #endif /* __KERNEL__ */ /* diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c index 3916925..2b35188 100644 --- a/fs/xfs/xfs_alloc_btree.c +++ b/fs/xfs/xfs_alloc_btree.c @@ -95,6 +95,8 @@ xfs_allocbt_alloc_block( return 0; } + xfs_alloc_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1, false); + xfs_trans_agbtree_delta(cur->bc_tp, 1); new->s = cpu_to_be32(bno); @@ -118,18 +120,8 @@ xfs_allocbt_free_block( if (error) return error; - /* - * Since blocks move to the free list without the coordination used in - * xfs_bmap_finish, we can't allow block to be available for - * reallocation and non-transaction writing (user data) until we know - * that the transaction that moved it to the free list is permanently - * on disk. We track the blocks by declaring these blocks as "busy"; - * the busy list is maintained on a per-ag basis and each transaction - * records which entries should be removed when the iclog commits to - * disk. If a busy block is allocated, the iclog is pushed up to the - * LSN that freed the block. - */ - xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1); + xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1, + XFS_ALLOC_BUSY_SKIP_DISCARD); xfs_trans_agbtree_delta(cur->bc_tp, -1); return 0; } diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index c863753..01d2072 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -490,6 +490,13 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) args.whichfork = XFS_ATTR_FORK; /* + * we have no control over the attribute names that userspace passes us + * to remove, so we have to allow the name lookup prior to attribute + * removal to fail. + */ + args.op_flags = XFS_DA_OP_OKNOENT; + + /* * Attach the dquots to the inode. */ error = xfs_qm_dqattach(dp, 0); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index fa00788..e546a33 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -89,36 +89,19 @@ xfs_bmap_add_attrfork_local( int *flags); /* inode logging flags */ /* - * Called by xfs_bmapi to update file extent records and the btree - * after allocating space (or doing a delayed allocation). - */ -STATIC int /* error */ -xfs_bmap_add_extent( - xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ - xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ - xfs_bmbt_irec_t *new, /* new data to add to file extents */ - xfs_fsblock_t *first, /* pointer to firstblock variable */ - xfs_bmap_free_t *flist, /* list of extents to be freed */ - int *logflagsp, /* inode logging flags */ - int whichfork, /* data or attr fork */ - int rsvd); /* OK to allocate reserved blocks */ - -/* * Called by xfs_bmap_add_extent to handle cases converting a delayed * allocation to a real allocation. */ STATIC int /* error */ xfs_bmap_add_extent_delay_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */ xfs_fsblock_t *first, /* pointer to firstblock variable */ xfs_bmap_free_t *flist, /* list of extents to be freed */ - int *logflagsp, /* inode logging flags */ - int rsvd); /* OK to allocate reserved blocks */ + int *logflagsp); /* inode logging flags */ /* * Called by xfs_bmap_add_extent to handle cases converting a hole @@ -127,10 +110,9 @@ xfs_bmap_add_extent_delay_real( STATIC int /* error */ xfs_bmap_add_extent_hole_delay( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ - int *logflagsp,/* inode logging flags */ - int rsvd); /* OK to allocate reserved blocks */ + int *logflagsp); /* inode logging flags */ /* * Called by xfs_bmap_add_extent to handle cases converting a hole @@ -139,7 +121,7 @@ xfs_bmap_add_extent_hole_delay( STATIC int /* error */ xfs_bmap_add_extent_hole_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t *cur, /* if null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ int *logflagsp, /* inode logging flags */ @@ -152,7 +134,7 @@ xfs_bmap_add_extent_hole_real( STATIC int /* error */ xfs_bmap_add_extent_unwritten_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ int *logflagsp); /* inode logging flags */ @@ -180,22 +162,6 @@ xfs_bmap_btree_to_extents( int whichfork); /* data or attr fork */ /* - * Called by xfs_bmapi to update file extent records and the btree - * after removing space (or undoing a delayed allocation). - */ -STATIC int /* error */ -xfs_bmap_del_extent( - xfs_inode_t *ip, /* incore inode pointer */ - xfs_trans_t *tp, /* current trans pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ - xfs_bmap_free_t *flist, /* list of extents to be freed */ - xfs_btree_cur_t *cur, /* if null, not a btree */ - xfs_bmbt_irec_t *new, /* new data to add to file extents */ - int *logflagsp,/* inode logging flags */ - int whichfork, /* data or attr fork */ - int rsvd); /* OK to allocate reserved blocks */ - -/* * Remove the entry "free" from the free item list. Prev points to the * previous entry, unless "free" is the head of the list. */ @@ -474,14 +440,13 @@ xfs_bmap_add_attrfork_local( STATIC int /* error */ xfs_bmap_add_extent( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ xfs_fsblock_t *first, /* pointer to firstblock variable */ xfs_bmap_free_t *flist, /* list of extents to be freed */ int *logflagsp, /* inode logging flags */ - int whichfork, /* data or attr fork */ - int rsvd) /* OK to use reserved data blocks */ + int whichfork) /* data or attr fork */ { xfs_btree_cur_t *cur; /* btree cursor or null */ xfs_filblks_t da_new; /* new count del alloc blocks used */ @@ -492,23 +457,27 @@ xfs_bmap_add_extent( xfs_extnum_t nextents; /* number of extents in file now */ XFS_STATS_INC(xs_add_exlist); + cur = *curp; ifp = XFS_IFORK_PTR(ip, whichfork); nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - ASSERT(idx <= nextents); da_old = da_new = 0; error = 0; + + ASSERT(*idx >= 0); + ASSERT(*idx <= nextents); + /* * This is the first extent added to a new/empty file. * Special case this one, so other routines get to assume there are * already extents in the list. */ if (nextents == 0) { - xfs_iext_insert(ip, 0, 1, new, + xfs_iext_insert(ip, *idx, 1, new, whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0); ASSERT(cur == NULL); - ifp->if_lastex = 0; + if (!isnullstartblock(new->br_startblock)) { XFS_IFORK_NEXT_SET(ip, whichfork, 1); logflags = XFS_ILOG_CORE | xfs_ilog_fext(whichfork); @@ -522,27 +491,25 @@ xfs_bmap_add_extent( if (cur) ASSERT((cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL) == 0); - if ((error = xfs_bmap_add_extent_hole_delay(ip, idx, new, - &logflags, rsvd))) - goto done; + error = xfs_bmap_add_extent_hole_delay(ip, idx, new, + &logflags); } /* * Real allocation off the end of the file. */ - else if (idx == nextents) { + else if (*idx == nextents) { if (cur) ASSERT((cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL) == 0); - if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur, new, - &logflags, whichfork))) - goto done; + error = xfs_bmap_add_extent_hole_real(ip, idx, cur, new, + &logflags, whichfork); } else { xfs_bmbt_irec_t prev; /* old extent at offset idx */ /* * Get the record referred to by idx. */ - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &prev); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &prev); /* * If it's a real allocation record, and the new allocation ends * after the start of the referred to record, then