diff options
| author | Ingo Molnar <mingo@elte.hu> | 2008-10-13 11:05:51 +0200 |
|---|---|---|
| committer | Ingo Molnar <mingo@elte.hu> | 2008-10-13 11:05:51 +0200 |
| commit | accba5f3965d6a9d1bf7c1e1a7995d17e9d521b6 (patch) | |
| tree | 8fb40782e79472ed882ff2098d4dd295557278ee /fs | |
| parent | 6852fd9b86d05063c6ef49d2e12e061cc7f6a105 (diff) | |
| parent | 4480f15b3306f43bbb0310d461142b4e897ca45b (diff) | |
| download | op-kernel-dev-accba5f3965d6a9d1bf7c1e1a7995d17e9d521b6.zip op-kernel-dev-accba5f3965d6a9d1bf7c1e1a7995d17e9d521b6.tar.gz | |
Merge branch 'linus' into oprofile-v2
Conflicts:
arch/x86/kernel/apic_32.c
arch/x86/oprofile/nmi_int.c
include/linux/pci_ids.h
Diffstat (limited to 'fs')
397 files changed, 15084 insertions, 12816 deletions
diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index 88e3787..e298fe1 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -119,6 +119,7 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) const struct file_operations v9fs_dir_operations = { .read = generic_read_dir, + .llseek = generic_file_llseek, .readdir = v9fs_dir_readdir, .open = v9fs_file_open, .release = v9fs_dir_release, diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index c95295c..e83aa5e 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -626,8 +626,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, return NULL; error: - if (fid) - p9_client_clunk(fid); + p9_client_clunk(fid); return ERR_PTR(result); } @@ -136,37 +136,51 @@ config EXT3_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. -config EXT4DEV_FS - tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)" - depends on EXPERIMENTAL +config EXT4_FS + tristate "The Extended 4 (ext4) filesystem" select JBD2 select CRC16 help - Ext4dev is a predecessor filesystem of the next generation - extended fs ext4, based on ext3 filesystem code. It will be - renamed ext4 fs later, once ext4dev is mature and stabilized. + This is the next generation of the ext3 filesystem. Unlike the change from ext2 filesystem to ext3 filesystem, - the on-disk format of ext4dev is not the same as ext3 any more: - it is based on extent maps and it supports 48-bit physical block - numbers. These combined on-disk format changes will allow - ext4dev/ext4 to handle more than 16 TB filesystem volumes -- - a hard limit that ext3 cannot overcome without changing the - on-disk format. - - Other than extent maps and 48-bit block numbers, ext4dev also is - likely to have other new features such as persistent preallocation, - high resolution time stamps, and larger file support etc. These - features will be added to ext4dev gradually. + the on-disk format of ext4 is not forwards compatible with + ext3; it is based on extent maps and it supports 48-bit + physical block numbers. The ext4 filesystem also supports delayed + allocation, persistent preallocation, high resolution time stamps, + and a number of other features to improve performance and speed + up fsck time. For more information, please see the web pages at + http://ext4.wiki.kernel.org. + + The ext4 filesystem will support mounting an ext3 + filesystem; while there will be some performance gains from + the delayed allocation and inode table readahead, the best + performance gains will require enabling ext4 features in the + filesystem, or formating a new filesystem as an ext4 + filesystem initially. To compile this file system support as a module, choose M here. The module will be called ext4dev. If unsure, say N. -config EXT4DEV_FS_XATTR - bool "Ext4dev extended attributes" - depends on EXT4DEV_FS +config EXT4DEV_COMPAT + bool "Enable ext4dev compatibility" + depends on EXT4_FS + help + Starting with 2.6.28, the name of the ext4 filesystem was + renamed from ext4dev to ext4. Unfortunately there are some + legacy userspace programs (such as klibc's fstype) have + "ext4dev" hardcoded. + + To enable backwards compatibility so that systems that are + still expecting to mount ext4 filesystems using ext4dev, + chose Y here. This feature will go away by 2.6.31, so + please arrange to get your userspace programs fixed! + +config EXT4_FS_XATTR + bool "Ext4 extended attributes" + depends on EXT4_FS default y help Extended attributes are name:value pairs associated with inodes by @@ -175,11 +189,11 @@ config EXT4DEV_FS_XATTR If unsure, say N. - You need this for POSIX ACL support on ext4dev/ext4. + You need this for POSIX ACL support on ext4. -config EXT4DEV_FS_POSIX_ACL - bool "Ext4dev POSIX Access Control Lists" - depends on EXT4DEV_FS_XATTR +config EXT4_FS_POSIX_ACL + bool "Ext4 POSIX Access Control Lists" + depends on EXT4_FS_XATTR select FS_POSIX_ACL help POSIX Access Control Lists (ACLs) support permissions for users and @@ -190,14 +204,14 @@ config EXT4DEV_FS_POSIX_ACL If you don't know what Access Control Lists are, say N -config EXT4DEV_FS_SECURITY - bool "Ext4dev Security Labels" - depends on EXT4DEV_FS_XATTR +config EXT4_FS_SECURITY + bool "Ext4 Security Labels" + depends on EXT4_FS_XATTR help Security labels support alternative access control models implemented by security modules like SELinux. This option enables an extended attribute handler for file security - labels in the ext4dev/ext4 filesystem. + labels in the ext4 filesystem. If you are not using a security module that requires using extended attributes for file security labels, say N. @@ -240,22 +254,22 @@ config JBD2 help This is a generic journaling layer for block devices that support both 32-bit and 64-bit block numbers. It is currently used by - the ext4dev/ext4 filesystem, but it could also be used to add + the ext4 filesystem, but it could also be used to add journal support to other file systems or block devices such as RAID or LVM. - If you are using ext4dev/ext4, you need to say Y here. If you are not - using ext4dev/ext4 then you will probably want to say N. + If you are using ext4, you need to say Y here. If you are not + using ext4 then you will probably want to say N. To compile this device as a module, choose M here. The module will be - called jbd2. If you are compiling ext4dev/ext4 into the kernel, + called jbd2. If you are compiling ext4 into the kernel, you cannot compile this code as a module. config JBD2_DEBUG - bool "JBD2 (ext4dev/ext4) debugging support" + bool "JBD2 (ext4) debugging support" depends on JBD2 && DEBUG_FS help - If you are using the ext4dev/ext4 journaled file system (or + If you are using the ext4 journaled file system (or potentially any other filesystem/device using JBD2), this option allows you to enable debugging output while the system is running, in order to help track down any problems you are having. @@ -270,9 +284,9 @@ config JBD2_DEBUG config FS_MBCACHE # Meta block cache for Extended Attributes (ext2/ext3/ext4) tristate - depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4DEV_FS_XATTR - default y if EXT2_FS=y || EXT3_FS=y || EXT4DEV_FS=y - default m if EXT2_FS=m || EXT3_FS=m || EXT4DEV_FS=m + depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4_FS_XATTR + default y if EXT2_FS=y || EXT3_FS=y || EXT4_FS=y + default m if EXT2_FS=m || EXT3_FS=m || EXT4_FS=m config REISERFS_FS tristate "Reiserfs support" @@ -1383,6 +1397,19 @@ config MINIX_FS partition (the one containing the directory /) cannot be compiled as a module. +config OMFS_FS + tristate "SonicBlue Optimized MPEG File System support" + depends on BLOCK + select CRC_ITU_T + help + This is the proprietary file system used by the Rio Karma music + player and ReplayTV DVR. Despite the name, this filesystem is not + more efficient than a standard FS for MPEG files, in fact likely + the opposite is true. Say Y if you have either of these devices + and wish to mount its disk. + + To compile this file system support as a module, choose M here: the + module will be called omfs. If unsure, say N. config HPFS_FS tristate "OS/2 HPFS file system support" @@ -1917,6 +1944,16 @@ config CIFS_WEAK_PW_HASH If unsure, say N. +config CIFS_UPCALL + bool "Kerberos/SPNEGO advanced session setup" + depends on CIFS && KEYS + help + Enables an upcall mechanism for CIFS which accesses + userspace helper utilities to provide SPNEGO packaged (RFC 4178) + Kerberos tickets which are needed to mount to certain secure servers + (for which more secure Kerberos authentication is required). If + unsure, say N. + config CIFS_XATTR bool "CIFS extended attributes" depends on CIFS @@ -1969,17 +2006,6 @@ config CIFS_EXPERIMENTAL (which is disabled by default). See the file fs/cifs/README for more details. If unsure, say N. -config CIFS_UPCALL - bool "Kerberos/SPNEGO advanced session setup (EXPERIMENTAL)" - depends on CIFS_EXPERIMENTAL - depends on KEYS - help - Enables an upcall mechanism for CIFS which accesses - userspace helper utilities to provide SPNEGO packaged (RFC 4178) - Kerberos tickets which are needed to mount to certain secure servers - (for which more secure Kerberos authentication is required). If - unsure, say N. - config CIFS_DFS_UPCALL bool "DFS feature support (EXPERIMENTAL)" depends on CIFS_EXPERIMENTAL diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 3263084..4a551af 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -30,7 +30,7 @@ config COMPAT_BINFMT_ELF config BINFMT_ELF_FDPIC bool "Kernel support for FDPIC ELF binaries" default y - depends on (FRV || BLACKFIN) + depends on (FRV || BLACKFIN || (SUPERH32 && !MMU)) help ELF FDPIC binaries are based on ELF, but allow the individual load segments of a binary to be located in memory independently of each diff --git a/fs/Makefile b/fs/Makefile index 3b2178b..de404b0 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -69,7 +69,7 @@ obj-$(CONFIG_DLM) += dlm/ # Do not add any filesystems before this line obj-$(CONFIG_REISERFS_FS) += reiserfs/ obj-$(CONFIG_EXT3_FS) += ext3/ # Before ext2 so root fs can be ext3 -obj-$(CONFIG_EXT4DEV_FS) += ext4/ # Before ext2 so root fs can be ext4dev +obj-$(CONFIG_EXT4_FS) += ext4/ # Before ext2 so root fs can be ext4dev obj-$(CONFIG_JBD) += jbd/ obj-$(CONFIG_JBD2) += jbd2/ obj-$(CONFIG_EXT2_FS) += ext2/ @@ -111,6 +111,7 @@ obj-$(CONFIG_ADFS_FS) += adfs/ obj-$(CONFIG_FUSE_FS) += fuse/ obj-$(CONFIG_UDF_FS) += udf/ obj-$(CONFIG_SUN_OPENPROMFS) += openpromfs/ +obj-$(CONFIG_OMFS_FS) += omfs/ obj-$(CONFIG_JFS_FS) += jfs/ obj-$(CONFIG_XFS_FS) += xfs/ obj-$(CONFIG_9P_FS) += 9p/ diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index fc1a8dc..85a30e9 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -197,6 +197,7 @@ out: const struct file_operations adfs_dir_operations = { .read = generic_read_dir, + .llseek = generic_file_llseek, .readdir = adfs_readdir, .fsync = file_fsync, }; diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 9e421ee..26f3b43 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -249,7 +249,7 @@ static void adfs_destroy_inode(struct inode *inode) kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; diff --git a/fs/affs/affs.h b/fs/affs/affs.h index 223b191..e9ec915 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -2,6 +2,7 @@ #include <linux/fs.h> #include <linux/buffer_head.h> #include <linux/amigaffs.h> +#include <linux/mutex.h> /* AmigaOS allows file names with up to 30 characters length. * Names longer than that will be silently truncated. If you @@ -98,7 +99,7 @@ struct affs_sb_info { gid_t s_gid; /* gid to override */ umode_t s_mode; /* mode to override */ struct buffer_head *s_root_bh; /* Cached root block. */ - struct semaphore s_bmlock; /* Protects bitmap access. */ + struct mutex s_bmlock; /* Protects bitmap access. */ struct affs_bm_info *s_bitmap; /* Bitmap infos. */ u32 s_bmap_count; /* # of bitmap blocks. */ u32 s_bmap_bits; /* # of bits in one bitmap blocks */ diff --git a/fs/affs/bitmap.c b/fs/affs/bitmap.c index c4a5ad0..dc5ef14 100644 --- a/fs/affs/bitmap.c +++ b/fs/affs/bitmap.c @@ -45,14 +45,14 @@ affs_count_free_blocks(struct super_block *sb) if (sb->s_flags & MS_RDONLY) return 0; - down(&AFFS_SB(sb)->s_bmlock); + mutex_lock(&AFFS_SB(sb)->s_bmlock); bm = AFFS_SB(sb)->s_bitmap; free = 0; for (i = AFFS_SB(sb)->s_bmap_count; i > 0; bm++, i--) free += bm->bm_free; - up(&AFFS_SB(sb)->s_bmlock); + mutex_unlock(&AFFS_SB(sb)->s_bmlock); return free; } @@ -76,7 +76,7 @@ affs_free_block(struct super_block *sb, u32 block) bit = blk % sbi->s_bmap_bits; bm = &sbi->s_bitmap[bmap]; - down(&sbi->s_bmlock); + mutex_lock(&sbi->s_bmlock); bh = sbi->s_bmap_bh; if (sbi->s_last_bmap != bmap) { @@ -105,19 +105,19 @@ affs_free_block(struct super_block *sb, u32 block) sb->s_dirt = 1; bm->bm_free++; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_free: affs_warning(sb,"affs_free_block","Trying to free block %u which is already free", block); - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_bh_read: affs_error(sb,"affs_free_block","Cannot read bitmap block %u", bm->bm_key); sbi->s_bmap_bh = NULL; sbi->s_last_bmap = ~0; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); return; err_range: @@ -168,7 +168,7 @@ affs_alloc_block(struct inode *inode, u32 goal) bmap = blk / sbi->s_bmap_bits; bm = &sbi->s_bitmap[bmap]; - down(&sbi->s_bmlock); + mutex_lock(&sbi->s_bmlock); if (bm->bm_free) goto find_bmap_bit; @@ -249,7 +249,7 @@ find_bit: mark_buffer_dirty(bh); sb->s_dirt = 1; - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); pr_debug("%d\n", blk); return blk; @@ -259,7 +259,7 @@ err_bh_read: sbi->s_bmap_bh = NULL; sbi->s_last_bmap = ~0; err_full: - up(&sbi->s_bmlock); + mutex_unlock(&sbi->s_bmlock); pr_debug("failed\n"); return 0; } diff --git a/fs/affs/dir.c b/fs/affs/dir.c index 6e3f282..7b36904 100644 --- a/fs/affs/dir.c +++ b/fs/affs/dir.c @@ -19,6 +19,7 @@ static int affs_readdir(struct file *, void *, filldir_t); const struct file_operations affs_dir_operations = { .read = generic_read_dir, + .llseek = generic_file_llseek, .readdir = affs_readdir, .fsync = file_fsync, }; diff --git a/fs/affs/file.c b/fs/affs/file.c index 6eac7bd..1377b12 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -46,8 +46,6 @@ const struct inode_operations affs_file_inode_operations = { static int affs_file_open(struct inode *inode, struct file *filp) { - if (atomic_read(&filp->f_count) != 1) - return 0; pr_debug("AFFS: open(%lu,%d)\n", inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); atomic_inc(&AFFS_I(inode)->i_opencnt); @@ -57,8 +55,6 @@ affs_file_open(struct inode *inode, struct file *filp) static int affs_file_release(struct inode *inode, struct file *filp) { - if (atomic_read(&filp->f_count) != 0) - return 0; pr_debug("AFFS: release(%lu, %d)\n", inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt)); diff --git a/fs/affs/super.c b/fs/affs/super.c index d214837..3a89094 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -90,7 +90,7 @@ static void affs_destroy_inode(struct inode *inode) kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; @@ -290,7 +290,7 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent) if (!sbi) return -ENOMEM; sb->s_fs_info = sbi; - init_MUTEX(&sbi->s_bmlock); + mutex_init(&sbi->s_bmlock); if (!parse_options(data,&uid,&gid,&i,&reserved,&root_block, &blocksize,&sbi->s_prefix, diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 7102824..3cb6920 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -469,8 +469,6 @@ extern bool afs_cm_incoming_call(struct afs_call *); extern const struct inode_operations afs_dir_inode_operations; extern const struct file_operations afs_dir_file_operations; -extern int afs_permission(struct inode *, int, struct nameidata *); - /* * file.c */ @@ -605,7 +603,7 @@ extern void afs_clear_permits(struct afs_vnode *); extern void afs_cache_permit(struct afs_vnode *, struct key *, long); extern void afs_zap_permits(struct rcu_head *); extern struct key *afs_request_key(struct afs_cell *); -extern int afs_permission(struct inode *, int, struct nameidata *); +extern int afs_permission(struct inode *, int); /* * server.c diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 2f55039..78db495 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -232,7 +232,7 @@ static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd) } mntget(newmnt); - err = do_add_mount(newmnt, nd, MNT_SHRINKABLE, &afs_vfsmounts); + err = do_add_mount(newmnt, &nd->path, MNT_SHRINKABLE, &afs_vfsmounts); switch (err) { case 0: path_put(&nd->path); diff --git a/fs/afs/security.c b/fs/afs/security.c index 3bcbece..3ef5043 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -284,7 +284,7 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask, struct nameidata *nd) +int afs_permission(struct inode *inode, int mask) { struct afs_vnode *vnode = AFS_FS_I(inode); afs_access_t uninitialized_var(access); diff --git a/fs/afs/super.c b/fs/afs/super.c index 7e3faee..250d8c4 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -27,7 +27,7 @@ #define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */ -static void afs_i_init_once(struct kmem_cache *cachep, void *foo); +static void afs_i_init_once(void *foo); static int afs_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt); @@ -449,7 +449,7 @@ static void afs_put_super(struct super_block *sb) /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(struct kmem_cache *cachep, void *_vnode) +static void afs_i_init_once(void *_vnode) { struct afs_vnode *vnode = _vnode; diff --git a/fs/afs/write.c b/fs/afs/write.c index 9a849ad..065b4e1 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -404,7 +404,7 @@ static int afs_write_back_from_locked_page(struct afs_writeback *wb, page = pages[loop]; if (page->index > wb->last) break; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) break; if (!PageDirty(page) || page_private(page) != (unsigned long) wb) { @@ -512,8 +512,8 @@ static void aio_fput_routine(struct work_struct *data) */ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) { - dprintk(KERN_DEBUG "aio_put(%p): f_count=%d\n", - req, atomic_read(&req->ki_filp->f_count)); + dprintk(KERN_DEBUG "aio_put(%p): f_count=%ld\n", + req, atomic_long_read(&req->ki_filp->f_count)); assert_spin_locked(&ctx->ctx_lock); @@ -528,7 +528,7 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) /* Must be done under the lock to serialise against cancellation. * Call this aio_fput as it duplicates fput via the fput_work. */ - if (unlikely(atomic_dec_and_test(&req->ki_filp->f_count))) { + if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) { get_ioctx(ctx); spin_lock(&fput_lock); list_add(&req->ki_list, &fput_head); @@ -51,7 +51,7 @@ int inode_change_ok(struct inode *inode, struct iattr *attr) } /* Check for setting the inode time. */ - if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) { + if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) { if (!is_owner_or_cap(inode)) goto error; } @@ -108,6 +108,11 @@ int notify_change(struct dentry * dentry, struct iattr * attr) struct timespec now; unsigned int ia_valid = attr->ia_valid; + if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_TIMES_SET)) { + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return -EPERM; + } + now = current_fs_time(inode->i_sb); attr->ia_ctime = now; diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index bcfb2dc..2a41c2a 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -36,6 +36,7 @@ const struct file_operations autofs4_root_operations = { .release = dcache_dir_close, .read = generic_read_dir, .readdir = dcache_readdir, + .llseek = dcache_dir_lseek, .ioctl = autofs4_root_ioctl, }; @@ -44,6 +45,7 @@ const struct file_operations autofs4_dir_operations = { .release = dcache_dir_close, .read = generic_read_dir, .readdir = dcache_readdir, + .llseek = dcache_dir_lseek, }; const struct inode_operations autofs4_indirect_root_inode_operations = { diff --git a/fs/bad_inode.c b/fs/bad_inode.c index f1c2ea8..5f1538c 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -243,8 +243,7 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer, return -EIO; } -static int bad_inode_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int bad_inode_permission(struct inode *inode, int mask) { return -EIO; } diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index e8717de..740f536 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -66,6 +66,7 @@ static struct kmem_cache *befs_inode_cachep; static const struct file_operations befs_dir_operations = { .read = generic_read_dir, .readdir = befs_readdir, + .llseek = generic_file_llseek, }; static const struct inode_operations befs_dir_inode_operations = { @@ -289,7 +290,7 @@ befs_destroy_inode(struct inode *inode) kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h index 70f5d3a..7109e45 100644 --- a/fs/bfs/bfs.h +++ b/fs/bfs/bfs.h @@ -16,8 +16,9 @@ struct bfs_sb_info { unsigned long si_freei; unsigned long si_lf_eblk; unsigned long si_lasti; - unsigned long * si_imap; - struct buffer_head * si_sbh; /* buffer header w/superblock */ + unsigned long *si_imap; + struct buffer_head *si_sbh; /* buffer header w/superblock */ + struct mutex bfs_lock; }; /* diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 034950c..ed8feb0 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -32,16 +32,17 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) struct inode *dir = f->f_path.dentry->d_inode; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(dir->i_sb); unsigned int offset; int block; - lock_kernel(); + mutex_lock(&info->bfs_lock); if (f->f_pos & (BFS_DIRENT_SIZE - 1)) { printf("Bad f_pos=%08lx for %s:%08lx\n", (unsigned long)f->f_pos, dir->i_sb->s_id, dir->i_ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return -EBADF; } @@ -61,7 +62,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) le16_to_cpu(de->ino), DT_UNKNOWN) < 0) { brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } } @@ -71,7 +72,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir) brelse(bh); } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } @@ -95,10 +96,10 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, inode = new_inode(s); if (!inode) return -ENOSPC; - lock_kernel(); + mutex_lock(&info->bfs_lock); ino = find_first_zero_bit(info->si_imap, info->si_lasti); if (ino > info->si_lasti) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); iput(inode); return -ENOSPC; } @@ -124,11 +125,11 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode, inode->i_ino); if (err) { inode_dec_link_count(inode); + mutex_unlock(&info->bfs_lock); iput(inode); - unlock_kernel(); return err; } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); d_instantiate(dentry, inode); return 0; } @@ -139,22 +140,23 @@ static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry, struct inode *inode = NULL; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(dir->i_sb); if (dentry->d_name.len > BFS_NAMELEN) return ERR_PTR(-ENAMETOOLONG); - lock_kernel(); + mutex_lock(&info->bfs_lock); bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); if (bh) { unsigned long ino = (unsigned long)le16_to_cpu(de->ino); brelse(bh); inode = bfs_iget(dir->i_sb, ino); if (IS_ERR(inode)) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return ERR_CAST(inode); } } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); d_add(dentry, inode); return NULL; } @@ -163,13 +165,14 @@ static int bfs_link(struct dentry *old, struct inode *dir, struct dentry *new) { struct inode *inode = old->d_inode; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); int err; - lock_kernel(); + mutex_lock(&info->bfs_lock); err = bfs_add_entry(dir, new->d_name.name, new->d_name.len, inode->i_ino); if (err) { - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return err; } inc_nlink(inode); @@ -177,19 +180,19 @@ static int bfs_link(struct dentry *old, struct inode *dir, mark_inode_dirty(inode); atomic_inc(&inode->i_count); d_instantiate(new, inode); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } static int bfs_unlink(struct inode *dir, struct dentry *dentry) { int error = -ENOENT; - struct inode *inode; + struct inode *inode = dentry->d_inode; struct buffer_head *bh; struct bfs_dirent *de; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); - inode = dentry->d_inode; - lock_kernel(); + mutex_lock(&info->bfs_lock); bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de); if (!bh || (le16_to_cpu(de->ino) != inode->i_ino)) goto out_brelse; @@ -210,7 +213,7 @@ static int bfs_unlink(struct inode *dir, struct dentry *dentry) out_brelse: brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return error; } @@ -220,6 +223,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode, *new_inode; struct buffer_head *old_bh, *new_bh; struct bfs_dirent *old_de, *new_de; + struct bfs_sb_info *info; int error = -ENOENT; old_bh = new_bh = NULL; @@ -227,7 +231,9 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (S_ISDIR(old_inode->i_mode)) return -EINVAL; - lock_kernel(); + info = BFS_SB(old_inode->i_sb); + + mutex_lock(&info->bfs_lock); old_bh = bfs_find_entry(old_dir, old_dentry->d_name.name, old_dentry->d_name.len, &old_de); @@ -264,7 +270,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, error = 0; end_rename: - unlock_kernel(); + mutex_unlock(&info->bfs_lock); brelse(old_bh); brelse(new_bh); return error; diff --git a/fs/bfs/file.c b/fs/bfs/file.c index b11e63e..6a02126 100644 --- a/fs/bfs/file.c +++ b/fs/bfs/file.c @@ -99,7 +99,7 @@ static int bfs_get_block(struct inode *inode, sector_t block, return -ENOSPC; /* The rest has to be protected against itself. */ - lock_kernel(); + mutex_lock(&info->bfs_lock); /* * If the last data block for this file is the last allocated @@ -151,7 +151,7 @@ static int bfs_get_block(struct inode *inode, sector_t block, mark_buffer_dirty(sbh); map_bh(bh_result, sb, phys); out: - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return err; } diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 8db6238..0ed57b5 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -104,6 +104,7 @@ static int bfs_write_inode(struct inode *inode, int unused) struct bfs_inode *di; struct buffer_head *bh; int block, off; + struct bfs_sb_info *info = BFS_SB(inode->i_sb); dprintf("ino=%08x\n", ino); @@ -112,13 +113,13 @@ static int bfs_write_inode(struct inode *inode, int unused) return -EIO; } - lock_kernel(); + mutex_lock(&info->bfs_lock); block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; bh = sb_bread(inode->i_sb, block); if (!bh) { printf("Unable to read inode %s:%08x\n", inode->i_sb->s_id, ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return -EIO; } @@ -145,7 +146,7 @@ static int bfs_write_inode(struct inode *inode, int unused) mark_buffer_dirty(bh); brelse(bh); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return 0; } @@ -170,7 +171,7 @@ static void bfs_delete_inode(struct inode *inode) inode->i_size = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; - lock_kernel(); + mutex_lock(&info->bfs_lock); mark_inode_dirty(inode); block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; @@ -178,7 +179,7 @@ static void bfs_delete_inode(struct inode *inode) if (!bh) { printf("Unable to read inode %s:%08lx\n", inode->i_sb->s_id, ino); - unlock_kernel(); + mutex_unlock(&info->bfs_lock); return; } off = (ino - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; @@ -204,14 +205,16 @@ static void bfs_delete_inode(struct inode *inode) info->si_lf_eblk = bi->i_sblock - 1; mark_buffer_dirty(info->si_sbh); } - unlock_kernel(); + mutex_unlock(&info->bfs_lock); clear_inode(inode); } static void bfs_put_super(struct super_block *s) { struct bfs_sb_info *info = BFS_SB(s); + brelse(info->si_sbh); + mutex_destroy(&info->bfs_lock); kfree(info->si_imap); kfree(info); s->s_fs_info = NULL; @@ -236,11 +239,13 @@ static int bfs_statfs(struct dentry *dentry, struct kstatfs *buf) static void bfs_write_super(struct super_block *s) { - lock_kernel(); + struct bfs_sb_info *info = BFS_SB(s); + + mutex_lock(&info->bfs_lock); if (!(s->s_flags & MS_RDONLY)) - mark_buffer_dirty(BFS_SB(s)->si_sbh); + mark_buffer_dirty(info->si_sbh); s->s_dirt = 0; - unlock_kernel(); + mutex_unlock(&info->bfs_lock); } static struct kmem_cache *bfs_inode_cachep; @@ -259,7 +264,7 @@ static void bfs_destroy_inode(struct inode *inode) kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct bfs_inode_info *bi = foo; @@ -380,7 +385,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) struct bfs_inode *di; int block = (i - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1; int off = (i - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; - unsigned long sblock, eblock; + unsigned long eblock; if (!off) { brelse(bh); @@ -399,7 +404,6 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) set_bit(i, info->si_imap); info->si_freeb -= BFS_FILEBLOCKS(di); - sblock = le32_to_cpu(di->i_sblock); eblock = le32_to_cpu(di->i_eblock); if (eblock > info->si_lf_eblk) info->si_lf_eblk = eblock; @@ -410,6 +414,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) s->s_dirt = 1; } dump_imap("read_super", s); + mutex_init(&info->bfs_lock); return 0; out: diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index ba4cddb..204cfd1 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -444,12 +444,6 @@ beyond_if: regs->gp = ex.a_gpvalue; #endif start_thread(regs, ex.a_entry, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } return 0; } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 3b6ff85..655ed8d 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1003,12 +1003,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) #endif start_thread(regs, elf_entry, bprm->p); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } retval = 0; out: kfree(loc); diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 1b59b1e..80c1f95 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -433,13 +433,6 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, entryaddr = interp_params.entry_addr ?: exec_params.entry_addr; start_thread(regs, entryaddr, current->mm->start_stack); - if (unlikely(current->ptrace & PT_PTRACED)) { - if (current->ptrace & PT_TRACE_EXEC) - ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP); - else - send_sig(SIGTRAP, current, 0); - } - retval = 0; error: @@ -477,6 +470,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, char __user *u_platform, *p; long hwcap; int loop; + int nr; /* reset for each csp adjustment */ /* we're going to shovel a whole load of stuff onto the stack */ #ifdef CONFIG_MMU @@ -549,10 +543,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, /* force 16 byte _final_ alignment here for generality */ #define DLINFO_ITEMS 13 - nitems = 1 + DLINFO_ITEMS + (k_platform ? 1 : 0); -#ifdef DLINFO_ARCH_ITEMS - nitems += DLINFO_ARCH_ITEMS; -#endif + nitems = 1 + DLINFO_ITEMS + (k_platform ? 1 : 0) + AT_VECTOR_SIZE_ARCH; csp = sp; sp -= nitems * 2 * sizeof(unsigned long); @@ -564,39 +555,46 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, sp -= sp & 15UL; /* put the ELF interpreter info on the stack */ -#define NEW_AUX_ENT(nr, id, val) \ +#define NEW_AUX_ENT(id, val) \ do { \ struct { unsigned long _id, _val; } __user *ent; \ \ ent = (void __user *) csp; \ __put_user((id), &ent[nr]._id); \ __put_user((val), &ent[nr]._val); \ + nr++; \ } while (0) + nr = 0; csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(0, AT_NULL, 0); + NEW_AUX_ENT(AT_NULL, 0); if (k_platform) { + nr = 0; csp -= 2 * sizeof(unsigned long); - NEW_AUX_ENT(0, AT_PLATFORM, + NEW_AUX_ENT(AT_PLATFORM, (elf_addr_t) (unsigned long) u_platform); } + nr = 0; csp -= DLINFO_ITEMS * 2 * sizeof(unsigned long); - NEW_AUX_ENT( 0, AT_HWCAP, hwcap); - NEW_AUX_ENT( 1, AT_PAGESZ, PAGE_SIZE); - NEW_AUX_ENT( 2, AT_CLKTCK, CLOCKS_PER_SEC); - NEW_AUX_ENT( 3, AT_PHDR, exec_params->ph_addr); - NEW_AUX_ENT( 4, AT_PHENT, sizeof(struct elf_phdr)); - NEW_AUX_ENT( 5, AT_PHNUM, exec_params->hdr.e_phnum); - NEW_AUX_ENT( 6, AT_BASE, interp_params->elfhdr_addr); - NEW_AUX_ENT( 7, AT_FLAGS, 0); - NEW_AUX_ENT( 8, AT_ENTRY, exec_params->entry_addr); - NEW_AUX_ENT( 9, AT_UID, (elf_addr_t) current->uid); - NEW_AUX_ENT(10, AT_EUID, (elf_addr_t) current->euid); - NEW_AUX_ENT(11, AT_GID, (elf_addr_t) current->gid); - NEW_AUX_ENT(12, AT_EGID, (elf_addr_t) current->egid); + NEW_AUX_ENT(AT_HWCAP, hwcap); + NEW_AUX_ENT(AT_PAGESZ, PAGE_SIZE); + NEW_AUX_ENT(AT_CLKTCK, CLOCKS_PER_SEC); + NEW_AUX_ENT(AT_PHDR, exec_params->ph_addr); + NEW_AUX_ENT(AT_PHENT, sizeof(struct elf_phdr)); + NEW_AUX_ENT(AT_PHNUM, exec_params->hdr.e_phnum); + NEW_AUX_ENT(AT_BASE, interp_params->elfhdr_addr); + NEW_AUX_ENT(AT_FLAGS, 0); + NEW_AUX_ENT(AT_ENTRY, exec_params->entry_addr); + NEW_AUX_ENT(AT_UID, (elf_addr_t) current->uid); + NEW_AUX_ENT(AT_EUID, (elf_addr_t) current->euid); + NEW_AUX_ENT(AT_GID, (elf_addr_t) current->gid); + NEW_AUX_ENT(AT_EGID, (elf_addr_t) current->egid); #ifdef ARCH_DLINFO + nr = 0; + csp -= AT_VECTOR_SIZE_ARCH * 2 * sizeof(unsigned long); + /* ARCH_DLINFO must come last so platform specific code can enforce * special alignment requirements on the AUXV if necessary (eg. PPC). */ diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 2cb1acd..dfc0197 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -914,15 +914,14 @@ static int load_flat_binary(struct linux_binprm * bprm, struct pt_regs * regs) /* Stash our initial stack pointer into the mm structure */ current->mm->start_stack = (unsigned long )sp; - +#ifdef FLAT_PLAT_INIT + FLAT_PLAT_INIT(regs); +#endif DBG_FLT("start_thread(regs=0x%x, entry=0x%x, start_stack=0x%x)\n", (int)regs, (int)start_addr, (int)current->mm->start_stack); start_thread(regs, start_addr, current->mm->start_stack); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); - return 0; } diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 7562053..8d7e88e 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -120,8 +120,6 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs) if (bprm->misc_bang) goto _ret; - bprm->misc_bang = 1; - /* to keep locking time low, we copy the interpreter string */ read_lock(&entries_lock); fmt = check_file(bprm); @@ -199,6 +197,8 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs) if (retval < 0) goto _error; + bprm->misc_bang = 1; + retval = search_binary_handler (bprm, regs); if (retval < 0) goto _error; diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c index fdc36bf..68be580 100644 --- a/fs/binfmt_som.c +++ b/fs/binfmt_som.c @@ -274,8 +274,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs) map_hpux_gateway_page(current,current->mm); start_thread_som(regs, som_entry, bprm->p); - if (current->ptrace & PT_PTRACED) - send_sig(SIGTRAP, current, 0); return 0; /* error cleanup */ diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 63e2ee6..19caf7c 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -107,7 +107,8 @@ void bio_integrity_free(struct bio *bio, struct bio_set *bs) BUG_ON(bip == NULL); /* A cloned bio doesn't own the integrity metadata */ - if (!bio_flagged(bio, BIO_CLONED) && bip->bip_buf != NULL) + if (!bio_flagged(bio, BIO_CLONED) && !bio_flagged(bio, BIO_FS_INTEGRITY) + && bip->bip_buf != NULL) kfree(bip->bip_buf); mempool_free(bip->bip_vec, bs->bvec_pools[bip->bip_pool]); @@ -150,6 +151,24 @@ int bio_integrity_add_page(struct bio *bio, struct page *page, } EXPORT_SYMBOL(bio_integrity_add_page); +static int bdev_integrity_enabled(struct block_device *bdev, int rw) +{ + struct blk_integrity *bi = bdev_get_integrity(bdev); + + if (bi == NULL) + return 0; + + if (rw == READ && bi->verify_fn != NULL && + (bi->flags & INTEGRITY_FLAG_READ)) + return 1; + + if (rw == WRITE && bi->generate_fn != NULL && + (bi->flags & INTEGRITY_FLAG_WRITE)) + return 1; + + return 0; +} + /** * bio_integrity_enabled - Check whether integrity can be passed * @bio: bio to check @@ -313,6 +332,14 @@ static void bio_integrity_generate(struct bio *bio) } } +static inline unsigned short blk_integrity_tuple_size(struct blk_integrity *bi) +{ + if (bi) + return bi->tuple_size; + + return 0; +} + /** * bio_integrity_prep - Prepare bio for integrity I/O * @bio: bio to prepare @@ -705,7 +732,6 @@ void __init bio_integrity_init_slab(void) bio_integrity_slab = KMEM_CACHE(bio_integrity_payload, SLAB_HWCACHE_ALIGN|SLAB_PANIC); } -EXPORT_SYMBOL(bio_integrity_init_slab); static int __init integrity_init(void) { @@ -30,7 +30,7 @@ static struct kmem_cache *bio_slab __read_mostly; -mempool_t *bio_split_pool __read_mostly; +static mempool_t *bio_split_pool __read_mostly; /* * if you change this list, also change bvec_alloc or things will @@ -60,28 +60,46 @@ struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct struct bio_vec *bvl; /* - * see comment near bvec_array define! + * If 'bs' is given, lookup the pool and do the mempool alloc. + * If not, this is a bio_kmalloc() allocation and just do a + * kzalloc() for the exact number of vecs right away. */ - switch (nr) { - case 1 : *idx = 0; break; - case 2 ... 4: *idx = 1; break; - case 5 ... 16: *idx = 2; break; - case 17 ... 64: *idx = 3; break; - case 65 ... 128: *idx = 4; break; - case 129 ... BIO_MAX_PAGES: *idx = 5; break; + if (bs) { + /* + * see comment near bvec_array define! + */ + switch (nr) { + case 1: + *idx = 0; + break; + case 2 ... 4: + *idx = 1; + break; + case 5 ... 16: + *idx = 2; + break; + case 17 ... 64: + *idx = 3; + break; + case 65 ... 128: + *idx = 4; + break; + case 129 ... BIO_MAX_PAGES: + *idx = 5; + break; default: return NULL; - } - /* - * idx now points to the pool we want to allocate from - */ - - bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); - if (bvl) { - struct biovec_slab *bp = bvec_slabs + *idx; + } - memset(bvl, 0, bp->nr_vecs * sizeof(struct bio_vec)); - } + /* + * idx now points to the pool we want to allocate from + */ + bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); + if (bvl) + memset(bvl, 0, + bvec_nr_vecs(*idx) * sizeof(struct bio_vec)); + } else + bvl = kzalloc(nr * sizeof(struct bio_vec), gfp_mask); return bvl; } @@ -110,10 +128,17 @@ static void bio_fs_destructor(struct bio *bio) bio_free(bio, fs_bio_set); } +static void bio_kmalloc_destructor(struct bio *bio) +{ + kfree(bio->bi_io_vec); + kfree(bio); +} + void bio_init(struct bio *bio) { memset(bio, 0, sizeof(*bio)); bio->bi_flags = 1 << BIO_UPTODATE; + bio->bi_comp_cpu = -1; atomic_set(&bio->bi_cnt, 1); } @@ -121,19 +146,25 @@ void bio_init(struct bio *bio) * bio_alloc_bioset - allocate a bio for I/O * @gfp_mask: the GFP_ mask given to the slab allocator * @nr_iovecs: number of iovecs to pre-allocate - * @bs: the bio_set to allocate from + * @bs: the bio_set to allocate from. If %NULL, just use kmalloc * * Description: - * bio_alloc_bioset will first try it's on mempool to satisfy the allocation. + * bio_alloc_bioset will first try its own mempool to satisfy the allocation. * If %__GFP_WAIT is set then we will block on the internal pool waiting - * for a &struct bio to become free. + * for a &struct bio to become free. If a %NULL @bs is passed in, we will + * fall back to just using @kmalloc to allocate the required memory. * * allocate bio and iovecs from the memory pools specified by the - * bio_set structure. + * bio_set structure, or @kmalloc if none given. **/ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) { - struct bio *bio = mempool_alloc(bs->bio_pool, gfp_mask); + struct bio *bio; + + if (bs) + bio = mempool_alloc(bs->bio_pool, gfp_mask); + else + bio = kmalloc(sizeof(*bio), gfp_mask); if (likely(bio)) { struct bio_vec *bvl = NULL; @@ -144,12 +175,15 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) { - mempool_free(bio, bs->bio_pool); + if (bs) + mempool_free(bio, bs->bio_pool); + else + kfree(bio); bio = NULL; goto out; } bio->bi_flags |= idx << BIO_POOL_OFFSET; - bio->bi_max_vecs = bvec_slabs[idx].nr_vecs; + bio->bi_max_vecs = bvec_nr_vecs(idx); } bio->bi_io_vec = bvl; } @@ -167,6 +201,23 @@ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) return bio; } +/* + * Like bio_alloc(), but doesn't use a mempool backing. This means that + * it CAN fail, but while bio_alloc() can only be used for allocations + * that have a short (finite) life span, bio_kmalloc() should be used + * for more permanent bio allocations (like allocating some bio's for + * initalization or setup purposes). + */ +struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) +{ + struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, NULL); + + if (bio) + bio->bi_destructor = bio_kmalloc_destructor; + + return bio; +} + void zero_fill_bio(struct bio *bio) { unsigned long flags; @@ -211,14 +262,6 @@ inline int bio_phys_segments(struct request_queue *q, struct bio *bio) return bio->bi_phys_segments; } -inline int bio_hw_segments(struct request_queue *q, struct bio *bio) -{ - if (unlikely(!bio_flagged(bio, BIO_SEG_VALID))) - blk_recount_segments(q, bio); - - return bio->bi_hw_segments; -} - /** * __bio_clone - clone a bio * @bio: destination bio @@ -353,8 +396,7 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page */ while (bio->bi_phys_segments >= q->max_phys_segments - || bio->bi_hw_segments >= q->max_hw_segments - || BIOVEC_VIRT_OVERSIZE(bio->bi_size)) { + || bio->bi_phys_segments >= q->max_hw_segments) { if (retried_segments) return 0; @@ -398,13 +440,11 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page } /* If we may be able to merge these biovecs, force a recount */ - if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec) || - BIOVEC_VIRT_MERGEABLE(bvec-1, bvec))) + if (bio->bi_vcnt && (BIOVEC_PHYS_MERGEABLE(bvec-1, bvec))) bio->bi_flags &= ~(1 << BIO_SEG_VALID); bio->bi_vcnt++; bio->bi_phys_segments++; - bio->bi_hw_segments++; done: bio->bi_size += len; return len; @@ -452,16 +492,19 @@ int bio_add_page(struct bio *bio, struct page *page, unsigned int len, struct bio_map_data { struct bio_vec *iovecs; - int nr_sgvecs; struct sg_iovec *sgvecs; + int nr_sgvecs; + int is_our_pages; }; static void bio_set_map_data(struct bio_map_data *bmd, struct bio *bio, - struct sg_iovec *iov, int iov_count) + struct sg_iovec *iov, int iov_count, + int is_our_pages) { memcpy(bmd->iovecs, bio->bi_io_vec, sizeof(struct bio_vec) * bio->bi_vcnt); memcpy(bmd->sgvecs, iov, sizeof(struct sg_iovec) * iov_count); bmd->nr_sgvecs = iov_count; + bmd->is_our_pages = is_our_pages; bio->bi_private = bmd; } @@ -472,20 +515,21 @@ static void bio_free_map_data(struct bio_map_data *bmd) kfree(bmd); } -static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count) +static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, + gfp_t gfp_mask) { - struct bio_map_data *bmd = kmalloc(sizeof(*bmd), GFP_KERNEL); + struct bio_map_data *bmd = kmalloc(sizeof(*bmd), gfp_mask); if (!bmd) return NULL; - bmd->iovecs = kmalloc(sizeof(struct bio_vec) * nr_segs, GFP_KERNEL); + bmd->iovecs = kmalloc(sizeof(struct bio_vec) * nr_segs, gfp_mask); if (!bmd->iovecs) { kfree(bmd); return NULL; } - bmd->sgvecs = kmalloc(sizeof(struct sg_iovec) * iov_count, GFP_KERNEL); + bmd->sgvecs = kmalloc(sizeof(struct sg_iovec) * iov_count, gfp_mask); if (bmd->sgvecs) return bmd; @@ -494,8 +538,9 @@ static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count) return NULL; } -static int __bio_copy_iov(struct bio *bio, struct sg_iovec *iov, int iov_count, - int uncopy) +static int __bio_copy_iov(struct bio *bio, struct bio_vec *iovecs, + struct sg_iovec *iov, int iov_count, int uncopy, + int do_free_page) { int ret = 0, i; struct bio_vec *bvec; @@ -505,7 +550,7 @@ static int __bio_copy_iov(struct bio *bio, struct sg_iovec *iov, int iov_count, __bio_for_each_segment(bvec, bio, i, 0) { char *bv_addr = page_address(bvec->bv_page); - unsigned int bv_len = bvec->bv_len; + unsigned int bv_len = iovecs[i].bv_len; while (bv_len && iov_idx < iov_count) { unsigned int bytes; @@ -538,7 +583,7 @@ static int __bio_copy_iov(struct bio *bio, struct sg_iovec *iov, int iov_count, } } - if (uncopy) + if (do_free_page) __free_page(bvec->bv_page); } @@ -555,10 +600,11 @@ static int __bio_copy_iov(struct bio *bio, struct sg_iovec *iov, int iov_count, int bio_uncopy_user(struct bio *bio) { struct bio_map_data *bmd = bio->bi_private; - int ret; - - ret = __bio_copy_iov(bio, bmd->sgvecs, bmd->nr_sgvecs, 1); + int ret = 0; + if (!bio_flagged(bio, BIO_NULL_MAPPED)) + ret = __bio_copy_iov(bio, bmd->iovecs, bmd->sgvecs, + bmd->nr_sgvecs, 1, bmd->is_our_pages); bio_free_map_data(bmd); bio_put(bio); return ret; @@ -567,16 +613,20 @@ int bio_uncopy_user(struct bio *bio) /** * bio_copy_user_iov - copy user data to bio * @q: destination block queue + * @map_data: pointer to the rq_map_data holding pages (if necessary) * @iov: the iovec. * @iov_count: number of elements in the iovec * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Prepares and returns a bio for indirect user io, bouncing data * to/from kernel pages as necessary. Must be paired with * call bio_uncopy_user() on io completion. */ -struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, - int iov_count, int write_to_vm) +struct bio *bio_copy_user_iov(struct request_queue *q, + struct rq_map_data *map_data, + struct sg_iovec *iov, int iov_count, + int write_to_vm, gfp_t gfp_mask) { struct bio_map_data *bmd; struct bio_vec *bvec; @@ -599,25 +649,38 @@ struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, len += iov[i].iov_len; } - bmd = bio_alloc_map_data(nr_pages, iov_count); + bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask); if (!bmd) return ERR_PTR(-ENOMEM); ret = -ENOMEM; - bio = bio_alloc(GFP_KERNEL, nr_pages); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) goto out_bmd; bio->bi_rw |= (!write_to_vm << BIO_RW); ret = 0; + i = 0; while (len) { - unsigned int bytes = PAGE_SIZE; + unsigned int bytes; + + if (map_data) + bytes = 1U << (PAGE_SHIFT + map_data->page_order); + else + bytes = PAGE_SIZE; if (bytes > len) bytes = len; - page = alloc_page(q->bounce_gfp | GFP_KERNEL); + if (map_data) { + if (i == map_data->nr_entries) { + ret = -ENOMEM; + break; + } + page = map_data->pages[i++]; + } else + page = alloc_page(q->bounce_gfp | gfp_mask); if (!page) { ret = -ENOMEM; break; @@ -636,16 +699,17 @@ struct bio *bio_copy_user_iov(struct request_queue *q, struct sg_iovec *iov, * success */ if (!write_to_vm) { - ret = __bio_copy_iov(bio, iov, iov_count, 0); + ret = __bio_copy_iov(bio, bio->bi_io_vec, iov, iov_count, 0, 0); if (ret) goto cleanup; } - bio_set_map_data(bmd, bio, iov, iov_count); + bio_set_map_data(bmd, bio, iov, iov_count, map_data ? 0 : 1); return bio; cleanup: - bio_for_each_segment(bvec, bio, i) - __free_page(bvec->bv_page); + if (!map_data) + bio_for_each_segment(bvec, bio, i) + __free_page(bvec->bv_page); bio_put(bio); out_bmd: @@ -656,29 +720,32 @@ out_bmd: /** * bio_copy_user - copy user data to bio * @q: destination block queue + * @map_data: pointer to the rq_map_data holding pages (if necessary) * @uaddr: start of user address * @len: length in bytes * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Prepares and returns a bio for indirect user io, bouncing data * to/from kernel pages as necessary. Must be paired with * call bio_uncopy_user() on io completion. */ -struct bio *bio_copy_user(struct request_queue *q, unsigned long uaddr, - unsigned int len, int write_to_vm) +struct bio *bio_copy_user(struct request_queue *q, struct rq_map_data *map_data, + unsigned long uaddr, unsigned int len, + int write_to_vm, gfp_t gfp_mask) { struct sg_iovec iov; iov.iov_base = (void __user *)uaddr; iov.iov_len = len; - return bio_copy_user_iov(q, &iov, 1, write_to_vm); + return bio_copy_user_iov(q, map_data, &iov, 1, write_to_vm, gfp_mask); } static struct bio *__bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { int i, j; int nr_pages = 0; @@ -704,12 +771,12 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, if (!nr_pages) return ERR_PTR(-EINVAL); - bio = bio_alloc(GFP_KERNEL, nr_pages); + bio = bio_alloc(gfp_mask, nr_pages); if (!bio) return ERR_PTR(-ENOMEM); ret = -ENOMEM; - pages = kcalloc(nr_pages, sizeof(struct page *), GFP_KERNEL); + pages = kcalloc(nr_pages, sizeof(struct page *), gfp_mask); if (!pages) goto out; @@ -721,12 +788,8 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, const int local_nr_pages = end - start; const int page_limit = cur_page + local_nr_pages; - down_read(¤t->mm->mmap_sem); - ret = get_user_pages(current, current->mm, uaddr, - local_nr_pages, - write_to_vm, 0, &pages[cur_page], NULL); - up_read(¤t->mm->mmap_sem); - + ret = get_user_pages_fast(uaddr, local_nr_pages, + write_to_vm, &pages[cur_page]); if (ret < local_nr_pages) { ret = -EFAULT; goto out_unmap; @@ -792,19 +855,21 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, * @uaddr: start of user address * @len: length in bytes * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Map the user space address into a bio suitable for io to a block * device. Returns an error pointer in case of error. */ struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, - unsigned long uaddr, unsigned int len, int write_to_vm) + unsigned long uaddr, unsigned int len, int write_to_vm, + gfp_t gfp_mask) { struct sg_iovec iov; iov.iov_base = (void __user *)uaddr; iov.iov_len = len; - return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm); + return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm, gfp_mask); } /** @@ -814,18 +879,19 @@ struct bio *bio_map_user(struct request_queue *q, struct block_device *bdev, * @iov: the iovec. * @iov_count: number of elements in the iovec * @write_to_vm: bool indicating writing to pages or not + * @gfp_mask: memory allocation flags * * Map the user space address into a bio suitable for io to a block * device. Returns an error pointer in case of error. */ struct bio *bio_map_user_iov(struct request_queue *q, struct block_device *bdev, struct sg_iovec *iov, int iov_count, - int write_to_vm) + int write_to_vm, gfp_t gfp_mask) { struct bio *bio; - bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm); - + bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm, + gfp_mask); if (IS_ERR(bio)) return bio; @@ -949,19 +1015,22 @@ static void bio_copy_kern_endio(struct bio *bio, int err) { struct bio_vec *bvec; const int read = bio_data_dir(bio) == READ; - char *p = bio->bi_private; + struct bio_map_data *bmd = bio->bi_private; int i; + char *p = bmd->sgvecs[0].iov_base; __bio_for_each_segment(bvec, bio, i, 0) { char *addr = page_address(bvec->bv_page); + int len = bmd->iovecs[i].bv_len; if (read && !err) - memcpy(p, addr, bvec->bv_len); + memcpy(p, addr, len); __free_page(bvec->bv_page); - p += bvec->bv_len; + p += len; } + bio_free_map_data(bmd); bio_put(bio); } @@ -979,38 +1048,13 @@ static void bio_copy_kern_endio(struct bio *bio, int err) struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, gfp_t gfp_mask, int reading) { - unsigned long kaddr = (unsigned long)data; - unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; - unsigned long start = kaddr >> PAGE_SHIFT; - const int nr_pages = end - start; struct bio *bio; struct bio_vec *bvec; - int i, ret; - - bio = bio_alloc(gfp_mask, nr_pages); - if (!bio) - return ERR_PTR(-ENOMEM); - - while (len) { - struct page *page; - unsigned int bytes = PAGE_SIZE; - - if (bytes > len) - bytes = len; - - page = alloc_page(q->bounce_gfp | gfp_mask); - if (!page) { - ret = -ENOMEM; - goto cleanup; - } - - if (bio_add_pc_page(q, bio, page, bytes, 0) < bytes) { - ret = -EINVAL; - goto cleanup; - } + int i; - len -= bytes; - } + bio = bio_copy_user(q, NULL, (unsigned long)data, len, 1, gfp_mask); + if (IS_ERR(bio)) + return bio; if (!reading) { void *p = data; @@ -1023,16 +1067,9 @@ struct bio *bio_copy_kern(struct request_queue *q, void *data, unsigned int len, } } - bio->bi_private = data; bio->bi_end_io = bio_copy_kern_endio; - return bio; -cleanup: - bio_for_each_segment(bvec, bio, i) - __free_page(bvec->bv_page); - bio_put(bio); - - return ERR_PTR(ret); + return bio; } /* @@ -1219,9 +1256,9 @@ static void bio_pair_end_2(struct bio *bi, int err) * split a bio - only worry about a bio with a single page * in it's iovec */ -struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) +struct bio_pair *bio_split(struct bio *bi, int first_sectors) { - struct bio_pair *bp = mempool_alloc(pool, GFP_NOIO); + struct bio_pair *bp = mempool_alloc(bio_split_pool, GFP_NOIO); if (!bp) return bp; @@ -1255,7 +1292,7 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) bp->bio2.bi_end_io = bio_pair_end_2; bp->bio1.bi_private = bi; - bp->bio2.bi_private = pool; + bp->bio2.bi_private = bio_split_pool; if (bio_integrity(bi)) bio_integrity_split(bi, bp, first_sectors); @@ -1263,6 +1300,42 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) return bp; } +/** + * bio_sector_offset - Find hardware sector offset in bio + * @bio: bio to inspect + * @index: bio_vec index + * @offset: offset in bv_page + * + * Return the number of hardware sectors between beginning of bio + * and an end point indicated by a bio_vec index and an offset + * within that vector's page. + */ +sector_t bio_sector_offset(struct bio *bio, unsigned short index, + unsigned int offset) +{ + unsigned int sector_sz = queue_hardsect_size(bio->bi_bdev->bd_disk->queue); + struct bio_vec *bv; + sector_t sectors; + int i; + + sectors = 0; + + if (index >= bio->bi_idx) + index = bio->bi_vcnt - 1; + + __bio_for_each_segment(bv, bio, i, 0) { + if (i == index) { + if (offset > bv->bv_offset) + sectors += (offset - bv->bv_offset) / sector_sz; + break; + } + + sectors += bv->bv_len / sector_sz; + } + + return sectors; +} +EXPORT_SYMBOL(bio_sector_offset); /* * create memory pools for biovec's in a bio_set. @@ -1365,6 +1438,7 @@ static int __init init_bio(void) subsys_initcall(init_bio); EXPORT_SYMBOL(bio_alloc); +EXPORT_SYMBOL(bio_kmalloc); EXPORT_SYMBOL(bio_put); EXPORT_SYMBOL(bio_free); EXPORT_SYMBOL(bio_endio); @@ -1372,7 +1446,6 @@ EXPORT_SYMBOL(bio_init); EXPORT_SYMBOL(__bio_clone); EXPORT_SYMBOL(bio_clone); EXPORT_SYMBOL(bio_phys_segments); -EXPORT_SYMBOL(bio_hw_segments); EXPORT_SYMBOL(bio_add_page); EXPORT_SYMBOL(bio_add_pc_page); EXPORT_SYMBOL(bio_get_nr_vecs); @@ -1382,7 +1455,6 @@ EXPORT_SYMBOL(bio_map_kern); EXPORT_SYMBOL(bio_copy_kern); EXPORT_SYMBOL(bio_pair_release); EXPORT_SYMBOL(bio_split); -EXPORT_SYMBOL(bio_split_pool); EXPORT_SYMBOL(bio_copy_user); EXPORT_SYMBOL(bio_uncopy_user); EXPORT_SYMBOL(bioset_create); diff --git a/fs/block_dev.c b/fs/block_dev.c index 10d8a0a..d84f0469 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -271,7 +271,7 @@ static void bdev_destroy_inode(struct inode *inode) kmem_cache_free(bdev_cachep, bdi); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; @@ -540,22 +540,6 @@ EXPORT_SYMBOL(bd_release); * /sys/block/sda/holders/dm-0 --> /sys/block/dm-0 */ -static struct kobject *bdev_get_kobj(struct block_device *bdev) -{ - if (bdev->bd_contains != bdev) - return kobject_get(&bdev->bd_part->dev.kobj); - else - return kobject_get(&bdev->bd_disk->dev.kobj); -} - -static struct kobject *bdev_get_holder(struct block_device *bdev) -{ - if (bdev->bd_contains != bdev) - return kobject_get(bdev->bd_part->holder_dir); - else - return kobject_get(bdev->bd_disk->holder_dir); -} - static int add_symlink(struct kobject *from, struct kobject *to) { if (!from || !to) @@ -604,11 +588,11 @@ static int bd_holder_grab_dirs(struct block_device *bdev, if (!bo->hdev) goto fail_put_sdir; - bo->sdev = bdev_get_kobj(bdev); + bo->sdev = kobject_get(&part_to_dev(bdev->bd_part)->kobj); if (!bo->sdev) goto fail_put_hdev; - bo->hdir = bdev_get_holder(bdev); + bo->hdir = kobject_get(bdev->bd_part->holder_dir); if (!bo->hdir) goto fail_put_sdev; @@ -868,6 +852,87 @@ struct block_device *open_by_devnum(dev_t dev, unsigned mode) EXPORT_SYMBOL(open_by_devnum); +/** + * flush_disk - invalidates all buffer-cache entries on a disk + * + * @bdev: struct block device to be flushed + * + * Invalidates all buffer-cache entries on a disk. It should be called + * when a disk has been changed -- either by a media change or online + * resize. + */ +static void flush_disk(struct block_device *bdev) +{ + if (__invalidate_device(bdev)) { + char name[BDEVNAME_SIZE] = ""; + + if (bdev->bd_disk) + disk_name(bdev->bd_disk, 0, name); + printk(KERN_WARNING "VFS: busy inodes on changed media or " + "resized disk %s\n", name); + } + + if (!bdev->bd_disk) + return; + if (disk_partitionable(bdev->bd_disk)) + bdev->bd_invalidated = 1; +} + +/** + * check_disk_size_change - checks for disk size change and adjusts bdev size. + * @disk: struct gendisk to check + * @bdev: struct bdev to adjust. + * + * This routine checks to see if the bdev size does not match the disk size + * and adjusts it if it differs. + */ +void check_disk_size_change(struct gendisk *disk, struct block_device *bdev) +{ + loff_t disk_size, bdev_size; + + disk_size = (loff_t)get_capacity(disk) << 9; + bdev_size = i_size_read(bdev->bd_inode); + if (disk_size != bdev_size) { + char name[BDEVNAME_SIZE]; + + disk_name(disk, 0, name); + printk(KERN_INFO + "%s: detected capacity change from %lld to %lld\n", + name, bdev_size, disk_size); + i_size_write(bdev->bd_inode, disk_size); + flush_disk(bdev); + } +} +EXPORT_SYMBOL(check_disk_size_change); + +/** + * revalidate_disk - wrapper for lower-level driver's revalidate_disk call-back + * @disk: struct gendisk to be revalidated + * + * This routine is a wrapper for lower-level driver's revalidate_disk + * call-backs. It is used to do common pre and post operations needed + * for all revalidate_disk operations. + */ +int revalidate_disk(struct gendisk *disk) +{ + struct block_device *bdev; + int ret = 0; + + if (disk->fops->revalidate_disk) + ret = disk->fops->revalidate_disk(disk); + + bdev = bdget_disk(disk, 0); + if (!bdev) + return ret; + + mutex_lock(&bdev->bd_mutex); + check_disk_size_change(disk, bdev); + mutex_unlock(&bdev->bd_mutex); + bdput(bdev); + return ret; +} +EXPORT_SYMBOL(revalidate_disk); + /* * This routine checks whether a removable media has been changed, * and invalidates all buffer-cache-entries in that case. This @@ -887,13 +952,9 @@ int check_disk_change(struct block_device *bdev) if (!bdops->media_changed(bdev->bd_disk)) return 0; - if (__invalidate_device(bdev)) - printk("VFS: busy inodes on changed media.\n"); - + flush_disk(bdev); if (bdops->revalidate_disk) bdops->revalidate_disk(bdev->bd_disk); - if (bdev->bd_disk->minors > 1) - bdev->bd_invalidated = 1; return 1; } @@ -927,10 +988,10 @@ static int __blkdev_put(struct block_device *bdev, int for_part); static int do_open(struct block_device *bdev, struct file *file, int for_part) { - struct module *owner = NULL; struct gendisk *disk; + struct hd_struct *part = NULL; int ret; - int part; + int partno; int perm = 0; if (file->f_mode & FMODE_READ) @@ -941,30 +1002,34 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) * hooks: /n/, see "layering violations". */ ret = devcgroup_inode_permission(bdev->bd_inode, perm); - if (ret != 0) + if (ret != 0) { + bdput(bdev); return ret; + } ret = -ENXIO; file->f_mapping = bdev->bd_inode->i_mapping; + lock_kernel(); - disk = get_gendisk(bdev->bd_dev, &part); - if (!disk) { - unlock_kernel(); - bdput(bdev); - return ret; - } - owner = disk->fops->owner; + + disk = get_gendisk(bdev->bd_dev, &partno); + if (!disk) + goto out_unlock_kernel; + part = disk_get_part(disk, partno); + if (!part) + goto out_unlock_kernel; mutex_lock_nested(&bdev->bd_mutex, for_part); if (!bdev->bd_openers) { bdev->bd_disk = disk; + bdev->bd_part = part; bdev->bd_contains = bdev; - if (!part) { + if (!partno) { struct backing_dev_info *bdi; if (disk->fops->open) { ret = disk->fops->open(bdev->bd_inode, file); if (ret) - goto out_first; + goto out_clear; } if (!bdev->bd_openers) { bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); @@ -976,36 +1041,36 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) if (bdev->bd_invalidated) rescan_partitions(disk, bdev); } else { - struct hd_struct *p; struct block_device *whole; whole = bdget_disk(disk, 0); ret = -ENOMEM; if (!whole) - goto out_first; + goto out_clear; BUG_ON(for_part); ret = __blkdev_get(whole, file->f_mode, file->f_flags, 1); if (ret) - goto out_first; + goto out_clear; bdev->bd_contains = whole; - p = disk->part[part - 1]; bdev->bd_inode->i_data.backing_dev_info = whole->bd_inode->i_data.backing_dev_info; - if (!(disk->flags & GENHD_FL_UP) || !p || !p->nr_sects) { + if (!(disk->flags & GENHD_FL_UP) || + !part || !part->nr_sects) { ret = -ENXIO; - goto out_first; + goto out_clear; } - kobject_get(&p->dev.kobj); - bdev->bd_part = p; - bd_set_size(bdev, (loff_t) p->nr_sects << 9); + bd_set_size(bdev, (loff_t)part->nr_sects << 9); } } else { + disk_put_part(part); put_disk(disk); - module_put(owner); + module_put(disk->fops->owner); + part = NULL; + disk = NULL; if (bdev->bd_contains == bdev) { if (bdev->bd_disk->fops->open) { ret = bdev->bd_disk->fops->open(bdev->bd_inode, file); if (ret) - goto out; + goto out_unlock_bdev; } if (bdev->bd_invalidated) rescan_partitions(bdev->bd_disk, bdev); @@ -1018,19 +1083,24 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part) unlock_kernel(); return 0; -out_first: + out_clear: bdev->bd_disk = NULL; + bdev->bd_part = NULL; bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) __blkdev_put(bdev->bd_contains, 1); bdev->bd_contains = NULL; - put_disk(disk); - module_put(owner); -out: + out_unlock_bdev: mutex_unlock(&bdev->bd_mutex); + out_unlock_kernel: unlock_kernel(); - if (ret) - bdput(bdev); + + disk_put_part(part); + if (disk) + module_put(disk->fops->owner); + put_disk(disk); + bdput(bdev); + return ret; } @@ -1115,11 +1185,8 @@ static int __blkdev_put(struct block_device *bdev, int for_part) put_disk(disk); module_put(owner); - - if (bdev->bd_contains != bdev) { - kobject_put(&bdev->bd_part->dev.kobj); - bdev->bd_part = NULL; - } + disk_put_part(bdev->bd_part); + bdev->bd_part = NULL; bdev->bd_disk = NULL; bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info; if (bdev != bdev->bd_contains) @@ -1195,10 +1262,9 @@ EXPORT_SYMBOL(ioctl_by_bdev); /** * lookup_bdev - lookup a struct block_device by name + * @pathname: special file representing the block device * - * @path: special file representing the block device - * - * Get a reference to the blockdevice at @path in the current + * Get a reference to the blockdevice at @pathname in the current * namespace if possible and return it. Return ERR_PTR(error) * otherwise. */ @@ -1234,6 +1300,7 @@ fail: bdev = ERR_PTR(error); goto out; } +EXPORT_SYMBOL(lookup_bdev); /** * open_bdev_excl - open a block device by name and set it up for use diff --git a/fs/buffer.c b/fs/buffer.c index d48caee..ac78d4c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -580,7 +580,7 @@ EXPORT_SYMBOL(mark_buffer_async_write); /* * The buffer's backing address_space's private_lock must be held */ -static inline void __remove_assoc_queue(struct buffer_head *bh) +static void __remove_assoc_queue(struct buffer_head *bh) { list_del_init(&bh->b_assoc_buffers); WARN_ON(!bh->b_assoc_map); @@ -706,7 +706,7 @@ static int __set_page_dirty(struct page *page, if (TestSetPageDirty(page)) return 0; - write_lock_irq(&mapping->tree_lock); + spin_lock_irq(&mapping->tree_lock); if (page->mapping) { /* Race with truncate? */ WARN_ON_ONCE(warn && !PageUptodate(page)); @@ -719,7 +719,7 @@ static int __set_page_dirty(struct page *page, radix_tree_tag_set(&mapping->page_tree, page_index(page), PAGECACHE_TAG_DIRTY); } - write_unlock_irq(&mapping->tree_lock); + spin_unlock_irq(&mapping->tree_lock); __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); return 1; @@ -1214,8 +1214,7 @@ void __brelse(struct buffer_head * buf) put_bh(buf); return; } - printk(KERN_ERR "VFS: brelse: Trying to free free buffer\n"); - WARN_ON(1); + WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n"); } /* @@ -1721,7 +1720,7 @@ static int __block_write_full_page(struct inode *inode, struct page *page, */ if (wbc->sync_mode != WB_SYNC_NONE || !wbc->nonblocking) { lock_buffer(bh); - } else if (test_set_buffer_locked(bh)) { + } else if (!trylock_buffer(bh)) { redirty_page_for_writepage(wbc, page); continue; } @@ -2097,6 +2096,52 @@ int generic_write_end(struct file *file, struct address_space *mapping, EXPORT_SYMBOL(generic_write_end); /* + * block_is_partially_uptodate checks whether buffers within a page are + * uptodate or not. + * + * Returns true if all buffers which correspond to a file portion + * we want to read are uptodate. + */ +int block_is_partially_uptodate(struct page *page, read_descriptor_t *desc, + unsigned long from) +{ + struct inode *inode = page->mapping->host; + unsigned block_start, block_end, blocksize; + unsigned to; + struct buffer_head *bh, *head; + int ret = 1; + + if (!page_has_buffers(page)) + return 0; + + blocksize = 1 << inode->i_blkbits; + to = min_t(unsigned, PAGE_CACHE_SIZE - from, desc->count); + to = from + to; + if (from < blocksize && to > PAGE_CACHE_SIZE - blocksize) + return 0; + + head = page_buffers(page); + bh = head; + block_start = 0; + do { + block_end = block_start + blocksize; + if (block_end > from && block_start < to) { + if (!buffer_uptodate(bh)) { + ret = 0; + break; + } + if (block_end >= to) + break; + } + block_start = block_end; + bh = bh->b_this_page; + } while (bh != head); + + return ret; +} +EXPORT_SYMBOL(block_is_partially_uptodate); + +/* * Generic "read page" function for block devices that have the normal * get_block functionality. This is most of the block device filesystems. * Reads the page asynchronously --- the unlock_buffer() and @@ -2881,14 +2926,17 @@ int submit_bh(int rw, struct buffer_head * bh) BUG_ON(!buffer_mapped(bh)); BUG_ON(!bh->b_end_io); - if (buffer_ordered(bh) && (rw == WRITE)) - rw = WRITE_BARRIER; + /* + * Mask in barrier bit for a write (could be either a WRITE or a + * WRITE_SYNC + */ + if (buffer_ordered(bh) && (rw & WRITE)) + rw |= WRITE_BARRIER; /* - * Only clear out a write error when rewriting, should this - * include WRITE_SYNC as well? + * Only clear out a write error when rewriting */ - if (test_set_buffer_req(bh) && (rw == WRITE || rw == WRITE_BARRIER)) + if (test_set_buffer_req(bh) && (rw & WRITE)) clear_buffer_write_io_error(bh); /* @@ -2955,7 +3003,7 @@ void ll_rw_block(int rw, int nr, struct buffer_head *bhs[]) if (rw == SWRITE || rw == SWRITE_SYNC) lock_buffer(bh); - else if (test_set_buffer_locked(bh)) + else if (!trylock_buffer(bh)) continue; if (rw == WRITE || rw == SWRITE || rw == SWRITE_SYNC) { @@ -3272,7 +3320,7 @@ int bh_submit_read(struct buffer_head *bh) EXPORT_SYMBOL(bh_submit_read); static void -init_buffer_head(struct kmem_cache *cachep, void *data) +init_buffer_head(void *data) { struct buffer_head *bh = data; diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 1f34652..06e521a 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,19 @@ +Version 1.54 +------------ +Fix premature write failure on congested networks (we would give up +on EAGAIN from the socket too quickly on large writes). +Cifs_mkdir and cifs_create now respect the setgid bit on parent dir. +Fix endian problems in acl (mode from/to cifs acl) on bigendian +architectures. Fix problems with preserving timestamps on copying open +files (e.g. "cp -a") to Windows servers. For mkdir and create honor setgid bit +on parent directory when server supports Unix Extensions but not POSIX +create. Update cifs.upcall version to handle new Kerberos sec flags +(this requires update of cifs.upcall program from Samba). Fix memory leak +on dns_upcall (resolving DFS referralls). Fix plain text password +authentication (requires setting SecurityFlags to 0x30030 to enable +lanman and plain text though). Fix writes to be at correct offset when +file is open with O_APPEND and file is on a directio (forcediretio) mount. + Version 1.53 ------------ DFS support added (Microsoft Distributed File System client support needed diff --git a/fs/cifs/README b/fs/cifs/README index 2bd6fe5..bd2343d 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -542,10 +542,20 @@ SecurityFlags Flags which control security negotiation and hashing mechanisms (as "must use") on the other hand does not make much sense. Default flags are 0x07007 - (NTLM, NTLMv2 and packet signing allowed). Maximum + (NTLM, NTLMv2 and packet signing allowed). The maximum allowable flags if you want to allow mounts to servers using weaker password hashes is 0x37037 (lanman, - plaintext, ntlm, ntlmv2, signing allowed): + plaintext, ntlm, ntlmv2, signing allowed). Some + SecurityFlags require the corresponding menuconfig + options to be enabled (lanman and plaintext require + CONFIG_CIFS_WEAK_PW_HASH for example). Enabling + plaintext authentication currently requires also + enabling lanman authentication in the security flags + because the cifs module only supports sending + laintext passwords using the older lanman dialect + form of the session setup SMB. (e.g. for authentication + using plain text passwords, set the SecurityFlags + to 0x30030): may use packet signing 0x00001 must use packet signing 0x01001 @@ -642,8 +652,30 @@ The statistics for the number of total SMBs and oplock breaks are different in that they represent all for that share, not just those for which the server returned success. -Also note that "cat /proc/fs/cifs/DebugData" will display information about +Also note that "cat /proc/fs/cifs/DebugData" will display information about the active sessions and the shares that are mounted. -Enabling Kerberos (extended security) works when CONFIG_CIFS_EXPERIMENTAL is -on but requires a user space helper (from the Samba project). NTLM and NTLMv2 and -LANMAN support do not require this helper. + +Enabling Kerberos (extended security) works but requires version 1.2 or later +of the helper program cifs.upcall to be present and to be configured in the +/etc/request-key.conf file. The cifs.upcall helper program is from the Samba +project(http://www.samba.org). NTLM and NTLMv2 and LANMAN support do not +require this helper. Note that NTLMv2 security (which does not require the +cifs.upcall helper program), instead of using Kerberos, is sufficient for +some use cases. + +Enabling DFS support (used to access shares transparently in an MS-DFS +global name space) requires that CONFIG_CIFS_EXPERIMENTAL be enabled. In +addition, DFS support for target shares which are specified as UNC +names which begin with host names (rather than IP addresses) requires +a user space helper (such as cifs.upcall) to be present in order to +translate host names to ip address, and the user space helper must also +be configured in the file /etc/request-key.conf + +To use cifs Kerberos and DFS support, the Linux keyutils package should be +installed and something like the following lines should be added to the +/etc/request-key.conf file: + +create cifs.spnego * * /usr/local/sbin/cifs.upcall %k +create dns_resolver * * /usr/local/sbin/cifs.upcall %k + + diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c index f58e41d..1b09f16 100644 --- a/fs/cifs/asn1.c +++ b/fs/cifs/asn1.c @@ -400,7 +400,7 @@ asn1_oid_decode(struct asn1_ctx *ctx, size = eoc - ctx->pointer + 1; /* first subid actually encodes first two subids */ - if (size < 2 || size > ULONG_MAX/sizeof(unsigned long)) + if (size < 2 || size > UINT_MAX/sizeof(unsigned long)) return 0; *oid = kmalloc(size * sizeof(unsigned long), GFP_ATOMIC); @@ -476,6 +476,7 @@ decode_negTokenInit(unsigned char *security_blob, int length, unsigned int cls, con, tag, oidlen, rc; bool use_ntlmssp = false; bool use_kerberos = false; + bool use_mskerberos = false; *secType = NTLM; /* BB eventually make Kerberos or NLTMSSP the default*/ @@ -483,6 +484,7 @@ decode_negTokenInit(unsigned char *security_blob, int length, asn1_open(&ctx, security_blob, length); + /* GSSAPI header */ if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { cFYI(1, ("Error decoding negTokenInit header")); return 0; @@ -490,156 +492,149 @@ decode_negTokenInit(unsigned char *security_blob, int length, || (tag != ASN1_EOC)) { cFYI(1, ("cls = %d con = %d tag = %d", cls, con, tag)); return 0; - } else { - /* remember to free obj->oid */ - rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag); - if (rc) { - if ((tag == ASN1_OJI) && (cls == ASN1_PRI)) { - rc = asn1_oid_decode(&ctx, end, &oid, &oidlen); - if (rc) { - rc = compare_oid(oid, oidlen, - SPNEGO_OID, - SPNEGO_OID_LEN); - kfree(oid); - } - } else - rc = 0; - } + } - if (!rc) { - cFYI(1, ("Error decoding negTokenInit header")); - return 0; - } + /* Check for SPNEGO OID -- remember to free obj->oid */ + rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag); + if (rc) { + if ((tag == ASN1_OJI) && (con == ASN1_PRI) && + (cls == ASN1_UNI)) { + rc = asn1_oid_decode(&ctx, end, &oid, &oidlen); + if (rc) { + rc = compare_oid(oid, oidlen, SPNEGO_OID, + SPNEGO_OID_LEN); + kfree(oid); + } + } else + rc = 0; + } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, ("Error decoding negTokenInit")); - return 0; - } else if ((cls != ASN1_CTX) || (con != ASN1_CON) - || (tag != ASN1_EOC)) { - cFYI(1, - ("cls = %d con = %d tag = %d end = %p (%d) exit 0", - cls, con, tag, end, *end)); - return 0; - } + /* SPNEGO OID not present or garbled -- bail out */ + if (!rc) { + cFYI(1, ("Error decoding negTokenInit header")); + return 0; + } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, ("Error decoding negTokenInit")); - return 0; - } else if ((cls != ASN1_UNI) || (con != ASN1_CON) - || (tag != ASN1_SEQ)) { - cFYI(1, - ("cls = %d con = %d tag = %d end = %p (%d) exit 1", - cls, con, tag, end, *end)); - return 0; - } + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding negTokenInit")); + return 0; + } else if ((cls != ASN1_CTX) || (con != ASN1_CON) + || (tag != ASN1_EOC)) { + cFYI(1, + ("cls = %d con = %d tag = %d end = %p (%d) exit 0", + cls, con, tag, end, *end)); + return 0; + } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, ("Error decoding 2nd part of negTokenInit")); - return 0; - } else if ((cls != ASN1_CTX) || (con != ASN1_CON) - || (tag != ASN1_EOC)) { - cFYI(1, - ("cls = %d con = %d tag = %d end = %p (%d) exit 0", - cls, con, tag, end, *end)); - return 0; - } + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding negTokenInit")); + return 0; + } else if ((cls != ASN1_UNI) || (con != ASN1_CON) + || (tag != ASN1_SEQ)) { + cFYI(1, + ("cls = %d con = %d tag = %d end = %p (%d) exit 1", + cls, con, tag, end, *end)); + return 0; + } - if (asn1_header_decode - (&ctx, &sequence_end, &cls, &con, &tag) == 0) { - cFYI(1, ("Error decoding 2nd part of negTokenInit")); - return 0; - } else if ((cls != ASN1_UNI) || (con != ASN1_CON) - || (tag != ASN1_SEQ)) { - cFYI(1, - ("cls = %d con = %d tag = %d end = %p (%d) exit 1", - cls, con, tag, end, *end)); - return 0; - } + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding 2nd part of negTokenInit")); + return 0; + } else if ((cls != ASN1_CTX) || (con != ASN1_CON) + || (tag != ASN1_EOC)) { + cFYI(1, + ("cls = %d con = %d tag = %d end = %p (%d) exit 0", + cls, con, tag, end, *end)); + return 0; + } - while (!asn1_eoc_decode(&ctx, sequence_end)) { - rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag); - if (!rc) { - cFYI(1, - ("Error decoding negTokenInit hdr exit2")); - return 0; - } - if ((tag == ASN1_OJI) && (con == ASN1_PRI)) { - if (asn1_oid_decode(&ctx, end, &oid, &oidlen)) { - - cFYI(1, - ("OID len = %d oid = 0x%lx 0x%lx " - "0x%lx 0x%lx", - oidlen, *oid, *(oid + 1), - *(oid + 2), *(oid + 3))); - - if (compare_oid(oid, oidlen, - MSKRB5_OID, - MSKRB5_OID_LEN)) - use_kerberos = true; - else if (compare_oid(oid, oidlen, - KRB5_OID, - KRB5_OID_LEN)) - use_kerberos = true; - else if (compare_oid(oid, oidlen, - NTLMSSP_OID, - NTLMSSP_OID_LEN)) - use_ntlmssp = true; - - kfree(oid); - } - } else { - cFYI(1, ("Should be an oid what is going on?")); - } - } + if (asn1_header_decode + (&ctx, &sequence_end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding 2nd part of negTokenInit")); + return 0; + } else if ((cls != ASN1_UNI) || (con != ASN1_CON) + || (tag != ASN1_SEQ)) { + cFYI(1, + ("cls = %d con = %d tag = %d end = %p (%d) exit 1", + cls, con, tag, end, *end)); + return 0; + } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, - ("Error decoding last part negTokenInit exit3")); - return 0; - } else if ((cls != ASN1_CTX) || (con != ASN1_CON)) { - /* tag = 3 indicating mechListMIC */ + while (!asn1_eoc_decode(&ctx, sequence_end)) { + rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag); + if (!rc) { cFYI(1, - ("Exit 4 cls = %d con = %d tag = %d end = %p (%d)", - cls, con, tag, end, *end)); + ("Error decoding negTokenInit hdr exit2")); return 0; } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, - ("Error decoding last part negTokenInit exit5")); - return 0; - } else if ((cls != ASN1_UNI) || (con != ASN1_CON) - || (tag != ASN1_SEQ)) { - cFYI(1, ("cls = %d con = %d tag = %d end = %p (%d)", - cls, con, tag, end, *end)); + if ((tag == ASN1_OJI) && (con == ASN1_PRI)) { + if (asn1_oid_decode(&ctx, end, &oid, &oidlen)) { + + cFYI(1, ("OID len = %d oid = 0x%lx 0x%lx " + "0x%lx 0x%lx", oidlen, *oid, + *(oid + 1), *(oid + 2), *(oid + 3))); + + if (compare_oid(oid, oidlen, MSKRB5_OID, + MSKRB5_OID_LEN) && + !use_kerberos) + use_mskerberos = true; + else if (compare_oid(oid, oidlen, KRB5_OID, + KRB5_OID_LEN) && + !use_mskerberos) + use_kerberos = true; + else if (compare_oid(oid, oidlen, NTLMSSP_OID, + NTLMSSP_OID_LEN)) + use_ntlmssp = true; + + kfree(oid); + } + } else { + cFYI(1, ("Should be an oid what is going on?")); } + } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, - ("Error decoding last part negTokenInit exit 7")); - return 0; - } else if ((cls != ASN1_CTX) || (con != ASN1_CON)) { - cFYI(1, - ("Exit 8 cls = %d con = %d tag = %d end = %p (%d)", - cls, con, tag, end, *end)); - return 0; - } - if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { - cFYI(1, - ("Error decoding last part negTokenInit exit9")); - return 0; - } else if ((cls != ASN1_UNI) || (con != ASN1_PRI) - || (tag != ASN1_GENSTR)) { - cFYI(1, - ("Exit10 cls = %d con = %d tag = %d end = %p (%d)", - cls, con, tag, end, *end)); - return 0; - } - cFYI(1, ("Need to call asn1_octets_decode() function for %s", - ctx.pointer)); /* is this UTF-8 or ASCII? */ + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding last part negTokenInit exit3")); + return 0; + } else if ((cls != ASN1_CTX) || (con != ASN1_CON)) { + /* tag = 3 indicating mechListMIC */ + cFYI(1, ("Exit 4 cls = %d con = %d tag = %d end = %p (%d)", + cls, con, tag, end, *end)); + return 0; + } + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding last part negTokenInit exit5")); + return 0; + } else if ((cls != ASN1_UNI) || (con != ASN1_CON) + || (tag != ASN1_SEQ)) { + cFYI(1, ("cls = %d con = %d tag = %d end = %p (%d)", + cls, con, tag, end, *end)); + } + + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding last part negTokenInit exit 7")); + return 0; + } else if ((cls != ASN1_CTX) || (con != ASN1_CON)) { + cFYI(1, ("Exit 8 cls = %d con = %d tag = %d end = %p (%d)", + cls, con, tag, end, *end)); + return 0; + } + if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) { + cFYI(1, ("Error decoding last part negTokenInit exit9")); + return 0; + } else if ((cls != ASN1_UNI) || (con != ASN1_PRI) + || (tag != ASN1_GENSTR)) { + cFYI(1, ("Exit10 cls = %d con = %d tag = %d end = %p (%d)", + cls, con, tag, end, *end)); + return 0; } + cFYI(1, ("Need to call asn1_octets_decode() function for %s", + ctx.pointer)); /* is this UTF-8 or ASCII? */ if (use_kerberos) *secType = Kerberos; + else if (use_mskerberos) + *secType = MSKerberos; else if (use_ntlmssp) *secType = NTLMSSP; diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index cc950f6..69a12aa 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -79,27 +79,25 @@ void cifs_dump_mids(struct TCP_Server_Info *server) spin_lock(&GlobalMid_Lock); list_for_each(tmp, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - if (mid_entry) { - cERROR(1, ("State: %d Cmd: %d Pid: %d Tsk: %p Mid %d", - mid_entry->midState, - (int)mid_entry->command, - mid_entry->pid, - mid_entry->tsk, - mid_entry->mid)); + cERROR(1, ("State: %d Cmd: %d Pid: %d Tsk: %p Mid %d", + mid_entry->midState, + (int)mid_entry->command, + mid_entry->pid, + mid_entry->tsk, + mid_entry->mid)); #ifdef CONFIG_CIFS_STATS2 - cERROR(1, ("IsLarge: %d buf: %p time rcv: %ld now: %ld", - mid_entry->largeBuf, - mid_entry->resp_buf, - mid_entry->when_received, - jiffies)); + cERROR(1, ("IsLarge: %d buf: %p time rcv: %ld now: %ld", + mid_entry->largeBuf, + mid_entry->resp_buf, + mid_entry->when_received, + jiffies)); #endif /* STATS2 */ - cERROR(1, ("IsMult: %d IsEnd: %d", mid_entry->multiRsp, - mid_entry->multiEnd)); - if (mid_entry->resp_buf) { - cifs_dump_detail(mid_entry->resp_buf); - cifs_dump_mem("existing buf: ", - mid_entry->resp_buf, 62); - } + cERROR(1, ("IsMult: %d IsEnd: %d", mid_entry->multiRsp, + mid_entry->multiEnd)); + if (mid_entry->resp_buf) { + cifs_dump_detail(mid_entry->resp_buf); + cifs_dump_mem("existing buf: ", + mid_entry->resp_buf, 62); } } spin_unlock(&GlobalMid_Lock); @@ -107,9 +105,7 @@ void cifs_dump_mids(struct TCP_Server_Info *server) #endif /* CONFIG_CIFS_DEBUG2 */ #ifdef CONFIG_PROC_FS -static int -cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, - int count, int *eof, void *data) +static int cifs_debug_data_proc_show(struct seq_file *m, void *v) { struct list_head *tmp; struct list_head *tmp1; @@ -117,23 +113,13 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, struct cifsSesInfo *ses; struct cifsTconInfo *tcon; int i; - int length = 0; - char *original_buf = buf; - - *beginBuffer = buf + offset; - length = - sprintf(buf, + seq_puts(m, "Display Internal CIFS Data Structures for Debugging\n" "---------------------------------------------------\n"); - buf += length; - length = sprintf(buf, "CIFS Version %s\n", CIFS_VERSION); - buf += length; - length = sprintf(buf, - "Active VFS Requests: %d\n", GlobalTotalActiveXid); - buf += length; - length = sprintf(buf, "Servers:"); - buf += length; + seq_printf(m, "CIFS Version %s\n", CIFS_VERSION); + seq_printf(m, "Active VFS Requests: %d\n", GlobalTotalActiveXid); + seq_printf(m, "Servers:"); i = 0; read_lock(&GlobalSMBSeslock); @@ -142,11 +128,10 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); if ((ses->serverDomain == NULL) || (ses->serverOS == NULL) || (ses->serverNOS == NULL)) { - buf += sprintf(buf, "\nentry for %s not fully " + seq_printf(m, "\nentry for %s not fully " "displayed\n\t", ses->serverName); } else { - length = - sprintf(buf, + seq_printf(m, "\n%d) Name: %s Domain: %s Mounts: %d OS:" " %s \n\tNOS: %s\tCapability: 0x%x\n\tSMB" " session status: %d\t", @@ -154,10 +139,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, atomic_read(&ses->inUse), ses->serverOS, ses->serverNOS, ses->capabilities, ses->status); - buf += length; } if (ses->server) { - buf += sprintf(buf, "TCP status: %d\n\tLocal Users To " + seq_printf(m, "TCP status: %d\n\tLocal Users To " "Server: %d SecMode: 0x%x Req On Wire: %d", ses->server->tcpStatus, atomic_read(&ses->server->socketUseCount), @@ -165,41 +149,34 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, atomic_read(&ses->server->inFlight)); #ifdef CONFIG_CIFS_STATS2 - buf += sprintf(buf, " In Send: %d In MaxReq Wait: %d", + seq_printf(m, " In Send: %d In MaxReq Wait: %d", atomic_read(&ses->server->inSend), atomic_read(&ses->server->num_waiters)); #endif - length = sprintf(buf, "\nMIDs:\n"); - buf += length; + seq_puts(m, "\nMIDs:\n"); spin_lock(&GlobalMid_Lock); list_for_each(tmp1, &ses->server->pending_mid_q) { mid_entry = list_entry(tmp1, struct mid_q_entry, qhead); - if (mid_entry) { - length = sprintf(buf, - "State: %d com: %d pid:" - " %d tsk: %p mid %d\n", - mid_entry->midState, - (int)mid_entry->command, - mid_entry->pid, - mid_entry->tsk, - mid_entry->mid); - buf += length; - } + seq_printf(m, "State: %d com: %d pid:" + " %d tsk: %p mid %d\n", + mid_entry->midState, + (int)mid_entry->command, + mid_entry->pid, + mid_entry->tsk, + mid_entry->mid); } spin_unlock(&GlobalMid_Lock); } } read_unlock(&GlobalSMBSeslock); - sprintf(buf, "\n"); - buf++; + seq_putc(m, '\n'); - length = sprintf(buf, "Shares:"); - buf += length; + seq_puts(m, "Shares:"); i = 0; read_lock(&GlobalSMBSeslock); @@ -208,62 +185,52 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, i++; tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType); - length = sprintf(buf, "\n%d) %s Uses: %d ", i, + seq_printf(m, "\n%d) %s Uses: %d ", i, tcon->treeName, atomic_read(&tcon->useCount)); - buf += length; if (tcon->nativeFileSystem) { - length = sprintf(buf, "Type: %s ", + seq_printf(m, "Type: %s ", tcon->nativeFileSystem); - buf += length; } - length = sprintf(buf, "DevInfo: 0x%x Attributes: 0x%x" + seq_printf(m, "DevInfo: 0x%x Attributes: 0x%x" "\nPathComponentMax: %d Status: %d", le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics), le32_to_cpu(tcon->fsAttrInfo.Attributes), le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength), tcon->tidStatus); - buf += length; if (dev_type == FILE_DEVICE_DISK) - length = sprintf(buf, " type: DISK "); + seq_puts(m, " type: DISK "); else if (dev_type == FILE_DEVICE_CD_ROM) - length = sprintf(buf, " type: CDROM "); + seq_puts(m, " type: CDROM "); else - length = - sprintf(buf, " type: %d ", dev_type); - buf += length; - if (tcon->tidStatus == CifsNeedReconnect) { - buf += sprintf(buf, "\tDISCONNECTED "); - length += 14; - } + seq_printf(m, " type: %d ", dev_type); + + if (tcon->tidStatus == CifsNeedReconnect) + seq_puts(m, "\tDISCONNECTED "); } read_unlock(&GlobalSMBSeslock); - length = sprintf(buf, "\n"); - buf += length; + seq_putc(m, '\n'); /* BB add code to dump additional info such as TCP session info now */ - /* Now calculate total size of returned data */ - length = buf - original_buf; - - if (offset + count >= length) - *eof = 1; - if (length < offset) { - *eof = 1; - return 0; - } else { - length = length - offset; - } - if (length > count) - length = count; + return 0; +} - return length; +static int cifs_debug_data_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_debug_data_proc_show, NULL); } -#ifdef CONFIG_CIFS_STATS +static const struct file_operations cifs_debug_data_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_debug_data_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; -static int -cifs_stats_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) +#ifdef CONFIG_CIFS_STATS +static ssize_t cifs_stats_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -307,236 +274,132 @@ cifs_stats_write(struct file *file, const char __user *buffer, return count; } -static int -cifs_stats_read(char *buf, char **beginBuffer, off_t offset, - int count, int *eof, void *data) +static int cifs_stats_proc_show(struct seq_file *m, void *v) { - int item_length, i, length; + int i; struct list_head *tmp; struct cifsTconInfo *tcon; - *beginBuffer = buf + offset; - - length = sprintf(buf, + seq_printf(m, "Resources in use\nCIFS Session: %d\n", sesInfoAllocCount.counter); - buf += length; - item_length = - sprintf(buf, "Share (unique mount targets): %d\n", + seq_printf(m, "Share (unique mount targets): %d\n", tconInfoAllocCount.counter); - length += item_length; - buf += item_length; - item_length = - sprintf(buf, "SMB Request/Response Buffer: %d Pool size: %d\n", + seq_printf(m, "SMB Request/Response Buffer: %d Pool size: %d\n", bufAllocCount.counter, cifs_min_rcv + tcpSesAllocCount.counter); - length += item_length; - buf += item_length; - item_length = - sprintf(buf, "SMB Small Req/Resp Buffer: %d Pool size: %d\n", + seq_printf(m, "SMB Small Req/Resp Buffer: %d Pool size: %d\n", smBufAllocCount.counter, cifs_min_small); - length += item_length; - buf += item_length; #ifdef CONFIG_CIFS_STATS2 - item_length = sprintf(buf, "Total Large %d Small %d Allocations\n", + seq_printf(m, "Total Large %d Small %d Allocations\n", atomic_read(&totBufAllocCount), atomic_read(&totSmBufAllocCount)); - length += item_length; - buf += item_length; #endif /* CONFIG_CIFS_STATS2 */ - item_length = - sprintf(buf, "Operations (MIDs): %d\n", - midCount.counter); - length += item_length; - buf += item_length; - item_length = sprintf(buf, + seq_printf(m, "Operations (MIDs): %d\n", midCount.counter); + seq_printf(m, "\n%d session %d share reconnects\n", tcpSesReconnectCount.counter, tconInfoReconnectCount.counter); - length += item_length; - buf += item_length; - item_length = sprintf(buf, + seq_printf(m, "Total vfs operations: %d maximum at one time: %d\n", GlobalCurrentXid, GlobalMaxActiveXid); - length += item_length; - buf += item_length; i = 0; read_lock(&GlobalSMBSeslock); list_for_each(tmp, &GlobalTreeConnectionList) { i++; tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); - item_length = sprintf(buf, "\n%d) %s", i, tcon->treeName); - buf += item_length; - length += item_length; - if (tcon->tidStatus == CifsNeedReconnect) { - buf += sprintf(buf, "\tDISCONNECTED "); - length += 14; - } - item_length = sprintf(buf, "\nSMBs: %d Oplock Breaks: %d", + seq_printf(m, "\n%d) %s", i, tcon->treeName); + if (tcon->tidStatus == CifsNeedReconnect) + seq_puts(m, "\tDISCONNECTED "); + seq_printf(m, "\nSMBs: %d Oplock Breaks: %d", atomic_read(&tcon->num_smbs_sent), atomic_read(&tcon->num_oplock_brks)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nReads: %d Bytes: %lld", + seq_printf(m, "\nReads: %d Bytes: %lld", atomic_read(&tcon->num_reads), (long long)(tcon->bytes_read)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nWrites: %d Bytes: %lld", + seq_printf(m, "\nWrites: %d Bytes: %lld", atomic_read(&tcon->num_writes), (long long)(tcon->bytes_written)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, + seq_printf(m, "\nLocks: %d HardLinks: %d Symlinks: %d", atomic_read(&tcon->num_locks), atomic_read(&tcon->num_hardlinks), atomic_read(&tcon->num_symlinks)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nOpens: %d Closes: %d Deletes: %d", + seq_printf(m, "\nOpens: %d Closes: %d Deletes: %d", atomic_read(&tcon->num_opens), atomic_read(&tcon->num_closes), atomic_read(&tcon->num_deletes)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nMkdirs: %d Rmdirs: %d", + seq_printf(m, "\nMkdirs: %d Rmdirs: %d", atomic_read(&tcon->num_mkdirs), atomic_read(&tcon->num_rmdirs)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nRenames: %d T2 Renames %d", + seq_printf(m, "\nRenames: %d T2 Renames %d", atomic_read(&tcon->num_renames), atomic_read(&tcon->num_t2renames)); - buf += item_length; - length += item_length; - item_length = sprintf(buf, "\nFindFirst: %d FNext %d FClose %d", + seq_printf(m, "\nFindFirst: %d FNext %d FClose %d", atomic_read(&tcon->num_ffirst), atomic_read(&tcon->num_fnext), atomic_read(&tcon->num_fclose)); - buf += item_length; - length += item_length; } read_unlock(&GlobalSMBSeslock); - buf += sprintf(buf, "\n"); - length++; - - if (offset + count >= length) - *eof = 1; - if (length < offset) { - *eof = 1; - return 0; - } else { - length = length - offset; - } - if (length > count) - length = count; + seq_putc(m, '\n'); + return 0; +} - return length; +static int cifs_stats_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_stats_proc_show, NULL); } + +static const struct file_operations cifs_stats_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_stats_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_stats_proc_write, +}; #endif /* STATS */ static struct proc_dir_entry *proc_fs_cifs; -read_proc_t cifs_txanchor_read; -static read_proc_t cifsFYI_read; -static write_proc_t cifsFYI_write; -static read_proc_t oplockEnabled_read; -static write_proc_t oplockEnabled_write; -static read_proc_t lookupFlag_read; -static write_proc_t lookupFlag_write; -static read_proc_t traceSMB_read; -static write_proc_t traceSMB_write; -static read_proc_t multiuser_mount_read; -static write_proc_t multiuser_mount_write; -static read_proc_t security_flags_read; -static write_proc_t security_flags_write; -/* static read_proc_t ntlmv2_enabled_read; -static write_proc_t ntlmv2_enabled_write; -static read_proc_t packet_signing_enabled_read; -static write_proc_t packet_signing_enabled_write;*/ -static read_proc_t experimEnabled_read; -static write_proc_t experimEnabled_write; -static read_proc_t linuxExtensionsEnabled_read; -static write_proc_t linuxExtensionsEnabled_write; +static const struct file_operations cifsFYI_proc_fops; +static const struct file_operations cifs_oplock_proc_fops; +static const struct file_operations cifs_lookup_cache_proc_fops; +static const struct file_operations traceSMB_proc_fops; +static const struct file_operations cifs_multiuser_mount_proc_fops; +static const struct file_operations cifs_security_flags_proc_fops; +static const struct file_operations cifs_experimental_proc_fops; +static const struct file_operations cifs_linux_ext_proc_fops; void cifs_proc_init(void) { - struct proc_dir_entry *pde; - proc_fs_cifs = proc_mkdir("fs/cifs", NULL); if (proc_fs_cifs == NULL) return; proc_fs_cifs->owner = THIS_MODULE; - create_proc_read_entry("DebugData", 0, proc_fs_cifs, - cifs_debug_data_read, NULL); + proc_create("DebugData", 0, proc_fs_cifs, &cifs_debug_data_proc_fops); #ifdef CONFIG_CIFS_STATS - pde = create_proc_read_entry("Stats", 0, proc_fs_cifs, - cifs_stats_read, NULL); - if (pde) - pde->write_proc = cifs_stats_write; + proc_create("Stats", 0, proc_fs_cifs, &cifs_stats_proc_fops); #endif /* STATS */ - pde = create_proc_read_entry("cifsFYI", 0, proc_fs_cifs, - cifsFYI_read, NULL); - if (pde) - pde->write_proc = cifsFYI_write; - - pde = - create_proc_read_entry("traceSMB", 0, proc_fs_cifs, - traceSMB_read, NULL); - if (pde) - pde->write_proc = traceSMB_write; - - pde = create_proc_read_entry("OplockEnabled", 0, proc_fs_cifs, - oplockEnabled_read, NULL); - if (pde) - pde->write_proc = oplockEnabled_write; - - pde = create_proc_read_entry("Experimental", 0, proc_fs_cifs, - experimEnabled_read, NULL); - if (pde) - pde->write_proc = experimEnabled_write; - - pde = create_proc_read_entry("LinuxExtensionsEnabled", 0, proc_fs_cifs, - linuxExtensionsEnabled_read, NULL); - if (pde) - pde->write_proc = linuxExtensionsEnabled_write; - - pde = - create_proc_read_entry("MultiuserMount", 0, proc_fs_cifs, - multiuser_mount_read, NULL); - if (pde) - pde->write_proc = multiuser_mount_write; - - pde = - create_proc_read_entry("SecurityFlags", 0, proc_fs_cifs, - security_flags_read, NULL); - if (pde) - pde->write_proc = security_flags_write; - - pde = - create_proc_read_entry("LookupCacheEnabled", 0, proc_fs_cifs, - lookupFlag_read, NULL); - if (pde) - pde->write_proc = lookupFlag_write; - -/* pde = - create_proc_read_entry("NTLMV2Enabled", 0, proc_fs_cifs, - ntlmv2_enabled_read, NULL); - if (pde) - pde->write_proc = ntlmv2_enabled_write; - - pde = - create_proc_read_entry("PacketSigningEnabled", 0, proc_fs_cifs, - packet_signing_enabled_read, NULL); - if (pde) - pde->write_proc = packet_signing_enabled_write;*/ + proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops); + proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops); + proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops); + proc_create("Experimental", 0, proc_fs_cifs, + &cifs_experimental_proc_fops); + proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, + &cifs_linux_ext_proc_fops); + proc_create("MultiuserMount", 0, proc_fs_cifs, + &cifs_multiuser_mount_proc_fops); + proc_create("SecurityFlags", 0, proc_fs_cifs, + &cifs_security_flags_proc_fops); + proc_create("LookupCacheEnabled", 0, proc_fs_cifs, + &cifs_lookup_cache_proc_fops); } void @@ -553,39 +416,26 @@ cifs_proc_clean(void) #endif remove_proc_entry("MultiuserMount", proc_fs_cifs); remove_proc_entry("OplockEnabled", proc_fs_cifs); -/* remove_proc_entry("NTLMV2Enabled",proc_fs_cifs); */ remove_proc_entry("SecurityFlags", proc_fs_cifs); -/* remove_proc_entry("PacketSigningEnabled", proc_fs_cifs); */ remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs); remove_proc_entry("Experimental", proc_fs_cifs); remove_proc_entry("LookupCacheEnabled", proc_fs_cifs); remove_proc_entry("fs/cifs", NULL); } -static int -cifsFYI_read(char *page, char **start, off_t off, int count, - int *eof, void *data) +static int cifsFYI_proc_show(struct seq_file *m, void *v) { - int len; - - len = sprintf(page, "%d\n", cifsFYI); - - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; + seq_printf(m, "%d\n", cifsFYI); + return 0; +} - return len; +static int cifsFYI_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifsFYI_proc_show, NULL); } -static int -cifsFYI_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifsFYI_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos) { char c; int rc; @@ -603,30 +453,28 @@ cifsFYI_write(struct file *file, const char __user *buffer, return count; } -static int -oplockEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", oplockEnabled); - - len -= off; - *start = page + off; +static const struct file_operations cifsFYI_proc_fops = { + .owner = THIS_MODULE, + .open = cifsFYI_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifsFYI_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_oplock_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", oplockEnabled); + return 0; +} - return len; +static int cifs_oplock_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_oplock_proc_show, NULL); } -static int -oplockEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_oplock_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -642,30 +490,28 @@ oplockEnabled_write(struct file *file, const char __user *buffer, return count; } -static int -experimEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", experimEnabled); +static const struct file_operations cifs_oplock_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_oplock_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_oplock_proc_write, +}; - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_experimental_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", experimEnabled); + return 0; +} - return len; +static int cifs_experimental_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_experimental_proc_show, NULL); } -static int -experimEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_experimental_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -683,29 +529,28 @@ experimEnabled_write(struct file *file, const char __user *buffer, return count; } -static int -linuxExtensionsEnabled_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", linuxExtEnabled); - len -= off; - *start = page + off; +static const struct file_operations cifs_experimental_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_experimental_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_experimental_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_linux_ext_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", linuxExtEnabled); + return 0; +} - return len; +static int cifs_linux_ext_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_linux_ext_proc_show, NULL); } -static int -linuxExtensionsEnabled_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_linux_ext_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -721,31 +566,28 @@ linuxExtensionsEnabled_write(struct file *file, const char __user *buffer, return count; } +static const struct file_operations cifs_linux_ext_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_linux_ext_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_linux_ext_proc_write, +}; -static int -lookupFlag_read(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int cifs_lookup_cache_proc_show(struct seq_file *m, void *v) { - int len; - - len = sprintf(page, "%d\n", lookupCacheEnabled); - - len -= off; - *start = page + off; - - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; + seq_printf(m, "%d\n", lookupCacheEnabled); + return 0; +} - return len; +static int cifs_lookup_cache_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_lookup_cache_proc_show, NULL); } -static int -lookupFlag_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_lookup_cache_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -760,30 +602,29 @@ lookupFlag_write(struct file *file, const char __user *buffer, return count; } -static int -traceSMB_read(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", traceSMB); - - len -= off; - *start = page + off; - if (len > count) - len = count; - else - *eof = 1; +static const struct file_operations cifs_lookup_cache_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_lookup_cache_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_lookup_cache_proc_write, +}; - if (len < 0) - len = 0; +static int traceSMB_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", traceSMB); + return 0; +} - return len; +static int traceSMB_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, traceSMB_proc_show, NULL); } -static int -traceSMB_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t traceSMB_proc_write(struct file *file, const char __user *buffer, + size_t count, loff_t *ppos) { char c; int rc; @@ -799,30 +640,28 @@ traceSMB_write(struct file *file, const char __user *buffer, return count; } -static int -multiuser_mount_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "%d\n", multiuser_mount); - - len -= off; - *start = page + off; +static const struct file_operations traceSMB_proc_fops = { + .owner = THIS_MODULE, + .open = traceSMB_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = traceSMB_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_multiuser_mount_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "%d\n", multiuser_mount); + return 0; +} - return len; +static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *fh) +{ + return single_open(fh, cifs_multiuser_mount_proc_show, NULL); } -static int -multiuser_mount_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_multiuser_mount_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { char c; int rc; @@ -838,30 +677,28 @@ multiuser_mount_write(struct file *file, const char __user *buffer, return count; } -static int -security_flags_read(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - int len; - - len = sprintf(page, "0x%x\n", extended_security); - - len -= off; - *start = page + off; +static const struct file_operations cifs_multiuser_mount_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_multiuser_mount_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_multiuser_mount_proc_write, +}; - if (len > count) - len = count; - else - *eof = 1; - - if (len < 0) - len = 0; +static int cifs_security_flags_proc_show(struct seq_file *m, void *v) +{ + seq_printf(m, "0x%x\n", extended_security); + return 0; +} - return len; +static int cifs_security_flags_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, cifs_security_flags_proc_show, NULL); } -static int -security_flags_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) + +static ssize_t cifs_security_flags_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { unsigned int flags; char flags_string[12]; @@ -917,6 +754,15 @@ security_flags_write(struct file *file, const char __user *buffer, /* BB should we turn on MAY flags for other MUST options? */ return count; } + +static const struct file_operations cifs_security_flags_proc_fops = { + .owner = THIS_MODULE, + .open = cifs_security_flags_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = cifs_security_flags_proc_write, +}; #else inline void cifs_proc_init(void) { diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index d82374c..d2c8eef 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -226,7 +226,7 @@ static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd, int err; mntget(newmnt); - err = do_add_mount(newmnt, nd, nd->path.mnt->mnt_flags, mntlist); + err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags, mntlist); switch (err) { case 0: path_put(&nd->path); diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 7013aaf..fcee929 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -66,11 +66,28 @@ struct key_type cifs_spnego_key_type = { .describe = user_describe, }; -#define MAX_VER_STR_LEN 9 /* length of longest version string e.g. - strlen(";ver=0xFF") */ -#define MAX_MECH_STR_LEN 13 /* length of longest security mechanism name, eg - in future could have strlen(";sec=ntlmsspi") */ -#define MAX_IPV6_ADDR_LEN 42 /* eg FEDC:BA98:7654:3210:FEDC:BA98:7654:3210/60 */ +/* length of longest version string e.g. strlen("ver=0xFF") */ +#define MAX_VER_STR_LEN 8 + +/* length of longest security mechanism name, eg in future could have + * strlen(";sec=ntlmsspi") */ +#define MAX_MECH_STR_LEN 13 + +/* max possible addr len eg FEDC:BA98:7654:3210:FEDC:BA98:7654:3210/60 */ +#define MAX_IPV6_ADDR_LEN 42 + +/* strlen of "host=" */ +#define HOST_KEY_LEN 5 + +/* strlen of ";ip4=" or ";ip6=" */ +#define IP_KEY_LEN 5 + +/* strlen of ";uid=0x" */ +#define UID_KEY_LEN 7 + +/* strlen of ";user=" */ +#define USER_KEY_LEN 6 + /* get a key struct with a SPNEGO security blob, suitable for session setup */ struct key * cifs_get_spnego_key(struct cifsSesInfo *sesInfo) @@ -81,11 +98,15 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) struct key *spnego_key; const char *hostname = server->hostname; - /* BB: come up with better scheme for determining length */ - /* length of fields (with semicolons): ver=0xyz ipv4= ipaddress host= - hostname sec=mechanism uid=0x uid */ - desc_len = MAX_VER_STR_LEN + 5 + MAX_IPV6_ADDR_LEN + 1 + 6 + - strlen(hostname) + MAX_MECH_STR_LEN + 8 + (sizeof(uid_t) * 2); + /* length of fields (with semicolons): ver=0xyz ip4=ipaddress + host=hostname sec=mechanism uid=0xFF user=username */ + desc_len = MAX_VER_STR_LEN + + HOST_KEY_LEN + strlen(hostname) + + IP_KEY_LEN + MAX_IPV6_ADDR_LEN + + MAX_MECH_STR_LEN + + UID_KEY_LEN + (sizeof(uid_t) * 2) + + USER_KEY_LEN + strlen(sesInfo->userName) + 1; + spnego_key = ERR_PTR(-ENOMEM); description = kzalloc(desc_len, GFP_KERNEL); if (description == NULL) @@ -110,9 +131,11 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) dp = description + strlen(description); - /* for now, only sec=krb5 is valid */ + /* for now, only sec=krb5 and sec=mskrb5 are valid */ if (server->secType == Kerberos) sprintf(dp, ";sec=krb5"); + else if (server->secType == MSKerberos) + sprintf(dp, ";sec=mskrb5"); else goto out; diff --git a/fs/cifs/cifs_spnego.h b/fs/cifs/cifs_spnego.h index 05a34b1..e4041ec 100644 --- a/fs/cifs/cifs_spnego.h +++ b/fs/cifs/cifs_spnego.h @@ -23,7 +23,7 @@ #ifndef _CIFS_SPNEGO_H #define _CIFS_SPNEGO_H -#define CIFS_SPNEGO_UPCALL_VERSION 1 +#define CIFS_SPNEGO_UPCALL_VERSION 2 /* * The version field should always be set to CIFS_SPNEGO_UPCALL_VERSION. diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 0e9fc2b..57ecdc8 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -56,7 +56,7 @@ int match_sid(struct cifs_sid *ctsid) struct cifs_sid *cwsid; if (!ctsid) - return (-1); + return -1; for (i = 0; i < NUM_WK_SIDS; ++i) { cwsid = &(wksidarr[i].cifssid); @@ -87,11 +87,11 @@ int match_sid(struct cifs_sid *ctsid) } cFYI(1, ("matching sid: %s\n", wksidarr[i].sidname)); - return (0); /* sids compare/match */ + return 0; /* sids compare/match */ } cFYI(1, ("No matching sid")); - return (-1); + return -1; } /* if the two SIDs (roughly equivalent to a UUID for a user or group) are @@ -102,16 +102,16 @@ 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 0; /* compare the revision */ if (ctsid->revision != cwsid->revision) - return (0); + return 0; /* compare all of the six auth values */ for (i = 0; i < 6; ++i) { if (ctsid->authority[i] != cwsid->authority[i]) - return (0); + return 0; } /* compare all of the subauth values if any */ @@ -121,11 +121,11 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid) if (num_subauth) { for (i = 0; i < num_subauth; ++i) { if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) - return (0); + return 0; } } - return (1); /* sids compare/match */ + return 1; /* sids compare/match */ } @@ -169,8 +169,7 @@ static void copy_sec_desc(const struct cifs_ntsd *pntsd, for (i = 0; i < 6; i++) ngroup_sid_ptr->authority[i] = group_sid_ptr->authority[i]; for (i = 0; i < 5; i++) - ngroup_sid_ptr->sub_auth[i] = - cpu_to_le32(group_sid_ptr->sub_auth[i]); + ngroup_sid_ptr->sub_auth[i] = group_sid_ptr->sub_auth[i]; return; } @@ -285,7 +284,7 @@ static __u16 fill_ace_for_sid(struct cifs_ace *pntace, size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth * 4); pntace->size = cpu_to_le16(size); - return (size); + return size; } @@ -426,7 +425,7 @@ static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, pndacl->size = cpu_to_le16(size + sizeof(struct cifs_acl)); pndacl->num_aces = cpu_to_le32(3); - return (0); + return 0; } @@ -510,7 +509,7 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len, sizeof(struct cifs_sid)); */ - return (0); + return 0; } @@ -527,7 +526,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, struct cifs_acl *ndacl_ptr = NULL; /* no need for SACL ptr */ if ((inode == NULL) || (pntsd == NULL) || (pnntsd == NULL)) - return (-EIO); + return -EIO; owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + le32_to_cpu(pntsd->osidoffset)); @@ -550,7 +549,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, /* copy security descriptor control portion and owner and group sid */ copy_sec_desc(pntsd, pnntsd, sidsoffset); - return (rc); + return rc; } @@ -629,11 +628,11 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, cFYI(DBG2, ("set ACL for %s from mode 0x%x", path, inode->i_mode)); if (!inode) - return (rc); + return rc; sb = inode->i_sb; if (sb == NULL) - return (rc); + return rc; cifs_sb = CIFS_SB(sb); xid = GetXid(); @@ -652,7 +651,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, if (rc != 0) { cERROR(1, ("Unable to open file to set ACL")); FreeXid(xid); - return (rc); + return rc; } } @@ -665,7 +664,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, FreeXid(xid); - return (rc); + return rc; } /* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */ @@ -715,7 +714,7 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) if (!pnntsd) { cERROR(1, ("Unable to allocate security descriptor")); kfree(pntsd); - return (-ENOMEM); + return -ENOMEM; } rc = build_sec_desc(pntsd, pnntsd, inode, nmode); @@ -732,6 +731,6 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) kfree(pntsd); } - return (rc); + return rc; } #endif /* CONFIG_CIFS_EXPERIMENTAL */ diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 4ff8939..bd5f13d 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -294,6 +294,7 @@ void calc_lanman_hash(struct cifsSesInfo *ses, char *lnm_session_key) if ((ses->server->secMode & SECMODE_PW_ENCRYPT) == 0) if (extended_security & CIFSSEC_MAY_PLNTXT) { + memset(lnm_session_key, 0, CIFS_SESS_KEY_SIZE); memcpy(lnm_session_key, password_with_pad, CIFS_ENCPWD_SIZE); return; @@ -310,9 +311,8 @@ void calc_lanman_hash(struct cifsSesInfo *ses, char *lnm_session_key) utf8 and other multibyte codepages each need their own strupper function since a byte at a time will ont work. */ - for (i = 0; i < CIFS_ENCPWD_SIZE; i++) { + for (i = 0; i < CIFS_ENCPWD_SIZE; i++) password_with_pad[i] = toupper(password_with_pad[i]); - } SMBencrypt(password_with_pad, ses->server->cryptKey, lnm_session_key); /* clear password before we return/free memory */ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 22857c6..25ecbd5 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -175,6 +175,8 @@ out_no_root: if (inode) iput(inode); + cifs_umount(sb, cifs_sb); + out_mount_failed: if (cifs_sb) { #ifdef CONFIG_CIFS_DFS_UPCALL @@ -267,7 +269,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int cifs_permission(struct inode *inode, int mask, struct nameidata *nd) +static int cifs_permission(struct inode *inode, int mask) { struct cifs_sb_info *cifs_sb; @@ -766,7 +768,7 @@ const struct file_operations cifs_dir_ops = { }; static void -cifs_init_once(struct kmem_cache *cachep, void *inode) +cifs_init_once(void *inode) { struct cifsInodeInfo *cifsi = inode; @@ -930,36 +932,34 @@ static int cifs_oplock_thread(void *dummyarg) schedule_timeout(39*HZ); } else { oplock_item = list_entry(GlobalOplock_Q.next, - struct oplock_q_entry, qhead); - if (oplock_item) { - cFYI(1, ("found oplock item to write out")); - pTcon = oplock_item->tcon; - inode = oplock_item->pinode; - netfid = oplock_item->netfid; - spin_unlock(&GlobalMid_Lock); - DeleteOplockQEntry(oplock_item); - /* can not grab inode sem here since it would + struct oplock_q_entry, qhead); + cFYI(1, ("found oplock item to write out")); + pTcon = oplock_item->tcon; + inode = oplock_item->pinode; + netfid = oplock_item->netfid; + spin_unlock(&GlobalMid_Lock); + DeleteOplockQEntry(oplock_item); + /* can not grab inode sem here since it would deadlock when oplock received on delete since vfs_unlink holds the i_mutex across the call */ - /* mutex_lock(&inode->i_mutex);*/ - if (S_ISREG(inode->i_mode)) { - rc = - filemap_fdatawrite(inode->i_mapping); - if (CIFS_I(inode)->clientCanCacheRead - == 0) { - waitrc = filemap_fdatawait(inode->i_mapping); - invalidate_remote_inode(inode); - } - if (rc == 0) - rc = waitrc; - } else - rc = 0; - /* mutex_unlock(&inode->i_mutex);*/ - if (rc) - CIFS_I(inode)->write_behind_rc = rc; - cFYI(1, ("Oplock flush inode %p rc %d", - inode, rc)); + /* mutex_lock(&inode->i_mutex);*/ + if (S_ISREG(inode->i_mode)) { + rc = filemap_fdatawrite(inode->i_mapping); + if (CIFS_I(inode)->clientCanCacheRead == 0) { + waitrc = filemap_fdatawait( + inode->i_mapping); + invalidate_remote_inode(inode); + } + if (rc == 0) + rc = waitrc; + } else + rc = 0; + /* mutex_unlock(&inode->i_mutex);*/ + if (rc) + CIFS_I(inode)->write_behind_rc = rc; + cFYI(1, ("Oplock flush inode %p rc %d", + inode, rc)); /* releasing stale oplock after recent reconnect of smb session using a now incorrect file @@ -967,15 +967,13 @@ static int cifs_oplock_thread(void *dummyarg) not bother sending an oplock release if session to server still is disconnected since oplock already released by the server in that case */ - if (pTcon->tidStatus != CifsNeedReconnect) { - rc = CIFSSMBLock(0, pTcon, netfid, - 0 /* len */ , 0 /* offset */, 0, - 0, LOCKING_ANDX_OPLOCK_RELEASE, - false /* wait flag */); - cFYI(1, ("Oplock release rc = %d", rc)); - } - } else - spin_unlock(&GlobalMid_Lock); + if (pTcon->tidStatus != CifsNeedReconnect) { + rc = CIFSSMBLock(0, pTcon, netfid, + 0 /* len */ , 0 /* offset */, 0, + 0, LOCKING_ANDX_OPLOCK_RELEASE, + false /* wait flag */); + cFYI(1, ("Oplock release rc = %d", rc)); + } set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(1); /* yield in case q were corrupt */ } @@ -1001,8 +999,7 @@ static int cifs_dnotify_thread(void *dummyarg) list_for_each(tmp, &GlobalSMBSessionList) { ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); - if (ses && ses->server && - atomic_read(&ses->server->inFlight)) + if (ses->server && atomic_read(&ses->server->inFlight)) wake_up_all(&ses->server->response_q); } read_unlock(&GlobalSMBSeslock); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 25a6cbd..f7b4a5c 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -41,7 +41,7 @@ extern int cifs_create(struct inode *, struct dentry *, int, struct nameidata *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, struct nameidata *); -extern int cifs_unlink(struct inode *, struct dentry *); +extern int cifs_unlink(struct inode *dir, struct dentry *dentry); extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *); extern int cifs_mknod(struct inode *, struct dentry *, int, dev_t); extern int cifs_mkdir(struct inode *, struct dentry *, int); @@ -101,5 +101,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* EXPERIMENTAL */ -#define CIFS_VERSION "1.53" +#define CIFS_VERSION "1.54" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 9cfcf32..0d22479 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -27,7 +27,7 @@ #define MAX_SES_INFO 2 #define MAX_TCON_INFO 4 -#define MAX_TREE_SIZE 2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1 +#define MAX_TREE_SIZE (2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1) #define MAX_SERVER_SIZE 15 #define MAX_SHARE_SIZE 64 /* used to be 20, this should still be enough */ #define MAX_USERNAME_SIZE 32 /* 32 is to allow for 15 char names + null @@ -80,7 +80,8 @@ enum securityEnum { NTLMv2, /* Legacy NTLM auth with NTLMv2 hash */ RawNTLMSSP, /* NTLMSSP without SPNEGO */ NTLMSSP, /* NTLMSSP via SPNEGO */ - Kerberos /* Kerberos via SPNEGO */ + Kerberos, /* Kerberos via SPNEGO */ + MSKerberos, /* MS Kerberos via SPNEGO */ }; enum protocolEnum { @@ -308,6 +309,7 @@ struct cifs_search_info { __u32 resume_key; char *ntwrk_buf_start; char *srch_entries_start; + char *last_entry; char *presume_name; unsigned int resume_name_len; bool endOfSearch:1; @@ -537,8 +539,8 @@ require use of the stronger protocol */ #endif /* WEAK_PW_HASH */ #define CIFSSEC_MUST_SEAL 0x40040 /* not supported yet */ -#define CIFSSEC_DEF CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 -#define CIFSSEC_MAX CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2 +#define CIFSSEC_DEF (CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2) +#define CIFSSEC_MAX (CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2) #define CIFSSEC_AUTH_MASK (CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 | CIFSSEC_MAY_LANMAN | CIFSSEC_MAY_PLNTXT | CIFSSEC_MAY_KRB5) /* ***************************************************************** diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 0f327c2..d2a073e 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -31,7 +31,7 @@ #else #define CIFS_PROT 0 #endif -#define POSIX_PROT CIFS_PROT+1 +#define POSIX_PROT (CIFS_PROT+1) #define BAD_PROT 0xFFFF /* SMB command codes */ @@ -262,7 +262,7 @@ */ #define CIFS_NO_HANDLE 0xFFFF -#define NO_CHANGE_64 cpu_to_le64(0xFFFFFFFFFFFFFFFFULL) +#define NO_CHANGE_64 0xFFFFFFFFFFFFFFFFULL #define NO_CHANGE_32 0xFFFFFFFFUL /* IPC$ in ASCII */ @@ -341,7 +341,7 @@ #define CREATE_COMPLETE_IF_OPLK 0x00000100 /* should be zero */ #define CREATE_NO_EA_KNOWLEDGE 0x00000200 #define CREATE_EIGHT_DOT_THREE 0x00000400 /* doc says this is obsolete - "open for recovery" flag - should + "open for recovery" flag should be zero in any case */ #define CREATE_OPEN_FOR_RECOVERY 0x00000400 #define CREATE_RANDOM_ACCESS 0x00000800 @@ -414,8 +414,8 @@ struct smb_hdr { __u8 WordCount; } __attribute__((packed)); /* given a pointer to an smb_hdr retrieve the value of byte count */ -#define BCC(smb_var) ( *(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) -#define BCC_LE(smb_var) ( *(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) +#define BCC(smb_var) (*(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) +#define BCC_LE(smb_var) (*(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount))) /* given a pointer to an smb_hdr retrieve the pointer to the byte area */ #define pByteArea(smb_var) ((unsigned char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount) + 2) diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index b9f5e93..0cff7fe 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -172,12 +172,15 @@ extern int CIFSSMBQFSUnixInfo(const int xid, struct cifsTconInfo *tcon); extern int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData); -extern int CIFSSMBSetTimes(const int xid, struct cifsTconInfo *tcon, +extern int CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, const char *fileName, const FILE_BASIC_INFO *data, const struct nls_table *nls_codepage, int remap_special_chars); -extern int CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, - const FILE_BASIC_INFO *data, __u16 fid); +extern int CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, + const FILE_BASIC_INFO *data, __u16 fid, + __u32 pid_of_opener); +extern int CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, + bool delete_file, __u16 fid, __u32 pid_of_opener); #if 0 extern int CIFSSMBSetAttrLegacy(int xid, struct cifsTconInfo *tcon, char *fileName, __u16 dos_attributes, @@ -191,9 +194,20 @@ extern int CIFSSMBSetEOF(const int xid, struct cifsTconInfo *tcon, extern int CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size, __u16 fileHandle, __u32 opener_pid, bool AllocSizeFlag); -extern int CIFSSMBUnixSetPerms(const int xid, struct cifsTconInfo *pTcon, - char *full_path, __u64 mode, __u64 uid, - __u64 gid, dev_t dev, + +struct cifs_unix_set_info_args { + __u64 ctime; + __u64 atime; + __u64 mtime; + __u64 mode; + __u64 uid; + __u64 gid; + dev_t device; +}; + +extern int CIFSSMBUnixSetInfo(const int xid, struct cifsTconInfo *pTcon, + char *fileName, + const struct cifs_unix_set_info_args *args, const struct nls_table *nls_codepage, int remap_special_chars); @@ -217,7 +231,7 @@ extern int CIFSSMBRename(const int xid, struct cifsTconInfo *tcon, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, - int netfid, char *target_name, + int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap_special_chars); extern int CIFSCreateHardLink(const int xid, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 4511b70..6f4ffe1 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -128,8 +128,7 @@ static void mark_open_files_invalid(struct cifsTconInfo *pTcon) write_lock(&GlobalSMBSeslock); list_for_each_safe(tmp, tmp1, &pTcon->openFileList) { open_file = list_entry(tmp, struct cifsFileInfo, tlist); - if (open_file) - open_file->invalidHandle = true; + open_file->invalidHandle = true; } write_unlock(&GlobalSMBSeslock); /* BB Add call to invalidate_inodes(sb) for all superblocks mounted @@ -686,11 +685,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) SecurityBlob, count - 16, &server->secType); - if (rc == 1) { + if (rc == 1) rc = 0; - } else { + else rc = -EINVAL; - } } } else server->capabilities &= ~CAP_EXTENDED_SECURITY; @@ -2019,7 +2017,7 @@ renameRetry: } int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, - int netfid, char *target_name, + int netfid, const char *target_name, const struct nls_table *nls_codepage, int remap) { struct smb_com_transaction2_sfi_req *pSMB = NULL; @@ -2073,7 +2071,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon, remap); } rename_info->target_name_len = cpu_to_le32(2 * len_of_str); - count = 12 /* sizeof(struct set_file_rename) */ + (2 * len_of_str) + 2; + count = 12 /* sizeof(struct set_file_rename) */ + (2 * len_of_str); byte_count += count; pSMB->DataCount = cpu_to_le16(count); pSMB->TotalDataCount = pSMB->DataCount; @@ -3616,6 +3614,8 @@ findFirstRetry: /* BB remember to free buffer if error BB */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); if (rc == 0) { + unsigned int lnoff; + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) psrch_inf->unicode = true; else @@ -3638,6 +3638,17 @@ findFirstRetry: le16_to_cpu(parms->SearchCount); psrch_inf->index_of_last_entry = 2 /* skip . and .. */ + psrch_inf->entries_in_buffer; + lnoff = le16_to_cpu(parms->LastNameOffset); + if (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE < + lnoff) { + cERROR(1, ("ignoring corrupt resume name")); + psrch_inf->last_entry = NULL; + return rc; + } + + psrch_inf->last_entry = psrch_inf->srch_entries_start + + lnoff; + *pnetfid = parms->SearchHandle; } else { cifs_buf_release(pSMB); @@ -3727,6 +3738,8 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, rc = validate_t2((struct smb_t2_rsp *)pSMBr); if (rc == 0) { + unsigned int lnoff; + /* BB fixme add lock for file (srch_info) struct here */ if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) psrch_inf->unicode = true; @@ -3753,6 +3766,16 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, le16_to_cpu(parms->SearchCount); psrch_inf->index_of_last_entry += psrch_inf->entries_in_buffer; + lnoff = le16_to_cpu(parms->LastNameOffset); + if (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE < + lnoff) { + cERROR(1, ("ignoring corrupt resume name")); + psrch_inf->last_entry = NULL; + return rc; + } else + psrch_inf->last_entry = + psrch_inf->srch_entries_start + lnoff; + /* cFYI(1,("fnxt2 entries in buf %d index_of_last %d", psrch_inf->entries_in_buffer, psrch_inf->index_of_last_entry)); */ @@ -3914,7 +3937,10 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr, bool is_unicode; struct dfs_referral_level_3 *ref; - is_unicode = pSMBr->hdr.Flags2 & SMBFLG2_UNICODE; + if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) + is_unicode = true; + else + is_unicode = false; *num_of_nodes = le16_to_cpu(pSMBr->NumberOfReferrals); if (*num_of_nodes < 1) { @@ -4814,8 +4840,8 @@ 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 -CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, - const FILE_BASIC_INFO *data, __u16 fid) +CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon, + const FILE_BASIC_INFO *data, __u16 fid, __u32 pid_of_opener) { struct smb_com_transaction2_sfi_req *pSMB = NULL; char *data_offset; @@ -4828,11 +4854,8 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, if (rc) return rc; - /* At this point there is no need to override the current pid - with the pid of the opener, but that could change if we someday - use an existing handle (rather than opening one on the fly) */ - /* pSMB->hdr.Pid = cpu_to_le16((__u16)pid_of_opener); - pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid_of_opener >> 16));*/ + pSMB->hdr.Pid = cpu_to_le16((__u16)pid_of_opener); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid_of_opener >> 16)); params = 6; pSMB->MaxSetupCount = 0; @@ -4878,11 +4901,66 @@ CIFSSMBSetFileTimes(const int xid, struct cifsTconInfo *tcon, return rc; } +int +CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon, + bool delete_file, __u16 fid, __u32 pid_of_opener) +{ + struct smb_com_transaction2_sfi_req *pSMB = NULL; + char *data_offset; + int rc = 0; + __u16 params, param_offset, offset, byte_count, count; + + cFYI(1, ("Set File Disposition (via SetFileInfo)")); + rc = small_smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB); + + if (rc) + return rc; + + pSMB->hdr.Pid = cpu_to_le16((__u16)pid_of_opener); + pSMB->hdr.PidHigh = cpu_to_le16((__u16)(pid_of_opener >> 16)); + + params = 6; + pSMB->MaxSetupCount = 0; + pSMB->Reserved = 0; + pSMB->Flags = 0; + pSMB->Timeout = 0; + pSMB->Reserved2 = 0; + param_offset = offsetof(struct smb_com_transaction2_sfi_req, Fid) - 4; + offset = param_offset + params; + + data_offset = (char *) (&pSMB->hdr.Protocol) + offset; + + count = 1; + pSMB->MaxParameterCount = cpu_to_le16(2); + /* BB find max SMB PDU from sess */ + pSMB->MaxDataCount = cpu_to_le16(1000); + pSMB->SetupCount = 1; + pSMB->Reserved3 = 0; + pSMB->SubCommand = cpu_to_le16(TRANS2_SET_FILE_INFORMATION); + byte_count = 3 /* pad */ + params + count; + pSMB->DataCount = cpu_to_le16(count); + pSMB->ParameterCount = cpu_to_le16(params); + pSMB->TotalDataCount = pSMB->DataCount; + pSMB->TotalParameterCount = pSMB->ParameterCount; + pSMB->ParameterOffset = cpu_to_le16(param_offset); + pSMB->DataOffset = cpu_to_le16(offset); + pSMB->Fid = fid; + pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_DISPOSITION_INFO); + pSMB->Reserved4 = 0; + pSMB->hdr.smb_buf_length += 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); + if (rc) + cFYI(1, ("Send error in SetFileDisposition = %d", rc)); + + return rc; +} int -CIFSSMBSetTimes(const int xid, struct cifsTconInfo *tcon, const char *fileName, - const FILE_BASIC_INFO *data, - const struct nls_table *nls_codepage, int remap) +CIFSSMBSetPathInfo(const int xid, struct cifsTconInfo *tcon, + const char *fileName, const FILE_BASIC_INFO *data, + const struct nls_table *nls_codepage, int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; TRANSACTION2_SPI_RSP *pSMBr = NULL; @@ -5011,10 +5089,9 @@ SetAttrLgcyRetry: #endif /* temporarily unneeded SetAttr legacy function */ int -CIFSSMBUnixSetPerms(const int xid, struct cifsTconInfo *tcon, - char *fileName, __u64 mode, __u64 uid, __u64 gid, - dev_t device, const struct nls_table *nls_codepage, - int remap) +CIFSSMBUnixSetInfo(const int xid, struct cifsTconInfo *tcon, char *fileName, + const struct cifs_unix_set_info_args *args, + const struct nls_table *nls_codepage, int remap) { TRANSACTION2_SPI_REQ *pSMB = NULL; TRANSACTION2_SPI_RSP *pSMBr = NULL; @@ -5023,6 +5100,7 @@ CIFSSMBUnixSetPerms(const int xid, struct cifsTconInfo *tcon, int bytes_returned = 0; FILE_UNIX_BASIC_INFO *data_offset; __u16 params, param_offset, offset, count, byte_count; + __u64 mode = args->mode; cFYI(1, ("In SetUID/GID/Mode")); setPermsRetry: @@ -5078,16 +5156,16 @@ setPermsRetry: set file size and do not want to truncate file size to zero accidently as happened on one Samba server beta by putting zero instead of -1 here */ - data_offset->EndOfFile = NO_CHANGE_64; - data_offset->NumOfBytes = NO_CHANGE_64; - data_offset->LastStatusChange = NO_CHANGE_64; - data_offset->LastAccessTime = NO_CHANGE_64; - data_offset->LastModificationTime = NO_CHANGE_64; - data_offset->Uid = cpu_to_le64(uid); - data_offset->Gid = cpu_to_le64(gid); + data_offset->EndOfFile = cpu_to_le64(NO_CHANGE_64); + data_offset->NumOfBytes = cpu_to_le64(NO_CHANGE_64); + data_offset->LastStatusChange = cpu_to_le64(args->ctime); + data_offset->LastAccessTime = cpu_to_le64(args->atime); + data_offset->LastModificationTime = cpu_to_le64(args->mtime); + data_offset->Uid = cpu_to_le64(args->uid); + data_offset->Gid = cpu_to_le64(args->gid); /* better to leave device as zero when it is */ - data_offset->DevMajor = cpu_to_le64(MAJOR(device)); - data_offset->DevMinor = cpu_to_le64(MINOR(device)); + data_offset->DevMajor = cpu_to_le64(MAJOR(args->device)); + data_offset->DevMinor = cpu_to_le64(MINOR(args->device)); data_offset->Permissions = cpu_to_le64(mode); if (S_ISREG(mode)) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index e8fa46c..4c13bcd 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -151,7 +151,7 @@ cifs_reconnect(struct TCP_Server_Info *server) } list_for_each(tmp, &GlobalTreeConnectionList) { tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); - if ((tcon) && (tcon->ses) && (tcon->ses->server == server)) + if ((tcon->ses) && (tcon->ses->server == server)) tcon->tidStatus = CifsNeedReconnect; } read_unlock(&GlobalSMBSeslock); @@ -173,14 +173,12 @@ cifs_reconnect(struct TCP_Server_Info *server) mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - if (mid_entry) { - if (mid_entry->midState == MID_REQUEST_SUBMITTED) { + if (mid_entry->midState == MID_REQUEST_SUBMITTED) { /* Mark other intransit requests as needing retry so we do not immediately mark the session bad again (ie after we reconnect below) as they timeout too */ - mid_entry->midState = MID_RETRY_NEEDED; - } + mid_entry->midState = MID_RETRY_NEEDED; } } spin_unlock(&GlobalMid_Lock); @@ -351,11 +349,9 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) current->flags |= PF_MEMALLOC; cFYI(1, ("Demultiplex PID: %d", task_pid_nr(current))); - write_lock(&GlobalSMBSeslock); - atomic_inc(&tcpSesAllocCount); - length = tcpSesAllocCount.counter; - write_unlock(&GlobalSMBSeslock); - if (length > 1) + + length = atomic_inc_return(&tcpSesAllocCount); + if (length > 1) mempool_resize(cifs_req_poolp, length + cifs_min_rcv, GFP_KERNEL); @@ -455,7 +451,7 @@ 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 = ntohl(smb_buffer->smb_buf_length); + pdu_length = be32_to_cpu((__force __be32)smb_buffer->smb_buf_length); smb_buffer->smb_buf_length = pdu_length; cFYI(1, ("rfc1002 length 0x%x", pdu_length+4)); @@ -745,14 +741,11 @@ multi_t2_fnd: coming home not much else we can do but free the memory */ } - write_lock(&GlobalSMBSeslock); - atomic_dec(&tcpSesAllocCount); - length = tcpSesAllocCount.counter; - /* last chance to mark ses pointers invalid if there are any pointing to this (e.g if a crazy root user tried to kill cifsd kernel thread explicitly this might happen) */ + write_lock(&GlobalSMBSeslock); list_for_each(tmp, &GlobalSMBSessionList) { ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); @@ -763,6 +756,8 @@ multi_t2_fnd: kfree(server->hostname); kfree(server); + + length = atomic_dec_return(&tcpSesAllocCount); if (length > 0) mempool_resize(cifs_req_poolp, length + cifs_min_rcv, GFP_KERNEL); @@ -1461,6 +1456,39 @@ get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path, return rc; } +#ifdef CONFIG_DEBUG_LOCK_ALLOC +static struct lock_class_key cifs_key[2]; +static struct lock_class_key cifs_slock_key[2]; + +static inline void +cifs_reclassify_socket4(struct socket *sock) +{ + struct sock *sk = sock->sk; + BUG_ON(sock_owned_by_user(sk)); + sock_lock_init_class_and_name(sk, "slock-AF_INET-CIFS", + &cifs_slock_key[0], "sk_lock-AF_INET-CIFS", &cifs_key[0]); +} + +static inline void +cifs_reclassify_socket6(struct socket *sock) +{ + struct sock *sk = sock->sk; + BUG_ON(sock_owned_by_user(sk)); + sock_lock_init_class_and_name(sk, "slock-AF_INET6-CIFS", + &cifs_slock_key[1], "sk_lock-AF_INET6-CIFS", &cifs_key[1]); +} +#else +static inline void +cifs_reclassify_socket4(struct socket *sock) +{ +} + +static inline void +cifs_reclassify_socket6(struct socket *sock) +{ +} +#endif + /* See RFC1001 section 14 on representation of Netbios names */ static void rfc1002mangle(char *target, char *source, unsigned int length) { @@ -1495,6 +1523,7 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, /* BB other socket options to set KEEPALIVE, NODELAY? */ cFYI(1, ("Socket created")); (*csocket)->sk->sk_allocation = GFP_NOFS; + cifs_reclassify_socket4(*csocket); } } @@ -1627,6 +1656,7 @@ ipv6_connect(struct sockaddr_in6 *psin_server, struct socket **csocket) /* BB other socket options to set KEEPALIVE, NODELAY? */ cFYI(1, ("ipv6 Socket created")); (*csocket)->sk->sk_allocation = GFP_NOFS; + cifs_reclassify_socket6(*csocket); } } @@ -3568,19 +3598,21 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, char ntlm_session_key[CIFS_SESS_KEY_SIZE]; bool ntlmv2_flag = false; int first_time = 0; + struct TCP_Server_Info *server = pSesInfo->server; /* what if server changes its buffer size after dropping the session? */ - if (pSesInfo->server->maxBuf == 0) /* no need to send on reconnect */ { + if (server->maxBuf == 0) /* no need to send on reconnect */ { rc = CIFSSMBNegotiate(xid, pSesInfo); - if (rc == -EAGAIN) /* retry only once on 1st time connection */ { + if (rc == -EAGAIN) { + /* retry only once on 1st time connection */ rc = CIFSSMBNegotiate(xid, pSesInfo); if (rc == -EAGAIN) rc = -EHOSTDOWN; } if (rc == 0) { spin_lock(&GlobalMid_Lock); - if (pSesInfo->server->tcpStatus != CifsExiting) - pSesInfo->server->tcpStatus = CifsGood; + if (server->tcpStatus != CifsExiting) + server->tcpStatus = CifsGood; else rc = -EHOSTDOWN; spin_unlock(&GlobalMid_Lock); @@ -3588,97 +3620,90 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, } first_time = 1; } - if (!rc) { - pSesInfo->flags = 0; - pSesInfo->capabilities = pSesInfo->server->capabilities; - if (linuxExtEnabled == 0) - pSesInfo->capabilities &= (~CAP_UNIX); + + if (rc) + goto ss_err_exit; + + pSesInfo->flags = 0; + pSesInfo->capabilities = server->capabilities; + if (linuxExtEnabled == 0) + pSesInfo->capabilities &= (~CAP_UNIX); /* pSesInfo->sequence_number = 0;*/ - cFYI(1, - ("Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", - pSesInfo->server->secMode, - pSesInfo->server->capabilities, - pSesInfo->server->timeAdj)); - if (experimEnabled < 2) - rc = CIFS_SessSetup(xid, pSesInfo, - first_time, nls_info); - else if (extended_security - && (pSesInfo->capabilities - & CAP_EXTENDED_SECURITY) - && (pSesInfo->server->secType == NTLMSSP)) { - rc = -EOPNOTSUPP; - } else if (extended_security - && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) - && (pSesInfo->server->secType == RawNTLMSSP)) { - cFYI(1, ("NTLMSSP sesssetup")); - rc = CIFSNTLMSSPNegotiateSessSetup(xid, - pSesInfo, - &ntlmv2_flag, - nls_info); - if (!rc) { - if (ntlmv2_flag) { - char *v2_response; - cFYI(1, ("more secure NTLM ver2 hash")); - if (CalcNTLMv2_partial_mac_key(pSesInfo, - nls_info)) { - rc = -ENOMEM; - goto ss_err_exit; - } else - v2_response = kmalloc(16 + 64 /* blob */, GFP_KERNEL); - if (v2_response) { - CalcNTLMv2_response(pSesInfo, - v2_response); - /* if (first_time) - cifs_calculate_ntlmv2_mac_key( - pSesInfo->server->mac_signing_key, - response, ntlm_session_key,*/ - kfree(v2_response); + cFYI(1, ("Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", + server->secMode, server->capabilities, server->timeAdj)); + + if (experimEnabled < 2) + rc = CIFS_SessSetup(xid, pSesInfo, first_time, nls_info); + else if (extended_security + && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) + && (server->secType == NTLMSSP)) { + rc = -EOPNOTSUPP; + } else if (extended_security + && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) + && (server->secType == RawNTLMSSP)) { + cFYI(1, ("NTLMSSP sesssetup")); + rc = CIFSNTLMSSPNegotiateSessSetup(xid, pSesInfo, &ntlmv2_flag, + nls_info); + if (!rc) { + if (ntlmv2_flag) { + char *v2_response; + cFYI(1, ("more secure NTLM ver2 hash")); + if (CalcNTLMv2_partial_mac_key(pSesInfo, + nls_info)) { + rc = -ENOMEM; + goto ss_err_exit; + } else + v2_response = kmalloc(16 + 64 /* blob*/, + GFP_KERNEL); + if (v2_response) { + CalcNTLMv2_response(pSesInfo, + v2_response); + /* if (first_time) + cifs_calculate_ntlmv2_mac_key */ + kfree(v2_response); /* BB Put dummy sig in SessSetup PDU? */ - } else { - rc = -ENOMEM; - goto ss_err_exit; - } - } else { - SMBNTencrypt(pSesInfo->password, - pSesInfo->server->cryptKey, - ntlm_session_key); - - if (first_time) - cifs_calculate_mac_key( - &pSesInfo->server->mac_signing_key, - ntlm_session_key, - pSesInfo->password); + rc = -ENOMEM; + goto ss_err_exit; } + + } else { + SMBNTencrypt(pSesInfo->password, + server->cryptKey, + ntlm_session_key); + + if (first_time) + cifs_calculate_mac_key( + &server->mac_signing_key, + ntlm_session_key, + pSesInfo->password); + } /* for better security the weaker lanman hash not sent in AuthSessSetup so we no longer calculate it */ - rc = CIFSNTLMSSPAuthSessSetup(xid, - pSesInfo, - ntlm_session_key, - ntlmv2_flag, - nls_info); - } - } else { /* old style NTLM 0.12 session setup */ - SMBNTencrypt(pSesInfo->password, - pSesInfo->server->cryptKey, - ntlm_session_key); + rc = CIFSNTLMSSPAuthSessSetup(xid, pSesInfo, + ntlm_session_key, + ntlmv2_flag, + nls_info); + } + } else { /* old style NTLM 0.12 session setup */ + SMBNTencrypt(pSesInfo->password, server->cryptKey, + ntlm_session_key); - if (first_time) - cifs_calculate_mac_key( - &pSesInfo->server->mac_signing_key, - ntlm_session_key, pSesInfo->password); + if (first_time) + cifs_calculate_mac_key(&server->mac_signing_key, + ntlm_session_key, + pSesInfo->password); - rc = CIFSSessSetup(xid, pSesInfo, - ntlm_session_key, nls_info); - } - if (rc) { - cERROR(1, ("Send error in SessSetup = %d", rc)); - } else { - cFYI(1, ("CIFS Session Established successfully")); + rc = CIFSSessSetup(xid, pSesInfo, ntlm_session_key, nls_info); + } + if (rc) { + cERROR(1, ("Send error in SessSetup = %d", rc)); + } else { + cFYI(1, ("CIFS Session Established successfully")); pSesInfo->status = CifsGood; - } } + ss_err_exit: return rc; } diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index fb69c1f..e962e75 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -226,23 +226,28 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, /* If Open reported that we actually created a file then we now have to set the mode if possible */ if ((pTcon->unix_ext) && (oplock & CIFS_CREATE_ACTION)) { + struct cifs_unix_set_info_args args = { + .mode = mode, + .ctime = NO_CHANGE_64, + .atime = NO_CHANGE_64, + .mtime = NO_CHANGE_64, + .device = 0, + }; + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { - CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, - (__u64)current->fsuid, - (__u64)current->fsgid, - 0 /* dev */, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = (__u64) current->fsuid; + if (inode->i_mode & S_ISGID) + args.gid = (__u64) inode->i_gid; + else + args.gid = (__u64) current->fsgid; } else { - CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, - (__u64)-1, - (__u64)-1, - 0 /* dev */, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = NO_CHANGE_64; + args.gid = NO_CHANGE_64; } + CIFSSMBUnixSetInfo(xid, pTcon, full_path, &args, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { /* BB implement mode setting via Windows security descriptors e.g. */ @@ -267,7 +272,12 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)) { newinode->i_uid = current->fsuid; - newinode->i_gid = current->fsgid; + if (inode->i_mode & S_ISGID) + newinode->i_gid = + inode->i_gid; + else + newinode->i_gid = + current->fsgid; } } } @@ -357,21 +367,24 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, if (full_path == NULL) rc = -ENOMEM; else if (pTcon->unix_ext) { - mode &= ~current->fs->umask; + struct cifs_unix_set_info_args args = { + .mode = mode & ~current->fs->umask, + .ctime = NO_CHANGE_64, + .atime = NO_CHANGE_64, + .mtime = NO_CHANGE_64, + .device = device_number, + }; if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { - rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, - mode, (__u64)current->fsuid, - (__u64)current->fsgid, - device_number, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = (__u64) current->fsuid; + args.gid = (__u64) current->fsgid; } else { - rc = CIFSSMBUnixSetPerms(xid, pTcon, - full_path, mode, (__u64)-1, (__u64)-1, - device_number, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = NO_CHANGE_64; + args.gid = NO_CHANGE_64; } + rc = CIFSSMBUnixSetInfo(xid, pTcon, full_path, + &args, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); if (!rc) { rc = cifs_get_inode_info_unix(&newinode, full_path, diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index f730ef3..1e0c1bd 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -29,38 +29,13 @@ #include "cifsproto.h" #include "cifs_debug.h" -static int dns_resolver_instantiate(struct key *key, const void *data, - size_t datalen) -{ - int rc = 0; - char *ip; - - ip = kmalloc(datalen+1, GFP_KERNEL); - if (!ip) - return -ENOMEM; - - memcpy(ip, data, datalen); - ip[datalen] = '\0'; - - rcu_assign_pointer(key->payload.data, ip); - - return rc; -} - -struct key_type key_type_dns_resolver = { - .name = "dns_resolver", - .def_datalen = sizeof(struct in_addr), - .describe = user_describe, - .instantiate = dns_resolver_instantiate, - .match = user_match, -}; - /* Checks if supplied name is IP address * returns: * 1 - name is IP * 0 - name is not IP */ -static int is_ip(const char *name) +static int +is_ip(const char *name) { int rc; struct sockaddr_in sin_server; @@ -82,6 +57,47 @@ static int is_ip(const char *name) return 0; } +static int +dns_resolver_instantiate(struct key *key, const void *data, + size_t datalen) +{ + int rc = 0; + char *ip; + + ip = kmalloc(datalen + 1, GFP_KERNEL); + if (!ip) + return -ENOMEM; + + memcpy(ip, data, datalen); + ip[datalen] = '\0'; + + /* make sure this looks like an address */ + if (!is_ip((const char *) ip)) { + kfree(ip); + return -EINVAL; + } + + key->type_data.x[0] = datalen; + rcu_assign_pointer(key->payload.data, ip); + + return rc; +} + +static void +dns_resolver_destroy(struct key *key) +{ + kfree(key->payload.data); +} + +struct key_type key_type_dns_resolver = { + .name = "dns_resolver", + .def_datalen = sizeof(struct in_addr), + .describe = user_describe, + .instantiate = dns_resolver_instantiate, + .destroy = dns_resolver_destroy, + .match = user_match, +}; + /* Resolves server name to ip address. * input: * unc - server UNC @@ -133,6 +149,7 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) rkey = request_key(&key_type_dns_resolver, name, ""); if (!IS_ERR(rkey)) { + len = rkey->type_data.x[0]; data = rkey->payload.data; } else { cERROR(1, ("%s: unable to resolve: %s", __func__, name)); @@ -141,11 +158,9 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) skip_upcall: if (data) { - len = strlen(data); - *ip_addr = kmalloc(len+1, GFP_KERNEL); + *ip_addr = kmalloc(len + 1, GFP_KERNEL); if (*ip_addr) { - memcpy(*ip_addr, data, len); - (*ip_addr)[len] = '\0'; + memcpy(*ip_addr, data, len + 1); if (!IS_ERR(rkey)) cFYI(1, ("%s: resolved: %s to %s", __func__, name, diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 0aac824..c4a8a06 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -107,7 +107,7 @@ static inline int cifs_open_inode_helper(struct inode *inode, struct file *file, /* want handles we can use to read with first in the list so we do not have to walk the - list to search for one in prepare_write */ + list to search for one in write_begin */ if ((file->f_flags & O_ACCMODE) == O_WRONLY) { list_add_tail(&pCifsFile->flist, &pCifsInode->openFileList); @@ -310,18 +310,19 @@ int cifs_open(struct inode *inode, struct file *file) /* time to set mode which we can not set earlier due to problems creating new read-only files */ if (pTcon->unix_ext) { - CIFSSMBUnixSetPerms(xid, pTcon, full_path, - inode->i_mode, - (__u64)-1, (__u64)-1, 0 /* dev */, + struct cifs_unix_set_info_args args = { + .mode = inode->i_mode, + .uid = NO_CHANGE_64, + .gid = NO_CHANGE_64, + .ctime = NO_CHANGE_64, + .atime = NO_CHANGE_64, + .mtime = NO_CHANGE_64, + .device = 0, + }; + CIFSSMBUnixSetInfo(xid, pTcon, full_path, &args, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - } else { - /* BB implement via Windows security descriptors eg - CIFSSMBWinSetPerms(xid, pTcon, full_path, mode, - -1, -1, local_nls); - in the meantime could set r/o dos attribute when - perms are eg: mode & 0222 == 0 */ } } @@ -832,6 +833,10 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, return -EBADF; open_file = (struct cifsFileInfo *) file->private_data; + rc = generic_write_checks(file, poffset, &write_size, 0); + if (rc) + return rc; + xid = GetXid(); if (*poffset > file->f_path.dentry->d_inode->i_size) @@ -910,7 +915,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, } static ssize_t cifs_write(struct file *file, const char *write_data, - size_t write_size, loff_t *poffset) + size_t write_size, loff_t *poffset) { int rc = 0; unsigned int bytes_written = 0; @@ -1060,6 +1065,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode) struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode) { struct cifsFileInfo *open_file; + bool any_available = false; int rc; /* Having a null inode here (because mapping->host was set to zero by @@ -1075,8 +1081,10 @@ struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode) read_lock(&GlobalSMBSeslock); refind_writable: list_for_each_entry(open_file, &cifs_inode->openFileList, flist) { - if (open_file->closePend) + if (open_file->closePend || + (!any_available && open_file->pid != current->tgid)) continue; + if (open_file->pfile && ((open_file->pfile->f_flags & O_RDWR) || (open_file->pfile->f_flags & O_WRONLY))) { @@ -1126,6 +1134,11 @@ refind_writable: of the loop here. */ } } + /* couldn't find useable FH with same pid, try any available */ + if (!any_available) { + any_available = true; + goto refind_writable; + } read_unlock(&GlobalSMBSeslock); return NULL; } @@ -1280,7 +1293,7 @@ retry: if (first < 0) lock_page(page); - else if (TestSetPageLocked(page)) + else if (!trylock_page(page)) break; if (unlikely(page->mapping != mapping)) { @@ -1442,49 +1455,52 @@ static int cifs_writepage(struct page *page, struct writeback_control *wbc) return rc; } -static int cifs_commit_write(struct file *file, struct page *page, - unsigned offset, unsigned to) +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 xid; - int rc = 0; - struct inode *inode = page->mapping->host; - loff_t position = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; - char *page_data; + int rc; + struct inode *inode = mapping->host; - xid = GetXid(); - cFYI(1, ("commit write for page %p up to position %lld for %d", - page, position, to)); - spin_lock(&inode->i_lock); - if (position > inode->i_size) - i_size_write(inode, position); + cFYI(1, ("write_end for page %p from pos %lld with %d bytes", + page, pos, copied)); + + if (!PageUptodate(page) && copied == PAGE_CACHE_SIZE) + SetPageUptodate(page); - spin_unlock(&inode->i_lock); if (!PageUptodate(page)) { - position = ((loff_t)page->index << PAGE_CACHE_SHIFT) + offset; - /* can not rely on (or let) writepage write this data */ - if (to < offset) { - cFYI(1, ("Illegal offsets, can not copy from %d to %d", - offset, to)); - FreeXid(xid); - return rc; - } + char *page_data; + unsigned offset = pos & (PAGE_CACHE_SIZE - 1); + int xid; + + xid = GetXid(); /* this is probably better than directly calling partialpage_write since in this function the file handle is known which we might as well leverage */ /* BB check if anything else missing out of ppw such as updating last write time */ page_data = kmap(page); - rc = cifs_write(file, page_data + offset, to-offset, - &position); - if (rc > 0) - rc = 0; - /* else if (rc < 0) should we set writebehind rc? */ + rc = cifs_write(file, page_data + offset, copied, &pos); + /* if (rc < 0) should we set writebehind rc? */ kunmap(page); + + FreeXid(xid); } else { + rc = copied; + pos += copied; set_page_dirty(page); } - FreeXid(xid); + if (rc > 0) { + spin_lock(&inode->i_lock); + if (pos > inode->i_size) + i_size_write(inode, pos); + spin_unlock(&inode->i_lock); + } + + unlock_page(page); + page_cache_release(page); + return rc; } @@ -2030,49 +2046,44 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file) return true; } -static int cifs_prepare_write(struct file *file, struct page *page, - unsigned from, unsigned to) +static int cifs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) { - int rc = 0; - loff_t i_size; - loff_t offset; + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + loff_t offset = pos & (PAGE_CACHE_SIZE - 1); + + cFYI(1, ("write_begin from %lld len %d", (long long)pos, len)); - cFYI(1, ("prepare write for page %p from %d to %d", page, from, to)); - if (PageUptodate(page)) + *pagep = __grab_cache_page(mapping, index); + if (!*pagep) + return -ENOMEM; + + if (PageUptodate(*pagep)) return 0; /* If we are writing a full page it will be up to date, no need to read from the server */ - if ((to == PAGE_CACHE_SIZE) && (from == 0)) { - SetPageUptodate(page); + if (len == PAGE_CACHE_SIZE && flags & AOP_FLAG_UNINTERRUPTIBLE) return 0; - } - offset = (loff_t)page->index << PAGE_CACHE_SHIFT; - i_size = i_size_read(page->mapping->host); + if ((file->f_flags & O_ACCMODE) != O_WRONLY) { + int rc; - if ((offset >= i_size) || - ((from == 0) && (offset + to) >= i_size)) { - /* - * We don't need to read data beyond the end of the file. - * zero it, and set the page uptodate - */ - simple_prepare_write(file, page, from, to); - SetPageUptodate(page); - } else if ((file->f_flags & O_ACCMODE) != O_WRONLY) { /* might as well read a page, it is fast enough */ - rc = cifs_readpage_worker(file, page, &offset); + rc = cifs_readpage_worker(file, *pagep, &offset); + + /* we do not need to pass errors back + e.g. if we do not have read access to the file + because cifs_write_end will attempt synchronous writes + -- shaggy */ } else { /* we could try using another file handle if there is one - but how would we lock it to prevent close of that handle racing with this read? In any case - this will be written out by commit_write so is fine */ + this will be written out by write_end so is fine */ } - /* we do not need to pass errors back - e.g. if we do not have read access to the file - because cifs_commit_write will do the right thing. -- shaggy */ - return 0; } @@ -2081,8 +2092,8 @@ const struct address_space_operations cifs_addr_ops = { .readpages = cifs_readpages, .writepage = cifs_writepage, .writepages = cifs_writepages, - .prepare_write = cifs_prepare_write, - .commit_write = cifs_commit_write, + .write_begin = cifs_write_begin, + .write_end = cifs_write_end, .set_page_dirty = __set_page_dirty_nobuffers, /* .sync_page = cifs_sync_page, */ /* .direct_IO = */ @@ -2097,8 +2108,8 @@ const struct address_space_operations cifs_addr_ops_smallbuf = { .readpage = cifs_readpage, .writepage = cifs_writepage, .writepages = cifs_writepages, - .prepare_write = cifs_prepare_write, - .commit_write = cifs_commit_write, + .write_begin = cifs_write_begin, + .write_end = cifs_write_end, .set_page_dirty = __set_page_dirty_nobuffers, /* .sync_page = cifs_sync_page, */ /* .direct_IO = */ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 2e904bd..a8c8333 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -546,7 +546,8 @@ int cifs_get_inode_info(struct inode **pinode, if ((inode->i_mode & S_IWUGO) == 0 && (attr & ATTR_READONLY) == 0) inode->i_mode |= (S_IWUGO & default_mode); - inode->i_mode &= ~S_IFMT; + + inode->i_mode &= ~S_IFMT; } /* clear write bits if ATTR_READONLY is set */ if (attr & ATTR_READONLY) @@ -649,6 +650,7 @@ struct inode *cifs_iget(struct super_block *sb, unsigned long ino) inode->i_fop = &simple_dir_operations; inode->i_uid = cifs_sb->mnt_uid; inode->i_gid = cifs_sb->mnt_gid; + } else if (rc) { _FreeXid(xid); iget_failed(inode); return ERR_PTR(rc); @@ -663,40 +665,201 @@ struct inode *cifs_iget(struct super_block *sb, unsigned long ino) return inode; } -int cifs_unlink(struct inode *inode, struct dentry *direntry) +static int +cifs_set_file_info(struct inode *inode, struct iattr *attrs, int xid, + char *full_path, __u32 dosattr) +{ + int rc; + int oplock = 0; + __u16 netfid; + __u32 netpid; + bool set_time = false; + struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + FILE_BASIC_INFO info_buf; + + if (attrs->ia_valid & ATTR_ATIME) { + set_time = true; + info_buf.LastAccessTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_atime)); + } else + info_buf.LastAccessTime = 0; + + if (attrs->ia_valid & ATTR_MTIME) { + set_time = true; + info_buf.LastWriteTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_mtime)); + } else + info_buf.LastWriteTime = 0; + + /* + * Samba throws this field away, but windows may actually use it. + * Do not set ctime unless other time stamps are changed explicitly + * (i.e. by utimes()) since we would then have a mix of client and + * server times. + */ + if (set_time && (attrs->ia_valid & ATTR_CTIME)) { + cFYI(1, ("CIFS - CTIME changed")); + info_buf.ChangeTime = + cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_ctime)); + } else + info_buf.ChangeTime = 0; + + info_buf.CreationTime = 0; /* don't change */ + info_buf.Attributes = cpu_to_le32(dosattr); + + /* + * If the file is already open for write, just use that fileid + */ + open_file = find_writable_file(cifsInode); + if (open_file) { + netfid = open_file->netfid; + netpid = open_file->pid; + goto set_via_filehandle; + } + + /* + * NT4 apparently returns success on this call, but it doesn't + * really work. + */ + if (!(pTcon->ses->flags & CIFS_SES_NT4)) { + rc = CIFSSMBSetPathInfo(xid, pTcon, full_path, + &info_buf, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc == 0) { + cifsInode->cifsAttrs = dosattr; + goto out; + } else if (rc != -EOPNOTSUPP && rc != -EINVAL) + goto out; + } + + cFYI(1, ("calling SetFileInfo since SetPathInfo for " + "times not supported by this server")); + rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, + SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, + CREATE_NOT_DIR, &netfid, &oplock, + NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (rc != 0) { + if (rc == -EIO) + rc = -EINVAL; + goto out; + } + + netpid = current->tgid; + +set_via_filehandle: + rc = CIFSSMBSetFileInfo(xid, pTcon, &info_buf, netfid, netpid); + if (!rc) + cifsInode->cifsAttrs = dosattr; + + if (open_file == NULL) + CIFSSMBClose(xid, pTcon, netfid); + else + atomic_dec(&open_file->wrtPending); +out: + return rc; +} + +/* + * open the given file (if it isn't already), set the DELETE_ON_CLOSE bit + * and rename it to a random name that hopefully won't conflict with + * anything else. + */ +static int +cifs_rename_pending_delete(char *full_path, struct inode *inode, int xid) +{ + int oplock = 0; + int rc; + __u16 netfid; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *tcon = cifs_sb->tcon; + __u32 dosattr; + FILE_BASIC_INFO *info_buf; + + rc = CIFSSMBOpen(xid, tcon, full_path, FILE_OPEN, + DELETE|FILE_WRITE_ATTRIBUTES, + CREATE_NOT_DIR|CREATE_DELETE_ON_CLOSE, + &netfid, &oplock, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc != 0) + goto out; + + /* set ATTR_HIDDEN and clear ATTR_READONLY */ + cifsInode = CIFS_I(inode); + dosattr = cifsInode->cifsAttrs & ~ATTR_READONLY; + if (dosattr == 0) + dosattr |= ATTR_NORMAL; + dosattr |= ATTR_HIDDEN; + + info_buf = kzalloc(sizeof(*info_buf), GFP_KERNEL); + if (info_buf == NULL) { + rc = -ENOMEM; + goto out_close; + } + info_buf->Attributes = cpu_to_le32(dosattr); + rc = CIFSSMBSetFileInfo(xid, tcon, info_buf, netfid, current->tgid); + kfree(info_buf); + if (rc != 0) + goto out_close; + cifsInode->cifsAttrs = dosattr; + + /* silly-rename the file */ + CIFSSMBRenameOpenFile(xid, tcon, netfid, NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + /* set DELETE_ON_CLOSE */ + rc = CIFSSMBSetFileDisposition(xid, tcon, true, netfid, current->tgid); + + /* + * some samba versions return -ENOENT when we try to set the file + * disposition here. Likely a samba bug, but work around it for now + */ + if (rc == -ENOENT) + rc = 0; + +out_close: + CIFSSMBClose(xid, tcon, netfid); +out: + return rc; +} + +int cifs_unlink(struct inode *dir, struct dentry *dentry) { int rc = 0; int xid; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; char *full_path = NULL; - struct cifsInodeInfo *cifsInode; - FILE_BASIC_INFO *pinfo_buf; + struct inode *inode = dentry->d_inode; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct super_block *sb = dir->i_sb; + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); + struct cifsTconInfo *tcon = cifs_sb->tcon; + struct iattr *attrs = NULL; + __u32 dosattr = 0, origattr = 0; - cFYI(1, ("cifs_unlink, inode = 0x%p", inode)); + cFYI(1, ("cifs_unlink, dir=0x%p, dentry=0x%p", dir, dentry)); xid = GetXid(); - if (inode) - cifs_sb = CIFS_SB(inode->i_sb); - else - cifs_sb = CIFS_SB(direntry->d_sb); - pTcon = cifs_sb->tcon; - - /* Unlink can be called from rename so we can not grab the sem here - since we deadlock otherwise */ -/* mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);*/ - full_path = build_path_from_dentry(direntry); -/* mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);*/ + /* Unlink can be called from rename so we can not take the + * sb->s_vfs_rename_mutex here */ + full_path = build_path_from_dentry(dentry); if (full_path == NULL) { FreeXid(xid); return -ENOMEM; } - if ((pTcon->ses->capabilities & CAP_UNIX) && + if ((tcon->ses->capabilities & CAP_UNIX) && (CIFS_UNIX_POSIX_PATH_OPS_CAP & - le64_to_cpu(pTcon->fsUnixInfo.Capability))) { - rc = CIFSPOSIXDelFile(xid, pTcon, full_path, + le64_to_cpu(tcon->fsUnixInfo.Capability))) { + rc = CIFSPOSIXDelFile(xid, tcon, full_path, SMB_POSIX_UNLINK_FILE_TARGET, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); cFYI(1, ("posix del rc %d", rc)); @@ -704,124 +867,60 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) goto psx_del_no_retry; } - rc = CIFSSMBDelFile(xid, pTcon, full_path, cifs_sb->local_nls, +retry_std_delete: + rc = CIFSSMBDelFile(xid, tcon, full_path, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + psx_del_no_retry: if (!rc) { - if (direntry->d_inode) - drop_nlink(direntry->d_inode); + if (inode) + drop_nlink(inode); } else if (rc == -ENOENT) { - d_drop(direntry); + d_drop(dentry); } else if (rc == -ETXTBSY) { - int oplock = 0; - __u16 netfid; - - rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, DELETE, - CREATE_NOT_DIR | CREATE_DELETE_ON_CLOSE, - &netfid, &oplock, NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - CIFSSMBRenameOpenFile(xid, pTcon, netfid, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - if (direntry->d_inode) - drop_nlink(direntry->d_inode); + rc = cifs_rename_pending_delete(full_path, inode, xid); + if (rc == 0) + drop_nlink(inode); + } else if (rc == -EACCES && dosattr == 0) { + attrs = kzalloc(sizeof(*attrs), GFP_KERNEL); + if (attrs == NULL) { + rc = -ENOMEM; + goto out_reval; } - } else if (rc == -EACCES) { - /* try only if r/o attribute set in local lookup data? */ - pinfo_buf = kzalloc(sizeof(FILE_BASIC_INFO), GFP_KERNEL); - if (pinfo_buf) { - /* ATTRS set to normal clears r/o bit */ - pinfo_buf->Attributes = cpu_to_le32(ATTR_NORMAL); - if (!(pTcon->ses->flags & CIFS_SES_NT4)) - rc = CIFSSMBSetTimes(xid, pTcon, full_path, - pinfo_buf, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - else - rc = -EOPNOTSUPP; - if (rc == -EOPNOTSUPP) { - int oplock = 0; - __u16 netfid; - /* rc = CIFSSMBSetAttrLegacy(xid, pTcon, - full_path, - (__u16)ATTR_NORMAL, - cifs_sb->local_nls); - For some strange reason it seems that NT4 eats the - old setattr call without actually setting the - attributes so on to the third attempted workaround - */ - - /* BB could scan to see if we already have it open - and pass in pid of opener to function */ - rc = CIFSSMBOpen(xid, pTcon, full_path, - FILE_OPEN, SYNCHRONIZE | - FILE_WRITE_ATTRIBUTES, 0, - &netfid, &oplock, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - rc = CIFSSMBSetFileTimes(xid, pTcon, - pinfo_buf, - netfid); - CIFSSMBClose(xid, pTcon, netfid); - } - } - kfree(pinfo_buf); - } - if (rc == 0) { - rc = CIFSSMBDelFile(xid, pTcon, full_path, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (!rc) { - if (direntry->d_inode) - drop_nlink(direntry->d_inode); - } else if (rc == -ETXTBSY) { - int oplock = 0; - __u16 netfid; - - rc = CIFSSMBOpen(xid, pTcon, full_path, - FILE_OPEN, DELETE, - CREATE_NOT_DIR | - CREATE_DELETE_ON_CLOSE, - &netfid, &oplock, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - CIFSSMBRenameOpenFile(xid, pTcon, - netfid, NULL, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - if (direntry->d_inode) - drop_nlink(direntry->d_inode); - } - /* BB if rc = -ETXTBUSY goto the rename logic BB */ - } - } - } - if (direntry->d_inode) { - cifsInode = CIFS_I(direntry->d_inode); - cifsInode->time = 0; /* will force revalidate to get info - when needed */ - direntry->d_inode->i_ctime = current_fs_time(inode->i_sb); + /* try to reset dos attributes */ + origattr = cifsInode->cifsAttrs; + if (origattr == 0) + origattr |= ATTR_NORMAL; + dosattr = origattr & ~ATTR_READONLY; + if (dosattr == 0) + dosattr |= ATTR_NORMAL; + dosattr |= ATTR_HIDDEN; + + rc = cifs_set_file_info(inode, attrs, xid, full_path, dosattr); + if (rc != 0) + goto out_reval; + + goto retry_std_delete; } + + /* undo the setattr if we errored out and it's needed */ + if (rc != 0 && dosattr != 0) + cifs_set_file_info(inode, attrs, xid, full_path, origattr); + +out_reval: if (inode) { - inode->i_ctime = inode->i_mtime = current_fs_time(inode->i_sb); cifsInode = CIFS_I(inode); - cifsInode->time = 0; /* force revalidate of dir as well */ + cifsInode->time = 0; /* will force revalidate to get info + when needed */ + inode->i_ctime = current_fs_time(sb); } + dir->i_ctime = dir->i_mtime = current_fs_time(sb); + cifsInode = CIFS_I(dir); + CIFS_I(dir)->time = 0; /* force revalidate of dir as well */ kfree(full_path); + kfree(attrs); FreeXid(xid); return rc; } @@ -866,7 +965,7 @@ static void posix_fill_in_inode(struct inode *tmp_inode, int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) { - int rc = 0; + int rc = 0, tmprc; int xid; struct cifs_sb_info *cifs_sb; struct cifsTconInfo *pTcon; @@ -928,6 +1027,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) kfree(pInfo); goto mkdir_get_info; } + /* Is an i_ino of zero legal? */ /* Are there sanity checks we can use to ensure that the server is really filling in that field? */ @@ -984,35 +1084,52 @@ mkdir_get_info: * failed to get it from the server or was set bogus */ if ((direntry->d_inode) && (direntry->d_inode->i_nlink < 2)) direntry->d_inode->i_nlink = 2; + mode &= ~current->fs->umask; + /* must turn on setgid bit if parent dir has it */ + if (inode->i_mode & S_ISGID) + mode |= S_ISGID; + if (pTcon->unix_ext) { + struct cifs_unix_set_info_args args = { + .mode = mode, + .ctime = NO_CHANGE_64, + .atime = NO_CHANGE_64, + .mtime = NO_CHANGE_64, + .device = 0, + }; if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { - CIFSSMBUnixSetPerms(xid, pTcon, full_path, - mode, - (__u64)current->fsuid, - (__u64)current->fsgid, - 0 /* dev_t */, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = (__u64)current->fsuid; + if (inode->i_mode & S_ISGID) + args.gid = (__u64)inode->i_gid; + else + args.gid = (__u64)current->fsgid; } else { - CIFSSMBUnixSetPerms(xid, pTcon, full_path, - mode, (__u64)-1, - (__u64)-1, 0 /* dev_t */, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + args.uid = NO_CHANGE_64; + args.gid = NO_CHANGE_64; } + CIFSSMBUnixSetInfo(xid, pTcon, full_path, &args, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); } else { if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) && (mode & S_IWUGO) == 0) { FILE_BASIC_INFO pInfo; + struct cifsInodeInfo *cifsInode; + u32 dosattrs; + memset(&pInfo, 0, sizeof(pInfo)); - pInfo.Attributes = cpu_to_le32(ATTR_READONLY); - CIFSSMBSetTimes(xid, pTcon, full_path, - &pInfo, cifs_sb->local_nls, + cifsInode = CIFS_I(newinode); + dosattrs = cifsInode->cifsAttrs|ATTR_READONLY; + pInfo.Attributes = cpu_to_le32(dosattrs); + tmprc = CIFSSMBSetPathInfo(xid, pTcon, + full_path, &pInfo, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (tmprc == 0) + cifsInode->cifsAttrs = dosattrs; } if (direntry->d_inode) { if (cifs_sb->mnt_cifs_flags & @@ -1024,8 +1141,12 @@ mkdir_get_info: CIFS_MOUNT_SET_UID) { direntry->d_inode->i_uid = current->fsuid; - direntry->d_inode->i_gid = - current->fsgid; + if (inode->i_mode & S_ISGID) + direntry->d_inode->i_gid = + inode->i_gid; + else + direntry->d_inode->i_gid = + current->fsgid; } } } @@ -1080,117 +1201,141 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) return rc; } +static int +cifs_do_rename(int xid, struct dentry *from_dentry, const char *fromPath, + struct dentry *to_dentry, const char *toPath) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(from_dentry->d_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + __u16 srcfid; + int oplock, rc; + + /* try path-based rename first */ + rc = CIFSSMBRename(xid, pTcon, fromPath, toPath, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + /* + * don't bother with rename by filehandle unless file is busy and + * source Note that cross directory moves do not work with + * rename by filehandle to various Windows servers. + */ + if (rc == 0 || rc != -ETXTBSY) + return rc; + + /* open the file to be renamed -- we need DELETE perms */ + rc = CIFSSMBOpen(xid, pTcon, fromPath, FILE_OPEN, DELETE, + CREATE_NOT_DIR, &srcfid, &oplock, NULL, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (rc == 0) { + rc = CIFSSMBRenameOpenFile(xid, pTcon, srcfid, + (const char *) to_dentry->d_name.name, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + CIFSSMBClose(xid, pTcon, srcfid); + } + + return rc; +} + int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, struct inode *target_inode, struct dentry *target_direntry) { - char *fromName; - char *toName; + char *fromName = NULL; + char *toName = NULL; struct cifs_sb_info *cifs_sb_source; struct cifs_sb_info *cifs_sb_target; struct cifsTconInfo *pTcon; + FILE_UNIX_BASIC_INFO *info_buf_source = NULL; + FILE_UNIX_BASIC_INFO *info_buf_target; int xid; - int rc = 0; - - xid = GetXid(); + int rc; cifs_sb_target = CIFS_SB(target_inode->i_sb); cifs_sb_source = CIFS_SB(source_inode->i_sb); pTcon = cifs_sb_source->tcon; + xid = GetXid(); + + /* + * BB: this might be allowed if same server, but different share. + * Consider adding support for this + */ if (pTcon != cifs_sb_target->tcon) { - FreeXid(xid); - return -EXDEV; /* BB actually could be allowed if same server, - but different share. - Might eventually add support for this */ + rc = -EXDEV; + goto cifs_rename_exit; } - /* we already have the rename sem so we do not need to grab it again - here to protect the path integrity */ + /* + * we already have the rename sem so we do not need to + * grab it again here to protect the path integrity + */ fromName = build_path_from_dentry(source_direntry); + if (fromName == NULL) { + rc = -ENOMEM; + goto cifs_rename_exit; + } + toName = build_path_from_dentry(target_direntry); - if ((fromName == NULL) || (toName == NULL)) { + if (toName == NULL) { rc = -ENOMEM; goto cifs_rename_exit; } - rc = CIFSSMBRename(xid, pTcon, fromName, toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + rc = cifs_do_rename(xid, source_direntry, fromName, + target_direntry, toName); + if (rc == -EEXIST) { - /* check if they are the same file because rename of hardlinked - files is a noop */ - FILE_UNIX_BASIC_INFO *info_buf_source; - FILE_UNIX_BASIC_INFO *info_buf_target; - - info_buf_source = - kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL); - if (info_buf_source != NULL) { + if (pTcon->unix_ext) { + /* + * Are src and dst hardlinks of same inode? We can + * only tell with unix extensions enabled + */ + info_buf_source = + kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), + GFP_KERNEL); + if (info_buf_source == NULL) + goto unlink_target; + info_buf_target = info_buf_source + 1; - if (pTcon->unix_ext) - rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, - info_buf_source, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & + rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, + info_buf_source, + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - /* else rc is still EEXIST so will fall through to - unlink the target and retry rename */ - if (rc == 0) { - rc = CIFSSMBUnixQPathInfo(xid, pTcon, toName, - info_buf_target, + if (rc != 0) + goto unlink_target; + + rc = CIFSSMBUnixQPathInfo(xid, pTcon, + toName, info_buf_target, cifs_sb_target->local_nls, /* remap based on source sb */ cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - } - if ((rc == 0) && - (info_buf_source->UniqueId == - info_buf_target->UniqueId)) { - /* do not rename since the files are hardlinked which - is a noop */ - } else { - /* we either can not tell the files are hardlinked - (as with Windows servers) or files are not - hardlinked so delete the target manually before - renaming to follow POSIX rather than Windows - semantics */ - cifs_unlink(target_inode, target_direntry); - rc = CIFSSMBRename(xid, pTcon, fromName, - toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags - & CIFS_MOUNT_MAP_SPECIAL_CHR); - } - kfree(info_buf_source); - } /* if we can not get memory just leave rc as EEXIST */ - } - - if (rc) - cFYI(1, ("rename rc %d", rc)); - - if ((rc == -EIO) || (rc == -EEXIST)) { - int oplock = 0; - __u16 netfid; - - /* BB FIXME Is Generic Read correct for rename? */ - /* if renaming directory - we should not say CREATE_NOT_DIR, - need to test renaming open directory, also GENERIC_READ - might not right be right access to request */ - rc = CIFSSMBOpen(xid, pTcon, fromName, FILE_OPEN, GENERIC_READ, - CREATE_NOT_DIR, &netfid, &oplock, NULL, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - rc = CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, - cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - CIFSSMBClose(xid, pTcon, netfid); - } + + if (rc == 0 && (info_buf_source->UniqueId == + info_buf_target->UniqueId)) + /* same file, POSIX says that this is a noop */ + goto cifs_rename_exit; + } /* else ... BB we could add the same check for Windows by + checking the UniqueId via FILE_INTERNAL_INFO */ +unlink_target: + /* + * we either can not tell the files are hardlinked (as with + * Windows servers) or files are not hardlinked. Delete the + * target manually before renaming to follow POSIX rather than + * Windows semantics + */ + cifs_unlink(target_inode, target_direntry); + rc = cifs_do_rename(xid, source_direntry, fromName, + target_direntry, toName); } cifs_rename_exit: + kfree(info_buf_source); kfree(fromName); kfree(toName); FreeXid(xid); @@ -1310,10 +1455,11 @@ int cifs_revalidate(struct dentry *direntry) /* if (S_ISDIR(direntry->d_inode->i_mode)) shrink_dcache_parent(direntry); */ if (S_ISREG(direntry->d_inode->i_mode)) { - if (direntry->d_inode->i_mapping) + if (direntry->d_inode->i_mapping) { wbrc = filemap_fdatawait(direntry->d_inode->i_mapping); if (wbrc) CIFS_I(direntry->d_inode)->write_behind_rc = wbrc; + } /* may eventually have to do this for open files too */ if (list_empty(&(cifsInode->openFileList))) { /* changed on server - flush read ahead pages */ @@ -1413,31 +1559,209 @@ out_busy: return -ETXTBSY; } -int cifs_setattr(struct dentry *direntry, struct iattr *attrs) +static int +cifs_set_file_size(struct inode *inode, struct iattr *attrs, + int xid, char *full_path) { + int rc; + struct cifsFileInfo *open_file; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + + /* + * To avoid spurious oplock breaks from server, in the case of + * inodes that we already have open, avoid doing path based + * setting of file size if we can do it by handle. + * This keeps our caching token (oplock) and avoids timeouts + * when the local oplock break takes longer to flush + * writebehind data than the SMB timeout for the SetPathInfo + * request would allow + */ + open_file = find_writable_file(cifsInode); + if (open_file) { + __u16 nfid = open_file->netfid; + __u32 npid = open_file->pid; + rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, nfid, + npid, false); + atomic_dec(&open_file->wrtPending); + 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); + cFYI(1, ("Wrt seteof rc %d", rc)); + } + } else + rc = -EINVAL; + + if (rc != 0) { + /* Set file size by pathname rather than by handle + either because no valid, writeable file handle for + it was found or because there was an error setting + it by handle */ + rc = CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, + false, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc)); + if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { + __u16 netfid; + int oplock = 0; + + rc = SMBLegacyOpen(xid, pTcon, full_path, + FILE_OPEN, GENERIC_WRITE, + CREATE_NOT_DIR, &netfid, &oplock, NULL, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + 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); + cFYI(1, ("wrt seteof rc %d", rc)); + CIFSSMBClose(xid, pTcon, netfid); + } + } + } + + if (rc == 0) { + rc = cifs_vmtruncate(inode, attrs->ia_size); + cifs_truncate_page(inode->i_mapping, inode->i_size); + } + + return rc; +} + +static int +cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) +{ + int rc; int xid; - struct cifs_sb_info *cifs_sb; - struct cifsTconInfo *pTcon; char *full_path = NULL; - int rc = -EACCES; - struct cifsFileInfo *open_file = NULL; - FILE_BASIC_INFO time_buf; - bool set_time = false; - bool set_dosattr = false; - __u64 mode = 0xFFFFFFFFFFFFFFFFULL; - __u64 uid = 0xFFFFFFFFFFFFFFFFULL; - __u64 gid = 0xFFFFFFFFFFFFFFFFULL; - struct cifsInodeInfo *cifsInode; struct inode *inode = direntry->d_inode; + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsTconInfo *pTcon = cifs_sb->tcon; + struct cifs_unix_set_info_args *args = NULL; + + cFYI(1, ("setattr_unix on file %s attrs->ia_valid=0x%x", + direntry->d_name.name, attrs->ia_valid)); + + xid = GetXid(); + + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) == 0) { + /* check if we have permission to change attrs */ + rc = inode_change_ok(inode, attrs); + if (rc < 0) + goto out; + else + rc = 0; + } + + full_path = build_path_from_dentry(direntry); + if (full_path == NULL) { + rc = -ENOMEM; + goto out; + } + + if ((attrs->ia_valid & ATTR_MTIME) || (attrs->ia_valid & ATTR_SIZE)) { + /* + Flush data before changing file size or changing the last + write time of the file on the server. If the + flush returns error, store it to report later and continue. + BB: This should be smarter. Why bother flushing pages that + will be truncated anyway? Also, should we error out here if + the flush returns error? + */ + rc = filemap_write_and_wait(inode->i_mapping); + if (rc != 0) { + cifsInode->write_behind_rc = rc; + rc = 0; + } + } + + if (attrs->ia_valid & ATTR_SIZE) { + rc = cifs_set_file_size(inode, attrs, xid, full_path); + if (rc != 0) + goto out; + } + + /* skip mode change if it's just for clearing setuid/setgid */ + if (attrs->ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) + attrs->ia_valid &= ~ATTR_MODE; + + args = kmalloc(sizeof(*args), GFP_KERNEL); + if (args == NULL) { + rc = -ENOMEM; + goto out; + } + + /* set up the struct */ + if (attrs->ia_valid & ATTR_MODE) + args->mode = attrs->ia_mode; + else + args->mode = NO_CHANGE_64; + + if (attrs->ia_valid & ATTR_UID) + args->uid = attrs->ia_uid; + else + args->uid = NO_CHANGE_64; + + if (attrs->ia_valid & ATTR_GID) + args->gid = attrs->ia_gid; + else + args->gid = NO_CHANGE_64; + + if (attrs->ia_valid & ATTR_ATIME) + args->atime = cifs_UnixTimeToNT(attrs->ia_atime); + else + args->atime = NO_CHANGE_64; + + if (attrs->ia_valid & ATTR_MTIME) + args->mtime = cifs_UnixTimeToNT(attrs->ia_mtime); + else + args->mtime = NO_CHANGE_64; + + if (attrs->ia_valid & ATTR_CTIME) + args->ctime = cifs_UnixTimeToNT(attrs->ia_ctime); + else + args->ctime = NO_CHANGE_64; + + args->device = 0; + rc = CIFSSMBUnixSetInfo(xid, pTcon, full_path, args, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (!rc) + rc = inode_setattr(inode, attrs); +out: + kfree(args); + kfree(full_path); + FreeXid(xid); + return rc; +} + +static int +cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) +{ + int xid; + struct inode *inode = direntry->d_inode; + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifsInodeInfo *cifsInode = CIFS_I(inode); + char *full_path = NULL; + int rc = -EACCES; + __u32 dosattr = 0; + __u64 mode = NO_CHANGE_64; xid = GetXid(); cFYI(1, ("setattr on file %s attrs->iavalid 0x%x", direntry->d_name.name, attrs->ia_valid)); - cifs_sb = CIFS_SB(inode->i_sb); - pTcon = cifs_sb->tcon; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) == 0) { /* check if we have permission to change attrs */ rc = inode_change_ok(inode, attrs); @@ -1453,7 +1777,6 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) FreeXid(xid); return -ENOMEM; } - cifsInode = CIFS_I(inode); if ((attrs->ia_valid & ATTR_MTIME) || (attrs->ia_valid & ATTR_SIZE)) { /* @@ -1472,78 +1795,8 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) } if (attrs->ia_valid & ATTR_SIZE) { - /* To avoid spurious oplock breaks from server, in the case of - inodes that we already have open, avoid doing path based - setting of file size if we can do it by handle. - This keeps our caching token (oplock) and avoids timeouts - when the local oplock break takes longer to flush - writebehind data than the SMB timeout for the SetPathInfo - request would allow */ - - open_file = find_writable_file(cifsInode); - if (open_file) { - __u16 nfid = open_file->netfid; - __u32 npid = open_file->pid; - rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, - nfid, npid, false); - atomic_dec(&open_file->wrtPending); - 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 /* 45 seconds */); - cFYI(1, ("Wrt seteof rc %d", rc)); - } - } else - rc = -EINVAL; - - if (rc != 0) { - /* Set file size by pathname rather than by handle - either because no valid, writeable file handle for - it was found or because there was an error setting - it by handle */ - rc = CIFSSMBSetEOF(xid, pTcon, full_path, - attrs->ia_size, false, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc)); - if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { - __u16 netfid; - int oplock = 0; - - rc = SMBLegacyOpen(xid, pTcon, full_path, - FILE_OPEN, GENERIC_WRITE, - CREATE_NOT_DIR, &netfid, &oplock, - NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - 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 /* 45 sec */); - cFYI(1, ("wrt seteof rc %d", rc)); - CIFSSMBClose(xid, pTcon, netfid); - } - - } - } - - /* Server is ok setting allocation size implicitly - no need - to call: - CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, true, - cifs_sb->local_nls); - */ - - if (rc == 0) { - rc = cifs_vmtruncate(inode, attrs->ia_size); - cifs_truncate_page(inode->i_mapping, inode->i_size); - } else + rc = cifs_set_file_size(inode, attrs, xid, full_path); + if (rc != 0) goto cifs_setattr_exit; } @@ -1554,21 +1807,8 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) * CIFSACL support + proper Windows to Unix idmapping, we may be * able to support this in the future. */ - if (!pTcon->unix_ext && - !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)) { + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)) attrs->ia_valid &= ~(ATTR_UID | ATTR_GID); - } else { - if (attrs->ia_valid & ATTR_UID) { - cFYI(1, ("UID changed to %d", attrs->ia_uid)); - uid = attrs->ia_uid; - } - if (attrs->ia_valid & ATTR_GID) { - cFYI(1, ("GID changed to %d", attrs->ia_gid)); - gid = attrs->ia_gid; - } - } - - time_buf.Attributes = 0; /* skip mode change if it's just for clearing setuid/setgid */ if (attrs->ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) @@ -1579,13 +1819,7 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) mode = attrs->ia_mode; } - if ((pTcon->unix_ext) - && (attrs->ia_valid & (ATTR_MODE | ATTR_GID | ATTR_UID))) - rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode, uid, gid, - 0 /* dev_t */, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - else if (attrs->ia_valid & ATTR_MODE) { + if (attrs->ia_valid & ATTR_MODE) { rc = 0; #ifdef CONFIG_CIFS_EXPERIMENTAL if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) @@ -1594,24 +1828,19 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) #endif if (((mode & S_IWUGO) == 0) && (cifsInode->cifsAttrs & ATTR_READONLY) == 0) { - set_dosattr = true; - time_buf.Attributes = cpu_to_le32(cifsInode->cifsAttrs | - ATTR_READONLY); + + dosattr = cifsInode->cifsAttrs | ATTR_READONLY; + /* fix up mode if we're not using dynperm */ if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) == 0) attrs->ia_mode = inode->i_mode & ~S_IWUGO; } else if ((mode & S_IWUGO) && (cifsInode->cifsAttrs & ATTR_READONLY)) { - /* If file is readonly on server, we would - not be able to write to it - so if any write - bit is enabled for user or group or other we - need to at least try to remove r/o dos attr */ - set_dosattr = true; - time_buf.Attributes = cpu_to_le32(cifsInode->cifsAttrs & - (~ATTR_READONLY)); - /* Windows ignores set to zero */ - if (time_buf.Attributes == 0) - time_buf.Attributes |= cpu_to_le32(ATTR_NORMAL); + + dosattr = cifsInode->cifsAttrs & ~ATTR_READONLY; + /* Attributes of 0 are ignored */ + if (dosattr == 0) + dosattr |= ATTR_NORMAL; /* reset local inode permissions to normal */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)) { @@ -1629,82 +1858,18 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) } } - if (attrs->ia_valid & ATTR_ATIME) { - set_time = true; - time_buf.LastAccessTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_atime)); - } else - time_buf.LastAccessTime = 0; - - if (attrs->ia_valid & ATTR_MTIME) { - set_time = true; - time_buf.LastWriteTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_mtime)); - } else - time_buf.LastWriteTime = 0; - /* Do not set ctime explicitly unless other time - stamps are changed explicitly (i.e. by utime() - since we would then have a mix of client and - server times */ + if (attrs->ia_valid & (ATTR_MTIME|ATTR_ATIME|ATTR_CTIME) || + ((attrs->ia_valid & ATTR_MODE) && dosattr)) { + rc = cifs_set_file_info(inode, attrs, xid, full_path, dosattr); + /* BB: check for rc = -EOPNOTSUPP and switch to legacy mode */ - if (set_time && (attrs->ia_valid & ATTR_CTIME)) { - set_time = true; - /* Although Samba throws this field away - it may be useful to Windows - but we do - not want to set ctime unless some other - timestamp is changing */ - cFYI(1, ("CIFS - CTIME changed")); - time_buf.ChangeTime = - cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_ctime)); - } else - time_buf.ChangeTime = 0; - - if (set_time || set_dosattr) { - time_buf.CreationTime = 0; /* do not change */ - /* In the future we should experiment - try setting timestamps - via Handle (SetFileInfo) instead of by path */ - if (!(pTcon->ses->flags & CIFS_SES_NT4)) - rc = CIFSSMBSetTimes(xid, pTcon, full_path, &time_buf, - cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - else - rc = -EOPNOTSUPP; - - if (rc == -EOPNOTSUPP) { - int oplock = 0; - __u16 netfid; - - cFYI(1, ("calling SetFileInfo since SetPathInfo for " - "times not supported by this server")); - /* BB we could scan to see if we already have it open - and pass in pid of opener to function */ - rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, - SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, - CREATE_NOT_DIR, &netfid, &oplock, - NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc == 0) { - rc = CIFSSMBSetFileTimes(xid, pTcon, &time_buf, - netfid); - CIFSSMBClose(xid, pTcon, netfid); - } else { - /* BB For even older servers we could convert time_buf - into old DOS style which uses two second - granularity */ - - /* rc = CIFSSMBSetTimesLegacy(xid, pTcon, full_path, - &time_buf, cifs_sb->local_nls); */ - } - } /* Even if error on time set, no sense failing the call if the server would set the time to a reasonable value anyway, and this check ensures that we are not being called from sys_utimes in which case we ought to fail the call back to the user when the server rejects the call */ if ((rc) && (attrs->ia_valid & - (ATTR_MODE | ATTR_GID | ATTR_UID | ATTR_SIZE))) + (ATTR_MODE | ATTR_GID | ATTR_UID | ATTR_SIZE))) rc = 0; } @@ -1718,6 +1883,21 @@ cifs_setattr_exit: return rc; } +int +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->tcon; + + if (pTcon->unix_ext) + return cifs_setattr_unix(direntry, attrs); + + return cifs_setattr_nounix(direntry, attrs); + + /* BB: add cifs_setattr_legacy for really old servers */ +} + #if 0 void cifs_delete_inode(struct inode *inode) { diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 4b17f8f..88786ba 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -150,8 +150,7 @@ cifs_buf_get(void) but it may be more efficient to always alloc same size albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ - ret_buf = (struct smb_hdr *) mempool_alloc(cifs_req_poolp, - GFP_KERNEL | GFP_NOFS); + ret_buf = mempool_alloc(cifs_req_poolp, GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -188,8 +187,7 @@ cifs_small_buf_get(void) but it may be more efficient to always alloc same size albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ - ret_buf = (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, - GFP_KERNEL | GFP_NOFS); + ret_buf = mempool_alloc(cifs_sm_req_poolp, GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ @@ -313,8 +311,6 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , buffer->Flags2 = SMBFLG2_KNOWS_LONG_NAMES; buffer->Pid = cpu_to_le16((__u16)current->tgid); buffer->PidHigh = cpu_to_le16((__u16)(current->tgid >> 16)); - spin_lock(&GlobalMid_Lock); - spin_unlock(&GlobalMid_Lock); if (treeCon) { buffer->Tid = treeCon->tid; if (treeCon->ses) { diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 83f3069..765adf1 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -640,6 +640,70 @@ static int is_dir_changed(struct file *file) } +static int cifs_save_resume_key(const char *current_entry, + struct cifsFileInfo *cifsFile) +{ + int rc = 0; + unsigned int len = 0; + __u16 level; + char *filename; + + if ((cifsFile == NULL) || (current_entry == NULL)) + return -EINVAL; + + level = cifsFile->srch_inf.info_level; + + if (level == SMB_FIND_FILE_UNIX) { + FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; + + filename = &pFindData->FileName[0]; + if (cifsFile->srch_inf.unicode) { + len = cifs_unicode_bytelen(filename); + } else { + /* BB should we make this strnlen of PATH_MAX? */ + len = strnlen(filename, PATH_MAX); + } + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; + } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { + FILE_DIRECTORY_INFO *pFindData = + (FILE_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_FULL_DIRECTORY_INFO) { + FILE_FULL_DIRECTORY_INFO *pFindData = + (FILE_FULL_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_ID_FULL_DIR_INFO) { + SEARCH_ID_FULL_DIR_INFO *pFindData = + (SEARCH_ID_FULL_DIR_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { + FILE_BOTH_DIRECTORY_INFO *pFindData = + (FILE_BOTH_DIRECTORY_INFO *)current_entry; + filename = &pFindData->FileName[0]; + len = le32_to_cpu(pFindData->FileNameLength); + cifsFile->srch_inf.resume_key = pFindData->FileIndex; + } else if (level == SMB_FIND_FILE_INFO_STANDARD) { + FIND_FILE_STANDARD_INFO *pFindData = + (FIND_FILE_STANDARD_INFO *)current_entry; + filename = &pFindData->FileName[0]; + /* one byte length, no name conversion */ + len = (unsigned int)pFindData->FileNameLength; + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; + } else { + cFYI(1, ("Unknown findfirst level %d", level)); + return -EINVAL; + } + cifsFile->srch_inf.resume_name_len = len; + cifsFile->srch_inf.presume_name = filename; + return rc; +} + /* find the corresponding entry in the search */ /* Note that the SMB server returns search entries for . and .. which complicates logic here if we choose to parse for them and we do not @@ -690,6 +754,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, else cifs_buf_release(cifsFile->srch_inf. ntwrk_buf_start); + cifsFile->srch_inf.ntwrk_buf_start = NULL; } rc = initiate_cifs_search(xid, file); if (rc) { @@ -702,6 +767,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, while ((index_to_find >= cifsFile->srch_inf.index_of_last_entry) && (rc == 0) && !cifsFile->srch_inf.endOfSearch) { cFYI(1, ("calling findnext2")); + cifs_save_resume_key(cifsFile->srch_inf.last_entry, cifsFile); rc = CIFSFindNext(xid, pTcon, cifsFile->netfid, &cifsFile->srch_inf); if (rc) @@ -918,69 +984,6 @@ static int cifs_filldir(char *pfindEntry, struct file *file, return rc; } -static int cifs_save_resume_key(const char *current_entry, - struct cifsFileInfo *cifsFile) -{ - int rc = 0; - unsigned int len = 0; - __u16 level; - char *filename; - - if ((cifsFile == NULL) || (current_entry == NULL)) - return -EINVAL; - - level = cifsFile->srch_inf.info_level; - - if (level == SMB_FIND_FILE_UNIX) { - FILE_UNIX_INFO *pFindData = (FILE_UNIX_INFO *)current_entry; - - filename = &pFindData->FileName[0]; - if (cifsFile->srch_inf.unicode) { - len = cifs_unicode_bytelen(filename); - } else { - /* BB should we make this strnlen of PATH_MAX? */ - len = strnlen(filename, PATH_MAX); - } - cifsFile->srch_inf.resume_key = pFindData->ResumeKey; - } else if (level == SMB_FIND_FILE_DIRECTORY_INFO) { - FILE_DIRECTORY_INFO *pFindData = - (FILE_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_FULL_DIRECTORY_INFO) { - FILE_FULL_DIRECTORY_INFO *pFindData = - (FILE_FULL_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_ID_FULL_DIR_INFO) { - SEARCH_ID_FULL_DIR_INFO *pFindData = - (SEARCH_ID_FULL_DIR_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_BOTH_DIRECTORY_INFO) { - FILE_BOTH_DIRECTORY_INFO *pFindData = - (FILE_BOTH_DIRECTORY_INFO *)current_entry; - filename = &pFindData->FileName[0]; - len = le32_to_cpu(pFindData->FileNameLength); - cifsFile->srch_inf.resume_key = pFindData->FileIndex; - } else if (level == SMB_FIND_FILE_INFO_STANDARD) { - FIND_FILE_STANDARD_INFO *pFindData = - (FIND_FILE_STANDARD_INFO *)current_entry; - filename = &pFindData->FileName[0]; - /* one byte length, no name conversion */ - len = (unsigned int)pFindData->FileNameLength; - cifsFile->srch_inf.resume_key = pFindData->ResumeKey; - } else { - cFYI(1, ("Unknown findfirst level %d", level)); - return -EINVAL; - } - cifsFile->srch_inf.resume_name_len = len; - cifsFile->srch_inf.presume_name = filename; - return rc; -} int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) { diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index ed150ef..2851d5d 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -409,6 +409,8 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, #ifdef CONFIG_CIFS_WEAK_PW_HASH char lnm_session_key[CIFS_SESS_KEY_SIZE]; + pSMB->req.hdr.Flags2 &= ~SMBFLG2_UNICODE; + /* no capabilities flags in old lanman negotiation */ pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_SESS_KEY_SIZE); @@ -505,7 +507,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, unicode_ssetup_strings(&bcc_ptr, ses, nls_cp); } else ascii_ssetup_strings(&bcc_ptr, ses, nls_cp); - } else if (type == Kerberos) { + } else if (type == Kerberos || type == MSKerberos) { #ifdef CONFIG_CIFS_UPCALL struct cifs_spnego_msg *msg; spnego_key = cifs_get_spnego_key(ses); @@ -516,6 +518,15 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, } msg = spnego_key->payload.data; + /* check version field to make sure that cifs.upcall is + sending us a response in an expected form */ + if (msg->version != CIFS_SPNEGO_UPCALL_VERSION) { + cERROR(1, ("incorrect version of cifs.upcall (expected" + " %d but got %d)", + CIFS_SPNEGO_UPCALL_VERSION, msg->version)); + rc = -EKEYREJECTED; + goto ssetup_exit; + } /* bail out if key is too long */ if (msg->sesskey_len > sizeof(ses->server->mac_signing_key.data.krb5)) { @@ -613,8 +624,10 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, ses, nls_cp); ssetup_exit: - if (spnego_key) + if (spnego_key) { + key_revoke(spnego_key); key_put(spnego_key); + } kfree(str_area); if (resp_buf_type == CIFS_SMALL_BUFFER) { cFYI(1, ("ssetup freeing small buf %p", iov[0].iov_base)); diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 000ac50..bf0e6d8 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -50,8 +50,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) return NULL; } - temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - GFP_KERNEL | GFP_NOFS); + temp = mempool_alloc(cifs_mid_poolp, GFP_NOFS); if (temp == NULL) return temp; else { @@ -265,6 +264,7 @@ smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, cFYI(1, ("Sending smb: total_len %d", total_len)); dump_smb(smb_buffer, len); + i = 0; while (total_len) { rc = kernel_sendmsg(ssocket, &smb_msg, &iov[first_vec], n_vec - first_vec, total_len); diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 3d2580e..c591622 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -137,9 +137,11 @@ exit: } -int coda_permission(struct inode *inode, int mask, struct nameidata *nd) +int coda_permission(struct inode *inode, int mask) { int error = 0; + + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; if (!mask) return 0; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 2f58dfc..830f51a 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode) kmem_cache_free(coda_inode_cachep, ITOC(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index c21a1f5..c513654 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -24,8 +24,7 @@ #include <linux/coda_psdev.h> /* pioctl ops */ -static int coda_ioctl_permission(struct inode *inode, int mask, - struct nameidata *nd); +static int coda_ioctl_permission(struct inode *inode, int mask); static int coda_pioctl(struct inode * inode, struct file * filp, unsigned int cmd, unsigned long user_data); @@ -42,8 +41,7 @@ const struct file_operations coda_ioctl_operations = { }; /* the coda pioctl inode ops */ -static int coda_ioctl_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int coda_ioctl_permission(struct inode *inode, int mask) { return 0; } @@ -51,7 +49,7 @@ static int coda_ioctl_permission(struct inode *inode, int mask, static int coda_pioctl(struct inode * inode, struct file * filp, unsigned int cmd, unsigned long user_data) { - struct nameidata nd; + struct path path; int error; struct PioctlData data; struct inode *target_inode = NULL; @@ -66,21 +64,21 @@ static int coda_pioctl(struct inode * inode, struct file * filp, * Look up the pathname. Note that the pathname is in * user memory, and namei takes care of this */ - if ( data.follow ) { - error = user_path_walk(data.path, &nd); + if (data.follow) { + error = user_path(data.path, &path); } else { - error = user_path_walk_link(data.path, &nd); + error = user_lpath(data.path, &path); } if ( error ) { return error; } else { - target_inode = nd.path.dentry->d_inode; + target_inode = path.dentry->d_inode; } /* return if it is not a Coda inode */ if ( target_inode->i_sb != inode->i_sb ) { - path_put(&nd.path); + path_put(&path); return -EINVAL; } @@ -89,7 +87,7 @@ static int coda_pioctl(struct inode * inode, struct file * filp, error = venus_pioctl(inode->i_sb, &(cnp->c_fid), cmd, &data); - path_put(&nd.path); + path_put(&path); return error; } diff --git a/fs/compat.c b/fs/compat.c index 106eba2..075d050 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -234,18 +234,18 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs * * The following statfs calls are copies of code from fs/open.c and * should be checked against those from time to time */ -asmlinkage long compat_sys_statfs(const char __user *path, struct compat_statfs __user *buf) +asmlinkage long compat_sys_statfs(const char __user *pathname, struct compat_statfs __user *buf) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.path.dentry, &tmp); + error = vfs_statfs(path.dentry, &tmp); if (!error) error = put_compat_statfs(buf, &tmp); - path_put(&nd.path); + path_put(&path); } return error; } @@ -299,21 +299,21 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat return 0; } -asmlinkage long compat_sys_statfs64(const char __user *path, compat_size_t sz, struct compat_statfs64 __user *buf) +asmlinkage long compat_sys_statfs64(const char __user *pathname, compat_size_t sz, struct compat_statfs64 __user *buf) { - struct nameidata nd; + struct path path; int error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct kstatfs tmp; - error = vfs_statfs(nd.path.dentry, &tmp); + error = vfs_statfs(path.dentry, &tmp); if (!error) error = put_compat_statfs64(buf, &tmp); - path_put(&nd.path); + path_put(&path); } return error; } @@ -792,8 +792,10 @@ static int compat_fillonedir(void *__buf, const char *name, int namlen, if (buf->result) return -EINVAL; d_ino = ino; - if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) + if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) { + buf->result = -EOVERFLOW; return -EOVERFLOW; + } buf->result++; dirent = buf->dirent; if (!access_ok(VERIFY_WRITE, dirent, @@ -862,8 +864,10 @@ static int compat_filldir(void *__buf, const char *name, int namlen, if (reclen > buf->count) return -EINVAL; d_ino = ino; - if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) + if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) { + buf->error = -EOVERFLOW; return -EOVERFLOW; + } dirent = buf->previous; if (dirent) { if (__put_user(offset, &dirent->d_off)) diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index da015c1..762d287 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -49,8 +49,10 @@ struct configfs_dirent { #define CONFIGFS_USET_DEFAULT 0x0080 #define CONFIGFS_USET_DROPPING 0x0100 #define CONFIGFS_USET_IN_MKDIR 0x0200 +#define CONFIGFS_USET_CREATING 0x0400 #define CONFIGFS_NOT_PINNED (CONFIGFS_ITEM_ATTR) +extern struct mutex configfs_symlink_mutex; extern spinlock_t configfs_dirent_lock; extern struct vfsmount * configfs_mount; @@ -66,6 +68,7 @@ extern void configfs_inode_exit(void); extern int configfs_create_file(struct config_item *, const struct configfs_attribute *); extern int configfs_make_dirent(struct configfs_dirent *, struct dentry *, void *, umode_t, int); +extern int configfs_dirent_is_ready(struct configfs_dirent *); extern int configfs_add_file(struct dentry *, const struct configfs_attribute *, int); extern void configfs_hash_and_remove(struct dentry * dir, const char * name); diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 179589b..8e93341 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -185,7 +185,7 @@ static int create_dir(struct config_item * k, struct dentry * p, error = configfs_dirent_exists(p->d_fsdata, d->d_name.name); if (!error) error = configfs_make_dirent(p->d_fsdata, d, k, mode, - CONFIGFS_DIR); + CONFIGFS_DIR | CONFIGFS_USET_CREATING); if (!error) { error = configfs_create(d, mode, init_dir); if (!error) { @@ -209,6 +209,9 @@ static int create_dir(struct config_item * k, struct dentry * p, * configfs_create_dir - create a directory for an config_item. * @item: config_itemwe're creating directory for. * @dentry: config_item's dentry. + * + * Note: user-created entries won't be allowed under this new directory + * until it is validated by configfs_dir_set_ready() */ static int configfs_create_dir(struct config_item * item, struct dentry *dentry) @@ -231,6 +234,44 @@ static int configfs_create_dir(struct config_item * item, struct dentry *dentry) return error; } +/* + * Allow userspace to create new entries under a new directory created with + * configfs_create_dir(), and under all of its chidlren directories recursively. + * @sd configfs_dirent of the new directory to validate + * + * Caller must hold configfs_dirent_lock. + */ +static void configfs_dir_set_ready(struct configfs_dirent *sd) +{ + struct configfs_dirent *child_sd; + + sd->s_type &= ~CONFIGFS_USET_CREATING; + list_for_each_entry(child_sd, &sd->s_children, s_sibling) + if (child_sd->s_type & CONFIGFS_USET_CREATING) + configfs_dir_set_ready(child_sd); +} + +/* + * Check that a directory does not belong to a directory hierarchy being + * attached and not validated yet. + * @sd configfs_dirent of the directory to check + * + * @return non-zero iff the directory was validated + * + * Note: takes configfs_dirent_lock, so the result may change from false to true + * in two consecutive calls, but never from true to false. + */ +int configfs_dirent_is_ready(struct configfs_dirent *sd) +{ + int ret; + + spin_lock(&configfs_dirent_lock); + ret = !(sd->s_type & CONFIGFS_USET_CREATING); + spin_unlock(&configfs_dirent_lock); + + return ret; +} + int configfs_create_link(struct configfs_symlink *sl, struct dentry *parent, struct dentry *dentry) @@ -283,6 +324,8 @@ static void remove_dir(struct dentry * d) * The only thing special about this is that we remove any files in * the directory before we remove the directory, and we've inlined * what used to be configfs_rmdir() below, instead of calling separately. + * + * Caller holds the mutex of the item's inode */ static void configfs_remove_dir(struct config_item * item) @@ -330,7 +373,19 @@ static struct dentry * configfs_lookup(struct inode *dir, struct configfs_dirent * parent_sd = dentry->d_parent->d_fsdata; struct configfs_dirent * sd; int found = 0; - int err = 0; + int err; + + /* + * Fake invisibility if dir belongs to a group/default groups hierarchy + * being attached + * + * This forbids userspace to read/write attributes of items which may + * not complete their initialization, since the dentries of the + * attributes won't be instantiated. + */ + err = -ENOENT; + if (!configfs_dirent_is_ready(parent_sd)) + goto out; list_for_each_entry(sd, &parent_sd->s_children, s_sibling) { if (sd->s_type & CONFIGFS_NOT_PINNED) { @@ -353,6 +408,7 @@ static struct dentry * configfs_lookup(struct inode *dir, return simple_lookup(dir, dentry, nd); } +out: return ERR_PTR(err); } @@ -370,13 +426,17 @@ static int configfs_detach_prep(struct dentry *dentry, struct mutex **wait_mutex struct configfs_dirent *sd; int ret; + /* Mark that we're trying to drop the group */ + parent_sd->s_type |= CONFIGFS_USET_DROPPING; + ret = -EBUSY; if (!list_empty(&parent_sd->s_links)) goto out; ret = 0; list_for_each_entry(sd, &parent_sd->s_children, s_sibling) { - if (sd->s_type & CONFIGFS_NOT_PINNED) + if (!sd->s_element || + (sd->s_type & CONFIGFS_NOT_PINNED)) continue; if (sd->s_type & CONFIGFS_USET_DEFAULT) { /* Abort if racing with mkdir() */ @@ -385,8 +445,6 @@ static int configfs_detach_prep(struct dentry *dentry, struct mutex **wait_mutex *wait_mutex = &sd->s_dentry->d_inode->i_mutex; return -EAGAIN; } - /* Mark that we're trying to drop the group */ - sd->s_type |= CONFIGFS_USET_DROPPING; /* * Yup, recursive. If there's a problem, blame @@ -414,12 +472,11 @@ static void configfs_detach_rollback(struct dentry *dentry) struct configfs_dirent *parent_sd = dentry->d_fsdata; struct configfs_dirent *sd; - list_for_each_entry(sd, &parent_sd->s_children, s_sibling) { - if (sd->s_type & CONFIGFS_USET_DEFAULT) { + parent_sd->s_type &= ~CONFIGFS_USET_DROPPING; + + list_for_each_entry(sd, &parent_sd->s_children, s_sibling) + if (sd->s_type & CONFIGFS_USET_DEFAULT) configfs_detach_rollback(sd->s_dentry); - sd->s_type &= ~CONFIGFS_USET_DROPPING; - } - } } static void detach_attrs(struct config_item * item) @@ -558,36 +615,21 @@ static int create_default_group(struct config_group *parent_group, static int populate_groups(struct config_group *group) { struct config_group *new_group; - struct dentry *dentry = group->cg_item.ci_dentry; int ret = 0; int i; if (group->default_groups) { - /* - * FYI, we're faking mkdir here - * I'm not sure we need this semaphore, as we're called - * from our parent's mkdir. That holds our parent's - * i_mutex, so afaik lookup cannot continue through our - * parent to find us, let alone mess with our tree. - * That said, taking our i_mutex is closer to mkdir - * emulation, and shouldn't hurt. - */ - mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); - for (i = 0; group->default_groups[i]; i++) { new_group = group->default_groups[i]; ret = create_default_group(group, new_group); - if (ret) + if (ret) { + detach_groups(group); break; + } } - - mutex_unlock(&dentry->d_inode->i_mutex); } - if (ret) - detach_groups(group); - return ret; } @@ -702,7 +744,15 @@ static int configfs_attach_item(struct config_item *parent_item, if (!ret) { ret = populate_attrs(item); if (ret) { + /* + * We are going to remove an inode and its dentry but + * the VFS may already have hit and used them. Thus, + * we must lock them as rmdir() would. + */ + mutex_lock(&dentry->d_inode->i_mutex); configfs_remove_dir(item); + dentry->d_inode->i_flags |= S_DEAD; + mutex_unlock(&dentry->d_inode->i_mutex); d_delete(dentry); } } @@ -710,6 +760,7 @@ static int configfs_attach_item(struct config_item *parent_item, return ret; } +/* Caller holds the mutex of the item's inode */ static void configfs_detach_item(struct config_item *item) { detach_attrs(item); @@ -728,16 +779,30 @@ static int configfs_attach_group(struct config_item *parent_item, sd = dentry->d_fsdata; sd->s_type |= CONFIGFS_USET_DIR; + /* + * FYI, we're faking mkdir in populate_groups() + * We must lock the group's inode to avoid races with the VFS + * which can already hit the inode and try to add/remove entries + * under it. + * + * We must also lock the inode to remove it safely in case of + * error, as rmdir() would. + */ + mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); ret = populate_groups(to_config_group(item)); if (ret) { configfs_detach_item(item); - d_delete(dentry); + dentry->d_inode->i_flags |= S_DEAD; } + mutex_unlock(&dentry->d_inode->i_mutex); + if (ret) + d_delete(dentry); } return ret; } +/* Caller holds the mutex of the group's inode */ static void configfs_detach_group(struct config_item *item) { detach_groups(to_config_group(item)); @@ -1035,7 +1100,7 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) struct configfs_subsystem *subsys; struct configfs_dirent *sd; struct config_item_type *type; - struct module *owner = NULL; + struct module *subsys_owner = NULL, *new_item_owner = NULL; char *name; if (dentry->d_parent == configfs_sb->s_root) { @@ -1044,6 +1109,16 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) } sd = dentry->d_parent->d_fsdata; + + /* + * Fake invisibility if dir belongs to a group/default groups hierarchy + * being attached + */ + if (!configfs_dirent_is_ready(sd)) { + ret = -ENOENT; + goto out; + } + if (!(sd->s_type & CONFIGFS_USET_DIR)) { ret = -EPERM; goto out; @@ -1062,10 +1137,25 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) goto out_put; } + /* + * The subsystem may belong to a different module than the item + * being created. We don't want to safely pin the new item but + * fail to pin the subsystem it sits under. + */ + if (!subsys->su_group.cg_item.ci_type) { + ret = -EINVAL; + goto out_put; + } + subsys_owner = subsys->su_group.cg_item.ci_type->ct_owner; + if (!try_module_get(subsys_owner)) { + ret = -EINVAL; + goto out_put; + } + name = kmalloc(dentry->d_name.len + 1, GFP_KERNEL); if (!name) { ret = -ENOMEM; - goto out_put; + goto out_subsys_put; } snprintf(name, dentry->d_name.len + 1, "%s", dentry->d_name.name); @@ -1094,10 +1184,10 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) kfree(name); if (ret) { /* - * If item == NULL, then link_obj() was never called. + * If ret != 0, then link_obj() was never called. * There are no extra references to clean up. */ - goto out_put; + goto out_subsys_put; } /* @@ -1111,8 +1201,8 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) goto out_unlink; } - owner = type->ct_owner; - if (!try_module_get(owner)) { + new_item_owner = type->ct_owner; + if (!try_module_get(new_item_owner)) { ret = -EINVAL; goto out_unlink; } @@ -1142,6 +1232,8 @@ static int configfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) spin_lock(&configfs_dirent_lock); sd->s_type &= ~CONFIGFS_USET_IN_MKDIR; + if (!ret) + configfs_dir_set_ready(dentry->d_fsdata); spin_unlock(&configfs_dirent_lock); out_unlink: @@ -1159,9 +1251,13 @@ out_unlink: mutex_unlock(&subsys->su_mutex); if (module_got) - module_put(owner); + module_put(new_item_owner); } +out_subsys_put: + if (ret) + module_put(subsys_owner); + out_put: /* * link_obj()/link_group() took a reference from child->parent, @@ -1180,7 +1276,7 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry) struct config_item *item; struct configfs_subsystem *subsys; struct configfs_dirent *sd; - struct module *owner = NULL; + struct module *subsys_owner = NULL, *dead_item_owner = NULL; int ret; if (dentry->d_parent == configfs_sb->s_root) @@ -1207,14 +1303,26 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry) return -EINVAL; } - spin_lock(&configfs_dirent_lock); + /* configfs_mkdir() shouldn't have allowed this */ + BUG_ON(!subsys->su_group.cg_item.ci_type); + subsys_owner = subsys->su_group.cg_item.ci_type->ct_owner; + + /* + * Ensure that no racing symlink() will make detach_prep() fail while + * the new link is temporarily attached + */ do { struct mutex *wait_mutex; + mutex_lock(&configfs_symlink_mutex); + spin_lock(&configfs_dirent_lock); ret = configfs_detach_prep(dentry, &wait_mutex); - if (ret) { + if (ret) configfs_detach_rollback(dentry); - spin_unlock(&configfs_dirent_lock); + spin_unlock(&configfs_dirent_lock); + mutex_unlock(&configfs_symlink_mutex); + + if (ret) { if (ret != -EAGAIN) { config_item_put(parent_item); return ret; @@ -1223,11 +1331,8 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry) /* Wait until the racing operation terminates */ mutex_lock(wait_mutex); mutex_unlock(wait_mutex); - - spin_lock(&configfs_dirent_lock); } } while (ret == -EAGAIN); - spin_unlock(&configfs_dirent_lock); /* Get a working ref for the duration of this function */ item = configfs_get_config_item(dentry); @@ -1236,7 +1341,7 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry) config_item_put(parent_item); if (item->ci_type) - owner = item->ci_type->ct_owner; + dead_item_owner = item->ci_type->ct_owner; if (sd->s_type & CONFIGFS_USET_DIR) { configfs_detach_group(item); @@ -1258,7 +1363,8 @@ static int configfs_rmdir(struct inode *dir, struct dentry *dentry) /* Drop our reference from above */ config_item_put(item); - module_put(owner); + module_put(dead_item_owner); + module_put(subsys_owner); return 0; } @@ -1314,13 +1420,24 @@ static int configfs_dir_open(struct inode *inode, struct file *file) { struct dentry * dentry = file->f_path.dentry; struct configfs_dirent * parent_sd = dentry->d_fsdata; + int err; mutex_lock(&dentry->d_inode->i_mutex); - file->private_data = configfs_new_dirent(parent_sd, NULL); + /* + * Fake invisibility if dir belongs to a group/default groups hierarchy + * being attached + */ + err = -ENOENT; + if (configfs_dirent_is_ready(parent_sd)) { + file->private_data = configfs_new_dirent(parent_sd, NULL); + if (IS_ERR(file->private_data)) + err = PTR_ERR(file->private_data); + else + err = 0; + } mutex_unlock(&dentry->d_inode->i_mutex); - return IS_ERR(file->private_data) ? PTR_ERR(file->private_data) : 0; - + return err; } static int configfs_dir_close(struct inode *inode, struct file *file) @@ -1491,6 +1608,10 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) if (err) { d_delete(dentry); dput(dentry); + } else { + spin_lock(&configfs_dirent_lock); + configfs_dir_set_ready(dentry->d_fsdata); + spin_unlock(&configfs_dirent_lock); } } @@ -1517,11 +1638,13 @@ void configfs_unregister_subsystem(struct configfs_subsystem *subsys) mutex_lock_nested(&configfs_sb->s_root->d_inode->i_mutex, I_MUTEX_PARENT); mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); + mutex_lock(&configfs_symlink_mutex); spin_lock(&configfs_dirent_lock); if (configfs_detach_prep(dentry, NULL)) { printk(KERN_ERR "configfs: Tried to unregister non-empty subsystem!\n"); } spin_unlock(&configfs_dirent_lock); + mutex_unlock(&configfs_symlink_mutex); configfs_detach_group(&group->cg_item); dentry->d_inode->i_flags |= S_DEAD; mutex_unlock(&dentry->d_inode->i_mutex); diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index 0004d18..bf74973 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -31,6 +31,9 @@ #include <linux/configfs.h> #include "configfs_internal.h" +/* Protects attachments of new symlinks */ +DEFINE_MUTEX(configfs_symlink_mutex); + static int item_depth(struct config_item * item) { struct config_item * p = item; @@ -73,11 +76,20 @@ static int create_link(struct config_item *parent_item, struct configfs_symlink *sl; int ret; + ret = -ENOENT; + if (!configfs_dirent_is_ready(target_sd)) + goto out; ret = -ENOMEM; sl = kmalloc(sizeof(struct configfs_symlink), GFP_KERNEL); if (sl) { sl->sl_target = config_item_get(item); spin_lock(&configfs_dirent_lock); + if (target_sd->s_type & CONFIGFS_USET_DROPPING) { + spin_unlock(&configfs_dirent_lock); + config_item_put(item); + kfree(sl); + return -ENOENT; + } list_add(&sl->sl_list, &target_sd->s_links); spin_unlock(&configfs_dirent_lock); ret = configfs_create_link(sl, parent_item->ci_dentry, @@ -91,6 +103,7 @@ static int create_link(struct config_item *parent_item, } } +out: return ret; } @@ -120,6 +133,7 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna { int ret; struct nameidata nd; + struct configfs_dirent *sd; struct config_item *parent_item; struct config_item *target_item; struct config_item_type *type; @@ -128,9 +142,19 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna if (dentry->d_parent == configfs_sb->s_root) goto out; + sd = dentry->d_parent->d_fsdata; + /* + * Fake invisibility if dir belongs to a group/default groups hierarchy + * being attached + */ + ret = -ENOENT; + if (!configfs_dirent_is_ready(sd)) + goto out; + parent_item = configfs_get_config_item(dentry->d_parent); type = parent_item->ci_type; + ret = -EPERM; if (!type || !type->ct_item_ops || !type->ct_item_ops->allow_link) goto out_put; @@ -141,7 +165,9 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna ret = type->ct_item_ops->allow_link(parent_item, target_item); if (!ret) { + mutex_lock(&configfs_symlink_mutex); ret = create_link(parent_item, target_item, dentry); + mutex_unlock(&configfs_symlink_mutex); if (ret && type->ct_item_ops->drop_link) type->ct_item_ops->drop_link(parent_item, target_item); diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 0c3b618..f40423e 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -43,58 +43,13 @@ static DEFINE_MUTEX(read_mutex); static int cramfs_iget5_test(struct inode *inode, void *opaque) { struct cramfs_inode *cramfs_inode = opaque; - - if (inode->i_ino != CRAMINO(cramfs_inode)) - return 0; /* does not match */ - - if (inode->i_ino != 1) - return 1; - - /* all empty directories, char, block, pipe, and sock, share inode #1 */ - - if ((inode->i_mode != cramfs_inode->mode) || - (inode->i_gid != cramfs_inode->gid) || - (inode->i_uid != cramfs_inode->uid)) - return 0; /* does not match */ - - if ((S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) && - (inode->i_rdev != old_decode_dev(cramfs_inode->size))) - return 0; /* does not match */ - - return 1; /* matches */ + return inode->i_ino == CRAMINO(cramfs_inode) && inode->i_ino != 1; } static int cramfs_iget5_set(struct inode *inode, void *opaque) { - static struct timespec zerotime; struct cramfs_inode *cramfs_inode = opaque; - inode->i_mode = cramfs_inode->mode; - inode->i_uid = cramfs_inode->uid; - inode->i_size = cramfs_inode->size; - inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1; - inode->i_gid = cramfs_inode->gid; - /* Struct copy intentional */ - inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime; inode->i_ino = CRAMINO(cramfs_inode); - /* inode->i_nlink is left 1 - arguably wrong for directories, - but it's the best we can do without reading the directory - contents. 1 yields the right result in GNU find, even - without -noleaf option. */ - if (S_ISREG(inode->i_mode)) { - inode->i_fop = &generic_ro_fops; - inode->i_data.a_ops = &cramfs_aops; - } else if (S_ISDIR(inode->i_mode)) { - inode->i_op = &cramfs_dir_inode_operations; - inode->i_fop = &cramfs_directory_operations; - } else if (S_ISLNK(inode->i_mode)) { - inode->i_op = &page_symlink_inode_operations; - inode->i_data.a_ops = &cramfs_aops; - } else { - inode->i_size = 0; - inode->i_blocks = 0; - init_special_inode(inode, inode->i_mode, - old_decode_dev(cramfs_inode->size)); - } return 0; } @@ -104,12 +59,48 @@ static struct inode *get_cramfs_inode(struct super_block *sb, struct inode *inode = iget5_locked(sb, CRAMINO(cramfs_inode), cramfs_iget5_test, cramfs_iget5_set, cramfs_inode); + static struct timespec zerotime; + if (inode && (inode->i_state & I_NEW)) { + inode->i_mode = cramfs_inode->mode; + inode->i_uid = cramfs_inode->uid; + inode->i_size = cramfs_inode->size; + inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1; + inode->i_gid = cramfs_inode->gid; + /* Struct copy intentional */ + inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime; + /* inode->i_nlink is left 1 - arguably wrong for directories, + but it's the best we can do without reading the directory + contents. 1 yields the right result in GNU find, even + without -noleaf option. */ + if (S_ISREG(inode->i_mode)) { + inode->i_fop = &generic_ro_fops; + inode->i_data.a_ops = &cramfs_aops; + } else if (S_ISDIR(inode->i_mode)) { + inode->i_op = &cramfs_dir_inode_operations; + inode->i_fop = &cramfs_directory_operations; + } else if (S_ISLNK(inode->i_mode)) { + inode->i_op = &page_symlink_inode_operations; + inode->i_data.a_ops = &cramfs_aops; + } else { + inode->i_size = 0; + inode->i_blocks = 0; + init_special_inode(inode, inode->i_mode, + old_decode_dev(cramfs_inode->size)); + } unlock_new_inode(inode); } return inode; } +static void cramfs_drop_inode(struct inode *inode) +{ + if (inode->i_ino == 1) + generic_delete_inode(inode); + else + generic_drop_inode(inode); +} + /* * We have our own block cache: don't fill up the buffer cache * with the rom-image, because the way the filesystem is set @@ -534,6 +525,7 @@ static const struct super_operations cramfs_ops = { .put_super = cramfs_put_super, .remount_fs = cramfs_remount, .statfs = cramfs_statfs, + .drop_inode = cramfs_drop_inode, }; static int cramfs_get_sb(struct file_system_type *fs_type, diff --git a/fs/dcache.c b/fs/dcache.c index 3818d6a..e7a1a99 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -487,6 +487,7 @@ restart: if (!cnt) break; } + cond_resched_lock(&dcache_lock); } } while (!list_empty(&tmp)) { @@ -1219,6 +1220,107 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) return new; } +/** + * d_add_ci - lookup or allocate new dentry with case-exact name + * @inode: the inode case-insensitive lookup has found + * @dentry: the negative dentry that was passed to the parent's lookup func + * @name: the case-exact name to be associated with the returned dentry + * + * This is to avoid filling the dcache with case-insensitive names to the + * same inode, only the actual correct case is stored in the dcache for + * case-insensitive filesystems. + * + * For a case-insensitive lookup match and if the the case-exact dentry + * already exists in in the dcache, use it and return it. + * + * If no entry exists with the exact case name, allocate new dentry with + * the exact case, and return the spliced entry. + */ +struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, + struct qstr *name) +{ + int error; + struct dentry *found; + struct dentry *new; + + /* Does a dentry matching the name exist already? */ + found = d_hash_and_lookup(dentry->d_parent, name); + /* If not, create it now and return */ + if (!found) { + new = d_alloc(dentry->d_parent, name); + if (!new) { + error = -ENOMEM; + goto err_out; + } + found = d_splice_alias(inode, new); + if (found) { + dput(new); + return found; + } + return new; + } + /* Matching dentry exists, check if it is negative. */ + if (found->d_inode) { + if (unlikely(found->d_inode != inode)) { + /* This can't happen because bad inodes are unhashed. */ + BUG_ON(!is_bad_inode(inode)); + BUG_ON(!is_bad_inode(found->d_inode)); + } + /* + * Already have the inode and the dentry attached, decrement + * the reference count to balance the iget() done + * earlier on. We found the dentry using d_lookup() so it + * cannot be disconnected and thus we do not need to worry + * about any NFS/disconnectedness issues here. + */ + iput(inode); + return found; + } + /* + * Negative dentry: instantiate it unless the inode is a directory and + * has a 'disconnected' dentry (i.e. IS_ROOT and DCACHE_DISCONNECTED), + * in which case d_move() that in place of the found dentry. + */ + if (!S_ISDIR(inode->i_mode)) { + /* Not a directory; everything is easy. */ + d_instantiate(found, inode); + return found; + } + spin_lock(&dcache_lock); + if (list_empty(&inode->i_dentry)) { + /* + * Directory without a 'disconnected' dentry; we need to do + * d_instantiate() by hand because it takes dcache_lock which + * we already hold. + */ + list_add(&found->d_alias, &inode->i_dentry); + found->d_inode = inode; + spin_unlock(&dcache_lock); + security_d_instantiate(found, inode); + return found; + } + /* + * Directory with a 'disconnected' dentry; get a reference to the + * 'disconnected' dentry. + */ + new = list_entry(inode->i_dentry.next, struct dentry, d_alias); + dget_locked(new); + spin_unlock(&dcache_lock); + /* Do security vodoo. */ + security_d_instantiate(found, inode); + /* Move new in place of found. */ + d_move(new, found); + /* Balance the iget() we did above. */ + iput(inode); + /* Throw away found. */ + dput(found); + /* Use new as the actual dentry. */ + return new; + +err_out: + iput(inode); + return ERR_PTR(error); +} /** * d_lookup - search for a dentry @@ -1293,6 +1395,10 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) if (dentry->d_parent != parent) goto next; + /* non-existing due to RCU? */ + if (d_unhashed(dentry)) + goto next; + /* * It is safe to compare names since d_move() cannot * change the qstr (protected by d_lock). @@ -1308,10 +1414,8 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) goto next; } - if (!d_unhashed(dentry)) { - atomic_inc(&dentry->d_count); - found = dentry; - } + atomic_inc(&dentry->d_count); + found = dentry; spin_unlock(&dentry->d_lock); break; next: @@ -2253,6 +2357,7 @@ EXPORT_SYMBOL(d_path); EXPORT_SYMBOL(d_prune_aliases); EXPORT_SYMBOL(d_rehash); EXPORT_SYMBOL(d_splice_alias); +EXPORT_SYMBOL(d_add_ci); EXPORT_SYMBOL(d_validate); EXPORT_SYMBOL(dget_locked); EXPORT_SYMBOL(dput); diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 285b64a..488eb42 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -29,7 +29,7 @@ #define DEVPTS_DEFAULT_MODE 0600 extern int pty_limit; /* Config limit on Unix98 ptys */ -static DEFINE_IDR(allocated_ptys); +static DEFINE_IDA(allocated_ptys); static DEFINE_MUTEX(allocated_ptys_lock); static struct vfsmount *devpts_mnt; @@ -180,24 +180,24 @@ static struct dentry *get_node(int num) int devpts_new_index(void) { int index; - int idr_ret; + int ida_ret; retry: - if (!idr_pre_get(&allocated_ptys, GFP_KERNEL)) { + if (!ida_pre_get(&allocated_ptys, GFP_KERNEL)) { return -ENOMEM; } mutex_lock(&allocated_ptys_lock); - idr_ret = idr_get_new(&allocated_ptys, NULL, &index); - if (idr_ret < 0) { + ida_ret = ida_get_new(&allocated_ptys, &index); + if (ida_ret < 0) { mutex_unlock(&allocated_ptys_lock); - if (idr_ret == -EAGAIN) + if (ida_ret == -EAGAIN) goto retry; return -EIO; } if (index >= pty_limit) { - idr_remove(&allocated_ptys, index); + ida_remove(&allocated_ptys, index); mutex_unlock(&allocated_ptys_lock); return -EIO; } @@ -208,7 +208,7 @@ retry: void devpts_kill_index(int idx) { mutex_lock(&allocated_ptys_lock); - idr_remove(&allocated_ptys, idx); + ida_remove(&allocated_ptys, idx); mutex_unlock(&allocated_ptys_lock); } diff --git a/fs/direct-io.c b/fs/direct-io.c index 9e81add..9606ee8 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -150,17 +150,11 @@ static int dio_refill_pages(struct dio *dio) int nr_pages; nr_pages = min(dio->total_pages - dio->curr_page, DIO_PAGES); - down_read(¤t->mm->mmap_sem); - ret = get_user_pages( - current, /* Task for fault acounting */ - current->mm, /* whose pages? */ + ret = get_user_pages_fast( dio->curr_user_address, /* Where from? */ nr_pages, /* How many pages? */ dio->rw == READ, /* Write to memory? */ - 0, /* force (?) */ - &dio->pages[0], - NULL); /* vmas */ - up_read(¤t->mm->mmap_sem); + &dio->pages[0]); /* Put results here */ if (ret < 0 && dio->blocks_available && (dio->rw & WRITE)) { struct page *page = ZERO_PAGE(0); diff --git a/fs/dlm/config.c b/fs/dlm/config.c index c4e7d72..fd9859f 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 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 @@ -14,6 +14,9 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/configfs.h> +#include <linux/in.h> +#include <linux/in6.h> +#include <net/ipv6.h> #include <net/sock.h> #include "config.h" @@ -30,16 +33,16 @@ static struct config_group *space_list; static struct config_group *comm_list; -static struct comm *local_comm; +static struct dlm_comm *local_comm; -struct clusters; -struct cluster; -struct spaces; -struct space; -struct comms; -struct comm; -struct nodes; -struct node; +struct dlm_clusters; +struct dlm_cluster; +struct dlm_spaces; +struct dlm_space; +struct dlm_comms; +struct dlm_comm; +struct dlm_nodes; +struct dlm_node; static struct config_group *make_cluster(struct config_group *, const char *); static void drop_cluster(struct config_group *, struct config_item *); @@ -68,17 +71,22 @@ static ssize_t show_node(struct config_item *i, struct configfs_attribute *a, static ssize_t store_node(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len); -static ssize_t comm_nodeid_read(struct comm *cm, char *buf); -static ssize_t comm_nodeid_write(struct comm *cm, const char *buf, size_t len); -static ssize_t comm_local_read(struct comm *cm, char *buf); -static ssize_t comm_local_write(struct comm *cm, const char *buf, size_t len); -static ssize_t comm_addr_write(struct comm *cm, const char *buf, size_t len); -static ssize_t node_nodeid_read(struct node *nd, char *buf); -static ssize_t node_nodeid_write(struct node *nd, const char *buf, size_t len); -static ssize_t node_weight_read(struct node *nd, char *buf); -static ssize_t node_weight_write(struct node *nd, const char *buf, size_t len); - -struct cluster { +static ssize_t comm_nodeid_read(struct dlm_comm *cm, char *buf); +static ssize_t comm_nodeid_write(struct dlm_comm *cm, const char *buf, + size_t len); +static ssize_t comm_local_read(struct dlm_comm *cm, char *buf); +static ssize_t comm_local_write(struct dlm_comm *cm, const char *buf, + size_t len); +static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, + size_t len); +static ssize_t node_nodeid_read(struct dlm_node *nd, char *buf); +static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf, + size_t len); +static ssize_t node_weight_read(struct dlm_node *nd, char *buf); +static ssize_t node_weight_write(struct dlm_node *nd, const char *buf, + size_t len); + +struct dlm_cluster { struct config_group group; unsigned int cl_tcp_port; unsigned int cl_buffer_size; @@ -109,11 +117,11 @@ enum { struct cluster_attribute { struct configfs_attribute attr; - ssize_t (*show)(struct cluster *, char *); - ssize_t (*store)(struct cluster *, const char *, size_t); + ssize_t (*show)(struct dlm_cluster *, char *); + ssize_t (*store)(struct dlm_cluster *, const char *, size_t); }; -static ssize_t cluster_set(struct cluster *cl, unsigned int *cl_field, +static ssize_t cluster_set(struct dlm_cluster *cl, unsigned int *cl_field, int *info_field, int check_zero, const char *buf, size_t len) { @@ -134,12 +142,12 @@ static ssize_t cluster_set(struct cluster *cl, unsigned int *cl_field, } #define CLUSTER_ATTR(name, check_zero) \ -static ssize_t name##_write(struct cluster *cl, const char *buf, size_t len) \ +static ssize_t name##_write(struct dlm_cluster *cl, const char *buf, size_t len) \ { \ return cluster_set(cl, &cl->cl_##name, &dlm_config.ci_##name, \ check_zero, buf, len); \ } \ -static ssize_t name##_read(struct cluster *cl, char *buf) \ +static ssize_t name##_read(struct dlm_cluster *cl, char *buf) \ { \ return snprintf(buf, PAGE_SIZE, "%u\n", cl->cl_##name); \ } \ @@ -181,8 +189,8 @@ enum { struct comm_attribute { struct configfs_attribute attr; - ssize_t (*show)(struct comm *, char *); - ssize_t (*store)(struct comm *, const char *, size_t); + ssize_t (*show)(struct dlm_comm *, char *); + ssize_t (*store)(struct dlm_comm *, const char *, size_t); }; static struct comm_attribute comm_attr_nodeid = { @@ -222,8 +230,8 @@ enum { struct node_attribute { struct configfs_attribute attr; - ssize_t (*show)(struct node *, char *); - ssize_t (*store)(struct node *, const char *, size_t); + ssize_t (*show)(struct dlm_node *, char *); + ssize_t (*store)(struct dlm_node *, const char *, size_t); }; static struct node_attribute node_attr_nodeid = { @@ -248,26 +256,26 @@ static struct configfs_attribute *node_attrs[] = { NULL, }; -struct clusters { +struct dlm_clusters { struct configfs_subsystem subsys; }; -struct spaces { +struct dlm_spaces { struct config_group ss_group; }; -struct space { +struct dlm_space { struct config_group group; struct list_head members; struct mutex members_lock; int members_count; }; -struct comms { +struct dlm_comms { struct config_group cs_group; }; -struct comm { +struct dlm_comm { struct config_item item; int nodeid; int local; @@ -275,11 +283,11 @@ struct comm { struct sockaddr_storage *addr[DLM_MAX_ADDR_COUNT]; }; -struct nodes { +struct dlm_nodes { struct config_group ns_group; }; -struct node { +struct dlm_node { struct config_item item; struct list_head list; /* space->members */ int nodeid; @@ -372,38 +380,40 @@ static struct config_item_type node_type = { .ct_owner = THIS_MODULE, }; -static struct cluster *to_cluster(struct config_item *i) +static struct dlm_cluster *config_item_to_cluster(struct config_item *i) { - return i ? container_of(to_config_group(i), struct cluster, group):NULL; + return i ? container_of(to_config_group(i), struct dlm_cluster, group) : + NULL; } -static struct space *to_space(struct config_item *i) +static struct dlm_space *config_item_to_space(struct config_item *i) { - return i ? container_of(to_config_group(i), struct space, group) : NULL; + return i ? container_of(to_config_group(i), struct dlm_space, group) : + NULL; } -static struct comm *to_comm(struct config_item *i) +static struct dlm_comm *config_item_to_comm(struct config_item *i) { - return i ? container_of(i, struct comm, item) : NULL; + return i ? container_of(i, struct dlm_comm, item) : NULL; } -static struct node *to_node(struct config_item *i) +static struct dlm_node *config_item_to_node(struct config_item *i) { - return i ? container_of(i, struct node, item) : NULL; + return i ? container_of(i, struct dlm_node, item) : NULL; } static struct config_group *make_cluster(struct config_group *g, const char *name) { - struct cluster *cl = NULL; - struct spaces *sps = NULL; - struct comms *cms = NULL; + struct dlm_cluster *cl = NULL; + struct dlm_spaces *sps = NULL; + struct dlm_comms *cms = NULL; void *gps = NULL; - cl = kzalloc(sizeof(struct cluster), GFP_KERNEL); + cl = kzalloc(sizeof(struct dlm_cluster), GFP_KERNEL); gps = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL); - sps = kzalloc(sizeof(struct spaces), GFP_KERNEL); - cms = kzalloc(sizeof(struct comms), GFP_KERNEL); + sps = kzalloc(sizeof(struct dlm_spaces), GFP_KERNEL); + cms = kzalloc(sizeof(struct dlm_comms), GFP_KERNEL); if (!cl || !gps || !sps || !cms) goto fail; @@ -443,7 +453,7 @@ static struct config_group *make_cluster(struct config_group *g, static void drop_cluster(struct config_group *g, struct config_item *i) { - struct cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct config_item *tmp; int j; @@ -461,20 +471,20 @@ static void drop_cluster(struct config_group *g, struct config_item *i) static void release_cluster(struct config_item *i) { - struct cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); kfree(cl->group.default_groups); kfree(cl); } static struct config_group *make_space(struct config_group *g, const char *name) { - struct space *sp = NULL; - struct nodes *nds = NULL; + struct dlm_space *sp = NULL; + struct dlm_nodes *nds = NULL; void *gps = NULL; - sp = kzalloc(sizeof(struct space), GFP_KERNEL); + sp = kzalloc(sizeof(struct dlm_space), GFP_KERNEL); gps = kcalloc(2, sizeof(struct config_group *), GFP_KERNEL); - nds = kzalloc(sizeof(struct nodes), GFP_KERNEL); + nds = kzalloc(sizeof(struct dlm_nodes), GFP_KERNEL); if (!sp || !gps || !nds) goto fail; @@ -500,7 +510,7 @@ static struct config_group *make_space(struct config_group *g, const char *name) static void drop_space(struct config_group *g, struct config_item *i) { - struct space *sp = to_space(i); + struct dlm_space *sp = config_item_to_space(i); struct config_item *tmp; int j; @@ -517,16 +527,16 @@ static void drop_space(struct config_group *g, struct config_item *i) static void release_space(struct config_item *i) { - struct space *sp = to_space(i); + struct dlm_space *sp = config_item_to_space(i); kfree(sp->group.default_groups); kfree(sp); } static struct config_item *make_comm(struct config_group *g, const char *name) { - struct comm *cm; + struct dlm_comm *cm; - cm = kzalloc(sizeof(struct comm), GFP_KERNEL); + cm = kzalloc(sizeof(struct dlm_comm), GFP_KERNEL); if (!cm) return ERR_PTR(-ENOMEM); @@ -539,7 +549,7 @@ static struct config_item *make_comm(struct config_group *g, const char *name) static void drop_comm(struct config_group *g, struct config_item *i) { - struct comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); if (local_comm == cm) local_comm = NULL; dlm_lowcomms_close(cm->nodeid); @@ -550,16 +560,16 @@ static void drop_comm(struct config_group *g, struct config_item *i) static void release_comm(struct config_item *i) { - struct comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); kfree(cm); } static struct config_item *make_node(struct config_group *g, const char *name) { - struct space *sp = to_space(g->cg_item.ci_parent); - struct node *nd; + struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent); + struct dlm_node *nd; - nd = kzalloc(sizeof(struct node), GFP_KERNEL); + nd = kzalloc(sizeof(struct dlm_node), GFP_KERNEL); if (!nd) return ERR_PTR(-ENOMEM); @@ -578,8 +588,8 @@ static struct config_item *make_node(struct config_group *g, const char *name) static void drop_node(struct config_group *g, struct config_item *i) { - struct space *sp = to_space(g->cg_item.ci_parent); - struct node *nd = to_node(i); + struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent); + struct dlm_node *nd = config_item_to_node(i); mutex_lock(&sp->members_lock); list_del(&nd->list); @@ -591,11 +601,11 @@ static void drop_node(struct config_group *g, struct config_item *i) static void release_node(struct config_item *i) { - struct node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); kfree(nd); } -static struct clusters clusters_root = { +static struct dlm_clusters clusters_root = { .subsys = { .su_group = { .cg_item = { @@ -625,7 +635,7 @@ void dlm_config_exit(void) static ssize_t show_cluster(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct cluster_attribute *cla = container_of(a, struct cluster_attribute, attr); return cla->show ? cla->show(cl, buf) : 0; @@ -635,7 +645,7 @@ static ssize_t store_cluster(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct cluster *cl = to_cluster(i); + struct dlm_cluster *cl = config_item_to_cluster(i); struct cluster_attribute *cla = container_of(a, struct cluster_attribute, attr); return cla->store ? cla->store(cl, buf, len) : -EINVAL; @@ -644,7 +654,7 @@ static ssize_t store_cluster(struct config_item *i, static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); struct comm_attribute *cma = container_of(a, struct comm_attribute, attr); return cma->show ? cma->show(cm, buf) : 0; @@ -653,29 +663,31 @@ static ssize_t show_comm(struct config_item *i, struct configfs_attribute *a, static ssize_t store_comm(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct comm *cm = to_comm(i); + struct dlm_comm *cm = config_item_to_comm(i); struct comm_attribute *cma = container_of(a, struct comm_attribute, attr); return cma->store ? cma->store(cm, buf, len) : -EINVAL; } -static ssize_t comm_nodeid_read(struct comm *cm, char *buf) +static ssize_t comm_nodeid_read(struct dlm_comm *cm, char *buf) { return sprintf(buf, "%d\n", cm->nodeid); } -static ssize_t comm_nodeid_write(struct comm *cm, const char *buf, size_t len) +static ssize_t comm_nodeid_write(struct dlm_comm *cm, const char *buf, + size_t len) { cm->nodeid = simple_strtol(buf, NULL, 0); return len; } -static ssize_t comm_local_read(struct comm *cm, char *buf) +static ssize_t comm_local_read(struct dlm_comm *cm, char *buf) { return sprintf(buf, "%d\n", cm->local); } -static ssize_t comm_local_write(struct comm *cm, const char *buf, size_t len) +static ssize_t comm_local_write(struct dlm_comm *cm, const char *buf, + size_t len) { cm->local= simple_strtol(buf, NULL, 0); if (cm->local && !local_comm) @@ -683,7 +695,7 @@ static ssize_t comm_local_write(struct comm *cm, const char *buf, size_t len) return len; } -static ssize_t comm_addr_write(struct comm *cm, const char *buf, size_t len) +static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, size_t len) { struct sockaddr_storage *addr; @@ -705,7 +717,7 @@ static ssize_t comm_addr_write(struct comm *cm, const char *buf, size_t len) static ssize_t show_node(struct config_item *i, struct configfs_attribute *a, char *buf) { - struct node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); struct node_attribute *nda = container_of(a, struct node_attribute, attr); return nda->show ? nda->show(nd, buf) : 0; @@ -714,29 +726,31 @@ static ssize_t show_node(struct config_item *i, struct configfs_attribute *a, static ssize_t store_node(struct config_item *i, struct configfs_attribute *a, const char *buf, size_t len) { - struct node *nd = to_node(i); + struct dlm_node *nd = config_item_to_node(i); struct node_attribute *nda = container_of(a, struct node_attribute, attr); return nda->store ? nda->store(nd, buf, len) : -EINVAL; } -static ssize_t node_nodeid_read(struct node *nd, char *buf) +static ssize_t node_nodeid_read(struct dlm_node *nd, char *buf) { return sprintf(buf, "%d\n", nd->nodeid); } -static ssize_t node_nodeid_write(struct node *nd, const char *buf, size_t len) +static ssize_t node_nodeid_write(struct dlm_node *nd, const char *buf, + size_t len) { nd->nodeid = simple_strtol(buf, NULL, 0); return len; } -static ssize_t node_weight_read(struct node *nd, char *buf) +static ssize_t node_weight_read(struct dlm_node *nd, char *buf) { return sprintf(buf, "%d\n", nd->weight); } -static ssize_t node_weight_write(struct node *nd, const char *buf, size_t len) +static ssize_t node_weight_write(struct dlm_node *nd, const char *buf, + size_t len) { nd->weight = simple_strtol(buf, NULL, 0); return len; @@ -746,7 +760,7 @@ static ssize_t node_weight_write(struct node *nd, const char *buf, size_t len) * Functions for the dlm to get the info that's been configured */ -static struct space *get_space(char *name) +static struct dlm_space *get_space(char *name) { struct config_item *i; @@ -757,18 +771,45 @@ static struct space *get_space(char *name) i = config_group_find_item(space_list, name); mutex_unlock(&space_list->cg_subsys->su_mutex); - return to_space(i); + return config_item_to_space(i); } -static void put_space(struct space *sp) +static void put_space(struct dlm_space *sp) { config_item_put(&sp->group.cg_item); } -static struct comm *get_comm(int nodeid, struct sockaddr_storage *addr) +static int addr_compare(struct sockaddr_storage *x, struct sockaddr_storage *y) +{ + switch (x->ss_family) { + case AF_INET: { + struct sockaddr_in *sinx = (struct sockaddr_in *)x; + struct sockaddr_in *siny = (struct sockaddr_in *)y; + if (sinx->sin_addr.s_addr != siny->sin_addr.s_addr) + return 0; + if (sinx->sin_port != siny->sin_port) + return 0; + break; + } + case AF_INET6: { + struct sockaddr_in6 *sinx = (struct sockaddr_in6 *)x; + struct sockaddr_in6 *siny = (struct sockaddr_in6 *)y; + if (!ipv6_addr_equal(&sinx->sin6_addr, &siny->sin6_addr)) + return 0; + if (sinx->sin6_port != siny->sin6_port) + return 0; + break; + } + default: + return 0; + } + return 1; +} + +static struct dlm_comm *get_comm(int nodeid, struct sockaddr_storage *addr) { struct config_item *i; - struct comm *cm = NULL; + struct dlm_comm *cm = NULL; int found = 0; if (!comm_list) @@ -777,7 +818,7 @@ static struct comm *get_comm(int nodeid, struct sockaddr_storage *addr) mutex_lock(&clusters_root.subsys.su_mutex); list_for_each_entry(i, &comm_list->cg_children, ci_entry) { - cm = to_comm(i); + cm = config_item_to_comm(i); if (nodeid) { if (cm->nodeid != nodeid) @@ -786,8 +827,7 @@ static struct comm *get_comm(int nodeid, struct sockaddr_storage *addr) config_item_get(i); break; } else { - if (!cm->addr_count || - memcmp(cm->addr[0], addr, sizeof(*addr))) + if (!cm->addr_count || !addr_compare(cm->addr[0], addr)) continue; found = 1; config_item_get(i); @@ -801,7 +841,7 @@ static struct comm *get_comm(int nodeid, struct sockaddr_storage *addr) return cm; } -static void put_comm(struct comm *cm) +static void put_comm(struct dlm_comm *cm) { config_item_put(&cm->item); } @@ -810,8 +850,8 @@ static void put_comm(struct comm *cm) int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out, int **new_out, int *new_count_out) { - struct space *sp; - struct node *nd; + struct dlm_space *sp; + struct dlm_node *nd; int i = 0, rv = 0, ids_count = 0, new_count = 0; int *ids, *new; @@ -874,8 +914,8 @@ int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out, int dlm_node_weight(char *lsname, int nodeid) { - struct space *sp; - struct node *nd; + struct dlm_space *sp; + struct dlm_node *nd; int w = -EEXIST; sp = get_space(lsname); @@ -897,7 +937,7 @@ int dlm_node_weight(char *lsname, int nodeid) int dlm_nodeid_to_addr(int nodeid, struct sockaddr_storage *addr) { - struct comm *cm = get_comm(nodeid, NULL); + struct dlm_comm *cm = get_comm(nodeid, NULL); if (!cm) return -EEXIST; if (!cm->addr_count) @@ -909,7 +949,7 @@ int dlm_nodeid_to_addr(int nodeid, struct sockaddr_storage *addr) int dlm_addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid) { - struct comm *cm = get_comm(0, addr); + struct dlm_comm *cm = get_comm(0, addr); if (!cm) return -EEXIST; *nodeid = cm->nodeid; diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 5a7ac33..868e4c9 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 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 @@ -441,8 +441,11 @@ struct dlm_ls { uint32_t ls_global_id; /* global unique lockspace ID */ uint32_t ls_exflags; int ls_lvblen; - int ls_count; /* reference count */ + int ls_count; /* refcount of processes in + the dlm using this ls */ + int ls_create_count; /* create/release refcount */ unsigned long ls_flags; /* LSFL_ */ + unsigned long ls_scan_time; struct kobject ls_kobj; struct dlm_rsbtable *ls_rsbtbl; diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 2d3d102..724ddac 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -363,6 +363,7 @@ static int search_rsb_list(struct list_head *head, char *name, int len, if (len == r->res_length && !memcmp(name, r->res_name, len)) goto found; } + *r_ret = NULL; return -EBADR; found: @@ -1782,7 +1783,8 @@ static void grant_pending_locks(struct dlm_rsb *r) list_for_each_entry_safe(lkb, s, &r->res_grantqueue, lkb_statequeue) { if (lkb->lkb_bastfn && lock_requires_bast(lkb, high, cw)) { - if (cw && high == DLM_LOCK_PR) + if (cw && high == DLM_LOCK_PR && + lkb->lkb_grmode == DLM_LOCK_PR) queue_bast(r, lkb, DLM_LOCK_CW); else queue_bast(r, lkb, high); diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 499e167..d910501 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2008 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 @@ -23,6 +23,7 @@ #include "lock.h" #include "recover.h" #include "requestqueue.h" +#include "user.h" static int ls_count; static struct mutex ls_lock; @@ -211,19 +212,41 @@ void dlm_lockspace_exit(void) kset_unregister(dlm_kset); } +static struct dlm_ls *find_ls_to_scan(void) +{ + struct dlm_ls *ls; + + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + if (time_after_eq(jiffies, ls->ls_scan_time + + dlm_config.ci_scan_secs * HZ)) { + spin_unlock(&lslist_lock); + return ls; + } + } + spin_unlock(&lslist_lock); + return NULL; +} + static int dlm_scand(void *data) { struct dlm_ls *ls; + int timeout_jiffies = dlm_config.ci_scan_secs * HZ; while (!kthread_should_stop()) { - list_for_each_entry(ls, &lslist, ls_list) { + ls = find_ls_to_scan(); + if (ls) { if (dlm_lock_recovery_try(ls)) { + ls->ls_scan_time = jiffies; dlm_scan_rsbs(ls); dlm_scan_timeout(ls); dlm_unlock_recovery(ls); + } else { + ls->ls_scan_time += HZ; } + } else { + schedule_timeout_interruptible(timeout_jiffies); } - schedule_timeout_interruptible(dlm_config.ci_scan_secs * HZ); } return 0; } @@ -246,23 +269,6 @@ static void dlm_scand_stop(void) kthread_stop(scand_task); } -static struct dlm_ls *dlm_find_lockspace_name(char *name, int namelen) -{ - struct dlm_ls *ls; - - spin_lock(&lslist_lock); - - list_for_each_entry(ls, &lslist, ls_list) { - if (ls->ls_namelen == namelen && - memcmp(ls->ls_name, name, namelen) == 0) - goto out; - } - ls = NULL; - out: - spin_unlock(&lslist_lock); - return ls; -} - struct dlm_ls *dlm_find_lockspace_global(uint32_t id) { struct dlm_ls *ls; @@ -327,6 +333,7 @@ static void remove_lockspace(struct dlm_ls *ls) for (;;) { spin_lock(&lslist_lock); if (ls->ls_count == 0) { + WARN_ON(ls->ls_create_count != 0); list_del(&ls->ls_list); spin_unlock(&lslist_lock); return; @@ -381,7 +388,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, uint32_t flags, int lvblen) { struct dlm_ls *ls; - int i, size, error = -ENOMEM; + int i, size, error; int do_unreg = 0; if (namelen > DLM_LOCKSPACE_LEN) @@ -393,12 +400,37 @@ static int new_lockspace(char *name, int namelen, void **lockspace, if (!try_module_get(THIS_MODULE)) return -EINVAL; - ls = dlm_find_lockspace_name(name, namelen); - if (ls) { - *lockspace = ls; + if (!dlm_user_daemon_available()) { + module_put(THIS_MODULE); + return -EUNATCH; + } + + error = 0; + + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + WARN_ON(ls->ls_create_count <= 0); + if (ls->ls_namelen != namelen) + continue; + if (memcmp(ls->ls_name, name, namelen)) + continue; + if (flags & DLM_LSFL_NEWEXCL) { + error = -EEXIST; + break; + } + ls->ls_create_count++; module_put(THIS_MODULE); - return -EEXIST; + error = 1; /* not an error, return 0 */ + break; } + spin_unlock(&lslist_lock); + + if (error < 0) + goto out; + if (error) + goto ret_zero; + + error = -ENOMEM; ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_KERNEL); if (!ls) @@ -408,6 +440,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, ls->ls_lvblen = lvblen; ls->ls_count = 0; ls->ls_flags = 0; + ls->ls_scan_time = jiffies; if (flags & DLM_LSFL_TIMEWARN) set_bit(LSFL_TIMEWARN, &ls->ls_flags); @@ -418,8 +451,9 @@ static int new_lockspace(char *name, int namelen, void **lockspace, ls->ls_allocation = GFP_KERNEL; /* ls_exflags are forced to match among nodes, and we don't - need to require all nodes to have TIMEWARN or FS set */ - ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS)); + need to require all nodes to have some flags set */ + ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS | + DLM_LSFL_NEWEXCL)); size = dlm_config.ci_rsbtbl_size; ls->ls_rsbtbl_size = size; @@ -510,6 +544,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, down_write(&ls->ls_in_recovery); spin_lock(&lslist_lock); + ls->ls_create_count = 1; list_add(&ls->ls_list, &lslist); spin_unlock(&lslist_lock); @@ -548,7 +583,7 @@ static int new_lockspace(char *name, int namelen, void **lockspace, dlm_create_debug_file(ls); log_debug(ls, "join complete"); - + ret_zero: *lockspace = ls; return 0; @@ -635,13 +670,34 @@ static int release_lockspace(struct dlm_ls *ls, int force) struct dlm_lkb *lkb; struct dlm_rsb *rsb; struct list_head *head; - int i; - int busy = lockspace_busy(ls); + int i, busy, rv; + + busy = lockspace_busy(ls); + + spin_lock(&lslist_lock); + if (ls->ls_create_count == 1) { + if (busy > force) + rv = -EBUSY; + else { + /* remove_lockspace takes ls off lslist */ + ls->ls_create_count = 0; + rv = 0; + } + } else if (ls->ls_create_count > 1) { + rv = --ls->ls_create_count; + } else { + rv = -EINVAL; + } + spin_unlock(&lslist_lock); - if (busy > force) - return -EBUSY; + if (rv) { + log_debug(ls, "release_lockspace no remove %d", rv); + return rv; + } + + dlm_device_deregister(ls); - if (force < 3) + if (force < 3 && dlm_user_daemon_available()) do_uevent(ls, 0); dlm_recoverd_stop(ls); @@ -720,15 +776,10 @@ static int release_lockspace(struct dlm_ls *ls, int force) dlm_clear_members(ls); dlm_clear_members_gone(ls); kfree(ls->ls_node_array); + log_debug(ls, "release_lockspace final free"); kobject_put(&ls->ls_kobj); /* The ls structure will be freed when the kobject is done with */ - mutex_lock(&ls_lock); - ls_count--; - if (!ls_count) - threads_stop(); - mutex_unlock(&ls_lock); - module_put(THIS_MODULE); return 0; } @@ -750,11 +801,38 @@ static int release_lockspace(struct dlm_ls *ls, int force) int dlm_release_lockspace(void *lockspace, int force) { struct dlm_ls *ls; + int error; ls = dlm_find_lockspace_local(lockspace); if (!ls) return -EINVAL; dlm_put_lockspace(ls); - return release_lockspace(ls, force); + + mutex_lock(&ls_lock); + error = release_lockspace(ls, force); + if (!error) + ls_count--; + else if (!ls_count) + threads_stop(); + mutex_unlock(&ls_lock); + + return error; +} + +void dlm_stop_lockspaces(void) +{ + struct dlm_ls *ls; + + restart: + spin_lock(&lslist_lock); + list_for_each_entry(ls, &lslist, ls_list) { + if (!test_bit(LSFL_RUNNING, &ls->ls_flags)) + continue; + spin_unlock(&lslist_lock); + log_error(ls, "no userland control daemon, stopping lockspace"); + dlm_ls_stop(ls); + goto restart; + } + spin_unlock(&lslist_lock); } diff --git a/fs/dlm/lockspace.h b/fs/dlm/lockspace.h index 891eabb..f879f87 100644 --- a/fs/dlm/lockspace.h +++ b/fs/dlm/lockspace.h @@ -20,6 +20,7 @@ struct dlm_ls *dlm_find_lockspace_global(uint32_t id); struct dlm_ls *dlm_find_lockspace_local(void *id); struct dlm_ls *dlm_find_lockspace_device(int minor); void dlm_put_lockspace(struct dlm_ls *ls); +void dlm_stop_lockspaces(void); #endif /* __LOCKSPACE_DOT_H__ */ diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 637018c..3962262 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -891,8 +891,10 @@ static void tcp_connect_to_sock(struct connection *con) goto out_err; memset(&saddr, 0, sizeof(saddr)); - if (dlm_nodeid_to_addr(con->nodeid, &saddr)) + if (dlm_nodeid_to_addr(con->nodeid, &saddr)) { + sock_release(sock); goto out_err; + } sock->sk->sk_user_data = con; con->rx_action = receive_from_sock; diff --git a/fs/dlm/user.c b/fs/dlm/user.c index f976f30..b3832c6 100644 --- a/fs/dlm/user.c +++ b/fs/dlm/user.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2006-2007 Red Hat, Inc. All rights reserved. + * Copyright (C) 2006-2008 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 @@ -15,7 +15,6 @@ #include <linux/poll.h> #include <linux/signal.h> #include <linux/spinlock.h> -#include <linux/smp_lock.h> #include <linux/dlm.h> #include <linux/dlm_device.h> @@ -27,6 +26,8 @@ static const char name_prefix[] = "dlm"; static const struct file_operations device_fops; +static atomic_t dlm_monitor_opened; +static int dlm_monitor_unused = 1; #ifdef CONFIG_COMPAT @@ -340,10 +341,15 @@ static int device_user_deadlock(struct dlm_user_proc *proc, return error; } -static int create_misc_device(struct dlm_ls *ls, char *name) +static int dlm_device_register(struct dlm_ls *ls, char *name) { int error, len; + /* The device is already registered. This happens when the + lockspace is created multiple times from userspace. */ + if (ls->ls_device.name) + return 0; + error = -ENOMEM; len = strlen(name) + strlen(name_prefix) + 2; ls->ls_device.name = kzalloc(len, GFP_KERNEL); @@ -363,6 +369,22 @@ fail: return error; } +int dlm_device_deregister(struct dlm_ls *ls) +{ + int error; + + /* The device is not registered. This happens when the lockspace + was never used from userspace, or when device_create_lockspace() + calls dlm_release_lockspace() after the register fails. */ + if (!ls->ls_device.name) + return 0; + + error = misc_deregister(&ls->ls_device); + if (!error) + kfree(ls->ls_device.name); + return error; +} + static int device_user_purge(struct dlm_user_proc *proc, struct dlm_purge_params *params) { @@ -397,7 +419,7 @@ static int device_create_lockspace(struct dlm_lspace_params *params) if (!ls) return -ENOENT; - error = create_misc_device(ls, params->name); + error = dlm_device_register(ls, params->name); dlm_put_lockspace(ls); if (error) @@ -421,31 +443,22 @@ static int device_remove_lockspace(struct dlm_lspace_params *params) if (!ls) return -ENOENT; - /* Deregister the misc device first, so we don't have - * a device that's not attached to a lockspace. If - * dlm_release_lockspace fails then we can recreate it - */ - error = misc_deregister(&ls->ls_device); - if (error) { - dlm_put_lockspace(ls); - goto out; - } - kfree(ls->ls_device.name); - if (params->flags & DLM_USER_LSFLG_FORCEFREE) force = 2; lockspace = ls->ls_local_handle; + dlm_put_lockspace(ls); - /* dlm_release_lockspace waits for references to go to zero, - so all processes will need to close their device for the ls - before the release will procede */ + /* The final dlm_release_lockspace waits for references to go to + zero, so all processes will need to close their device for the + ls before the release will proceed. release also calls the + device_deregister above. Converting a positive return value + from release to zero means that userspace won't know when its + release was the final one, but it shouldn't need to know. */ - dlm_put_lockspace(ls); error = dlm_release_lockspace(lockspace, force); - if (error) - create_misc_device(ls, ls->ls_name); - out: + if (error > 0) + error = 0; return error; } @@ -527,8 +540,10 @@ static ssize_t device_write(struct file *file, const char __user *buf, k32buf = (struct dlm_write_request32 *)kbuf; kbuf = kmalloc(count + 1 + (sizeof(struct dlm_write_request) - sizeof(struct dlm_write_request32)), GFP_KERNEL); - if (!kbuf) + if (!kbuf) { + kfree(k32buf); return -ENOMEM; + } if (proc) set_bit(DLM_PROC_FLAGS_COMPAT, &proc->flags); @@ -539,8 +554,10 @@ static ssize_t device_write(struct file *file, const char __user *buf, /* do we really need this? can a write happen after a close? */ if ((kbuf->cmd == DLM_USER_LOCK || kbuf->cmd == DLM_USER_UNLOCK) && - test_bit(DLM_PROC_FLAGS_CLOSING, &proc->flags)) - return -EINVAL; + (proc && test_bit(DLM_PROC_FLAGS_CLOSING, &proc->flags))) { + error = -EINVAL; + goto out_free; + } sigfillset(&allsigs); sigprocmask(SIG_BLOCK, &allsigs, &tmpsig); @@ -619,17 +636,13 @@ static int device_open(struct inode *inode, struct file *file) struct dlm_user_proc *proc; struct dlm_ls *ls; - lock_kernel(); ls = dlm_find_lockspace_device(iminor(inode)); - if (!ls) { - unlock_kernel(); + if (!ls) return -ENOENT; - } proc = kzalloc(sizeof(struct dlm_user_proc), GFP_KERNEL); if (!proc) { dlm_put_lockspace(ls); - unlock_kernel(); return -ENOMEM; } @@ -641,7 +654,6 @@ static int device_open(struct inode *inode, struct file *file) spin_lock_init(&proc->locks_spin); init_waitqueue_head(&proc->wait); file->private_data = proc; - unlock_kernel(); return 0; } @@ -874,9 +886,28 @@ static unsigned int device_poll(struct file *file, poll_table *wait) return 0; } +int dlm_user_daemon_available(void) +{ + /* dlm_controld hasn't started (or, has started, but not + properly populated configfs) */ + + if (!dlm_our_nodeid()) + return 0; + + /* This is to deal with versions of dlm_controld that don't + know about the monitor device. We assume that if the + dlm_controld was started (above), but the monitor device + was never opened, that it's an old version. dlm_controld + should open the monitor device before populating configfs. */ + + if (dlm_monitor_unused) + return 1; + + return atomic_read(&dlm_monitor_opened) ? 