we're filling @@ -557,22 +524,18 @@ xfs_bmap_add_extent( if (cur) ASSERT(cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL); - if ((error = xfs_bmap_add_extent_delay_real(ip, - idx, &cur, new, &da_new, first, flist, - &logflags, rsvd))) - goto done; - } else if (new->br_state == XFS_EXT_NORM) { - ASSERT(new->br_state == XFS_EXT_NORM); - if ((error = xfs_bmap_add_extent_unwritten_real( - ip, idx, &cur, new, &logflags))) - goto done; + error = xfs_bmap_add_extent_delay_real(ip, + idx, &cur, new, &da_new, + first, flist, &logflags); } else { - ASSERT(new->br_state == XFS_EXT_UNWRITTEN); - if ((error = xfs_bmap_add_extent_unwritten_real( - ip, idx, &cur, new, &logflags))) + ASSERT(new->br_state == XFS_EXT_NORM || + new->br_state == XFS_EXT_UNWRITTEN); + + error = xfs_bmap_add_extent_unwritten_real(ip, + idx, &cur, new, &logflags); + if (error) goto done; } - ASSERT(*curp == cur || *curp == NULL); } /* * Otherwise we're filling in a hole with an allocation. @@ -581,13 +544,15 @@ xfs_bmap_add_extent( if (cur) ASSERT((cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL) == 0); - if ((error = xfs_bmap_add_extent_hole_real(ip, idx, cur, - new, &logflags, whichfork))) - goto done; + error = xfs_bmap_add_extent_hole_real(ip, idx, cur, + new, &logflags, whichfork); } } + if (error) + goto done; ASSERT(*curp == cur || *curp == NULL); + /* * Convert to a btree if necessary. */ @@ -615,7 +580,7 @@ xfs_bmap_add_extent( ASSERT(nblks <= da_old); if (nblks < da_old) xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS, - (int64_t)(da_old - nblks), rsvd); + (int64_t)(da_old - nblks), 0); } /* * Clear out the allocated field, done with it now in any case. @@ -640,14 +605,13 @@ done: STATIC int /* error */ xfs_bmap_add_extent_delay_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */ xfs_fsblock_t *first, /* pointer to firstblock variable */ xfs_bmap_free_t *flist, /* list of extents to be freed */ - int *logflagsp, /* inode logging flags */ - int rsvd) /* OK to use reserved data block allocation */ + int *logflagsp) /* inode logging flags */ { xfs_btree_cur_t *cur; /* btree cursor */ int diff; /* temp value */ @@ -673,7 +637,7 @@ xfs_bmap_add_extent_delay_real( */ cur = *curp; ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); - ep = xfs_iext_get_ext(ifp, idx); + ep = xfs_iext_get_ext(ifp, *idx); xfs_bmbt_get_all(ep, &PREV); new_endoff = new->br_startoff + new->br_blockcount; ASSERT(PREV.br_startoff <= new->br_startoff); @@ -692,9 +656,9 @@ xfs_bmap_add_extent_delay_real( * Check and set flags if this segment has a left neighbor. * Don't set contiguous if the combined extent would be too large. */ - if (idx > 0) { + if (*idx > 0) { state |= BMAP_LEFT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &LEFT); if (isnullstartblock(LEFT.br_startblock)) state |= BMAP_LEFT_DELAY; @@ -712,9 +676,9 @@ xfs_bmap_add_extent_delay_real( * Don't set contiguous if the combined extent would be too large. * Also check for all-three-contiguous being too large. */ - if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { + if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { state |= BMAP_RIGHT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT); if (isnullstartblock(RIGHT.br_startblock)) state |= BMAP_RIGHT_DELAY; @@ -745,14 +709,14 @@ xfs_bmap_add_extent_delay_real( * Filling in all of a previously delayed allocation extent. * The left and right neighbors are both contiguous with new. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), LEFT.br_blockcount + PREV.br_blockcount + RIGHT.br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - xfs_iext_remove(ip, idx, 2, state); - ip->i_df.if_lastex = idx - 1; + xfs_iext_remove(ip, *idx + 1, 2, state); ip->i_d.di_nextents--; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -784,13 +748,14 @@ xfs_bmap_add_extent_delay_real( * Filling in all of a previously delayed allocation extent. * The left neighbor is contiguous, the right is not. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), LEFT.br_blockcount + PREV.br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx - 1; - xfs_iext_remove(ip, idx, 1, state); + xfs_iext_remove(ip, *idx + 1, 1, state); if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -814,14 +779,13 @@ xfs_bmap_add_extent_delay_real( * Filling in all of a previously delayed allocation extent. * The right neighbor is contiguous, the left is not. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startblock(ep, new->br_startblock); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount + RIGHT.br_blockcount); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx; - xfs_iext_remove(ip, idx + 1, 1, state); + xfs_iext_remove(ip, *idx + 1, 1, state); if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -837,6 +801,7 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_blockcount, PREV.br_state))) goto done; } + *dnew = 0; break; @@ -846,11 +811,10 @@ xfs_bmap_add_extent_delay_real( * Neither the left nor right neighbors are contiguous with * the new one. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startblock(ep, new->br_startblock); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx; ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -866,6 +830,7 @@ xfs_bmap_add_extent_delay_real( goto done; XFS_WANT_CORRUPTED_GOTO(i == 1, done); } + *dnew = 0; break; @@ -874,17 +839,16 @@ xfs_bmap_add_extent_delay_real( * Filling in the first part of a previous delayed allocation. * The left neighbor is contiguous. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + trace_xfs_bmap_pre_update(ip, *idx - 1, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx - 1), LEFT.br_blockcount + new->br_blockcount); xfs_bmbt_set_startoff(ep, PREV.br_startoff + new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx - 1, state, _THIS_IP_); temp = PREV.br_blockcount - new->br_blockcount; - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); - ip->i_df.if_lastex = idx - 1; if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -904,7 +868,9 @@ xfs_bmap_add_extent_delay_real( temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), startblockval(PREV.br_startblock)); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + --*idx; *dnew = temp; break; @@ -913,12 +879,11 @@ xfs_bmap_add_extent_delay_real( * Filling in the first part of a previous delayed allocation. * The left neighbor is not contiguous. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startoff(ep, new_endoff); temp = PREV.br_blockcount - new->br_blockcount; xfs_bmbt_set_blockcount(ep, temp); - xfs_iext_insert(ip, idx, 1, new, state); - ip->i_df.if_lastex = idx; + xfs_iext_insert(ip, *idx, 1, new, state); ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -946,9 +911,10 @@ xfs_bmap_add_extent_delay_real( temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), startblockval(PREV.br_startblock) - (cur ? cur->bc_private.b.allocated : 0)); - ep = xfs_iext_get_ext(ifp, idx + 1); + ep = xfs_iext_get_ext(ifp, *idx + 1); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx + 1, state, _THIS_IP_); + *dnew = temp; break; @@ -958,15 +924,13 @@ xfs_bmap_add_extent_delay_real( * The right neighbor is contiguous with the new allocation. */ temp = PREV.br_blockcount - new->br_blockcount; - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); - trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx + 1, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); - xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1), + xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx + 1), new->br_startoff, new->br_startblock, new->br_blockcount + RIGHT.br_blockcount, RIGHT.br_state); - trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_); - ip->i_df.if_lastex = idx + 1; + trace_xfs_bmap_post_update(ip, *idx + 1, state, _THIS_IP_); if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -983,10 +947,14 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_state))) goto done; } + temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), startblockval(PREV.br_startblock)); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + ++*idx; *dnew = temp; break; @@ -996,10 +964,9 @@ xfs_bmap_add_extent_delay_real( * The right neighbor is not contiguous. */ temp = PREV.br_blockcount - new->br_blockcount; - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); - xfs_iext_insert(ip, idx + 1, 1, new, state); - ip->i_df.if_lastex = idx + 1; + xfs_iext_insert(ip, *idx + 1, 1, new, state); ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1027,9 +994,11 @@ xfs_bmap_add_extent_delay_real( temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), startblockval(PREV.br_startblock) - (cur ? cur->bc_private.b.allocated : 0)); - ep = xfs_iext_get_ext(ifp, idx); + ep = xfs_iext_get_ext(ifp, *idx); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + ++*idx; *dnew = temp; break; @@ -1056,7 +1025,7 @@ xfs_bmap_add_extent_delay_real( */ temp = new->br_startoff - PREV.br_startoff; temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff; - trace_xfs_bmap_pre_update(ip, idx, 0, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, 0, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); /* truncate PREV */ LEFT = *new; RIGHT.br_state = PREV.br_state; @@ -1065,8 +1034,7 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_startoff = new_endoff; RIGHT.br_blockcount = temp2; /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */ - xfs_iext_insert(ip, idx + 1, 2, &LEFT, state); - ip->i_df.if_lastex = idx + 1; + xfs_iext_insert(ip, *idx + 1, 2, &LEFT, state); ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1097,7 +1065,7 @@ xfs_bmap_add_extent_delay_real( (cur ? cur->bc_private.b.allocated : 0)); if (diff > 0 && xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS, - -((int64_t)diff), rsvd)) { + -((int64_t)diff), 0)) { /* * Ick gross gag me with a spoon. */ @@ -1109,7 +1077,7 @@ xfs_bmap_add_extent_delay_real( if (!diff || !xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS, - -((int64_t)diff), rsvd)) + -((int64_t)diff), 0)) break; } if (temp2) { @@ -1118,18 +1086,20 @@ xfs_bmap_add_extent_delay_real( if (!diff || !xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS, - -((int64_t)diff), rsvd)) + -((int64_t)diff), 0)) break; } } } - ep = xfs_iext_get_ext(ifp, idx); + ep = xfs_iext_get_ext(ifp, *idx); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); - trace_xfs_bmap_pre_update(ip, idx + 2, state, _THIS_IP_); - xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx + 2), + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx + 2, state, _THIS_IP_); + xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx + 2), nullstartblock((int)temp2)); - trace_xfs_bmap_post_update(ip, idx + 2, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx + 2, state, _THIS_IP_); + + ++*idx; *dnew = temp + temp2; break; @@ -1161,7 +1131,7 @@ done: STATIC int /* error */ xfs_bmap_add_extent_unwritten_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t **curp, /* if *curp is null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ int *logflagsp) /* inode logging flags */ @@ -1188,7 +1158,7 @@ xfs_bmap_add_extent_unwritten_real( error = 0; cur = *curp; ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); - ep = xfs_iext_get_ext(ifp, idx); + ep = xfs_iext_get_ext(ifp, *idx); xfs_bmbt_get_all(ep, &PREV); newext = new->br_state; oldext = (newext == XFS_EXT_UNWRITTEN) ? @@ -1211,9 +1181,9 @@ xfs_bmap_add_extent_unwritten_real( * Check and set flags if this segment has a left neighbor. * Don't set contiguous if the combined extent would be too large. */ - if (idx > 0) { + if (*idx > 0) { state |= BMAP_LEFT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &LEFT); if (isnullstartblock(LEFT.br_startblock)) state |= BMAP_LEFT_DELAY; @@ -1231,9 +1201,9 @@ xfs_bmap_add_extent_unwritten_real( * Don't set contiguous if the combined extent would be too large. * Also check for all-three-contiguous being too large. */ - if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { + if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { state |= BMAP_RIGHT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT); if (isnullstartblock(RIGHT.br_startblock)) state |= BMAP_RIGHT_DELAY; } @@ -1262,14 +1232,15 @@ xfs_bmap_add_extent_unwritten_real( * Setting all of a previous oldext extent to newext. * The left and right neighbors are both contiguous with new. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), LEFT.br_blockcount + PREV.br_blockcount + RIGHT.br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - xfs_iext_remove(ip, idx, 2, state); - ip->i_df.if_lastex = idx - 1; + xfs_iext_remove(ip, *idx + 1, 2, state); ip->i_d.di_nextents -= 2; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1305,13 +1276,14 @@ xfs_bmap_add_extent_unwritten_real( * Setting all of a previous oldext extent to newext. * The left neighbor is contiguous, the right is not. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), LEFT.br_blockcount + PREV.br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx - 1; - xfs_iext_remove(ip, idx, 1, state); + xfs_iext_remove(ip, *idx + 1, 1, state); ip->i_d.di_nextents--; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1341,13 +1313,12 @@ xfs_bmap_add_extent_unwritten_real( * Setting all of a previous oldext extent to newext. * The right neighbor is contiguous, the left is not. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount + RIGHT.br_blockcount); xfs_bmbt_set_state(ep, newext); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx; - xfs_iext_remove(ip, idx + 1, 1, state); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + xfs_iext_remove(ip, *idx + 1, 1, state); ip->i_d.di_nextents--; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1378,11 +1349,10 @@ xfs_bmap_add_extent_unwritten_real( * Neither the left nor right neighbors are contiguous with * the new one. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_state(ep, newext); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx; if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -1404,21 +1374,22 @@ xfs_bmap_add_extent_unwritten_real( * Setting the first part of a previous oldext extent to newext. * The left neighbor is contiguous. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + trace_xfs_bmap_pre_update(ip, *idx - 1, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx - 1), LEFT.br_blockcount + new->br_blockcount); xfs_bmbt_set_startoff(ep, PREV.br_startoff + new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx - 1, state, _THIS_IP_); - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startblock(ep, new->br_startblock + new->br_blockcount); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount - new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + --*idx; - ip->i_df.if_lastex = idx - 1; if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -1449,17 +1420,16 @@ xfs_bmap_add_extent_unwritten_real( * Setting the first part of a previous oldext extent to newext. * The left neighbor is not contiguous. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); ASSERT(ep && xfs_bmbt_get_state(ep) == oldext); xfs_bmbt_set_startoff(ep, new_endoff); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount - new->br_blockcount); xfs_bmbt_set_startblock(ep, new->br_startblock + new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - xfs_iext_insert(ip, idx, 1, new, state); - ip->i_df.if_lastex = idx; + xfs_iext_insert(ip, *idx, 1, new, state); ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1488,17 +1458,19 @@ xfs_bmap_add_extent_unwritten_real( * Setting the last part of a previous oldext extent to newext. * The right neighbor is contiguous with the new allocation. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); - trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount - new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); - xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1), + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + ++*idx; + + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx), new->br_startoff, new->br_startblock, new->br_blockcount + RIGHT.br_blockcount, newext); - trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ip->i_df.if_lastex = idx + 1; if (cur == NULL) rval = XFS_ILOG_DEXT; else { @@ -1528,13 +1500,14 @@ xfs_bmap_add_extent_unwritten_real( * Setting the last part of a previous oldext extent to newext. * The right neighbor is not contiguous. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, PREV.br_blockcount - new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + ++*idx; + xfs_iext_insert(ip, *idx, 1, new, state); - xfs_iext_insert(ip, idx + 1, 1, new, state); - ip->i_df.if_lastex = idx + 1; ip->i_d.di_nextents++; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1568,10 +1541,10 @@ xfs_bmap_add_extent_unwritten_real( * newext. Contiguity is impossible here. * One extent becomes three extents. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, new->br_startoff - PREV.br_startoff); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); r[0] = *new; r[1].br_startoff = new_endoff; @@ -1579,8 +1552,10 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startoff + PREV.br_blockcount - new_endoff; r[1].br_startblock = new->br_startblock + new->br_blockcount; r[1].br_state = oldext; - xfs_iext_insert(ip, idx + 1, 2, &r[0], state); - ip->i_df.if_lastex = idx + 1; + + ++*idx; + xfs_iext_insert(ip, *idx, 2, &r[0], state); + ip->i_d.di_nextents += 2; if (cur == NULL) rval = XFS_ILOG_CORE | XFS_ILOG_DEXT; @@ -1650,12 +1625,10 @@ done: STATIC int /* error */ xfs_bmap_add_extent_hole_delay( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ - int *logflagsp, /* inode logging flags */ - int rsvd) /* OK to allocate reserved blocks */ + int *logflagsp) /* inode logging flags */ { - xfs_bmbt_rec_host_t *ep; /* extent record for idx */ xfs_ifork_t *ifp; /* inode fork pointer */ xfs_bmbt_irec_t left; /* left neighbor extent entry */ xfs_filblks_t newlen=0; /* new indirect size */ @@ -1665,16 +1638,15 @@ xfs_bmap_add_extent_hole_delay( xfs_filblks_t temp=0; /* temp for indirect calculations */ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); - ep = xfs_iext_get_ext(ifp, idx); state = 0; ASSERT(isnullstartblock(new->br_startblock)); /* * Check and set flags if this segment has a left neighbor */ - if (idx > 0) { + if (*idx > 0) { state |= BMAP_LEFT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left); if (isnullstartblock(left.br_startblock)) state |= BMAP_LEFT_DELAY; @@ -1684,9 +1656,9 @@ xfs_bmap_add_extent_hole_delay( * Check and set flags if the current (right) segment exists. * If it doesn't exist, we're converting the hole at end-of-file. */ - if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) { + if (*idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) { state |= BMAP_RIGHT_VALID; - xfs_bmbt_get_all(ep, &right); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right); if (isnullstartblock(right.br_startblock)) state |= BMAP_RIGHT_DELAY; @@ -1719,21 +1691,21 @@ xfs_bmap_add_extent_hole_delay( * on the left and on the right. * Merge all three into a single extent record. */ + --*idx; temp = left.br_blockcount + new->br_blockcount + right.br_blockcount; - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp); oldlen = startblockval(left.br_startblock) + startblockval(new->br_startblock) + startblockval(right.br_startblock); newlen = xfs_bmap_worst_indlen(ip, temp); - xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1), + xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx), nullstartblock((int)newlen)); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - xfs_iext_remove(ip, idx, 1, state); - ip->i_df.if_lastex = idx - 1; + xfs_iext_remove(ip, *idx + 1, 1, state); break; case BMAP_LEFT_CONTIG: @@ -1742,17 +1714,17 @@ xfs_bmap_add_extent_hole_delay( * on the left. * Merge the new allocation with the left neighbor. */ + --*idx; temp = left.br_blockcount + new->br_blockcount; - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp); + + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp); oldlen = startblockval(left.br_startblock) + startblockval(new->br_startblock); newlen = xfs_bmap_worst_indlen(ip, temp); - xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1), + xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx), nullstartblock((int)newlen)); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); - - ip->i_df.if_lastex = idx - 1; + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); break; case BMAP_RIGHT_CONTIG: @@ -1761,16 +1733,15 @@ xfs_bmap_add_extent_hole_delay( * on the right. * Merge the new allocation with the right neighbor. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); temp = new->br_blockcount + right.br_blockcount; oldlen = startblockval(new->br_startblock) + startblockval(right.br_startblock); newlen = xfs_bmap_worst_indlen(ip, temp); - xfs_bmbt_set_allf(ep, new->br_startoff, + xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx), + new->br_startoff, nullstartblock((int)newlen), temp, right.br_state); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); - - ip->i_df.if_lastex = idx; + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); break; case 0: @@ -1780,14 +1751,13 @@ xfs_bmap_add_extent_hole_delay( * Insert a new entry. */ oldlen = newlen = 0; - xfs_iext_insert(ip, idx, 1, new, state); - ip->i_df.if_lastex = idx; + xfs_iext_insert(ip, *idx, 1, new, state); break; } if (oldlen != newlen) { ASSERT(oldlen > newlen); xfs_icsb_modify_counters(ip->i_mount, XFS_SBS_FDBLOCKS, - (int64_t)(oldlen - newlen), rsvd); + (int64_t)(oldlen - newlen), 0); /* * Nothing to do for disk quota accounting here. */ @@ -1803,13 +1773,12 @@ xfs_bmap_add_extent_hole_delay( STATIC int /* error */ xfs_bmap_add_extent_hole_real( xfs_inode_t *ip, /* incore inode pointer */ - xfs_extnum_t idx, /* extent number to update/insert */ + xfs_extnum_t *idx, /* extent number to update/insert */ xfs_btree_cur_t *cur, /* if null, not a btree */ xfs_bmbt_irec_t *new, /* new data to add to file extents */ int *logflagsp, /* inode logging flags */ int whichfork) /* data or attr fork */ { - xfs_bmbt_rec_host_t *ep; /* pointer to extent entry ins. point */ int error; /* error return value */ int i; /* temp state */ xfs_ifork_t *ifp; /* inode fork pointer */ @@ -1819,8 +1788,7 @@ xfs_bmap_add_extent_hole_real( int state; /* state bits, accessed thru macros */ ifp = XFS_IFORK_PTR(ip, whichfork); - ASSERT(idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)); - ep = xfs_iext_get_ext(ifp, idx); + ASSERT(*idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)); state = 0; if (whichfork == XFS_ATTR_FORK) @@ -1829,9 +1797,9 @@ xfs_bmap_add_extent_hole_real( /* * Check and set flags if this segment has a left neighbor. */ - if (idx > 0) { + if (*idx > 0) { state |= BMAP_LEFT_VALID; - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left); if (isnullstartblock(left.br_startblock)) state |= BMAP_LEFT_DELAY; } @@ -1840,9 +1808,9 @@ xfs_bmap_add_extent_hole_real( * Check and set flags if this segment has a current value. * Not true if we're inserting into the "hole" at eof. */ - if (idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) { + if (*idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) { state |= BMAP_RIGHT_VALID; - xfs_bmbt_get_all(ep, &right); + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right); if (isnullstartblock(right.br_startblock)) state |= BMAP_RIGHT_DELAY; } @@ -1879,14 +1847,15 @@ xfs_bmap_add_extent_hole_real( * left and on the right. * Merge all three into a single extent record. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), left.br_blockcount + new->br_blockcount + right.br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + + xfs_iext_remove(ip, *idx + 1, 1, state); - xfs_iext_remove(ip, idx, 1, state); - ifp->if_lastex = idx - 1; XFS_IFORK_NEXT_SET(ip, whichfork, XFS_IFORK_NEXTENTS(ip, whichfork) - 1); if (cur == NULL) { @@ -1921,12 +1890,12 @@ xfs_bmap_add_extent_hole_real( * on the left. * Merge the new allocation with the left neighbor. */ - trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_); - xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), + --*idx; + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), left.br_blockcount + new->br_blockcount); - trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ifp->if_lastex = idx - 1; if (cur == NULL) { rval = xfs_ilog_fext(whichfork); } else { @@ -1952,13 +1921,13 @@ xfs_bmap_add_extent_hole_real( * on the right. * Merge the new allocation with the right neighbor. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); - xfs_bmbt_set_allf(ep, new->br_startoff, new->br_startblock, + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); + xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx), + new->br_startoff, new->br_startblock, new->br_blockcount + right.br_blockcount, right.br_state); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); - ifp->if_lastex = idx; if (cur == NULL) { rval = xfs_ilog_fext(whichfork); } else { @@ -1984,8 +1953,7 @@ xfs_bmap_add_extent_hole_real( * real allocation. * Insert a new entry. */ - xfs_iext_insert(ip, idx, 1, new, state); - ifp->if_lastex = idx; + xfs_iext_insert(ip, *idx, 1, new, state); XFS_IFORK_NEXT_SET(ip, whichfork, XFS_IFORK_NEXTENTS(ip, whichfork) + 1); if (cur == NULL) { @@ -2833,13 +2801,12 @@ STATIC int /* error */ xfs_bmap_del_extent( xfs_inode_t *ip, /* incore inode pointer */ xfs_trans_t *tp, /* current transaction pointer */ - xfs_extnum_t idx, /* extent number to update/delete */ + xfs_extnum_t *idx, /* extent number to update/delete */ xfs_bmap_free_t *flist, /* list of extents to be freed */ xfs_btree_cur_t *cur, /* if null, not a btree */ xfs_bmbt_irec_t *del, /* data to remove from extents */ int *logflagsp, /* inode logging flags */ - int whichfork, /* data or attr fork */ - int rsvd) /* OK to allocate reserved blocks */ + int whichfork) /* data or attr fork */ { xfs_filblks_t da_new; /* new delay-alloc indirect blocks */ xfs_filblks_t da_old; /* old delay-alloc indirect blocks */ @@ -2870,10 +2837,10 @@ xfs_bmap_del_extent( mp = ip->i_mount; ifp = XFS_IFORK_PTR(ip, whichfork); - ASSERT((idx >= 0) && (idx < ifp->if_bytes / + ASSERT((*idx >= 0) && (*idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))); ASSERT(del->br_blockcount > 0); - ep = xfs_iext_get_ext(ifp, idx); + ep = xfs_iext_get_ext(ifp, *idx); xfs_bmbt_get_all(ep, &got); ASSERT(got.br_startoff <= del->br_startoff); del_endoff = del->br_startoff + del->br_blockcount; @@ -2947,11 +2914,12 @@ xfs_bmap_del_extent( /* * Matches the whole extent. Delete the entry. */ - xfs_iext_remove(ip, idx, 1, + xfs_iext_remove(ip, *idx, 1, whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0); - ifp->if_lastex = idx; + --*idx; if (delay) break; + XFS_IFORK_NEXT_SET(ip, whichfork, XFS_IFORK_NEXTENTS(ip, whichfork) - 1); flags |= XFS_ILOG_CORE; @@ -2968,21 +2936,20 @@ xfs_bmap_del_extent( /* * Deleting the first part of the extent. */ - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_startoff(ep, del_endoff); temp = got.br_blockcount - del->br_blockcount; xfs_bmbt_set_blockcount(ep, temp); - ifp->if_lastex = idx; if (delay) { temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), da_old); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); da_new = temp; break; } xfs_bmbt_set_startblock(ep, del_endblock); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); if (!cur) { flags |= xfs_ilog_fext(whichfork); break; @@ -2998,18 +2965,17 @@ xfs_bmap_del_extent( * Deleting the last part of the extent. */ temp = got.br_blockcount - del->br_blockcount; - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); - ifp->if_lastex = idx; if (delay) { temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp), da_old); xfs_bmbt_set_startblock(ep, nullstartblock((int)temp)); - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); da_new = temp; break; } - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); if (!cur) { flags |= xfs_ilog_fext(whichfork); break; @@ -3026,7 +2992,7 @@ xfs_bmap_del_extent( * Deleting the middle of the extent. */ temp = del->br_startoff - got.br_startoff; - trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_); + trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_); xfs_bmbt_set_blockcount(ep, temp); new.br_startoff = del_endoff; temp2 = got_endoff - del_endoff; @@ -3113,9 +3079,9 @@ xfs_bmap_del_extent( } } } - trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_); - xfs_iext_insert(ip, idx + 1, 1, &new, state); - ifp->if_lastex = idx + 1; + trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_); + xfs_iext_insert(ip, *idx + 1, 1, &new, state); + ++*idx; break; } /* @@ -3142,7 +3108,7 @@ xfs_bmap_del_extent( ASSERT(da_old >= da_new); if (da_old > da_new) { xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, - (int64_t)(da_old - da_new), rsvd); + (int64_t)(da_old - da_new), 0); } done: *logflagsp = flags; @@ -4562,29 +4528,24 @@ xfs_bmapi( if (rt) { error = xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS, - -((int64_t)extsz), (flags & - XFS_BMAPI_RSVBLOCKS)); + -((int64_t)extsz), 0); } else { error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, - -((int64_t)alen), (flags & - XFS_BMAPI_RSVBLOCKS)); + -((int64_t)alen), 0); } if (!error) { error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, - -((int64_t)indlen), (flags & - XFS_BMAPI_RSVBLOCKS)); + -((int64_t)indlen), 0); if (error && rt) xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS, - (int64_t)extsz, (flags & - XFS_BMAPI_RSVBLOCKS)); + (int64_t)extsz, 0); else if (error) xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, - (int64_t)alen, (flags & - XFS_BMAPI_RSVBLOCKS)); + (int64_t)alen, 0); } if (error) { @@ -4701,13 +4662,12 @@ xfs_bmapi( if (!wasdelay && (flags & XFS_BMAPI_PREALLOC)) got.br_state = XFS_EXT_UNWRITTEN; } - error = xfs_bmap_add_extent(ip, lastx, &cur, &got, + error = xfs_bmap_add_extent(ip, &lastx, &cur, &got, firstblock, flist, &tmp_logflags, - whichfork, (flags & XFS_BMAPI_RSVBLOCKS)); + whichfork); logflags |= tmp_logflags; if (error) goto error0; - lastx = ifp->if_lastex; ep = xfs_iext_get_ext(ifp, lastx); nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); xfs_bmbt_get_all(ep, &got); @@ -4803,13 +4763,12 @@ xfs_bmapi( mval->br_state = (mval->br_state == XFS_EXT_UNWRITTEN) ? XFS_EXT_NORM : XFS_EXT_UNWRITTEN; - error = xfs_bmap_add_extent(ip, lastx, &cur, mval, + error = xfs_bmap_add_extent(ip, &lastx, &cur, mval, firstblock, flist, &tmp_logflags, - whichfork, (flags & XFS_BMAPI_RSVBLOCKS)); + whichfork); logflags |= tmp_logflags; if (error) goto error0; - lastx = ifp->if_lastex; ep = xfs_iext_get_ext(ifp, lastx); nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); xfs_bmbt_get_all(ep, &got); @@ -4868,14 +4827,14 @@ xfs_bmapi( /* * Else go on to the next record. */ - ep = xfs_iext_get_ext(ifp, ++lastx); prev = got; - if (lastx >= nextents) - eof = 1; - else + if (++lastx < nextents) { + ep = xfs_iext_get_ext(ifp, lastx); xfs_bmbt_get_all(ep, &got); + } else { + eof = 1; + } } - ifp->if_lastex = lastx; *nmap = n; /* * Transform from btree to extents, give it cur. @@ -4984,7 +4943,6 @@ xfs_bmapi_single( ASSERT(!isnullstartblock(got.br_startblock)); ASSERT(bno < got.br_startoff + got.br_blockcount); *fsb = got.br_startblock + (bno - got.br_startoff); - ifp->if_lastex = lastx; return 0; } @@ -5026,7 +4984,6 @@ xfs_bunmapi( int tmp_logflags; /* partial logging flags */ int wasdel; /* was a delayed alloc extent */ int whichfork; /* data or attribute fork */ - int rsvd; /* OK to allocate reserved blocks */ xfs_fsblock_t sum; trace_xfs_bunmap(ip, bno, len, flags, _RET_IP_); @@ -5044,7 +5001,7 @@ xfs_bunmapi( mp = ip->i_mount; if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); - rsvd = (flags & XFS_BMAPI_RSVBLOCKS) != 0; + ASSERT(len > 0); ASSERT(nexts >= 0); ASSERT(ifp->if_ext_max == @@ -5160,9 +5117,9 @@ xfs_bunmapi( del.br_blockcount = mod; } del.br_state = XFS_EXT_UNWRITTEN; - error = xfs_bmap_add_extent(ip, lastx, &cur, &del, + error = xfs_bmap_add_extent(ip, &lastx, &cur, &del, firstblock, flist, &logflags, - XFS_DATA_FORK, 0); + XFS_DATA_FORK); if (error) goto error0; goto nodelete; @@ -5188,9 +5145,12 @@ xfs_bunmapi( */ ASSERT(bno >= del.br_blockcount); bno -= del.br_blockcount; - if (bno < got.br_startoff) { - if (--lastx >= 0) - xfs_bmbt_get_all(--ep, &got); + if (got.br_startoff > bno) { + if (--lastx >= 0) { + ep = xfs_iext_get_ext(ifp, + lastx); + xfs_bmbt_get_all(ep, &got); + } } continue; } else if (del.br_state == XFS_EXT_UNWRITTEN) { @@ -5214,18 +5174,19 @@ xfs_bunmapi( prev.br_startoff = start; } prev.br_state = XFS_EXT_UNWRITTEN; - error = xfs_bmap_add_extent(ip, lastx - 1, &cur, + lastx--; + error = xfs_bmap_add_extent(ip, &lastx, &cur, &prev, firstblock, flist, &logflags, - XFS_DATA_FORK, 0); + XFS_DATA_FORK); if (error) goto error0; goto nodelete; } else { ASSERT(del.br_state == XFS_EXT_NORM); del.br_state = XFS_EXT_UNWRITTEN; - error = xfs_bmap_add_extent(ip, lastx, &cur, + error = xfs_bmap_add_extent(ip, &lastx, &cur, &del, firstblock, flist, &logflags, - XFS_DATA_FORK, 0); + XFS_DATA_FORK); if (error) goto error0; goto nodelete; @@ -5240,13 +5201,13 @@ xfs_bunmapi( rtexts = XFS_FSB_TO_B(mp, del.br_blockcount); do_div(rtexts, mp->m_sb.sb_rextsize); xfs_mod_incore_sb(mp, XFS_SBS_FREXTENTS, - (int64_t)rtexts, rsvd); + (int64_t)rtexts, 0); (void)xfs_trans_reserve_quota_nblks(NULL, ip, -((long)del.br_blockcount), 0, XFS_QMOPT_RES_RTBLKS); } else { xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, - (int64_t)del.br_blockcount, rsvd); + (int64_t)del.br_blockcount, 0); (void)xfs_trans_reserve_quota_nblks(NULL, ip, -((long)del.br_blockcount), 0, XFS_QMOPT_RES_REGBLKS); @@ -5277,31 +5238,29 @@ xfs_bunmapi( error = XFS_ERROR(ENOSPC); goto error0; } - error = xfs_bmap_del_extent(ip, tp, lastx, flist, cur, &del, - &tmp_logflags, whichfork, rsvd); + error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del, + &tmp_logflags, whichfork); logflags |= tmp_logflags; if (error) goto error0; bno = del.br_startoff - 1; nodelete: - lastx = ifp->if_lastex; /* * If not done go on to the next (previous) record. - * Reset ep in case the extents array was re-alloced. */ - ep = xfs_iext_get_ext(ifp, lastx); if (bno != (xfs_fileoff_t)-1 && bno >= start) { - if (lastx >= XFS_IFORK_NEXTENTS(ip, whichfork) || - xfs_bmbt_get_startoff(ep) > bno) { - if (--lastx >= 0) - ep = xfs_iext_get_ext(ifp, lastx); - } - if (lastx >= 0) + if (lastx >= 0) { + ep = xfs_iext_get_ext(ifp, lastx); + if (xfs_bmbt_get_startoff(ep) > bno) { + if (--lastx >= 0) + ep = xfs_iext_get_ext(ifp, + lastx); + } xfs_bmbt_get_all(ep, &got); + } extno++; } } - ifp->if_lastex = lastx; *done = bno == (xfs_fileoff_t)-1 || bno < start || lastx < 0; ASSERT(ifp->if_ext_max == XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t)); diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 3651191..c62234b 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -69,7 +69,6 @@ typedef struct xfs_bmap_free #define XFS_BMAPI_ENTIRE 0x004 /* return entire extent, not trimmed */ #define XFS_BMAPI_METADATA 0x008 /* mapping metadata not user data */ #define XFS_BMAPI_ATTRFORK 0x010 /* use attribute fork not data */ -#define XFS_BMAPI_RSVBLOCKS 0x020 /* OK to alloc. reserved data blocks */ #define XFS_BMAPI_PREALLOC 0x040 /* preallocation op: unwritten space */ #define XFS_BMAPI_IGSTATE 0x080 /* Ignore state - */ /* combine contig. space */ @@ -87,7 +86,6 @@ typedef struct xfs_bmap_free { XFS_BMAPI_ENTIRE, "ENTIRE" }, \ { XFS_BMAPI_METADATA, "METADATA" }, \ { XFS_BMAPI_ATTRFORK, "ATTRFORK" }, \ - { XFS_BMAPI_RSVBLOCKS, "RSVBLOCKS" }, \ { XFS_BMAPI_PREALLOC, "PREALLOC" }, \ { XFS_BMAPI_IGSTATE, "IGSTATE" }, \ { XFS_BMAPI_CONTIG, "CONTIG" }, \ diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index be628677..9a84a85 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -202,7 +202,7 @@ xfs_swap_extents( xfs_inode_t *tip, /* tmp inode */ xfs_swapext_t *sxp) { - xfs_mount_t *mp; + xfs_mount_t *mp = ip->i_mount; xfs_trans_t *tp; xfs_bstat_t *sbp = &sxp->sx_stat; xfs_ifork_t *tempifp, *ifp, *tifp; @@ -212,16 +212,12 @@ xfs_swap_extents( int taforkblks = 0; __uint64_t tmp; - mp = ip->i_mount; - tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL); if (!tempifp) { error = XFS_ERROR(ENOMEM); goto out; } - sbp = &sxp->sx_stat; - /* * we have to do two separate lock calls here to keep lockdep * happy. If we try to get all the locks in one call, lock will diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index cb9b6d1..3631783 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -253,16 +253,21 @@ xfs_iget_cache_hit( rcu_read_lock(); spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~XFS_INEW; - ip->i_flags |= XFS_IRECLAIMABLE; - __xfs_inode_set_reclaim_tag(pag, ip); + ip->i_flags &= ~(XFS_INEW | XFS_IRECLAIM); + ASSERT(ip->i_flags & XFS_IRECLAIMABLE); trace_xfs_iget_reclaim_fail(ip); goto out_error; } spin_lock(&pag->pag_ici_lock); spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM); + + /* + * Clear the per-lifetime state in the inode as we are now + * effectively a new inode and need to return to the initial + * state before reuse occurs. + */ + ip->i_flags &= ~XFS_IRECLAIM_RESET_FLAGS; ip->i_flags |= XFS_INEW; __xfs_inode_clear_reclaim_tag(mp, pag, ip); inode->i_state = I_NEW; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index a37480a..a098a20 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -920,7 +920,6 @@ xfs_iread_extents( /* * We know that the size is valid (it's checked in iformat_btree) */ - ifp->if_lastex = NULLEXTNUM; ifp->if_bytes = ifp->if_real_bytes = 0; ifp->if_flags |= XFS_IFEXTENTS; xfs_iext_add(ifp, 0, nextents); @@ -1354,7 +1353,7 @@ xfs_itruncate_start( return 0; } last_byte = xfs_file_last_byte(ip); - trace_xfs_itruncate_start(ip, flags, new_size, toss_start, last_byte); + trace_xfs_itruncate_start(ip, new_size, flags, toss_start, last_byte); if (last_byte > toss_start) { if (flags & XFS_ITRUNC_DEFINITE) { xfs_tosspages(ip, toss_start, @@ -1470,7 +1469,7 @@ xfs_itruncate_finish( * file but the log buffers containing the free and reallocation * don't, then we'd end up with garbage in the blocks being freed. * As long as we make the new_size permanent before actually - * freeing any blocks it doesn't matter if they get writtten to. + * freeing any blocks it doesn't matter if they get written to. * * The callers must signal into us whether or not the size * setting here must be synchronous. There are a few cases @@ -2558,12 +2557,9 @@ xfs_iflush_fork( case XFS_DINODE_FMT_EXTENTS: ASSERT((ifp->if_flags & XFS_IFEXTENTS) || !(iip->ili_format.ilf_fields & extflag[whichfork])); - ASSERT((xfs_iext_get_ext(ifp, 0) != NULL) || - (ifp->if_bytes == 0)); - ASSERT((xfs_iext_get_ext(ifp, 0) == NULL) || - (ifp->if_bytes > 0)); if ((iip->ili_format.ilf_fields & extflag[whichfork]) && (ifp->if_bytes > 0)) { + ASSERT(xfs_iext_get_ext(ifp, 0)); ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0); (void)xfs_iextents_copy(ip, (xfs_bmbt_rec_t *)cp, whichfork); @@ -3112,6 +3108,8 @@ xfs_iext_get_ext( xfs_extnum_t idx) /* index of target extent */ { ASSERT(idx >= 0); + ASSERT(idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); + if ((ifp->if_flags & XFS_IFEXTIREC) && (idx == 0)) { return ifp->if_u1.if_ext_irec->er_extbuf; } else if (ifp->if_flags & XFS_IFEXTIREC) { @@ -3191,7 +3189,6 @@ xfs_iext_add( } ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; ifp->if_real_bytes = 0; - ifp->if_lastex = nextents + ext_diff; } /* * Otherwise use a linear (direct) extent list. @@ -3886,8 +3883,10 @@ xfs_iext_idx_to_irec( xfs_extnum_t page_idx = *idxp; /* extent index in target list */ ASSERT(ifp->if_flags & XFS_IFEXTIREC); - ASSERT(page_idx >= 0 && page_idx <= - ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)); + ASSERT(page_idx >= 0); + ASSERT(page_idx <= ifp->if_bytes / sizeof(xfs_bmbt_rec_t)); + ASSERT(page_idx < ifp->if_bytes / sizeof(xfs_bmbt_rec_t) || realloc); + nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; erp_idx = 0; low = 0; diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index ff4e2a3..964cfea 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -67,7 +67,6 @@ typedef struct xfs_ifork { short if_broot_bytes; /* bytes allocated for root */ unsigned char if_flags; /* per-fork flags */ unsigned char if_ext_max; /* max # of extent records */ - xfs_extnum_t if_lastex; /* last if_extents used */ union { xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */ xfs_ext_irec_t *if_ext_irec; /* irec map file exts */ @@ -385,6 +384,16 @@ static inline void xfs_ifunlock(xfs_inode_t *ip) #define XFS_IDIRTY_RELEASE 0x0040 /* dirty release already seen */ /* + * Per-lifetime flags need to be reset when re-using a reclaimable inode during + * inode lookup. Thi prevents unintended behaviour on the new inode from + * ocurring. + */ +#define XFS_IRECLAIM_RESET_FLAGS \ + (XFS_IRECLAIMABLE | XFS_IRECLAIM | \ + XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \ + XFS_IFILESTREAM); + +/* * Flags for inode locking. * Bit ranges: 1<<1 - 1<<16-1 -- iolock/ilock modes (bitfield) * 1<<16 - 1<<32-1 -- lockdep annotation (integers) diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 576fdfe..09983a3 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -970,7 +970,6 @@ xfs_iflush_abort( { xfs_inode_log_item_t *iip = ip->i_itemp; - iip = ip->i_itemp; if (iip) { struct xfs_ail *ailp = iip->ili_item.li_ailp; if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index b612ce4..41d5b8f 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1372,8 +1372,17 @@ xlog_sync(xlog_t *log, XFS_BUF_ASYNC(bp); bp->b_flags |= XBF_LOG_BUFFER; - if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) + if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) { + /* + * If we have an external log device, flush the data device + * before flushing the log to make sure all meta data + * written back from the AIL actually made it to disk + * before writing out the new log tail LSN in the log buffer. + */ + if (log->l_mp->m_logdev_targp != log->l_mp->m_ddev_targp) + xfs_blkdev_issue_flush(log->l_mp->m_ddev_targp); XFS_BUF_ORDERED(bp); + } ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1); ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize); @@ -1449,6 +1458,13 @@ xlog_dealloc_log(xlog_t *log) xlog_cil_destroy(log); + /* + * always need to ensure that the extra buffer does not point to memory + * owned by another log buffer before we free it. + */ + xfs_buf_set_empty(log->l_xbuf, log->l_iclog_size); + xfs_buf_free(log->l_xbuf); + iclog = log->l_iclog; for (i=0; i<log->l_iclog_bufs; i++) { xfs_buf_free(iclog->ic_bp); @@ -1458,7 +1474,6 @@ xlog_dealloc_log(xlog_t *log) } spinlock_destroy(&log->l_icloglock); - xfs_buf_free(log->l_xbuf); log->l_mp->m_log = NULL; kmem_free(log); } /* xlog_dealloc_log */ @@ -3248,13 +3263,6 @@ xfs_log_ticket_get( return ticket; } -xlog_tid_t -xfs_log_get_trans_ident( - struct xfs_trans *tp) -{ - return tp->t_ticket->t_tid; -} - /* * Allocate and initialise a new log ticket. */ diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 3bd3291..78c9039 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -189,8 +189,6 @@ void xlog_iodone(struct xfs_buf *); struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket); void xfs_log_ticket_put(struct xlog_ticket *ticket); -xlog_tid_t xfs_log_get_trans_ident(struct xfs_trans *tp); - void xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp, struct xfs_log_vec *log_vector, xfs_lsn_t *commit_lsn, int flags); diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index 9ca59be..c7755d5 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -29,6 +29,7 @@ #include "xfs_mount.h" #include "xfs_error.h" #include "xfs_alloc.h" +#include "xfs_discard.h" /* * Perform initial CIL structure initialisation. If the CIL is not @@ -361,19 +362,28 @@ xlog_cil_committed( int abort) { struct xfs_cil_ctx *ctx = args; - struct xfs_busy_extent *busyp, *n; + struct xfs_mount *mp = ctx->cil->xc_log->l_mp; xfs_trans_committed_bulk(ctx->cil->xc_log->l_ailp, ctx->lv_chain, ctx->start_lsn, abort); - list_for_each_entry_safe(busyp, n, &ctx->busy_extents, list) - xfs_alloc_busy_clear(ctx->cil->xc_log->l_mp, busyp); + xfs_alloc_busy_sort(&ctx->busy_extents); + xfs_alloc_busy_clear(mp, &ctx->busy_extents, + (mp->m_flags & XFS_MOUNT_DISCARD) && !abort); spin_lock(&ctx->cil->xc_cil_lock); list_del(&ctx->committing); spin_unlock(&ctx->cil->xc_cil_lock); xlog_cil_free_logvec(ctx->lv_chain); + + if (!list_empty(&ctx->busy_extents)) { + ASSERT(mp->m_flags & XFS_MOUNT_DISCARD); + + xfs_discard_extents(mp, &ctx->busy_extents); + xfs_alloc_busy_clear(mp, &ctx->busy_extents, false); + } + kmem_free(ctx); } diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 5864850..2d3b6a4 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -146,6 +146,8 @@ static inline uint xlog_get_client_id(__be32 i) shutdown */ #define XLOG_TAIL_WARN 0x10 /* log tail verify warning issued */ +typedef __uint32_t xlog_tid_t; + #ifdef __KERNEL__ /* * Below are states for covering allocation transactions. diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 5cc464a..04142ca 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -205,6 +205,35 @@ xlog_bread( } /* + * Read at an offset into the buffer. Returns with the buffer in it's original + * state regardless of the result of the read. + */ +STATIC int +xlog_bread_offset( + xlog_t *log, + xfs_daddr_t blk_no, /* block to read from */ + int nbblks, /* blocks to read */ + xfs_buf_t *bp, + xfs_caddr_t offset) +{ + xfs_caddr_t orig_offset = XFS_BUF_PTR(bp); + int orig_len = bp->b_buffer_length; + int error, error2; + + error = XFS_BUF_SET_PTR(bp, offset, BBTOB(nbblks)); + if (error) + return error; + + error = xlog_bread_noalign(log, blk_no, nbblks, bp); + + /* must reset buffer pointer even on error */ + error2 = XFS_BUF_SET_PTR(bp, orig_offset, orig_len); + if (error) + return error; + return error2; +} + +/* * Write out the buffer at the given block for the given number of blocks. * The buffer is kept locked across the write and is returned locked. * This can only be used for synchronous log writes. @@ -1229,20 +1258,12 @@ xlog_write_log_records( */ ealign = round_down(end_block, sectbb); if (j == 0 && (start_block + endcount > ealign)) { - offset = XFS_BUF_PTR(bp); - balign = BBTOB(ealign - start_block); - error = XFS_BUF_SET_PTR(bp, offset + balign, - BBTOB(sectbb)); + offset = XFS_BUF_PTR(bp) + BBTOB(ealign - start_block); + error = xlog_bread_offset(log, ealign, sectbb, + bp, offset); if (error) break; - error = xlog_bread_noalign(log, ealign, sectbb, bp); - if (error) - break; - - error = XFS_BUF_SET_PTR(bp, offset, bufblks); - if (error) - break; } offset = xlog_align(log, start_block, endcount, bp); @@ -3448,19 +3469,9 @@ xlog_do_recovery_pass( * - order is important. */ wrapped_hblks = hblks - split_hblks; - error = XFS_BUF_SET_PTR(hbp, - offset + BBTOB(split_hblks), - BBTOB(hblks - split_hblks)); - if (error) - goto bread_err2; - - error = xlog_bread_noalign(log, 0, - wrapped_hblks, hbp); - if (error) - goto bread_err2; - - error = XFS_BUF_SET_PTR(hbp, offset, - BBTOB(hblks)); + error = xlog_bread_offset(log, 0, + wrapped_hblks, hbp, + offset + BBTOB(split_hblks)); if (error) goto bread_err2; } @@ -3511,19 +3522,9 @@ xlog_do_recovery_pass( * _first_, then the log start (LR header end) * - order is important. */ - error = XFS_BUF_SET_PTR(dbp, - offset + BBTOB(split_bblks), - BBTOB(bblks - split_bblks)); - if (error) - goto bread_err2; - - error = xlog_bread_noalign(log, wrapped_hblks, - bblks - split_bblks, - dbp); - if (error) - goto bread_err2; - - error = XFS_BUF_SET_PTR(dbp, offset, h_size); + error = xlog_bread_offset(log, 0, + bblks - split_bblks, hbp, + offset + BBTOB(split_bblks)); if (error) goto bread_err2; } diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index bb3f9a7..b49b823 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1900,7 +1900,7 @@ xfs_mod_incore_sb_batch( uint nmsb, int rsvd) { - xfs_mod_sb_t *msbp = &msb[0]; + xfs_mod_sb_t *msbp; int error = 0; /* @@ -1910,7 +1910,7 @@ xfs_mod_incore_sb_batch( * changes will be atomic. */ spin_lock(&mp->m_sb_lock); - for (msbp = &msbp[0]; msbp < (msb + nmsb); msbp++) { + for (msbp = msb; msbp < (msb + nmsb); msbp++) { ASSERT(msbp->msb_field < XFS_SBS_ICOUNT || msbp->msb_field > XFS_SBS_FDBLOCKS); diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 19af0ab..3d68bb2 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -224,6 +224,7 @@ typedef struct xfs_mount { #define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem operations, typically for disk errors in metadata */ +#define XFS_MOUNT_DISCARD (1ULL << 5) /* discard unused blocks */ #define XFS_MOUNT_RETERR (1ULL << 6) /* return alignment errors to user */ #define XFS_MOUNT_NOALIGN (1ULL << 7) /* turn off stripe alignment diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 7692279..7c7bc2b 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -608,10 +608,8 @@ STATIC void xfs_trans_free( struct xfs_trans *tp) { - struct xfs_busy_extent *busyp, *n; - - list_for_each_entry_safe(busyp, n, &tp->t_busy, list) - xfs_alloc_busy_clear(tp->t_mountp, busyp); + xfs_alloc_busy_sort(&tp->t_busy); + xfs_alloc_busy_clear(tp->t_mountp, &tp->t_busy, false); atomic_dec(&tp->t_mountp->m_active_trans); xfs_trans_free_dqinfo(tp); diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h index 26d1867..65584b5 100644 --- a/fs/xfs/xfs_types.h +++ b/fs/xfs/xfs_types.h @@ -73,8 +73,6 @@ typedef __int32_t xfs_tid_t; /* transaction identifier */ typedef __uint32_t xfs_dablk_t; /* dir/attr block number (in file) */ typedef __uint32_t xfs_dahash_t; /* dir/attr hash value */ -typedef __uint32_t xlog_tid_t; /* transaction ID type */ - /* * These types are 64 bits on disk but are either 32 or 64 bits in memory. * Disk based types: diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index b7a5fe7..6197207 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -960,8 +960,11 @@ xfs_release( * be exposed to that problem. */ truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED); - if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) - xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE); + if (truncated) { + xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE); + if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) + xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE); + } } if (ip->i_d.di_nlink == 0) |