1 : 0; +} + static int ctl_device_open(struct inode *inode, struct file *file) { - cycle_kernel_lock(); file->private_data = NULL; return 0; } @@ -886,6 +917,20 @@ static int ctl_device_close(struct inode *inode, struct file *file) return 0; } +static int monitor_device_open(struct inode *inode, struct file *file) +{ + atomic_inc(&dlm_monitor_opened); + dlm_monitor_unused = 0; + return 0; +} + +static int monitor_device_close(struct inode *inode, struct file *file) +{ + if (atomic_dec_and_test(&dlm_monitor_opened)) + dlm_stop_lockspaces(); + return 0; +} + static const struct file_operations device_fops = { .open = device_open, .release = device_close, @@ -909,19 +954,42 @@ static struct miscdevice ctl_device = { .minor = MISC_DYNAMIC_MINOR, }; +static const struct file_operations monitor_device_fops = { + .open = monitor_device_open, + .release = monitor_device_close, + .owner = THIS_MODULE, +}; + +static struct miscdevice monitor_device = { + .name = "dlm-monitor", + .fops = &monitor_device_fops, + .minor = MISC_DYNAMIC_MINOR, +}; + int __init dlm_user_init(void) { int error; + atomic_set(&dlm_monitor_opened, 0); + error = misc_register(&ctl_device); - if (error) + if (error) { log_print("misc_register failed for control device"); + goto out; + } + error = misc_register(&monitor_device); + if (error) { + log_print("misc_register failed for monitor device"); + misc_deregister(&ctl_device); + } + out: return error; } void dlm_user_exit(void) { misc_deregister(&ctl_device); + misc_deregister(&monitor_device); } diff --git a/fs/dlm/user.h b/fs/dlm/user.h index d38e9f3e..35eb6a1 100644 --- a/fs/dlm/user.h +++ b/fs/dlm/user.h @@ -1,5 +1,5 @@ /* - * Copyright (C) 2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2006-2008 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 @@ -12,5 +12,7 @@ void dlm_user_add_ast(struct dlm_lkb *lkb, int type); int dlm_user_init(void); void dlm_user_exit(void); +int dlm_device_deregister(struct dlm_ls *ls); +int dlm_user_daemon_available(void); #endif @@ -1793,6 +1793,21 @@ static int vfs_quota_on_remount(struct super_block *sb, int type) return ret; } +int vfs_quota_on_path(struct super_block *sb, int type, int format_id, + struct path *path) +{ + int error = security_quota_on(path->dentry); + if (error) + return error; + /* Quota file not on the same filesystem? */ + if (path->mnt->mnt_sb != sb) + error = -EXDEV; + else + error = vfs_quota_on_inode(path->dentry->d_inode, type, + format_id); + return error; +} + /* Actual function called from quotactl() */ int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path, int remount) @@ -1804,19 +1819,10 @@ int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path, return vfs_quota_on_remount(sb, type); error = path_lookup(path, LOOKUP_FOLLOW, &nd); - if (error < 0) - return error; - error = security_quota_on(nd.path.dentry); - if (error) - goto out_path; - /* Quota file not on the same filesystem? */ - if (nd.path.mnt->mnt_sb != sb) - error = -EXDEV; - else - error = vfs_quota_on_inode(nd.path.dentry->d_inode, type, - format_id); -out_path: - path_put(&nd.path); + if (!error) { + error = vfs_quota_on_path(sb, type, format_id, &nd.path); + path_put(&nd.path); + } return error; } @@ -2185,6 +2191,7 @@ EXPORT_SYMBOL(unregister_quota_format); EXPORT_SYMBOL(dqstats); EXPORT_SYMBOL(dq_data_lock); EXPORT_SYMBOL(vfs_quota_on); +EXPORT_SYMBOL(vfs_quota_on_path); EXPORT_SYMBOL(vfs_quota_on_mount); EXPORT_SYMBOL(vfs_quota_off); EXPORT_SYMBOL(vfs_quota_sync); diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 7b99917f..06db79d 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -475,8 +475,8 @@ int ecryptfs_encrypt_page(struct page *page) { struct inode *ecryptfs_inode; struct ecryptfs_crypt_stat *crypt_stat; - char *enc_extent_virt = NULL; - struct page *enc_extent_page; + char *enc_extent_virt; + struct page *enc_extent_page = NULL; loff_t extent_offset; int rc = 0; @@ -492,14 +492,14 @@ int ecryptfs_encrypt_page(struct page *page) page->index); goto out; } - enc_extent_virt = kmalloc(PAGE_CACHE_SIZE, GFP_USER); - if (!enc_extent_virt) { + enc_extent_page = alloc_page(GFP_USER); + if (!enc_extent_page) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Error allocating memory for " "encrypted extent\n"); goto out; } - enc_extent_page = virt_to_page(enc_extent_virt); + enc_extent_virt = kmap(enc_extent_page); for (extent_offset = 0; extent_offset < (PAGE_CACHE_SIZE / crypt_stat->extent_size); extent_offset++) { @@ -527,7 +527,10 @@ int ecryptfs_encrypt_page(struct page *page) } } out: - kfree(enc_extent_virt); + if (enc_extent_page) { + kunmap(enc_extent_page); + __free_page(enc_extent_page); + } return rc; } @@ -609,8 +612,8 @@ int ecryptfs_decrypt_page(struct page *page) { struct inode *ecryptfs_inode; struct ecryptfs_crypt_stat *crypt_stat; - char *enc_extent_virt = NULL; - struct page *enc_extent_page; + char *enc_extent_virt; + struct page *enc_extent_page = NULL; unsigned long extent_offset; int rc = 0; @@ -627,14 +630,14 @@ int ecryptfs_decrypt_page(struct page *page) page->index); goto out; } - enc_extent_virt = kmalloc(PAGE_CACHE_SIZE, GFP_USER); - if (!enc_extent_virt) { + enc_extent_page = alloc_page(GFP_USER); + if (!enc_extent_page) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Error allocating memory for " "encrypted extent\n"); goto out; } - enc_extent_page = virt_to_page(enc_extent_virt); + enc_extent_virt = kmap(enc_extent_page); for (extent_offset = 0; extent_offset < (PAGE_CACHE_SIZE / crypt_stat->extent_size); extent_offset++) { @@ -662,7 +665,10 @@ int ecryptfs_decrypt_page(struct page *page) } } out: - kfree(enc_extent_virt); + if (enc_extent_page) { + kunmap(enc_extent_page); + __free_page(enc_extent_page); + } return rc; } diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index d755455..89209f0 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -465,7 +465,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, int rc; struct dentry *lower_dentry; struct dentry *lower_dir_dentry; - umode_t mode; char *encoded_symname; int encoded_symlen; struct ecryptfs_crypt_stat *crypt_stat = NULL; @@ -473,7 +472,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, lower_dentry = ecryptfs_dentry_to_lower(dentry); dget(lower_dentry); lower_dir_dentry = lock_parent(lower_dentry); - mode = S_IALLUGO; encoded_symlen = ecryptfs_encode_filename(crypt_stat, symname, strlen(symname), &encoded_symname); @@ -482,7 +480,7 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, goto out_lock; } rc = vfs_symlink(lower_dir_dentry->d_inode, lower_dentry, - encoded_symname, mode); + encoded_symname); kfree(encoded_symname); if (rc || !lower_dentry->d_inode) goto out_lock; @@ -830,22 +828,9 @@ out: } static int -ecryptfs_permission(struct inode *inode, int mask, struct nameidata *nd) +ecryptfs_permission(struct inode *inode, int mask) { - int rc; - - if (nd) { - struct vfsmount *vfsmnt_save = nd->path.mnt; - struct dentry *dentry_save = nd->path.dentry; - - nd->path.mnt = ecryptfs_dentry_to_lower_mnt(nd->path.dentry); - nd->path.dentry = ecryptfs_dentry_to_lower(nd->path.dentry); - rc = permission(ecryptfs_inode_to_lower(inode), mask, nd); - nd->path.mnt = vfsmnt_save; - nd->path.dentry = dentry_save; - } else - rc = permission(ecryptfs_inode_to_lower(inode), mask, NULL); - return rc; + return inode_permission(ecryptfs_inode_to_lower(inode), mask); } /** diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 6f403cf..448dfd5 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -578,7 +578,7 @@ static struct file_system_type ecryptfs_fs_type = { * Initializes the ecryptfs_inode_info_cache when it is created */ static void -inode_info_init_once(struct kmem_cache *cachep, void *vptr) +inode_info_init_once(void *vptr) { struct ecryptfs_inode_info *ei = (struct ecryptfs_inode_info *)vptr; @@ -589,7 +589,7 @@ static struct ecryptfs_cache_info { struct kmem_cache **cache; const char *name; size_t size; - void (*ctor)(struct kmem_cache *cache, void *obj); + void (*ctor)(void *obj); } ecryptfs_cache_infos[] = { { .cache = &ecryptfs_auth_tok_list_item_cache, diff --git a/fs/efs/namei.c b/fs/efs/namei.c index 3a404e7..291abb1 100644 --- a/fs/efs/namei.c +++ b/fs/efs/namei.c @@ -74,8 +74,7 @@ struct dentry *efs_lookup(struct inode *dir, struct dentry *dentry, struct namei } unlock_kernel(); - d_add(dentry, inode); - return NULL; + return d_splice_alias(inode, dentry); } static struct inode *efs_nfs_get_inode(struct super_block *sb, u64 ino, diff --git a/fs/efs/super.c b/fs/efs/super.c index d733531..567b134 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -70,7 +70,7 @@ static void efs_destroy_inode(struct inode *inode) kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 0c87474..7cc0eb7 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1041,10 +1041,7 @@ retry: } /* - * It opens an eventpoll file descriptor. The "size" parameter is there - * for historical reasons, when epoll was using an hash instead of an - * RB tree. With the current implementation, the "size" parameter is ignored - * (besides sanity checks). + * Open an eventpoll file descriptor. */ asmlinkage long sys_epoll_create1(int flags) { @@ -32,6 +32,7 @@ #include <linux/swap.h> #include <linux/string.h> #include <linux/init.h> +#include <linux/pagemap.h> #include <linux/highmem.h> #include <linux/spinlock.h> #include <linux/key.h> @@ -42,13 +43,13 @@ #include <linux/module.h> #include <linux/namei.h> #include <linux/proc_fs.h> -#include <linux/ptrace.h> #include <linux/mount.h> #include <linux/security.h> #include <linux/syscalls.h> #include <linux/tsacct_kern.h> #include <linux/cn_proc.h> #include <linux/audit.h> +#include <linux/tracehook.h> #include <asm/uaccess.h> #include <asm/mmu_context.h> @@ -106,11 +107,17 @@ static inline void put_binfmt(struct linux_binfmt * fmt) */ asmlinkage long sys_uselib(const char __user * library) { - struct file * file; + struct file *file; struct nameidata nd; - int error; - - error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); + char *tmp = getname(library); + int error = PTR_ERR(tmp); + + if (!IS_ERR(tmp)) { + error = path_lookup_open(AT_FDCWD, tmp, + LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + putname(tmp); + } if (error) goto out; @@ -118,7 +125,11 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) goto exit; - error = vfs_permission(&nd, MAY_READ | MAY_EXEC); + error = -EACCES; + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto exit; + + error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN); if (error) goto exit; @@ -656,38 +667,43 @@ EXPORT_SYMBOL(setup_arg_pages); struct file *open_exec(const char *name) { struct nameidata nd; - int err; struct file *file; + int err; - err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC); - file = ERR_PTR(err); - - if (!err) { - struct inode *inode = nd.path.dentry->d_inode; - file = ERR_PTR(-EACCES); - if (S_ISREG(inode->i_mode)) { - int err = vfs_permission(&nd, MAY_EXEC); - file = ERR_PTR(err); - if (!err) { - file = nameidata_to_filp(&nd, - O_RDONLY|O_LARGEFILE); - if (!IS_ERR(file)) { - err = deny_write_access(file); - if (err) { - fput(file); - file = ERR_PTR(err); - } - } -out: - return file; - } - } - release_open_intent(&nd); - path_put(&nd.path); + err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, + FMODE_READ|FMODE_EXEC); + if (err) + goto out; + + err = -EACCES; + if (!S_ISREG(nd.path.dentry->d_inode->i_mode)) + goto out_path_put; + + if (nd.path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_put; + + err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN); + if (err) + goto out_path_put; + + file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE); + if (IS_ERR(file)) + return file; + + err = deny_write_access(file); + if (err) { + fput(file); + goto out; } - goto out; -} + return file; + + out_path_put: + release_open_intent(&nd); + path_put(&nd.path); + out: + return ERR_PTR(err); +} EXPORT_SYMBOL(open_exec); int kernel_read(struct file *file, unsigned long offset, @@ -736,11 +752,11 @@ static int exec_mmap(struct mm_struct *mm) tsk->active_mm = mm; activate_mm(active_mm, mm); task_unlock(tsk); - mm_update_next_owner(old_mm); arch_pick_mmap_layout(mm); if (old_mm) { up_read(&old_mm->mmap_sem); BUG_ON(active_mm != old_mm); + mm_update_next_owner(old_mm); mmput(old_mm); return 0; } @@ -1071,13 +1087,8 @@ EXPORT_SYMBOL(prepare_binprm); static int unsafe_exec(struct task_struct *p) { - int unsafe = 0; - if (p->ptrace & PT_PTRACED) { - if (p->ptrace & PT_PTRACE_CAP) - unsafe |= LSM_UNSAFE_PTRACE_CAP; - else - unsafe |= LSM_UNSAFE_PTRACE; - } + int unsafe = tracehook_unsafe_exec(p); + if (atomic_read(&p->fs->count) > 1 || atomic_read(&p->files->count) > 1 || atomic_read(&p->sighand->count) > 1) @@ -1214,6 +1225,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) read_unlock(&binfmt_lock); retval = fn(bprm, regs); if (retval >= 0) { + tracehook_report_exec(fmt, bprm, regs); put_binfmt(fmt); allow_write_access(bprm->file); if (bprm->file) diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index e58669e..ae8c4f8 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -294,7 +294,7 @@ ext2_check_acl(struct inode *inode, int mask) } int -ext2_permission(struct inode *inode, int mask, struct nameidata *nd) +ext2_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext2_check_acl); } diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h index 0bde85b..b42cf57 100644 --- a/fs/ext2/acl.h +++ b/fs/ext2/acl.h @@ -58,7 +58,7 @@ static inline int ext2_acl_count(size_t size) #define EXT2_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext2_permission (struct inode *, int, struct nameidata *); +extern int ext2_permission (struct inode *, int); extern int ext2_acl_chmod (struct inode *); extern int ext2_init_acl (struct inode *, struct inode *); diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 47d88da..bae998c 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -133,6 +133,8 @@ extern void ext2_truncate (struct inode *); extern int ext2_setattr (struct dentry *, struct iattr *); extern void ext2_set_inode_flags(struct inode *inode); extern void ext2_get_inode_flags(struct ext2_inode_info *); +extern int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len); int __ext2_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata); diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 5f2fa9c..45ed071 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -86,4 +86,5 @@ const struct inode_operations ext2_file_inode_operations = { #endif .setattr = ext2_setattr, .permission = ext2_permission, + .fiemap = ext2_fiemap, }; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 384fc0d..7658b33 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -31,6 +31,7 @@ #include <linux/writeback.h> #include <linux/buffer_head.h> #include <linux/mpage.h> +#include <linux/fiemap.h> #include "ext2.h" #include "acl.h" #include "xip.h" @@ -704,6 +705,13 @@ int ext2_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_ } +int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + return generic_block_fiemap(inode, fieinfo, start, len, + ext2_get_block); +} + static int ext2_writepage(struct page *page, struct writeback_control *wbc) { return block_write_full_page(page, ext2_get_block, wbc); @@ -791,6 +799,7 @@ const struct address_space_operations ext2_aops = { .direct_IO = ext2_direct_IO, .writepages = ext2_writepages, .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; const struct address_space_operations ext2_aops_xip = { diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 31308a3..fd88c7b4 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -159,7 +159,7 @@ static void ext2_destroy_inode(struct inode *inode) kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index a754d18..b60bb24 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -299,7 +299,7 @@ ext3_check_acl(struct inode *inode, int mask) } int -ext3_permission(struct inode *inode, int mask, struct nameidata *nd) +ext3_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext3_check_acl); } diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h index 0d1e627..42da16b 100644 --- a/fs/ext3/acl.h +++ b/fs/ext3/acl.h @@ -58,7 +58,7 @@ static inline int ext3_acl_count(size_t size) #define EXT3_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext3_permission (struct inode *, int, struct nameidata *); +extern int ext3_permission (struct inode *, int); extern int ext3_acl_chmod (struct inode *); extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); diff --git a/fs/ext3/file.c b/fs/ext3/file.c index acc4913..3be1e06 100644 --- a/fs/ext3/file.c +++ b/fs/ext3/file.c @@ -134,5 +134,6 @@ const struct inode_operations ext3_file_inode_operations = { .removexattr = generic_removexattr, #endif .permission = ext3_permission, + .fiemap = ext3_fiemap, }; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 3bf07d7..ebfec4d 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -36,6 +36,7 @@ #include <linux/mpage.h> #include <linux/uio.h> #include <linux/bio.h> +#include <linux/fiemap.h> #include "xattr.h" #include "acl.h" @@ -981,6 +982,13 @@ out: return ret; } +int ext3_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + return generic_block_fiemap(inode, fieinfo, start, len, + ext3_get_block); +} + /* * `handle' can be NULL if create is zero */ @@ -1767,44 +1775,47 @@ static int ext3_journalled_set_page_dirty(struct page *page) } static const struct address_space_operations ext3_ordered_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_ordered_writepage, - .sync_page = block_sync_page, - .write_begin = ext3_write_begin, - .write_end = ext3_ordered_write_end, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, - .direct_IO = ext3_direct_IO, - .migratepage = buffer_migrate_page, + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_ordered_writepage, + .sync_page = block_sync_page, + .write_begin = ext3_write_begin, + .write_end = ext3_ordered_write_end, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, + .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; static const struct address_space_operations ext3_writeback_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_writeback_writepage, - .sync_page = block_sync_page, - .write_begin = ext3_write_begin, - .write_end = ext3_writeback_write_end, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, - .direct_IO = ext3_direct_IO, - .migratepage = buffer_migrate_page, + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_writeback_writepage, + .sync_page = block_sync_page, + .write_begin = ext3_write_begin, + .write_end = ext3_writeback_write_end, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, + .direct_IO = ext3_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; static const struct address_space_operations ext3_journalled_aops = { - .readpage = ext3_readpage, - .readpages = ext3_readpages, - .writepage = ext3_journalled_writepage, - .sync_page = block_sync_page, - .write_begin = ext3_write_begin, - .write_end = ext3_journalled_write_end, - .set_page_dirty = ext3_journalled_set_page_dirty, - .bmap = ext3_bmap, - .invalidatepage = ext3_invalidatepage, - .releasepage = ext3_releasepage, + .readpage = ext3_readpage, + .readpages = ext3_readpages, + .writepage = ext3_journalled_writepage, + .sync_page = block_sync_page, + .write_begin = ext3_write_begin, + .write_end = ext3_journalled_write_end, + .set_page_dirty = ext3_journalled_set_page_dirty, + .bmap = ext3_bmap, + .invalidatepage = ext3_invalidatepage, + .releasepage = ext3_releasepage, + .is_partially_uptodate = block_is_partially_uptodate, }; void ext3_set_aops(struct inode *inode) diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 615788c..f38a5af 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -472,7 +472,7 @@ static void ext3_destroy_inode(struct inode *inode) kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; @@ -2810,8 +2810,9 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, journal_unlock_updates(EXT3_SB(sb)->s_journal); } + err = vfs_quota_on_path(sb, type, format_id, &nd.path); path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path, remount); + return err; } /* Read data from quotafile - avoid pagecache and such because we cannot afford diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index ac6fa8c..a8ff003 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -2,12 +2,12 @@ # Makefile for the linux ext4-filesystem routines. # -obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o +obj-$(CONFIG_EXT4_FS) += ext4.o -ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ +ext4-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ ext4_jbd2.o migrate.o mballoc.o -ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o -ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o -ext4dev-$(CONFIG_EXT4DEV_FS_SECURITY) += xattr_security.o +ext4-$(CONFIG_EXT4_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o +ext4-$(CONFIG_EXT4_FS_POSIX_ACL) += acl.o +ext4-$(CONFIG_EXT4_FS_SECURITY) += xattr_security.o diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 3c8dab8..694ed6f 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -40,34 +40,35 @@ ext4_acl_from_disk(const void *value, size_t size) acl = posix_acl_alloc(count, GFP_NOFS); if (!acl) return ERR_PTR(-ENOMEM); - for (n=0; n < count; n++) { + for (n = 0; n < count; n++) { ext4_acl_entry *entry = (ext4_acl_entry *)value; if ((char *)value + sizeof(ext4_acl_entry_short) > end) goto fail; acl->a_entries[n].e_tag = le16_to_cpu(entry->e_tag); acl->a_entries[n].e_perm = le16_to_cpu(entry->e_perm); - switch(acl->a_entries[n].e_tag) { - case ACL_USER_OBJ: - case ACL_GROUP_OBJ: - case ACL_MASK: - case ACL_OTHER: - value = (char *)value + - sizeof(ext4_acl_entry_short); - acl->a_entries[n].e_id = ACL_UNDEFINED_ID; - break; - - case ACL_USER: - case ACL_GROUP: - value = (char *)value + sizeof(ext4_acl_entry); - if ((char *)value > end) - goto fail; - acl->a_entries[n].e_id = - le32_to_cpu(entry->e_id); - break; - - default: + + switch (acl->a_entries[n].e_tag) { + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + value = (char *)value + + sizeof(ext4_acl_entry_short); + acl->a_entries[n].e_id = ACL_UNDEFINED_ID; + break; + + case ACL_USER: + case ACL_GROUP: + value = (char *)value + sizeof(ext4_acl_entry); + if ((char *)value > end) goto fail; + acl->a_entries[n].e_id = + le32_to_cpu(entry->e_id); + break; + + default: + goto fail; } } if (value != end) @@ -96,27 +97,26 @@ ext4_acl_to_disk(const struct posix_acl *acl, size_t *size) return ERR_PTR(-ENOMEM); ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION); e = (char *)ext_acl + sizeof(ext4_acl_header); - for (n=0; n < acl->a_count; n++) { + for (n = 0; n < acl->a_count; n++) { ext4_acl_entry *entry = (ext4_acl_entry *)e; entry->e_tag = cpu_to_le16(acl->a_entries[n].e_tag); entry->e_perm = cpu_to_le16(acl->a_entries[n].e_perm); - switch(acl->a_entries[n].e_tag) { - case ACL_USER: - case ACL_GROUP: - entry->e_id = - cpu_to_le32(acl->a_entries[n].e_id); - e += sizeof(ext4_acl_entry); - break; - - case ACL_USER_OBJ: - case ACL_GROUP_OBJ: - case ACL_MASK: - case ACL_OTHER: - e += sizeof(ext4_acl_entry_short); - break; - - default: - goto fail; + switch (acl->a_entries[n].e_tag) { + case ACL_USER: + case ACL_GROUP: + entry->e_id = cpu_to_le32(acl->a_entries[n].e_id); + e += sizeof(ext4_acl_entry); + break; + + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + e += sizeof(ext4_acl_entry_short); + break; + + default: + goto fail; } } return (char *)ext_acl; @@ -167,23 +167,23 @@ ext4_get_acl(struct inode *inode, int type) if (!test_opt(inode->i_sb, POSIX_ACL)) return NULL; - switch(type) { - case ACL_TYPE_ACCESS: - acl = ext4_iget_acl(inode, &ei->i_acl); - if (acl != EXT4_ACL_NOT_CACHED) - return acl; - name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; - break; - - case ACL_TYPE_DEFAULT: - acl = ext4_iget_acl(inode, &ei->i_default_acl); - if (acl != EXT4_ACL_NOT_CACHED) - return acl; - name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; - break; - - default: - return ERR_PTR(-EINVAL); + switch (type) { + case ACL_TYPE_ACCESS: + acl = ext4_iget_acl(inode, &ei->i_acl); + if (acl != EXT4_ACL_NOT_CACHED) + return acl; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; + break; + + case ACL_TYPE_DEFAULT: + acl = ext4_iget_acl(inode, &ei->i_default_acl); + if (acl != EXT4_ACL_NOT_CACHED) + return acl; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; + break; + + default: + return ERR_PTR(-EINVAL); } retval = ext4_xattr_get(inode, name_index, "", NULL, 0); if (retval > 0) { @@ -201,14 +201,14 @@ ext4_get_acl(struct inode *inode, int type) kfree(value); if (!IS_ERR(acl)) { - switch(type) { - case ACL_TYPE_ACCESS: - ext4_iset_acl(inode, &ei->i_acl, acl); - break; - - case ACL_TYPE_DEFAULT: - ext4_iset_acl(inode, &ei->i_default_acl, acl); - break; + switch (type) { + case ACL_TYPE_ACCESS: + ext4_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext4_iset_acl(inode, &ei->i_default_acl, acl); + break; } } return acl; @@ -232,31 +232,31 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type, if (S_ISLNK(inode->i_mode)) return -EOPNOTSUPP; - switch(type) { - case ACL_TYPE_ACCESS: - name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; - if (acl) { - mode_t mode = inode->i_mode; - error = posix_acl_equiv_mode(acl, &mode); - if (error < 0) - return error; - else { - inode->i_mode = mode; - ext4_mark_inode_dirty(handle, inode); - if (error == 0) - acl = NULL; - } + switch (type) { + case ACL_TYPE_ACCESS: + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; + if (acl) { + mode_t mode = inode->i_mode; + error = posix_acl_equiv_mode(acl, &mode); + if (error < 0) + return error; + else { + inode->i_mode = mode; + ext4_mark_inode_dirty(handle, inode); + if (error == 0) + acl = NULL; } - break; + } + break; - case ACL_TYPE_DEFAULT: - name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; - if (!S_ISDIR(inode->i_mode)) - return acl ? -EACCES : 0; - break; + case ACL_TYPE_DEFAULT: + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; + if (!S_ISDIR(inode->i_mode)) + return acl ? -EACCES : 0; + break; - default: - return -EINVAL; + default: + return -EINVAL; } if (acl) { value = ext4_acl_to_disk(acl, &size); @@ -269,14 +269,14 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type, kfree(value); if (!error) { - switch(type) { - case ACL_TYPE_ACCESS: - ext4_iset_acl(inode, &ei->i_acl, acl); - break; - - case ACL_TYPE_DEFAULT: - ext4_iset_acl(inode, &ei->i_default_acl, acl); - break; + switch (type) { + case ACL_TYPE_ACCESS: + ext4_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext4_iset_acl(inode, &ei->i_default_acl, acl); + break; } } return error; @@ -299,7 +299,7 @@ ext4_check_acl(struct inode *inode, int mask) } int -ext4_permission(struct inode *inode, int mask, struct nameidata *nd) +ext4_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, ext4_check_acl); } diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index 26a5c1a..cb45257 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -51,18 +51,18 @@ static inline int ext4_acl_count(size_t size) } } -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL /* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl if the ACL has not been cached */ #define EXT4_ACL_NOT_CACHED ((void *)-1) /* acl.c */ -extern int ext4_permission (struct inode *, int, struct nameidata *); -extern int ext4_acl_chmod (struct inode *); -extern int ext4_init_acl (handle_t *, struct inode *, struct inode *); +extern int ext4_permission(struct inode *, int); +extern int ext4_acl_chmod(struct inode *); +extern int ext4_init_acl(handle_t *, struct inode *, struct inode *); -#else /* CONFIG_EXT4DEV_FS_POSIX_ACL */ +#else /* CONFIG_EXT4_FS_POSIX_ACL */ #include <linux/sched.h> #define ext4_permission NULL @@ -77,5 +77,5 @@ ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) { return 0; } -#endif /* CONFIG_EXT4DEV_FS_POSIX_ACL */ +#endif /* CONFIG_EXT4_FS_POSIX_ACL */ diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 495ab21..bd2ece2 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -83,6 +83,7 @@ static int ext4_group_used_meta_blocks(struct super_block *sb, } return used_blocks; } + /* Initializes an uninitialized block bitmap if given, and returns the * number of blocks free in the group. */ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, @@ -132,7 +133,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, */ group_blocks = ext4_blocks_count(sbi->s_es) - le32_to_cpu(sbi->s_es->s_first_data_block) - - (EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count -1)); + (EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count - 1)); } else { group_blocks = EXT4_BLOCKS_PER_GROUP(sb); } @@ -200,20 +201,20 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, * @bh: pointer to the buffer head to store the block * group descriptor */ -struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, +struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb, ext4_group_t block_group, - struct buffer_head ** bh) + struct buffer_head **bh) { unsigned long group_desc; unsigned long offset; - struct ext4_group_desc * desc; + struct ext4_group_desc *desc; struct ext4_sb_info *sbi = EXT4_SB(sb); if (block_group >= sbi->s_groups_count) { - ext4_error (sb, "ext4_get_group_desc", - "block_group >= groups_count - " - "block_group = %lu, groups_count = %lu", - block_group, sbi->s_groups_count); + ext4_error(sb, "ext4_get_group_desc", + "block_group >= groups_count - " + "block_group = %lu, groups_count = %lu", + block_group, sbi->s_groups_count); return NULL; } @@ -222,10 +223,10 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); if (!sbi->s_group_desc[group_desc]) { - ext4_error (sb, "ext4_get_group_desc", - "Group descriptor not loaded - " - "block_group = %lu, group_desc = %lu, desc = %lu", - block_group, group_desc, offset); + ext4_error(sb, "ext4_get_group_desc", + "Group descriptor not loaded - " + "block_group = %lu, group_desc = %lu, desc = %lu", + block_group, group_desc, offset); return NULL; } @@ -302,8 +303,8 @@ err_out: struct buffer_head * ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) { - struct ext4_group_desc * desc; - struct buffer_head * bh = NULL; + struct ext4_group_desc *desc; + struct buffer_head *bh = NULL; ext4_fsblk_t bitmap_blk; desc = ext4_get_group_desc(sb, block_group, NULL); @@ -314,25 +315,30 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) if (unlikely(!bh)) { ext4_error(sb, __func__, "Cannot read block bitmap - " - "block_group = %d, block_bitmap = %llu", - (int)block_group, (unsigned long long)bitmap_blk); + "block_group = %lu, block_bitmap = %llu", + block_group, bitmap_blk); return NULL; } - if (bh_uptodate_or_lock(bh)) + if (buffer_uptodate(bh) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) return bh; + lock_buffer(bh); + spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { ext4_init_block_bitmap(sb, bh, block_group, desc); set_buffer_uptodate(bh); unlock_buffer(bh); + spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group)); return bh; } + spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (bh_submit_read(bh) < 0) { put_bh(bh); ext4_error(sb, __func__, "Cannot read block bitmap - " - "block_group = %d, block_bitmap = %llu", - (int)block_group, (unsigned long long)bitmap_blk); + "block_group = %lu, block_bitmap = %llu", + block_group, bitmap_blk); return NULL; } ext4_valid_block_bitmap(sb, desc, block_group, bh); @@ -342,301 +348,6 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) */ return bh; } -/* - * The reservation window structure operations - * -------------------------------------------- - * Operations include: - * dump, find, add, remove, is_empty, find_next_reservable_window, etc. - * - * We use a red-black tree to represent per-filesystem reservation - * windows. - * - */ - -/** - * __rsv_window_dump() -- Dump the filesystem block allocation reservation map - * @rb_root: root of per-filesystem reservation rb tree - * @verbose: verbose mode - * @fn: function which wishes to dump the reservation map - * - * If verbose is turned on, it will print the whole block reservation - * windows(start, end). Otherwise, it will only print out the "bad" windows, - * those windows that overlap with their immediate neighbors. - */ -#if 1 -static void __rsv_window_dump(struct rb_root *root, int verbose, - const char *fn) -{ - struct rb_node *n; - struct ext4_reserve_window_node *rsv, *prev; - int bad; - -restart: - n = rb_first(root); - bad = 0; - prev = NULL; - - printk("Block Allocation Reservation Windows Map (%s):\n", fn); - while (n) { - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - if (verbose) - printk("reservation window 0x%p " - "start: %llu, end: %llu\n", - rsv, rsv->rsv_start, rsv->rsv_end); - if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { - printk("Bad reservation %p (start >= end)\n", - rsv); - bad = 1; - } - if (prev && prev->rsv_end >= rsv->rsv_start) { - printk("Bad reservation %p (prev->end >= start)\n", - rsv); - bad = 1; - } - if (bad) { - if (!verbose) { - printk("Restarting reservation walk in verbose mode\n"); - verbose = 1; - goto restart; - } - } - n = rb_next(n); - prev = rsv; - } - printk("Window map complete.\n"); - BUG_ON(bad); -} -#define rsv_window_dump(root, verbose) \ - __rsv_window_dump((root), (verbose), __func__) -#else -#define rsv_window_dump(root, verbose) do {} while (0) -#endif - -/** - * goal_in_my_reservation() - * @rsv: inode's reservation window - * @grp_goal: given goal block relative to the allocation block group - * @group: the current allocation block group - * @sb: filesystem super block - * - * Test if the given goal block (group relative) is within the file's - * own block reservation window range. - * - * If the reservation window is outside the goal allocation group, return 0; - * grp_goal (given goal block) could be -1, which means no specific - * goal block. In this case, always return 1. - * If the goal block is within the reservation window, return 1; - * otherwise, return 0; - */ -static int -goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal, - ext4_group_t group, struct super_block *sb) -{ - ext4_fsblk_t group_first_block, group_last_block; - - group_first_block = ext4_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - if ((rsv->_rsv_start > group_last_block) || - (rsv->_rsv_end < group_first_block)) - return 0; - if ((grp_goal >= 0) && ((grp_goal + group_first_block < rsv->_rsv_start) - || (grp_goal + group_first_block > rsv->_rsv_end))) - return 0; - return 1; -} - -/** - * search_reserve_window() - * @rb_root: root of reservation tree - * @goal: target allocation block - * - * Find the reserved window which includes the goal, or the previous one - * if the goal is not in any window. - * Returns NULL if there are no windows or if all windows start after the goal. - */ -static struct ext4_reserve_window_node * -search_reserve_window(struct rb_root *root, ext4_fsblk_t goal) -{ - struct rb_node *n = root->rb_node; - struct ext4_reserve_window_node *rsv; - - if (!n) - return NULL; - - do { - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - - if (goal < rsv->rsv_start) - n = n->rb_left; - else if (goal > rsv->rsv_end) - n = n->rb_right; - else - return rsv; - } while (n); - /* - * We've fallen off the end of the tree: the goal wasn't inside - * any particular node. OK, the previous node must be to one - * side of the interval containing the goal. If it's the RHS, - * we need to back up one. - */ - if (rsv->rsv_start > goal) { - n = rb_prev(&rsv->rsv_node); - rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); - } - return rsv; -} - -/** - * ext4_rsv_window_add() -- Insert a window to the block reservation rb tree. - * @sb: super block - * @rsv: reservation window to add - * - * Must be called with rsv_lock hold. - */ -void ext4_rsv_window_add(struct super_block *sb, - struct ext4_reserve_window_node *rsv) -{ - struct rb_root *root = &EXT4_SB(sb)->s_rsv_window_root; - struct rb_node *node = &rsv->rsv_node; - ext4_fsblk_t start = rsv->rsv_start; - - struct rb_node ** p = &root->rb_node; - struct rb_node * parent = NULL; - struct ext4_reserve_window_node *this; - - while (*p) - { - parent = *p; - this = rb_entry(parent, struct ext4_reserve_window_node, rsv_node); - - if (start < this->rsv_start) - p = &(*p)->rb_left; - else if (start > this->rsv_end) - p = &(*p)->rb_right; - else { - rsv_window_dump(root, 1); - BUG(); - } - } - - rb_link_node(node, parent, p); - rb_insert_color(node, root); -} - -/** - * ext4_rsv_window_remove() -- unlink a window from the reservation rb tree - * @sb: super block - * @rsv: reservation window to remove - * - * Mark the block reservation window as not allocated, and unlink it - * from the filesystem reservation window rb tree. Must be called with - * rsv_lock hold. - */ -static void rsv_window_remove(struct super_block *sb, - struct ext4_reserve_window_node *rsv) -{ - rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_alloc_hit = 0; - rb_erase(&rsv->rsv_node, &EXT4_SB(sb)->s_rsv_window_root); -} - -/* - * rsv_is_empty() -- Check if the reservation window is allocated. - * @rsv: given reservation window to check - * - * returns 1 if the end block is EXT4_RESERVE_WINDOW_NOT_ALLOCATED. - */ -static inline int rsv_is_empty(struct ext4_reserve_window *rsv) -{ - /* a valid reservation end block could not be 0 */ - return rsv->_rsv_end == EXT4_RESERVE_WINDOW_NOT_ALLOCATED; -} - -/** - * ext4_init_block_alloc_info() - * @inode: file inode structure - * - * Allocate and initialize the reservation window structure, and - * link the window to the ext4 inode structure at last - * - * The reservation window structure is only dynamically allocated - * and linked to ext4 inode the first time the open file - * needs a new block. So, before every ext4_new_block(s) call, for - * regular files, we should check whether the reservation window - * structure exists or not. In the latter case, this function is called. - * Fail to do so will result in block reservation being turned off for that - * open file. - * - * This function is called from ext4_get_blocks_handle(), also called - * when setting the reservation window size through ioctl before the file - * is open for write (needs block allocation). - * - * Needs down_write(i_data_sem) protection prior to call this function. - */ -void ext4_init_block_alloc_info(struct inode *inode) -{ - struct ext4_inode_info *ei = EXT4_I(inode); - struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; - struct super_block *sb = inode->i_sb; - - block_i = kmalloc(sizeof(*block_i), GFP_NOFS); - if (block_i) { - struct ext4_reserve_window_node *rsv = &block_i->rsv_window_node; - - rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - - /* - * if filesystem is mounted with NORESERVATION, the goal - * reservation window size is set to zero to indicate - * block reservation is off - */ - if (!test_opt(sb, RESERVATION)) - rsv->rsv_goal_size = 0; - else - rsv->rsv_goal_size = EXT4_DEFAULT_RESERVE_BLOCKS; - rsv->rsv_alloc_hit = 0; - block_i->last_alloc_logical_block = 0; - block_i->last_alloc_physical_block = 0; - } - ei->i_block_alloc_info = block_i; -} - -/** - * ext4_discard_reservation() - * @inode: inode - * - * Discard(free) block reservation window on last file close, or truncate - * or at last iput(). - * - * It is being called in three cases: - * ext4_release_file(): last writer close the file - * ext4_clear_inode(): last iput(), when nobody link to this file. - * ext4_truncate(): when the block indirect map is about to change. - * - */ -void ext4_discard_reservation(struct inode *inode) -{ - struct ext4_inode_info *ei = EXT4_I(inode); - struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; - struct ext4_reserve_window_node *rsv; - spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; - - ext4_mb_discard_inode_preallocations(inode); - - if (!block_i) - return; - - rsv = &block_i->rsv_window_node; - if (!rsv_is_empty(&rsv->rsv_window)) { - spin_lock(rsv_lock); - if (!rsv_is_empty(&rsv->rsv_window)) - rsv_window_remove(inode->i_sb, rsv); - spin_unlock(rsv_lock); - } -} /** * ext4_free_blocks_sb() -- Free given blocks and update quota @@ -645,6 +356,13 @@ void ext4_discard_reservation(struct inode *inode) * @block: start physcial block to free * @count: number of blocks to free * @pdquot_freed_blocks: pointer to quota + * + * XXX This function is only used by the on-line resizing code, which + * should probably be fixed up to call the mballoc variant. There + * this needs to be cleaned up later; in fact, I'm not convinced this + * is 100% correct in the face of the mballoc code. The online resizing + * code needs to be fixed up to more tightly (and correctly) interlock + * with the mballoc code. */ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, ext4_fsblk_t block, unsigned long count, @@ -656,8 +374,8 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, ext4_grpblk_t bit; unsigned long i; unsigned long overflow; - struct ext4_group_desc * desc; - struct ext4_super_block * es; + struct ext4_group_desc *desc; + struct ext4_super_block *es; struct ext4_sb_info *sbi; int err = 0, ret; ext4_grpblk_t group_freed; @@ -668,13 +386,13 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, if (block < le32_to_cpu(es->s_first_data_block) || block + count < block || block + count > ext4_blocks_count(es)) { - ext4_error (sb, "ext4_free_blocks", - "Freeing blocks not in datazone - " - "block = %llu, count = %lu", block, count); + ext4_error(sb, "ext4_free_blocks", + "Freeing blocks not in datazone - " + "block = %llu, count = %lu", block, count); goto error_return; } - ext4_debug ("freeing block(s) %llu-%llu\n", block, block + count - 1); + ext4_debug("freeing block(s) %llu-%llu\n", block, block + count - 1); do_more: overflow = 0; @@ -691,7 +409,7 @@ do_more: 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); + desc = ext4_get_group_desc(sb, block_group, &gd_bh); if (!desc) goto error_return; @@ -700,10 +418,10 @@ do_more: 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, "ext4_free_blocks", - "Freeing blocks in system zones - " - "Block = %llu, count = %lu", - block, count); + ext4_error(sb, "ext4_free_blocks", + "Freeing blocks in system zones - " + "Block = %llu, count = %lu", + block, count); goto error_return; } @@ -845,7 +563,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t block, unsigned long count, int metadata) { - struct super_block * sb; + struct super_block *sb; unsigned long dquot_freed_blocks; /* this isn't the right place to decide whether block is metadata @@ -856,748 +574,52 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, sb = inode->i_sb; - if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info) - ext4_free_blocks_sb(handle, sb, block, count, - &dquot_freed_blocks); - else - ext4_mb_free_blocks(handle, inode, block, count, - metadata, &dquot_freed_blocks); + ext4_mb_free_blocks(handle, inode, block, count, + metadata, &dquot_freed_blocks); if (dquot_freed_blocks) DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); return; } -/** - * ext4_test_allocatable() - * @nr: given allocation block group - * @bh: bufferhead contains the bitmap of the given block group - * - * For ext4 allocations, we must not reuse any blocks which are - * allocated in the bitmap buffer's "last committed data" copy. This - * prevents deletes from freeing up the page for reuse until we have - * committed the delete transaction. - * - * If we didn't do this, then deleting something and reallocating it as - * data would allow the old block to be overwritten before the - * transaction committed (because we force data to disk before commit). - * This would lead to corruption if we crashed between overwriting the - * data and committing the delete. - * - * @@@ We may want to make this allocation behaviour conditional on - * data-writes at some point, and disable it for metadata allocations or - * sync-data inodes. - */ -static int ext4_test_allocatable(ext4_grpblk_t nr, struct buffer_head *bh) -{ - int ret; - struct journal_head *jh = bh2jh(bh); - - if (ext4_test_bit(nr, bh->b_data)) - return 0; - - jbd_lock_bh_state(bh); - if (!jh->b_committed_data) - ret = 1; - else - ret = !ext4_test_bit(nr, jh->b_committed_data); - jbd_unlock_bh_state(bh); - return ret; -} - -/** - * bitmap_search_next_usable_block() - * @start: the starting block (group relative) of the search - * @bh: bufferhead contains the block group bitmap - * @maxblocks: the ending block (group relative) of the reservation - * - * The bitmap search --- search forward alternately through the actual - * bitmap on disk and the last-committed copy in journal, until we find a - * bit free in both bitmaps. - */ -static ext4_grpblk_t -bitmap_search_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, - ext4_grpblk_t maxblocks) +int ext4_claim_free_blocks(struct ext4_sb_info *sbi, + s64 nblocks) { - ext4_grpblk_t next; - struct journal_head *jh = bh2jh(bh); - - while (start < maxblocks) { - next = ext4_find_next_zero_bit(bh->b_data, maxblocks, start); - if (next >= maxblocks) - return -1; - if (ext4_test_allocatable(next, bh)) - return next; - jbd_lock_bh_state(bh); - if (jh->b_committed_data) - start = ext4_find_next_zero_bit(jh->b_committed_data, - maxblocks, next); - jbd_unlock_bh_state(bh); - } - return -1; -} + s64 free_blocks, dirty_blocks; + s64 root_blocks = 0; + struct percpu_counter *fbc = &sbi->s_freeblocks_counter; + struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter; -/** - * find_next_usable_block() - * @start: the starting block (group relative) to find next - * allocatable block in bitmap. - * @bh: bufferhead contains the block group bitmap - * @maxblocks: the ending block (group relative) for the search - * - * Find an allocatable block in a bitmap. We honor both the bitmap and - * its last-committed copy (if that exists), and perform the "most - * appropriate allocation" algorithm of looking for a free block near - * the initial goal; then for a free byte somewhere in the bitmap; then - * for any free bit in the bitmap. - */ -static ext4_grpblk_t -find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, - ext4_grpblk_t maxblocks) -{ - ext4_grpblk_t here, next; - char *p, *r; - - if (start > 0) { - /* - * The goal was occupied; search forward for a free - * block within the next XX blocks. - * - * end_goal is more or less random, but it has to be - * less than EXT4_BLOCKS_PER_GROUP. Aligning up to the - * next 64-bit boundary is simple.. - */ - ext4_grpblk_t end_goal = (start + 63) & ~63; - if (end_goal > maxblocks) - end_goal = maxblocks; - here = ext4_find_next_zero_bit(bh->b_data, end_goal, start); - if (here < end_goal && ext4_test_allocatable(here, bh)) - return here; - ext4_debug("Bit not found near goal\n"); - } - - here = start; - if (here < 0) - here = 0; - - p = ((char *)bh->b_data) + (here >> 3); - r = memscan(p, 0, ((maxblocks + 7) >> 3) - (here >> 3)); - next = (r - ((char *)bh->b_data)) << 3; - - if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh)) - return next; - - /* - * The bitmap search --- search forward alternately through the actual - * bitmap and the last-committed copy until we find a bit free in - * both - */ - here = bitmap_search_next_usable_block(here, bh, maxblocks); - return here; -} - -/** - * claim_block() - * @block: the free block (group relative) to allocate - * @bh: the bufferhead containts the block group bitmap - * - * We think we can allocate this block in this bitmap. Try to set the bit. - * If that succeeds then check that nobody has allocated and then freed the - * block since we saw that is was not marked in b_committed_data. If it _was_ - * allocated and freed then clear the bit in the bitmap again and return - * zero (failure). - */ -static inline int -claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh) -{ - struct journal_head *jh = bh2jh(bh); - int ret; - - if (ext4_set_bit_atomic(lock, block, bh->b_data)) - return 0; - jbd_lock_bh_state(bh); - if (jh->b_committed_data && ext4_test_bit(block,jh->b_committed_data)) { - ext4_clear_bit_atomic(lock, block, bh->b_data); - ret = 0; - } else { - ret = 1; - } - jbd_unlock_bh_state(bh); - return ret; -} + free_blocks = percpu_counter_read_positive(fbc); + dirty_blocks = percpu_counter_read_positive(dbc); -/** - * ext4_try_to_allocate() - * @sb: superblock - * @handle: handle to this transaction - * @group: given allocation block group - * @bitmap_bh: bufferhead holds the block bitmap - * @grp_goal: given target block within the group - * @count: target number of blocks to allocate - * @my_rsv: reservation window - * - * Attempt to allocate blocks within a give range. Set the range of allocation - * first, then find the first free bit(s) from the bitmap (within the range), - * and at last, allocate the blocks by claiming the found free bit as allocated. - * - * To set the range of this allocation: - * if there is a reservation window, only try to allocate block(s) from the - * file's own reservation window; - * Otherwise, the allocation range starts from the give goal block, ends at - * the block group's last block. - * - * If we failed to allocate the desired block then we may end up crossing to a - * new bitmap. In that case we must release write access to the old one via - * ext4_journal_release_buffer(), else we'll run out of credits. - */ -static ext4_grpblk_t -ext4_try_to_allocate(struct super_block *sb, handle_t *handle, - ext4_group_t group, struct buffer_head *bitmap_bh, - ext4_grpblk_t grp_goal, unsigned long *count, - struct ext4_reserve_window *my_rsv) -{ - ext4_fsblk_t group_first_block; - ext4_grpblk_t start, end; - unsigned long num = 0; - - /* we do allocation within the reservation window if we have a window */ - if (my_rsv) { - group_first_block = ext4_group_first_block_no(sb, group); - if (my_rsv->_rsv_start >= group_first_block) - start = my_rsv->_rsv_start - group_first_block; - else - /* reservation window cross group boundary */ - start = 0; - end = my_rsv->_rsv_end - group_first_block + 1; - if (end > EXT4_BLOCKS_PER_GROUP(sb)) - /* reservation window crosses group boundary */ - end = EXT4_BLOCKS_PER_GROUP(sb); - if ((start <= grp_goal) && (grp_goal < end)) - start = grp_goal; - else - grp_goal = -1; - } else { - if (grp_goal > 0) - start = grp_goal; - else - start = 0; - end = EXT4_BLOCKS_PER_GROUP(sb); - } - - BUG_ON(start > EXT4_BLOCKS_PER_GROUP(sb)); - -repeat: - if (grp_goal < 0 || !ext4_test_allocatable(grp_goal, bitmap_bh)) { - grp_goal = find_next_usable_block(start, bitmap_bh, end); - if (grp_goal < 0) - goto fail_access; - if (!my_rsv) { - int i; - - for (i = 0; i < 7 && grp_goal > start && - ext4_test_allocatable(grp_goal - 1, - bitmap_bh); - i++, grp_goal--) - ; - } - } - start = grp_goal; - - if (!claim_block(sb_bgl_lock(EXT4_SB(sb), group), - grp_goal, bitmap_bh)) { - /* - * The block was allocated by another thread, or it was - * allocated and then freed by another thread - */ - start++; - grp_goal++; - if (start >= end) - goto fail_access; - goto repeat; - } - num++; - grp_goal++; - while (num < *count && grp_goal < end - && ext4_test_allocatable(grp_goal, bitmap_bh) - && claim_block(sb_bgl_lock(EXT4_SB(sb), group), - grp_goal, bitmap_bh)) { - num++; - grp_goal++; - } - *count = num; - return grp_goal - num; -fail_access: - *count = num; - return -1; -} - -/** - * find_next_reservable_window(): - * find a reservable space within the given range. - * It does not allocate the reservation window for now: - * alloc_new_reservation() will do the work later. - * - * @search_head: the head of the searching list; - * This is not necessarily the list head of the whole filesystem - * - * We have both head and start_block to assist the search - * for the reservable space. The list starts from head, - * but we will shift to the place where start_block is, - * then start from there, when looking for a reservable space. - * - * @size: the target new reservation window size - * - * @group_first_block: the first block we consider to start - * the real search from - * - * @last_block: - * the maximum block number that our goal reservable space - * could start from. This is normally the last block in this - * group. The search will end when we found the start of next - * possible reservable space is out of this boundary. - * This could handle the cross boundary reservation window - * request. - * - * basically we search from the given range, rather than the whole - * reservation double linked list, (start_block, last_block) - * to find a free region that is of my size and has not - * been reserved. - * - */ -static int find_next_reservable_window( - struct ext4_reserve_window_node *search_head, - struct ext4_reserve_window_node *my_rsv, - struct super_block * sb, - ext4_fsblk_t start_block, - ext4_fsblk_t last_block) -{ - struct rb_node *next; - struct ext4_reserve_window_node *rsv, *prev; - ext4_fsblk_t cur; - int size = my_rsv->rsv_goal_size; - - /* TODO: make the start of the reservation window byte-aligned */ - /* cur = *start_block & ~7;*/ - cur = start_block; - rsv = search_head; - if (!rsv) - return -1; - - while (1) { - if (cur <= rsv->rsv_end) - cur = rsv->rsv_end + 1; - - /* TODO? - * in the case we could not find a reservable space - * that is what is expected, during the re-search, we could - * remember what's the largest reservable space we could have - * and return that one. - * - * For now it will fail if we could not find the reservable - * space with expected-size (or more)... - */ - if (cur > last_block) - return -1; /* fail */ - - prev = rsv; - next = rb_next(&rsv->rsv_node); - rsv = rb_entry(next,struct ext4_reserve_window_node,rsv_node); + if (!capable(CAP_SYS_RESOURCE) && + sbi->s_resuid != current->fsuid && + (sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid))) + root_blocks = ext4_r_blocks_count(sbi->s_es); - /* - * Reached the last reservation, we can just append to the - * previous one. - */ - if (!next) - break; - - if (cur + size <= rsv->rsv_start) { - /* - * Found a reserveable space big enough. We could - * have a reservation across the group boundary here - */ - break; + if (free_blocks - (nblocks + root_blocks + dirty_blocks) < + EXT4_FREEBLOCKS_WATERMARK) { + free_blocks = percpu_counter_sum(fbc); + dirty_blocks = percpu_counter_sum(dbc); + if (dirty_blocks < 0) { + printk(KERN_CRIT "Dirty block accounting " + "went wrong %lld\n", + dirty_blocks); } } - /* - * we come here either : - * when we reach the end of the whole list, - * and there is empty reservable space after last entry in the list. - * append it to the end of the list. - * - * or we found one reservable space in the middle of the list, - * return the reservation window that we could append to. - * succeed. + /* Check whether we have space after + * accounting for current dirty blocks */ + if (free_blocks < ((root_blocks + nblocks) + dirty_blocks)) + /* we don't have free space */ + return -ENOSPC; - if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window))) - rsv_window_remove(sb, my_rsv); - - /* - * Let's book the whole avaliable window for now. We will check the - * disk bitmap later and then, if there are free blocks then we adjust - * the window size if it's larger than requested. - * Otherwise, we will remove this node from the tree next time - * call find_next_reservable_window. - */ - my_rsv->rsv_start = cur; - my_rsv->rsv_end = cur + size - 1; - my_rsv->rsv_alloc_hit = 0; - - if (prev != my_rsv) - ext4_rsv_window_add(sb, my_rsv); - + /* Add the blocks to nblocks */ + percpu_counter_add(dbc, nblocks); return 0; } /** - * alloc_new_reservation()--allocate a new reservation window - * - * To make a new reservation, we search part of the filesystem - * reservation list (the list that inside the group). We try to - * allocate a new reservation window near the allocation goal, - * or the beginning of the group, if there is no goal. - * - * We first find a reservable space after the goal, then from - * there, we check the bitmap for the first free block after - * it. If there is no free block until the end of group, then the - * whole group is full, we failed. Otherwise, check if the free - * block is inside the expected reservable space, if so, we - * succeed. - * If the first free block is outside the reservable space, then - * start from the first free block, we search for next available - * space, and go on. - * - * on succeed, a new reservation will be found and inserted into the list - * It contains at least one free block, and it does not overlap with other - * reservation windows. - * - * failed: we failed to find a reservation window in this group - * - * @rsv: the reservation - * - * @grp_goal: The goal (group-relative). It is where the search for a - * free reservable space should start from. - * if we have a grp_goal(grp_goal >0 ), then start from there, - * no grp_goal(grp_goal = -1), we start from the first block - * of the group. - * - * @sb: the super block - * @group: the group we are trying to allocate in - * @bitmap_bh: the block group block bitmap - * - */ -static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv, - ext4_grpblk_t grp_goal, struct super_block *sb, - ext4_group_t group, struct buffer_head *bitmap_bh) -{ - struct ext4_reserve_window_node *search_head; - ext4_fsblk_t group_first_block, group_end_block, start_block; - ext4_grpblk_t first_free_block; - struct rb_root *fs_rsv_root = &EXT4_SB(sb)->s_rsv_window_root; - unsigned long size; - int ret; - spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; - - group_first_block = ext4_group_first_block_no(sb, group); - group_end_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - if (grp_goal < 0) - start_block = group_first_block; - else - start_block = grp_goal + group_first_block; - - size = my_rsv->rsv_goal_size; - - if (!rsv_is_empty(&my_rsv->rsv_window)) { - /* - * if the old reservation is cross group boundary - * and if the goal is inside the old reservation window, - * we will come here when we just failed to allocate from - * the first part of the window. We still have another part - * that belongs to the next group. In this case, there is no - * point to discard our window and try to allocate a new one - * in this group(which will fail). we should - * keep the reservation window, just simply move on. - * - * Maybe we could shift the start block of the reservation - * window to the first block of next group. - */ - - if ((my_rsv->rsv_start <= group_end_block) && - (my_rsv->rsv_end > group_end_block) && - (start_block >= my_rsv->rsv_start)) - return -1; - - if ((my_rsv->rsv_alloc_hit > - (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) { - /* - * if the previously allocation hit ratio is - * greater than 1/2, then we double the size of - * the reservation window the next time, - * otherwise we keep the same size window - */ - size = size * 2; - if (size > EXT4_MAX_RESERVE_BLOCKS) - size = EXT4_MAX_RESERVE_BLOCKS; - my_rsv->rsv_goal_size= size; - } - } - - spin_lock(rsv_lock); - /* - * shift the search start to the window near the goal block - */ - search_head = search_reserve_window(fs_rsv_root, start_block); - - /* - * find_next_reservable_window() simply finds a reservable window - * inside the given range(start_block, group_end_block). - * - * To make sure the reservation window has a free bit inside it, we - * need to check the bitmap after we found a reservable window. - */ -retry: - ret = find_next_reservable_window(search_head, my_rsv, sb, - start_block, group_end_block); - - if (ret == -1) { - if (!rsv_is_empty(&my_rsv->rsv_window)) - rsv_window_remove(sb, my_rsv); - spin_unlock(rsv_lock); - return -1; - } - - /* - * On success, find_next_reservable_window() returns the - * reservation window where there is a reservable space after it. - * Before we reserve this reservable space, we need - * to make sure there is at least a free block inside this region. - * - * searching the first free bit on the block bitmap and copy of - * last committed bitmap alternatively, until we found a allocatable - * block. Search start from the start block of the reservable space - * we just found. - */ - spin_unlock(rsv_lock); - first_free_block = bitmap_search_next_usable_block( - my_rsv->rsv_start - group_first_block, - bitmap_bh, group_end_block - group_first_block + 1); - - if (first_free_block < 0) { - /* - * no free block left on the bitmap, no point - * to reserve the space. return failed. - */ - spin_lock(rsv_lock); - if (!rsv_is_empty(&my_rsv->rsv_window)) - rsv_window_remove(sb, my_rsv); - spin_unlock(rsv_lock); - return -1; /* failed */ - } - - start_block = first_free_block + group_first_block; - /* - * check if the first free block is within the - * free space we just reserved - */ - if (start_block >= my_rsv->rsv_start && start_block <= my_rsv->rsv_end) - return 0; /* success */ - /* - * if the first free bit we found is out of the reservable space - * continue search for next reservable space, - * start from where the free block is, - * we also shift the list head to where we stopped last time - */ - search_head = my_rsv; - spin_lock(rsv_lock); - goto retry; -} - -/** - * try_to_extend_reservation() - * @my_rsv: given reservation window - * @sb: super block - * @size: the delta to extend - * - * Attempt to expand the reservation window large enough to have - * required number of free blocks - * - * Since ext4_try_to_allocate() will always allocate blocks within - * the reservation window range, if the window size is too small, - * multiple blocks allocation has to stop at the end of the reservation - * window. To make this more efficient, given the total number of - * blocks needed and the current size of the window, we try to - * expand the reservation window size if necessary on a best-effort - * basis before ext4_new_blocks() tries to allocate blocks, - */ -static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, - struct super_block *sb, int size) -{ - struct ext4_reserve_window_node *next_rsv; - struct rb_node *next; - spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; - - if (!spin_trylock(rsv_lock)) - return; - - next = rb_next(&my_rsv->rsv_node); - - if (!next) - my_rsv->rsv_end += size; - else { - next_rsv = rb_entry(next, struct ext4_reserve_window_node, rsv_node); - - if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) - my_rsv->rsv_end += size; - else - my_rsv->rsv_end = next_rsv->rsv_start - 1; - } - spin_unlock(rsv_lock); -} - -/** - * ext4_try_to_allocate_with_rsv() - * @sb: superblock - * @handle: handle to this transaction - * @group: given allocation block group - * @bitmap_bh: bufferhead holds the block bitmap - * @grp_goal: given target block within the group - * @count: target number of blocks to allocate - * @my_rsv: reservation window - * @errp: pointer to store the error code - * - * This is the main function used to allocate a new block and its reservation - * window. - * - * Each time when a new block allocation is need, first try to allocate from - * its own reservation. If it does not have a reservation window, instead of - * looking for a free bit on bitmap first, then look up the reservation list to - * see if it is inside somebody else's reservation window, we try to allocate a - * reservation window for it starting from the goal first. Then do the block - * allocation within the reservation window. - * - * This will avoid keeping on searching the reservation list again and - * again when somebody is looking for a free block (without - * reservation), and there are lots of free blocks, but they are all - * being reserved. - * - * We use a red-black tree for the per-filesystem reservation list. - * - */ -static ext4_grpblk_t -ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, - ext4_group_t group, struct buffer_head *bitmap_bh, - ext4_grpblk_t grp_goal, - struct ext4_reserve_window_node * my_rsv, - unsigned long *count, int *errp) -{ - ext4_fsblk_t group_first_block, group_last_block; - ext4_grpblk_t ret = 0; - int fatal; - unsigned long num = *count; - - *errp = 0; - - /* - * Make sure we use undo access for the bitmap, because it is critical - * that we do the frozen_data COW on bitmap buffers in all cases even - * if the buffer is in BJ_Forget state in the committing transaction. - */ - BUFFER_TRACE(bitmap_bh, "get undo access for new block"); - fatal = ext4_journal_get_undo_access(handle, bitmap_bh); - if (fatal) { - *errp = fatal; - return -1; - } - - /* - * we don't deal with reservation when - * filesystem is mounted without reservation - * or the file is not a regular file - * or last attempt to allocate a block with reservation turned on failed - */ - if (my_rsv == NULL ) { - ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, - grp_goal, count, NULL); - goto out; - } - /* - * grp_goal is a group relative block number (if there is a goal) - * 0 <= grp_goal < EXT4_BLOCKS_PER_GROUP(sb) - * first block is a filesystem wide block number - * first block is the block number of the first block in this group - */ - group_first_block = ext4_group_first_block_no(sb, group); - group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); - - /* - * Basically we will allocate a new block from inode's reservation - * window. - * - * We need to allocate a new reservation window, if: - * a) inode does not have a reservation window; or - * b) last attempt to allocate a block from existing reservation - * failed; or - * c) we come here with a goal and with a reservation window - * - * We do not need to allocate a new reservation window if we come here - * at the beginning with a goal and the goal is inside the window, or - * we don't have a goal but already have a reservation window. - * then we could go to allocate from the reservation window directly. - */ - while (1) { - if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) || - !goal_in_my_reservation(&my_rsv->rsv_window, - grp_goal, group, sb)) { - if (my_rsv->rsv_goal_size < *count) - my_rsv->rsv_goal_size = *count; - ret = alloc_new_reservation(my_rsv, grp_goal, sb, - group, bitmap_bh); - if (ret < 0) - break; /* failed */ - - if (!goal_in_my_reservation(&my_rsv->rsv_window, - grp_goal, group, sb)) - grp_goal = -1; - } else if (grp_goal >= 0) { - int curr = my_rsv->rsv_end - - (grp_goal + group_first_block) + 1; - - if (curr < *count) - try_to_extend_reservation(my_rsv, sb, - *count - curr); - } - - if ((my_rsv->rsv_start > group_last_block) || - (my_rsv->rsv_end < group_first_block)) { - rsv_window_dump(&EXT4_SB(sb)->s_rsv_window_root, 1); - BUG(); - } - ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, - grp_goal, &num, &my_rsv->rsv_window); - if (ret >= 0) { - my_rsv->rsv_alloc_hit += num; - *count = num; - break; /* succeed */ - } - num = *count; - } -out: - if (ret >= 0) { - BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for " - "bitmap block"); - fatal = ext4_journal_dirty_metadata(handle, bitmap_bh); - if (fatal) { - *errp = fatal; - return -1; - } - return ret; - } - - BUFFER_TRACE(bitmap_bh, "journal_release_buffer"); - ext4_journal_release_buffer(handle, bitmap_bh); - return ret; -} - -/** * ext4_has_free_blocks() * @sbi: in-core super block structure. * @nblocks: number of neeed blocks @@ -1607,26 +629,34 @@ out: * On success, return nblocks */ ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, - ext4_fsblk_t nblocks) + s64 nblocks) { - ext4_fsblk_t free_blocks; - ext4_fsblk_t root_blocks = 0; + s64 free_blocks, dirty_blocks; + s64 root_blocks = 0; + struct percpu_counter *fbc = &sbi->s_freeblocks_counter; + struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter; - free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + free_blocks = percpu_counter_read_positive(fbc); + dirty_blocks = percpu_counter_read_positive(dbc); if (!capable(CAP_SYS_RESOURCE) && sbi->s_resuid != current->fsuid && (sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid))) root_blocks = ext4_r_blocks_count(sbi->s_es); -#ifdef CONFIG_SMP - if (free_blocks - root_blocks < FBC_BATCH) - free_blocks = - percpu_counter_sum_and_set(&sbi->s_freeblocks_counter); -#endif - if (free_blocks - root_blocks < nblocks) - return free_blocks - root_blocks; + + if (free_blocks - (nblocks + root_blocks + dirty_blocks) < + EXT4_FREEBLOCKS_WATERMARK) { + free_blocks = percpu_counter_sum(fbc); + dirty_blocks = percpu_counter_sum(dbc); + } + if (free_blocks <= (root_blocks + dirty_blocks)) + /* we don't have free space */ + return 0; + + if (free_blocks - (root_blocks + dirty_blocks) < nblocks) + return free_blocks - (root_blocks + dirty_blocks); return nblocks; - } +} /** @@ -1651,303 +681,6 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) return jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); } -/** - * ext4_old_new_blocks() -- core block bitmap based block allocation function - * - * @handle: handle to this transaction - * @inode: file inode - * @goal: given target block(filesystem wide) - * @count: target number of blocks to allocate - * @errp: error code - * - * ext4_old_new_blocks uses a goal block to assist allocation and look up - * the block bitmap directly to do block allocation. It tries to - * allocate block(s) from the block group contains the goal block first. If - * that fails, it will try to allocate block(s) from other block groups - * without any specific goal block. - * - * This function is called when -o nomballoc mount option is enabled - * - */ -ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, unsigned long *count, int *errp) -{ - struct buffer_head *bitmap_bh = NULL; - struct buffer_head *gdp_bh; - ext4_group_t group_no; - ext4_group_t goal_group; - ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ - ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ - ext4_fsblk_t ret_block; /* filesyetem-wide allocated block */ - ext4_group_t bgi; /* blockgroup iteration index */ - int fatal = 0, err; - int performed_allocation = 0; - ext4_grpblk_t free_blocks; /* number of free blocks in a group */ - struct super_block *sb; - struct ext4_group_desc *gdp; - struct ext4_super_block *es; - struct ext4_sb_info *sbi; - struct ext4_reserve_window_node *my_rsv = NULL; - struct ext4_block_alloc_info *block_i; - unsigned short windowsz = 0; - ext4_group_t ngroups; - unsigned long num = *count; - - sb = inode->i_sb; - if (!sb) { - *errp = -ENODEV; - printk("ext4_new_block: nonexistent device"); - return 0; - } - - sbi = EXT4_SB(sb); - if (!EXT4_I(inode)->i_delalloc_reserved_flag) { - /* - * With delalloc we already reserved the blocks - */ - *count = ext4_has_free_blocks(sbi, *count); - } - if (*count == 0) { - *errp = -ENOSPC; - return 0; /*return with ENOSPC error */ - } - num = *count; - - /* - * Check quota for allocation of this block. - */ - if (DQUOT_ALLOC_BLOCK(inode, num)) { - *errp = -EDQUOT; - return 0; - } - - sbi = EXT4_SB(sb); - es = EXT4_SB(sb)->s_es; - ext4_debug("goal=%llu.\n", goal); - /* - * Allocate a block from reservation only when - * filesystem is mounted with reservation(default,-o reservation), and - * it's a regular file, and - * the desired window size is greater than 0 (One could use ioctl - * command EXT4_IOC_SETRSVSZ to set the window size to 0 to turn off - * reservation on that particular file) - */ - block_i = EXT4_I(inode)->i_block_alloc_info; - if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) - my_rsv = &block_i->rsv_window_node; - - /* - * First, test whether the goal block is free. - */ - if (goal < le32_to_cpu(es->s_first_data_block) || - goal >= ext4_blocks_count(es)) - goal = le32_to_cpu(es->s_first_data_block); - ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk); - goal_group = group_no; -retry_alloc: - gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) - goto io_error; - - free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); - /* - * if there is not enough free blocks to make a new resevation - * turn off reservation for this allocation - */ - if (my_rsv && (free_blocks < windowsz) - && (rsv_is_empty(&my_rsv->rsv_window))) - my_rsv = NULL; - - if (free_blocks > 0) { - bitmap_bh = ext4_read_block_bitmap(sb, group_no); - if (!bitmap_bh) - goto io_error; - grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, - group_no, bitmap_bh, grp_target_blk, - my_rsv, &num, &fatal); - if (fatal) - goto out; - if (grp_alloc_blk >= 0) - goto allocated; - } - - ngroups = EXT4_SB(sb)->s_groups_count; - smp_rmb(); - - /* - * Now search the rest of the groups. We assume that - * group_no and gdp correctly point to the last group visited. - */ - for (bgi = 0; bgi < ngroups; bgi++) { - group_no++; - if (group_no >= ngroups) - group_no = 0; - gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) - goto io_error; - free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); - /* - * skip this group if the number of - * free blocks is less than half of the reservation - * window size. - */ - if (free_blocks <= (windowsz/2)) - continue; - - brelse(bitmap_bh); - bitmap_bh = ext4_read_block_bitmap(sb, group_no); - if (!bitmap_bh) - goto io_error; - /* - * try to allocate block(s) from this group, without a goal(-1). - */ - grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, - group_no, bitmap_bh, -1, my_rsv, - &num, &fatal); - if (fatal) - goto out; - if (grp_alloc_blk >= 0) - goto allocated; - } - /* - * We may end up a bogus ealier ENOSPC error due to - * filesystem is "full" of reservations, but - * there maybe indeed free blocks avaliable on disk - * In this case, we just forget about the reservations - * just do block allocation as without reservations. - */ - if (my_rsv) { - my_rsv = NULL; - windowsz = 0; - group_no = goal_group; - goto retry_alloc; - } - /* No space left on the device */ - *errp = -ENOSPC; - goto out; - -allocated: - - ext4_debug("using block group %lu(%d)\n", - group_no, gdp->bg_free_blocks_count); - - BUFFER_TRACE(gdp_bh, "get_write_access"); - fatal = ext4_journal_get_write_access(handle, gdp_bh); - if (fatal) - goto out; - - ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); - - if (in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || - in_range(ext4_inode_bitmap(sb, gdp), ret_block, num) || - in_range(ret_block, ext4_inode_table(sb, gdp), - EXT4_SB(sb)->s_itb_per_group) || - in_range(ret_block + num - 1, ext4_inode_table(sb, gdp), - EXT4_SB(sb)->s_itb_per_group)) { - ext4_error(sb, "ext4_new_block", - "Allocating block in system zone - " - "blocks from %llu, length %lu", - ret_block, num); - /* - * claim_block marked the blocks we allocated - * as in use. So we may want to selectively - * mark some of the blocks as free - */ - goto retry_alloc; - } - - performed_allocation = 1; - -#ifdef CONFIG_JBD2_DEBUG - { - struct buffer_head *debug_bh; - - /* Record bitmap buffer state in the newly allocated block */ - debug_bh = sb_find_get_block(sb, ret_block); - if (debug_bh) { - BUFFER_TRACE(debug_bh, "state when allocated"); - BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap state"); - brelse(debug_bh); - } - } - jbd_lock_bh_state(bitmap_bh); - spin_lock(sb_bgl_lock(sbi, group_no)); - if (buffer_jbd(bitmap_bh) && bh2jh(bitmap_bh)->b_committed_data) { - int i; - - for (i = 0; i < num; i++) { - if (ext4_test_bit(grp_alloc_blk+i, - bh2jh(bitmap_bh)->b_committed_data)) { - printk("%s: block was unexpectedly set in " - "b_committed_data\n", __func__); - } - } - } - ext4_debug("found bit %d\n", grp_alloc_blk); - spin_unlock(sb_bgl_lock(sbi, group_no)); - jbd_unlock_bh_state(bitmap_bh); -#endif - - if (ret_block + num - 1 >= ext4_blocks_count(es)) { - ext4_error(sb, "ext4_new_block", - "block(%llu) >= blocks count(%llu) - " - "block_group = %lu, es == %p ", ret_block, - ext4_blocks_count(es), group_no, es); - goto out; - } - - /* - * It is up to the caller to add the new buffer to a journal - * list of some description. We don't know in advance whether - * the caller wants to use it as metadata or data. - */ - spin_lock(sb_bgl_lock(sbi, group_no)); - if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) - gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT); - le16_add_cpu(&gdp->bg_free_blocks_count, -num); - gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp); - spin_unlock(sb_bgl_lock(sbi, group_no)); - if (!EXT4_I(inode)->i_delalloc_reserved_flag) - percpu_counter_sub(&sbi->s_freeblocks_counter, num); - - if (sbi->s_log_groups_per_flex) { - ext4_group_t flex_group = ext4_flex_group(sbi, group_no); - spin_lock(sb_bgl_lock(sbi, flex_group)); - sbi->s_flex_groups[flex_group].free_blocks -= num; - spin_unlock(sb_bgl_lock(sbi, flex_group)); - } - - BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); - err = ext4_journal_dirty_metadata(handle, gdp_bh); - if (!fatal) - fatal = err; - - sb->s_dirt = 1; - if (fatal) - goto out; - - *errp = 0; - brelse(bitmap_bh); - DQUOT_FREE_BLOCK(inode, *count-num); - *count = num; - return ret_block; - -io_error: - *errp = -EIO; -out: - if (fatal) { - *errp = fatal; - ext4_std_error(sb, fatal); - } - /* - * Undo the block allocation - */ - if (!performed_allocation) - DQUOT_FREE_BLOCK(inode, *count); - brelse(bitmap_bh); - return 0; -} - #define EXT4_META_BLOCK 0x1 static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode, @@ -1957,10 +690,6 @@ static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode, struct ext4_allocation_request ar; ext4_fsblk_t ret; - if (!test_opt(inode->i_sb, MBALLOC)) { - return ext4_old_new_blocks(handle, inode, goal, count, errp); - } - memset(&ar, 0, sizeof(ar)); /* Fill with neighbour allocated blocks */ @@ -2002,7 +731,7 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, /* * Account for the allocated meta blocks */ - if (!(*errp)) { + if (!(*errp) && EXT4_I(inode)->i_delalloc_reserved_flag) { spin_lock(&EXT4_I(inode)->i_block_reservation_lock); EXT4_I(inode)->i_allocated_meta_blocks += *count; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); @@ -2087,10 +816,9 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) bitmap_count += x; } brelse(bitmap_bh); - printk("ext4_count_free_blocks: stored = %llu" - ", computed = %llu, %llu\n", - ext4_free_blocks_count(es), - desc_count, bitmap_count); + printk(KERN_DEBUG "ext4_count_free_blocks: stored = %llu" + ", computed = %llu, %llu\n", ext4_free_blocks_count(es), + desc_count, bitmap_count); return bitmap_count; #else desc_count = 0; @@ -2177,8 +905,9 @@ unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group) if (!EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG) || metagroup < first_meta_bg) - return ext4_bg_num_gdb_nometa(sb,group); + return ext4_bg_num_gdb_nometa(sb, group); return ext4_bg_num_gdb_meta(sb,group); } + diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index d37ea67..0a7a666 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -15,17 +15,17 @@ static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; -unsigned long ext4_count_free (struct buffer_head * map, unsigned int numchars) +unsigned long ext4_count_free(struct buffer_head *map, unsigned int numchars) { unsigned int i; unsigned long sum = 0; if (!map) - return (0); + return 0; for (i = 0; i < numchars; i++) sum += nibblemap[map->b_data[i] & 0xf] + nibblemap[(map->b_data[i] >> 4) & 0xf]; - return (sum); + return sum; } #endif /* EXT4FS_DEBUG */ diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index d3d23d7..3ca6a2b 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -33,10 +33,10 @@ static unsigned char ext4_filetype_table[] = { }; static int ext4_readdir(struct file *, void *, filldir_t); -static int ext4_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir); -static int ext4_release_dir (struct inode * inode, - struct file * filp); +static int ext4_dx_readdir(struct file *filp, + void *dirent, filldir_t filldir); +static int ext4_release_dir(struct inode *inode, + struct file *filp); const struct file_operations ext4_dir_operations = { .llseek = generic_file_llseek, @@ -61,12 +61,12 @@ static unsigned char get_dtype(struct super_block *sb, int filetype) } -int ext4_check_dir_entry (const char * function, struct inode * dir, - struct ext4_dir_entry_2 * de, - struct buffer_head * bh, - unsigned long offset) +int ext4_check_dir_entry(const char *function, struct inode *dir, + struct ext4_dir_entry_2 *de, + struct buffer_head *bh, + unsigned long offset) { - const char * error_msg = NULL; + const char *error_msg = NULL; const int rlen = ext4_rec_len_from_disk(de->rec_len); if (rlen < EXT4_DIR_REC_LEN(1)) @@ -82,7 +82,7 @@ int ext4_check_dir_entry (const char * function, struct inode * dir, error_msg = "inode out of bounds"; if (error_msg != NULL) - ext4_error (dir->i_sb, function, + ext4_error(dir->i_sb, function, "bad entry in directory #%lu: %s - " "offset=%lu, inode=%lu, rec_len=%d, name_len=%d", dir->i_ino, error_msg, offset, @@ -91,8 +91,8 @@ int ext4_check_dir_entry (const char * function, struct inode * dir, return error_msg == NULL ? 1 : 0; } -static int ext4_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext4_readdir(struct file *filp, + void *dirent, filldir_t filldir) { int error = 0; unsigned long offset; @@ -102,6 +102,7 @@ static int ext4_readdir(struct file * filp, int err; struct inode *inode = filp->f_path.dentry->d_inode; int ret = 0; + int dir_has_error = 0; sb = inode->i_sb; @@ -148,9 +149,13 @@ static int ext4_readdir(struct file * filp, * of recovering data when there's a bad sector */ if (!bh) { - ext4_error (sb, "ext4_readdir", - "directory #%lu contains a hole at offset %lu", - inode->i_ino, (unsigned long)filp->f_pos); + if (!dir_has_error) { + ext4_error(sb, __func__, "directory #%lu " + "contains a hole at offset %Lu", + inode->i_ino, + (unsigned long long) filp->f_pos); + dir_has_error = 1; + } /* corrupt size? Maybe no more blocks to read */ if (filp->f_pos > inode->i_blocks << 9) break; @@ -187,14 +192,14 @@ revalidate: while (!error && filp->f_pos < inode->i_size && offset < sb->s_blocksize) { de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); - if (!ext4_check_dir_entry ("ext4_readdir", inode, de, - bh, offset)) { + if (!ext4_check_dir_entry("ext4_readdir", inode, de, + bh, offset)) { /* * On error, skip the f_pos to the next block */ filp->f_pos = (filp->f_pos | (sb->s_blocksize - 1)) + 1; - brelse (bh); + brelse(bh); ret = stored; goto out; } @@ -218,12 +223,12 @@ revalidate: break; if (version != filp->f_version) goto revalidate; - stored ++; + stored++; } filp->f_pos += ext4_rec_len_from_disk(de->rec_len); } offset = 0; - brelse (bh); + brelse(bh); } out: return ret; @@ -290,9 +295,9 @@ static void free_rb_tree_fname(struct rb_root *root) parent = rb_parent(n); fname = rb_entry(n, struct fname, rb_hash); while (fname) { - struct fname * old = fname; + struct fname *old = fname; fname = fname->next; - kfree (old); + kfree(old); } if (!parent) root->rb_node = NULL; @@ -331,7 +336,7 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, struct ext4_dir_entry_2 *dirent) { struct rb_node **p, *parent = NULL; - struct fname * fname, *new_fn; + struct fname *fname, *new_fn; struct dir_private_info *info; int len; @@ -388,19 +393,20 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, * for all entres on the fname linked list. (Normally there is only * one entry on the linked list, unless there are 62 bit hash collisions.) */ -static int call_filldir(struct file * filp, void * dirent, +static int call_filldir(struct file *filp, void *dirent, filldir_t filldir, struct fname *fname) { struct dir_private_info *info = filp->private_data; loff_t curr_pos; struct inode *inode = filp->f_path.dentry->d_inode; - struct super_block * sb; + struct super_block *sb; int error; sb = inode->i_sb; if (!fname) { - printk("call_filldir: called with null fname?!?\n"); + printk(KERN_ERR "ext4: call_filldir: called with " + "null fname?!?\n"); return 0; } curr_pos = hash2pos(fname->hash, fname->minor_hash); @@ -411,7 +417,7 @@ static int call_filldir(struct file * filp, void * dirent, get_dtype(sb, fname->file_type)); if (error) { filp->f_pos = curr_pos; - info->extra_fname = fname->next; + info->extra_fname = fname; return error; } fname = fname->next; @@ -419,8 +425,8 @@ static int call_filldir(struct file * filp, void * dirent, return 0; } -static int ext4_dx_readdir(struct file * filp, - void * dirent, filldir_t filldir) +static int ext4_dx_readdir(struct file *filp, + void *dirent, filldir_t filldir) { struct dir_private_info *info = filp->private_data; struct inode *inode = filp->f_path.dentry->d_inode; @@ -450,11 +456,21 @@ static int ext4_dx_readdir(struct file * filp, * If there are any leftover names on the hash collision * chain, return them first. */ - if (info->extra_fname && - call_filldir(filp, dirent, filldir, info->extra_fname)) - goto finished; + if (info->extra_fname) { + if (call_filldir(filp, dirent, filldir, info->extra_fname)) + goto finished; - if (!info->curr_node) + info->extra_fname = NULL; + info->curr_node = rb_next(info->curr_node); + if (!info->curr_node) { + if (info->next_hash == ~0) { + filp->f_pos = EXT4_HTREE_EOF; + goto finished; + } + info->curr_hash = info->next_hash; + info->curr_minor_hash = 0; + } + } else if (!info->curr_node) info->curr_node = rb_first(&info->root); while (1) { @@ -501,7 +517,7 @@ finished: return 0; } -static int ext4_release_dir (struct inode * inode, struct file * filp) +static int ext4_release_dir(struct inode *inode, struct file *filp) { if (filp->private_data) ext4_htree_free_dir_info(filp->private_data); diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 303e41c..6690a41 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -44,9 +44,9 @@ #ifdef EXT4FS_DEBUG #define ext4_debug(f, a...) \ do { \ - printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ + printk(KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ __FILE__, __LINE__, __func__); \ - printk (KERN_DEBUG f, ## a); \ + printk(KERN_DEBUG f, ## a); \ } while (0) #else #define ext4_debug(f, a...) do {} while (0) @@ -128,7 +128,7 @@ struct ext4_allocation_request { #else # define EXT4_BLOCK_SIZE(s) (EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size) #endif -#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof (__u32)) +#define EXT4_ADDR_PER_BLOCK(s) (EXT4_BLOCK_SIZE(s) / sizeof(__u32)) #ifdef __KERNEL__ # define EXT4_BLOCK_SIZE_BITS(s) ((s)->s_blocksize_bits) #else @@ -245,7 +245,7 @@ struct flex_groups { #define EXT4_RESERVED_FL 0x80000000 /* reserved for ext4 lib */ #define EXT4_FL_USER_VISIBLE 0x000BDFFF /* User visible flags */ -#define EXT4_FL_USER_MODIFIABLE 0x000380FF /* User modifiable flags */ +#define EXT4_FL_USER_MODIFIABLE 0x000B80FF /* User modifiable flags */ /* * Inode dynamic state flags @@ -291,8 +291,6 @@ struct ext4_new_group_data { #define EXT4_IOC_SETFLAGS FS_IOC_SETFLAGS #define EXT4_IOC_GETVERSION _IOR('f', 3, long) #define EXT4_IOC_SETVERSION _IOW('f', 4, long) -#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) -#define EXT4_IOC_GROUP_ADD _IOW('f', 8,struct ext4_new_group_input) #define EXT4_IOC_GETVERSION_OLD FS_IOC_GETVERSION #define EXT4_IOC_SETVERSION_OLD FS_IOC_SETVERSION #ifdef CONFIG_JBD2_DEBUG @@ -300,7 +298,10 @@ struct ext4_new_group_data { #endif #define EXT4_IOC_GETRSVSZ _IOR('f', 5, long) #define EXT4_IOC_SETRSVSZ _IOW('f', 6, long) -#define EXT4_IOC_MIGRATE _IO('f', 7) +#define EXT4_IOC_GROUP_EXTEND _IOW('f', 7, unsigned long) +#define EXT4_IOC_GROUP_ADD _IOW('f', 8, struct ext4_new_group_input) +#define EXT4_IOC_MIGRATE _IO('f', 9) + /* note ioctl 11 reserved for filesystem-independent FIEMAP ioctl */ /* * ioctl commands in 32 bit emulation @@ -538,8 +539,9 @@ do { \ #define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ -#define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ #define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */ +#define EXT4_MOUNT_DATA_ERR_ABORT 0x10000000 /* Abort on file data write */ + /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt @@ -667,7 +669,7 @@ struct ext4_super_block { }; #ifdef __KERNEL__ -static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb) +static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb) { return sb->s_fs_info; } @@ -725,11 +727,11 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) */ #define EXT4_HAS_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask)) #define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask)) #define EXT4_HAS_INCOMPAT_FEATURE(sb,mask) \ - ( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) ) + (EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask)) #define EXT4_SET_COMPAT_FEATURE(sb,mask) \ EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask) #define EXT4_SET_RO_COMPAT_FEATURE(sb,mask) \ @@ -789,6 +791,8 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) #define EXT4_DEF_RESUID 0 #define EXT4_DEF_RESGID 0 +#define EXT4_DEF_INODE_READAHEAD_BLKS 32 + /* * Default mount options */ @@ -954,6 +958,24 @@ ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no) void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, unsigned long *blockgrpp, ext4_grpblk_t *offsetp); +extern struct proc_dir_entry *ext4_proc_root; + +#ifdef CONFIG_PROC_FS +extern const struct file_operations ext4_ui_proc_fops; + +#define EXT4_PROC_HANDLER(name, var) \ +do { \ + proc = proc_create_data(name, mode, sbi->s_proc, \ + &ext4_ui_proc_fops, &sbi->s_##var); \ + if (proc == NULL) { \ + printk(KERN_ERR "EXT4-fs: can't create %s\n", name); \ + goto err_out; \ + } \ +} while (0) +#else +#define EXT4_PROC_HANDLER(name, var) +#endif + /* * Function prototypes */ @@ -981,23 +1003,20 @@ extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, extern ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, ext4_fsblk_t goal, unsigned long *count, int *errp); -extern ext4_fsblk_t ext4_old_new_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 ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, - ext4_fsblk_t nblocks); -extern void ext4_free_blocks (handle_t *handle, struct inode *inode, + s64 nblocks); +extern void ext4_free_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t block, unsigned long count, int metadata); -extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, - ext4_fsblk_t block, unsigned long count, +extern void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, unsigned long *pdquot_freed_blocks); -extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *); -extern void ext4_check_blocks_bitmap (struct super_block *); +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, ext4_group_t block_group, struct buffer_head ** bh); extern int ext4_should_retry_alloc(struct super_block *sb, int *retries); -extern void ext4_init_block_alloc_info(struct inode *); -extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv); /* dir.c */ extern int ext4_check_dir_entry(const char *, struct inode *, @@ -1009,20 +1028,20 @@ extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, extern void ext4_htree_free_dir_info(struct dir_private_info *p); /* fsync.c */ -extern int ext4_sync_file (struct file *, struct dentry *, int); +extern int ext4_sync_file(struct file *, struct dentry *, int); /* hash.c */ extern int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo); /* ialloc.c */ -extern struct inode * ext4_new_inode (handle_t *, struct inode *, int); -extern void ext4_free_inode (handle_t *, struct inode *); -extern struct inode * ext4_orphan_get (struct super_block *, unsigned long); -extern unsigned long ext4_count_free_inodes (struct super_block *); -extern unsigned long ext4_count_dirs (struct super_block *); -extern void ext4_check_inodes_bitmap (struct super_block *); -extern unsigned long ext4_count_free (struct buffer_head *, unsigned); +extern struct inode * ext4_new_inode(handle_t *, struct inode *, int); +extern void ext4_free_inode(handle_t *, struct inode *); +extern struct inode * ext4_orphan_get(struct super_block *, unsigned long); +extern unsigned long ext4_count_free_inodes(struct super_block *); +extern unsigned long ext4_count_dirs(struct super_block *); +extern void ext4_check_inodes_bitmap(struct super_block *); +extern unsigned long ext4_count_free(struct buffer_head *, unsigned); /* mballoc.c */ extern long ext4_mb_stats; @@ -1032,7 +1051,7 @@ extern int ext4_mb_release(struct super_block *); extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *, struct ext4_allocation_request *, int *); extern int ext4_mb_reserve_blocks(struct super_block *, int); -extern void ext4_mb_discard_inode_preallocations(struct inode *); +extern void ext4_discard_preallocations(struct inode *); extern int __init init_ext4_mballoc(void); extern void exit_ext4_mballoc(void); extern void ext4_mb_free_blocks(handle_t *, struct inode *, @@ -1044,46 +1063,47 @@ extern void ext4_mb_update_group_info(struct ext4_group_info *grp, /* inode.c */ -void ext4_da_release_space(struct inode *inode, int used, int to_free); int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, struct buffer_head *bh, ext4_fsblk_t blocknr); struct buffer_head *ext4_getblk(handle_t *, struct inode *, ext4_lblk_t, int, int *); struct buffer_head *ext4_bread(handle_t *, struct inode *, ext4_lblk_t, int, int *); +int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create); int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned long maxblocks, struct buffer_head *bh_result, int create, int extend_disksize); extern struct inode *ext4_iget(struct super_block *, unsigned long); -extern int ext4_write_inode (struct inode *, int); -extern int ext4_setattr (struct dentry *, struct iattr *); +extern int ext4_write_inode(struct inode *, int); +extern int ext4_setattr(struct dentry *, struct iattr *); extern int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -extern void ext4_delete_inode (struct inode *); -extern int ext4_sync_inode (handle_t *, struct inode *); -extern void ext4_discard_reservation (struct inode *); +extern void ext4_delete_inode(struct inode *); +extern int ext4_sync_inode(handle_t *, struct inode *); extern void ext4_dirty_inode(struct inode *); 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 void ext4_truncate(struct inode *); extern void ext4_set_inode_flags(struct inode *); extern void ext4_get_inode_flags(struct ext4_inode_info *); extern void ext4_set_aops(struct inode *inode); extern int ext4_writepage_trans_blocks(struct inode *); +extern int ext4_meta_trans_blocks(struct inode *, int nrblocks, int idxblocks); +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_page_mkwrite(struct vm_area_struct *vma, struct page *page); /* ioctl.c */ extern long ext4_ioctl(struct file *, unsigned int, unsigned long); -extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long); +extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long); /* migrate.c */ -extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int, - unsigned long); +extern int ext4_ext_migrate(struct inode *); /* namei.c */ extern int ext4_orphan_add(handle_t *, struct inode *); extern int ext4_orphan_del(handle_t *, struct inode *); @@ -1098,14 +1118,14 @@ extern int ext4_group_extend(struct super_block *sb, ext4_fsblk_t n_blocks_count); /* super.c */ -extern void ext4_error (struct super_block *, const char *, const char *, ...) +extern void ext4_error(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void __ext4_std_error (struct super_block *, const char *, int); -extern void ext4_abort (struct super_block *, const char *, const char *, ...) +extern void __ext4_std_error(struct super_block *, const char *, int); +extern void ext4_abort(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext4_warning (struct super_block *, const char *, const char *, ...) +extern void ext4_warning(struct super_block *, const char *, const char *, ...) __attribute__ ((format (printf, 3, 4))); -extern void ext4_update_dynamic_rev (struct super_block *sb); +extern void ext4_update_dynamic_rev(struct super_block *sb); extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb, __u32 compat); extern int ext4_update_rocompat_feature(handle_t *handle, @@ -1178,7 +1198,7 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size) static inline struct ext4_group_info *ext4_get_group_info(struct super_block *sb, - ext4_group_t group) + ext4_group_t group) { struct ext4_group_info ***grp_info; long indexv, indexh; @@ -1206,6 +1226,28 @@ do { \ __ext4_std_error((sb), __func__, (errno)); \ } while (0) +#ifdef CONFIG_SMP +/* Each CPU can accumulate FBC_BATCH blocks in their local + * counters. So we need to make sure we have free blocks more + * than FBC_BATCH * nr_cpu_ids. Also add a window of 4 times. + */ +#define EXT4_FREEBLOCKS_WATERMARK (4 * (FBC_BATCH * nr_cpu_ids)) +#else +#define EXT4_FREEBLOCKS_WATERMARK 0 +#endif + +static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize) +{ + /* + * XXX: replace with spinlock if seen contended -bzzz + */ + down_write(&EXT4_I(inode)->i_data_sem); + if (newsize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = newsize; + up_write(&EXT4_I(inode)->i_data_sem); + return ; +} + /* * Inodes and files operations */ @@ -1228,6 +1270,8 @@ extern const struct inode_operations ext4_fast_symlink_inode_operations; /* extents.c */ extern int ext4_ext_tree_init(handle_t *handle, struct inode *); extern int ext4_ext_writepage_trans_blocks(struct inode *, int); +extern int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, + int chunk); extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 6c166c0..bec7ce5 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -124,6 +124,19 @@ struct ext4_ext_path { #define EXT4_EXT_CACHE_GAP 1 #define EXT4_EXT_CACHE_EXTENT 2 +/* + * to be called by ext4_ext_walk_space() + * negative retcode - error + * 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 *, + struct ext4_ext_cache *, + struct ext4_extent *, void *); + +#define EXT_CONTINUE 0 +#define EXT_BREAK 1 +#define EXT_REPEAT 2 #define EXT_MAX_BLOCK 0xffffffff @@ -216,12 +229,16 @@ extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks); extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *); extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t); extern int ext4_extent_tree_init(handle_t *, struct inode *); -extern int ext4_ext_calc_credits_for_insert(struct inode *, struct ext4_ext_path *); +extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode, + int num, + struct ext4_ext_path *path); extern int ext4_ext_try_to_merge(struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *); extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *); extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *); +extern int ext4_ext_walk_space(struct inode *, ext4_lblk_t, ext4_lblk_t, + ext_prepare_callback, void *); extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t, struct ext4_ext_path *); extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *, diff --git a/fs/ext4/ext4_i.h b/fs/ext4/ext4_i.h index ef7409f..5c124c0 100644 --- a/fs/ext4/ext4_i.h +++ b/fs/ext4/ext4_i.h @@ -33,38 +33,6 @@ typedef __u32 ext4_lblk_t; /* data type for block group number */ typedef unsigned long ext4_group_t; -struct ext4_reserve_window { - ext4_fsblk_t _rsv_start; /* First byte reserved */ - ext4_fsblk_t _rsv_end; /* Last byte reserved or 0 */ -}; - -struct ext4_reserve_window_node { - struct rb_node rsv_node; - __u32 rsv_goal_size; - __u32 rsv_alloc_hit; - struct ext4_reserve_window rsv_window; -}; - -struct ext4_block_alloc_info { - /* information about reservation window */ - struct ext4_reserve_window_node rsv_window_node; - /* - * was i_next_alloc_block in ext4_inode_info - * is the logical (file-relative) number of the - * most-recently-allocated block in this file. - * We use this for detecting linearly ascending allocation requests. - */ - ext4_lblk_t last_alloc_logical_block; - /* - * Was i_next_alloc_goal in ext4_inode_info - * is the *physical* companion to i_next_alloc_block. - * it the physical block number of the block which was most-recentl - * allocated to this file. This give us the goal (target) for the next - * allocation when we detect linearly ascending requests. - */ - ext4_fsblk_t last_alloc_physical_block; -}; - #define rsv_start rsv_window._rsv_start #define rsv_end rsv_window._rsv_end @@ -97,11 +65,8 @@ struct ext4_inode_info { ext4_group_t i_block_group; __u32 i_state; /* Dynamic state flags for ext4 */ - /* block reservation info */ - struct ext4_block_alloc_info *i_block_alloc_info; - ext4_lblk_t i_dir_start_lookup; -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR /* * Extended attributes can be read independently of the main file * data. Taking i_mutex even when reading would cause contention @@ -111,7 +76,7 @@ struct ext4_inode_info { */ struct rw_semaphore xattr_sem; #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL struct posix_acl *i_acl; struct posix_acl *i_default_acl; #endif diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index eb8bc3a..b455c68 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -51,6 +51,14 @@ EXT4_XATTR_TRANS_BLOCKS - 2 + \ 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) +/* + * Define the number of metadata blocks we need to account to modify data. + * + * This include super block, inode block, quota blocks and xattr blocks + */ +#define EXT4_META_TRANS_BLOCKS(sb) (EXT4_XATTR_TRANS_BLOCKS + \ + 2*EXT4_QUOTA_TRANS_BLOCKS(sb)) + /* Delete operations potentially hit one directory's namespace plus an * entire inode, plus arbitrary amounts of bitmap/indirection data. Be * generous. We can grow the delete transaction later if necessary. */ diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h index 6300226..6a0b40d 100644 --- a/fs/ext4/ext4_sb.h +++ b/fs/ext4/ext4_sb.h @@ -40,8 +40,8 @@ struct ext4_sb_info { unsigned long s_blocks_last; /* Last seen block count */ loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */ struct buffer_head * s_sbh; /* Buffer containing the super block */ - struct ext4_super_block * s_es; /* Pointer to the super block in the buffer */ - struct buffer_head ** s_group_desc; + struct ext4_super_block *s_es; /* Pointer to the super block in the buffer */ + struct buffer_head **s_group_desc; unsigned long s_mount_opt; ext4_fsblk_t s_sb_block; uid_t s_resuid; @@ -52,6 +52,7 @@ struct ext4_sb_info { int s_desc_per_block_bits; int s_inode_size; int s_first_ino; + unsigned int s_inode_readahead_blks; spinlock_t s_next_gen_lock; u32 s_next_generation; u32 s_hash_seed[4]; @@ -59,16 +60,17 @@ struct ext4_sb_info { struct percpu_counter s_freeblocks_counter; struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; + struct percpu_counter s_dirtyblocks_counter; struct blockgroup_lock s_blockgroup_lock; + struct proc_dir_entry *s_proc; /* root of the per fs reservation window tree */ spinlock_t s_rsv_window_lock; struct rb_root s_rsv_window_root; - struct ext4_reserve_window_node s_rsv_window_head; /* Journaling */ - struct inode * s_journal_inode; - struct journal_s * s_journal; + struct inode *s_journal_inode; + struct journal_s *s_journal; struct list_head s_orphan; unsigned long s_commit_interval; struct block_device *journal_bdev; @@ -106,12 +108,12 @@ struct ext4_sb_info { /* tunables */ unsigned long s_stripe; - unsigned long s_mb_stream_request; - unsigned long s_mb_max_to_scan; - unsigned long s_mb_min_to_scan; - unsigned long s_mb_stats; - unsigned long s_mb_order2_reqs; - unsigned long s_mb_group_prealloc; + unsigned int s_mb_stream_request; + unsigned int s_mb_max_to_scan; + unsigned int s_mb_min_to_scan; + unsigned int s_mb_stats; + unsigned int s_mb_order2_reqs; + unsigned int s_mb_group_prealloc; /* where last allocation was done - for stream allocation */ unsigned long s_mb_last_group; unsigned long s_mb_last_start; @@ -121,7 +123,6 @@ struct ext4_sb_info { int s_mb_history_cur; int s_mb_history_max; int s_mb_history_num; - struct proc_dir_entry *s_mb_proc; spinlock_t s_mb_history_lock; int s_mb_history_filter; diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 42c4c0c..ea2ce3c 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -40,6 +40,7 @@ #include <linux/slab.h> #include <linux/falloc.h> #include <asm/uaccess.h> +#include <linux/fiemap.h> #include "ext4_jbd2.h" #include "ext4_extents.h" @@ -99,7 +100,7 @@ static int ext4_ext_journal_restart(handle_t *handle, int needed) if (handle->h_buffer_credits > needed) return 0; err = ext4_journal_extend(handle, needed); - if (err) + if (err <= 0) return err; return ext4_journal_restart(handle, needed); } @@ -383,8 +384,8 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) ext_debug("\n"); } #else -#define ext4_ext_show_path(inode,path) -#define ext4_ext_show_leaf(inode,path) +#define ext4_ext_show_path(inode, path) +#define ext4_ext_show_leaf(inode, path) #endif void ext4_ext_drop_refs(struct ext4_ext_path *path) @@ -440,9 +441,10 @@ ext4_ext_binsearch_idx(struct inode *inode, for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) { if (k != 0 && le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) { - printk("k=%d, ix=0x%p, first=0x%p\n", k, - ix, EXT_FIRST_INDEX(eh)); - printk("%u <= %u\n", + printk(KERN_DEBUG "k=%d, ix=0x%p, " + "first=0x%p\n", k, + ix, EXT_FIRST_INDEX(eh)); + printk(KERN_DEBUG "%u <= %u\n", le32_to_cpu(ix->ei_block), le32_to_cpu(ix[-1].ei_block)); } @@ -1441,7 +1443,7 @@ unsigned int ext4_ext_check_overlap(struct inode *inode, /* * get the next allocated block if the extent in the path - * is before the requested block(s) + * is before the requested block(s) */ if (b2 < b1) { b2 = ext4_ext_next_allocated_block(path); @@ -1475,7 +1477,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *newext) { - struct ext4_extent_header * eh; + struct ext4_extent_header *eh; struct ext4_extent *ex, *fex; struct ext4_extent *nearex; /* nearest extent */ struct ext4_ext_path *npath = NULL; @@ -1625,6 +1627,113 @@ cleanup: return err; } +int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, + ext4_lblk_t num, ext_prepare_callback func, + void *cbdata) +{ + struct ext4_ext_path *path = NULL; + struct ext4_ext_cache cbex; + struct ext4_extent *ex; + ext4_lblk_t next, start = 0, end = 0; + ext4_lblk_t last = block + num; + int depth, exists, err = 0; + + BUG_ON(func == NULL); + BUG_ON(inode == NULL); + + while (block < last && block != EXT_MAX_BLOCK) { + num = last - block; + /* find extent for this block */ + path = ext4_ext_find_extent(inode, block, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + break; + } + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + ex = path[depth].p_ext; + next = ext4_ext_next_allocated_block(path); + + exists = 0; + if (!ex) { + /* there is no extent yet, so try to allocate + * all requested space */ + start = block; + end = block + num; + } else if (le32_to_cpu(ex->ee_block) > block) { + /* need to allocate space before found extent */ + start = block; + end = le32_to_cpu(ex->ee_block); + if (block + num < end) + end = block + num; + } else if (block >= le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex)) { + /* need to allocate space after found extent */ + start = block; + end = block + num; + if (end >= next) + end = next; + } else if (block >= le32_to_cpu(ex->ee_block)) { + /* + * some part of requested space is covered + * by found extent + */ + start = block; + end = le32_to_cpu(ex->ee_block) + + ext4_ext_get_actual_len(ex); + if (block + num < end) + end = block + num; + exists = 1; + } else { + BUG(); + } + BUG_ON(end <= start); + + if (!exists) { + cbex.ec_block = start; + cbex.ec_len = end - start; + cbex.ec_start = 0; + cbex.ec_type = EXT4_EXT_CACHE_GAP; + } else { + cbex.ec_block = le32_to_cpu(ex->ee_block); + cbex.ec_len = ext4_ext_get_actual_len(ex); + cbex.ec_start = ext_pblock(ex); + cbex.ec_type = EXT4_EXT_CACHE_EXTENT; + } + + BUG_ON(cbex.ec_len == 0); + err = func(inode, path, &cbex, ex, cbdata); + ext4_ext_drop_refs(path); + + if (err < 0) + break; + + if (err == EXT_REPEAT) + continue; + else if (err == EXT_BREAK) { + err = 0; + break; + } + + if (ext_depth(inode) != depth) { + /* depth was changed. we have to realloc path */ + kfree(path); + path = NULL; + } + + block = cbex.ec_block + cbex.ec_len; + } + + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + + return err; +} + static void ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block, __u32 len, ext4_fsblk_t start, int type) @@ -1747,54 +1856,61 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, } /* - * ext4_ext_calc_credits_for_insert: - * This routine returns max. credits that the extent tree can consume. - * It should be OK for low-performance paths like ->writepage() - * To allow many writing processes to fit into a single transaction, - * the caller should calculate credits under i_data_sem and - * pass the actual path. + * ext4_ext_calc_credits_for_single_extent: + * This routine returns max. credits that needed to insert an extent + * to the extent tree. + * When pass the actual path, the caller should calculate credits + * under i_data_sem. */ -int ext4_ext_calc_credits_for_insert(struct inode *inode, +int ext4_ext_calc_credits_for_single_extent(struct inode *inode, int nrblocks, struct ext4_ext_path *path) { - int depth, needed; - if (path) { + int depth = ext_depth(inode); + int ret = 0; + /* probably there is space in leaf? */ - depth = ext_depth(inode); if (le16_to_cpu(path[depth].p_hdr->eh_entries) - < le16_to_cpu(path[depth].p_hdr->eh_max)) - return 1; - } - - /* - * given 32-bit logical block (4294967296 blocks), max. tree - * can be 4 levels in depth -- 4 * 340^4 == 53453440000. - * Let's also add one more level for imbalance. - */ - depth = 5; + < le16_to_cpu(path[depth].p_hdr->eh_max)) { - /* allocation of new data block(s) */ - needed = 2; + /* + * There are some space in the leaf tree, no + * need to account for leaf block credit + * + * bitmaps and block group descriptor blocks + * and other metadat blocks still need to be + * accounted. + */ + /* 1 bitmap, 1 block group descriptor */ + ret = 2 + EXT4_META_TRANS_BLOCKS(inode->i_sb); + } + } - /* - * tree can be full, so it would need to grow in depth: - * we need one credit to modify old root, credits for - * new root will be added in split accounting - */ - needed += 1; + return ext4_chunk_trans_blocks(inode, nrblocks); +} - /* - * Index split can happen, we would need: - * allocate intermediate indexes (bitmap + group) - * + change two blocks at each level, but root (already included) - */ - needed += (depth * 2) + (depth * 2); +/* + * How many index/leaf blocks need to change/allocate to modify nrblocks? + * + * if nrblocks are fit in a single extent (chunk flag is 1), then + * in the worse case, each tree level index/leaf need to be changed + * if the tree split due to insert a new extent, then the old tree + * index/leaf need to be updated too + * + * If the nrblocks are discontiguous, they could cause + * the whole tree split more than once, but this is really rare. + */ +int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) +{ + int index; + int depth = ext_depth(inode); - /* any allocation modifies superblock */ - needed += 1; + if (chunk) + index = depth * 2; + else + index = depth * 3; - return needed; + return index; } static int ext4_remove_blocks(handle_t *handle, struct inode *inode, @@ -1910,16 +2026,18 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, BUG_ON(b != ex_ee_block + ex_ee_len - 1); } - /* at present, extent can't cross block group: */ - /* leaf + bitmap + group desc + sb + inode */ - credits = 5; + /* + * 3 for leaf, sb, and inode plus 2 (bmap and group + * descriptor) for each block group; assume two block + * groups plus ex_ee_len/blocks_per_block_group for + * the worst case + */ + credits = 7 + 2*(ex_ee_len/EXT4_BLOCKS_PER_GROUP(inode->i_sb)); if (ex == EXT_FIRST_EXTENT(eh)) { correct_index = 1; credits += (ext_depth(inode)) + 1; } -#ifdef CONFIG_QUOTA credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); -#endif err = ext4_ext_journal_restart(handle, credits); if (err) @@ -2133,7 +2251,7 @@ void ext4_ext_init(struct super_block *sb) */ if (test_opt(sb, EXTENTS)) { - printk("EXT4-fs: file extents enabled"); + printk(KERN_INFO "EXT4-fs: file extents enabled"); #ifdef AGGRESSIVE_TEST printk(", aggressive tests"); #endif @@ -2323,7 +2441,10 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, unsigned int newdepth; /* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */ if (allocated <= EXT4_EXT_ZERO_LEN) { - /* Mark first half uninitialized. + /* + * iblock == ee_block is handled by the zerouout + * at the beginning. + * Mark first half uninitialized. * Mark second half initialized and zero out the * initialized extent */ @@ -2346,7 +2467,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ex->ee_len = orig_ex.ee_len; ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); - /* zeroed the full extent */ + /* blocks available from iblock */ return allocated; } else if (err) @@ -2374,6 +2495,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, 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); @@ -2403,6 +2525,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); /* zeroed the full extent */ + /* blocks available from iblock */ return allocated; } else if (err) @@ -2418,23 +2541,22 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, */ orig_ex.ee_len = cpu_to_le16(ee_len - ext4_ext_get_actual_len(ex3)); - if (newdepth != depth) { - depth = newdepth; - ext4_ext_drop_refs(path); - path = ext4_ext_find_extent(inode, iblock, 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; + depth = newdepth; + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, iblock, 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 = max_blocks; /* If extent has less than EXT4_EXT_ZERO_LEN and we are trying @@ -2452,6 +2574,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle, ext4_ext_store_pblock(ex, ext_pblock(&orig_ex)); ext4_ext_dirty(handle, inode, path + depth); /* zero out the first half */ + /* blocks available from iblock */ return allocated; } } @@ -2682,11 +2805,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, goto out2; } /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary. + * Okay, we need to do block allocation. */ - if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) - ext4_init_block_alloc_info(inode); /* find neighbour allocated blocks */ ar.lleft = iblock; @@ -2746,7 +2866,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, /* free data blocks we just allocated */ /* not a good idea to call discard here directly, * but otherwise we'd need to call it every free() */ - ext4_mb_discard_inode_preallocations(inode); + ext4_discard_preallocations(inode); ext4_free_blocks(handle, inode, ext_pblock(&newex), ext4_ext_get_actual_len(&newex), 0); goto out2; @@ -2796,7 +2916,7 @@ void ext4_ext_truncate(struct inode *inode) /* * probably first extent we're gonna free will be last in block */ - err = ext4_writepage_trans_blocks(inode) + 3; + err = ext4_writepage_trans_blocks(inode); handle = ext4_journal_start(inode, err); if (IS_ERR(handle)) return; @@ -2810,7 +2930,7 @@ void ext4_ext_truncate(struct inode *inode) down_write(&EXT4_I(inode)->i_data_sem); ext4_ext_invalidate_cache(inode); - ext4_mb_discard_inode_preallocations(inode); + ext4_discard_preallocations(inode); /* * TODO: optimization is possible here. @@ -2849,27 +2969,6 @@ out_stop: ext4_journal_stop(handle); } -/* - * ext4_ext_writepage_trans_blocks: - * calculate max number of blocks we could modify - * in order to allocate new block for an inode - */ -int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) -{ - int needed; - - needed = ext4_ext_calc_credits_for_insert(inode, NULL); - - /* caller wants to allocate num blocks, but note it includes sb */ - needed = needed * num - (num - 1); - -#ifdef CONFIG_QUOTA - needed += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); -#endif - - return needed; -} - static void ext4_falloc_update_inode(struct inode *inode, int mode, loff_t new_size, int update_ctime) { @@ -2884,10 +2983,11 @@ static void ext4_falloc_update_inode(struct inode *inode, * Update only when preallocation was requested beyond * the file size. */ - if (!(mode & FALLOC_FL_KEEP_SIZE) && - new_size > i_size_read(inode)) { - i_size_write(inode, new_size); - EXT4_I(inode)->i_disksize = new_size; + if (!(mode & FALLOC_FL_KEEP_SIZE)) { + if (new_size > i_size_read(inode)) + i_size_write(inode, new_size); + if (new_size > EXT4_I(inode)->i_disksize) + ext4_update_i_disksize(inode, new_size); } } @@ -2930,10 +3030,9 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits) - block; /* - * credits to insert 1 extent into extent tree + buffers to be able to - * modify 1 super block, 1 block bitmap and 1 group descriptor. + * credits to insert 1 extent into extent tree */ - credits = EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + 3; + credits = ext4_chunk_trans_blocks(inode, max_blocks); mutex_lock(&inode->i_mutex); retry: while (ret >= 0 && ret < max_blocks) { @@ -2980,3 +3079,143 @@ retry: mutex_unlock(&inode->i_mutex); return ret > 0 ? ret2 : ret; } + +/* + * Callback function called for each extent to gather FIEMAP information. + */ +int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path, + struct ext4_ext_cache *newex, struct ext4_extent *ex, + void *data) +{ + struct fiemap_extent_info *fieinfo = data; + unsigned long blksize_bits = inode->i_sb->s_blocksize_bits; + __u64 logical; + __u64 physical; + __u64 length; + __u32 flags = 0; + int error; + + logical = (__u64)newex->ec_block << blksize_bits; + + if (newex->ec_type == EXT4_EXT_CACHE_GAP) { + pgoff_t offset; + struct page *page; + struct buffer_head *bh = NULL; + + offset = logical >> PAGE_SHIFT; + page = find_get_page(inode->i_mapping, offset); + if (!page || !page_has_buffers(page)) + return EXT_CONTINUE; + + bh = page_buffers(page); + + if (!bh) + return EXT_CONTINUE; + + if (buffer_delay(bh)) { + flags |= FIEMAP_EXTENT_DELALLOC; + page_cache_release(page); + } else { + page_cache_release(page); + return EXT_CONTINUE; + } + } + + physical = (__u64)newex->ec_start << blksize_bits; + length = (__u64)newex->ec_len << blksize_bits; + + if (ex && ext4_ext_is_uninitialized(ex)) + flags |= FIEMAP_EXTENT_UNWRITTEN; + + /* + * If this extent reaches EXT_MAX_BLOCK, it must be last. + * + * Or if ext4_ext_next_allocated_block is EXT_MAX_BLOCK, + * this also indicates no more allocated blocks. + * + * XXX this might miss a single-block extent at EXT_MAX_BLOCK + */ + if (logical + length - 1 == EXT_MAX_BLOCK || + ext4_ext_next_allocated_block(path) == EXT_MAX_BLOCK) + flags |= FIEMAP_EXTENT_LAST; + + error = fiemap_fill_next_extent(fieinfo, logical, physical, + length, flags); + if (error < 0) + return error; + if (error == 1) + return EXT_BREAK; + + return EXT_CONTINUE; +} + +/* fiemap flags we can handle specified here */ +#define EXT4_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) + +int ext4_xattr_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo) +{ + __u64 physical = 0; + __u64 length; + __u32 flags = FIEMAP_EXTENT_LAST; + int blockbits = inode->i_sb->s_blocksize_bits; + int error = 0; + + /* in-inode? */ + if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) { + struct ext4_iloc iloc; + int offset; /* offset of xattr in inode */ + + error = ext4_get_inode_loc(inode, &iloc); + if (error) + return error; + physical = iloc.bh->b_blocknr << blockbits; + offset = EXT4_GOOD_OLD_INODE_SIZE + + EXT4_I(inode)->i_extra_isize; + physical += offset; + length = EXT4_SB(inode->i_sb)->s_inode_size - offset; + flags |= FIEMAP_EXTENT_DATA_INLINE; + } else { /* external block */ + physical = EXT4_I(inode)->i_file_acl << blockbits; + length = inode->i_sb->s_blocksize; + } + + if (physical) + error = fiemap_fill_next_extent(fieinfo, 0, physical, + length, flags); + return (error < 0 ? error : 0); +} + +int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + __u64 start, __u64 len) +{ + ext4_lblk_t start_blk; + ext4_lblk_t len_blks; + int error = 0; + + /* fallback to generic here if not in extents fmt */ + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) + return generic_block_fiemap(inode, fieinfo, start, len, + ext4_get_block); + + if (fiemap_check_flags(fieinfo, EXT4_FIEMAP_FLAGS)) + return -EBADR; + + if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) { + error = ext4_xattr_fiemap(inode, fieinfo); + } else { + start_blk = start >> inode->i_sb->s_blocksize_bits; + len_blks = len >> inode->i_sb->s_blocksize_bits; + + /* + * Walk the extent tree gathering extent information. + * ext4_ext_fiemap_cb will push extents back to user. + */ + down_write(&EXT4_I(inode)->i_data_sem); + error = ext4_ext_walk_space(inode, start_blk, len_blks, + ext4_ext_fiemap_cb, fieinfo); + up_write(&EXT4_I(inode)->i_data_sem); + } + + return error; +} + diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 430eb79..6bd11fb 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -31,14 +31,14 @@ * from ext4_file_open: open gets called at every open, but release * gets called only when /all/ the files are closed. */ -static int ext4_release_file (struct inode * inode, struct file * filp) +static int ext4_release_file(struct inode *inode, struct file *filp) { /* if we are the last writer on the inode, drop the block reservation */ if ((filp->f_mode & FMODE_WRITE) && (atomic_read(&inode->i_writecount) == 1)) { down_write(&EXT4_I(inode)->i_data_sem); - ext4_discard_reservation(inode); + ext4_discard_preallocations(inode); up_write(&EXT4_I(inode)->i_data_sem); } if (is_dx(inode) && filp->private_data) @@ -140,6 +140,9 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) return 0; } +extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + __u64 start, __u64 len); + const struct file_operations ext4_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -162,7 +165,7 @@ const struct inode_operations ext4_file_inode_operations = { .truncate = ext4_truncate, .setattr = ext4_setattr, .getattr = ext4_getattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -170,5 +173,6 @@ const struct inode_operations ext4_file_inode_operations = { #endif .permission = ext4_permission, .fallocate = ext4_fallocate, + .fiemap = ext4_fiemap, }; diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index a45c373..5afe437 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -28,6 +28,7 @@ #include <linux/writeback.h> #include <linux/jbd2.h> #include <linux/blkdev.h> +#include <linux/marker.h> #include "ext4.h" #include "ext4_jbd2.h" @@ -43,7 +44,7 @@ * inode to disk. */ -int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) +int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; @@ -51,6 +52,10 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) J_ASSERT(ext4_journal_current_handle() == NULL); + trace_mark(ext4_sync_file, "dev %s datasync %d ino %ld parent %ld", + inode->i_sb->s_id, datasync, inode->i_ino, + dentry->d_parent->d_inode->i_ino); + /* * data=writeback: * The caller's filemap_fdatawrite()/wait will sync the data. diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c index 1d6329d..556ca8e 100644 --- a/fs/ext4/hash.c +++ b/fs/ext4/hash.c @@ -27,7 +27,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[]) sum += DELTA; b0 += ((b1 << 4)+a) ^ (b1+sum) ^ ((b1 >> 5)+b); b1 += ((b0 << 4)+c) ^ (b0+sum) ^ ((b0 >> 5)+d); - } while(--n); + } while (--n); buf[0] += b0; buf[1] += b1; @@ -35,7 +35,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[]) /* The old legacy hash */ -static __u32 dx_hack_hash (const char *name, int len) +static __u32 dx_hack_hash(const char *name, int len) { __u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; while (len--) { @@ -59,7 +59,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) val = pad; if (len > num*4) len = num * 4; - for (i=0; i < len; i++) { + for (i = 0; i < len; i++) { if ((i % 4) == 0) val = pad; val = msg[i] + (val << 8); @@ -104,7 +104,7 @@ int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) /* Check to see if the seed is all zero's */ if (hinfo->seed) { - for (i=0; i < 4; i++) { + for (i = 0; i < 4; i++) { if (hinfo->seed[i]) break; } diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index a92eb30..fe34d74 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -97,34 +97,46 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh, * Return buffer_head of bitmap on success or NULL. */ static struct buffer_head * -read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) +ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group) { struct ext4_group_desc *desc; struct buffer_head *bh = NULL; + ext4_fsblk_t bitmap_blk; desc = ext4_get_group_desc(sb, block_group, NULL); if (!desc) - goto error_out; + return NULL; + bitmap_blk = ext4_inode_bitmap(sb, desc); + bh = sb_getblk(sb, bitmap_blk); + if (unlikely(!bh)) { + ext4_error(sb, __func__, + "Cannot read inode bitmap - " + "block_group = %lu, inode_bitmap = %llu", + block_group, bitmap_blk); + return NULL; + } + if (buffer_uptodate(bh) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT))) + return bh; + + lock_buffer(bh); + spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) { - bh = sb_getblk(sb, ext4_inode_bitmap(sb, desc)); - if (!buffer_uptodate(bh)) { - lock_buffer(bh); - if (!buffer_uptodate(bh)) { - ext4_init_inode_bitmap(sb, bh, block_group, - desc); - set_buffer_uptodate(bh); - } - unlock_buffer(bh); - } - } else { - bh = sb_bread(sb, ext4_inode_bitmap(sb, desc)); + ext4_init_inode_bitmap(sb, bh, block_group, desc); + set_buffer_uptodate(bh); + unlock_buffer(bh); + spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group)); + return bh; } - if (!bh) - ext4_error(sb, "read_inode_bitmap", + spin_unlock(sb_bgl_lock(EXT4_SB(sb), block_group)); + if (bh_submit_read(bh) < 0) { + put_bh(bh); + ext4_error(sb, __func__, "Cannot read inode bitmap - " "block_group = %lu, inode_bitmap = %llu", - block_group, ext4_inode_bitmap(sb, desc)); -error_out: + block_group, bitmap_blk); + return NULL; + } return bh; } @@ -144,39 +156,40 @@ error_out: * though), and then we'd have two inodes sharing the * same inode number and space on the harddisk. */ -void ext4_free_inode (handle_t *handle, struct inode * inode) +void ext4_free_inode(handle_t *handle, struct inode *inode) { - struct super_block * sb = inode->i_sb; + struct super_block *sb = inode->i_sb; int is_directory; unsigned long ino; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; ext4_group_t block_group; unsigned long bit; - struct ext4_group_desc * gdp; - struct ext4_super_block * es; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; struct ext4_sb_info *sbi; int fatal = 0, err; ext4_group_t flex_group; if (atomic_read(&inode->i_count) > 1) { - printk ("ext4_free_inode: inode has count=%d\n", - atomic_read(&inode->i_count)); + printk(KERN_ERR "ext4_free_inode: inode has count=%d\n", + atomic_read(&inode->i_count)); return; } if (inode->i_nlink) { - printk ("ext4_free_inode: inode has nlink=%d\n", - inode->i_nlink); + printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n", + inode->i_nlink); return; } if (!sb) { - printk("ext4_free_inode: inode on nonexistent device\n"); + printk(KERN_ERR "ext4_free_inode: inode on " + "nonexistent device\n"); return; } sbi = EXT4_SB(sb); ino = inode->i_ino; - ext4_debug ("freeing inode %lu\n", ino); + ext4_debug("freeing inode %lu\n", ino); /* * Note: we must free any quota before locking the superblock, @@ -190,17 +203,17 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) is_directory = S_ISDIR(inode->i_mode); /* Do this BEFORE marking the inode not in use or returning an error */ - clear_inode (inode); + clear_inode(inode); es = EXT4_SB(sb)->s_es; if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { - ext4_error (sb, "ext4_free_inode", - "reserved or nonexistent inode %lu", ino); + ext4_error(sb, "ext4_free_inode", + "reserved or nonexistent inode %lu", ino); goto error_return; } block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); - bitmap_bh = read_inode_bitmap(sb, block_group); + bitmap_bh = ext4_read_inode_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; @@ -212,10 +225,10 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) /* Ok, now we can actually update the inode bitmaps.. */ if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), bit, bitmap_bh->b_data)) - ext4_error (sb, "ext4_free_inode", - "bit already cleared for inode %lu", ino); + ext4_error(sb, "ext4_free_inode", + "bit already cleared for inode %lu", ino); else { - gdp = ext4_get_group_desc (sb, block_group, &bh2); + gdp = ext4_get_group_desc(sb, block_group, &bh2); BUFFER_TRACE(bh2, "get_write_access"); fatal = ext4_journal_get_write_access(handle, bh2); @@ -277,7 +290,7 @@ static int find_group_dir(struct super_block *sb, struct inode *parent, avefreei = freei / ngroups; for (group = 0; group < ngroups; group++) { - desc = ext4_get_group_desc (sb, group, NULL); + desc = ext4_get_group_desc(sb, group, NULL); if (!desc || !desc->bg_free_inodes_count) continue; if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) @@ -341,7 +354,7 @@ find_close_to_parent: goto found_flexbg; } - if (best_flex < 0 || + if (flex_group[best_flex].free_inodes == 0 || (flex_group[i].free_blocks > flex_group[best_flex].free_blocks && flex_group[i].free_inodes)) @@ -566,16 +579,16 @@ static int find_group_other(struct super_block *sb, struct inode *parent, * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) +struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode) { struct super_block *sb; struct buffer_head *bitmap_bh = NULL; struct buffer_head *bh2; ext4_group_t group = 0; unsigned long ino = 0; - struct inode * inode; - struct ext4_group_desc * gdp = NULL; - struct ext4_super_block * es; + struct inode *inode; + struct ext4_group_desc *gdp = NULL; + struct ext4_super_block *es; struct ext4_inode_info *ei; struct ext4_sb_info *sbi; int ret2, err = 0; @@ -603,7 +616,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) } if (S_ISDIR(mode)) { - if (test_opt (sb, OLDALLOC)) + if (test_opt(sb, OLDALLOC)) ret2 = find_group_dir(sb, dir, &group); else ret2 = find_group_orlov(sb, dir, &group); @@ -623,7 +636,7 @@ got_group: goto fail; brelse(bitmap_bh); - bitmap_bh = read_inode_bitmap(sb, group); + bitmap_bh = ext4_read_inode_bitmap(sb, group); if (!bitmap_bh) goto fail; @@ -728,7 +741,7 @@ got: /* When marking the block group with * ~EXT4_BG_INODE_UNINIT we don't want to depend - * on the value of bg_itable_unsed even though + * on the value of bg_itable_unused even though * mke2fs could have initialized the same for us. * Instead we calculated the value below */ @@ -773,7 +786,7 @@ got: } inode->i_uid = current->fsuid; - if (test_opt (sb, GRPID)) + if (test_opt(sb, GRPID)) inode->i_gid = dir->i_gid; else if (dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; @@ -806,7 +819,6 @@ got: ei->i_flags &= ~EXT4_DIRSYNC_FL; ei->i_file_acl = 0; ei->i_dtime = 0; - ei->i_block_alloc_info = NULL; ei->i_block_group = group; ext4_set_inode_flags(inode); @@ -822,7 +834,7 @@ got: ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize; ret = inode; - if(DQUOT_ALLOC_INODE(inode)) { + if (DQUOT_ALLOC_INODE(inode)) { err = -EDQUOT; goto fail_drop; } @@ -831,7 +843,7 @@ got: if (err) goto fail_free_drop; - err = ext4_init_security(handle,inode, dir); + err = ext4_init_security(handle, inode, dir); if (err) goto fail_free_drop; @@ -891,7 +903,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); - bitmap_bh = read_inode_bitmap(sb, block_group); + bitmap_bh = ext4_read_inode_bitmap(sb, block_group); if (!bitmap_bh) { ext4_warning(sb, __func__, "inode bitmap error for orphan %lu", ino); @@ -949,7 +961,7 @@ error: return ERR_PTR(err); } -unsigned long ext4_count_free_inodes (struct super_block * sb) +unsigned long ext4_count_free_inodes(struct super_block *sb) { unsigned long desc_count; struct ext4_group_desc *gdp; @@ -964,12 +976,12 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) bitmap_count = 0; gdp = NULL; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - gdp = ext4_get_group_desc (sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); brelse(bitmap_bh); - bitmap_bh = read_inode_bitmap(sb, i); + bitmap_bh = ext4_read_inode_bitmap(sb, i); if (!bitmap_bh) continue; @@ -979,13 +991,14 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) bitmap_count += x; } brelse(bitmap_bh); - printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n", - le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); + printk(KERN_DEBUG "ext4_count_free_inodes: " + "stored = %u, computed = %lu, %lu\n", + le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); return desc_count; #else desc_count = 0; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - gdp = ext4_get_group_desc (sb, i, NULL); + gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; desc_count += le16_to_cpu(gdp->bg_free_inodes_count); @@ -996,13 +1009,13 @@ unsigned long ext4_count_free_inodes (struct super_block * sb) } /* Called at mount-time, super-block is locked */ -unsigned long ext4_count_dirs (struct super_block * sb) +unsigned long ext4_count_dirs(struct super_block * sb) { unsigned long count = 0; ext4_group_t i; for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { - struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL); + struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL); if (!gdp) continue; count += le16_to_cpu(gdp->bg_used_dirs_count); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8ca2763..9b4ec9d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -41,6 +41,8 @@ #include "acl.h" #include "ext4_extents.h" +#define MPAGE_DA_EXTENT_TAIL 0x01 + static inline int ext4_begin_ordered_truncate(struct inode *inode, loff_t new_size) { @@ -188,9 +190,10 @@ static int ext4_journal_test_restart(handle_t *handle, struct inode *inode) /* * Called at the last iput() if i_nlink is zero. */ -void ext4_delete_inode (struct inode * inode) +void ext4_delete_inode(struct inode *inode) { handle_t *handle; + int err; if (ext4_should_order_data(inode)) ext4_begin_ordered_truncate(inode, 0); @@ -199,8 +202,9 @@ void ext4_delete_inode (struct inode * inode) if (is_bad_inode(inode)) goto no_delete; - handle = start_transaction(inode); + handle = ext4_journal_start(inode, blocks_for_truncate(inode)+3); if (IS_ERR(handle)) { + ext4_std_error(inode->i_sb, PTR_ERR(handle)); /* * If we're going to skip the normal cleanup, we still need to * make sure that the in-core orphan linked list is properly @@ -213,8 +217,34 @@ void ext4_delete_inode (struct inode * inode) if (IS_SYNC(inode)) handle->h_sync = 1; inode->i_size = 0; + err = ext4_mark_inode_dirty(handle, inode); + if (err) { + ext4_warning(inode->i_sb, __func__, + "couldn't mark inode dirty (err %d)", err); + goto stop_handle; + } if (inode->i_blocks) ext4_truncate(inode); + + /* + * ext4_ext_truncate() doesn't reserve any slop when it + * restarts journal transactions; therefore there may not be + * enough credits left in the handle to remove the inode from + * the orphan list and set the dtime field. + */ + if (handle->h_buffer_credits < 3) { + err = ext4_journal_extend(handle, 3); + if (err > 0) + err = ext4_journal_restart(handle, 3); + if (err != 0) { + ext4_warning(inode->i_sb, __func__, + "couldn't extend journal (err %d)", err); + stop_handle: + ext4_journal_stop(handle); + goto no_delete; + } + } + /* * Kill off the orphan record which ext4_truncate created. * AKPM: I think this can be inside the above `if'. @@ -300,11 +330,11 @@ static int ext4_block_to_path(struct inode *inode, int final = 0; if (i_block < 0) { - ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0"); + ext4_warning(inode->i_sb, "ext4_block_to_path", "block < 0"); } else if (i_block < direct_blocks) { offsets[n++] = i_block; final = direct_blocks; - } else if ( (i_block -= direct_blocks) < indirect_blocks) { + } else if ((i_block -= direct_blocks) < indirect_blocks) { offsets[n++] = EXT4_IND_BLOCK; offsets[n++] = i_block; final = ptrs; @@ -370,14 +400,14 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth, *err = 0; /* i_data is not going away, no lock needed */ - add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets); + add_chain(chain, NULL, EXT4_I(inode)->i_data + *offsets); if (!p->key) goto no_block; while (--depth) { bh = sb_bread(sb, le32_to_cpu(p->key)); if (!bh) goto failure; - add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); + add_chain(++p, bh, (__le32 *)bh->b_data + *++offsets); /* Reader: end */ if (!p->key) goto no_block; @@ -413,7 +443,7 @@ no_block: static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) { struct ext4_inode_info *ei = EXT4_I(inode); - __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; + __le32 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data; __le32 *p; ext4_fsblk_t bg_start; ext4_fsblk_t last_block; @@ -456,18 +486,9 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block, Indirect *partial) { - struct ext4_block_alloc_info *block_i; - - block_i = EXT4_I(inode)->i_block_alloc_info; - /* - * try the heuristic for sequential allocation, - * failing that at least try to get decent locality. + * XXX need to get goal block from mballoc's data structures */ - if (block_i && (block == block_i->last_alloc_logical_block + 1) - && (block_i->last_alloc_physical_block != 0)) { - return block_i->last_alloc_physical_block + 1; - } return ext4_find_near(inode, partial); } @@ -600,7 +621,7 @@ allocated: *err = 0; return ret; failed_out: - for (i = 0; i <index; i++) + for (i = 0; i < index; i++) ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); return ret; } @@ -673,7 +694,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, branch[n].p = (__le32 *) bh->b_data + offsets[n]; branch[n].key = cpu_to_le32(new_blocks[n]); *branch[n].p = branch[n].key; - if ( n == indirect_blks) { + if (n == indirect_blks) { current_block = new_blocks[n]; /* * End of chain, update the last new metablock of @@ -700,7 +721,7 @@ failed: BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget"); ext4_journal_forget(handle, branch[i].bh); } - for (i = 0; i <indirect_blks; i++) + for (i = 0; i < indirect_blks; i++) ext4_free_blocks(handle, inode, new_blocks[i], 1, 0); ext4_free_blocks(handle, inode, new_blocks[i], num, 0); @@ -727,10 +748,8 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, { int i; int err = 0; - struct ext4_block_alloc_info *block_i; ext4_fsblk_t current_block; - block_i = EXT4_I(inode)->i_block_alloc_info; /* * If we're splicing into a [td]indirect block (as opposed to the * inode) then we need to get write access to the [td]indirect block @@ -753,18 +772,7 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode, if (num == 0 && blks > 1) { current_block = le32_to_cpu(where->key) + 1; for (i = 1; i < blks; i++) - *(where->p + i ) = cpu_to_le32(current_block++); - } - - /* - * update the most recently allocated logical & physical block - * in i_block_alloc_info, to assist find the proper goal block for next - * allocation - */ - if (block_i) { - block_i->last_alloc_logical_block = block + blks - 1; - block_i->last_alloc_physical_block = - le32_to_cpu(where[num].key) + blks - 1; + *(where->p + i) = cpu_to_le32(current_block++); } /* We are done with atomic stuff, now do the rest of housekeeping */ @@ -884,12 +892,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, goto cleanup; /* - * Okay, we need to do block allocation. Lazily initialize the block - * allocation info here if necessary + * Okay, we need to do block allocation. */ - if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) - ext4_init_block_alloc_info(inode); - goal = ext4_find_goal(inode, iblock, partial); /* the number of blocks need to allocate for [d,t]indirect blocks */ @@ -952,23 +956,75 @@ out: return err; } -/* Maximum number of blocks we map for direct IO at once. */ -#define DIO_MAX_BLOCKS 4096 /* - * Number of credits we need for writing DIO_MAX_BLOCKS: - * We need sb + group descriptor + bitmap + inode -> 4 - * For B blocks with A block pointers per block we need: - * 1 (triple ind.) + (B/A/A + 2) (doubly ind.) + (B/A + 2) (indirect). - * If we plug in 4096 for B and 256 for A (for 1KB block size), we get 25. + * Calculate the number of metadata blocks need to reserve + * to allocate @blocks for non extent file based file */ -#define DIO_CREDITS 25 +static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) +{ + int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); + int ind_blks, dind_blks, tind_blks; + + /* number of new indirect blocks needed */ + ind_blks = (blocks + icap - 1) / icap; + dind_blks = (ind_blks + icap - 1) / icap; + + tind_blks = 1; + + return ind_blks + dind_blks + tind_blks; +} /* + * Calculate the number of metadata blocks need to reserve + * to allocate given number of blocks + */ +static int ext4_calc_metadata_amount(struct inode *inode, int blocks) +{ + if (!blocks) + return 0; + + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_calc_metadata_amount(inode, blocks); + + return ext4_indirect_calc_metadata_amount(inode, blocks); +} + +static void ext4_da_update_reserve_space(struct inode *inode, int used) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + int total, mdb, mdb_free; + + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + /* recalculate the number of metablocks still need to be reserved */ + total = EXT4_I(inode)->i_reserved_data_blocks - used; + mdb = ext4_calc_metadata_amount(inode, total); + + /* figure out how many metablocks to release */ + BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); + mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; + + if (mdb_free) { + /* Account for allocated meta_blocks */ + mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + + /* update fs dirty blocks counter */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); + EXT4_I(inode)->i_allocated_meta_blocks = 0; + EXT4_I(inode)->i_reserved_meta_blocks = mdb; + } + + /* update per-inode reservations */ + BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); + EXT4_I(inode)->i_reserved_data_blocks -= used; + + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); +} + +/* + * The ext4_get_blocks_wrap() function try to look up the requested blocks, + * and returns if the blocks are already mapped. * - * - * ext4_ext4 get_block() wrapper function - * It will do a look up first, and returns if the blocks already mapped. * Otherwise it takes the write lock of the i_data_sem and allocate blocks * and store the allocated blocks in the result buffer head and mark it * mapped. @@ -1069,26 +1125,30 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, * which were deferred till now */ if ((retval > 0) && buffer_delay(bh)) - ext4_da_release_space(inode, retval, 0); + ext4_da_update_reserve_space(inode, retval); } up_write((&EXT4_I(inode)->i_data_sem)); return retval; } -static int ext4_get_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) +/* Maximum number of blocks we map for direct IO at once. */ +#define DIO_MAX_BLOCKS 4096 + +int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) { handle_t *handle = ext4_journal_current_handle(); int ret = 0, started = 0; unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + int dio_credits; if (create && !handle) { /* Direct IO write... */ if (max_blocks > DIO_MAX_BLOCKS) max_blocks = DIO_MAX_BLOCKS; - handle = ext4_journal_start(inode, DIO_CREDITS + - 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb)); + dio_credits = ext4_chunk_trans_blocks(inode, max_blocks); + handle = ext4_journal_start(inode, dio_credits); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; @@ -1156,7 +1216,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, BUFFER_TRACE(bh, "call get_create_access"); fatal = ext4_journal_get_create_access(handle, bh); if (!fatal && !buffer_uptodate(bh)) { - memset(bh->b_data,0,inode->i_sb->s_blocksize); + memset(bh->b_data, 0, inode->i_sb->s_blocksize); set_buffer_uptodate(bh); } unlock_buffer(bh); @@ -1181,7 +1241,7 @@ err: struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, ext4_lblk_t block, int create, int *err) { - struct buffer_head * bh; + struct buffer_head *bh; bh = ext4_getblk(handle, inode, block, create, err); if (!bh) @@ -1197,13 +1257,13 @@ struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, return NULL; } -static int walk_page_buffers( handle_t *handle, - struct buffer_head *head, - unsigned from, - unsigned to, - int *partial, - int (*fn)( handle_t *handle, - struct buffer_head *bh)) +static int walk_page_buffers(handle_t *handle, + struct buffer_head *head, + unsigned from, + unsigned to, + int *partial, + int (*fn)(handle_t *handle, + struct buffer_head *bh)) { struct buffer_head *bh; unsigned block_start, block_end; @@ -1211,9 +1271,9 @@ static int walk_page_buffers( handle_t *handle, int err, ret = 0; struct buffer_head *next; - for ( bh = head, block_start = 0; - ret == 0 && (bh != head || !block_start); - block_start = block_end, bh = next) + for (bh = head, block_start = 0; + ret == 0 && (bh != head || !block_start); + block_start = block_end, bh = next) { next = bh->b_this_page; block_end = block_start + blocksize; @@ -1266,23 +1326,23 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) { - struct inode *inode = mapping->host; + struct inode *inode = mapping->host; int ret, needed_blocks = ext4_writepage_trans_blocks(inode); handle_t *handle; int retries = 0; - struct page *page; + struct page *page; pgoff_t index; - unsigned from, to; + unsigned from, to; index = pos >> PAGE_CACHE_SHIFT; - from = pos & (PAGE_CACHE_SIZE - 1); - to = from + len; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; retry: - handle = ext4_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } page = __grab_cache_page(mapping, index); @@ -1302,9 +1362,16 @@ retry: } if (ret) { - unlock_page(page); + unlock_page(page); ext4_journal_stop(handle); - page_cache_release(page); + page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + */ + if (pos + len > inode->i_size) + vmtruncate(inode, inode->i_size); } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -1336,25 +1403,23 @@ static int ext4_ordered_write_end(struct file *file, { handle_t *handle = ext4_journal_current_handle(); struct inode *inode = mapping->host; - unsigned from, to; int ret = 0, ret2; - from = pos & (PAGE_CACHE_SIZE - 1); - to = from + len; - ret = ext4_jbd2_file_inode(handle, inode); if (ret == 0) { - /* - * generic_write_end() will run mark_inode_dirty() if i_size - * changes. So let's piggyback the i_disksize mark_inode_dirty - * into that. - */ loff_t new_i_size; new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); + } + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; @@ -1379,8 +1444,14 @@ static int ext4_writeback_write_end(struct file *file, loff_t new_i_size; new_i_size = pos + copied; - if (new_i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = new_i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); + } ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1405,6 +1476,7 @@ static int ext4_journalled_write_end(struct file *file, int ret = 0, ret2; int partial = 0; unsigned from, to; + loff_t new_i_size; from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; @@ -1419,11 +1491,12 @@ static int ext4_journalled_write_end(struct file *file, to, &partial, write_end_fn); if (!partial) SetPageUptodate(page); - if (pos+copied > inode->i_size) + new_i_size = pos + copied; + if (new_i_size > inode->i_size) i_size_write(inode, pos+copied); EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - if (inode->i_size > EXT4_I(inode)->i_disksize) { - EXT4_I(inode)->i_disksize = inode->i_size; + if (new_i_size > EXT4_I(inode)->i_disksize) { + ext4_update_i_disksize(inode, new_i_size); ret2 = ext4_mark_inode_dirty(handle, inode); if (!ret) ret = ret2; @@ -1437,39 +1510,10 @@ static int ext4_journalled_write_end(struct file *file, return ret ? ret : copied; } -/* - * Calculate the number of metadata blocks need to reserve - * to allocate @blocks for non extent file based file - */ -static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) -{ - int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); - int ind_blks, dind_blks, tind_blks; - - /* number of new indirect blocks needed */ - ind_blks = (blocks + icap - 1) / icap; - - dind_blks = (ind_blks + icap - 1) / icap; - - tind_blks = 1; - - return ind_blks + dind_blks + tind_blks; -} - -/* - * Calculate the number of metadata blocks need to reserve - * to allocate given number of blocks - */ -static int ext4_calc_metadata_amount(struct inode *inode, int blocks) -{ - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_calc_metadata_amount(inode, blocks); - - return ext4_indirect_calc_metadata_amount(inode, blocks); -} static int ext4_da_reserve_space(struct inode *inode, int nrblocks) { + int retries = 0; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); unsigned long md_needed, mdblocks, total = 0; @@ -1478,6 +1522,7 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) * in order to allocate nrblocks * worse case is one extent per block */ +repeat: spin_lock(&EXT4_I(inode)->i_block_reservation_lock); total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; mdblocks = ext4_calc_metadata_amount(inode, total); @@ -1486,14 +1531,14 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; total = md_needed + nrblocks; - if (ext4_has_free_blocks(sbi, total) < total) { + if (ext4_claim_free_blocks(sbi, total)) { spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + if (ext4_should_retry_alloc(inode->i_sb, &retries)) { + yield(); + goto repeat; + } return -ENOSPC; } - - /* reduce fs free blocks counter */ - percpu_counter_sub(&sbi->s_freeblocks_counter, total); - EXT4_I(inode)->i_reserved_data_blocks += nrblocks; EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; @@ -1501,35 +1546,49 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) return 0; /* success */ } -void ext4_da_release_space(struct inode *inode, int used, int to_free) +static void ext4_da_release_space(struct inode *inode, int to_free) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); int total, mdb, mdb_free, release; + if (!to_free) + return; /* Nothing to release, exit */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + + if (!EXT4_I(inode)->i_reserved_data_blocks) { + /* + * if there is no reserved blocks, but we try to free some + * then the counter is messed up somewhere. + * but since this function is called from invalidate + * page, it's harmless to return without any action + */ + printk(KERN_INFO "ext4 delalloc try to release %d reserved " + "blocks for inode %lu, but there is no reserved " + "data blocks\n", to_free, inode->i_ino); + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return; + } + /* recalculate the number of metablocks still need to be reserved */ - total = EXT4_I(inode)->i_reserved_data_blocks - used - to_free; + total = EXT4_I(inode)->i_reserved_data_blocks - to_free; mdb = ext4_calc_metadata_amount(inode, total); /* figure out how many metablocks to release */ BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - /* Account for allocated meta_blocks */ - mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; - release = to_free + mdb_free; - /* update fs free blocks counter for truncate case */ - percpu_counter_add(&sbi->s_freeblocks_counter, release); + /* update fs dirty blocks counter for truncate case */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, release); /* update per-inode reservations */ - BUG_ON(used + to_free > EXT4_I(inode)->i_reserved_data_blocks); - EXT4_I(inode)->i_reserved_data_blocks -= (used + to_free); + BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks); + EXT4_I(inode)->i_reserved_data_blocks -= to_free; BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); EXT4_I(inode)->i_reserved_meta_blocks = mdb; - EXT4_I(inode)->i_allocated_meta_blocks = 0; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); } @@ -1551,7 +1610,7 @@ static void ext4_da_page_release_reservation(struct page *page, } curr_off = next_off; } while ((bh = bh->b_this_page) != head); - ext4_da_release_space(page->mapping->host, 0, to_release); + ext4_da_release_space(page->mapping->host, to_release); } /* @@ -1564,11 +1623,14 @@ struct mpage_da_data { unsigned long first_page, next_page; /* extent of pages */ get_block_t *get_block; struct writeback_control *wbc; + int io_done; + long pages_written; + int retval; }; /* * mpage_da_submit_io - walks through extent of pages and try to write - * them with __mpage_writepage() + * them with writepage() call back * * @mpd->inode: inode * @mpd->first_page: first page of the extent @@ -1583,18 +1645,11 @@ struct mpage_da_data { static int mpage_da_submit_io(struct mpage_da_data *mpd) { struct address_space *mapping = mpd->inode->i_mapping; - struct mpage_data mpd_pp = { - .bio = NULL, - .last_block_in_bio = 0, - .get_block = mpd->get_block, - .use_writepage = 1, - }; int ret = 0, err, nr_pages, i; unsigned long index, end; struct pagevec pvec; BUG_ON(mpd->next_page <= mpd->first_page); - pagevec_init(&pvec, 0); index = mpd->first_page; end = mpd->next_page - 1; @@ -1612,8 +1667,9 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd) break; index++; - err = __mpage_writepage(page, mpd->wbc, &mpd_pp); - + err = mapping->a_ops->writepage(page, mpd->wbc); + if (!err) + mpd->pages_written++; /* * In error case, we have to continue because * remaining pages are still locked @@ -1624,9 +1680,6 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd) } pagevec_release(&pvec); } - if (mpd_pp.bio) - mpage_bio_submit(WRITE, mpd_pp.bio); - return ret; } @@ -1649,7 +1702,7 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, int blocks = exbh->b_size >> inode->i_blkbits; sector_t pblock = exbh->b_blocknr, cur_logical; struct buffer_head *head, *bh; - unsigned long index, end; + pgoff_t index, end; struct pagevec pvec; int nr_pages, i; @@ -1692,6 +1745,13 @@ static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, if (buffer_delay(bh)) { bh->b_blocknr = pblock; clear_buffer_delay(bh); + bh->b_bdev = inode->i_sb->s_bdev; + } else if (buffer_unwritten(bh)) { + bh->b_blocknr = pblock; + clear_buffer_unwritten(bh); + set_buffer_mapped(bh); + set_buffer_new(bh); + bh->b_bdev = inode->i_sb->s_bdev; } else if (buffer_mapped(bh)) BUG_ON(bh->b_blocknr != pblock); @@ -1719,6 +1779,57 @@ static inline void __unmap_underlying_blocks(struct inode *inode, unmap_underlying_metadata(bdev, bh->b_blocknr + i); } +static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd, + sector_t logical, long blk_cnt) +{ + int nr_pages, i; + pgoff_t index, end; + struct pagevec pvec; + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; + + index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + end = (logical + blk_cnt - 1) >> + (PAGE_CACHE_SHIFT - inode->i_blkbits); + while (index <= end) { + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + index = page->index; + if (index > end) + break; + index++; + + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + block_invalidatepage(page, 0); + ClearPageUptodate(page); + unlock_page(page); + } + } + return; +} + +static void ext4_print_free_blocks(struct inode *inode) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + printk(KERN_EMERG "Total free blocks count %lld\n", + ext4_count_free_blocks(inode->i_sb)); + printk(KERN_EMERG "Free/Dirty block details\n"); + printk(KERN_EMERG "free_blocks=%lld\n", + percpu_counter_sum(&sbi->s_freeblocks_counter)); + printk(KERN_EMERG "dirty_blocks=%lld\n", + percpu_counter_sum(&sbi->s_dirtyblocks_counter)); + printk(KERN_EMERG "Block reservation details\n"); + printk(KERN_EMERG "i_reserved_data_blocks=%lu\n", + EXT4_I(inode)->i_reserved_data_blocks); + printk(KERN_EMERG "i_reserved_meta_blocks=%lu\n", + EXT4_I(inode)->i_reserved_meta_blocks); + return; +} + /* * mpage_da_map_blocks - go through given space * @@ -1727,54 +1838,87 @@ static inline void __unmap_underlying_blocks(struct inode *inode, * * The function skips space we know is already mapped to disk blocks. * - * The function ignores errors ->get_block() returns, thus real - * error handling is postponed to __mpage_writepage() */ -static void mpage_da_map_blocks(struct mpage_da_data *mpd) +static int mpage_da_map_blocks(struct mpage_da_data *mpd) { - struct buffer_head *lbh = &mpd->lbh; - int err = 0, remain = lbh->b_size; - sector_t next = lbh->b_blocknr; + int err = 0; struct buffer_head new; + struct buffer_head *lbh = &mpd->lbh; + sector_t next; /* * We consider only non-mapped and non-allocated blocks */ if (buffer_mapped(lbh) && !buffer_delay(lbh)) - return; + return 0; + new.b_state = lbh->b_state; + new.b_blocknr = 0; + new.b_size = lbh->b_size; + next = lbh->b_blocknr; + /* + * If we didn't accumulate anything + * to write simply return + */ + if (!new.b_size) + return 0; + err = mpd->get_block(mpd->inode, next, &new, 1); + if (err) { - while (remain) { - new.b_state = lbh->b_state; - new.b_blocknr = 0; - new.b_size = remain; - err = mpd->get_block(mpd->inode, next, &new, 1); - if (err) { - /* - * Rather than implement own error handling - * here, we just leave remaining blocks - * unallocated and try again with ->writepage() - */ - break; - } - BUG_ON(new.b_size == 0); + /* If get block returns with error + * we simply return. Later writepage + * will redirty the page and writepages + * will find the dirty page again + */ + if (err == -EAGAIN) + return 0; - if (buffer_new(&new)) - __unmap_underlying_blocks(mpd->inode, &new); + if (err == -ENOSPC && + ext4_count_free_blocks(mpd->inode->i_sb)) { + mpd->retval = err; + return 0; + } /* - * If blocks are delayed marked, we need to - * put actual blocknr and drop delayed bit + * get block failure will cause us + * to loop in writepages. Because + * a_ops->writepage won't be able to + * make progress. The page will be redirtied + * by writepage and writepages will again + * try to write the same. */ - if (buffer_delay(lbh)) - mpage_put_bnr_to_bhs(mpd, next, &new); - - /* go for the remaining blocks */ - next += new.b_size >> mpd->inode->i_blkbits; - remain -= new.b_size; + printk(KERN_EMERG "%s block allocation failed for inode %lu " + "at logical offset %llu with max blocks " + "%zd with error %d\n", + __func__, mpd->inode->i_ino, + (unsigned long long)next, + lbh->b_size >> mpd->inode->i_blkbits, err); + printk(KERN_EMERG "This should not happen.!! " + "Data will be lost\n"); + if (err == -ENOSPC) { + ext4_print_free_blocks(mpd->inode); + } + /* invlaidate all the pages */ + ext4_da_block_invalidatepages(mpd, next, + lbh->b_size >> mpd->inode->i_blkbits); + return err; } + BUG_ON(new.b_size == 0); + + if (buffer_new(&new)) + __unmap_underlying_blocks(mpd->inode, &new); + + /* + * If blocks are delayed marked, we need to + * put actual blocknr and drop delayed bit + */ + if (buffer_delay(lbh) || buffer_unwritten(lbh)) + mpage_put_bnr_to_bhs(mpd, next, &new); + + return 0; } -#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | (1 << BH_Delay)) +#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | \ + (1 << BH_Delay) | (1 << BH_Unwritten)) /* * mpage_add_bh_to_extent - try to add one more block to extent of blocks @@ -1788,41 +1932,61 @@ static void mpage_da_map_blocks(struct mpage_da_data *mpd) static void mpage_add_bh_to_extent(struct mpage_da_data *mpd, sector_t logical, struct buffer_head *bh) { - struct buffer_head *lbh = &mpd->lbh; sector_t next; + size_t b_size = bh->b_size; + struct buffer_head *lbh = &mpd->lbh; + int nrblocks = lbh->b_size >> mpd->inode->i_blkbits; - next = lbh->b_blocknr + (lbh->b_size >> mpd->inode->i_blkbits); - + /* check if thereserved journal credits might overflow */ + if (!(EXT4_I(mpd->inode)->i_flags & EXT4_EXTENTS_FL)) { + if (nrblocks >= EXT4_MAX_TRANS_DATA) { + /* + * With non-extent format we are limited by the journal + * credit available. Total credit needed to insert + * nrblocks contiguous blocks is dependent on the + * nrblocks. So limit nrblocks. + */ + goto flush_it; + } else if ((nrblocks + (b_size >> mpd->inode->i_blkbits)) > + EXT4_MAX_TRANS_DATA) { + /* + * Adding the new buffer_head would make it cross the + * allowed limit for which we have journal credit + * reserved. So limit the new bh->b_size + */ + b_size = (EXT4_MAX_TRANS_DATA - nrblocks) << + mpd->inode->i_blkbits; + /* we will do mpage_da_submit_io in the next loop */ + } + } /* * First block in the extent */ if (lbh->b_size == 0) { lbh->b_blocknr = logical; - lbh->b_size = bh->b_size; + lbh->b_size = b_size; lbh->b_state = bh->b_state & BH_FLAGS; return; } + next = lbh->b_blocknr + nrblocks; /* * Can we merge the block to our big extent? */ if (logical == next && (bh->b_state & BH_FLAGS) == lbh->b_state) { - lbh->b_size += bh->b_size; + lbh->b_size += b_size; return; } +flush_it: /* * We couldn't merge the block to our extent, so we * need to flush current extent and start new one */ - mpage_da_map_blocks(mpd); - - /* - * Now start a new extent - */ - lbh->b_size = bh->b_size; - lbh->b_state = bh->b_state & BH_FLAGS; - lbh->b_blocknr = logical; + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); + mpd->io_done = 1; + return; } /* @@ -1842,17 +2006,35 @@ static int __mpage_da_writepage(struct page *page, struct buffer_head *bh, *head, fake; sector_t logical; + if (mpd->io_done) { + /* + * Rest of the page in the page_vec + * redirty then and skip then. We will + * try to to write them again after + * starting a new transaction + */ + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return MPAGE_DA_EXTENT_TAIL; + } /* * Can we merge this page to current extent? */ if (mpd->next_page != page->index) { /* * Nope, we can't. So, we map non-allocated blocks - * and start IO on them using __mpage_writepage() + * and start IO on them using writepage() */ if (mpd->next_page != mpd->first_page) { - mpage_da_map_blocks(mpd); - mpage_da_submit_io(mpd); + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); + /* + * skip rest of the page in the page_vec + */ + mpd->io_done = 1; + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return MPAGE_DA_EXTENT_TAIL; } /* @@ -1883,6 +2065,8 @@ static int __mpage_da_writepage(struct page *page, set_buffer_dirty(bh); set_buffer_uptodate(bh); mpage_add_bh_to_extent(mpd, logical, bh); + if (mpd->io_done) + return MPAGE_DA_EXTENT_TAIL; } else { /* * Page with regular buffer heads, just add all dirty ones @@ -1891,8 +2075,12 @@ static int __mpage_da_writepage(struct page *page, bh = head; do { BUG_ON(buffer_locked(bh)); - if (buffer_dirty(bh)) + if (buffer_dirty(bh) && + (!buffer_mapped(bh) || buffer_delay(bh))) { mpage_add_bh_to_extent(mpd, logical, bh); + if (mpd->io_done) + return MPAGE_DA_EXTENT_TAIL; + } logical++; } while ((bh = bh->b_this_page) != head); } @@ -1911,46 +2099,39 @@ static int __mpage_da_writepage(struct page *page, * * This is a library function, which implements the writepages() * address_space_operation. - * - * In order to avoid duplication of logic that deals with partial pages, - * multiple bio per page, etc, we find non-allocated blocks, allocate - * them with minimal calls to ->get_block() and re-use __mpage_writepage() - * - * It's important that we call __mpage_writepage() only once for each - * involved page, otherwise we'd have to implement more complicated logic - * to deal with pages w/o PG_lock or w/ PG_writeback and so on. - * - * See comments to mpage_writepages() */ static int mpage_da_writepages(struct address_space *mapping, struct writeback_control *wbc, - get_block_t get_block) + struct mpage_da_data *mpd) { - struct mpage_da_data mpd; + long to_write; int ret; - if (!get_block) + if (!mpd->get_block) return generic_writepages(mapping, wbc); - mpd.wbc = wbc; - mpd.inode = mapping->host; - mpd.lbh.b_size = 0; - mpd.lbh.b_state = 0; - mpd.lbh.b_blocknr = 0; - mpd.first_page = 0; - mpd.next_page = 0; - mpd.get_block = get_block; + mpd->lbh.b_size = 0; + mpd->lbh.b_state = 0; + mpd->lbh.b_blocknr = 0; + mpd->first_page = 0; + mpd->next_page = 0; + mpd->io_done = 0; + mpd->pages_written = 0; + mpd->retval = 0; - ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd); + to_write = wbc->nr_to_write; + + ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, mpd); /* * Handle last extent of pages */ - if (mpd.next_page != mpd.first_page) { - mpage_da_map_blocks(&mpd); - mpage_da_submit_io(&mpd); + if (!mpd->io_done && mpd->next_page != mpd->first_page) { + if (mpage_da_map_blocks(mpd) == 0) + mpage_da_submit_io(mpd); } + wbc->nr_to_write = to_write - mpd->pages_written; return ret; } @@ -2003,18 +2184,24 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, handle_t *handle = NULL; handle = ext4_journal_current_handle(); - if (!handle) { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, 0, 0, 0); - BUG_ON(!ret); - } else { - ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, - bh_result, create, 0, EXT4_DELALLOC_RSVED); - } - + BUG_ON(!handle); + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, create, 0, EXT4_DELALLOC_RSVED); if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + if (ext4_should_order_data(inode)) { + int retval; + retval = ext4_jbd2_file_inode(handle, inode); + if (retval) + /* + * Failed to add inode for ordered + * mode. Don't update file size + */ + return retval; + } + /* * Update on-disk size along with block allocation * we don't use 'extend_disksize' as size may change @@ -2024,18 +2211,9 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, if (disksize > i_size_read(inode)) disksize = i_size_read(inode); if (disksize > EXT4_I(inode)->i_disksize) { - /* - * XXX: replace with spinlock if seen contended -bzzz - */ - down_write(&EXT4_I(inode)->i_data_sem); - if (disksize > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = disksize; - up_write(&EXT4_I(inode)->i_data_sem); - - if (EXT4_I(inode)->i_disksize == disksize) { - ret = ext4_mark_inode_dirty(handle, inode); - return ret; - } + ext4_update_i_disksize(inode, disksize); + ret = ext4_mark_inode_dirty(handle, inode); + return ret; } ret = 0; } @@ -2155,84 +2333,114 @@ static int ext4_da_writepage(struct page *page, } /* - * For now just follow the DIO way to estimate the max credits - * needed to write out EXT4_MAX_WRITEBACK_PAGES. - * todo: need to calculate the max credits need for - * extent based files, currently the DIO credits is based on - * indirect-blocks mapping way. - * - * Probably should have a generic way to calculate credits - * for DIO, writepages, and truncate + * This is called via ext4_da_writepages() to + * calulate the total number of credits to reserve to fit + * a single extent allocation into a single transaction, + * ext4_da_writpeages() will loop calling this before + * the block allocation. */ -#define EXT4_MAX_WRITEBACK_PAGES DIO_MAX_BLOCKS -#define EXT4_MAX_WRITEBACK_CREDITS DIO_CREDITS + +static int ext4_da_writepages_trans_blocks(struct inode *inode) +{ + int max_blocks = EXT4_I(inode)->i_reserved_data_blocks; + + /* + * With non-extent format the journal credit needed to + * insert nrblocks contiguous block is dependent on + * number of contiguous block. So we will limit + * number of contiguous block to a sane value + */ + if (!(inode->i_flags & EXT4_EXTENTS_FL) && + (max_blocks > EXT4_MAX_TRANS_DATA)) + max_blocks = EXT4_MAX_TRANS_DATA; + + return ext4_chunk_trans_blocks(inode, max_blocks); +} static int ext4_da_writepages(struct address_space *mapping, - struct writeback_control *wbc) + struct writeback_control *wbc) { - struct inode *inode = mapping->host; handle_t *handle = NULL; - int needed_blocks; - int ret = 0; - long to_write; loff_t range_start = 0; + struct mpage_da_data mpd; + struct inode *inode = mapping->host; + int needed_blocks, ret = 0, nr_to_writebump = 0; + long to_write, pages_skipped = 0; + struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb); /* * No pages to write? This is mainly a kludge to avoid starting * a transaction for special inodes like journal inode on last iput() * because that could violate lock ordering on umount */ - if (!mapping->nrpages) + if (!mapping->nrpages || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) return 0; - /* - * Estimate the worse case needed credits to write out - * EXT4_MAX_BUF_BLOCKS pages + * Make sure nr_to_write is >= sbi->s_mb_stream_request + * This make sure small files blocks are allocated in + * single attempt. This ensure that small files + * get less fragmented. */ - needed_blocks = EXT4_MAX_WRITEBACK_CREDITS; + if (wbc->nr_to_write < sbi->s_mb_stream_request) { + nr_to_writebump = sbi->s_mb_stream_request - wbc->nr_to_write; + wbc->nr_to_write = sbi->s_mb_stream_request; + } - to_write = wbc->nr_to_write; - if (!wbc->range_cyclic) { + if (!wbc->range_cyclic) /* * If range_cyclic is not set force range_cont * and save the old writeback_index */ wbc->range_cont = 1; - range_start = wbc->range_start; - } - while (!ret && to_write) { + range_start = wbc->range_start; + pages_skipped = wbc->pages_skipped; + + mpd.wbc = wbc; + mpd.inode = mapping->host; + +restart_loop: + to_write = wbc->nr_to_write; + while (!ret && to_write > 0) { + + /* + * we insert one extent at a time. So we need + * credit needed for single extent allocation. + * journalled mode is currently not supported + * by delalloc + */ + BUG_ON(ext4_should_journal_data(inode)); + needed_blocks = ext4_da_writepages_trans_blocks(inode); + /* start a new transaction*/ handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { ret = PTR_ERR(handle); + printk(KERN_EMERG "%s: jbd2_start: " + "%ld pages, ino %lu; err %d\n", __func__, + wbc->nr_to_write, inode->i_ino, ret); + dump_stack(); goto out_writepages; } - if (ext4_should_order_data(inode)) { - /* - * With ordered mode we need to add - * the inode to the journal handle - * when we do block allocation. - */ - ret = ext4_jbd2_file_inode(handle, inode); - if (ret) { - ext4_journal_stop(handle); - goto out_writepages; - } + to_write -= wbc->nr_to_write; - } - /* - * set the max dirty pages could be write at a time - * to fit into the reserved transaction credits - */ - if (wbc->nr_to_write > EXT4_MAX_WRITEBACK_PAGES) - wbc->nr_to_write = EXT4_MAX_WRITEBACK_PAGES; + mpd.get_block = ext4_da_get_block_write; + ret = mpage_da_writepages(mapping, wbc, &mpd); - to_write -= wbc->nr_to_write; - ret = mpage_da_writepages(mapping, wbc, - ext4_da_get_block_write); ext4_journal_stop(handle); - if (wbc->nr_to_write) { + + if (mpd.retval == -ENOSPC) + jbd2_journal_force_commit_nested(sbi->s_journal); + + /* reset the retry count */ + if (ret == MPAGE_DA_EXTENT_TAIL) { + /* + * got one extent now try with + * rest of the pages + */ + to_write += wbc->nr_to_write; + ret = 0; + } else if (wbc->nr_to_write) { /* * There is no more writeout needed * or we requested for a noblocking writeout @@ -2244,13 +2452,48 @@ static int ext4_da_writepages(struct address_space *mapping, wbc->nr_to_write = to_write; } -out_writepages: - wbc->nr_to_write = to_write; - if (range_start) + if (wbc->range_cont && (pages_skipped != wbc->pages_skipped)) { + /* We skipped pages in this loop */ wbc->range_start = range_start; + wbc->nr_to_write = to_write + + wbc->pages_skipped - pages_skipped; + wbc->pages_skipped = pages_skipped; + goto restart_loop; + } + +out_writepages: + wbc->nr_to_write = to_write - nr_to_writebump; + wbc->range_start = range_start; return ret; } +#define FALL_BACK_TO_NONDELALLOC 1 +static int ext4_nonda_switch(struct super_block *sb) +{ + s64 free_blocks, dirty_blocks; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + /* + * switch to non delalloc mode if we are running low + * on free block. The free block accounting via percpu + * counters can get slightly wrong with FBC_BATCH getting + * accumulated on each CPU without updating global counters + * Delalloc need an accurate free block accounting. So switch + * to non delalloc when we are near to error range. + */ + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + dirty_blocks = percpu_counter_read_positive(&sbi->s_dirtyblocks_counter); + if (2 * free_blocks < 3 * dirty_blocks || + free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) { + /* + * free block count is less that 150% of dirty blocks + * or free blocks is less that watermark + */ + return 1; + } + return 0; +} + static int ext4_da_write_begin(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned flags, struct page **pagep, void **fsdata) @@ -2266,6 +2509,12 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping, from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; + if (ext4_nonda_switch(inode->i_sb)) { + *fsdata = (void *)FALL_BACK_TO_NONDELALLOC; + return ext4_write_begin(file, mapping, pos, + len, flags, pagep, fsdata); + } + *fsdata = (void *)0; retry: /* * With delayed allocation, we don't log the i_disksize update @@ -2280,8 +2529,11 @@ retry: } page = __grab_cache_page(mapping, index); - if (!page) - return -ENOMEM; + if (!page) { + ext4_journal_stop(handle); + ret = -ENOMEM; + goto out; + } *pagep = page; ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, @@ -2290,6 +2542,13 @@ retry: unlock_page(page); ext4_journal_stop(handle); page_cache_release(page); + /* + * block_write_begin may have instantiated a few blocks + * outside i_size. Trim these off again. Don't need + * i_size_read because we hold i_mutex. + */ + if (pos + len > inode->i_size) + vmtruncate(inode, inode->i_size); } if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) @@ -2313,7 +2572,7 @@ static int ext4_da_should_update_i_disksize(struct page *page, bh = page_buffers(page); idx = offset >> inode->i_blkbits; - for (i=0; i < idx; i++) + for (i = 0; i < idx; i++) bh = bh->b_this_page; if (!buffer_mapped(bh) || (buffer_delay(bh))) @@ -2331,9 +2590,22 @@ static int ext4_da_write_end(struct file *file, handle_t *handle = ext4_journal_current_handle(); loff_t new_i_size; unsigned long start, end; + int write_mode = (int)(unsigned long)fsdata; + + if (write_mode == FALL_BACK_TO_NONDELALLOC) { + if (ext4_should_order_data(inode)) { + return ext4_ordered_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else if (ext4_should_writeback_data(inode)) { + return ext4_writeback_write_end(file, mapping, pos, + len, copied, page, fsdata); + } else { + BUG(); + } + } start = pos & (PAGE_CACHE_SIZE - 1); - end = start + copied -1; + end = start + copied - 1; /* * generic_write_end() will run mark_inode_dirty() if i_size @@ -2357,6 +2629,11 @@ static int ext4_da_write_end(struct file *file, EXT4_I(inode)->i_disksize = new_i_size; } up_write(&EXT4_I(inode)->i_data_sem); + /* We need to mark inode dirty even if + * new_i_size is less that inode->i_size + * bu greater than i_disksize.(hint delalloc) + */ + ext4_mark_inode_dirty(handle, inode); } } ret2 = generic_write_end(file, mapping, pos, len, copied, @@ -2448,7 +2725,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) return 0; } - return generic_block_bmap(mapping,block,ext4_get_block); + return generic_block_bmap(mapping, block, ext4_get_block); } static int bget_one(handle_t *handle, struct buffer_head *bh) @@ -2806,59 +3083,63 @@ static int ext4_journalled_set_page_dirty(struct page *page) } static const struct address_space_operations ext4_ordered_aops = { - .readpage = ext4_readpage, - .readpages = ext4_readpages, - .writepage = ext4_normal_writepage, - .sync_page = block_sync_page, - .write_begin = ext4_write_begin, - .write_end = ext4_ordered_write_end, - .bmap = ext4_bmap, - .invalidatepage = ext4_invalidatepage, - .releasepage = ext4_releasepage, - .direct_IO = ext4_direct_IO, - .migratepage = buffer_migrate_page, + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_normal_writepage, + .sync_page = block_sync_page, + .write_begin = ext4_write_begin, + .write_end = ext4_ordered_write_end, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; static const struct address_space_operations ext4_writeback_aops = { - .readpage = ext4_readpage, - .readpages = ext4_readpages, - .writepage = ext4_normal_writepage, - .sync_page = block_sync_page, - .write_begin = ext4_write_begin, - .write_end = ext4_writeback_write_end, - .bmap = ext4_bmap, - .invalidatepage = ext4_invalidatepage, - .releasepage = ext4_releasepage, - .direct_IO = ext4_direct_IO, - .migratepage = buffer_migrate_page, + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_normal_writepage, + .sync_page = block_sync_page, + .write_begin = ext4_write_begin, + .write_end = ext4_writeback_write_end, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; static const struct address_space_operations ext4_journalled_aops = { - .readpage = ext4_readpage, - .readpages = ext4_readpages, - .writepage = ext4_journalled_writepage, - .sync_page = block_sync_page, - .write_begin = ext4_write_begin, - .write_end = ext4_journalled_write_end, - .set_page_dirty = ext4_journalled_set_page_dirty, - .bmap = ext4_bmap, - .invalidatepage = ext4_invalidatepage, - .releasepage = ext4_releasepage, + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_journalled_writepage, + .sync_page = block_sync_page, + .write_begin = ext4_write_begin, + .write_end = ext4_journalled_write_end, + .set_page_dirty = ext4_journalled_set_page_dirty, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .is_partially_uptodate = block_is_partially_uptodate, }; static const struct address_space_operations ext4_da_aops = { - .readpage = ext4_readpage, - .readpages = ext4_readpages, - .writepage = ext4_da_writepage, - .writepages = ext4_da_writepages, - .sync_page = block_sync_page, - .write_begin = ext4_da_write_begin, - .write_end = ext4_da_write_end, - .bmap = ext4_bmap, - .invalidatepage = ext4_da_invalidatepage, - .releasepage = ext4_releasepage, - .direct_IO = ext4_direct_IO, - .migratepage = buffer_migrate_page, + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_da_writepage, + .writepages = ext4_da_writepages, + .sync_page = block_sync_page, + .write_begin = ext4_da_write_begin, + .write_end = ext4_da_write_end, + .bmap = ext4_bmap, + .invalidatepage = ext4_da_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, }; void ext4_set_aops(struct inode *inode) @@ -3050,7 +3331,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, if (!partial->key && *partial->p) /* Writer: end */ goto no_top; - for (p=partial; p>chain && all_zeroes((__le32*)p->bh->b_data,p->p); p--) + for (p = partial; (p > chain) && all_zeroes((__le32 *) p->bh->b_data, p->p); p--) ; /* * OK, we've found the last block that must survive. The rest of our @@ -3069,7 +3350,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth, } /* Writer: end */ - while(partial > p) { + while (partial > p) { brelse(partial->bh); partial--; } @@ -3261,9 +3542,9 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, /* This zaps the entire block. Bottom up. */ BUFFER_TRACE(bh, "free child branches"); ext4_free_branches(handle, inode, bh, - (__le32*)bh->b_data, - (__le32*)bh->b_data + addr_per_block, - depth); + (__le32 *) bh->b_data, + (__le32 *) bh->b_data + addr_per_block, + depth); /* * We've probably journalled the indirect block several @@ -3430,6 +3711,9 @@ void ext4_truncate(struct inode *inode) * modify the block allocation tree. */ down_write(&ei->i_data_sem); + + ext4_discard_preallocations(inode); + /* * The orphan list entry will now protect us from any crash which * occurs before the truncate completes, so it is now safe to propagate @@ -3499,8 +3783,6 @@ do_indirects: ; } - ext4_discard_reservation(inode); - up_write(&ei->i_data_sem); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); @@ -3525,41 +3807,6 @@ out_stop: ext4_journal_stop(handle); } -static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, - unsigned long ino, struct ext4_iloc *iloc) -{ - ext4_group_t block_group; - unsigned long offset; - ext4_fsblk_t block; - struct ext4_group_desc *gdp; - - if (!ext4_valid_inum(sb, ino)) { - /* - * This error is already checked for in namei.c unless we are - * looking at an NFS filehandle, in which case no error - * report is needed - */ - return 0; - } - - block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); - gdp = ext4_get_group_desc(sb, block_group, NULL); - if (!gdp) - return 0; - - /* - * Figure out the offset within the block group inode table - */ - offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * - EXT4_INODE_SIZE(sb); - block = ext4_inode_table(sb, gdp) + - (offset >> EXT4_BLOCK_SIZE_BITS(sb)); - - iloc->block_group = block_group; - iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); - return block; -} - /* * ext4_get_inode_loc returns with an extra refcount against the inode's * underlying buffer_head on success. If 'in_mem' is true, we have all @@ -3569,23 +3816,49 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, static int __ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc, int in_mem) { - ext4_fsblk_t block; - struct buffer_head *bh; + struct ext4_group_desc *gdp; + struct buffer_head *bh; + struct super_block *sb = inode->i_sb; + ext4_fsblk_t block; + int inodes_per_block, inode_offset; + + iloc->bh = 0; + if (!ext4_valid_inum(sb, inode->i_ino)) + return -EIO; - block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc); - if (!block) + iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb); + gdp = ext4_get_group_desc(sb, iloc->block_group, NULL); + if (!gdp) return -EIO; - bh = sb_getblk(inode->i_sb, block); + /* + * Figure out the offset within the block group inode table + */ + inodes_per_block = (EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb)); + inode_offset = ((inode->i_ino - 1) % + EXT4_INODES_PER_GROUP(sb)); + block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block); + iloc->offset = (inode_offset % inodes_per_block) * EXT4_INODE_SIZE(sb); + + bh = sb_getblk(sb, block); if (!bh) { - ext4_error (inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, "ext4_get_inode_loc", "unable to read " + "inode block - inode=%lu, block=%llu", + inode->i_ino, block); return -EIO; } if (!buffer_uptodate(bh)) { lock_buffer(bh); + + /* + * If the buffer has the write error flag, we have failed + * to write out another inode in the same block. In this + * case, we don't have to read the block because we may + * read the old inode data successfully. + */ + if (buffer_write_io_error(bh) && !buffer_uptodate(bh)) + set_buffer_uptodate(bh); + if (buffer_uptodate(bh)) { /* someone brought it uptodate while we waited */ unlock_buffer(bh); @@ -3599,28 +3872,12 @@ static int __ext4_get_inode_loc(struct inode *inode, */ if (in_mem) { struct buffer_head *bitmap_bh; - struct ext4_group_desc *desc; - int inodes_per_buffer; - int inode_offset, i; - ext4_group_t block_group; - int start; - - block_group = (inode->i_ino - 1) / - EXT4_INODES_PER_GROUP(inode->i_sb); - inodes_per_buffer = bh->b_size / - EXT4_INODE_SIZE(inode->i_sb); - inode_offset = ((inode->i_ino - 1) % - EXT4_INODES_PER_GROUP(inode->i_sb)); - start = inode_offset & ~(inodes_per_buffer - 1); + int i, start; - /* Is the inode bitmap in cache? */ - desc = ext4_get_group_desc(inode->i_sb, - block_group, NULL); - if (!desc) - goto make_io; + start = inode_offset & ~(inodes_per_block - 1); - bitmap_bh = sb_getblk(inode->i_sb, - ext4_inode_bitmap(inode->i_sb, desc)); + /* Is the inode bitmap in cache? */ + bitmap_bh = sb_getblk(sb, ext4_inode_bitmap(sb, gdp)); if (!bitmap_bh) goto make_io; @@ -3633,14 +3890,14 @@ static int __ext4_get_inode_loc(struct inode *inode, brelse(bitmap_bh); goto make_io; } - for (i = start; i < start + inodes_per_buffer; i++) { + for (i = start; i < start + inodes_per_block; i++) { if (i == inode_offset) continue; if (ext4_test_bit(i, bitmap_bh->b_data)) break; } brelse(bitmap_bh); - if (i == start + inodes_per_buffer) { + if (i == start + inodes_per_block) { /* all other inodes are free, so skip I/O */ memset(bh->b_data, 0, bh->b_size); set_buffer_uptodate(bh); @@ -3651,6 +3908,36 @@ static int __ext4_get_inode_loc(struct inode *inode, make_io: /* + * If we need to do any I/O, try to pre-readahead extra + * blocks from the inode table. + */ + if (EXT4_SB(sb)->s_inode_readahead_blks) { + ext4_fsblk_t b, end, table; + unsigned num; + + table = ext4_inode_table(sb, gdp); + /* Make sure s_inode_readahead_blks is a power of 2 */ + while (EXT4_SB(sb)->s_inode_readahead_blks & + (EXT4_SB(sb)->s_inode_readahead_blks-1)) + EXT4_SB(sb)->s_inode_readahead_blks = + (EXT4_SB(sb)->s_inode_readahead_blks & + (EXT4_SB(sb)->s_inode_readahead_blks-1)); + b = block & ~(EXT4_SB(sb)->s_inode_readahead_blks-1); + if (table > b) + b = table; + end = b + EXT4_SB(sb)->s_inode_readahead_blks; + num = EXT4_INODES_PER_GROUP(sb); + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_GDT_CSUM)) + num -= le16_to_cpu(gdp->bg_itable_unused); + table += num / inodes_per_block; + if (end > table) + end = table; + while (b <= end) + sb_breadahead(sb, b++); + } + + /* * There are other valid inodes in the buffer, this inode * has in-inode xattrs, or we don't have this inode in memory. * Read the block from disk. @@ -3660,10 +3947,9 @@ make_io: submit_bh(READ_META, bh); wait_on_buffer(bh); if (!buffer_uptodate(bh)) { - ext4_error(inode->i_sb, "ext4_get_inode_loc", - "unable to read inode block - " - "inode=%lu, block=%llu", - inode->i_ino, block); + ext4_error(sb, __func__, + "unable to read inode block - inode=%lu, " + "block=%llu", inode->i_ino, block); brelse(bh); return -EIO; } @@ -3755,11 +4041,10 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) return inode; ei = EXT4_I(inode); -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL ei->i_acl = EXT4_ACL_NOT_CACHED; ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif - ei->i_block_alloc_info = NULL; ret = __ext4_get_inode_loc(inode, &iloc, 0); if (ret < 0) @@ -3769,7 +4054,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) inode->i_mode = le16_to_cpu(raw_inode->i_mode); inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); - if(!(test_opt (inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; } @@ -3787,7 +4072,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) if (inode->i_mode == 0 || !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { /* this inode is deleted */ - brelse (bh); + brelse(bh); ret = -ESTALE; goto bad_inode; } @@ -3820,7 +4105,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > EXT4_INODE_SIZE(inode->i_sb)) { - brelse (bh); + brelse(bh); ret = -EIO; goto bad_inode; } @@ -3873,7 +4158,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) init_special_inode(inode, inode->i_mode, new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); } - brelse (iloc.bh); + brelse(iloc.bh); ext4_set_inode_flags(inode); unlock_new_inode(inode); return inode; @@ -3955,14 +4240,14 @@ static int ext4_do_update_inode(handle_t *handle, ext4_get_inode_flags(ei); raw_inode->i_mode = cpu_to_le16(inode->i_mode); - if(!(test_opt(inode->i_sb, NO_UID32))) { + if (!(test_opt(inode->i_sb, NO_UID32))) { raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid)); raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid)); /* * Fix up interoperability with old kernels. Otherwise, old inodes get * re-used with the upper 16 bits of the uid/gid intact */ - if(!ei->i_dtime) { + if (!ei->i_dtime) { raw_inode->i_uid_high = cpu_to_le16(high_16_bits(inode->i_uid)); raw_inode->i_gid_high = @@ -4050,7 +4335,7 @@ static int ext4_do_update_inode(handle_t *handle, ei->i_state &= ~EXT4_STATE_NEW; out_brelse: - brelse (bh); + brelse(bh); ext4_std_error(inode->i_sb, err); return err; } @@ -4258,57 +4543,129 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, return 0; } +static int ext4_indirect_trans_blocks(struct inode *inode, int nrblocks, + int chunk) +{ + int indirects; + + /* if nrblocks are contiguous */ + if (chunk) { + /* + * With N contiguous data blocks, it need at most + * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) indirect blocks + * 2 dindirect blocks + * 1 tindirect block + */ + indirects = nrblocks / EXT4_ADDR_PER_BLOCK(inode->i_sb); + return indirects + 3; + } + /* + * if nrblocks are not contiguous, worse case, each block touch + * a indirect block, and each indirect block touch a double indirect + * block, plus a triple indirect block + */ + indirects = nrblocks * 2 + 1; + return indirects; +} + +static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) +{ + if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)) + return ext4_indirect_trans_blocks(inode, nrblocks, 0); + return ext4_ext_index_trans_blocks(inode, nrblocks, 0); +} /* - * How many blocks doth make a writepage()? - * - * With N blocks per page, it may be: - * N data blocks - * 2 indirect block - * 2 dindirect - * 1 tindirect - * N+5 bitmap blocks (from the above) - * N+5 group descriptor summary blocks - * 1 inode block - * 1 superblock. - * 2 * EXT4_SINGLEDATA_TRANS_BLOCKS for the quote files + * Account for index blocks, block groups bitmaps and block group + * descriptor blocks if modify datablocks and index blocks + * worse case, the indexs blocks spread over different block groups * - * 3 * (N + 5) + 2 + 2 * EXT4_SINGLEDATA_TRANS_BLOCKS + * If datablocks are discontiguous, they are possible to spread over + * different block groups too. If they are contiugous, with flexbg, + * they could still across block group boundary. * - * With ordered or writeback data it's the same, less the N data blocks. + * Also account for superblock, inode, quota and xattr blocks + */ +int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) +{ + int groups, gdpblocks; + int idxblocks; + int ret = 0; + + /* + * How many index blocks need to touch to modify nrblocks? + * The "Chunk" flag indicating whether the nrblocks is + * physically contiguous on disk + * + * For Direct IO and fallocate, they calls get_block to allocate + * one single extent at a time, so they could set the "Chunk" flag + */ + idxblocks = ext4_index_trans_blocks(inode, nrblocks, chunk); + + ret = idxblocks; + + /* + * Now let's see how many group bitmaps and group descriptors need + * to account + */ + groups = idxblocks; + if (chunk) + groups += 1; + else + groups += nrblocks; + + gdpblocks = groups; + if (groups > EXT4_SB(inode->i_sb)->s_groups_count) + groups = EXT4_SB(inode->i_sb)->s_groups_count; + if (groups > EXT4_SB(inode->i_sb)->s_gdb_count) + gdpblocks = EXT4_SB(inode->i_sb)->s_gdb_count; + + /* bitmaps and block group descriptor blocks */ + ret += groups + gdpblocks; + + /* Blocks for super block, inode, quota and xattr blocks */ + ret += EXT4_META_TRANS_BLOCKS(inode->i_sb); + + return ret; +} + +/* + * Calulate the total number of credits to reserve to fit + * the modification of a single pages into a single transaction, + * which may include multiple chunks of block allocations. * - * If the inode's direct blocks can hold an integral number of pages then a - * page cannot straddle two indirect blocks, and we can only touch one indirect - * and dindirect block, and the "5" above becomes "3". + * This could be called via ext4_write_begin() * - * This still overestimates under most circumstances. If we were to pass the - * start and end offsets in here as well we could do block_to_path() on each - * block and work out the exact number of indirects which are touched. Pah. + * We need to consider the worse case, when + * one new block per extent. */ - int ext4_writepage_trans_blocks(struct inode *inode) { int bpp = ext4_journal_blocks_per_page(inode); - int indirects = (EXT4_NDIR_BLOCKS % bpp) ? 5 : 3; int ret; - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_writepage_trans_blocks(inode, bpp); + ret = ext4_meta_trans_blocks(inode, bpp, 0); + /* Account for data blocks for journalled mode */ if (ext4_should_journal_data(inode)) - ret = 3 * (bpp + indirects) + 2; - else - ret = 2 * (bpp + indirects) + 2; - -#ifdef CONFIG_QUOTA - /* We know that structure was already allocated during DQUOT_INIT so - * we will be updating only the data blocks + inodes */ - ret += 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); -#endif - + ret += bpp; return ret; } /* + * Calculate the journal credits for a chunk of data modification. + * + * This is called from DIO, fallocate or whoever calling + * ext4_get_blocks_wrap() to map/allocate a chunk of contigous disk blocks. + * + * journal buffers for data blocks are not included here, as DIO + * and fallocate do no need to journal data buffers. + */ +int ext4_chunk_trans_blocks(struct inode *inode, int nrblocks) +{ + return ext4_meta_trans_blocks(inode, nrblocks, 1); +} + +/* * The caller must have previously called ext4_reserve_inode_write(). * Give this, we know that the caller already has write access to iloc->bh. */ @@ -4581,6 +4938,7 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) loff_t size; unsigned long len; int ret = -EINVAL; + void *fsdata; struct file *file = vma->vm_file; struct inode *inode = file->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; @@ -4619,11 +4977,11 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) * on the same page though */ ret = mapping->a_ops->write_begin(file, mapping, page_offset(page), - len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); + len, AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata); if (ret < 0) goto out_unlock; ret = mapping->a_ops->write_end(file, mapping, page_offset(page), - len, len, page, NULL); + len, len, page, fsdata); if (ret < 0) goto out_unlock; ret = 0; diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 7a6c2f1..dc99b47 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -23,9 +23,8 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) struct inode *inode = filp->f_dentry->d_inode; struct ext4_inode_info *ei = EXT4_I(inode); unsigned int flags; - unsigned short rsv_window_size; - ext4_debug ("cmd = %u, arg = %lu\n", cmd, arg); + ext4_debug("cmd = %u, arg = %lu\n", cmd, arg); switch (cmd) { case EXT4_IOC_GETFLAGS: @@ -34,7 +33,7 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return put_user(flags, (int __user *) arg); case EXT4_IOC_SETFLAGS: { handle_t *handle = NULL; - int err; + int err, migrate = 0; struct ext4_iloc iloc; unsigned int oldflags; unsigned int jflag; @@ -82,6 +81,17 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (!capable(CAP_SYS_RESOURCE)) goto flags_out; } + if (oldflags & EXT4_EXTENTS_FL) { + /* We don't support clearning extent flags */ + if (!(flags & EXT4_EXTENTS_FL)) { + err = -EOPNOTSUPP; + goto flags_out; + } + } else if (flags & EXT4_EXTENTS_FL) { + /* migrate the file */ + migrate = 1; + flags &= ~EXT4_EXTENTS_FL; + } handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) { @@ -109,6 +119,10 @@ flags_err: if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) err = ext4_change_inode_journal_flag(inode, jflag); + if (err) + goto flags_out; + if (migrate) + err = ext4_ext_migrate(inode); flags_out: mutex_unlock(&inode->i_mutex); mnt_drop_write(filp->f_path.mnt); @@ -175,53 +189,10 @@ setversion_out: return ret; } #endif - case EXT4_IOC_GETRSVSZ: - if (test_opt(inode->i_sb, RESERVATION) - && S_ISREG(inode->i_mode) - && ei->i_block_alloc_info) { - rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size; - return put_user(rsv_window_size, (int __user *)arg); - } - return -ENOTTY; - case EXT4_IOC_SETRSVSZ: { - int err; - - if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) - return -ENOTTY; - - if (!is_owner_or_cap(inode)) - return -EACCES; - - if (get_user(rsv_window_size, (int __user *)arg)) - return -EFAULT; - - err = mnt_want_write(filp->f_path.mnt); - if (err) - return err; - - if (rsv_window_size > EXT4_MAX_RESERVE_BLOCKS) - rsv_window_size = EXT4_MAX_RESERVE_BLOCKS; - - /* - * need to allocate reservation structure for this inode - * before set the window size - */ - down_write(&ei->i_data_sem); - if (!ei->i_block_alloc_info) - ext4_init_block_alloc_info(inode); - - if (ei->i_block_alloc_info){ - struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; - rsv->rsv_goal_size = rsv_window_size; - } - up_write(&ei->i_data_sem); - mnt_drop_write(filp->f_path.mnt); - return 0; - } case EXT4_IOC_GROUP_EXTEND: { ext4_fsblk_t n_blocks_count; struct super_block *sb = inode->i_sb; - int err; + int err, err2; if (!capable(CAP_SYS_RESOURCE)) return -EPERM; @@ -235,8 +206,10 @@ setversion_out: err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err == 0) + err = err2; mnt_drop_write(filp->f_path.mnt); return err; @@ -244,7 +217,7 @@ setversion_out: case EXT4_IOC_GROUP_ADD: { struct ext4_new_group_data input; struct super_block *sb = inode->i_sb; - int err; + int err, err2; if (!capable(CAP_SYS_RESOURCE)) return -EPERM; @@ -259,15 +232,36 @@ setversion_out: err = ext4_group_add(sb, &input); jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err == 0) + err = err2; mnt_drop_write(filp->f_path.mnt); return err; } case EXT4_IOC_MIGRATE: - return ext4_ext_migrate(inode, filp, cmd, arg); + { + int err; + if (!is_owner_or_cap(inode)) + return -EACCES; + + err = mnt_want_write(filp->f_path.mnt); + if (err) + return err; + /* + * inode_mutex prevent write and truncate on the file. + * Read still goes through. We take i_data_sem in + * ext4_ext_swap_inode_data before we switch the + * inode format to prevent read. + */ + mutex_lock(&(inode->i_mutex)); + err = ext4_ext_migrate(inode); + mutex_unlock(&(inode->i_mutex)); + mnt_drop_write(filp->f_path.mnt); + return err; + } default: return -ENOTTY; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 8d141a2..b580714 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -477,9 +477,10 @@ static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap) b2 = (unsigned char *) bitmap; for (i = 0; i < e4b->bd_sb->s_blocksize; i++) { if (b1[i] != b2[i]) { - printk("corruption in group %lu at byte %u(%u):" - " %x in copy != %x on disk/prealloc\n", - e4b->bd_group, i, i * 8, b1[i], b2[i]); + printk(KERN_ERR "corruption in group %lu " + "at byte %u(%u): %x in copy != %x " + "on disk/prealloc\n", + e4b->bd_group, i, i * 8, b1[i], b2[i]); BUG(); } } @@ -533,9 +534,6 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file, void *buddy; void *buddy2; - if (!test_opt(sb, MBALLOC)) - return 0; - { static int mb_check_counter; if (mb_check_counter++ % 100 != 0) @@ -784,16 +782,21 @@ static int ext4_mb_init_cache(struct page *page, char *incore) if (bh[i] == NULL) goto out; - if (bh_uptodate_or_lock(bh[i])) + if (buffer_uptodate(bh[i]) && + !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))) continue; + lock_buffer(bh[i]); + spin_lock(sb_bgl_lock(EXT4_SB(sb), first_group + i)); if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { ext4_init_block_bitmap(sb, bh[i], first_group + i, desc); set_buffer_uptodate(bh[i]); unlock_buffer(bh[i]); + spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i)); continue; } + spin_unlock(sb_bgl_lock(EXT4_SB(sb), first_group + i)); get_bh(bh[i]); bh[i]->b_end_io = end_buffer_read_sync; submit_bh(READ, bh[i]); @@ -2166,9 +2169,10 @@ static void ext4_mb_history_release(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - remove_proc_entry("mb_groups", sbi->s_mb_proc); - remove_proc_entry("mb_history", sbi->s_mb_proc); - + if (sbi->s_proc != NULL) { + remove_proc_entry("mb_groups", sbi->s_proc); + remove_proc_entry("mb_history", sbi->s_proc); + } kfree(sbi->s_mb_history); } @@ -2177,10 +2181,10 @@ static void ext4_mb_history_init(struct super_block *sb) struct ext4_sb_info *sbi = EXT4_SB(sb); int i; - if (sbi->s_mb_proc != NULL) { - proc_create_data("mb_history", S_IRUGO, sbi->s_mb_proc, + if (sbi->s_proc != NULL) { + proc_create_data("mb_history", S_IRUGO, sbi->s_proc, &ext4_mb_seq_history_fops, sb); - proc_create_data("mb_groups", S_IRUGO, sbi->s_mb_proc, + proc_create_data("mb_groups", S_IRUGO, sbi->s_proc, &ext4_mb_seq_groups_fops, sb); } @@ -2477,24 +2481,19 @@ err_freesgi: int ext4_mb_init(struct super_block *sb, int needs_recovery) { struct ext4_sb_info *sbi = EXT4_SB(sb); - unsigned i; + unsigned i, j; unsigned offset; unsigned max; int ret; - if (!test_opt(sb, MBALLOC)) - return 0; - i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short); sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL); if (sbi->s_mb_offsets == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); return -ENOMEM; } sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL); if (sbi->s_mb_maxs == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_maxs); return -ENOMEM; } @@ -2517,7 +2516,6 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) /* init file for buddy data */ ret = ext4_mb_init_backend(sb); if (ret != 0) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); return ret; @@ -2537,26 +2535,25 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT; sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC; - i = sizeof(struct ext4_locality_group) * NR_CPUS; - sbi->s_locality_groups = kmalloc(i, GFP_KERNEL); + sbi->s_locality_groups = alloc_percpu(struct ext4_locality_group); if (sbi->s_locality_groups == NULL) { - clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); return -ENOMEM; } - for (i = 0; i < NR_CPUS; i++) { + for_each_possible_cpu(i) { struct ext4_locality_group *lg; - lg = &sbi->s_locality_groups[i]; + lg = per_cpu_ptr(sbi->s_locality_groups, i); mutex_init(&lg->lg_mutex); - INIT_LIST_HEAD(&lg->lg_prealloc_list); + for (j = 0; j < PREALLOC_TB_SIZE; j++) + INIT_LIST_HEAD(&lg->lg_prealloc_list[j]); spin_lock_init(&lg->lg_prealloc_lock); } ext4_mb_init_per_dev_proc(sb); ext4_mb_history_init(sb); - printk("EXT4-fs: mballoc enabled\n"); + printk(KERN_INFO "EXT4-fs: mballoc enabled\n"); return 0; } @@ -2585,9 +2582,6 @@ int ext4_mb_release(struct super_block *sb) struct ext4_group_info *grinfo; struct ext4_sb_info *sbi = EXT4_SB(sb); - if (!test_opt(sb, MBALLOC)) - return 0; - /* release freed, non-committed blocks */ spin_lock(&sbi->s_md_lock); list_splice_init(&sbi->s_closed_transaction, @@ -2643,8 +2637,7 @@ int ext4_mb_release(struct super_block *sb) atomic_read(&sbi->s_mb_discarded)); } - kfree(sbi->s_locality_groups); - + free_percpu(sbi->s_locality_groups); ext4_mb_history_release(sb); ext4_mb_destroy_per_dev_proc(sb); @@ -2717,118 +2710,46 @@ ext4_mb_free_committed_blocks(struct super_block *sb) #define EXT4_MB_STREAM_REQ "stream_req" #define EXT4_MB_GROUP_PREALLOC "group_prealloc" - - -#define MB_PROC_FOPS(name) \ -static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v) \ -{ \ - struct ext4_sb_info *sbi = m->private; \ - \ - seq_printf(m, "%ld\n", sbi->s_mb_##name); \ - return 0; \ -} \ - \ -static int ext4_mb_##name##_proc_open(struct inode *inode, struct file *file)\ -{ \ - return single_open(file, ext4_mb_##name##_proc_show, PDE(inode)->data);\ -} \ - \ -static ssize_t ext4_mb_##name##_proc_write(struct file *file, \ - const char __user *buf, size_t cnt, loff_t *ppos) \ -{ \ - struct ext4_sb_info *sbi = PDE(file->f_path.dentry->d_inode)->data;\ - char str[32]; \ - long value; \ - if (cnt >= sizeof(str)) \ - return -EINVAL; \ - if (copy_from_user(str, buf, cnt)) \ - return -EFAULT; \ - value = simple_strtol(str, NULL, 0); \ - if (value <= 0) \ - return -ERANGE; \ - sbi->s_mb_##name = value; \ - return cnt; \ -} \ - \ -static const struct file_operations ext4_mb_##name##_proc_fops = { \ - .owner = THIS_MODULE, \ - .open = ext4_mb_##name##_proc_open, \ - .read = seq_read, \ - .llseek = seq_lseek, \ - .release = single_release, \ - .write = ext4_mb_##name##_proc_write, \ -}; - -MB_PROC_FOPS(stats); -MB_PROC_FOPS(max_to_scan); -MB_PROC_FOPS(min_to_scan); -MB_PROC_FOPS(order2_reqs); -MB_PROC_FOPS(stream_request); -MB_PROC_FOPS(group_prealloc); - -#define MB_PROC_HANDLER(name, var) \ -do { \ - proc = proc_create_data(name, mode, sbi->s_mb_proc, \ - &ext4_mb_##var##_proc_fops, sbi); \ - if (proc == NULL) { \ - printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \ - goto err_out; \ - } \ -} while (0) - static int ext4_mb_init_per_dev_proc(struct super_block *sb) { mode_t mode = S_IFREG | S_IRUGO | S_IWUSR; struct ext4_sb_info *sbi = EXT4_SB(sb); struct proc_dir_entry *proc; - char devname[64]; - if (proc_root_ext4 == NULL) { - sbi->s_mb_proc = NULL; + if (sbi->s_proc == NULL) return -EINVAL; - } - bdevname(sb->s_bdev, devname); - sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4); - - MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats); - MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan); - MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan); - MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs); - MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request); - MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc); + EXT4_PROC_HANDLER(EXT4_MB_STATS_NAME, mb_stats); + EXT4_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, mb_max_to_scan); + EXT4_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, mb_min_to_scan); + EXT4_PROC_HANDLER(EXT4_MB_ORDER2_REQ, mb_order2_reqs); + EXT4_PROC_HANDLER(EXT4_MB_STREAM_REQ, mb_stream_request); + EXT4_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, mb_group_prealloc); return 0; err_out: - printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname); - remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); - remove_proc_entry(devname, proc_root_ext4); - sbi->s_mb_proc = NULL; - + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc); return -ENOMEM; } static int ext4_mb_destroy_per_dev_proc(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); - char devname[64]; - if (sbi->s_mb_proc == NULL) + if (sbi->s_proc == NULL) return -EINVAL; - bdevname(sb->s_bdev, devname); - remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc); - remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc); - remove_proc_entry(devname, proc_root_ext4); + remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc); + remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc); + remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc); + remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc); return 0; } @@ -2850,11 +2771,6 @@ int __init init_ext4_mballoc(void) kmem_cache_destroy(ext4_pspace_cachep); return -ENOMEM; } -#ifdef CONFIG_PROC_FS - proc_root_ext4 = proc_mkdir("fs/ext4", NULL); - if (proc_root_ext4 == NULL) - printk(KERN_ERR "EXT4-fs: Unable to create fs/ext4\n"); -#endif return 0; } @@ -2863,9 +2779,6 @@ void exit_ext4_mballoc(void) /* XXX: synchronize_rcu(); */ kmem_cache_destroy(ext4_pspace_cachep); kmem_cache_destroy(ext4_ac_cachep); -#ifdef CONFIG_PROC_FS - remove_proc_entry("fs/ext4", NULL); -#endif } @@ -2875,7 +2788,7 @@ void exit_ext4_mballoc(void) */ static noinline_for_stack int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, - handle_t *handle) + handle_t *handle, unsigned long reserv_blks) { struct buffer_head *bitmap_bh = NULL; struct ext4_super_block *es; @@ -2964,15 +2877,16 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len); gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); - + percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); /* - * free blocks account has already be reduced/reserved - * at write_begin() time for delayed allocation - * do not double accounting + * Now reduce the dirty block count also. Should not go negative */ if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED)) - percpu_counter_sub(&sbi->s_freeblocks_counter, - ac->ac_b_ex.fe_len); + /* release all the reserved blocks if non delalloc */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, reserv_blks); + else + percpu_counter_sub(&sbi->s_dirtyblocks_counter, + ac->ac_b_ex.fe_len); if (sbi->s_log_groups_per_flex) { ext4_group_t flex_group = ext4_flex_group(sbi, @@ -3260,6 +3174,7 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac, struct ext4_prealloc_space *pa) { unsigned int len = ac->ac_o_ex.fe_len; + ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart, &ac->ac_b_ex.fe_group, &ac->ac_b_ex.fe_start); @@ -3277,14 +3192,45 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac, } /* + * Return the prealloc space that have minimal distance + * from the goal block. @cpa is the prealloc + * space that is having currently known minimal distance + * from the goal block. + */ +static struct ext4_prealloc_space * +ext4_mb_check_group_pa(ext4_fsblk_t goal_block, + struct ext4_prealloc_space *pa, + struct ext4_prealloc_space *cpa) +{ + ext4_fsblk_t cur_distance, new_distance; + + if (cpa == NULL) { + atomic_inc(&pa->pa_count); + return pa; + } + cur_distance = abs(goal_block - cpa->pa_pstart); + new_distance = abs(goal_block - pa->pa_pstart); + + if (cur_distance < new_distance) + return cpa; + + /* drop the previous reference */ + atomic_dec(&cpa->pa_count); + atomic_inc(&pa->pa_count); + return pa; +} + +/* * search goal blocks in preallocated space */ static noinline_for_stack int ext4_mb_use_preallocated(struct ext4_allocation_context *ac) { + int order, i; struct ext4_inode_info *ei = EXT4_I(ac->ac_inode); struct ext4_locality_group *lg; - struct ext4_prealloc_space *pa; + struct ext4_prealloc_space *pa, *cpa = NULL; + ext4_fsblk_t goal_block; /* only data can be preallocated */ if (!(ac->ac_flags & EXT4_MB_HINT_DATA)) @@ -3322,22 +3268,38 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac) lg = ac->ac_lg; if (lg == NULL) return 0; + order = fls(ac->ac_o_ex.fe_len) - 1; + if (order > PREALLOC_TB_SIZE - 1) + /* The max size of hash table is PREALLOC_TB_SIZE */ + order = PREALLOC_TB_SIZE - 1; + + goal_block = ac->ac_g_ex.fe_group * EXT4_BLOCKS_PER_GROUP(ac->ac_sb) + + ac->ac_g_ex.fe_start + + le32_to_cpu(EXT4_SB(ac->ac_sb)->s_es->s_first_data_block); + /* + * search for the prealloc space that is having + * minimal distance from the goal block. + */ + for (i = order; i < PREALLOC_TB_SIZE; i++) { + rcu_read_lock(); + list_for_each_entry_rcu(pa, &lg->lg_prealloc_list[i], + pa_inode_list) { + spin_lock(&pa->pa_lock); + if (pa->pa_deleted == 0 && + pa->pa_free >= ac->ac_o_ex.fe_len) { - rcu_read_lock(); - list_for_each_entry_rcu(pa, &lg->lg_prealloc_list, pa_inode_list) { - spin_lock(&pa->pa_lock); - if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) { - atomic_inc(&pa->pa_count); - ext4_mb_use_group_pa(ac, pa); + cpa = ext4_mb_check_group_pa(goal_block, + pa, cpa); + } spin_unlock(&pa->pa_lock); - ac->ac_criteria = 20; - rcu_read_unlock(); - return 1; } - spin_unlock(&pa->pa_lock); + rcu_read_unlock(); + } + if (cpa) { + ext4_mb_use_group_pa(ac, cpa); + ac->ac_criteria = 20; + return 1; } - rcu_read_unlock(); - return 0; } @@ -3560,6 +3522,7 @@ ext4_mb_new_group_pa(struct ext4_allocation_context *ac) pa->pa_free = pa->pa_len; atomic_set(&pa->pa_count, 1); spin_lock_init(&pa->pa_lock); + INIT_LIST_HEAD(&pa->pa_inode_list); pa->pa_deleted = 0; pa->pa_linear = 1; @@ -3580,10 +3543,10 @@ ext4_mb_new_group_pa(struct ext4_allocation_context *ac) list_add(&pa->pa_group_list, &grp->bb_prealloc_list); ext4_unlock_group(sb, ac->ac_b_ex.fe_group); - spin_lock(pa->pa_obj_lock); - list_add_tail_rcu(&pa->pa_inode_list, &lg->lg_prealloc_list); - spin_unlock(pa->pa_obj_lock); - + /* + * We will later add the new pa to the right bucket + * after updating the pa_free in ext4_mb_release_context + */ return 0; } @@ -3733,20 +3696,23 @@ ext4_mb_discard_group_preallocations(struct super_block *sb, bitmap_bh = ext4_read_block_bitmap(sb, group); if (bitmap_bh == NULL) { - /* error handling here */ - ext4_mb_release_desc(&e4b); - BUG_ON(bitmap_bh == NULL); + ext4_error(sb, __func__, "Error in reading block " + "bitmap for %lu\n", group); + return 0; } err = ext4_mb_load_buddy(sb, group, &e4b); - BUG_ON(err != 0); /* error handling here */ + if (err) { + ext4_error(sb, __func__, "Error in loading buddy " + "information for %lu\n", group); + put_bh(bitmap_bh); + return 0; + } if (needed == 0) needed = EXT4_BLOCKS_PER_GROUP(sb) + 1; - grp = ext4_get_group_info(sb, group); INIT_LIST_HEAD(&list); - ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); repeat: ext4_lock_group(sb, group); @@ -3828,7 +3794,7 @@ out: * * FIXME!! Make sure it is valid at all the call sites */ -void ext4_mb_discard_inode_preallocations(struct inode *inode) +void ext4_discard_preallocations(struct inode *inode) { struct ext4_inode_info *ei = EXT4_I(inode); struct super_block *sb = inode->i_sb; @@ -3840,7 +3806,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode) struct ext4_buddy e4b; int err; - if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) { + if (!S_ISREG(inode->i_mode)) { /*BUG_ON(!list_empty(&ei->i_prealloc_list));*/ return; } @@ -3903,13 +3869,18 @@ repeat: ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL); err = ext4_mb_load_buddy(sb, group, &e4b); - BUG_ON(err != 0); /* error handling here */ + if (err) { + ext4_error(sb, __func__, "Error in loading buddy " + "information for %lu\n", group); + continue; + } bitmap_bh = ext4_read_block_bitmap(sb, group); if (bitmap_bh == NULL) { - /* error handling here */ + ext4_error(sb, __func__, "Error in reading block " + "bitmap for %lu\n", group); ext4_mb_release_desc(&e4b); - BUG_ON(bitmap_bh == NULL); + continue; } ext4_lock_group(sb, group); @@ -4033,8 +4004,7 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac) * per cpu locality group is to reduce the contention between block * request from multiple CPUs. */ - ac->ac_lg = &sbi->s_locality_groups[get_cpu()]; - put_cpu(); + ac->ac_lg = per_cpu_ptr(sbi->s_locality_groups, raw_smp_processor_id()); /* we're going to use group allocation */ ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC; @@ -4112,22 +4082,168 @@ ext4_mb_initialize_context(struct ext4_allocation_context *ac, } +static noinline_for_stack void +ext4_mb_discard_lg_preallocations(struct super_block *sb, + struct ext4_locality_group *lg, + int order, int total_entries) +{ + ext4_group_t group = 0; + struct ext4_buddy e4b; + struct list_head discard_list; + struct ext4_prealloc_space *pa, *tmp; + struct ext4_allocation_context *ac; + + mb_debug("discard locality group preallocation\n"); + + INIT_LIST_HEAD(&discard_list); + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); + + spin_lock(&lg->lg_prealloc_lock); + list_for_each_entry_rcu(pa, &lg->lg_prealloc_list[order], + pa_inode_list) { + spin_lock(&pa->pa_lock); + if (atomic_read(&pa->pa_count)) { + /* + * This is the pa that we just used + * for block allocation. So don't + * free that + */ + spin_unlock(&pa->pa_lock); + continue; + } + if (pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + /* only lg prealloc space */ + BUG_ON(!pa->pa_linear); + + /* seems this one can be freed ... */ + pa->pa_deleted = 1; + spin_unlock(&pa->pa_lock); + + list_del_rcu(&pa->pa_inode_list); + list_add(&pa->u.pa_tmp_list, &discard_list); + + total_entries--; + if (total_entries <= 5) { + /* + * we want to keep only 5 entries + * allowing it to grow to 8. This + * mak sure we don't call discard + * soon for this list. + */ + break; + } + } + spin_unlock(&lg->lg_prealloc_lock); + + list_for_each_entry_safe(pa, tmp, &discard_list, u.pa_tmp_list) { + + ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL); + if (ext4_mb_load_buddy(sb, group, &e4b)) { + ext4_error(sb, __func__, "Error in loading buddy " + "information for %lu\n", group); + continue; + } + ext4_lock_group(sb, group); + list_del(&pa->pa_group_list); + ext4_mb_release_group_pa(&e4b, pa, ac); + ext4_unlock_group(sb, group); + + ext4_mb_release_desc(&e4b); + list_del(&pa->u.pa_tmp_list); + call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback); + } + if (ac) + kmem_cache_free(ext4_ac_cachep, ac); +} + +/* + * We have incremented pa_count. So it cannot be freed at this + * point. Also we hold lg_mutex. So no parallel allocation is + * possible from this lg. That means pa_free cannot be updated. + * + * A parallel ext4_mb_discard_group_preallocations is possible. + * which can cause the lg_prealloc_list to be updated. + */ + +static void ext4_mb_add_n_trim(struct ext4_allocation_context *ac) +{ + int order, added = 0, lg_prealloc_count = 1; + struct super_block *sb = ac->ac_sb; + struct ext4_locality_group *lg = ac->ac_lg; + struct ext4_prealloc_space *tmp_pa, *pa = ac->ac_pa; + + order = fls(pa->pa_free) - 1; + if (order > PREALLOC_TB_SIZE - 1) + /* The max size of hash table is PREALLOC_TB_SIZE */ + order = PREALLOC_TB_SIZE - 1; + /* Add the prealloc space to lg */ + rcu_read_lock(); + list_for_each_entry_rcu(tmp_pa, &lg->lg_prealloc_list[order], + pa_inode_list) { + spin_lock(&tmp_pa->pa_lock); + if (tmp_pa->pa_deleted) { + spin_unlock(&pa->pa_lock); + continue; + } + if (!added && pa->pa_free < tmp_pa->pa_free) { + /* Add to the tail of the previous entry */ + list_add_tail_rcu(&pa->pa_inode_list, + &tmp_pa->pa_inode_list); + added = 1; + /* + * we want to count the total + * number of entries in the list + */ + } + spin_unlock(&tmp_pa->pa_lock); + lg_prealloc_count++; + } + if (!added) + list_add_tail_rcu(&pa->pa_inode_list, + &lg->lg_prealloc_list[order]); + rcu_read_unlock(); + + /* Now trim the list to be not more than 8 elements */ + if (lg_prealloc_count > 8) { + ext4_mb_discard_lg_preallocations(sb, lg, + order, lg_prealloc_count); + return; + } + return ; +} + /* * release all resource we used in allocation */ static int ext4_mb_release_context(struct ext4_allocation_context *ac) { - if (ac->ac_pa) { - if (ac->ac_pa->pa_linear) { + struct ext4_prealloc_space *pa = ac->ac_pa; + if (pa) { + if (pa->pa_linear) { /* see comment in ext4_mb_use_group_pa() */ - spin_lock(&ac->ac_pa->pa_lock); - ac->ac_pa->pa_pstart += ac->ac_b_ex.fe_len; - ac->ac_pa->pa_lstart += ac->ac_b_ex.fe_len; - ac->ac_pa->pa_free -= ac->ac_b_ex.fe_len; - ac->ac_pa->pa_len -= ac->ac_b_ex.fe_len; - spin_unlock(&ac->ac_pa->pa_lock); + spin_lock(&pa->pa_lock); + pa->pa_pstart += ac->ac_b_ex.fe_len; + pa->pa_lstart += ac->ac_b_ex.fe_len; + pa->pa_free -= ac->ac_b_ex.fe_len; + pa->pa_len -= ac->ac_b_ex.fe_len; + spin_unlock(&pa->pa_lock); + /* + * 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. + */ + if (likely(pa->pa_free)) { + spin_lock(pa->pa_obj_lock); + list_del_rcu(&pa->pa_inode_list); + spin_unlock(pa->pa_obj_lock); + ext4_mb_add_n_trim(ac); + } } - ext4_mb_put_pa(ac, ac->ac_sb, ac->ac_pa); + ext4_mb_put_pa(ac, ac->ac_sb, pa); } if (ac->ac_bitmap_page) page_cache_release(ac->ac_bitmap_page); @@ -4162,33 +4278,32 @@ static int ext4_mb_discard_preallocations(struct super_block *sb, int needed) ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, struct ext4_allocation_request *ar, int *errp) { + int freed; struct ext4_allocation_context *ac = NULL; struct ext4_sb_info *sbi; struct super_block *sb; ext4_fsblk_t block = 0; - int freed; - int inquota; + unsigned long inquota; + unsigned long reserv_blks = 0; sb = ar->inode->i_sb; sbi = EXT4_SB(sb); - if (!test_opt(sb, MBALLOC)) { - block = ext4_old_new_blocks(handle, ar->inode, ar->goal, - &(ar->len), errp); - return block; - } if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag) { /* * With delalloc we already reserved the blocks */ - ar->len = ext4_has_free_blocks(sbi, ar->len); - } - - if (ar->len == 0) { - *errp = -ENOSPC; - return 0; + while (ar->len && ext4_claim_free_blocks(sbi, ar->len)) { + /* let others to free the space */ + yield(); + ar->len = ar->len >> 1; + } + if (!ar->len) { + *errp = -ENOSPC; + return 0; + } + reserv_blks = ar->len; } - while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) { ar->flags |= EXT4_MB_HINT_NOPREALLOC; ar->len--; @@ -4234,7 +4349,7 @@ repeat: } if (likely(ac->ac_status == AC_STATUS_FOUND)) { - *errp = ext4_mb_mark_diskspace_used(ac, handle); + *errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_blks); if (*errp == -EAGAIN) { ac->ac_b_ex.fe_group = 0; ac->ac_b_ex.fe_start = 0; @@ -4420,11 +4535,15 @@ do_more: count -= overflow; } bitmap_bh = ext4_read_block_bitmap(sb, block_group); - if (!bitmap_bh) + if (!bitmap_bh) { + err = -EIO; goto error_return; + } gdp = ext4_get_group_desc(sb, block_group, &gd_bh); - if (!gdp) + if (!gdp) { + err = -EIO; goto error_return; + } if (in_range(ext4_block_bitmap(sb, gdp), block, count) || in_range(ext4_inode_bitmap(sb, gdp), block, count) || diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index bfe6add..b3b4828 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -164,11 +164,17 @@ struct ext4_free_extent { * Locality group: * we try to group all related changes together * so that writeback can flush/allocate them together as well + * Size of lg_prealloc_list hash is determined by MB_DEFAULT_GROUP_PREALLOC + * (512). We store prealloc space into the hash based on the pa_free blocks + * order value.ie, fls(pa_free)-1; */ +#define PREALLOC_TB_SIZE 10 struct ext4_locality_group { /* for allocator */ - struct mutex lg_mutex; /* to serialize allocates */ - struct list_head lg_prealloc_list;/* list of preallocations */ + /* to serialize allocates */ + struct mutex lg_mutex; + /* list of preallocations */ + struct list_head lg_prealloc_list[PREALLOC_TB_SIZE]; spinlock_t lg_prealloc_lock; }; @@ -251,7 +257,6 @@ static void ext4_mb_store_history(struct ext4_allocation_context *ac); #define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) -static struct proc_dir_entry *proc_root_ext4; struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index b9e077b..f2a9cf4 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -53,7 +53,8 @@ static int finish_range(handle_t *handle, struct inode *inode, * credit. But below we try to not accumalate too much * of them by restarting the journal. */ - needed = ext4_ext_calc_credits_for_insert(inode, path); + needed = ext4_ext_calc_credits_for_single_extent(inode, + lb->last_block - lb->first_block + 1, path); /* * Make sure the credit we accumalated is not really high @@ -446,8 +447,7 @@ static int free_ext_block(handle_t *handle, struct inode *inode) } -int ext4_ext_migrate(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +int ext4_ext_migrate(struct inode *inode) { handle_t *handle; int retval = 0, i; @@ -515,12 +515,6 @@ int ext4_ext_migrate(struct inode *inode, struct file *filp, * when we add extents we extent the journal */ /* - * inode_mutex prevent write and truncate on the file. Read still goes - * through. We take i_data_sem in ext4_ext_swap_inode_data before we - * switch the inode format to prevent read. - */ - mutex_lock(&(inode->i_mutex)); - /* * Even though we take i_mutex we can still cause block allocation * via mmap write to holes. If we have allocated new blocks we fail * migrate. New block allocation will clear EXT4_EXT_MIGRATE flag. @@ -622,7 +616,6 @@ err_out: tmp_inode->i_nlink = 0; ext4_journal_stop(handle); - mutex_unlock(&(inode->i_mutex)); if (tmp_inode) iput(tmp_inode); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 387ad98..92db9e9 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -151,34 +151,36 @@ struct dx_map_entry static inline ext4_lblk_t dx_get_block(struct dx_entry *entry); static void dx_set_block(struct dx_entry *entry, ext4_lblk_t value); -static inline unsigned dx_get_hash (struct dx_entry *entry); -static void dx_set_hash (struct dx_entry *entry, unsigned value); -static unsigned dx_get_count (struct dx_entry *entries); -static unsigned dx_get_limit (struct dx_entry *entries); -static void dx_set_count (struct dx_entry *entries, unsigned value); -static void dx_set_limit (struct dx_entry *entries, unsigned value); -static unsigned dx_root_limit (struct inode *dir, unsigned infosize); -static unsigned dx_node_limit (struct inode *dir); -static struct dx_frame *dx_probe(struct dentry *dentry, +static inline unsigned dx_get_hash(struct dx_entry *entry); +static void dx_set_hash(struct dx_entry *entry, unsigned value); +static unsigned dx_get_count(struct dx_entry *entries); +static unsigned dx_get_limit(struct dx_entry *entries); +static void dx_set_count(struct dx_entry *entries, unsigned value); +static void dx_set_limit(struct dx_entry *entries, unsigned value); +static unsigned dx_root_limit(struct inode *dir, unsigned infosize); +static unsigned dx_node_limit(struct inode *dir); +static struct dx_frame *dx_probe(const struct qstr *d_name, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame, int *err); -static void dx_release (struct dx_frame *frames); -static int dx_make_map (struct ext4_dir_entry_2 *de, int size, - struct dx_hash_info *hinfo, struct dx_map_entry map[]); +static void dx_release(struct dx_frame *frames); +static int dx_make_map(struct ext4_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry map[]); static void dx_sort_map(struct dx_map_entry *map, unsigned count); -static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to, +static struct ext4_dir_entry_2 *dx_move_dirents(char *from, char *to, struct dx_map_entry *offsets, int count); -static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size); +static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size); static void dx_insert_block(struct dx_frame *frame, u32 hash, ext4_lblk_t block); static int ext4_htree_next_block(struct inode *dir, __u32 hash, struct dx_frame *frame, struct dx_frame *frames, __u32 *start_hash); -static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, - struct ext4_dir_entry_2 **res_dir, int *err); +static struct buffer_head * ext4_dx_find_entry(struct inode *dir, + const struct qstr *d_name, + struct ext4_dir_entry_2 **res_dir, + int *err); static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode); @@ -207,44 +209,44 @@ static inline void dx_set_block(struct dx_entry *entry, ext4_lblk_t value) entry->block = cpu_to_le32(value); } -static inline unsigned dx_get_hash (struct dx_entry *entry) +static inline unsigned dx_get_hash(struct dx_entry *entry) { return le32_to_cpu(entry->hash); } -static inline void dx_set_hash (struct dx_entry *entry, unsigned value) +static inline void dx_set_hash(struct dx_entry *entry, unsigned value) { entry->hash = cpu_to_le32(value); } -static inline unsigned dx_get_count (struct dx_entry *entries) +static inline unsigned dx_get_count(struct dx_entry *entries) { return le16_to_cpu(((struct dx_countlimit *) entries)->count); } -static inline unsigned dx_get_limit (struct dx_entry *entries) +static inline unsigned dx_get_limit(struct dx_entry *entries) { return le16_to_cpu(((struct dx_countlimit *) entries)->limit); } -static inline void dx_set_count (struct dx_entry *entries, unsigned value) +static inline void dx_set_count(struct dx_entry *entries, unsigned value) { ((struct dx_countlimit *) entries)->count = cpu_to_le16(value); } -static inline void dx_set_limit (struct dx_entry *entries, unsigned value) +static inline void dx_set_limit(struct dx_entry *entries, unsigned value) { ((struct dx_countlimit *) entries)->limit = cpu_to_le16(value); } -static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) +static inline unsigned dx_root_limit(struct inode *dir, unsigned infosize) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) - EXT4_DIR_REC_LEN(2) - infosize; return entry_space / sizeof(struct dx_entry); } -static inline unsigned dx_node_limit (struct inode *dir) +static inline unsigned dx_node_limit(struct inode *dir) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0); return entry_space / sizeof(struct dx_entry); @@ -254,12 +256,12 @@ static inline unsigned dx_node_limit (struct inode *dir) * Debug */ #ifdef DX_DEBUG -static void dx_show_index (char * label, struct dx_entry *entries) +static void dx_show_index(char * label, struct dx_entry *entries) { int i, n = dx_get_count (entries); - printk("%s index ", label); + printk(KERN_DEBUG "%s index ", label); for (i = 0; i < n; i++) { - printk("%x->%lu ", i? dx_get_hash(entries + i) : + printk("%x->%lu ", i ? dx_get_hash(entries + i) : 0, (unsigned long)dx_get_block(entries + i)); } printk("\n"); @@ -306,7 +308,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, struct dx_entry *entries, int levels) { unsigned blocksize = dir->i_sb->s_blocksize; - unsigned count = dx_get_count (entries), names = 0, space = 0, i; + unsigned count = dx_get_count(entries), names = 0, space = 0, i; unsigned bcount = 0; struct buffer_head *bh; int err; @@ -325,11 +327,12 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, names += stats.names; space += stats.space; bcount += stats.bcount; - brelse (bh); + brelse(bh); } if (bcount) - printk("%snames %u, fullness %u (%u%%)\n", levels?"":" ", - names, space/bcount,(space/bcount)*100/blocksize); + printk(KERN_DEBUG "%snames %u, fullness %u (%u%%)\n", + levels ? "" : " ", names, space/bcount, + (space/bcount)*100/blocksize); return (struct stats) { names, space, bcount}; } #endif /* DX_DEBUG */ @@ -344,7 +347,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, * back to userspace. */ static struct dx_frame * -dx_probe(struct dentry *dentry, struct inode *dir, +dx_probe(const struct qstr *d_name, struct inode *dir, struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err) { unsigned count, indirect; @@ -355,8 +358,6 @@ dx_probe(struct dentry *dentry, struct inode *dir, u32 hash; frame->bh = NULL; - if (dentry) - dir = dentry->d_parent->d_inode; if (!(bh = ext4_bread (NULL,dir, 0, 0, err))) goto fail; root = (struct dx_root *) bh->b_data; @@ -372,8 +373,8 @@ dx_probe(struct dentry *dentry, struct inode *dir, } hinfo->hash_version = root->info.hash_version; hinfo->seed = EXT4_SB(dir->i_sb)->s_hash_seed; - if (dentry) - ext4fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); + if (d_name) + ext4fs_dirhash(d_name->name, d_name->len, hinfo); hash = hinfo->hash; if (root->info.unused_flags & 1) { @@ -406,7 +407,7 @@ dx_probe(struct dentry *dentry, struct inode *dir, goto fail; } - dxtrace (printk("Look up %x", hash)); + dxtrace(printk("Look up %x", hash)); while (1) { count = dx_get_count(entries); @@ -555,7 +556,7 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash, 0, &err))) return err; /* Failure */ p++; - brelse (p->bh); + brelse(p->bh); p->bh = bh; p->at = p->entries = ((struct dx_node *) bh->b_data)->entries; } @@ -593,7 +594,7 @@ static int htree_dirblock_to_tree(struct file *dir_file, /* On error, skip the f_pos to the next block. */ dir_file->f_pos = (dir_file->f_pos | (dir->i_sb->s_blocksize - 1)) + 1; - brelse (bh); + brelse(bh); return count; } ext4fs_dirhash(de->name, de->name_len, hinfo); @@ -635,8 +636,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, int ret, err; __u32 hashval; - dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, - start_minor_hash)); + dxtrace(printk(KERN_DEBUG "In htree_fill_tree, start hash: %x:%x\n", + start_hash, start_minor_hash)); dir = dir_file->f_path.dentry->d_inode; if (!(EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) { hinfo.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; @@ -648,7 +649,7 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, } hinfo.hash = start_hash; hinfo.minor_hash = 0; - frame = dx_probe(NULL, dir_file->f_path.dentry->d_inode, &hinfo, frames, &err); + frame = dx_probe(NULL, dir, &hinfo, frames, &err); if (!frame) return err; @@ -694,8 +695,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, break; } dx_release(frames); - dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", - count, *next_hash)); + dxtrace(printk(KERN_DEBUG "Fill tree: returned %d entries, " + "next hash: %x\n", count, *next_hash)); return count; errout: dx_release(frames); @@ -802,17 +803,17 @@ static inline int ext4_match (int len, const char * const name, /* * Returns 0 if not found, -1 on failure, and 1 on success */ -static inline int search_dirblock(struct buffer_head * bh, +static inline int search_dirblock(struct buffer_head *bh, struct inode *dir, - struct dentry *dentry, + const struct qstr *d_name, unsigned long offset, struct ext4_dir_entry_2 ** res_dir) { struct ext4_dir_entry_2 * de; char * dlimit; int de_len; - const char *name = dentry->d_name.name; - int namelen = dentry->d_name.len; + const char *name = d_name->name; + int namelen = d_name->len; de = (struct ext4_dir_entry_2 *) bh->b_data; dlimit = bh->b_data + dir->i_sb->s_blocksize; @@ -851,12 +852,13 @@ static inline int search_dirblock(struct buffer_head * bh, * The returned buffer_head has ->b_count elevated. The caller is expected * to brelse() it when appropriate. */ -static struct buffer_head * ext4_find_entry (struct dentry *dentry, +static struct buffer_head * ext4_find_entry (struct inode *dir, + const struct qstr *d_name, struct ext4_dir_entry_2 ** res_dir) { - struct super_block * sb; - struct buffer_head * bh_use[NAMEI_RA_SIZE]; - struct buffer_head * bh, *ret = NULL; + struct super_block *sb; + struct buffer_head *bh_use[NAMEI_RA_SIZE]; + struct buffer_head *bh, *ret = NULL; ext4_lblk_t start, block, b; int ra_max = 0; /* Number of bh's in the readahead buffer, bh_use[] */ @@ -865,16 +867,15 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry, int num = 0; ext4_lblk_t nblocks; int i, err; - struct inode *dir = dentry->d_parent->d_inode; int namelen; *res_dir = NULL; sb = dir->i_sb; - namelen = dentry->d_name.len; + namelen = d_name->len; if (namelen > EXT4_NAME_LEN) return NULL; if (is_dx(dir)) { - bh = ext4_dx_find_entry(dentry, res_dir, &err); + bh = ext4_dx_find_entry(dir, d_name, res_dir, &err); /* * On success, or if the error was file not found, * return. Otherwise, fall back to doing a search the @@ -882,7 +883,8 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry, */ if (bh || (err != ERR_BAD_DX_DIR)) return bh; - dxtrace(printk("ext4_find_entry: dx failed, falling back\n")); + dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, " + "falling back\n")); } nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); start = EXT4_I(dir)->i_dir_start_lookup; @@ -926,7 +928,7 @@ restart: brelse(bh); goto next; } - i = search_dirblock(bh, dir, dentry, + i = search_dirblock(bh, dir, d_name, block << EXT4_BLOCK_SIZE_BITS(sb), res_dir); if (i == 1) { EXT4_I(dir)->i_dir_start_lookup = block; @@ -956,11 +958,11 @@ restart: cleanup_and_exit: /* Clean up the read-ahead blocks */ for (; ra_ptr < ra_max; ra_ptr++) - brelse (bh_use[ra_ptr]); + brelse(bh_use[ra_ptr]); return ret; } -static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, +static struct buffer_head * ext4_dx_find_entry(struct inode *dir, const struct qstr *d_name, struct ext4_dir_entry_2 **res_dir, int *err) { struct super_block * sb; @@ -971,14 +973,13 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, struct buffer_head *bh; ext4_lblk_t block; int retval; - int namelen = dentry->d_name.len; - const u8 *name = dentry->d_name.name; - struct inode *dir = dentry->d_parent->d_inode; + int namelen = d_name->len; + const u8 *name = d_name->name; sb = dir->i_sb; /* NFS may look up ".." - look at dx_root directory block */ if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){ - if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err))) + if (!(frame = dx_probe(d_name, dir, &hinfo, frames, err))) return NULL; } else { frame = frames; @@ -1010,7 +1011,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, return bh; } } - brelse (bh); + brelse(bh); /* Check to see if we should continue to search */ retval = ext4_htree_next_block(dir, hash, frame, frames, NULL); @@ -1025,25 +1026,25 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, *err = -ENOENT; errout: - dxtrace(printk("%s not found\n", name)); + dxtrace(printk(KERN_DEBUG "%s not found\n", name)); dx_release (frames); return NULL; } -static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { - struct inode * inode; - struct ext4_dir_entry_2 * de; - struct buffer_head * bh; + struct inode *inode; + struct ext4_dir_entry_2 *de; + struct buffer_head *bh; if (dentry->d_name.len > EXT4_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); - bh = ext4_find_entry(dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); inode = NULL; if (bh) { unsigned long ino = le32_to_cpu(de->inode); - brelse (bh); + brelse(bh); if (!ext4_valid_inum(dir->i_sb, ino)) { ext4_error(dir->i_sb, "ext4_lookup", "bad inode number: %lu", ino); @@ -1062,15 +1063,14 @@ struct dentry *ext4_get_parent(struct dentry *child) unsigned long ino; struct dentry *parent; struct inode *inode; - struct dentry dotdot; + static const struct qstr dotdot = { + .name = "..", + .len = 2, + }; struct ext4_dir_entry_2 * de; struct buffer_head *bh; - dotdot.d_name.name = ".."; - dotdot.d_name.len = 2; - dotdot.d_parent = child; /* confusing, isn't it! */ - - bh = ext4_find_entry(&dotdot, &de); + bh = ext4_find_entry(child->d_inode, &dotdot, &de); inode = NULL; if (!bh) return ERR_PTR(-ENOENT); @@ -1201,10 +1201,10 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, /* create map in the end of data2 block */ map = (struct dx_map_entry *) (data2 + blocksize); - count = dx_make_map ((struct ext4_dir_entry_2 *) data1, + count = dx_make_map((struct ext4_dir_entry_2 *) data1, blocksize, hinfo, map); map -= count; - dx_sort_map (map, count); + dx_sort_map(map, count); /* Split the existing block in the middle, size-wise */ size = 0; move = 0; @@ -1225,7 +1225,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, /* Fancy dance to stay within two buffers */ de2 = dx_move_dirents(data1, data2, map + split, count - split); - de = dx_pack_dirents(data1,blocksize); + de = dx_pack_dirents(data1, blocksize); de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de); de2->rec_len = ext4_rec_len_to_disk(data2 + blocksize - (char *) de2); dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1)); @@ -1237,15 +1237,15 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, swap(*bh, bh2); de = de2; } - dx_insert_block (frame, hash2 + continued, newblock); - err = ext4_journal_dirty_metadata (handle, bh2); + dx_insert_block(frame, hash2 + continued, newblock); + err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; - err = ext4_journal_dirty_metadata (handle, frame->bh); + err = ext4_journal_dirty_metadata(handle, frame->bh); if (err) goto journal_error; - brelse (bh2); - dxtrace(dx_show_index ("frame", frame->entries)); + brelse(bh2); + dxtrace(dx_show_index("frame", frame->entries)); return de; journal_error: @@ -1271,7 +1271,7 @@ errout: */ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, struct inode *inode, struct ext4_dir_entry_2 *de, - struct buffer_head * bh) + struct buffer_head *bh) { struct inode *dir = dentry->d_parent->d_inode; const char *name = dentry->d_name.name; @@ -1288,11 +1288,11 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, while ((char *) de <= top) { if (!ext4_check_dir_entry("ext4_add_entry", dir, de, bh, offset)) { - brelse (bh); + brelse(bh); return -EIO; } - if (ext4_match (namelen, name, de)) { - brelse (bh); + if (ext4_match(namelen, name, de)) { + brelse(bh); return -EEXIST; } nlen = EXT4_DIR_REC_LEN(de->name_len); @@ -1329,7 +1329,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, } else de->inode = 0; de->name_len = namelen; - memcpy (de->name, name, namelen); + memcpy(de->name, name, namelen); /* * XXX shouldn't update any times until successful * completion of syscall, but too many callers depend @@ -1377,7 +1377,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, struct fake_dirent *fde; blocksize = dir->i_sb->s_blocksize; - dxtrace(printk("Creating index\n")); + dxtrace(printk(KERN_DEBUG "Creating index\n")); retval = ext4_journal_get_write_access(handle, bh); if (retval) { ext4_std_error(dir->i_sb, retval); @@ -1386,7 +1386,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, } root = (struct dx_root *) bh->b_data; - bh2 = ext4_append (handle, dir, &block, &retval); + bh2 = ext4_append(handle, dir, &block, &retval); if (!(bh2)) { brelse(bh); return retval; @@ -1412,9 +1412,9 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, root->info.info_length = sizeof(root->info); root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; entries = root->entries; - dx_set_block (entries, 1); - dx_set_count (entries, 1); - dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info))); + dx_set_block(entries, 1); + dx_set_count(entries, 1); + dx_set_limit(entries, dx_root_limit(dir, sizeof(root->info))); /* Initialize as for dx_probe */ hinfo.hash_version = root->info.hash_version; @@ -1443,14 +1443,14 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry, * may not sleep between calling this and putting something into * the entry, as someone else might have used it while you slept. */ -static int ext4_add_entry (handle_t *handle, struct dentry *dentry, - struct inode *inode) +static int ext4_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode) { struct inode *dir = dentry->d_parent->d_inode; unsigned long offset; - struct buffer_head * bh; + struct buffer_head *bh; struct ext4_dir_entry_2 *de; - struct super_block * sb; + struct super_block *sb; int retval; int dx_fallback=0; unsigned blocksize; @@ -1500,13 +1500,13 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct dx_frame frames[2], *frame; struct dx_entry *entries, *at; struct dx_hash_info hinfo; - struct buffer_head * bh; + struct buffer_head *bh; struct inode *dir = dentry->d_parent->d_inode; - struct super_block * sb = dir->i_sb; + struct super_block *sb = dir->i_sb; struct ext4_dir_entry_2 *de; int err; - frame = dx_probe(dentry, NULL, &hinfo, frames, &err); + frame = dx_probe(&dentry->d_name, dir, &hinfo, frames, &err); if (!frame) return err; entries = frame->entries; @@ -1527,7 +1527,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, } /* Block full, should compress but for now just split */ - dxtrace(printk("using %u of %u node entries\n", + dxtrace(printk(KERN_DEBUG "using %u of %u node entries\n", dx_get_count(entries), dx_get_limit(entries))); /* Need to split index? */ if (dx_get_count(entries) == dx_get_limit(entries)) { @@ -1559,7 +1559,8 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, if (levels) { unsigned icount1 = icount/2, icount2 = icount - icount1; unsigned hash2 = dx_get_hash(entries + icount1); - dxtrace(printk("Split index %i/%i\n", icount1, icount2)); + dxtrace(printk(KERN_DEBUG "Split index %i/%i\n", + icount1, icount2)); BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ err = ext4_journal_get_write_access(handle, @@ -1567,11 +1568,11 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, if (err) goto journal_error; - memcpy ((char *) entries2, (char *) (entries + icount1), - icount2 * sizeof(struct dx_entry)); - dx_set_count (entries, icount1); - dx_set_count (entries2, icount2); - dx_set_limit (entries2, dx_node_limit(dir)); + memcpy((char *) entries2, (char *) (entries + icount1), + icount2 * sizeof(struct dx_entry)); + dx_set_count(entries, icount1); + dx_set_count(entries2, icount2); + dx_set_limit(entries2, dx_node_limit(dir)); /* Which index block gets the new entry? */ if (at - entries >= icount1) { @@ -1579,16 +1580,17 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, frame->entries = entries = entries2; swap(frame->bh, bh2); } - dx_insert_block (frames + 0, hash2, newblock); - dxtrace(dx_show_index ("node", frames[1].entries)); - dxtrace(dx_show_index ("node", + dx_insert_block(frames + 0, hash2, newblock); + dxtrace(dx_show_index("node", frames[1].entries)); + dxtrace(dx_show_index("node", ((struct dx_node *) bh2->b_data)->entries)); err = ext4_journal_dirty_metadata(handle, bh2); if (err) goto journal_error; brelse (bh2); } else { - dxtrace(printk("Creating second level index...\n")); + dxtrace(printk(KERN_DEBUG + "Creating second level index...\n")); memcpy((char *) entries2, (char *) entries, icount * sizeof(struct dx_entry)); dx_set_limit(entries2, dx_node_limit(dir)); @@ -1630,12 +1632,12 @@ cleanup: * ext4_delete_entry deletes a directory entry by merging it with the * previous entry */ -static int ext4_delete_entry (handle_t *handle, - struct inode * dir, - struct ext4_dir_entry_2 * de_del, - struct buffer_head * bh) +static int ext4_delete_entry(handle_t *handle, + struct inode *dir, + struct ext4_dir_entry_2 *de_del, + struct buffer_head *bh) { - struct ext4_dir_entry_2 * de, * pde; + struct ext4_dir_entry_2 *de, *pde; int i; i = 0; @@ -1716,11 +1718,11 @@ static int ext4_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext4_create (struct inode * dir, struct dentry * dentry, int mode, - struct nameidata *nd) +static int ext4_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) { handle_t *handle; - struct inode * inode; + struct inode *inode; int err, retries = 0; retry: @@ -1747,8 +1749,8 @@ retry: return err; } -static int ext4_mknod (struct inode * dir, struct dentry *dentry, - int mode, dev_t rdev) +static int ext4_mknod(struct inode *dir, struct dentry *dentry, + int mode, dev_t rdev) { handle_t *handle; struct inode *inode; @@ -1767,11 +1769,11 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, mode); + inode = ext4_new_inode(handle, dir, mode); err = PTR_ERR(inode); if (!IS_ERR(inode)) { init_special_inode(inode, inode->i_mode, rdev); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR inode->i_op = &ext4_special_inode_operations; #endif err = ext4_add_nondir(handle, dentry, inode); @@ -1782,12 +1784,12 @@ retry: return err; } -static int ext4_mkdir(struct inode * dir, struct dentry * dentry, int mode) +static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode) { handle_t *handle; - struct inode * inode; - struct buffer_head * dir_block; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *dir_block; + struct ext4_dir_entry_2 *de; int err, retries = 0; if (EXT4_DIR_LINK_MAX(dir)) @@ -1803,7 +1805,7 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, S_IFDIR | mode); + inode = ext4_new_inode(handle, dir, S_IFDIR | mode); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; @@ -1811,7 +1813,7 @@ retry: inode->i_op = &ext4_dir_inode_operations; inode->i_fop = &ext4_dir_operations; inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; - dir_block = ext4_bread (handle, inode, 0, 1, &err); + dir_block = ext4_bread(handle, inode, 0, 1, &err); if (!dir_block) goto out_clear_inode; BUFFER_TRACE(dir_block, "get_write_access"); @@ -1820,26 +1822,26 @@ retry: de->inode = cpu_to_le32(inode->i_ino); de->name_len = 1; de->rec_len = ext4_rec_len_to_disk(EXT4_DIR_REC_LEN(de->name_len)); - strcpy (de->name, "."); + strcpy(de->name, "."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); de = ext4_next_entry(de); de->inode = cpu_to_le32(dir->i_ino); de->rec_len = ext4_rec_len_to_disk(inode->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1)); de->name_len = 2; - strcpy (de->name, ".."); + strcpy(de->name, ".."); ext4_set_de_type(dir->i_sb, de, S_IFDIR); inode->i_nlink = 2; BUFFER_TRACE(dir_block, "call ext4_journal_dirty_metadata"); ext4_journal_dirty_metadata(handle, dir_block); - brelse (dir_block); + brelse(dir_block); ext4_mark_inode_dirty(handle, inode); - err = ext4_add_entry (handle, dentry, inode); + err = ext4_add_entry(handle, dentry, inode); if (err) { out_clear_inode: clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); - iput (inode); + iput(inode); goto out_stop; } ext4_inc_count(handle, dir); @@ -1856,17 +1858,17 @@ out_stop: /* * routine to check that the specified directory is empty (for rmdir) */ -static int empty_dir (struct inode * inode) +static int empty_dir(struct inode *inode) { unsigned long offset; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de, * de1; - struct super_block * sb; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de, *de1; + struct super_block *sb; int err = 0; sb = inode->i_sb; if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) || - !(bh = ext4_bread (NULL, inode, 0, 0, &err))) { + !(bh = ext4_bread(NULL, inode, 0, 0, &err))) { if (err) ext4_error(inode->i_sb, __func__, "error %d reading directory #%lu offset 0", @@ -1881,23 +1883,23 @@ static int empty_dir (struct inode * inode) de1 = ext4_next_entry(de); if (le32_to_cpu(de->inode) != inode->i_ino || !le32_to_cpu(de1->inode) || - strcmp (".", de->name) || - strcmp ("..", de1->name)) { - ext4_warning (inode->i_sb, "empty_dir", - "bad directory (dir #%lu) - no `.' or `..'", - inode->i_ino); - brelse (bh); + strcmp(".", de->name) || + strcmp("..", de1->name)) { + ext4_warning(inode->i_sb, "empty_dir", + "bad directory (dir #%lu) - no `.' or `..'", + inode->i_ino); + brelse(bh); return 1; } offset = ext4_rec_len_from_disk(de->rec_len) + ext4_rec_len_from_disk(de1->rec_len); de = ext4_next_entry(de1); - while (offset < inode->i_size ) { + while (offset < inode->i_size) { if (!bh || (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { err = 0; - brelse (bh); - bh = ext4_bread (NULL, inode, + brelse(bh); + bh = ext4_bread(NULL, inode, offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err); if (!bh) { if (err) @@ -1917,13 +1919,13 @@ static int empty_dir (struct inode * inode) continue; } if (le32_to_cpu(de->inode)) { - brelse (bh); + brelse(bh); return 0; } offset += ext4_rec_len_from_disk(de->rec_len); de = ext4_next_entry(de); } - brelse (bh); + brelse(bh); return 1; } @@ -1954,8 +1956,8 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode) * ->i_nlink. For, say it, character device. Not a regular file, * not a directory, not a symlink and ->i_nlink > 0. */ - J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || - S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); + J_ASSERT((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access"); err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); @@ -2069,12 +2071,12 @@ out_brelse: goto out_err; } -static int ext4_rmdir (struct inode * dir, struct dentry *dentry) +static int ext4_rmdir(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de; handle_t *handle; /* Initialize quotas before so that eventual writes go in @@ -2085,7 +2087,7 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) return PTR_ERR(handle); retval = -ENOENT; - bh = ext4_find_entry (dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_rmdir; @@ -2099,16 +2101,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) goto end_rmdir; retval = -ENOTEMPTY; - if (!empty_dir (inode)) + if (!empty_dir(inode)) goto end_rmdir; retval = ext4_delete_entry(handle, dir, de, bh); if (retval) goto end_rmdir; if (!EXT4_DIR_LINK_EMPTY(inode)) - ext4_warning (inode->i_sb, "ext4_rmdir", - "empty directory has too many links (%d)", - inode->i_nlink); + ext4_warning(inode->i_sb, "ext4_rmdir", + "empty directory has too many links (%d)", + inode->i_nlink); inode->i_version++; clear_nlink(inode); /* There's no need to set i_disksize: the fact that i_nlink is @@ -2124,16 +2126,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry) end_rmdir: ext4_journal_stop(handle); - brelse (bh); + brelse(bh); return retval; } -static int ext4_unlink(struct inode * dir, struct dentry *dentry) +static int ext4_unlink(struct inode *dir, struct dentry *dentry) { int retval; - struct inode * inode; - struct buffer_head * bh; - struct ext4_dir_entry_2 * de; + struct inode *inode; + struct buffer_head *bh; + struct ext4_dir_entry_2 *de; handle_t *handle; /* Initialize quotas before so that eventual writes go @@ -2147,7 +2149,7 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) handle->h_sync = 1; retval = -ENOENT; - bh = ext4_find_entry (dentry, &de); + bh = ext4_find_entry(dir, &dentry->d_name, &de); if (!bh) goto end_unlink; @@ -2158,9 +2160,9 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) goto end_unlink; if (!inode->i_nlink) { - ext4_warning (inode->i_sb, "ext4_unlink", - "Deleting nonexistent file (%lu), %d", - inode->i_ino, inode->i_nlink); + ext4_warning(inode->i_sb, "ext4_unlink", + "Deleting nonexistent file (%lu), %d", + inode->i_ino, inode->i_nlink); inode->i_nlink = 1; } retval = ext4_delete_entry(handle, dir, de, bh); @@ -2178,15 +2180,15 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry) end_unlink: ext4_journal_stop(handle); - brelse (bh); + brelse(bh); return retval; } -static int ext4_symlink (struct inode * dir, - struct dentry *dentry, const char * symname) +static int ext4_symlink(struct inode *dir, + struct dentry *dentry, const char *symname) { handle_t *handle; - struct inode * inode; + struct inode *inode; int l, err, retries = 0; l = strlen(symname)+1; @@ -2203,12 +2205,12 @@ retry: if (IS_DIRSYNC(dir)) handle->h_sync = 1; - inode = ext4_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); + inode = ext4_new_inode(handle, dir, S_IFLNK|S_IRWXUGO); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_stop; - if (l > sizeof (EXT4_I(inode)->i_data)) { + if (l > sizeof(EXT4_I(inode)->i_data)) { inode->i_op = &ext4_symlink_inode_operations; ext4_set_aops(inode); /* @@ -2221,14 +2223,14 @@ retry: if (err) { clear_nlink(inode); ext4_mark_inode_dirty(handle, inode); - iput (inode); + iput(inode); goto out_stop; } } else { /* clear the extent format for fast symlink */ EXT4_I(inode)->i_flags &= ~EXT4_EXTENTS_FL; inode->i_op = &ext4_fast_symlink_inode_operations; - memcpy((char*)&EXT4_I(inode)->i_data,symname,l); + memcpy((char *)&EXT4_I(inode)->i_data, symname, l); inode->i_size = l-1; } EXT4_I(inode)->i_disksize = inode->i_size; @@ -2240,8 +2242,8 @@ out_stop: return err; } -static int ext4_link (struct dentry * old_dentry, - struct inode * dir, struct dentry *dentry) +static int ext4_link(struct dentry *old_dentry, + struct inode *dir, struct dentry *dentry) { handle_t *handle; struct inode *inode = old_dentry->d_inode; @@ -2284,13 +2286,13 @@ retry: * Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, - struct inode * new_dir,struct dentry *new_dentry) +static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) { handle_t *handle; - struct inode * old_inode, * new_inode; - struct buffer_head * old_bh, * new_bh, * dir_bh; - struct ext4_dir_entry_2 * old_de, * new_de; + struct inode *old_inode, *new_inode; + struct buffer_head *old_bh, *new_bh, *dir_bh; + struct ext4_dir_entry_2 *old_de, *new_de; int retval; old_bh = new_bh = dir_bh = NULL; @@ -2308,7 +2310,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) handle->h_sync = 1; - old_bh = ext4_find_entry (old_dentry, &old_de); + old_bh = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de); /* * Check for inode number is _not_ due to possible IO errors. * We might rmdir the source, keep it as pwd of some process @@ -2321,32 +2323,32 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, goto end_rename; new_inode = new_dentry->d_inode; - new_bh = ext4_find_entry (new_dentry, &new_de); + new_bh = ext4_find_entry(new_dir, &new_dentry->d_name, &new_de); if (new_bh) { if (!new_inode) { - brelse (new_bh); + brelse(new_bh); new_bh = NULL; } } if (S_ISDIR(old_inode->i_mode)) { if (new_inode) { retval = -ENOTEMPTY; - if (!empty_dir (new_inode)) + if (!empty_dir(new_inode)) goto end_rename; } retval = -EIO; - dir_bh = ext4_bread (handle, old_inode, 0, 0, &retval); + dir_bh = ext4_bread(handle, old_inode, 0, 0, &retval); if (!dir_bh) goto end_rename; if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) goto end_rename; retval = -EMLINK; - if (!new_inode && new_dir!=old_dir && + if (!new_inode && new_dir != old_dir && new_dir->i_nlink >= EXT4_LINK_MAX) goto end_rename; } if (!new_bh) { - retval = ext4_add_entry (handle, new_dentry, old_inode); + retval = ext4_add_entry(handle, new_dentry, old_inode); if (retval) goto end_rename; } else { @@ -2388,7 +2390,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, struct buffer_head *old_bh2; struct ext4_dir_entry_2 *old_de2; - old_bh2 = ext4_find_entry(old_dentry, &old_de2); + old_bh2 = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de2); if (old_bh2) { retval = ext4_delete_entry(handle, old_dir, old_de2, old_bh2); @@ -2433,9 +2435,9 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, retval = 0; end_rename: - brelse (dir_bh); - brelse (old_bh); - brelse (new_bh); + brelse(dir_bh); + brelse(old_bh); + brelse(new_bh); ext4_journal_stop(handle); return retval; } @@ -2454,7 +2456,7 @@ const struct inode_operations ext4_dir_inode_operations = { .mknod = ext4_mknod, .rename = ext4_rename, .setattr = ext4_setattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -2465,7 +2467,7 @@ const struct inode_operations ext4_dir_inode_operations = { const struct inode_operations ext4_special_inode_operations = { .setattr = ext4_setattr, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index f000fbe..b6ec184 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -73,7 +73,7 @@ static int verify_group_input(struct super_block *sb, "Inode bitmap not in group (block %llu)", (unsigned long long)input->inode_bitmap); else if (outside(input->inode_table, start, end) || - outside(itend - 1, start, end)) + outside(itend - 1, start, end)) ext4_warning(sb, __func__, "Inode table not in group (blocks %llu-%llu)", (unsigned long long)input->inode_table, itend - 1); @@ -104,7 +104,7 @@ static int verify_group_input(struct super_block *sb, (unsigned long long)input->inode_bitmap, start, metaend - 1); else if (inside(input->inode_table, start, metaend) || - inside(itend - 1, start, metaend)) + inside(itend - 1, start, metaend)) ext4_warning(sb, __func__, "Inode table (%llu-%llu) overlaps" "GDT table (%llu-%llu)", @@ -158,9 +158,9 @@ static int extend_or_restart_transaction(handle_t *handle, int thresh, if (err) { if ((err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA))) return err; - if ((err = ext4_journal_get_write_access(handle, bh))) + if ((err = ext4_journal_get_write_access(handle, bh))) return err; - } + } return 0; } @@ -418,9 +418,9 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, /* * If we are not using the primary superblock/GDT copy don't resize, - * because the user tools have no way of handling this. Probably a - * bad time to do it anyways. - */ + * because the user tools have no way of handling this. Probably a + * bad time to do it anyways. + */ if (EXT4_SB(sb)->s_sbh->b_blocknr != le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) { ext4_warning(sb, __func__, @@ -507,14 +507,14 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, return 0; exit_inode: - //ext4_journal_release_buffer(handle, iloc.bh); + /* ext4_journal_release_buffer(handle, iloc.bh); */ brelse(iloc.bh); exit_dindj: - //ext4_journal_release_buffer(handle, dind); + /* ext4_journal_release_buffer(handle, dind); */ exit_primary: - //ext4_journal_release_buffer(handle, *primary); + /* ext4_journal_release_buffer(handle, *primary); */ exit_sbh: - //ext4_journal_release_buffer(handle, *primary); + /* ext4_journal_release_buffer(handle, *primary); */ exit_dind: brelse(dind); exit_bh: @@ -773,7 +773,8 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) if (reserved_gdb || gdb_off == 0) { if (!EXT4_HAS_COMPAT_FEATURE(sb, - EXT4_FEATURE_COMPAT_RESIZE_INODE)){ + EXT4_FEATURE_COMPAT_RESIZE_INODE) + || !le16_to_cpu(es->s_reserved_gdt_blocks)) { ext4_warning(sb, __func__, "No reserved GDT blocks, can't resize"); return -EPERM; @@ -818,12 +819,12 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) if ((err = ext4_journal_get_write_access(handle, sbi->s_sbh))) goto exit_journal; - /* - * We will only either add reserved group blocks to a backup group - * or remove reserved blocks for the first group in a new group block. - * Doing both would be mean more complex code, and sane people don't - * use non-sparse filesystems anymore. This is already checked above. - */ + /* + * We will only either add reserved group blocks to a backup group + * or remove reserved blocks for the first group in a new group block. + * Doing both would be mean more complex code, and sane people don't + * use non-sparse filesystems anymore. This is already checked above. + */ if (gdb_off) { primary = sbi->s_group_desc[gdb_num]; if ((err = ext4_journal_get_write_access(handle, primary))) @@ -835,24 +836,24 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) } else if ((err = add_new_gdb(handle, inode, input, &primary))) goto exit_journal; - /* - * OK, now we've set up the new group. Time to make it active. - * - * Current kernels don't lock all allocations via lock_super(), - * so we have to be safe wrt. concurrent accesses the group - * data. So we need to be careful to set all of the relevant - * group descriptor data etc. *before* we enable the group. - * - * The key field here is sbi->s_groups_count: as long as - * that retains its old value, nobody is going to access the new - * group. - * - * So first we update all the descriptor metadata for the new - * group; then we update the total disk blocks count; then we - * update the groups count to enable the group; then finally we - * update the free space counts so that the system can start - * using the new disk blocks. - */ + /* + * OK, now we've set up the new group. Time to make it active. + * + * Current kernels don't lock all allocations via lock_super(), + * so we have to be safe wrt. concurrent accesses the group + * data. So we need to be careful to set all of the relevant + * group descriptor data etc. *before* we enable the group. + * + * The key field here is sbi->s_groups_count: as long as + * that retains its old value, nobody is going to access the new + * group. + * + * So first we update all the descriptor metadata for the new + * group; then we update the total disk blocks count; then we + * update the groups count to enable the group; then finally we + * update the free space counts so that the system can start + * using the new disk blocks. + */ /* Update group descriptor block for new group */ gdp = (struct ext4_group_desc *)((char *)primary->b_data + @@ -869,11 +870,10 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) * We can allocate memory for mb_alloc based on the new group * descriptor */ - if (test_opt(sb, MBALLOC)) { - err = ext4_mb_add_more_groupinfo(sb, input->group, gdp); - if (err) - goto exit_journal; - } + err = ext4_mb_add_more_groupinfo(sb, input->group, gdp); + if (err) + goto exit_journal; + /* * Make the new blocks and inodes valid next. We do this before * increasing the group count so that once the group is enabled, @@ -928,6 +928,15 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) percpu_counter_add(&sbi->s_freeinodes_counter, EXT4_INODES_PER_GROUP(sb)); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) { + ext4_group_t flex_group; + flex_group = ext4_flex_group(sbi, input->group); + sbi->s_flex_groups[flex_group].free_blocks += + input->free_blocks_count; + sbi->s_flex_groups[flex_group].free_inodes += + EXT4_INODES_PER_GROUP(sb); + } + ext4_journal_dirty_metadata(handle, sbi->s_sbh); sb->s_dirt = 1; @@ -946,7 +955,8 @@ exit_put: return err; } /* ext4_group_add */ -/* Extend the filesystem to the new number of blocks specified. This entry +/* + * Extend the filesystem to the new number of blocks specified. This entry * point is only used to extend the current filesystem to the end of the last * existing group. It can be accessed via ioctl, or by "remount,resize=<size>" * for emergencies (because it has no dependencies on reserved blocks). @@ -962,7 +972,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, ext4_group_t o_groups_count; ext4_grpblk_t last; ext4_grpblk_t add; - struct buffer_head * bh; + struct buffer_head *bh; handle_t *handle; int err; unsigned long freed_blocks; @@ -1024,7 +1034,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, o_blocks_count + add, add); /* See if the device is actually as big as what was requested */ - bh = sb_bread(sb, o_blocks_count + add -1); + bh = sb_bread(sb, o_blocks_count + add - 1); if (!bh) { ext4_warning(sb, __func__, "can't read last block, resize aborted"); @@ -1075,8 +1085,15 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, /* * Mark mballoc pages as not up to date so that they will be updated * next time they are loaded by ext4_mb_load_buddy. + * + * XXX Bad, Bad, BAD!!! We should not be overloading the + * Uptodate flag, particularly on thte bitmap bh, as way of + * hinting to ext4_mb_load_buddy() that it needs to be + * overloaded. A user could take a LVM snapshot, then do an + * on-line fsck, and clear the uptodate flag, and this would + * not be a bug in userspace, but a bug in the kernel. FIXME!!! */ - if (test_opt(sb, MBALLOC)) { + { struct ext4_sb_info *sbi = EXT4_SB(sb); struct inode *inode = sbi->s_buddy_cache; int blocks_per_page; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1cb371d..fb940c2 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -34,6 +34,8 @@ #include <linux/namei.h> #include <linux/quotaops.h> #include <linux/seq_file.h> +#include <linux/proc_fs.h> +#include <linux/marker.h> #include <linux/log2.h> #include <linux/crc16.h> #include <asm/uaccess.h> @@ -45,24 +47,25 @@ #include "namei.h" #include "group.h" +struct proc_dir_entry *ext4_proc_root; + static int ext4_load_journal(struct super_block *, struct ext4_super_block *, unsigned long journal_devnum); static int ext4_create_journal(struct super_block *, struct ext4_super_block *, unsigned int); -static void ext4_commit_super (struct super_block * sb, - struct ext4_super_block * es, - int sync); -static void ext4_mark_recovery_complete(struct super_block * sb, - struct ext4_super_block * es); -static void ext4_clear_journal_err(struct super_block * sb, - struct ext4_super_block * es); +static void ext4_commit_super(struct super_block *sb, + struct ext4_super_block *es, int sync); +static void ext4_mark_recovery_complete(struct super_block *sb, + struct ext4_super_block *es); +static void ext4_clear_journal_err(struct super_block *sb, + struct ext4_super_block *es); static int ext4_sync_fs(struct super_block *sb, int wait); -static const char *ext4_decode_error(struct super_block * sb, int errno, +static const char *ext4_decode_error(struct super_block *sb, int errno, char nbuf[16]); -static int ext4_remount (struct super_block * sb, int * flags, char * data); -static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf); +static int ext4_remount(struct super_block *sb, int *flags, char *data); +static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf); static void ext4_unlockfs(struct super_block *sb); -static void ext4_write_super (struct super_block * sb); +static void ext4_write_super(struct super_block *sb); static void ext4_write_super_lockfs(struct super_block *sb); @@ -211,15 +214,15 @@ static void ext4_handle_error(struct super_block *sb) if (sb->s_flags & MS_RDONLY) return; - if (!test_opt (sb, ERRORS_CONT)) { + if (!test_opt(sb, ERRORS_CONT)) { journal_t *journal = EXT4_SB(sb)->s_journal; EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; if (journal) jbd2_journal_abort(journal, -EIO); } - if (test_opt (sb, ERRORS_RO)) { - printk (KERN_CRIT "Remounting filesystem read-only\n"); + if (test_opt(sb, ERRORS_RO)) { + printk(KERN_CRIT "Remounting filesystem read-only\n"); sb->s_flags |= MS_RDONLY; } ext4_commit_super(sb, es, 1); @@ -228,13 +231,13 @@ static void ext4_handle_error(struct super_block *sb) sb->s_id); } -void ext4_error (struct super_block * sb, const char * function, - const char * fmt, ...) +void ext4_error(struct super_block *sb, const char *function, + const char *fmt, ...) { va_list args; va_start(args, fmt); - printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ", sb->s_id, function); vprintk(fmt, args); printk("\n"); va_end(args); @@ -242,7 +245,7 @@ void ext4_error (struct super_block * sb, const char * function, ext4_handle_error(sb); } -static const char *ext4_decode_error(struct super_block * sb, int errno, +static const char *ext4_decode_error(struct super_block *sb, int errno, char nbuf[16]) { char *errstr = NULL; @@ -278,8 +281,7 @@ static const char *ext4_decode_error(struct super_block * sb, int errno, /* __ext4_std_error decodes expected errors from journaling functions * automatically and invokes the appropriate error response. */ -void __ext4_std_error (struct super_block * sb, const char * function, - int errno) +void __ext4_std_error(struct super_block *sb, const char *function, int errno) { char nbuf[16]; const char *errstr; @@ -292,8 +294,8 @@ void __ext4_std_error (struct super_block * sb, const char * function, return; errstr = ext4_decode_error(sb, errno, nbuf); - printk (KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n", - sb->s_id, function, errstr); + printk(KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n", + sb->s_id, function, errstr); ext4_handle_error(sb); } @@ -308,15 +310,15 @@ void __ext4_std_error (struct super_block * sb, const char * function, * case we take the easy way out and panic immediately. */ -void ext4_abort (struct super_block * sb, const char * function, - const char * fmt, ...) +void ext4_abort(struct super_block *sb, const char *function, + const char *fmt, ...) { va_list args; - printk (KERN_CRIT "ext4_abort called.\n"); + printk(KERN_CRIT "ext4_abort called.\n"); va_start(args, fmt); - printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ", sb->s_id, function); vprintk(fmt, args); printk("\n"); va_end(args); @@ -334,8 +336,8 @@ void ext4_abort (struct super_block * sb, const char * function, jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO); } -void ext4_warning (struct super_block * sb, const char * function, - const char * fmt, ...) +void ext4_warning(struct super_block *sb, const char *function, + const char *fmt, ...) { va_list args; @@ -496,7 +498,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi) } } -static void ext4_put_super (struct super_block * sb) +static void ext4_put_super(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; @@ -505,15 +507,18 @@ static void ext4_put_super (struct super_block * sb) ext4_mb_release(sb); ext4_ext_release(sb); ext4_xattr_put_super(sb); - jbd2_journal_destroy(sbi->s_journal); + if (jbd2_journal_destroy(sbi->s_journal) < 0) + ext4_abort(sb, __func__, "Couldn't clean up the journal"); sbi->s_journal = NULL; if (!(sb->s_flags & MS_RDONLY)) { EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); es->s_state = cpu_to_le16(sbi->s_mount_state); - BUFFER_TRACE(sbi->s_sbh, "marking dirty"); - mark_buffer_dirty(sbi->s_sbh); ext4_commit_super(sb, es, 1); } + if (sbi->s_proc) { + remove_proc_entry("inode_readahead_blks", sbi->s_proc); + remove_proc_entry(sb->s_id, ext4_proc_root); + } for (i = 0; i < sbi->s_gdb_count; i++) brelse(sbi->s_group_desc[i]); @@ -522,6 +527,7 @@ static void ext4_put_super (struct super_block * sb) percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); + percpu_counter_destroy(&sbi->s_dirtyblocks_counter); brelse(sbi->s_sbh); #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) @@ -564,12 +570,12 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS); if (!ei) return NULL; -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL ei->i_acl = EXT4_ACL_NOT_CACHED; ei->i_default_acl = EXT4_ACL_NOT_CACHED; #endif - ei->i_block_alloc_info = NULL; ei->vfs_inode.i_version = 1; + ei->vfs_inode.i_data.writeback_index = 0; memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); INIT_LIST_HEAD(&ei->i_prealloc_list); spin_lock_init(&ei->i_prealloc_lock); @@ -595,12 +601,12 @@ static void ext4_destroy_inode(struct inode *inode) kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; INIT_LIST_HEAD(&ei->i_orphan); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR init_rwsem(&ei->xattr_sem); #endif init_rwsem(&ei->i_data_sem); @@ -626,8 +632,7 @@ static void destroy_inodecache(void) static void ext4_clear_inode(struct inode *inode) { - struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info; -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (EXT4_I(inode)->i_acl && EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) { posix_acl_release(EXT4_I(inode)->i_acl); @@ -639,22 +644,20 @@ static void ext4_clear_inode(struct inode *inode) EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED; } #endif - ext4_discard_reservation(inode); - EXT4_I(inode)->i_block_alloc_info = NULL; - if (unlikely(rsv)) - kfree(rsv); + ext4_discard_preallocations(inode); jbd2_journal_release_jbd_inode(EXT4_SB(inode->i_sb)->s_journal, &EXT4_I(inode)->jinode); } -static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb) +static inline void ext4_show_quota_options(struct seq_file *seq, + struct super_block *sb) { #if defined(CONFIG_QUOTA) struct ext4_sb_info *sbi = EXT4_SB(sb); if (sbi->s_jquota_fmt) seq_printf(seq, ",jqfmt=%s", - (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0"); + (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold" : "vfsv0"); if (sbi->s_qf_names[USRQUOTA]) seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]); @@ -718,7 +721,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",debug"); if (test_opt(sb, OLDALLOC)) seq_puts(seq, ",oldalloc"); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR if (test_opt(sb, XATTR_USER) && !(def_mount_opts & EXT4_DEFM_XATTR_USER)) seq_puts(seq, ",user_xattr"); @@ -727,7 +730,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nouser_xattr"); } #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (test_opt(sb, POSIX_ACL) && !(def_mount_opts & EXT4_DEFM_ACL)) seq_puts(seq, ",acl"); if (!test_opt(sb, POSIX_ACL) && (def_mount_opts & EXT4_DEFM_ACL)) @@ -752,8 +755,6 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nobh"); if (!test_opt(sb, EXTENTS)) seq_puts(seq, ",noextents"); - if (!test_opt(sb, MBALLOC)) - seq_puts(seq, ",nomballoc"); if (test_opt(sb, I_VERSION)) seq_puts(seq, ",i_version"); if (!test_opt(sb, DELALLOC)) @@ -773,6 +774,13 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) seq_puts(seq, ",data=writeback"); + if (sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS) + seq_printf(seq, ",inode_readahead_blks=%u", + sbi->s_inode_readahead_blks); + + if (test_opt(sb, DATA_ERR_ABORT)) + seq_puts(seq, ",data_err=abort"); + ext4_show_quota_options(seq, sb); return 0; } @@ -822,8 +830,8 @@ static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid, } #ifdef CONFIG_QUOTA -#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") -#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) +#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group") +#define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) static int ext4_dquot_initialize(struct inode *inode, int type); static int ext4_dquot_drop(struct inode *inode); @@ -902,11 +910,13 @@ enum { Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, + Opt_data_err_abort, Opt_data_err_ignore, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version, Opt_mballoc, Opt_nomballoc, Opt_stripe, Opt_delalloc, Opt_nodelalloc, + Opt_inode_readahead_blks }; static match_table_t tokens = { @@ -947,6 +957,8 @@ static match_table_t tokens = { {Opt_data_journal, "data=journal"}, {Opt_data_ordered, "data=ordered"}, {Opt_data_writeback, "data=writeback"}, + {Opt_data_err_abort, "data_err=abort"}, + {Opt_data_err_ignore, "data_err=ignore"}, {Opt_offusrjquota, "usrjquota="}, {Opt_usrjquota, "usrjquota=%s"}, {Opt_offgrpjquota, "grpjquota="}, @@ -967,6 +979,7 @@ static match_table_t tokens = { {Opt_resize, "resize"}, {Opt_delalloc, "delalloc"}, {Opt_nodelalloc, "nodelalloc"}, + {Opt_inode_readahead_blks, "inode_readahead_blks=%u"}, {Opt_err, NULL}, }; @@ -981,7 +994,7 @@ static ext4_fsblk_t get_sb_block(void **data) /*todo: use simple_strtoll with >32bit ext4 */ sb_block = simple_strtoul(options, &options, 0); if (*options && *options != ',') { - printk("EXT4-fs: Invalid sb specification: %s\n", + printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n", (char *) *data); return 1; } @@ -991,12 +1004,12 @@ static ext4_fsblk_t get_sb_block(void **data) return sb_block; } -static int parse_options (char *options, struct super_block *sb, - unsigned int *inum, unsigned long *journal_devnum, - ext4_fsblk_t *n_blocks_count, int is_remount) +static int parse_options(char *options, struct super_block *sb, + unsigned int *inum, unsigned long *journal_devnum, + ext4_fsblk_t *n_blocks_count, int is_remount) { struct ext4_sb_info *sbi = EXT4_SB(sb); - char * p; + char *p; substring_t args[MAX_OPT_ARGS]; int data_opt = 0; int option; @@ -1009,7 +1022,7 @@ static int parse_options (char *options, struct super_block *sb, if (!options) return 1; - while ((p = strsep (&options, ",")) != NULL) { + while ((p = strsep(&options, ",")) != NULL) { int token; if (!*p) continue; @@ -1017,16 +1030,16 @@ static int parse_options (char *options, struct super_block *sb, token = match_token(p, tokens, args); switch (token) { case Opt_bsd_df: - clear_opt (sbi->s_mount_opt, MINIX_DF); + clear_opt(sbi->s_mount_opt, MINIX_DF); break; case Opt_minix_df: - set_opt (sbi->s_mount_opt, MINIX_DF); + set_opt(sbi->s_mount_opt, MINIX_DF); break; case Opt_grpid: - set_opt (sbi->s_mount_opt, GRPID); + set_opt(sbi->s_mount_opt, GRPID); break; case Opt_nogrpid: - clear_opt (sbi->s_mount_opt, GRPID); + clear_opt(sbi->s_mount_opt, GRPID); break; case Opt_resuid: if (match_int(&args[0], &option)) @@ -1043,49 +1056,50 @@ static int parse_options (char *options, struct super_block *sb, /* *sb_block = match_int(&args[0]); */ break; case Opt_err_panic: - clear_opt (sbi->s_mount_opt, ERRORS_CONT); - clear_opt (sbi->s_mount_opt, ERRORS_RO); - set_opt (sbi->s_mount_opt, ERRORS_PANIC); + clear_opt(sbi->s_mount_opt, ERRORS_CONT); + clear_opt(sbi->s_mount_opt, ERRORS_RO); + set_opt(sbi->s_mount_opt, ERRORS_PANIC); break; case Opt_err_ro: - clear_opt (sbi->s_mount_opt, ERRORS_CONT); - clear_opt (sbi->s_mount_opt, ERRORS_PANIC); - set_opt (sbi->s_mount_opt, ERRORS_RO); + clear_opt(sbi->s_mount_opt, ERRORS_CONT); + clear_opt(sbi->s_mount_opt, ERRORS_PANIC); + set_opt(sbi->s_mount_opt, ERRORS_RO); break; case Opt_err_cont: - clear_opt (sbi->s_mount_opt, ERRORS_RO); - clear_opt (sbi->s_mount_opt, ERRORS_PANIC); - set_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt(sbi->s_mount_opt, ERRORS_RO); + clear_opt(sbi->s_mount_opt, ERRORS_PANIC); + set_opt(sbi->s_mount_opt, ERRORS_CONT); break; case Opt_nouid32: - set_opt (sbi->s_mount_opt, NO_UID32); + set_opt(sbi->s_mount_opt, NO_UID32); break; case Opt_nocheck: - clear_opt (sbi->s_mount_opt, CHECK); + clear_opt(sbi->s_mount_opt, CHECK); break; case Opt_debug: - set_opt (sbi->s_mount_opt, DEBUG); + set_opt(sbi->s_mount_opt, DEBUG); break; case Opt_oldalloc: - set_opt (sbi->s_mount_opt, OLDALLOC); + set_opt(sbi->s_mount_opt, OLDALLOC); break; case Opt_orlov: - clear_opt (sbi->s_mount_opt, OLDALLOC); + clear_opt(sbi->s_mount_opt, OLDALLOC); break; -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR case Opt_user_xattr: - set_opt (sbi->s_mount_opt, XATTR_USER); + set_opt(sbi->s_mount_opt, XATTR_USER); break; case Opt_nouser_xattr: - clear_opt (sbi->s_mount_opt, XATTR_USER); + clear_opt(sbi->s_mount_opt, XATTR_USER); break; #else case Opt_user_xattr: case Opt_nouser_xattr: - printk("EXT4 (no)user_xattr options not supported\n"); + printk(KERN_ERR "EXT4 (no)user_xattr options " + "not supported\n"); break; #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL case Opt_acl: set_opt(sbi->s_mount_opt, POSIX_ACL); break; @@ -1095,7 +1109,8 @@ static int parse_options (char *options, struct super_block *sb, #else case Opt_acl: case Opt_noacl: - printk("EXT4 (no)acl options not supported\n"); + printk(KERN_ERR "EXT4 (no)acl options " + "not supported\n"); break; #endif case Opt_reservation: @@ -1115,7 +1130,7 @@ static int parse_options (char *options, struct super_block *sb, "journal on remount\n"); return 0; } - set_opt (sbi->s_mount_opt, UPDATE_JOURNAL); + set_opt(sbi->s_mount_opt, UPDATE_JOURNAL); break; case Opt_journal_inum: if (is_remount) { @@ -1145,7 +1160,7 @@ static int parse_options (char *options, struct super_block *sb, set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM); break; case Opt_noload: - set_opt (sbi->s_mount_opt, NOLOAD); + set_opt(sbi->s_mount_opt, NOLOAD); break; case Opt_commit: if (match_int(&args[0], &option)) @@ -1178,6 +1193,12 @@ static int parse_options (char *options, struct super_block *sb, sbi->s_mount_opt |= data_opt; } break; + case Opt_data_err_abort: + set_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; + case Opt_data_err_ignore: + clear_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; #ifdef CONFIG_QUOTA case Opt_usrjquota: qtype = USRQUOTA; @@ -1189,8 +1210,8 @@ set_qf_name: sb_any_quota_suspended(sb)) && !sbi->s_qf_names[qtype]) { printk(KERN_ERR - "EXT4-fs: Cannot change journaled " - "quota options when quota turned on.\n"); + "EXT4-fs: Cannot change journaled " + "quota options when quota turned on.\n"); return 0; } qname = match_strdup(&args[0]); @@ -1331,7 +1352,7 @@ set_qf_format: "on this filesystem, use tune2fs\n"); return 0; } - set_opt (sbi->s_mount_opt, EXTENTS); + set_opt(sbi->s_mount_opt, EXTENTS); break; case Opt_noextents: /* @@ -1348,7 +1369,7 @@ set_qf_format: "-o noextents options\n"); return 0; } - clear_opt (sbi->s_mount_opt, EXTENTS); + clear_opt(sbi->s_mount_opt, EXTENTS); break; case Opt_i_version: set_opt(sbi->s_mount_opt, I_VERSION); @@ -1357,12 +1378,6 @@ set_qf_format: case Opt_nodelalloc: clear_opt(sbi->s_mount_opt, DELALLOC); break; - case Opt_mballoc: - set_opt(sbi->s_mount_opt, MBALLOC); - break; - case Opt_nomballoc: - clear_opt(sbi->s_mount_opt, MBALLOC); - break; case Opt_stripe: if (match_int(&args[0], &option)) return 0; @@ -1373,10 +1388,17 @@ set_qf_format: case Opt_delalloc: set_opt(sbi->s_mount_opt, DELALLOC); break; + case Opt_inode_readahead_blks: + if (match_int(&args[0], &option)) + return 0; + if (option < 0 || option > (1 << 30)) + return 0; + sbi->s_inode_readahead_blks = option; + break; default: - printk (KERN_ERR - "EXT4-fs: Unrecognized mount option \"%s\" " - "or missing value\n", p); + printk(KERN_ERR + "EXT4-fs: Unrecognized mount option \"%s\" " + "or missing value\n", p); return 0; } } @@ -1423,31 +1445,31 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, int res = 0; if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) { - printk (KERN_ERR "EXT4-fs warning: revision level too high, " - "forcing read-only mode\n"); + printk(KERN_ERR "EXT4-fs warning: revision level too high, " + "forcing read-only mode\n"); res = MS_RDONLY; } if (read_only) return res; if (!(sbi->s_mount_state & EXT4_VALID_FS)) - printk (KERN_WARNING "EXT4-fs warning: mounting unchecked fs, " - "running e2fsck is recommended\n"); + printk(KERN_WARNING "EXT4-fs warning: mounting unchecked fs, " + "running e2fsck is recommended\n"); else if ((sbi->s_mount_state & EXT4_ERROR_FS)) - printk (KERN_WARNING - "EXT4-fs warning: mounting fs with errors, " - "running e2fsck is recommended\n"); + printk(KERN_WARNING + "EXT4-fs warning: mounting fs with errors, " + "running e2fsck is recommended\n"); 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)) - printk (KERN_WARNING - "EXT4-fs warning: maximal mount count reached, " - "running e2fsck is recommended\n"); + printk(KERN_WARNING + "EXT4-fs warning: maximal mount count reached, " + "running e2fsck is recommended\n"); else if (le32_to_cpu(es->s_checkinterval) && (le32_to_cpu(es->s_lastcheck) + le32_to_cpu(es->s_checkinterval) <= get_seconds())) - printk (KERN_WARNING - "EXT4-fs warning: checktime reached, " - "running e2fsck is recommended\n"); + printk(KERN_WARNING + "EXT4-fs warning: checktime reached, " + "running e2fsck is recommended\n"); #if 0 /* @@@ We _will_ want to clear the valid bit if we find * inconsistencies, to force a fsck at reboot. But for @@ -1473,15 +1495,9 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt); - printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id); - if (EXT4_SB(sb)->s_journal->j_inode == NULL) { - char b[BDEVNAME_SIZE]; - - printk("external journal on %s\n", - bdevname(EXT4_SB(sb)->s_journal->j_dev, b)); - } else { - printk("internal journal\n"); - } + printk(KERN_INFO "EXT4 FS on %s, %s journal on %s\n", + sb->s_id, EXT4_SB(sb)->s_journal->j_inode ? "internal" : + "external", EXT4_SB(sb)->s_journal->j_devname); return res; } @@ -1504,16 +1520,18 @@ static int ext4_fill_flex_info(struct super_block *sb) sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; groups_per_flex = 1 << sbi->s_log_groups_per_flex; - flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) / - groups_per_flex; - sbi->s_flex_groups = kmalloc(flex_group_count * + /* We allocate both existing and potentially added groups */ + flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) + + ((sbi->s_es->s_reserved_gdt_blocks +1 ) << + EXT4_DESC_PER_BLOCK_BITS(sb))) / + groups_per_flex; + sbi->s_flex_groups = kzalloc(flex_group_count * sizeof(struct flex_groups), GFP_KERNEL); if (sbi->s_flex_groups == NULL) { - printk(KERN_ERR "EXT4-fs: not enough memory\n"); + printk(KERN_ERR "EXT4-fs: not enough memory for " + "%lu flex groups\n", flex_group_count); goto failed; } - memset(sbi->s_flex_groups, 0, flex_group_count * - sizeof(struct flex_groups)); gdp = ext4_get_group_desc(sb, 1, &bh); block_bitmap = ext4_block_bitmap(sb, gdp) - 1; @@ -1585,7 +1603,7 @@ static int ext4_check_descriptors(struct super_block *sb) if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) flexbg_flag = 1; - ext4_debug ("Checking group descriptors"); + ext4_debug("Checking group descriptors"); for (i = 0; i < sbi->s_groups_count; i++) { struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL); @@ -1597,16 +1615,14 @@ static int ext4_check_descriptors(struct super_block *sb) (EXT4_BLOCKS_PER_GROUP(sb) - 1); block_bitmap = ext4_block_bitmap(sb, gdp); - if (block_bitmap < first_block || block_bitmap > last_block) - { + if (block_bitmap < first_block || block_bitmap > last_block) { printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " "Block bitmap for group %lu not in group " "(block %llu)!", i, block_bitmap); return 0; } inode_bitmap = ext4_inode_bitmap(sb, gdp); - if (inode_bitmap < first_block || inode_bitmap > last_block) - { + if (inode_bitmap < first_block || inode_bitmap > last_block) { printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " "Inode bitmap for group %lu not in group " "(block %llu)!", i, inode_bitmap); @@ -1614,26 +1630,30 @@ static int ext4_check_descriptors(struct super_block *sb) } inode_table = ext4_inode_table(sb, gdp); if (inode_table < first_block || - inode_table + sbi->s_itb_per_group - 1 > last_block) - { + inode_table + sbi->s_itb_per_group - 1 > last_block) { printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " "Inode table for group %lu not in group " "(block %llu)!", i, inode_table); return 0; } + spin_lock(sb_bgl_lock(sbi, i)); if (!ext4_group_desc_csum_verify(sbi, i, gdp)) { printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: " "Checksum for group %lu failed (%u!=%u)\n", i, le16_to_cpu(ext4_group_desc_csum(sbi, i, gdp)), le16_to_cpu(gdp->bg_checksum)); - return 0; + if (!(sb->s_flags & MS_RDONLY)) { + spin_unlock(sb_bgl_lock(sbi, i)); + return 0; + } } + spin_unlock(sb_bgl_lock(sbi, i)); if (!flexbg_flag) first_block += EXT4_BLOCKS_PER_GROUP(sb); } ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb)); - sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb)); + sbi->s_es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); return 1; } @@ -1654,8 +1674,8 @@ static int ext4_check_descriptors(struct super_block *sb) * e2fsck was run on this filesystem, and it must have already done the orphan * inode cleanup for us, so we can safely abort without any further action. */ -static void ext4_orphan_cleanup (struct super_block * sb, - struct ext4_super_block * es) +static void ext4_orphan_cleanup(struct super_block *sb, + struct ext4_super_block *es) { unsigned int s_flags = sb->s_flags; int nr_orphans = 0, nr_truncates = 0; @@ -1715,9 +1735,9 @@ static void ext4_orphan_cleanup (struct super_block * sb, DQUOT_INIT(inode); if (inode->i_nlink) { printk(KERN_DEBUG - "%s: truncating inode %lu to %Ld bytes\n", + "%s: truncating inode %lu to %lld bytes\n", __func__, inode->i_ino, inode->i_size); - jbd_debug(2, "truncating inode %lu to %Ld bytes\n", + jbd_debug(2, "truncating inode %lu to %lld bytes\n", inode->i_ino, inode->i_size); ext4_truncate(inode); nr_truncates++; @@ -1732,7 +1752,7 @@ static void ext4_orphan_cleanup (struct super_block * sb, iput(inode); /* The delete magic happens here! */ } -#define PLURAL(x) (x), ((x)==1) ? "" : "s" +#define PLURAL(x) (x), ((x) == 1) ? "" : "s" if (nr_orphans) printk(KERN_INFO "EXT4-fs: %s: %d orphan inode%s deleted\n", @@ -1899,12 +1919,12 @@ static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi) return 0; } -static int ext4_fill_super (struct super_block *sb, void *data, int silent) +static int ext4_fill_super(struct super_block *sb, void *data, int silent) __releases(kernel_lock) __acquires(kernel_lock) { - struct buffer_head * bh; + struct buffer_head *bh; struct ext4_super_block *es = NULL; struct ext4_sb_info *sbi; ext4_fsblk_t block; @@ -1915,6 +1935,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) unsigned long journal_devnum = 0; unsigned long def_mount_opts; struct inode *root; + char *cp; int ret = -EINVAL; int blocksize; int db_count; @@ -1931,10 +1952,15 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_mount_opt = 0; sbi->s_resuid = EXT4_DEF_RESUID; sbi->s_resgid = EXT4_DEF_RESGID; + sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS; sbi->s_sb_block = sb_block; unlock_kernel(); + /* Cleanup superblock name */ + for (cp = sb->s_id; (cp = strchr(cp, '/'));) + *cp = '!'; + blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE); if (!blocksize) { printk(KERN_ERR "EXT4-fs: unable to set blocksize\n"); @@ -1953,7 +1979,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) } if (!(bh = sb_bread(sb, logical_sb_block))) { - printk (KERN_ERR "EXT4-fs: unable to read superblock\n"); + printk(KERN_ERR "EXT4-fs: unable to read superblock\n"); goto out_fail; } /* @@ -1974,11 +2000,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, GRPID); if (def_mount_opts & EXT4_DEFM_UID16) set_opt(sbi->s_mount_opt, NO_UID32); -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR if (def_mount_opts & EXT4_DEFM_XATTR_USER) set_opt(sbi->s_mount_opt, XATTR_USER); #endif -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL if (def_mount_opts & EXT4_DEFM_ACL) set_opt(sbi->s_mount_opt, POSIX_ACL); #endif @@ -2013,11 +2039,6 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) ext4_warning(sb, __func__, "extents feature not enabled on this filesystem, " "use tune2fs.\n"); - /* - * turn on mballoc code by default in ext4 filesystem - * Use -o nomballoc to turn it off - */ - set_opt(sbi->s_mount_opt, MBALLOC); /* * enable delayed allocation by default @@ -2026,8 +2047,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, DELALLOC); - if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum, - NULL, 0)) + if (!parse_options((char *) data, sb, &journal_inum, &journal_devnum, + NULL, 0)) goto failed_mount; sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | @@ -2042,16 +2063,6 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) "running e2fsck is recommended\n"); /* - * Since ext4 is still considered development code, we require - * that the TEST_FILESYS flag in s->flags be set. - */ - if (!(le32_to_cpu(es->s_flags) & EXT2_FLAGS_TEST_FILESYS)) { - printk(KERN_WARNING "EXT4-fs: %s: not marked " - "OK to use with test code.\n", sb->s_id); - goto failed_mount; - } - - /* * Check feature flags regardless of the revision level, since we * previously didn't change the revision level when setting the flags, * so there is a chance incompat flags are set on a rev 0 filesystem. @@ -2102,7 +2113,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount; } - brelse (bh); + brelse(bh); logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; offset = do_div(logical_sb_block, blocksize); bh = sb_bread(sb, logical_sb_block); @@ -2114,8 +2125,8 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) es = (struct ext4_super_block *)(((char *)bh->b_data) + offset); sbi->s_es = es; if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) { - printk (KERN_ERR - "EXT4-fs: Magic mismatch, very weird !\n"); + printk(KERN_ERR + "EXT4-fs: Magic mismatch, very weird !\n"); goto failed_mount; } } @@ -2132,9 +2143,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) || (!is_power_of_2(sbi->s_inode_size)) || (sbi->s_inode_size > blocksize)) { - printk (KERN_ERR - "EXT4-fs: unsupported inode size: %d\n", - sbi->s_inode_size); + printk(KERN_ERR + "EXT4-fs: unsupported inode size: %d\n", + sbi->s_inode_size); goto failed_mount; } if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) @@ -2166,20 +2177,20 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_mount_state = le16_to_cpu(es->s_state); sbi->s_addr_per_block_bits = ilog2(EXT4_ADDR_PER_BLOCK(sb)); sbi->s_desc_per_block_bits = ilog2(EXT4_DESC_PER_BLOCK(sb)); - for (i=0; i < 4; i++) + for (i = 0; i < 4; i++) sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); sbi->s_def_hash_version = es->s_def_hash_version; if (sbi->s_blocks_per_group > blocksize * 8) { - printk (KERN_ERR - "EXT4-fs: #blocks per group too big: %lu\n", - sbi->s_blocks_per_group); + printk(KERN_ERR + "EXT4-fs: #blocks per group too big: %lu\n", + sbi->s_blocks_per_group); goto failed_mount; } if (sbi->s_inodes_per_group > blocksize * 8) { - printk (KERN_ERR - "EXT4-fs: #inodes per group too big: %lu\n", - sbi->s_inodes_per_group); + printk(KERN_ERR + "EXT4-fs: #inodes per group too big: %lu\n", + sbi->s_inodes_per_group); goto failed_mount; } @@ -2213,26 +2224,36 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) sbi->s_groups_count = blocks_count; db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / EXT4_DESC_PER_BLOCK(sb); - sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *), + sbi->s_group_desc = kmalloc(db_count * sizeof(struct buffer_head *), GFP_KERNEL); if (sbi->s_group_desc == NULL) { - printk (KERN_ERR "EXT4-fs: not enough memory\n"); + printk(KERN_ERR "EXT4-fs: not enough memory\n"); goto failed_mount; } +#ifdef CONFIG_PROC_FS + if (ext4_proc_root) + sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root); + + if (sbi->s_proc) + proc_create_data("inode_readahead_blks", 0644, sbi->s_proc, + &ext4_ui_proc_fops, + &sbi->s_inode_readahead_blks); +#endif + bgl_lock_init(&sbi->s_blockgroup_lock); for (i = 0; i < db_count; i++) { block = descriptor_loc(sb, logical_sb_block, i); sbi->s_group_desc[i] = sb_bread(sb, block); if (!sbi->s_group_desc[i]) { - printk (KERN_ERR "EXT4-fs: " - "can't read group descriptor %d\n", i); + printk(KERN_ERR "EXT4-fs: " + "can't read group descriptor %d\n", i); db_count = i; goto failed_mount2; } } - if (!ext4_check_descriptors (sb)) { + if (!ext4_check_descriptors(sb)) { printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n"); goto failed_mount2; } @@ -2258,24 +2279,14 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) err = percpu_counter_init(&sbi->s_dirs_counter, ext4_count_dirs(sb)); } + if (!err) { + err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0); + } if (err) { printk(KERN_ERR "EXT4-fs: insufficient memory\n"); goto failed_mount3; } - /* per fileystem reservation list head & lock */ - spin_lock_init(&sbi->s_rsv_window_lock); - sbi->s_rsv_window_root = RB_ROOT; - /* Add a single, static dummy reservation to the start of the - * reservation window list --- it gives us a placeholder for - * append-at-start-of-list which makes the allocation logic - * _much_ simpler. */ - sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; - sbi->s_rsv_window_head.rsv_alloc_hit = 0; - sbi->s_rsv_window_head.rsv_goal_size = 0; - ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); - sbi->s_stripe = ext4_get_stripe_size(sbi); /* @@ -2308,11 +2319,11 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) EXT4_SB(sb)->s_journal->j_failed_commit) { printk(KERN_CRIT "EXT4-fs error (device %s): " "ext4_fill_super: Journal transaction " - "%u is corrupt\n", sb->s_id, + "%u is corrupt\n", sb->s_id, EXT4_SB(sb)->s_journal->j_failed_commit); - if (test_opt (sb, ERRORS_RO)) { - printk (KERN_CRIT - "Mounting filesystem read-only\n"); + if (test_opt(sb, ERRORS_RO)) { + printk(KERN_CRIT + "Mounting filesystem read-only\n"); sb->s_flags |= MS_RDONLY; EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; es->s_state |= cpu_to_le16(EXT4_ERROR_FS); @@ -2332,9 +2343,9 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount3; } else { if (!silent) - printk (KERN_ERR - "ext4: No journal on filesystem on %s\n", - sb->s_id); + printk(KERN_ERR + "ext4: No journal on filesystem on %s\n", + sb->s_id); goto failed_mount3; } @@ -2418,7 +2429,7 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto failed_mount4; } - ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY); + ext4_setup_super(sb, es, sb->s_flags & MS_RDONLY); /* determine the minimum size of new large inodes, if present */ if (sbi->s_inode_size > EXT4_GOOD_OLD_INODE_SIZE) { @@ -2457,12 +2468,12 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) ext4_orphan_cleanup(sb, es); EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS; if (needs_recovery) - printk (KERN_INFO "EXT4-fs: recovery complete.\n"); + printk(KERN_INFO "EXT4-fs: recovery complete.\n"); ext4_mark_recovery_complete(sb, es); - printk (KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n", - test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal": - test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": - "writeback"); + printk(KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n", + test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal": + test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": + "writeback"); if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) { printk(KERN_WARNING "EXT4-fs: Ignoring delalloc option - " @@ -2472,7 +2483,12 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) printk(KERN_INFO "EXT4-fs: delayed allocation enabled\n"); ext4_ext_init(sb); - ext4_mb_init(sb, needs_recovery); + err = ext4_mb_init(sb, needs_recovery); + if (err) { + printk(KERN_ERR "EXT4-fs: failed to initalize mballoc (%d)\n", + err); + goto failed_mount4; + } lock_kernel(); return 0; @@ -2490,11 +2506,16 @@ failed_mount3: percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); + percpu_counter_destroy(&sbi->s_dirtyblocks_counter); failed_mount2: for (i = 0; i < db_count; i++) brelse(sbi->s_group_desc[i]); kfree(sbi->s_group_desc); failed_mount: + if (sbi->s_proc) { + remove_proc_entry("inode_readahead_blks", sbi->s_proc); + remove_proc_entry(sb->s_id, ext4_proc_root); + } #ifdef CONFIG_QUOTA for (i = 0; i < MAXQUOTAS; i++) kfree(sbi->s_qf_names[i]); @@ -2528,6 +2549,10 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) journal->j_flags |= JBD2_BARRIER; else journal->j_flags &= ~JBD2_BARRIER; + if (test_opt(sb, DATA_ERR_ABORT)) + journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR; + else + journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR; spin_unlock(&journal->j_state_lock); } @@ -2553,7 +2578,7 @@ static journal_t *ext4_get_journal(struct super_block *sb, return NULL; } - jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", + jbd_debug(2, "Journal inode found at %p: %lld bytes\n", journal_inode, journal_inode->i_size); if (!S_ISREG(journal_inode->i_mode)) { printk(KERN_ERR "EXT4-fs: invalid journal inode.\n"); @@ -2575,14 +2600,14 @@ static journal_t *ext4_get_journal(struct super_block *sb, static journal_t *ext4_get_dev_journal(struct super_block *sb, dev_t j_dev) { - struct buffer_head * bh; + struct buffer_head *bh; journal_t *journal; ext4_fsblk_t start; ext4_fsblk_t len; int hblock, blocksize; ext4_fsblk_t sb_block; unsigned long offset; - struct ext4_super_block * es; + struct ext4_super_block *es; struct block_device *bdev; bdev = ext4_blkdev_get(j_dev); @@ -2697,8 +2722,8 @@ static int ext4_load_journal(struct super_block *sb, "unavailable, cannot proceed.\n"); return -EROFS; } - printk (KERN_INFO "EXT4-fs: write access will " - "be enabled during recovery.\n"); + printk(KERN_INFO "EXT4-fs: write access will " + "be enabled during recovery.\n"); } } @@ -2716,6 +2741,11 @@ static int ext4_load_journal(struct super_block *sb, return -EINVAL; } + if (journal->j_flags & JBD2_BARRIER) + printk(KERN_INFO "EXT4-fs: barriers enabled\n"); + else + printk(KERN_INFO "EXT4-fs: barriers disabled\n"); + if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { err = jbd2_journal_update_format(journal); if (err) { @@ -2751,8 +2781,8 @@ static int ext4_load_journal(struct super_block *sb, return 0; } -static int ext4_create_journal(struct super_block * sb, - struct ext4_super_block * es, +static int ext4_create_journal(struct super_block *sb, + struct ext4_super_block *es, unsigned int journal_inum) { journal_t *journal; @@ -2793,21 +2823,41 @@ static int ext4_create_journal(struct super_block * sb, return 0; } -static void ext4_commit_super (struct super_block * sb, - struct ext4_super_block * es, - int sync) +static void ext4_commit_super(struct super_block *sb, + struct ext4_super_block *es, int sync) { struct buffer_head *sbh = EXT4_SB(sb)->s_sbh; if (!sbh) return; + if (buffer_write_io_error(sbh)) { + /* + * Oh, dear. A previous attempt to write the + * superblock failed. This could happen because the + * USB device was yanked out. Or it could happen to + * be a transient write error and maybe the block will + * be remapped. Nothing we can do but to retry the + * write and hope for the best. + */ + printk(KERN_ERR "ext4: previous I/O error to " + "superblock detected for %s.\n", sb->s_id); + clear_buffer_write_io_error(sbh); + set_buffer_uptodate(sbh); + } es->s_wtime = cpu_to_le32(get_seconds()); ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb)); es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); BUFFER_TRACE(sbh, "marking dirty"); mark_buffer_dirty(sbh); - if (sync) + if (sync) { sync_dirty_buffer(sbh); + if (buffer_write_io_error(sbh)) { + printk(KERN_ERR "ext4: I/O error while writing " + "superblock for %s.\n", sb->s_id); + clear_buffer_write_io_error(sbh); + set_buffer_uptodate(sbh); + } + } } @@ -2816,13 +2866,15 @@ static void ext4_commit_super (struct super_block * sb, * remounting) the filesystem readonly, then we will end up with a * consistent fs on disk. Record that fact. */ -static void ext4_mark_recovery_complete(struct super_block * sb, - struct ext4_super_block * es) +static void ext4_mark_recovery_complete(struct super_block *sb, + struct ext4_super_block *es) { journal_t *journal = EXT4_SB(sb)->s_journal; jbd2_journal_lock_updates(journal); - jbd2_journal_flush(journal); + if (jbd2_journal_flush(journal) < 0) + goto out; + lock_super(sb); if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) && sb->s_flags & MS_RDONLY) { @@ -2831,6 +2883,8 @@ static void ext4_mark_recovery_complete(struct super_block * sb, ext4_commit_super(sb, es, 1); } unlock_super(sb); + +out: jbd2_journal_unlock_updates(journal); } @@ -2839,8 +2893,8 @@ static void ext4_mark_recovery_complete(struct super_block * sb, * has recorded an error from a previous lifetime, move that error to the * main filesystem now. */ -static void ext4_clear_journal_err(struct super_block * sb, - struct ext4_super_block * es) +static void ext4_clear_journal_err(struct super_block *sb, + struct ext4_super_block *es) { journal_t *journal; int j_errno; @@ -2865,7 +2919,7 @@ static void ext4_clear_journal_err(struct super_block * sb, EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; es->s_state |= cpu_to_le16(EXT4_ERROR_FS); - ext4_commit_super (sb, es, 1); + ext4_commit_super(sb, es, 1); jbd2_journal_clear_err(journal); } @@ -2898,7 +2952,7 @@ int ext4_force_commit(struct super_block *sb) * This implicitly triggers the writebehind on sync(). */ -static void ext4_write_super (struct super_block * sb) +static void ext4_write_super(struct super_block *sb) { if (mutex_trylock(&sb->s_lock) != 0) BUG(); @@ -2909,6 +2963,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait) { tid_t target; + trace_mark(ext4_sync_fs, "dev %s wait %d", sb->s_id, wait); sb->s_dirt = 0; if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { if (wait) @@ -2930,7 +2985,13 @@ static void ext4_write_super_lockfs(struct super_block *sb) /* Now we set up the journal barrier. */ jbd2_journal_lock_updates(journal); - jbd2_journal_flush(journal); + + /* + * We don't want to clear needs_recovery flag when we failed + * to flush the journal. + */ + if (jbd2_journal_flush(journal) < 0) + return; /* Journal blocked and flushed, clear needs_recovery flag. */ EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); @@ -2954,13 +3015,14 @@ static void ext4_unlockfs(struct super_block *sb) } } -static int ext4_remount (struct super_block * sb, int * flags, char * data) +static int ext4_remount(struct super_block *sb, int *flags, char *data) { - struct ext4_super_block * es; + struct ext4_super_block *es; struct ext4_sb_info *sbi = EXT4_SB(sb); ext4_fsblk_t n_blocks_count = 0; unsigned long old_sb_flags; struct ext4_mount_options old_opts; + ext4_group_t g; int err; #ifdef CONFIG_QUOTA int i; @@ -3039,6 +3101,26 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data) } /* + * Make sure the group descriptor checksums + * are sane. If they aren't, refuse to + * remount r/w. + */ + for (g = 0; g < sbi->s_groups_count; g++) { + struct ext4_group_desc *gdp = + ext4_get_group_desc(sb, g, NULL); + + if (!ext4_group_desc_csum_verify(sbi, g, gdp)) { + printk(KERN_ERR + "EXT4-fs: ext4_remount: " + "Checksum for group %lu failed (%u!=%u)\n", + g, le16_to_cpu(ext4_group_desc_csum(sbi, g, gdp)), + le16_to_cpu(gdp->bg_checksum)); + err = -EINVAL; + goto restore_opts; + } + } + + /* * If we have an unprocessed orphan list hanging * around from a previously readonly bdev mount, * require a full umount/remount for now. @@ -3063,7 +3145,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data) sbi->s_mount_state = le16_to_cpu(es->s_state); if ((err = ext4_group_extend(sb, es, n_blocks_count))) goto restore_opts; - if (!ext4_setup_super (sb, es, 0)) + if (!ext4_setup_super(sb, es, 0)) sb->s_flags &= ~MS_RDONLY; } } @@ -3093,7 +3175,7 @@ restore_opts: return err; } -static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) +static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_sb; struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -3143,7 +3225,8 @@ 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); + buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) - + percpu_counter_sum_positive(&sbi->s_dirtyblocks_counter); ext4_free_blocks_count_set(es, buf->f_bfree); buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es); if (buf->f_bfree < ext4_r_blocks_count(es)) @@ -3331,12 +3414,12 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, } /* Journaling quota? */ if (EXT4_SB(sb)->s_qf_names[type]) { - /* Quotafile not of fs root? */ + /* Quotafile not in fs root? */ if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode) printk(KERN_WARNING "EXT4-fs: Quota file not on filesystem root. " "Journaled quota will not work.\n"); - } + } /* * When we journal data on quota file, we have to flush journal to see @@ -3348,12 +3431,17 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, * otherwise be livelocked... */ jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); - jbd2_journal_flush(EXT4_SB(sb)->s_journal); + err = jbd2_journal_flush(EXT4_SB(sb)->s_journal); jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + if (err) { + path_put(&nd.path); + return err; + } } + err = vfs_quota_on_path(sb, type, format_id, &nd.path); path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path, remount); + return err; } /* Read data from quotafile - avoid pagecache and such because we cannot afford @@ -3412,7 +3500,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type, handle_t *handle = journal_current_handle(); if (!handle) { - printk(KERN_WARNING "EXT4-fs: Quota write (off=%Lu, len=%Lu)" + printk(KERN_WARNING "EXT4-fs: Quota write (off=%llu, len=%llu)" " cancelled because transaction is not started.\n", (unsigned long long)off, (unsigned long long)len); return -EIO; @@ -3473,18 +3561,82 @@ static int ext4_get_sb(struct file_system_type *fs_type, return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); } +#ifdef CONFIG_PROC_FS +static int ext4_ui_proc_show(struct seq_file *m, void *v) +{ + unsigned int *p = m->private; + + seq_printf(m, "%u\n", *p); + return 0; +} + +static int ext4_ui_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, ext4_ui_proc_show, PDE(inode)->data); +} + +static ssize_t ext4_ui_proc_write(struct file *file, const char __user *buf, + size_t cnt, loff_t *ppos) +{ + unsigned int *p = PDE(file->f_path.dentry->d_inode)->data; + char str[32]; + unsigned long value; + + if (cnt >= sizeof(str)) + return -EINVAL; + if (copy_from_user(str, buf, cnt)) + return -EFAULT; + value = simple_strtol(str, NULL, 0); + if (value < 0) + return -ERANGE; + *p = value; + return cnt; +} + +const struct file_operations ext4_ui_proc_fops = { + .owner = THIS_MODULE, + .open = ext4_ui_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = ext4_ui_proc_write, +}; +#endif + +static struct file_system_type ext4_fs_type = { + .owner = THIS_MODULE, + .name = "ext4", + .get_sb = ext4_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +#ifdef CONFIG_EXT4DEV_COMPAT +static int ext4dev_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data, struct vfsmount *mnt) +{ + printk(KERN_WARNING "EXT4-fs: Update your userspace programs " + "to mount using ext4\n"); + printk(KERN_WARNING "EXT4-fs: ext4dev backwards compatibility " + "will go away by 2.6.31\n"); + return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); +} + static struct file_system_type ext4dev_fs_type = { .owner = THIS_MODULE, .name = "ext4dev", - .get_sb = ext4_get_sb, + .get_sb = ext4dev_get_sb, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; +MODULE_ALIAS("ext4dev"); +#endif static int __init init_ext4_fs(void) { int err; + ext4_proc_root = proc_mkdir("fs/ext4", NULL); err = init_ext4_mballoc(); if (err) return err; @@ -3495,9 +3647,16 @@ static int __init init_ext4_fs(void) err = init_inodecache(); if (err) goto out1; - err = register_filesystem(&ext4dev_fs_type); + err = register_filesystem(&ext4_fs_type); if (err) goto out; +#ifdef CONFIG_EXT4DEV_COMPAT + err = register_filesystem(&ext4dev_fs_type); + if (err) { + unregister_filesystem(&ext4_fs_type); + goto out; + } +#endif return 0; out: destroy_inodecache(); @@ -3510,10 +3669,14 @@ out2: static void __exit exit_ext4_fs(void) { + unregister_filesystem(&ext4_fs_type); +#ifdef CONFIG_EXT4DEV_COMPAT unregister_filesystem(&ext4dev_fs_type); +#endif destroy_inodecache(); exit_ext4_xattr(); exit_ext4_mballoc(); + remove_proc_entry("fs/ext4", NULL); } MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c index e917864..00740cb 100644 --- a/fs/ext4/symlink.c +++ b/fs/ext4/symlink.c @@ -23,10 +23,10 @@ #include "ext4.h" #include "xattr.h" -static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd) +static void *ext4_follow_link(struct dentry *dentry, struct nameidata *nd) { struct ext4_inode_info *ei = EXT4_I(dentry->d_inode); - nd_set_link(nd, (char*)ei->i_data); + nd_set_link(nd, (char *) ei->i_data); return NULL; } @@ -34,7 +34,7 @@ const struct inode_operations ext4_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = page_follow_link_light, .put_link = page_put_link, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, @@ -45,7 +45,7 @@ const struct inode_operations ext4_symlink_inode_operations = { const struct inode_operations ext4_fast_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = ext4_follow_link, -#ifdef CONFIG_EXT4DEV_FS_XATTR +#ifdef CONFIG_EXT4_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, .listxattr = ext4_listxattr, diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 93c5fdc..80626d5 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -99,12 +99,12 @@ static struct mb_cache *ext4_xattr_cache; static struct xattr_handler *ext4_xattr_handler_map[] = { [EXT4_XATTR_INDEX_USER] = &ext4_xattr_user_handler, -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL [EXT4_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext4_xattr_acl_access_handler, [EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext4_xattr_acl_default_handler, #endif [EXT4_XATTR_INDEX_TRUSTED] = &ext4_xattr_trusted_handler, -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY [EXT4_XATTR_INDEX_SECURITY] = &ext4_xattr_security_handler, #endif }; @@ -112,11 +112,11 @@ static struct xattr_handler *ext4_xattr_handler_map[] = { struct xattr_handler *ext4_xattr_handlers[] = { &ext4_xattr_user_handler, &ext4_xattr_trusted_handler, -#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL +#ifdef CONFIG_EXT4_FS_POSIX_ACL &ext4_xattr_acl_access_handler, &ext4_xattr_acl_default_handler, #endif -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY &ext4_xattr_security_handler, #endif NULL @@ -959,6 +959,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, struct ext4_xattr_block_find bs = { .s = { .not_found = -ENODATA, }, }; + unsigned long no_expand; int error; if (!name) @@ -966,6 +967,9 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, if (strlen(name) > 255) return -ERANGE; down_write(&EXT4_I(inode)->xattr_sem); + no_expand = EXT4_I(inode)->i_state & EXT4_STATE_NO_EXPAND; + EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND; + error = ext4_get_inode_loc(inode, &is.iloc); if (error) goto cleanup; @@ -1042,6 +1046,8 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, cleanup: brelse(is.iloc.bh); brelse(bs.bh); + if (no_expand == 0) + EXT4_I(inode)->i_state &= ~EXT4_STATE_NO_EXPAND; up_write(&EXT4_I(inode)->xattr_sem); return error; } @@ -1512,7 +1518,7 @@ static inline void ext4_xattr_hash_entry(struct ext4_xattr_header *header, char *name = entry->e_name; int n; - for (n=0; n < entry->e_name_len; n++) { + for (n = 0; n < entry->e_name_len; n++) { hash = (hash << NAME_HASH_SHIFT) ^ (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^ *name++; diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h index 5992fe9..8ede88b 100644 --- a/fs/ext4/xattr.h +++ b/fs/ext4/xattr.h @@ -51,8 +51,8 @@ struct ext4_xattr_entry { (((name_len) + EXT4_XATTR_ROUND + \ sizeof(struct ext4_xattr_entry)) & ~EXT4_XATTR_ROUND) #define EXT4_XATTR_NEXT(entry) \ - ( (struct ext4_xattr_entry *)( \ - (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len)) ) + ((struct ext4_xattr_entry *)( \ + (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len))) #define EXT4_XATTR_SIZE(size) \ (((size) + EXT4_XATTR_ROUND) & ~EXT4_XATTR_ROUND) @@ -63,7 +63,7 @@ struct ext4_xattr_entry { EXT4_I(inode)->i_extra_isize)) #define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1)) -# ifdef CONFIG_EXT4DEV_FS_XATTR +# ifdef CONFIG_EXT4_FS_XATTR extern struct xattr_handler ext4_xattr_user_handler; extern struct xattr_handler ext4_xattr_trusted_handler; @@ -88,7 +88,7 @@ extern void exit_ext4_xattr(void); extern struct xattr_handler *ext4_xattr_handlers[]; -# else /* CONFIG_EXT4DEV_FS_XATTR */ +# else /* CONFIG_EXT4_FS_XATTR */ static inline int ext4_xattr_get(struct inode *inode, int name_index, const char *name, @@ -141,9 +141,9 @@ ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize, #define ext4_xattr_handlers NULL -# endif /* CONFIG_EXT4DEV_FS_XATTR */ +# endif /* CONFIG_EXT4_FS_XATTR */ -#ifdef CONFIG_EXT4DEV_FS_SECURITY +#ifdef CONFIG_EXT4_FS_SECURITY extern int ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir); #else diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 3a9ecac8..3222f51 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -36,7 +36,7 @@ static inline int fat_max_cache(struct inode *inode) static struct kmem_cache *fat_cache_cachep; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct fat_cache *cache = (struct fat_cache *)foo; diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c index 302e95c..fb98b3d 100644 --- a/fs/fat/fatent.c +++ b/fs/fat/fatent.c @@ -6,6 +6,7 @@ #include <linux/module.h> #include <linux/fs.h> #include <linux/msdos_fs.h> +#include <linux/blkdev.h> struct fatent_operations { void (*ent_blocknr)(struct super_block *, int, int *, sector_t *); @@ -535,6 +536,7 @@ int fat_free_clusters(struct inode *inode, int cluster) struct fat_entry fatent; struct buffer_head *bhs[MAX_BUF_PER_PAGE]; int i, err, nr_bhs; + int first_cl = cluster; nr_bhs = 0; fatent_init(&fatent); @@ -551,6 +553,18 @@ int fat_free_clusters(struct inode *inode, int cluster) goto error; } + /* + * Issue discard for the sectors we no longer care about, + * batching contiguous clusters into one request + */ + if (cluster != fatent.entry + 1) { + int nr_clus = fatent.entry - first_cl + 1; + + sb_issue_discard(sb, fat_clus_to_blknr(sbi, first_cl), + nr_clus * sbi->sec_per_clus); + first_cl = cluster; + } + ops->ent_put(&fatent, FAT_ENT_FREE); if (sbi->free_clusters != -1) { sbi->free_clusters++; diff --git a/fs/fat/file.c b/fs/fat/file.c index c672df4..ddde370 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -15,6 +15,8 @@ #include <linux/writeback.h> #include <linux/backing-dev.h> #include <linux/blkdev.h> +#include <linux/fsnotify.h> +#include <linux/security.h> int fat_generic_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) @@ -64,6 +66,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, /* Equivalent to a chmod() */ ia.ia_valid = ATTR_MODE | ATTR_CTIME; + ia.ia_ctime = current_fs_time(inode->i_sb); if (is_dir) { ia.ia_mode = MSDOS_MKMODE(attr, S_IRWXUGO & ~sbi->options.fs_dmask) @@ -90,11 +93,21 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, } } + /* + * The security check is questionable... We single + * out the RO attribute for checking by the security + * module, just because it maps to a file mode. + */ + err = security_inode_setattr(filp->f_path.dentry, &ia); + if (err) + goto up; + /* This MUST be done before doing anything irreversible... */ - err = notify_change(filp->f_path.dentry, &ia); + err = fat_setattr(filp->f_path.dentry, &ia); if (err) goto up; + fsnotify_change(filp->f_path.dentry, ia.ia_valid); if (sbi->options.sys_immutable) { if (attr & ATTR_SYS) inode->i_flags |= S_IMMUTABLE; @@ -300,6 +313,8 @@ static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode) return 0; } +#define TIMES_SET_FLAGS (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET) + int fat_setattr(struct dentry *dentry, struct iattr *attr) { struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); @@ -323,9 +338,9 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) /* Check for setting the inode time. */ ia_valid = attr->ia_valid; - if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) { + if (ia_valid & TIMES_SET_FLAGS) { if (fat_allow_set_time(sbi, inode)) - attr->ia_valid &= ~(ATTR_MTIME_SET | ATTR_ATIME_SET); + attr->ia_valid &= ~TIMES_SET_FLAGS; } error = inode_change_ok(inode, attr); diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 23676f9..80ff338 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -498,7 +498,7 @@ static void fat_destroy_inode(struct inode *inode) kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; @@ -562,26 +562,23 @@ static int fat_write_inode(struct inode *inode, int wait) struct buffer_head *bh; struct msdos_dir_entry *raw_entry; loff_t i_pos; - int err = 0; + int err; retry: i_pos = MSDOS_I(inode)->i_pos; if (inode->i_ino == MSDOS_ROOT_INO || !i_pos) return 0; - lock_super(sb); 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); - err = -EIO; - goto out; + return -EIO; } spin_lock(&sbi->inode_hash_lock); if (i_pos != MSDOS_I(inode)->i_pos) { spin_unlock(&sbi->inode_hash_lock); brelse(bh); - unlock_super(sb); goto retry; } @@ -607,11 +604,10 @@ retry: } spin_unlock(&sbi->inode_hash_lock); mark_buffer_dirty(bh); + err = 0; if (wait) err = sync_dirty_buffer(bh); brelse(bh); -out: - unlock_super(sb); return err; } @@ -49,82 +49,6 @@ static int get_close_on_exec(unsigned int fd) return res; } -/* - * locate_fd finds a free file descriptor in the open_fds fdset, - * expanding the fd arrays if necessary. Must be called with the - * file_lock held for write. - */ - -static int locate_fd(unsigned int orig_start, int cloexec) -{ - struct files_struct *files = current->files; - unsigned int newfd; - unsigned int start; - int error; - struct fdtable *fdt; - - spin_lock(&files->file_lock); - - error = -EINVAL; - if (orig_start >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; - -repeat: - fdt = files_fdtable(files); - /* - * Someone might have closed fd's in the range - * orig_start..fdt->next_fd - */ - start = orig_start; - if (start < files->next_fd) - start = files->next_fd; - - newfd = start; - if (start < fdt->max_fds) - newfd = find_next_zero_bit(fdt->open_fds->fds_bits, - fdt->max_fds, start); - - error = -EMFILE; - if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; - - error = expand_files(files, newfd); - if (error < 0) - goto out; - - /* - * If we needed to expand the fs array we - * might have blocked - try again. - */ - if (error) - goto repeat; - - if (start <= files->next_fd) - files->next_fd = newfd + 1; - - FD_SET(newfd, fdt->open_fds); - if (cloexec) - FD_SET(newfd, fdt->close_on_exec); - else - FD_CLR(newfd, fdt->close_on_exec); - error = newfd; - -out: - spin_unlock(&files->file_lock); - return error; -} - -static int dupfd(struct file *file, unsigned int start, int cloexec) -{ - int fd = locate_fd(start, cloexec); - if (fd >= 0) - fd_install(fd, file); - else - fput(file); - - return fd; -} - asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) { int err = -EBADF; @@ -135,35 +59,39 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) if ((flags & ~O_CLOEXEC) != 0) return -EINVAL; - spin_lock(&files->file_lock); - if (!(file = fcheck(oldfd))) - goto out_unlock; - err = newfd; - if (newfd == oldfd) - goto out_unlock; - err = -EBADF; - if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out_unlock; - get_file(file); /* We are now finished with oldfd */ + if (unlikely(oldfd == newfd)) + return -EINVAL; + spin_lock(&files->file_lock); err = expand_files(files, newfd); - if (err < 0) - goto out_fput; - - /* To avoid races with open() and dup(), we will mark the fd as - * in-use in the open-file bitmap throughout the entire dup2() - * process. This is quite safe: do_close() uses the fd array - * entry, not the bitmap, to decide what work needs to be - * done. --sct */ - /* Doesn't work. open() might be there first. --AV */ - - /* Yes. It's a race. In user space. Nothing sane to do */ + file = fcheck(oldfd); + if (unlikely(!file)) + goto Ebadf; + if (unlikely(err < 0)) { + if (err == -EMFILE) + goto Ebadf; + goto out_unlock; + } + /* + * We need to detect attempts to do dup2() over allocated but still + * not finished descriptor. NB: OpenBSD avoids that at the price of + * extra work in their equivalent of fget() - they insert struct + * file immediately after grabbing descriptor, mark it larval if + * more work (e.g. actual opening) is needed and make sure that + * fget() treats larval files as absent. Potentially interesting, + * but while extra work in fget() is trivial, locking implications + * and amount of surgery on open()-related paths in VFS are not. + * FreeBSD fails with -EBADF in the same situation, NetBSD "solution" + * deadlocks in rather amusing ways, AFAICS. All of that is out of + * scope of POSIX or SUS, since neither considers shared descriptor + * tables and this condition does not arise without those. + */ err = -EBUSY; fdt = files_fdtable(files); tofree = fdt->fd[newfd]; if (!tofree && FD_ISSET(newfd, fdt->open_fds)) - goto out_fput; - + goto out_unlock; + get_file(file); rcu_assign_pointer(fdt->fd[newfd], file); FD_SET(newfd, fdt->open_fds); if (flags & O_CLOEXEC) @@ -174,31 +102,41 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) if (tofree) filp_close(tofree, files); - err = newfd; -out: - return err; -out_unlock: - spin_unlock(&files->file_lock); - goto out; -out_fput: + return newfd; + +Ebadf: + err = -EBADF; +out_unlock: spin_unlock(&files->file_lock); - fput(file); - goto out; + return err; } asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) { + if (unlikely(newfd == oldfd)) { /* corner case */ + struct files_struct *files = current->files; + rcu_read_lock(); + if (!fcheck_files(files, oldfd)) + oldfd = -EBADF; + rcu_read_unlock(); + return oldfd; + } return sys_dup3(oldfd, newfd, 0); } asmlinkage long sys_dup(unsigned int fildes) { int ret = -EBADF; - struct file * file = fget(fildes); - - if (file) - ret = dupfd(file, 0, 0); + struct file *file = fget(fildes); + + if (file) { + ret = get_unused_fd(); + if (ret >= 0) + fd_install(ret, file); + else + fput(file); + } return ret; } @@ -321,8 +259,13 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, switch (cmd) { case F_DUPFD: case F_DUPFD_CLOEXEC: - get_file(filp); - err = dupfd(filp, arg, cmd == F_DUPFD_CLOEXEC); + if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + break; + err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0); + if (err >= 0) { + get_file(filp); + fd_install(err, filp); + } break; case F_GETFD: err = get_close_on_exec(fd) ? FD_CLOEXEC : 0; @@ -57,7 +57,7 @@ static int fifo_open(struct inode *inode, struct file *filp) * POSIX.1 says that O_NONBLOCK means return with the FIFO * opened, even when there is no process writing the FIFO. */ - filp->f_op = &read_fifo_fops; + filp->f_op = &read_pipefifo_fops; pipe->r_counter++; if (pipe->readers++ == 0) wake_up_partner(inode); @@ -86,7 +86,7 @@ static int fifo_open(struct inode *inode, struct file *filp) if ((filp->f_flags & O_NONBLOCK) && !pipe->readers) goto err; - filp->f_op = &write_fifo_fops; + filp->f_op = &write_pipefifo_fops; pipe->w_counter++; if (!pipe->writers++) wake_up_partner(inode); @@ -105,7 +105,7 @@ static int fifo_open(struct inode *inode, struct file *filp) * This implementation will NEVER block on a O_RDWR open, since * the process can at least talk to itself. */ - filp->f_op = &rdwr_fifo_fops; + filp->f_op = &rdwr_pipefifo_fops; pipe->readers++; pipe->writers++; @@ -151,5 +151,5 @@ err_nocleanup: * depending on the access mode of the file... */ const struct file_operations def_fifo_fops = { - .open = fifo_open, /* will set read or write pipe_fops */ + .open = fifo_open, /* will set read_ or write_pipefifo_fops */ }; @@ -6,6 +6,7 @@ * Manage the dynamic fd arrays in the process files_struct. */ +#include <linux/module.h> #include <linux/fs.h> #include <linux/mm.h> #include <linux/time.h> @@ -250,9 +251,18 @@ int expand_files(struct files_struct *files, int nr) struct fdtable *fdt; fdt = files_fdtable(files); + + /* + * N.B. For clone tasks sharing a files structure, this test + * will limit the total number of files that can be opened. + */ + if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) + return -EMFILE; + /* Do we need to expand? */ if (nr < fdt->max_fds) return 0; + /* Can we expand? */ if (nr >= sysctl_nr_open) return -EMFILE; @@ -423,3 +433,63 @@ struct files_struct init_files = { }, .file_lock = __SPIN_LOCK_UNLOCKED(init_task.file_lock), }; + +/* + * allocate a file descriptor, mark it busy. + */ +int alloc_fd(unsigned start, unsigned flags) +{ + struct files_struct *files = current->files; + unsigned int fd; + int error; + struct fdtable *fdt; + + spin_lock(&files->file_lock); +repeat: + fdt = files_fdtable(files); + fd = start; + if (fd < files->next_fd) + fd = files->next_fd; + + if (fd < fdt->max_fds) + fd = find_next_zero_bit(fdt->open_fds->fds_bits, + fdt->max_fds, fd); + + error = expand_files(files, fd); + if (error < 0) + goto out; + + /* + * If we needed to expand the fs array we + * might have blocked - try again. + */ + if (error) + goto repeat; + + if (start <= files->next_fd) + files->next_fd = fd + 1; + + FD_SET(fd, fdt->open_fds); + if (flags & O_CLOEXEC) + FD_SET(fd, fdt->close_on_exec); + else + FD_CLR(fd, fdt->close_on_exec); + error = fd; +#if 1 + /* Sanity check */ + if (rcu_dereference(fdt->fd[fd]) != NULL) { + printk(KERN_WARNING "alloc_fd: slot %d not NULL!\n", fd); + rcu_assign_pointer(fdt->fd[fd], NULL); + } +#endif + +out: + spin_unlock(&files->file_lock); + return error; +} + +int get_unused_fd(void) +{ + return alloc_fd(0, 0); +} +EXPORT_SYMBOL(get_unused_fd); diff --git a/fs/file_table.c b/fs/file_table.c index 8308422..f45a449 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -120,7 +120,7 @@ struct file *get_empty_filp(void) tsk = current; INIT_LIST_HEAD(&f->f_u.fu_list); - atomic_set(&f->f_count, 1); + atomic_long_set(&f->f_count, 1); rwlock_init(&f->f_owner.lock); f->f_uid = tsk->fsuid; f->f_gid = tsk->fsgid; @@ -219,7 +219,7 @@ EXPORT_SYMBOL(init_file); void fput(struct file *file) { - if (atomic_dec_and_test(&file->f_count)) + if (atomic_long_dec_and_test(&file->f_count)) __fput(file); } @@ -294,7 +294,7 @@ struct file *fget(unsigned int fd) rcu_read_lock(); file = fcheck_files(files, fd); if (file) { - if (!atomic_inc_not_zero(&file->f_count)) { + if (!atomic_long_inc_not_zero(&file->f_count)) { /* File object ref couldn't be taken */ rcu_read_unlock(); return NULL; @@ -326,7 +326,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed) rcu_read_lock(); file = fcheck_files(files, fd); if (file) { - if (atomic_inc_not_zero(&file->f_count)) + if (atomic_long_inc_not_zero(&file->f_count)) *fput_needed = 1; else /* Didn't get the reference, someone's freed */ @@ -341,7 +341,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed) void put_filp(struct file *file) { - if (atomic_dec_and_test(&file->f_count)) { + if (atomic_long_dec_and_test(&file->f_count)) { security_file_free(file); file_kill(file); file_free(file); diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 51d0035..fd03330 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -898,7 +898,7 @@ static int fuse_access(struct inode *inode, int mask) return PTR_ERR(req); memset(&inarg, 0, sizeof(inarg)); - inarg.mask = mask; + inarg.mask = mask & (MAY_READ | MAY_WRITE | MAY_EXEC); req->in.h.opcode = FUSE_ACCESS; req->in.h.nodeid = get_node_id(inode); req->in.numargs = 1; @@ -927,7 +927,7 @@ static int fuse_access(struct inode *inode, int mask) * access request is sent. Execute permission is still checked * locally based on file mode. */ -static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) +static int fuse_permission(struct inode *inode, int mask) { struct fuse_conn *fc = get_fuse_conn(inode); bool refreshed = false; @@ -962,7 +962,7 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) exist. So if permissions are revoked this won't be noticed immediately, only after the attribute timeout has expired */ - } else if (nd && (nd->flags & (LOOKUP_ACCESS | LOOKUP_CHDIR))) { + } else if (mask & MAY_ACCESS) { err = fuse_access(inode, mask); } else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { if (!(inode->i_mode & S_IXUGO)) { diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 67ff2c6..2bada6b 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -893,7 +893,7 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov, if (count == 0) goto out; - err = remove_suid(file->f_path.dentry); + err = file_remove_suid(file); if (err) goto out; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d2f7d6..d2249f1 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -956,7 +956,7 @@ static inline void unregister_fuseblk(void) } #endif -static void fuse_inode_init_once(struct kmem_cache *cachep, void *foo) +static void fuse_inode_init_once(void *foo) { struct inode * inode = foo; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 13391e5..c962283 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -1265,6 +1265,8 @@ static void blocking_cb(struct gfs2_sbd *sdp, struct lm_lockname *name, holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time; if (time_before(now, holdtime)) delay = holdtime - now; + if (test_bit(GLF_REPLY_PENDING, &gl->gl_flags)) + delay = gl->gl_ops->go_min_hold_time; spin_lock(&gl->gl_spin); handle_callback(gl, state, 1, delay); @@ -1578,8 +1580,6 @@ static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags) *p++ = 'a'; if (flags & GL_EXACT) *p++ = 'E'; - if (flags & GL_ATIME) - *p++ = 'a'; if (flags & GL_NOCACHE) *p++ = 'c'; if (test_bit(HIF_HOLDER, &iflags)) @@ -1816,15 +1816,17 @@ restart: if (gl) { gi->gl = hlist_entry(gl->gl_list.next, struct gfs2_glock, gl_list); - if (gi->gl) - gfs2_glock_hold(gi->gl); + } else { + gi->gl = hlist_entry(gl_hash_table[gi->hash].hb_list.first, + struct gfs2_glock, gl_list); } + if (gi->gl) + gfs2_glock_hold(gi->gl); read_unlock(gl_lock_addr(gi->hash)); if (gl) gfs2_glock_put(gl); - if (gl && gi->gl == NULL) - gi->hash++; while (gi->gl == NULL) { + gi->hash++; if (gi->hash >= GFS2_GL_HASH_SIZE) return 1; read_lock(gl_lock_addr(gi->hash)); @@ -1833,7 +1835,6 @@ restart: if (gi->gl) gfs2_glock_hold(gi->gl); read_unlock(gl_lock_addr(gi->hash)); - gi->hash++; } if (gi->sdp != gi->gl->gl_sbd) diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 971d92a..695c6b1 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -24,7 +24,6 @@ #define GL_ASYNC 0x00000040 #define GL_EXACT 0x00000080 #define GL_SKIP 0x00000100 -#define GL_ATIME 0x00000200 #define GL_NOCACHE 0x00000400 #define GLR_TRYFAILED 13 diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 448697a..f566ec1 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -386,20 +386,21 @@ struct gfs2_statfs_change_host { #define GFS2_DATA_ORDERED 2 struct gfs2_args { - char ar_lockproto[GFS2_LOCKNAME_LEN]; /* Name of the Lock Protocol */ - char ar_locktable[GFS2_LOCKNAME_LEN]; /* Name of the Lock Table */ - char ar_hostdata[GFS2_LOCKNAME_LEN]; /* Host specific data */ - int ar_spectator; /* Don't get a journal because we're always RO */ - int ar_ignore_local_fs; /* Don't optimize even if local_fs is 1 */ - int ar_localflocks; /* Let the VFS do flock|fcntl locks for us */ - int ar_localcaching; /* Local-style caching (dangerous on multihost) */ - int ar_debug; /* Oops on errors instead of trying to be graceful */ - int ar_upgrade; /* Upgrade ondisk/multihost format */ - unsigned int ar_num_glockd; /* Number of glockd threads */ - int ar_posix_acl; /* Enable posix acls */ - int ar_quota; /* off/account/on */ - int ar_suiddir; /* suiddir support */ - int ar_data; /* ordered/writeback */ + char ar_lockproto[GFS2_LOCKNAME_LEN]; /* Name of the Lock Protocol */ + char ar_locktable[GFS2_LOCKNAME_LEN]; /* Name of the Lock Table */ + char ar_hostdata[GFS2_LOCKNAME_LEN]; /* Host specific data */ + unsigned int ar_spectator:1; /* Don't get a journal */ + unsigned int ar_ignore_local_fs:1; /* Ignore optimisations */ + unsigned int ar_localflocks:1; /* Let the VFS do flock|fcntl */ + unsigned int ar_localcaching:1; /* Local caching */ + unsigned int ar_debug:1; /* Oops on errors */ + unsigned int ar_upgrade:1; /* Upgrade ondisk format */ + unsigned int ar_posix_acl:1; /* Enable posix acls */ + unsigned int ar_quota:2; /* off/account/on */ + unsigned int ar_suiddir:1; /* suiddir support */ + unsigned int ar_data:2; /* ordered/writeback */ + unsigned int ar_meta:1; /* mount metafs */ + unsigned int ar_num_glockd; /* Number of glockd threads */ }; struct gfs2_tune { @@ -419,7 +420,6 @@ struct gfs2_tune { unsigned int gt_quota_scale_den; /* Denominator */ unsigned int gt_quota_cache_secs; unsigned int gt_quota_quantum; /* Secs between syncs to quota file */ - unsigned int gt_atime_quantum; /* Min secs between atime updates */ unsigned int gt_new_files_jdata; unsigned int gt_max_readahead; /* Max bytes to read-ahead from disk */ unsigned int gt_stall_secs; /* Detects trouble! */ @@ -432,7 +432,7 @@ enum { SDF_JOURNAL_CHECKED = 0, SDF_JOURNAL_LIVE = 1, SDF_SHUTDOWN = 2, - SDF_NOATIME = 3, + SDF_NOBARRIERS = 3, }; #define GFS2_FSNAME_LEN 256 @@ -461,7 +461,6 @@ struct gfs2_sb_host { struct gfs2_sbd { struct super_block *sd_vfs; - struct super_block *sd_vfs_meta; struct kobject sd_kobj; unsigned long sd_flags; /* SDF_... */ struct gfs2_sb_host sd_sb; @@ -499,7 +498,9 @@ struct gfs2_sbd { /* Inode Stuff */ - struct inode *sd_master_dir; + struct dentry *sd_master_dir; + struct dentry *sd_root_dir; + struct inode *sd_jindex; struct inode *sd_inum_inode; struct inode *sd_statfs_inode; @@ -634,7 +635,6 @@ struct gfs2_sbd { /* Debugging crud */ unsigned long sd_last_warning; - struct vfsmount *sd_gfs2mnt; struct dentry *debugfs_dir; /* debugfs directory */ struct dentry *debugfs_dentry_glocks; /* for debugfs */ }; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 6da0ab35..7cee695 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -18,6 +18,7 @@ #include <linux/crc32.h> #include <linux/lm_interface.h> #include <linux/security.h> +#include <linux/time.h> #include "gfs2.h" #include "incore.h" @@ -249,6 +250,7 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) { struct gfs2_dinode_host *di = &ip->i_di; 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))) @@ -275,8 +277,10 @@ static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) di->di_size = be64_to_cpu(str->di_size); i_size_write(&ip->i_inode, di->di_size); gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); - ip->i_inode.i_atime.tv_sec = be64_to_cpu(str->di_atime); - ip->i_inode.i_atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); + 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); @@ -448,7 +452,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) struct qstr qstr; struct inode *inode; gfs2_str2qstr(&qstr, name); - inode = gfs2_lookupi(dip, &qstr, 1, NULL); + inode = gfs2_lookupi(dip, &qstr, 1); /* gfs2_lookupi has inconsistent callers: vfs * related routines expect NULL for no entry found, * gfs2_lookup_simple callers expect ENOENT @@ -477,7 +481,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) */ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, - int is_root, struct nameidata *nd) + int is_root) { struct super_block *sb = dir->i_sb; struct gfs2_inode *dip = GFS2_I(dir); @@ -1033,13 +1037,11 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (bh) brelse(bh); - if (!inode) - return ERR_PTR(-ENOMEM); return inode; fail_gunlock2: gfs2_glock_dq_uninit(ghs + 1); - if (inode) + if (inode && !IS_ERR(inode)) iput(inode); fail_gunlock: gfs2_glock_dq(ghs); @@ -1140,54 +1142,6 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, 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 - */ - -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; - struct qstr dotdot; - int error = 0; - - gfs2_str2qstr(&dotdot, ".."); - - 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, &dotdot, 1, NULL); - if (IS_ERR(tmp)) { - error = PTR_ERR(tmp); - break; - } - - iput(dir); - dir = tmp; - } - - iput(dir); - - return error; -} - /** * gfs2_readlinki - return the contents of a symlink * @ip: the symlink's inode @@ -1207,8 +1161,8 @@ int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len) unsigned int x; int error; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &i_gh); - error = gfs2_glock_nq_atime(&i_gh); + 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); return error; @@ -1243,101 +1197,6 @@ out: return error; } -/** - * gfs2_glock_nq_atime - Acquire a hold on an inode's glock, and - * conditionally update the inode's atime - * @gh: the holder to acquire - * - * Tests atime (access time) for gfs2_read, gfs2_readdir and gfs2_mmap - * Update if the difference between the current time and the inode's current - * atime is greater than an interval specified at mount. - * - * Returns: errno - */ - -int gfs2_glock_nq_atime(struct gfs2_holder *gh) -{ - struct gfs2_glock *gl = gh->gh_gl; - struct gfs2_sbd *sdp = gl->gl_sbd; - struct gfs2_inode *ip = gl->gl_object; - s64 quantum = gfs2_tune_get(sdp, gt_atime_quantum); - unsigned int state; - int flags; - int error; - struct timespec tv = CURRENT_TIME; - - if (gfs2_assert_warn(sdp, gh->gh_flags & GL_ATIME) || - gfs2_assert_warn(sdp, !(gh->gh_flags & GL_ASYNC)) || - gfs2_assert_warn(sdp, gl->gl_ops == &gfs2_inode_glops)) - return -EINVAL; - - state = gh->gh_state; - flags = gh->gh_flags; - - error = gfs2_glock_nq(gh); - if (error) - return error; - - if (test_bit(SDF_NOATIME, &sdp->sd_flags) || - (sdp->sd_vfs->s_flags & MS_RDONLY)) - return 0; - - if (tv.tv_sec - ip->i_inode.i_atime.tv_sec >= quantum) { - gfs2_glock_dq(gh); - gfs2_holder_reinit(LM_ST_EXCLUSIVE, gh->gh_flags & ~LM_FLAG_ANY, - gh); - error = gfs2_glock_nq(gh); - if (error) - return error; - - /* Verify that atime hasn't been updated while we were - trying to get exclusive lock. */ - - tv = CURRENT_TIME; - if (tv.tv_sec - ip->i_inode.i_atime.tv_sec >= quantum) { - struct buffer_head *dibh; - struct gfs2_dinode *di; - - error = gfs2_trans_begin(sdp, RES_DINODE, 0); - if (error == -EROFS) - return 0; - if (error) - goto fail; - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto fail_end_trans; - - ip->i_inode.i_atime = tv; - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - di = (struct gfs2_dinode *)dibh->b_data; - di->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); - di->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); - brelse(dibh); - - gfs2_trans_end(sdp); - } - - /* If someone else has asked for the glock, - unlock and let them have it. Then reacquire - in the original state. */ - if (gfs2_glock_is_blocking(gl)) { - gfs2_glock_dq(gh); - gfs2_holder_reinit(state, flags, gh); - return gfs2_glock_nq(gh); - } - } - - return 0; - -fail_end_trans: - gfs2_trans_end(sdp); -fail: - gfs2_glock_dq(gh); - return error; -} - static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) { diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 6074c25..2d43f69 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -83,7 +83,7 @@ int gfs2_inode_refresh(struct gfs2_inode *ip); int gfs2_dinode_dealloc(struct gfs2_inode *inode); int gfs2_change_nlink(struct gfs2_inode *ip, int diff); struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, - int is_root, struct nameidata *nd); + int is_root); struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, unsigned int mode, dev_t dev); int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, @@ -91,9 +91,7 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, const struct gfs2_inode *ip); int gfs2_permission(struct inode *inode, int mask); -int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to); int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len); -int gfs2_glock_nq_atime(struct gfs2_holder *gh); int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c index 09d78c2..0c4cbe6 100644 --- a/fs/gfs2/locking/dlm/mount.c +++ b/fs/gfs2/locking/dlm/mount.c @@ -144,7 +144,8 @@ static int gdlm_mount(char *table_name, char *host_data, error = dlm_new_lockspace(ls->fsname, strlen(ls->fsname), &ls->dlm_lockspace, - DLM_LSFL_FS | (nodir ? DLM_LSFL_NODIR : 0), + DLM_LSFL_FS | DLM_LSFL_NEWEXCL | + (nodir ? DLM_LSFL_NODIR : 0), GDLM_LVB_SIZE); if (error) { log_error("dlm_new_lockspace error %d", error); diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 6c6af9f..ad30585 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -18,6 +18,7 @@ #include <linux/delay.h> #include <linux/kthread.h> #include <linux/freezer.h> +#include <linux/bio.h> #include "gfs2.h" #include "incore.h" @@ -584,7 +585,6 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) memset(bh->b_data, 0, bh->b_size); set_buffer_uptodate(bh); clear_buffer_dirty(bh); - unlock_buffer(bh); gfs2_ail1_empty(sdp, 0); tail = current_tail(sdp); @@ -601,8 +601,23 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) hash = gfs2_disk_hash(bh->b_data, sizeof(struct gfs2_log_header)); lh->lh_hash = cpu_to_be32(hash); - set_buffer_dirty(bh); - if (sync_dirty_buffer(bh)) + bh->b_end_io = end_buffer_write_sync; + if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags)) + goto skip_barrier; + get_bh(bh); + submit_bh(WRITE_BARRIER | (1 << BIO_RW_META), bh); + wait_on_buffer(bh); + if (buffer_eopnotsupp(bh)) { + clear_buffer_eopnotsupp(bh); + set_buffer_uptodate(bh); + set_bit(SDF_NOBARRIERS, &sdp->sd_flags); + lock_buffer(bh); +skip_barrier: + get_bh(bh); + submit_bh(WRITE_SYNC | (1 << BIO_RW_META), bh); + wait_on_buffer(bh); + } + if (!buffer_uptodate(bh)) gfs2_io_error_bh(sdp, bh); brelse(bh); diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index bcc668d..bb2cc30 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -24,7 +24,7 @@ #include "util.h" #include "glock.h" -static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo) +static void gfs2_init_inode_once(void *foo) { struct gfs2_inode *ip = foo; @@ -33,7 +33,7 @@ static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo) ip->i_alloc = NULL; } -static void gfs2_init_glock_once(struct kmem_cache *cachep, void *foo) +static void gfs2_init_glock_once(void *foo) { struct gfs2_glock *gl = foo; diff --git a/fs/gfs2/mount.c b/fs/gfs2/mount.c index b941f9f..df48333 100644 --- a/fs/gfs2/mount.c +++ b/fs/gfs2/mount.c @@ -42,6 +42,7 @@ enum { Opt_nosuiddir, Opt_data_writeback, Opt_data_ordered, + Opt_meta, Opt_err, }; @@ -66,6 +67,7 @@ static match_table_t tokens = { {Opt_nosuiddir, "nosuiddir"}, {Opt_data_writeback, "data=writeback"}, {Opt_data_ordered, "data=ordered"}, + {Opt_meta, "meta"}, {Opt_err, NULL} }; @@ -239,6 +241,11 @@ int gfs2_mount_args(struct gfs2_sbd *sdp, char *data_arg, int remount) case Opt_data_ordered: args->ar_data = GFS2_DATA_ORDERED; break; + case Opt_meta: + if (remount && args->ar_meta != 1) + goto cant_remount; + args->ar_meta = 1; + break; case Opt_err: default: fs_info(sdp, "unknown option: %s\n", o); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index e64a1b0..2756381 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -512,8 +512,8 @@ static int gfs2_readpage(struct file *file, struct page *page) int error; unlock_page(page); - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - error = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + error = gfs2_glock_nq(&gh); if (unlikely(error)) goto out; error = AOP_TRUNCATED_PAGE; @@ -594,8 +594,8 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, struct gfs2_holder gh; int ret; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - ret = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + ret = gfs2_glock_nq(&gh); if (unlikely(ret)) goto out_uninit; if (!gfs2_is_stuffed(ip)) @@ -636,8 +636,8 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, unsigned to = from + len; struct page *page; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &ip->i_gh); - error = gfs2_glock_nq_atime(&ip->i_gh); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); + error = gfs2_glock_nq(&ip->i_gh); if (unlikely(error)) goto out_uninit; @@ -975,7 +975,7 @@ static int gfs2_ok_for_dio(struct gfs2_inode *ip, int rw, loff_t offset) if (gfs2_is_stuffed(ip)) return 0; - if (offset > i_size_read(&ip->i_inode)) + if (offset >= i_size_read(&ip->i_inode)) return 0; return 1; } @@ -1000,8 +1000,8 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, * unfortunately have the option of only flushing a range like * the VFS does. */ - gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, GL_ATIME, &gh); - rv = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, &gh); + rv = gfs2_glock_nq(&gh); if (rv) return rv; rv = gfs2_ok_for_dio(ip, rw, offset); diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c index 990d9f4..9cda853 100644 --- a/fs/gfs2/ops_export.c +++ b/fs/gfs2/ops_export.c @@ -134,7 +134,7 @@ static struct dentry *gfs2_get_parent(struct dentry *child) struct dentry *dentry; gfs2_str2qstr(&dotdot, ".."); - inode = gfs2_lookupi(child->d_inode, &dotdot, 1, NULL); + inode = gfs2_lookupi(child->d_inode, &dotdot, 1); if (!inode) return ERR_PTR(-ENOENT); diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index e9a366d..3a747f8 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -89,8 +89,8 @@ static int gfs2_readdir(struct file *file, void *dirent, filldir_t filldir) u64 offset = file->f_pos; int error; - gfs2_holder_init(dip->i_gl, LM_ST_SHARED, GL_ATIME, &d_gh); - error = gfs2_glock_nq_atime(&d_gh); + gfs2_holder_init(dip->i_gl, LM_ST_SHARED, 0, &d_gh); + error = gfs2_glock_nq(&d_gh); if (error) { gfs2_holder_uninit(&d_gh); return error; @@ -153,8 +153,8 @@ static int gfs2_get_flags(struct file *filp, u32 __user *ptr) int error; u32 fsflags; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - error = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + error = gfs2_glock_nq(&gh); if (error) return error; @@ -351,8 +351,8 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) struct gfs2_alloc *al; int ret; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME, &gh); - ret = gfs2_glock_nq_atime(&gh); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); if (ret) goto out; @@ -434,8 +434,8 @@ static int gfs2_mmap(struct file *file, struct vm_area_struct *vma) struct gfs2_holder i_gh; int error; - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &i_gh); - error = gfs2_glock_nq_atime(&i_gh); + 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); return error; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index b4d1d64..b117fcf 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -40,6 +40,44 @@ #define DO 0 #define UNDO 1 +static const u32 gfs2_old_fs_formats[] = { + 0 +}; + +static const u32 gfs2_old_multihost_formats[] = { + 0 +}; + +/** + * gfs2_tune_init - Fill a gfs2_tune structure with default values + * @gt: tune + * + */ + +static void gfs2_tune_init(struct gfs2_tune *gt) +{ + spin_lock_init(>->gt_spin); + + gt->gt_demote_secs = 300; + gt->gt_incore_log_blocks = 1024; + gt->gt_log_flush_secs = 60; + gt->gt_recoverd_secs = 60; + gt->gt_logd_secs = 1; + gt->gt_quotad_secs = 5; + gt->gt_quota_simul_sync = 64; + gt->gt_quota_warn_period = 10; + gt->gt_quota_scale_num = 1; + gt->gt_quota_scale_den = 1; + gt->gt_quota_cache_secs = 300; + gt->gt_quota_quantum = 60; + gt->gt_new_files_jdata = 0; + gt->gt_max_readahead = 1 << 18; + gt->gt_stall_secs = 600; + gt->gt_complain_secs = 10; + gt->gt_statfs_quantum = 30; + gt->gt_statfs_slow = 0; +} + static struct gfs2_sbd *init_sbd(struct super_block *sb) { struct gfs2_sbd *sdp; @@ -96,21 +134,271 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) return sdp; } -static void init_vfs(struct super_block *sb, unsigned noatime) + +/** + * gfs2_check_sb - Check superblock + * @sdp: the filesystem + * @sb: The superblock + * @silent: Don't print a message if the check fails + * + * Checks the version code of the FS is one that we understand how to + * read and that the sizes of the various on-disk structures have not + * changed. + */ + +static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) { - struct gfs2_sbd *sdp = sb->s_fs_info; + unsigned int x; - sb->s_magic = GFS2_MAGIC; - sb->s_op = &gfs2_super_ops; - sb->s_export_op = &gfs2_export_ops; - sb->s_time_gran = 1; - sb->s_maxbytes = MAX_LFS_FILESIZE; + if (sb->sb_magic != GFS2_MAGIC || + sb->sb_type != GFS2_METATYPE_SB) { + if (!silent) + printk(KERN_WARNING "GFS2: not a GFS2 filesystem\n"); + return -EINVAL; + } + + /* If format numbers match exactly, we're done. */ + + if (sb->sb_fs_format == GFS2_FORMAT_FS && + sb->sb_multihost_format == GFS2_FORMAT_MULTI) + return 0; + + if (sb->sb_fs_format != GFS2_FORMAT_FS) { + for (x = 0; gfs2_old_fs_formats[x]; x++) + if (gfs2_old_fs_formats[x] == sb->sb_fs_format) + break; + + if (!gfs2_old_fs_formats[x]) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_WARNING + "GFS2: I don't know how to upgrade this FS\n"); + return -EINVAL; + } + } + + if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) { + for (x = 0; gfs2_old_multihost_formats[x]; x++) + if (gfs2_old_multihost_formats[x] == + sb->sb_multihost_format) + break; + + if (!gfs2_old_multihost_formats[x]) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_WARNING + "GFS2: I don't know how to upgrade this FS\n"); + return -EINVAL; + } + } + + if (!sdp->sd_args.ar_upgrade) { + printk(KERN_WARNING + "GFS2: code version (%u, %u) is incompatible " + "with ondisk format (%u, %u)\n", + GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, + sb->sb_fs_format, sb->sb_multihost_format); + printk(KERN_INFO + "GFS2: Use the \"upgrade\" mount option to upgrade " + "the FS\n"); + printk(KERN_INFO "GFS2: See the manual for more details\n"); + return -EINVAL; + } + + return 0; +} + +static void end_bio_io_page(struct bio *bio, int error) +{ + struct page *page = bio->bi_private; - if (sb->s_flags & (MS_NOATIME | MS_NODIRATIME)) - set_bit(noatime, &sdp->sd_flags); + if (!error) + SetPageUptodate(page); + else + printk(KERN_WARNING "gfs2: error %d reading superblock\n", error); + unlock_page(page); +} + +static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) +{ + const struct gfs2_sb *str = buf; + + sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); + sb->sb_type = be32_to_cpu(str->sb_header.mh_type); + sb->sb_format = be32_to_cpu(str->sb_header.mh_format); + sb->sb_fs_format = be32_to_cpu(str->sb_fs_format); + sb->sb_multihost_format = be32_to_cpu(str->sb_multihost_format); + sb->sb_bsize = be32_to_cpu(str->sb_bsize); + sb->sb_bsize_shift = be32_to_cpu(str->sb_bsize_shift); + sb->sb_master_dir.no_addr = be64_to_cpu(str->sb_master_dir.no_addr); + sb->sb_master_dir.no_formal_ino = be64_to_cpu(str->sb_master_dir.no_formal_ino); + sb->sb_root_dir.no_addr = be64_to_cpu(str->sb_root_dir.no_addr); + sb->sb_root_dir.no_formal_ino = be64_to_cpu(str->sb_root_dir.no_formal_ino); + + memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); + memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); +} + +/** + * gfs2_read_super - Read the gfs2 super block from disk + * @sdp: The GFS2 super block + * @sector: The location of the super block + * @error: The error code to return + * + * This uses the bio functions to read the super block from disk + * because we want to be 100% sure that we never read cached data. + * A super block is read twice only during each GFS2 mount and is + * never written to by the filesystem. The first time its read no + * locks are held, and the only details which are looked at are those + * relating to the locking protocol. Once locking is up and working, + * the sb is read again under the lock to establish the location of + * the master directory (contains pointers to journals etc) and the + * root directory. + * + * Returns: 0 on success or error + */ + +static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) +{ + struct super_block *sb = sdp->sd_vfs; + struct gfs2_sb *p; + struct page *page; + struct bio *bio; + + page = alloc_page(GFP_NOFS); + if (unlikely(!page)) + return -ENOBUFS; + + ClearPageUptodate(page); + ClearPageDirty(page); + lock_page(page); + + bio = bio_alloc(GFP_NOFS, 1); + if (unlikely(!bio)) { + __free_page(page); + return -ENOBUFS; + } - /* Don't let the VFS update atimes. GFS2 handles this itself. */ - sb->s_flags |= MS_NOATIME | MS_NODIRATIME; + bio->bi_sector = sector * (sb->s_blocksize >> 9); + bio->bi_bdev = sb->s_bdev; + bio_add_page(bio, page, PAGE_SIZE, 0); + + bio->bi_end_io = end_bio_io_page; + bio->bi_private = page; + submit_bio(READ_SYNC | (1 << BIO_RW_META), bio); + wait_on_page_locked(page); + bio_put(bio); + if (!PageUptodate(page)) { + __free_page(page); + return -EIO; + } + p = kmap(page); + gfs2_sb_in(&sdp->sd_sb, p); + kunmap(page); + __free_page(page); + return 0; +} +/** + * gfs2_read_sb - Read super block + * @sdp: The GFS2 superblock + * @gl: the glock for the superblock (assumed to be held) + * @silent: Don't print message if mount fails + * + */ + +static int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent) +{ + u32 hash_blocks, ind_blocks, leaf_blocks; + u32 tmp_blocks; + unsigned int x; + int error; + + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + 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; + sdp->sd_diptrs = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode)) / sizeof(u64); + sdp->sd_inptrs = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_meta_header)) / sizeof(u64); + sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header); + sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2; + sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1; + sdp->sd_hash_ptrs = sdp->sd_hash_bsize / sizeof(u64); + sdp->sd_qc_per_block = (sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_meta_header)) / + sizeof(struct gfs2_quota_change); + + /* Compute maximum reservation required to add a entry to a directory */ + + hash_blocks = DIV_ROUND_UP(sizeof(u64) * (1 << GFS2_DIR_MAX_DEPTH), + sdp->sd_jbsize); + + ind_blocks = 0; + for (tmp_blocks = hash_blocks; tmp_blocks > sdp->sd_diptrs;) { + tmp_blocks = DIV_ROUND_UP(tmp_blocks, sdp->sd_inptrs); + ind_blocks += tmp_blocks; + } + + leaf_blocks = 2 + GFS2_DIR_MAX_DEPTH; + + sdp->sd_max_dirres = hash_blocks + ind_blocks + leaf_blocks; + + sdp->sd_heightsize[0] = sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode); + sdp->sd_heightsize[1] = sdp->sd_sb.sb_bsize * sdp->sd_diptrs; + for (x = 2;; x++) { + u64 space, d; + u32 m; + + space = sdp->sd_heightsize[x - 1] * sdp->sd_inptrs; + d = space; + m = do_div(d, sdp->sd_inptrs); + + if (d != sdp->sd_heightsize[x - 1] || m) + break; + sdp->sd_heightsize[x] = space; + } + sdp->sd_max_height = x; + sdp->sd_heightsize[x] = ~0; + gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT); + + sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize - + sizeof(struct gfs2_dinode); + sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs; + for (x = 2;; x++) { + u64 space, d; + u32 m; + + space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs; + d = space; + m = do_div(d, sdp->sd_inptrs); + + if (d != sdp->sd_jheightsize[x - 1] || m) + break; + sdp->sd_jheightsize[x] = space; + } + sdp->sd_max_jheight = x; + sdp->sd_jheightsize[x] = ~0; + gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT); + + return 0; } static int init_names(struct gfs2_sbd *sdp, int silent) @@ -224,51 +512,59 @@ fail: return error; } -static inline struct inode *gfs2_lookup_root(struct super_block *sb, - u64 no_addr) +static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr, + u64 no_addr, const char *name) { - return gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0); + struct gfs2_sbd *sdp = sb->s_fs_info; + struct dentry *dentry; + struct inode *inode; + + inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0); + if (IS_ERR(inode)) { + fs_err(sdp, "can't read in %s inode: %ld\n", name, PTR_ERR(inode)); + return PTR_ERR(inode); + } + dentry = d_alloc_root(inode); + if (!dentry) { + fs_err(sdp, "can't alloc %s dentry\n", name); + iput(inode); + return -ENOMEM; + } + dentry->d_op = &gfs2_dops; + *dptr = dentry; + return 0; } -static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) +static int init_sb(struct gfs2_sbd *sdp, int silent) { struct super_block *sb = sdp->sd_vfs; struct gfs2_holder sb_gh; u64 no_addr; - struct inode *inode; - int error = 0; + int ret; - if (undo) { - if (sb->s_root) { - dput(sb->s_root); - sb->s_root = NULL; - } - return 0; + ret = gfs2_glock_nq_num(sdp, GFS2_SB_LOCK, &gfs2_meta_glops, + LM_ST_SHARED, 0, &sb_gh); + if (ret) { + fs_err(sdp, "can't acquire superblock glock: %d\n", ret); + return ret; } - error = gfs2_glock_nq_num(sdp, GFS2_SB_LOCK, &gfs2_meta_glops, - LM_ST_SHARED, 0, &sb_gh); - if (error) { - fs_err(sdp, "can't acquire superblock glock: %d\n", error); - return error; - } - - error = gfs2_read_sb(sdp, sb_gh.gh_gl, silent); - if (error) { - fs_err(sdp, "can't read superblock: %d\n", error); + ret = gfs2_read_sb(sdp, sb_gh.gh_gl, silent); + if (ret) { + fs_err(sdp, "can't read superblock: %d\n", ret); goto out; } /* Set up the buffer cache and SB for real */ if (sdp->sd_sb.sb_bsize < bdev_hardsect_size(sb->s_bdev)) { - error = -EINVAL; + ret = -EINVAL; fs_err(sdp, "FS block size (%u) is too small for device " "block size (%u)\n", sdp->sd_sb.sb_bsize, bdev_hardsect_size(sb->s_bdev)); goto out; } if (sdp->sd_sb.sb_bsize > PAGE_SIZE) { - error = -EINVAL; + ret = -EINVAL; fs_err(sdp, "FS block size (%u) is too big for machine " "page size (%u)\n", sdp->sd_sb.sb_bsize, (unsigned int)PAGE_SIZE); @@ -278,26 +574,21 @@ static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) /* Get the root inode */ no_addr = sdp->sd_sb.sb_root_dir.no_addr; - if (sb->s_type == &gfs2meta_fs_type) - no_addr = sdp->sd_sb.sb_master_dir.no_addr; - inode = gfs2_lookup_root(sb, no_addr); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); - fs_err(sdp, "can't read in root inode: %d\n", error); + ret = gfs2_lookup_root(sb, &sdp->sd_root_dir, no_addr, "root"); + if (ret) goto out; - } - sb->s_root = d_alloc_root(inode); - if (!sb->s_root) { - fs_err(sdp, "can't get root dentry\n"); - error = -ENOMEM; - iput(inode); - } else - sb->s_root->d_op = &gfs2_dops; - + /* Get the master inode */ + no_addr = sdp->sd_sb.sb_master_dir.no_addr; + ret = gfs2_lookup_root(sb, &sdp->sd_master_dir, no_addr, "master"); + if (ret) { + dput(sdp->sd_root_dir); + goto out; + } + sb->s_root = dget(sdp->sd_args.ar_meta ? sdp->sd_master_dir : sdp->sd_root_dir); out: gfs2_glock_dq_uninit(&sb_gh); - return error; + return ret; } /** @@ -372,6 +663,7 @@ static void gfs2_lm_others_may_mount(struct gfs2_sbd *sdp) static int init_journal(struct gfs2_sbd *sdp, int undo) { + struct inode *master = sdp->sd_master_dir->d_inode; struct gfs2_holder ji_gh; struct task_struct *p; struct gfs2_inode *ip; @@ -383,7 +675,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) goto fail_recoverd; } - sdp->sd_jindex = gfs2_lookup_simple(sdp->sd_master_dir, "jindex"); + sdp->sd_jindex = gfs2_lookup_simple(master, "jindex"); if (IS_ERR(sdp->sd_jindex)) { fs_err(sdp, "can't lookup journal index: %d\n", error); return PTR_ERR(sdp->sd_jindex); @@ -506,25 +798,17 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) { int error = 0; struct gfs2_inode *ip; - struct inode *inode; + struct inode *master = sdp->sd_master_dir->d_inode; if (undo) goto fail_qinode; - inode = gfs2_lookup_root(sdp->sd_vfs, sdp->sd_sb.sb_master_dir.no_addr); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); - fs_err(sdp, "can't read in master directory: %d\n", error); - goto fail; - } - sdp->sd_master_dir = inode; - error = init_journal(sdp, undo); if (error) - goto fail_master; + goto fail; /* Read in the master inode number inode */ - sdp->sd_inum_inode = gfs2_lookup_simple(sdp->sd_master_dir, "inum"); + sdp->sd_inum_inode = gfs2_lookup_simple(master, "inum"); if (IS_ERR(sdp->sd_inum_inode)) { error = PTR_ERR(sdp->sd_inum_inode); fs_err(sdp, "can't read in inum inode: %d\n", error); @@ -533,7 +817,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) /* Read in the master statfs inode */ - sdp->sd_statfs_inode = gfs2_lookup_simple(sdp->sd_master_dir, "statfs"); + sdp->sd_statfs_inode = gfs2_lookup_simple(master, "statfs"); if (IS_ERR(sdp->sd_statfs_inode)) { error = PTR_ERR(sdp->sd_statfs_inode); fs_err(sdp, "can't read in statfs inode: %d\n", error); @@ -541,7 +825,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) } /* Read in the resource index inode */ - sdp->sd_rindex = gfs2_lookup_simple(sdp->sd_master_dir, "rindex"); + sdp->sd_rindex = gfs2_lookup_simple(master, "rindex"); if (IS_ERR(sdp->sd_rindex)) { error = PTR_ERR(sdp->sd_rindex); fs_err(sdp, "can't get resource index inode: %d\n", error); @@ -552,7 +836,7 @@ static int init_inodes(struct gfs2_sbd *sdp, int undo) sdp->sd_rindex_uptodate = 0; /* Read in the quota inode */ - sdp->sd_quota_inode = gfs2_lookup_simple(sdp->sd_master_dir, "quota"); + sdp->sd_quota_inode = gfs2_lookup_simple(master, "quota"); if (IS_ERR(sdp->sd_quota_inode)) { error = PTR_ERR(sdp->sd_quota_inode); fs_err(sdp, "can't get quota file inode: %d\n", error); @@ -571,8 +855,6 @@ fail_inum: iput(sdp->sd_inum_inode); fail_journal: init_journal(sdp, UNDO); -fail_master: - iput(sdp->sd_master_dir); fail: return error; } @@ -583,6 +865,7 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo) char buf[30]; int error = 0; struct gfs2_inode *ip; + struct inode *master = sdp->sd_master_dir->d_inode; if (sdp->sd_args.ar_spectator) return 0; @@ -590,7 +873,7 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo) if (undo) goto fail_qc_gh; - pn = gfs2_lookup_simple(sdp->sd_master_dir, "per_node"); + pn = gfs2_lookup_simple(master, "per_node"); if (IS_ERR(pn)) { error = PTR_ERR(pn); fs_err(sdp, "can't find per_node directory: %d\n", error); @@ -800,7 +1083,11 @@ static int fill_super(struct super_block *sb, void *data, int silent) goto fail; } - init_vfs(sb, SDF_NOATIME); + sb->s_magic = GFS2_MAGIC; + sb->s_op = &gfs2_super_ops; + sb->s_export_op = &gfs2_export_ops; + sb->s_time_gran = 1; + sb->s_maxbytes = MAX_LFS_FILESIZE; /* Set up the buffer cache and fill in some fake block size values to allow us to read-in the on-disk superblock. */ @@ -828,7 +1115,7 @@ static int fill_super(struct super_block *sb, void *data, int silent) if (error) goto fail_lm; - error = init_sb(sdp, silent, DO); + error = init_sb(sdp, silent); if (error) goto fail_locking; @@ -869,7 +1156,11 @@ fail_per_node: fail_inodes: init_inodes(sdp, UNDO); fail_sb: - init_sb(sdp, 0, UNDO); + if (sdp->sd_root_dir) + dput(sdp->sd_root_dir); + if (sdp->sd_master_dir) + dput(sdp->sd_master_dir); + sb->s_root = NULL; fail_locking: init_locking(sdp, &mount_gh, UNDO); fail_lm: @@ -887,151 +1178,63 @@ fail: } static int gfs2_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) + const char *dev_name, void *data, struct vfsmount *mnt) { - struct super_block *sb; - struct gfs2_sbd *sdp; - int error = get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt); - if (error) - goto out; - sb = mnt->mnt_sb; - sdp = sb->s_fs_info; - sdp->sd_gfs2mnt = mnt; -out: - return error; + return get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt); } -static int fill_super_meta(struct super_block *sb, struct super_block *new, - void *data, int silent) +static struct super_block *get_gfs2_sb(const char *dev_name) { - struct gfs2_sbd *sdp = sb->s_fs_info; - struct inode *inode; - int error = 0; - - new->s_fs_info = sdp; - sdp->sd_vfs_meta = sb; - - init_vfs(new, SDF_NOATIME); - - /* Get the master inode */ - inode = igrab(sdp->sd_master_dir); - - new->s_root = d_alloc_root(inode); - if (!new->s_root) { - fs_err(sdp, "can't get root dentry\n"); - error = -ENOMEM; - iput(inode); - } else - new->s_root->d_op = &gfs2_dops; - - return error; -} - -static int set_bdev_super(struct super_block *s, void *data) -{ - s->s_bdev = data; - s->s_dev = s->s_bdev->bd_dev; - return 0; -} - -static int test_bdev_super(struct super_block *s, void *data) -{ - return s->s_bdev == data; -} - -static struct super_block* get_gfs2_sb(const char *dev_name) -{ - struct kstat stat; + struct super_block *sb; struct nameidata nd; - struct super_block *sb = NULL, *s; int error; error = path_lookup(dev_name, LOOKUP_FOLLOW, &nd); if (error) { - printk(KERN_WARNING "GFS2: path_lookup on %s returned error\n", - dev_name); - goto out; - } - error = vfs_getattr(nd.path.mnt, nd.path.dentry, &stat); - - list_for_each_entry(s, &gfs2_fs_type.fs_supers, s_instances) { - if ((S_ISBLK(stat.mode) && s->s_dev == stat.rdev) || - (S_ISDIR(stat.mode) && - s == nd.path.dentry->d_inode->i_sb)) { - sb = s; - goto free_nd; - } + printk(KERN_WARNING "GFS2: path_lookup on %s returned error %d\n", + dev_name, error); + return NULL; } - - printk(KERN_WARNING "GFS2: Unrecognized block device or " - "mount point %s\n", dev_name); - -free_nd: + sb = nd.path.dentry->d_inode->i_sb; + if (sb && (sb->s_type == &gfs2_fs_type)) + atomic_inc(&sb->s_active); + else + sb = NULL; path_put(&nd.path); -out: return sb; } static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, struct vfsmount *mnt) { - int error = 0; - struct super_block *sb = NULL, *new; + struct super_block *sb = NULL; struct gfs2_sbd *sdp; sb = get_gfs2_sb(dev_name); if (!sb) { printk(KERN_WARNING "GFS2: gfs2 mount does not exist\n"); - error = -ENOENT; - goto error; + return -ENOENT; } sdp = sb->s_fs_info; - if (sdp->sd_vfs_meta) { - printk(KERN_WARNING "GFS2: gfs2meta mount already exists\n"); - error = -EBUSY; - goto error; - } - down(&sb->s_bdev->bd_mount_sem); - new = sget(fs_type, test_bdev_super, set_bdev_super, sb->s_bdev); - up(&sb->s_bdev->bd_mount_sem); - if (IS_ERR(new)) { - error = PTR_ERR(new); - goto error; - } - new->s_flags = flags; - strlcpy(new->s_id, sb->s_id, sizeof(new->s_id)); - sb_set_blocksize(new, sb->s_blocksize); - error = fill_super_meta(sb, new, data, flags & MS_SILENT ? 1 : 0); - if (error) { - up_write(&new->s_umount); - deactivate_super(new); - goto error; - } - - new->s_flags |= MS_ACTIVE; - - /* Grab a reference to the gfs2 mount point */ - atomic_inc(&sdp->sd_gfs2mnt->mnt_count); - return simple_set_mnt(mnt, new); -error: - return error; + mnt->mnt_sb = sb; + mnt->mnt_root = dget(sdp->sd_master_dir); + return 0; } static void gfs2_kill_sb(struct super_block *sb) { - if (sb->s_fs_info) { - gfs2_delete_debugfs_file(sb->s_fs_info); - gfs2_meta_syncfs(sb->s_fs_info); + struct gfs2_sbd *sdp = sb->s_fs_info; + if (sdp) { + gfs2_meta_syncfs(sdp); + dput(sdp->sd_root_dir); + dput(sdp->sd_master_dir); + sdp->sd_root_dir = NULL; + sdp->sd_master_dir = NULL; } + shrink_dcache_sb(sb); kill_block_super(sb); -} - -static void gfs2_kill_sb_meta(struct super_block *sb) -{ - struct gfs2_sbd *sdp = sb->s_fs_info; - generic_shutdown_super(sb); - sdp->sd_vfs_meta = NULL; - atomic_dec(&sdp->sd_gfs2mnt->mnt_count); + if (sdp) + gfs2_delete_debugfs_file(sdp); } struct file_system_type gfs2_fs_type = { @@ -1046,7 +1249,6 @@ struct file_system_type gfs2meta_fs_type = { .name = "gfs2meta", .fs_flags = FS_REQUIRES_DEV, .get_sb = gfs2_get_sb_meta, - .kill_sb = gfs2_kill_sb_meta, .owner = THIS_MODULE, }; diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 1e252df..534e1e2 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -74,7 +74,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, return PTR_ERR(inode); } - inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd); + inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode) { if (!IS_ERR(inode)) { gfs2_holder_uninit(ghs); @@ -109,7 +109,7 @@ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, dentry->d_op = &gfs2_dops; - inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd); + inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode && IS_ERR(inode)) return ERR_CAST(inode); @@ -159,9 +159,13 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, 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_m(2, ghs); + error = gfs2_glock_nq(ghs); /* parent */ if (error) - goto out; + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC); if (error) @@ -245,8 +249,10 @@ out_alloc: if (alloc_required) gfs2_alloc_put(dip); out_gunlock: - gfs2_glock_dq_m(2, ghs); -out: + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_glock_dq(ghs); +out_parent: gfs2_holder_uninit(ghs); gfs2_holder_uninit(ghs + 1); if (!error) { @@ -302,7 +308,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) error = gfs2_unlink_ok(dip, &dentry->d_name, ip); if (error) - goto out_rgrp; + goto out_gunlock; error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0); if (error) @@ -316,6 +322,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) out_end_trans: gfs2_trans_end(sdp); +out_gunlock: gfs2_glock_dq(ghs + 2); out_rgrp: gfs2_holder_uninit(ghs + 2); @@ -485,7 +492,6 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) struct gfs2_holder ri_gh; int error; - error = gfs2_rindex_hold(sdp, &ri_gh); if (error) return error; @@ -495,9 +501,17 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - error = gfs2_glock_nq_m(3, ghs); + error = gfs2_glock_nq(ghs); /* parent */ if (error) - goto out; + 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) @@ -523,11 +537,15 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) gfs2_trans_end(sdp); out_gunlock: - gfs2_glock_dq_m(3, ghs); -out: - gfs2_holder_uninit(ghs); - gfs2_holder_uninit(ghs + 1); + 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; } @@ -571,6 +589,54 @@ static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, 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; + struct qstr dotdot; + int error = 0; + + gfs2_str2qstr(&dotdot, ".."); + + 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, &dotdot, 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 @@ -589,7 +655,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, 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; + struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }; struct gfs2_rgrpd *nrgd; unsigned int num_gh; int dir_rename = 0; @@ -603,19 +669,20 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, return 0; } - /* Make sure we aren't trying to move a dirctory into it's subdir */ - - if (S_ISDIR(ip->i_inode.i_mode) && odip != ndip) { - dir_rename = 1; - error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, 0, - &r_gh); + if (odip != ndip) { + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, + 0, &r_gh); if (error) goto out; - error = gfs2_ok_to_move(ip, ndip); - if (error) - goto out_gunlock_r; + 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; @@ -639,9 +706,11 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); } - error = gfs2_glock_nq_m(num_gh, ghs); - if (error) - goto out_uninit; + for (x = 0; x < num_gh; x++) { + error = gfs2_glock_nq(ghs + x); + if (error) + goto out_gunlock; + } /* Check out the old directory */ @@ -804,12 +873,12 @@ out_alloc: if (alloc_required) gfs2_alloc_put(ndip); out_gunlock: - gfs2_glock_dq_m(num_gh, ghs); -out_uninit: - for (x = 0; x < num_gh; x++) + while (x--) { + gfs2_glock_dq(ghs + x); gfs2_holder_uninit(ghs + x); + } out_gunlock_r: - if (dir_rename) + if (r_gh.gh_gl) gfs2_glock_dq_uninit(&r_gh); out: return error; @@ -915,12 +984,6 @@ int gfs2_permission(struct inode *inode, int mask) return error; } -static int gfs2_iop_permission(struct inode *inode, int mask, - struct nameidata *nd) -{ - return gfs2_permission(inode, mask); -} - static int setattr_size(struct inode *inode, struct iattr *attr) { struct gfs2_inode *ip = GFS2_I(inode); @@ -1150,7 +1213,7 @@ static int gfs2_removexattr(struct dentry *dentry, const char *name) } const struct inode_operations gfs2_file_iops = { - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, @@ -1169,7 +1232,7 @@ const struct inode_operations gfs2_dir_iops = { .rmdir = gfs2_rmdir, .mknod = gfs2_mknod, .rename = gfs2_rename, - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, @@ -1181,7 +1244,7 @@ const struct inode_operations gfs2_dir_iops = { const struct inode_operations gfs2_symlink_iops = { .readlink = gfs2_readlink, .follow_link = gfs2_follow_link, - .permission = gfs2_iop_permission, + .permission = gfs2_permission, .setattr = gfs2_setattr, .getattr = gfs2_getattr, .setxattr = gfs2_setxattr, diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index f66ea0f..d5355d9 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -20,6 +20,7 @@ #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> #include <linux/lm_interface.h> +#include <linux/time.h> #include "gfs2.h" #include "incore.h" @@ -38,6 +39,7 @@ #include "dir.h" #include "eattr.h" #include "bmap.h" +#include "meta_io.h" /** * gfs2_write_inode - Make sure the inode is stable on the disk @@ -50,16 +52,74 @@ static int gfs2_write_inode(struct inode *inode, int sync) { struct gfs2_inode *ip = GFS2_I(inode); - - /* Check this is a "normal" inode */ - if (test_bit(GIF_USER, &ip->i_flags)) { - if (current->flags & PF_MEMALLOC) - return 0; - if (sync) - gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + struct gfs2_sbd *sdp = GFS2_SB(inode); + struct gfs2_holder gh; + struct buffer_head *bh; + struct timespec atime; + struct gfs2_dinode *di; + int ret = 0; + + /* Check this is a "normal" inode, etc */ + if (!test_bit(GIF_USER, &ip->i_flags) || + (current->flags & PF_MEMALLOC)) + return 0; + ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + if (ret) + goto do_flush; + ret = gfs2_trans_begin(sdp, RES_DINODE, 0); + if (ret) + goto do_unlock; + ret = gfs2_meta_inode_buffer(ip, &bh); + if (ret == 0) { + di = (struct gfs2_dinode *)bh->b_data; + atime.tv_sec = be64_to_cpu(di->di_atime); + atime.tv_nsec = be32_to_cpu(di->di_atime_nsec); + if (timespec_compare(&inode->i_atime, &atime) > 0) { + gfs2_trans_add_bh(ip->i_gl, bh, 1); + gfs2_dinode_out(ip, bh->b_data); + } + brelse(bh); } + gfs2_trans_end(sdp); +do_unlock: + gfs2_glock_dq_uninit(&gh); +do_flush: + if (sync != 0) + gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + return ret; +} - return 0; +/** + * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one + * @sdp: the filesystem + * + * Returns: errno + */ + +static int gfs2_make_fs_ro(struct gfs2_sbd *sdp) +{ + struct gfs2_holder t_gh; + int error; + + gfs2_quota_sync(sdp); + gfs2_statfs_sync(sdp); + + error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE, + &t_gh); + if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) + return error; + + gfs2_meta_syncfs(sdp); + gfs2_log_shutdown(sdp); + + clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + + if (t_gh.gh_gl) + gfs2_glock_dq_uninit(&t_gh); + + gfs2_quota_cleanup(sdp); + + return error; } /** @@ -73,12 +133,6 @@ static void gfs2_put_super(struct super_block *sb) struct gfs2_sbd *sdp = sb->s_fs_info; int error; - if (!sdp) - return; - - if (!strncmp(sb->s_type->name, "gfs2meta", 8)) - return; /* Nothing to do */ - /* Unfreeze the filesystem, if we need to */ mutex_lock(&sdp->sd_freeze_lock); @@ -101,7 +155,6 @@ static void gfs2_put_super(struct super_block *sb) /* Release stuff */ - iput(sdp->sd_master_dir); iput(sdp->sd_jindex); iput(sdp->sd_inum_inode); iput(sdp->sd_statfs_inode); @@ -152,6 +205,7 @@ static void gfs2_write_super(struct super_block *sb) * * Flushes the log to disk. */ + static int gfs2_sync_fs(struct super_block *sb, int wait) { sb->s_dirt = 0; @@ -270,14 +324,6 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data) } } - if (*flags & (MS_NOATIME | MS_NODIRATIME)) - set_bit(SDF_NOATIME, &sdp->sd_flags); - else - clear_bit(SDF_NOATIME, &sdp->sd_flags); - - /* Don't let the VFS update atimes. GFS2 handles this itself. */ - *flags |= MS_NOATIME | MS_NODIRATIME; - return error; } @@ -295,6 +341,7 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data) * inode's blocks, or alternatively pass the baton on to another * node for later deallocation. */ + static void gfs2_drop_inode(struct inode *inode) { struct gfs2_inode *ip = GFS2_I(inode); @@ -333,6 +380,16 @@ static void gfs2_clear_inode(struct inode *inode) } } +static int is_ancestor(const struct dentry *d1, const struct dentry *d2) +{ + do { + if (d1 == d2) + return 1; + d1 = d1->d_parent; + } while (!IS_ROOT(d1)); + return 0; +} + /** * gfs2_show_options - Show mount options for /proc/mounts * @s: seq_file structure @@ -346,6 +403,8 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) struct gfs2_sbd *sdp = mnt->mnt_sb->s_fs_info; struct gfs2_args *args = &sdp->sd_args; + if (is_ancestor(mnt->mnt_root, sdp->sd_master_dir)) + seq_printf(s, ",meta"); if (args->ar_lockproto[0]) seq_printf(s, ",lockproto=%s", args->ar_lockproto); if (args->ar_locktable[0]) @@ -414,6 +473,7 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) * conversion on the iopen lock, but we can change that later. This * is safe, just less efficient. */ + static void gfs2_delete_inode(struct inode *inode) { struct gfs2_sbd *sdp = inode->i_sb->s_fs_info; @@ -478,8 +538,6 @@ out: clear_inode(inode); } - - static struct inode *gfs2_alloc_inode(struct super_block *sb) { struct gfs2_inode *ip; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 63a8a90..c3ba3d9 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -33,313 +33,6 @@ #include "trans.h" #include "util.h" -static const u32 gfs2_old_fs_formats[] = { - 0 -}; - -static const u32 gfs2_old_multihost_formats[] = { - 0 -}; - -/** - * gfs2_tune_init - Fill a gfs2_tune structure with default values - * @gt: tune - * - */ - -void gfs2_tune_init(struct gfs2_tune *gt) -{ - spin_lock_init(>->gt_spin); - - gt->gt_demote_secs = 300; - gt->gt_incore_log_blocks = 1024; - gt->gt_log_flush_secs = 60; - gt->gt_recoverd_secs = 60; - gt->gt_logd_secs = 1; - gt->gt_quotad_secs = 5; - gt->gt_quota_simul_sync = 64; - gt->gt_quota_warn_period = 10; - gt->gt_quota_scale_num = 1; - gt->gt_quota_scale_den = 1; - gt->gt_quota_cache_secs = 300; - gt->gt_quota_quantum = 60; - gt->gt_atime_quantum = 3600; - gt->gt_new_files_jdata = 0; - gt->gt_max_readahead = 1 << 18; - gt->gt_stall_secs = 600; - gt->gt_complain_secs = 10; - gt->gt_statfs_quantum = 30; - gt->gt_statfs_slow = 0; -} - -/** - * gfs2_check_sb - Check superblock - * @sdp: the filesystem - * @sb: The superblock - * @silent: Don't print a message if the check fails - * - * Checks the version code of the FS is one that we understand how to - * read and that the sizes of the various on-disk structures have not - * changed. - */ - -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) -{ - unsigned int x; - - if (sb->sb_magic != GFS2_MAGIC || - sb->sb_type != GFS2_METATYPE_SB) { - if (!silent) - printk(KERN_WARNING "GFS2: not a GFS2 filesystem\n"); - return -EINVAL; - } - - /* If format numbers match exactly, we're done. */ - - if (sb->sb_fs_format == GFS2_FORMAT_FS && - sb->sb_multihost_format == GFS2_FORMAT_MULTI) - return 0; - - if (sb->sb_fs_format != GFS2_FORMAT_FS) { - for (x = 0; gfs2_old_fs_formats[x]; x++) - if (gfs2_old_fs_formats[x] == sb->sb_fs_format) - break; - - if (!gfs2_old_fs_formats[x]) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_WARNING - "GFS2: I don't know how to upgrade this FS\n"); - return -EINVAL; - } - } - - if (sb->sb_multihost_format != GFS2_FORMAT_MULTI) { - for (x = 0; gfs2_old_multihost_formats[x]; x++) - if (gfs2_old_multihost_formats[x] == - sb->sb_multihost_format) - break; - - if (!gfs2_old_multihost_formats[x]) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_WARNING - "GFS2: I don't know how to upgrade this FS\n"); - return -EINVAL; - } - } - - if (!sdp->sd_args.ar_upgrade) { - printk(KERN_WARNING - "GFS2: code version (%u, %u) is incompatible " - "with ondisk format (%u, %u)\n", - GFS2_FORMAT_FS, GFS2_FORMAT_MULTI, - sb->sb_fs_format, sb->sb_multihost_format); - printk(KERN_INFO - "GFS2: Use the \"upgrade\" mount option to upgrade " - "the FS\n"); - printk(KERN_INFO "GFS2: See the manual for more details\n"); - return -EINVAL; - } - - return 0; -} - - -static void end_bio_io_page(struct bio *bio, int error) -{ - struct page *page = bio->bi_private; - - if (!error) - SetPageUptodate(page); - else - printk(KERN_WARNING "gfs2: error %d reading superblock\n", error); - unlock_page(page); -} - -static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) -{ - const struct gfs2_sb *str = buf; - - sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); - sb->sb_type = be32_to_cpu(str->sb_header.mh_type); - sb->sb_format = be32_to_cpu(str->sb_header.mh_format); - sb->sb_fs_format = be32_to_cpu(str->sb_fs_format); - sb->sb_multihost_format = be32_to_cpu(str->sb_multihost_format); - sb->sb_bsize = be32_to_cpu(str->sb_bsize); - sb->sb_bsize_shift = be32_to_cpu(str->sb_bsize_shift); - sb->sb_master_dir.no_addr = be64_to_cpu(str->sb_master_dir.no_addr); - sb->sb_master_dir.no_formal_ino = be64_to_cpu(str->sb_master_dir.no_formal_ino); - sb->sb_root_dir.no_addr = be64_to_cpu(str->sb_root_dir.no_addr); - sb->sb_root_dir.no_formal_ino = be64_to_cpu(str->sb_root_dir.no_formal_ino); - - memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); - memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); -} - -/** - * gfs2_read_super - Read the gfs2 super block from disk - * @sdp: The GFS2 super block - * @sector: The location of the super block - * @error: The error code to return - * - * This uses the bio functions to read the super block from disk - * because we want to be 100% sure that we never read cached data. - * A super block is read twice only during each GFS2 mount and is - * never written to by the filesystem. The first time its read no - * locks are held, and the only details which are looked at are those - * relating to the locking protocol. Once locking is up and working, - * the sb is read again under the lock to establish the location of - * the master directory (contains pointers to journals etc) and the - * root directory. - * - * Returns: 0 on success or error - */ - -int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) -{ - struct super_block *sb = sdp->sd_vfs; - struct gfs2_sb *p; - struct page *page; - struct bio *bio; - - page = alloc_page(GFP_NOFS); - if (unlikely(!page)) - return -ENOBUFS; - - ClearPageUptodate(page); - ClearPageDirty(page); - lock_page(page); - - bio = bio_alloc(GFP_NOFS, 1); - if (unlikely(!bio)) { - __free_page(page); - return -ENOBUFS; - } - - bio->bi_sector = sector * (sb->s_blocksize >> 9); - bio->bi_bdev = sb->s_bdev; - bio_add_page(bio, page, PAGE_SIZE, 0); - - bio->bi_end_io = end_bio_io_page; - bio->bi_private = page; - submit_bio(READ_SYNC | (1 << BIO_RW_META), bio); - wait_on_page_locked(page); - bio_put(bio); - if (!PageUptodate(page)) { - __free_page(page); - return -EIO; - } - p = kmap(page); - gfs2_sb_in(&sdp->sd_sb, p); - kunmap(page); - __free_page(page); - return 0; -} - -/** - * gfs2_read_sb - Read super block - * @sdp: The GFS2 superblock - * @gl: the glock for the superblock (assumed to be held) - * @silent: Don't print message if mount fails - * - */ - -int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent) -{ - u32 hash_blocks, ind_blocks, leaf_blocks; - u32 tmp_blocks; - unsigned int x; - int error; - - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); - 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; - sdp->sd_diptrs = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode)) / sizeof(u64); - sdp->sd_inptrs = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_meta_header)) / sizeof(u64); - sdp->sd_jbsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header); - sdp->sd_hash_bsize = sdp->sd_sb.sb_bsize / 2; - sdp->sd_hash_bsize_shift = sdp->sd_sb.sb_bsize_shift - 1; - sdp->sd_hash_ptrs = sdp->sd_hash_bsize / sizeof(u64); - sdp->sd_qc_per_block = (sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_meta_header)) / - sizeof(struct gfs2_quota_change); - - /* Compute maximum reservation required to add a entry to a directory */ - - hash_blocks = DIV_ROUND_UP(sizeof(u64) * (1 << GFS2_DIR_MAX_DEPTH), - sdp->sd_jbsize); - - ind_blocks = 0; - for (tmp_blocks = hash_blocks; tmp_blocks > sdp->sd_diptrs;) { - tmp_blocks = DIV_ROUND_UP(tmp_blocks, sdp->sd_inptrs); - ind_blocks += tmp_blocks; - } - - leaf_blocks = 2 + GFS2_DIR_MAX_DEPTH; - - sdp->sd_max_dirres = hash_blocks + ind_blocks + leaf_blocks; - - sdp->sd_heightsize[0] = sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode); - sdp->sd_heightsize[1] = sdp->sd_sb.sb_bsize * sdp->sd_diptrs; - for (x = 2;; x++) { - u64 space, d; - u32 m; - - space = sdp->sd_heightsize[x - 1] * sdp->sd_inptrs; - d = space; - m = do_div(d, sdp->sd_inptrs); - - if (d != sdp->sd_heightsize[x - 1] || m) - break; - sdp->sd_heightsize[x] = space; - } - sdp->sd_max_height = x; - sdp->sd_heightsize[x] = ~0; - gfs2_assert(sdp, sdp->sd_max_height <= GFS2_MAX_META_HEIGHT); - - sdp->sd_jheightsize[0] = sdp->sd_sb.sb_bsize - - sizeof(struct gfs2_dinode); - sdp->sd_jheightsize[1] = sdp->sd_jbsize * sdp->sd_diptrs; - for (x = 2;; x++) { - u64 space, d; - u32 m; - - space = sdp->sd_jheightsize[x - 1] * sdp->sd_inptrs; - d = space; - m = do_div(d, sdp->sd_inptrs); - - if (d != sdp->sd_jheightsize[x - 1] || m) - break; - sdp->sd_jheightsize[x] = space; - } - sdp->sd_max_jheight = x; - sdp->sd_jheightsize[x] = ~0; - gfs2_assert(sdp, sdp->sd_max_jheight <= GFS2_MAX_META_HEIGHT); - - return 0; -} - /** * gfs2_jindex_hold - Grab a lock on the jindex * @sdp: The GFS2 superblock @@ -389,7 +82,7 @@ int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) break; INIT_LIST_HEAD(&jd->extent_list); - jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1, NULL); + jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1); if (!jd->jd_inode || IS_ERR(jd->jd_inode)) { if (!jd->jd_inode) error = -ENOENT; @@ -581,39 +274,6 @@ fail: return error; } -/** - * gfs2_make_fs_ro - Turn a Read-Write FS into a Read-Only one - * @sdp: the filesystem - * - * Returns: errno - */ - -int gfs2_make_fs_ro(struct gfs2_sbd *sdp) -{ - struct gfs2_holder t_gh; - int error; - - gfs2_quota_sync(sdp); - gfs2_statfs_sync(sdp); - - error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, GL_NOCACHE, - &t_gh); - if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) - return error; - - gfs2_meta_syncfs(sdp); - gfs2_log_shutdown(sdp); - - clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); - - if (t_gh.gh_gl) - gfs2_glock_dq_uninit(&t_gh); - - gfs2_quota_cleanup(sdp); - - return error; -} - static void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf) { const struct gfs2_statfs_change *str = buf; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index 44361ec..50a4c9b 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -12,11 +12,6 @@ #include "incore.h" -void gfs2_tune_init(struct gfs2_tune *gt); - -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent); -int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent); -int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector); void gfs2_lm_unmount(struct gfs2_sbd *sdp); static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp) @@ -40,7 +35,6 @@ int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename, struct gfs2_inode **ipp); int gfs2_make_fs_rw(struct gfs2_sbd *sdp); -int gfs2_make_fs_ro(struct gfs2_sbd *sdp); int gfs2_statfs_init(struct gfs2_sbd *sdp); void gfs2_statfs_change(struct gfs2_sbd *sdp, diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 7484655..7e1879f 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -269,14 +269,6 @@ ARGS_ATTR(quota, "%u\n"); ARGS_ATTR(suiddir, "%d\n"); ARGS_ATTR(data, "%d\n"); -/* one oddball doesn't fit the macro mold */ -static ssize_t noatime_show(struct gfs2_sbd *sdp, char *buf) -{ - return snprintf(buf, PAGE_SIZE, "%d\n", - !!test_bit(SDF_NOATIME, &sdp->sd_flags)); -} -static struct args_attr args_attr_noatime = __ATTR_RO(noatime); - static struct attribute *args_attrs[] = { &args_attr_lockproto.attr, &args_attr_locktable.attr, @@ -292,7 +284,6 @@ static struct attribute *args_attrs[] = { &args_attr_quota.attr, &args_attr_suiddir.attr, &args_attr_data.attr, - &args_attr_noatime.attr, NULL, }; @@ -407,7 +398,6 @@ TUNE_ATTR(incore_log_blocks, 0); TUNE_ATTR(log_flush_secs, 0); TUNE_ATTR(quota_warn_period, 0); TUNE_ATTR(quota_quantum, 0); -TUNE_ATTR(atime_quantum, 0); TUNE_ATTR(max_readahead, 0); TUNE_ATTR(complain_secs, 0); TUNE_ATTR(statfs_slow, 0); @@ -427,7 +417,6 @@ static struct attribute *tune_attrs[] = { &tune_attr_log_flush_secs.attr, &tune_attr_quota_warn_period.attr, &tune_attr_quota_quantum.attr, - &tune_attr_atime_quantum.attr, &tune_attr_max_readahead.attr, &tune_attr_complain_secs.attr, &tune_attr_statfs_slow.attr, diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index dc4ec64..7e19835 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -511,8 +511,7 @@ void hfs_clear_inode(struct inode *inode) } } -static int hfs_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int hfs_permission(struct inode *inode, int mask) { if (S_ISREG(inode->i_mode) && mask & MAY_EXEC) return 0; @@ -523,8 +522,6 @@ static int hfs_file_open(struct inode *inode, struct file *file) { if (HFS_IS_RSRC(inode)) inode = HFS_I(inode)->rsrc_inode; - if (atomic_read(&file->f_count) != 1) - return 0; atomic_inc(&HFS_I(inode)->opencnt); return 0; } @@ -535,8 +532,6 @@ static int hfs_file_release(struct inode *inode, struct file *file) if (HFS_IS_RSRC(inode)) inode = HFS_I(inode)->rsrc_inode; - if (atomic_read(&file->f_count) != 0) - return 0; if (atomic_dec_and_test(&HFS_I(inode)->opencnt)) { mutex_lock(&inode->i_mutex); hfs_file_truncate(inode); diff --git a/fs/hfs/super.c b/fs/hfs/super.c index ac2ec5e..4abb104 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -432,7 +432,7 @@ static struct file_system_type hfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfs_init_once(struct kmem_cache *cachep, void *p) +static void hfs_init_once(void *p) { struct hfs_inode_info *i = p; diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index cc3b5e2..b085d64 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -238,7 +238,7 @@ static void hfsplus_set_perms(struct inode *inode, struct hfsplus_perm *perms) perms->dev = cpu_to_be32(HFSPLUS_I(inode).dev); } -static int hfsplus_permission(struct inode *inode, int mask, struct nameidata *nd) +static int hfsplus_permission(struct inode *inode, int mask) { /* MAY_EXEC is also used for lookup, if no x bit is set allow lookup, * open_exec has the same test, so it's still not executable, if a x bit @@ -254,8 +254,6 @@ static int hfsplus_file_open(struct inode *inode, struct file *file) { if (HFSPLUS_IS_RSRC(inode)) inode = HFSPLUS_I(inode).rsrc_inode; - if (atomic_read(&file->f_count) != 1) - return 0; atomic_inc(&HFSPLUS_I(inode).opencnt); return 0; } @@ -266,8 +264,6 @@ static int hfsplus_file_release(struct inode *inode, struct file *file) if (HFSPLUS_IS_RSRC(inode)) inode = HFSPLUS_I(inode).rsrc_inode; - if (atomic_read(&file->f_count) != 0) - return 0; if (atomic_dec_and_test(&HFSPLUS_I(inode).opencnt)) { mutex_lock(&inode->i_mutex); hfsplus_file_truncate(inode); diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 3859118..e834e57 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -485,7 +485,7 @@ static struct file_system_type hfsplus_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfsplus_init_once(struct kmem_cache *cachep, void *p) +static void hfsplus_init_once(void *p) { struct hfsplus_inode_info *i = p; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 5222345..d6ecabf 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -822,7 +822,7 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from, return err; } -int hostfs_permission(struct inode *ino, int desired, struct nameidata *nd) +int hostfs_permission(struct inode *ino, int desired) { char *name; int r = 0, w = 0, x = 0, err; diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index d256559..d9c59a7 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -415,7 +415,7 @@ again: d_drop(dentry); spin_lock(&dentry->d_lock); if (atomic_read(&dentry->d_count) > 1 || - permission(inode, MAY_WRITE, NULL) || + generic_permission(inode, MAY_WRITE, NULL) || !S_ISREG(inode->i_mode) || get_write_access(inode)) { spin_unlock(&dentry->d_lock); diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index f63a699..b8ae9c9 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -173,7 +173,7 @@ static void hpfs_destroy_inode(struct inode *inode) kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 65077aa..2b3d182 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -655,20 +655,13 @@ static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd) return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd); } -int hppfs_permission(struct inode *inode, int mask, struct nameidata *nd) -{ - return generic_permission(inode, mask, NULL); -} - static const struct inode_operations hppfs_dir_iops = { .lookup = hppfs_lookup, - .permission = hppfs_permission, }; static const struct inode_operations hppfs_link_iops = { .readlink = hppfs_readlink, .follow_link = hppfs_follow_link, - .permission = hppfs_permission, }; static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index dbd01d2..3f58923 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -705,7 +705,7 @@ static const struct address_space_operations hugetlbfs_aops = { }; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; @@ -166,6 +166,7 @@ static struct inode *alloc_inode(struct super_block *sb) mapping_set_gfp_mask(mapping, GFP_HIGHUSER_PAGECACHE); mapping->assoc_mapping = NULL; mapping->backing_dev_info = &default_backing_dev_info; + mapping->writeback_index = 0; /* * If the block_device provides a backing_dev_info for client @@ -209,7 +210,7 @@ void inode_init_once(struct inode *inode) INIT_LIST_HEAD(&inode->i_dentry); INIT_LIST_HEAD(&inode->i_devices); INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC); - rwlock_init(&inode->i_data.tree_lock); + spin_lock_init(&inode->i_data.tree_lock); spin_lock_init(&inode->i_data.i_mmap_lock); INIT_LIST_HEAD(&inode->i_data.private_list); spin_lock_init(&inode->i_data.private_lock); @@ -224,7 +225,7 @@ void inode_init_once(struct inode *inode) EXPORT_SYMBOL(inode_init_once); -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct inode * inode = (struct inode *) foo; diff --git a/fs/inotify_user.c b/fs/inotify_user.c index fe79c25d..d85c7d9 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -323,7 +323,7 @@ out: } /* - * remove_kevent - cleans up and ultimately frees the given kevent + * remove_kevent - cleans up the given kevent * * Caller must hold dev->ev_mutex. */ @@ -334,7 +334,13 @@ static void remove_kevent(struct inotify_device *dev, dev->event_count--; dev->queue_size -= sizeof(struct inotify_event) + kevent->event.len; +} +/* + * free_kevent - frees the given kevent. + */ +static void free_kevent(struct inotify_kernel_event *kevent) +{ kfree(kevent->name); kmem_cache_free(event_cachep, kevent); } @@ -350,24 +356,25 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev) struct inotify_kernel_event *kevent; kevent = inotify_dev_get_event(dev); remove_kevent(dev, kevent); + free_kevent(kevent); } } /* - * find_inode - resolve a user-given path to a specific inode and return a nd + * find_inode - resolve a user-given path to a specific inode */ -static int find_inode(const char __user *dirname, struct nameidata *nd, +static int find_inode(const char __user *dirname, struct path *path, unsigned flags) { int error; - error = __user_walk(dirname, flags, nd); + error = user_path_at(AT_FDCWD, dirname, flags, path); if (error) return error; /* you can only watch an inode if you have read permissions on it */ - error = vfs_permission(nd, MAY_READ); + error = inode_permission(path->dentry->d_inode, MAY_READ); if (error) - path_put(&nd->path); + path_put(path); return error; } @@ -433,17 +440,15 @@ static ssize_t inotify_read(struct file *file, char __user *buf, dev = file->private_data; while (1) { - int events; prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE); mutex_lock(&dev->ev_mutex); - events = !list_empty(&dev->events); - mutex_unlock(&dev->ev_mutex); - if (events) { + if (!list_empty(&dev->events)) { ret = 0; break; } + mutex_unlock(&dev->ev_mutex); if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; @@ -462,7 +467,6 @@ static ssize_t inotify_read(struct file *file, char __user *buf, if (ret) return ret; - mutex_lock(&dev->ev_mutex); while (1) { struct inotify_kernel_event *kevent; @@ -481,6 +485,13 @@ static ssize_t inotify_read(struct file *file, char __user *buf, } break; } + remove_kevent(dev, kevent); + + /* + * Must perform the copy_to_user outside the mutex in order + * to avoid a lock order reversal with mmap_sem. + */ + mutex_unlock(&dev->ev_mutex); if (copy_to_user(buf, &kevent->event, event_size)) { ret = -EFAULT; @@ -498,7 +509,9 @@ static ssize_t inotify_read(struct file *file, char __user *buf, count -= kevent->event.len; } - remove_kevent(dev, kevent); + free_kevent(kevent); + + mutex_lock(&dev->ev_mutex); } mutex_unlock(&dev->ev_mutex); @@ -650,11 +663,11 @@ asmlinkage long sys_inotify_init(void) return sys_inotify_init1(0); } -asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) +asmlinkage long sys_inotify_add_watch(int fd, const char __user *pathname, u32 mask) { struct inode *inode; struct inotify_device *dev; - struct nameidata nd; + struct path path; struct file *filp; int ret, fput_needed; unsigned flags = 0; @@ -674,12 +687,12 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) if (mask & IN_ONLYDIR) flags |= LOOKUP_DIRECTORY; - ret = find_inode(path, &nd, flags); + ret = find_inode(pathname, &path, flags); if (unlikely(ret)) goto fput_and_out; - /* inode held in place by reference to nd; dev by fget on fd */ - inode = nd.path.dentry->d_inode; + /* inode held in place by reference to path; dev by fget on fd */ + inode = path.dentry->d_inode; dev = filp->private_data; mutex_lock(&dev->up_mutex); @@ -688,7 +701,7 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) ret = create_watch(dev, inode, mask); mutex_unlock(&dev->up_mutex); - path_put(&nd.path); + path_put(&path); fput_and_out: fput_light(filp, fput_needed); return ret; @@ -13,9 +13,14 @@ #include <linux/security.h> #include <linux/module.h> #include <linux/uaccess.h> +#include <linux/writeback.h> +#include <linux/buffer_head.h> #include <asm/ioctls.h> +/* So that the fiemap access checks can't overflow on 32 bit machines. */ +#define FIEMAP_MAX_EXTENTS (UINT_MAX / sizeof(struct fiemap_extent)) + /** * vfs_ioctl - call filesystem specific ioctl methods * @filp: open file to invoke ioctl method on @@ -71,6 +76,276 @@ static int ioctl_fibmap(struct file *filp, int __user *p) return put_user(res, p); } +/** + * fiemap_fill_next_extent - Fiemap helper function + * @fieinfo: Fiemap context passed into ->fiemap + * @logical: Extent logical start offset, in bytes + * @phys: Extent physical start offset, in bytes + * @len: Extent length, in bytes + * @flags: FIEMAP_EXTENT flags that describe this extent + * + * Called from file system ->fiemap callback. Will populate extent + * info as passed in via arguments and copy to user memory. On + * success, extent count on fieinfo is incremented. + * + * Returns 0 on success, -errno on error, 1 if this was the last + * extent that will fit in user array. + */ +#define SET_UNKNOWN_FLAGS (FIEMAP_EXTENT_DELALLOC) +#define SET_NO_UNMOUNTED_IO_FLAGS (FIEMAP_EXTENT_DATA_ENCRYPTED) +#define SET_NOT_ALIGNED_FLAGS (FIEMAP_EXTENT_DATA_TAIL|FIEMAP_EXTENT_DATA_INLINE) +int fiemap_fill_next_extent(struct fiemap_extent_info *fieinfo, u64 logical, + u64 phys, u64 len, u32 flags) +{ + struct fiemap_extent extent; + struct fiemap_extent *dest = fieinfo->fi_extents_start; + + /* only count the extents */ + if (fieinfo->fi_extents_max == 0) { + fieinfo->fi_extents_mapped++; + return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0; + } + + if (fieinfo->fi_extents_mapped >= fieinfo->fi_extents_max) + return 1; + + if (flags & SET_UNKNOWN_FLAGS) + flags |= FIEMAP_EXTENT_UNKNOWN; + if (flags & SET_NO_UNMOUNTED_IO_FLAGS) + flags |= FIEMAP_EXTENT_ENCODED; + if (flags & SET_NOT_ALIGNED_FLAGS) + flags |= FIEMAP_EXTENT_NOT_ALIGNED; + + memset(&extent, 0, sizeof(extent)); + extent.fe_logical = logical; + extent.fe_physical = phys; + extent.fe_length = len; + extent.fe_flags = flags; + + dest += fieinfo->fi_extents_mapped; + if (copy_to_user(dest, &extent, sizeof(extent))) + return -EFAULT; + + fieinfo->fi_extents_mapped++; + if (fieinfo->fi_extents_mapped == fieinfo->fi_extents_max) + return 1; + return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0; +} +EXPORT_SYMBOL(fiemap_fill_next_extent); + +/** + * fiemap_check_flags - check validity of requested flags for fiemap + * @fieinfo: Fiemap context passed into ->fiemap + * @fs_flags: Set of fiemap flags that the file system understands + * + * Called from file system ->fiemap callback. This will compute the + * intersection of valid fiemap flags and those that the fs supports. That + * value is then compared against the user supplied flags. In case of bad user + * flags, the invalid values will be written into the fieinfo structure, and + * -EBADR is returned, which tells ioctl_fiemap() to return those values to + * userspace. For this reason, a return code of -EBADR should be preserved. + * + * Returns 0 on success, -EBADR on bad flags. + */ +int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags) +{ + u32 incompat_flags; + + incompat_flags = fieinfo->fi_flags & ~(FIEMAP_FLAGS_COMPAT & fs_flags); + if (incompat_flags) { + fieinfo->fi_flags = incompat_flags; + return -EBADR; + } + return 0; +} +EXPORT_SYMBOL(fiemap_check_flags); + +static int fiemap_check_ranges(struct super_block *sb, + u64 start, u64 len, u64 *new_len) +{ + *new_len = len; + + if (len == 0) + return -EINVAL; + + if (start > sb->s_maxbytes) + return -EFBIG; + + /* + * Shrink request scope to what the fs can actually handle. + */ + if ((len > sb->s_maxbytes) || + (sb->s_maxbytes - len) < start) + *new_len = sb->s_maxbytes - start; + + return 0; +} + +static int ioctl_fiemap(struct file *filp, unsigned long arg) +{ + struct fiemap fiemap; + struct fiemap_extent_info fieinfo = { 0, }; + struct inode *inode = filp->f_path.dentry->d_inode; + struct super_block *sb = inode->i_sb; + u64 len; + int error; + + if (!inode->i_op->fiemap) + return -EOPNOTSUPP; + + if (copy_from_user(&fiemap, (struct fiemap __user *)arg, + sizeof(struct fiemap))) + return -EFAULT; + + if (fiemap.fm_extent_count > FIEMAP_MAX_EXTENTS) + return -EINVAL; + + error = fiemap_check_ranges(sb, fiemap.fm_start, fiemap.fm_length, + &len); + if (error) + return error; + + fieinfo.fi_flags = fiemap.fm_flags; + fieinfo.fi_extents_max = fiemap.fm_extent_count; + fieinfo.fi_extents_start = (struct fiemap_extent *)(arg + sizeof(fiemap)); + + if (fiemap.fm_extent_count != 0 && + !access_ok(VERIFY_WRITE, fieinfo.fi_extents_start, + fieinfo.fi_extents_max * sizeof(struct fiemap_extent))) + return -EFAULT; + + if (fieinfo.fi_flags & FIEMAP_FLAG_SYNC) + filemap_write_and_wait(inode->i_mapping); + + error = inode->i_op->fiemap(inode, &fieinfo, fiemap.fm_start, len); + fiemap.fm_flags = fieinfo.fi_flags; + fiemap.fm_mapped_extents = fieinfo.fi_extents_mapped; + if (copy_to_user((char *)arg, &fiemap, sizeof(fiemap))) + error = -EFAULT; + + return error; +} + +#ifdef CONFIG_BLOCK + +#define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits) +#define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits); + +/* + * @inode - the inode to map + * @arg - the pointer to userspace where we copy everything to + * @get_block - the fs's get_block function + * + * This does FIEMAP for block based inodes. Basically it will just loop + * through get_block until we hit the number of extents we want to map, or we + * go past the end of the file and hit a hole. + * + * If it is possible to have data blocks beyond a hole past @inode->i_size, then + * please do not use this function, it will stop at the first unmapped block + * beyond i_size + */ +int generic_block_fiemap(struct inode *inode, + struct fiemap_extent_info *fieinfo, u64 start, + u64 len, get_block_t *get_block) +{ + struct buffer_head tmp; + unsigned int start_blk; + long long length = 0, map_len = 0; + u64 logical = 0, phys = 0, size = 0; + u32 flags = FIEMAP_EXTENT_MERGED; + int ret = 0; + + if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC))) + return ret; + + start_blk = logical_to_blk(inode, start); + + /* guard against change */ + mutex_lock(&inode->i_mutex); + + length = (long long)min_t(u64, len, i_size_read(inode)); + map_len = length; + + do { + /* + * we set b_size to the total size we want so it will map as + * many contiguous blocks as possible at once + */ + memset(&tmp, 0, sizeof(struct buffer_head)); + tmp.b_size = map_len; + + ret = get_block(inode, start_blk, &tmp, 0); + if (ret) + break; + + /* HOLE */ + if (!buffer_mapped(&tmp)) { + /* + * first hole after going past the EOF, this is our + * last extent + */ + if (length <= 0) { + flags = FIEMAP_EXTENT_MERGED|FIEMAP_EXTENT_LAST; + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, + flags); + break; + } + + length -= blk_to_logical(inode, 1); + + /* if we have holes up to/past EOF then we're done */ + if (length <= 0) + break; + + start_blk++; + } else { + if (length <= 0 && size) { + ret = fiemap_fill_next_extent(fieinfo, logical, + phys, size, + flags); + if (ret) + break; + } + + logical = blk_to_logical(inode, start_blk); + phys = blk_to_logical(inode, tmp.b_blocknr); + size = tmp.b_size; + flags = FIEMAP_EXTENT_MERGED; + + length -= tmp.b_size; + start_blk += logical_to_blk(inode, size); + + /* + * if we are past the EOF we need to loop again to see + * if there is a hole so we can mark this extent as the + * last one, and if not keep mapping things until we + * find a hole, or we run out of slots in the extent + * array + */ + if (length <= 0) + continue; + + ret = fiemap_fill_next_extent(fieinfo, logical, phys, + size, flags); + if (ret) + break; + } + cond_resched(); + } while (1); + + mutex_unlock(&inode->i_mutex); + + /* if ret is 1 then we just hit the end of the extent array */ + if (ret == 1) + ret = 0; + + return ret; +} +EXPORT_SYMBOL(generic_block_fiemap); + +#endif /* CONFIG_BLOCK */ + static int file_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -80,6 +355,8 @@ static int file_ioctl(struct file *filp, unsigned int cmd, switch (cmd) { case FIBMAP: return ioctl_fibmap(filp, p); + case FS_IOC_FIEMAP: + return ioctl_fiemap(filp, arg); case FIGETBSZ: return put_user(inode->i_sb->s_blocksize, p); case FIONREAD: diff --git a/fs/ioprio.c b/fs/ioprio.c index c4a1c3c..da3cc46 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -115,11 +115,11 @@ asmlinkage long sys_ioprio_set(int which, int who, int ioprio) pgrp = task_pgrp(current); else pgrp = find_vpid(who); - do_each_pid_task(pgrp, PIDTYPE_PGID, p) { + do_each_pid_thread(pgrp, PIDTYPE_PGID, p) { ret = set_task_ioprio(p, ioprio); if (ret) break; - } while_each_pid_task(pgrp, PIDTYPE_PGID, p); + } while_each_pid_thread(pgrp, PIDTYPE_PGID, p); break; case IOPRIO_WHO_USER: if (!who) @@ -204,7 +204,7 @@ asmlinkage long sys_ioprio_get(int which, int who) pgrp = task_pgrp(current); else pgrp = find_vpid(who); - do_each_pid_task(pgrp, PIDTYPE_PGID, p) { + do_each_pid_thread(pgrp, PIDTYPE_PGID, p) { tmpio = get_task_ioprio(p); if (tmpio < 0) continue; @@ -212,7 +212,7 @@ asmlinkage long sys_ioprio_get(int which, int who) ret = tmpio; else ret = ioprio_best(ret, tmpio); - } while_each_pid_task(pgrp, PIDTYPE_PGID, p); + } while_each_pid_thread(pgrp, PIDTYPE_PGID, p); break; case IOPRIO_WHO_USER: if (!who) diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 044a254..26948a6 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode) kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct iso_inode_info *ei = foo; diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 2eccbfa..ae08c05 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -63,7 +63,7 @@ static void release_buffer_page(struct buffer_head *bh) goto nope; /* OK, it's a truncated page */ - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto nope; page_cache_get(page); @@ -221,7 +221,7 @@ write_out_data: * blocking lock_buffer(). */ if (buffer_dirty(bh)) { - if (test_set_buffer_locked(bh)) { + if (!trylock_buffer(bh)) { BUFFER_TRACE(bh, "needs blocking lock"); spin_unlock(&journal->j_list_lock); /* Write out all data to prevent deadlocks */ @@ -446,7 +446,7 @@ void journal_commit_transaction(journal_t *journal) spin_lock(&journal->j_list_lock); } if (unlikely(!buffer_uptodate(bh))) { - if (TestSetPageLocked(bh->b_page)) { + if (!trylock_page(bh->b_page)) { spin_unlock(&journal->j_list_lock); lock_page(bh->b_page); spin_lock(&journal->j_list_lock); diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 8dee320..0540ca2 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -291,7 +291,7 @@ handle_t *journal_start(journal_t *journal, int nblocks) goto out; } - lock_acquire(&handle->h_lockdep_map, 0, 0, 0, 2, _THIS_IP_); + lock_map_acquire(&handle->h_lockdep_map); out: return handle; @@ -1448,7 +1448,7 @@ int journal_stop(handle_t *handle) spin_unlock(&journal->j_state_lock); } - lock_release(&handle->h_lockdep_map, 1, _THIS_IP_); + lock_map_release(&handle->h_lockdep_map); jbd_free_handle(handle); return err; diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 91389c8..9203c33 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -20,6 +20,7 @@ #include <linux/time.h> #include <linux/fs.h> #include <linux/jbd2.h> +#include <linux/marker.h> #include <linux/errno.h> #include <linux/slab.h> @@ -93,7 +94,8 @@ static int __try_to_free_cp_buf(struct journal_head *jh) int ret = 0; struct buffer_head *bh = jh2bh(jh); - if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) { + if (jh->b_jlist == BJ_None && !buffer_locked(bh) && + !buffer_dirty(bh) && !buffer_write_io_error(bh)) { JBUFFER_TRACE(jh, "remove from checkpoint list"); ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); @@ -126,14 +128,29 @@ void __jbd2_log_wait_for_space(journal_t *journal) /* * Test again, another process may have checkpointed while we - * were waiting for the checkpoint lock + * were waiting for the checkpoint lock. If there are no + * outstanding transactions there is nothing to checkpoint and + * we can't make progress. Abort the journal in this case. */ spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); nblocks = jbd_space_needed(journal); if (__jbd2_log_space_left(journal) < nblocks) { + int chkpt = journal->j_checkpoint_transactions != NULL; + + spin_unlock(&journal->j_list_lock); spin_unlock(&journal->j_state_lock); - jbd2_log_do_checkpoint(journal); + if (chkpt) { + jbd2_log_do_checkpoint(journal); + } else { + printk(KERN_ERR "%s: no transactions\n", + __func__); + jbd2_journal_abort(journal, 0); + } + spin_lock(&journal->j_state_lock); + } else { + spin_unlock(&journal->j_list_lock); } mutex_unlock(&journal->j_checkpoint_mutex); } @@ -160,21 +177,25 @@ static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh) * buffers. Note that we take the buffers in the opposite ordering * from the one in which they were submitted for IO. * + * Return 0 on success, and return <0 if some buffers have failed + * to be written out. + * * Called with j_list_lock held. */ -static void __wait_cp_io(journal_t *journal, transaction_t *transaction) +static int __wait_cp_io(journal_t *journal, transaction_t *transaction) { struct journal_head *jh; struct buffer_head *bh; tid_t this_tid; int released = 0; + int ret = 0; this_tid = transaction->t_tid; restart: /* Did somebody clean up the transaction in the meanwhile? */ if (journal->j_checkpoint_transactions != transaction || transaction->t_tid != this_tid) - return; + return ret; while (!released && transaction->t_checkpoint_io_list) { jh = transaction->t_checkpoint_io_list; bh = jh2bh(jh); @@ -194,6 +215,9 @@ restart: spin_lock(&journal->j_list_lock); goto restart; } + if (unlikely(buffer_write_io_error(bh))) + ret = -EIO; + /* * Now in whatever state the buffer currently is, we know that * it has been written out and so we can drop it from the list @@ -203,6 +227,8 @@ restart: jbd2_journal_remove_journal_head(bh); __brelse(bh); } + + return ret; } #define NR_BATCH 64 @@ -226,7 +252,8 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) * Try to flush one buffer from the checkpoint list to disk. * * Return 1 if something happened which requires us to abort the current - * scan of the checkpoint list. + * scan of the checkpoint list. Return <0 if the buffer has failed to + * be written out. * * Called with j_list_lock held and drops it if 1 is returned * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it @@ -258,6 +285,9 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, jbd2_log_wait_commit(journal, tid); ret = 1; } else if (!buffer_dirty(bh)) { + ret = 1; + if (unlikely(buffer_write_io_error(bh))) + ret = -EIO; J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); __jbd2_journal_remove_checkpoint(jh); @@ -265,7 +295,6 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, jbd_unlock_bh_state(bh); jbd2_journal_remove_journal_head(bh); __brelse(bh); - ret = 1; } else { /* * Important: we are about to write the buffer, and @@ -298,6 +327,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, * to disk. We submit larger chunks of data at once. * * The journal should be locked before calling this function. + * Called with j_checkpoint_mutex held. */ int jbd2_log_do_checkpoint(journal_t *journal) { @@ -313,6 +343,8 @@ int jbd2_log_do_checkpoint(journal_t *journal) * journal straight away. */ result = jbd2_cleanup_journal_tail(journal); + trace_mark(jbd2_checkpoint, "dev %s need_checkpoint %d", + journal->j_devname, result); jbd_debug(1, "cleanup_journal_tail returned %d\n", result); if (result <= 0) return result; @@ -321,6 +353,7 @@ int jbd2_log_do_checkpoint(journal_t *journal) * OK, we need to start writing disk blocks. Take one transaction * and write it. */ + result = 0; spin_lock(&journal->j_list_lock); if (!journal->j_checkpoint_transactions) goto out; @@ -339,7 +372,7 @@ restart: int batch_count = 0; struct buffer_head *bhs[NR_BATCH]; struct journal_head *jh; - int retry = 0; + int retry = 0, err; while (!retry && transaction->t_checkpoint_list) { struct buffer_head *bh; @@ -353,6 +386,8 @@ restart: } retry = __process_buffer(journal, jh, bhs, &batch_count, transaction); + if (retry < 0 && !result) + result = retry; if (!retry && (need_resched() || spin_needbreak(&journal->j_list_lock))) { spin_unlock(&journal->j_list_lock); @@ -377,14 +412,18 @@ restart: * Now we have cleaned up the first transaction's checkpoint * list. Let's clean up the second one */ - __wait_cp_io(journal, transaction); + err = __wait_cp_io(journal, transaction); + if (!result) + result = err; } out: spin_unlock(&journal->j_list_lock); - result = jbd2_cleanup_journal_tail(journal); if (result < 0) - return result; - return 0; + jbd2_journal_abort(journal, result); + else + result = jbd2_cleanup_journal_tail(journal); + + return (result < 0) ? result : 0; } /* @@ -400,8 +439,9 @@ out: * This is the only part of the journaling code which really needs to be * aware of transaction aborts. Checkpointing involves writing to the * main filesystem area rather than to the journal, so it can proceed - * even in abort state, but we must not update the journal superblock if - * we have an abort error outstanding. + * even in abort state, but we must not update the super block if + * checkpointing may have failed. Otherwise, we would lose some metadata + * buffers which should be written-back to the filesystem. */ int jbd2_cleanup_journal_tail(journal_t *journal) @@ -410,6 +450,9 @@ int jbd2_cleanup_journal_tail(journal_t *journal) tid_t first_tid; unsigned long blocknr, freed; + if (is_journal_aborted(journal)) + return 1; + /* OK, work out the oldest transaction remaining in the log, and * the log block it starts at. * diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index f8b3be8..0abe02c 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -16,6 +16,7 @@ #include <linux/time.h> #include <linux/fs.h> #include <linux/jbd2.h> +#include <linux/marker.h> #include <linux/errno.h> #include <linux/slab.h> #include <linux/mm.h> @@ -67,7 +68,7 @@ static void release_buffer_page(struct buffer_head *bh) goto nope; /* OK, it's a truncated page */ - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto nope; page_cache_get(page); @@ -126,8 +127,7 @@ static int journal_submit_commit_record(journal_t *journal, JBUFFER_TRACE(descriptor, "submit commit block"); lock_buffer(bh); - get_bh(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); set_buffer_uptodate(bh); bh->b_end_io = journal_end_buffer_io_sync; @@ -147,12 +147,9 @@ static int journal_submit_commit_record(journal_t *journal, * to remember if we sent a barrier request */ if (ret == -EOPNOTSUPP && barrier_done) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING - "JBD: barrier-based sync failed on %s - " - "disabling barriers\n", - bdevname(journal->j_dev, b)); + "JBD: barrier-based sync failed on %s - " + "disabling barriers\n", journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags &= ~JBD2_BARRIER; spin_unlock(&journal->j_state_lock); @@ -160,7 +157,7 @@ static int journal_submit_commit_record(journal_t *journal, /* And try again, without the barrier */ lock_buffer(bh); set_buffer_uptodate(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); ret = submit_bh(WRITE, bh); } *cbh = bh; @@ -262,8 +259,18 @@ static int journal_finish_inode_data_buffers(journal_t *journal, jinode->i_flags |= JI_COMMIT_RUNNING; spin_unlock(&journal->j_list_lock); err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping); - if (!ret) - ret = err; + if (err) { + /* + * Because AS_EIO is cleared by + * wait_on_page_writeback_range(), set it again so + * that user process can get -EIO from fsync(). + */ + set_bit(AS_EIO, + &jinode->i_vfs_inode->i_mapping->flags); + + if (!ret) + ret = err; + } spin_lock(&journal->j_list_lock); jinode->i_flags &= ~JI_COMMIT_RUNNING; wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); @@ -361,6 +368,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) commit_transaction = journal->j_running_transaction; J_ASSERT(commit_transaction->t_state == T_RUNNING); + trace_mark(jbd2_start_commit, "dev %s transaction %d", + journal->j_devname, commit_transaction->t_tid); jbd_debug(1, "JBD: starting commit of transaction %d\n", commit_transaction->t_tid); @@ -495,9 +504,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) jh = commit_transaction->t_buffers; /* If we're in abort mode, we just un-journal the buffer and - release it for background writing. */ + release it. */ if (is_journal_aborted(journal)) { + clear_buffer_jbddirty(jh2bh(jh)); JBUFFER_TRACE(jh, "journal is aborting: refile"); jbd2_journal_refile_buffer(journal, jh); /* If that was the last one, we need to clean up @@ -670,8 +680,14 @@ start_journal_io: * commit block, which happens below in such setting. */ err = journal_finish_inode_data_buffers(journal, commit_transaction); - if (err) - jbd2_journal_abort(journal, err); + if (err) { + printk(KERN_WARNING + "JBD2: Detected IO errors while flushing file data " + "on %s\n", journal->j_devname); + if (journal->j_flags & JBD2_ABORT_ON_SYNCDATA_ERR) + jbd2_journal_abort(journal, err); + err = 0; + } /* Lo and behold: we have just managed to send a transaction to the log. Before we can commit it, wait for the IO so far to @@ -770,6 +786,9 @@ wait_for_iobuf: /* AKPM: bforget here */ } + if (err) + jbd2_journal_abort(journal, err); + jbd_debug(3, "JBD: commit phase 5\n"); if (!JBD2_HAS_INCOMPAT_FEATURE(journal, @@ -868,6 +887,8 @@ restart_loop: if (buffer_jbddirty(bh)) { JBUFFER_TRACE(jh, "add to new checkpointing trans"); __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); @@ -974,6 +995,9 @@ restart_loop: } spin_unlock(&journal->j_list_lock); + trace_mark(jbd2_end_commit, "dev %s transaction %d head %d", + journal->j_devname, commit_transaction->t_tid, + journal->j_tail_sequence); jbd_debug(1, "JBD: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index b26c6d9..783de11 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -68,7 +68,6 @@ EXPORT_SYMBOL(jbd2_journal_set_features); EXPORT_SYMBOL(jbd2_journal_create); EXPORT_SYMBOL(jbd2_journal_load); EXPORT_SYMBOL(jbd2_journal_destroy); -EXPORT_SYMBOL(jbd2_journal_update_superblock); EXPORT_SYMBOL(jbd2_journal_abort); EXPORT_SYMBOL(jbd2_journal_errno); EXPORT_SYMBOL(jbd2_journal_ack_err); @@ -598,13 +597,9 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, if (ret) *retp = ret; else { - char b[BDEVNAME_SIZE]; - printk(KERN_ALERT "%s: journal block not found " "at offset %lu on %s\n", - __func__, - blocknr, - bdevname(journal->j_dev, b)); + __func__, blocknr, journal->j_devname); err = -EIO; __journal_abort_soft(journal, err); } @@ -902,10 +897,7 @@ static struct proc_dir_entry *proc_jbd2_stats; static void jbd2_stats_proc_init(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); - journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats); + journal->j_proc_entry = proc_mkdir(journal->j_devname, proc_jbd2_stats); if (journal->j_proc_entry) { proc_create_data("history", S_IRUGO, journal->j_proc_entry, &jbd2_seq_history_fops, journal); @@ -916,12 +908,9 @@ static void jbd2_stats_proc_init(journal_t *journal) static void jbd2_stats_proc_exit(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); remove_proc_entry("info", journal->j_proc_entry); remove_proc_entry("history", journal->j_proc_entry); - remove_proc_entry(name, proc_jbd2_stats); + remove_proc_entry(journal->j_devname, proc_jbd2_stats); } static void journal_init_stats(journal_t *journal) @@ -1019,6 +1008,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, { journal_t *journal = journal_init_common(); struct buffer_head *bh; + char *p; int n; if (!journal) @@ -1040,6 +1030,10 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_fs_dev = fs_dev; journal->j_blk_offset = start; journal->j_maxlen = len; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; jbd2_stats_proc_init(journal); bh = __getblk(journal->j_dev, start, journal->j_blocksize); @@ -1062,6 +1056,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) { struct buffer_head *bh; journal_t *journal = journal_init_common(); + char *p; int err; int n; unsigned long long blocknr; @@ -1071,6 +1066,12 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev; journal->j_inode = inode; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; + p = journal->j_devname + strlen(journal->j_devname); + sprintf(p, ":%lu", journal->j_inode->i_ino); jbd_debug(1, "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n", journal, inode->i_sb->s_id, inode->i_ino, @@ -1254,6 +1255,22 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait) goto out; } + if (buffer_write_io_error(bh)) { + /* + * Oh, dear. A previous attempt to write the journal + * superblock failed. This could happen because the + * USB device was yanked out. Or it could happen to + * be a transient write error and maybe the block will + * be remapped. Nothing we can do but to retry the + * write and hope for the best. + */ + printk(KERN_ERR "JBD2: previous I/O error detected " + "for journal superblock update for %s.\n", + journal->j_devname); + clear_buffer_write_io_error(bh); + set_buffer_uptodate(bh); + } + spin_lock(&journal->j_state_lock); jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n", journal->j_tail, journal->j_tail_sequence, journal->j_errno); @@ -1265,9 +1282,16 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait) BUFFER_TRACE(bh, "marking dirty"); mark_buffer_dirty(bh); - if (wait) + if (wait) { sync_dirty_buffer(bh); - else + if (buffer_write_io_error(bh)) { + printk(KERN_ERR "JBD2: I/O error detected " + "when updating journal superblock for %s.\n", + journal->j_devname); + clear_buffer_write_io_error(bh); + set_buffer_uptodate(bh); + } + } else ll_rw_block(SWRITE, 1, &bh); out: @@ -1427,9 +1451,12 @@ recovery_error: * * Release a journal_t structure once it is no longer in use by the * journaled object. + * Return <0 if we couldn't clean up the journal. */ -void jbd2_journal_destroy(journal_t *journal) +int jbd2_journal_destroy(journal_t *journal) { + int err = 0; + /* Wait for the commit thread to wake up and die. */ journal_kill_thread(journal); @@ -1452,11 +1479,16 @@ void jbd2_journal_destroy(journal_t *journal) J_ASSERT(journal->j_checkpoint_transactions == NULL); spin_unlock(&journal->j_list_lock); - /* We can now mark the journal as empty. */ - journal->j_tail = 0; - journal->j_tail_sequence = ++journal->j_transaction_sequence; if (journal->j_sb_buffer) { - jbd2_journal_update_superblock(journal, 1); + if (!is_journal_aborted(journal)) { + /* We can now mark the journal as empty. */ + journal->j_tail = 0; + journal->j_tail_sequence = + ++journal->j_transaction_sequence; + jbd2_journal_update_superblock(journal, 1); + } else { + err = -EIO; + } brelse(journal->j_sb_buffer); } @@ -1468,6 +1500,8 @@ void jbd2_journal_destroy(journal_t *journal) jbd2_journal_destroy_revoke(journal); kfree(journal->j_wbuf); kfree(journal); + + return err; } @@ -1693,10 +1727,16 @@ int jbd2_journal_flush(journal_t *journal) spin_lock(&journal->j_list_lock); while (!err && journal->j_checkpoint_transactions != NULL) { spin_unlock(&journal->j_list_lock); + mutex_lock(&journal->j_checkpoint_mutex); err = jbd2_log_do_checkpoint(journal); + mutex_unlock(&journal->j_checkpoint_mutex); spin_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); + + if (is_journal_aborted(journal)) + return -EIO; + jbd2_cleanup_journal_tail(journal); /* Finally, mark the journal as really needing no recovery. @@ -1718,7 +1758,7 @@ int jbd2_journal_flush(journal_t *journal) J_ASSERT(journal->j_head == journal->j_tail); J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence); spin_unlock(&journal->j_state_lock); - return err; + return 0; } /** @@ -1762,23 +1802,6 @@ int jbd2_journal_wipe(journal_t *journal, int write) } /* - * journal_dev_name: format a character string to describe on what - * device this journal is present. - */ - -static const char *journal_dev_name(journal_t *journal, char *buffer) -{ - struct block_device *bdev; - - if (journal->j_inode) - bdev = journal->j_inode->i_sb->s_bdev; - else - bdev = journal->j_dev; - - return bdevname(bdev, buffer); -} - -/* * Journal abort has very specific semantics, which we describe * for journal abort. * @@ -1794,13 +1817,12 @@ static const char *journal_dev_name(journal_t *journal, char *buffer) void __jbd2_journal_abort_hard(journal_t *journal) { transaction_t *transaction; - char b[BDEVNAME_SIZE]; if (journal->j_flags & JBD2_ABORT) return; printk(KERN_ERR "Aborting journal on device %s.\n", - journal_dev_name(journal, b)); + journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags |= JBD2_ABORT; diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 058f50f..7306328 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -225,7 +225,7 @@ do { \ */ int jbd2_journal_recover(journal_t *journal) { - int err; + int err, err2; journal_superblock_t * sb; struct recovery_info info; @@ -263,7 +263,10 @@ int jbd2_journal_recover(journal_t *journal) journal->j_transaction_sequence = ++info.end_transaction; jbd2_journal_clear_revoke(journal); - sync_blockdev(journal->j_fs_dev); + err2 = sync_blockdev(journal->j_fs_dev); + if (!err) + err = err2; + return err; } diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 4f7cadb..e5d5405 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -301,7 +301,7 @@ handle_t *jbd2_journal_start(journal_t *journal, int nblocks) goto out; } - lock_acquire(&handle->h_lockdep_map, 0, 0, 0, 2, _THIS_IP_); + lock_map_acquire(&handle->h_lockdep_map); out: return handle; } @@ -1279,7 +1279,7 @@ int jbd2_journal_stop(handle_t *handle) spin_unlock(&journal->j_state_lock); } - lock_release(&handle->h_lockdep_map, 1, _THIS_IP_); + lock_map_release(&handle->h_lockdep_map); jbd2_free_handle(handle); return err; diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 4c80404..d987137 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -314,7 +314,7 @@ static int jffs2_check_acl(struct inode *inode, int mask) return -EAGAIN; } -int jffs2_permission(struct inode *inode, int mask, struct nameidata *nd) +int jffs2_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, jffs2_check_acl); } diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 0bb7f003..8ca058a 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -28,7 +28,7 @@ struct jffs2_acl_header { #define JFFS2_ACL_NOT_CACHED ((void *)-1) -extern int jffs2_permission(struct inode *, int, struct nameidata *); +extern int jffs2_permission(struct inode *, int); extern int jffs2_acl_chmod(struct inode *); extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *); extern int jffs2_init_acl_post(struct inode *); diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index c0c141f..cd219ef 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -38,7 +38,7 @@ const struct file_operations jffs2_dir_operations = { .read = generic_read_dir, .readdir = jffs2_readdir, - .ioctl = jffs2_ioctl, + .unlocked_ioctl=jffs2_ioctl, .fsync = jffs2_fsync }; diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 5e92034..5a98aa8 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -46,7 +46,7 @@ const struct file_operations jffs2_file_operations = .aio_read = generic_file_aio_read, .write = do_sync_write, .aio_write = generic_file_aio_write, - .ioctl = jffs2_ioctl, + .unlocked_ioctl=jffs2_ioctl, .mmap = generic_file_readonly_mmap, .fsync = jffs2_fsync, .splice_read = generic_file_splice_read, diff --git a/fs/jffs2/ioctl.c b/fs/jffs2/ioctl.c index e217721..9d41f43 100644 --- a/fs/jffs2/ioctl.c +++ b/fs/jffs2/ioctl.c @@ -12,8 +12,7 @@ #include <linux/fs.h> #include "nodelist.h" -int jffs2_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, - unsigned long arg) +long jffs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { /* Later, this will provide for lsattr.jffs2 and chattr.jffs2, which will include compression support etc. */ diff --git a/fs/jffs2/jffs2_fs_i.h b/fs/jffs2/jffs2_fs_i.h index 31559f4..4c41db9 100644 --- a/fs/jffs2/jffs2_fs_i.h +++ b/fs/jffs2/jffs2_fs_i.h @@ -12,7 +12,6 @@ #ifndef _JFFS2_FS_I #define _JFFS2_FS_I -#include <linux/version.h> #include <linux/rbtree.h> #include <linux/posix_acl.h> #include <linux/mutex.h> diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 2cc866c..5e194a5 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -167,7 +167,7 @@ int jffs2_fsync(struct file *, struct dentry *, int); int jffs2_do_readpage_unlock (struct inode *inode, struct page *pg); /* ioctl.c */ -int jffs2_ioctl(struct inode *, struct file *, unsigned int, unsigned long); +long jffs2_ioctl(struct file *, unsigned int, unsigned long); /* symlink.c */ extern const struct inode_operations jffs2_symlink_inode_operations; diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c index 629af01..6caf1e1e 100644 --- a/fs/jffs2/summary.c +++ b/fs/jffs2/summary.c @@ -23,6 +23,8 @@ int jffs2_sum_init(struct jffs2_sb_info *c) { + uint32_t sum_size = max_t(uint32_t, c->sector_size, MAX_SUMMARY_SIZE); + c->summary = kzalloc(sizeof(struct jffs2_summary), GFP_KERNEL); if (!c->summary) { @@ -30,7 +32,7 @@ int jffs2_sum_init(struct jffs2_sb_info *c) return -ENOMEM; } - c->summary->sum_buf = vmalloc(c->sector_size); + c->summary->sum_buf = kmalloc(sum_size, GFP_KERNEL); if (!c->summary->sum_buf) { JFFS2_WARNING("Can't allocate buffer for writing out summary information!\n"); @@ -49,7 +51,7 @@ void jffs2_sum_exit(struct jffs2_sb_info *c) jffs2_sum_disable_collecting(c->summary); - vfree(c->summary->sum_buf); + kfree(c->summary->sum_buf); c->summary->sum_buf = NULL; kfree(c->summary); @@ -665,7 +667,7 @@ crc_err: /* Write summary data to flash - helper function for jffs2_sum_write_sumnode() */ static int jffs2_sum_write_data(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - uint32_t infosize, uint32_t datasize, int padsize) + uint32_t infosize, uint32_t datasize, int padsize) { struct jffs2_raw_summary isum; union jffs2_sum_mem *temp; @@ -676,6 +678,26 @@ static int jffs2_sum_write_data(struct jffs2_sb_info *c, struct jffs2_eraseblock int ret; size_t retlen; + if (padsize + datasize > MAX_SUMMARY_SIZE) { + /* It won't fit in the buffer. Abort summary for this jeb */ + jffs2_sum_disable_collecting(c->summary); + + JFFS2_WARNING("Summary too big (%d data, %d pad) in eraseblock at %08x\n", + datasize, padsize, jeb->offset); + /* Non-fatal */ + return 0; + } + /* Is there enough space for summary? */ + if (padsize < 0) { + /* don't try to write out summary for this jeb */ + jffs2_sum_disable_collecting(c->summary); + + JFFS2_WARNING("Not enough space for summary, padsize = %d\n", + padsize); + /* Non-fatal */ + return 0; + } + memset(c->summary->sum_buf, 0xff, datasize); memset(&isum, 0, sizeof(isum)); @@ -821,7 +843,7 @@ int jffs2_sum_write_sumnode(struct jffs2_sb_info *c) { int datasize, infosize, padsize; struct jffs2_eraseblock *jeb; - int ret; + int ret = 0; dbg_summary("called\n"); @@ -841,16 +863,6 @@ int jffs2_sum_write_sumnode(struct jffs2_sb_info *c) infosize += padsize; datasize += padsize; - /* Is there enough space for summary? */ - if (padsize < 0) { - /* don't try to write out summary for this jeb */ - jffs2_sum_disable_collecting(c->summary); - - JFFS2_WARNING("Not enough space for summary, padsize = %d\n", padsize); - spin_lock(&c->erase_completion_lock); - return 0; - } - ret = jffs2_sum_write_data(c, jeb, infosize, datasize, padsize); spin_lock(&c->erase_completion_lock); return ret; diff --git a/fs/jffs2/summary.h b/fs/jffs2/summary.h index 8bf34f2..60207a2 100644 --- a/fs/jffs2/summary.h +++ b/fs/jffs2/summary.h @@ -13,6 +13,12 @@ #ifndef JFFS2_SUMMARY_H #define JFFS2_SUMMARY_H +/* Limit summary size to 64KiB so that we can kmalloc it. If the summary + is larger than that, we have to just ditch it and avoid using summary + for the eraseblock in question... and it probably doesn't hurt us much + anyway. */ +#define MAX_SUMMARY_SIZE 65536 + #include <linux/uio.h> #include <linux/jffs2.h> diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 7da69ea..efd4012 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode) kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } -static void jffs2_i_init_once(struct kmem_cache *cachep, void *foo) +static void jffs2_i_init_once(void *foo) { struct jffs2_inode_info *f = foo; diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c index 4d84bdc..d3e5c33 100644 --- a/fs/jfs/acl.c +++ b/fs/jfs/acl.c @@ -140,7 +140,7 @@ static int jfs_check_acl(struct inode *inode, int mask) return -EAGAIN; } -int jfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int jfs_permission(struct inode *inode, int mask) { return generic_permission(inode, mask, jfs_check_acl); } diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h index 455fa42..88475f1 100644 --- a/fs/jfs/jfs_acl.h +++ b/fs/jfs/jfs_acl.h @@ -20,7 +20,7 @@ #ifdef CONFIG_JFS_POSIX_ACL -int jfs_permission(struct inode *, int, struct nameidata *); +int jfs_permission(struct inode *, int); int jfs_init_acl(tid_t, struct inode *, struct inode *); int jfs_setattr(struct dentry *, struct iattr *); diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 854ff0e..c350057 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -182,7 +182,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) #endif -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct metapage *mp = (struct metapage *)foo; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 359c091..3630718 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -760,7 +760,7 @@ static struct file_system_type jfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo; @@ -216,8 +216,8 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, s->s_flags = MS_NOUSER; s->s_maxbytes = ~0ULL; - s->s_blocksize = 1024; - s->s_blocksize_bits = 10; + s->s_blocksize = PAGE_SIZE; + s->s_blocksize_bits = PAGE_SHIFT; s->s_magic = magic; s->s_op = ops ? ops : &simple_super_operations; s->s_time_gran = 1; diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 39944463..4a714f64 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -83,7 +83,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; - int rc = rpc_success; + __be32 rc = rpc_success; dprintk("lockd: TEST4 called\n"); resp->cookie = argp->cookie; @@ -116,7 +116,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; - int rc = rpc_success; + __be32 rc = rpc_success; dprintk("lockd: LOCK called\n"); diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 76019d2..76262c1 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -112,7 +112,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; - int rc = rpc_success; + __be32 rc = rpc_success; dprintk("lockd: TEST called\n"); resp->cookie = argp->cookie; @@ -146,7 +146,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, { struct nlm_host *host; struct nlm_file *file; - int rc = rpc_success; + __be32 rc = rpc_success; dprintk("lockd: LOCK called\n"); @@ -201,7 +201,7 @@ EXPORT_SYMBOL(locks_init_lock); * Initialises the fields of the file lock which are invariant for * free file_locks. */ -static void init_once(struct kmem_cache *cache, void *foo) +static void init_once(void *foo) { struct file_lock *lock = (struct file_lock *) foo; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 523d737..d1d1eb8 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -68,7 +68,7 @@ static void minix_destroy_inode(struct inode *inode) kmem_cache_free(minix_inode_cachep, minix_i(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; @@ -31,7 +31,6 @@ #include <linux/file.h> #include <linux/fcntl.h> #include <linux/device_cgroup.h> -#include <asm/namei.h> #include <asm/uaccess.h> #define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE]) @@ -185,6 +184,8 @@ int generic_permission(struct inode *inode, int mask, { umode_t mode = inode->i_mode; + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; + if (current->fsuid == inode->i_uid) mode >>= 6; else { @@ -203,7 +204,7 @@ int generic_permission(struct inode *inode, int mask, /* * If the DACs are ok we don't need any capability check. */ - if (((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask)) + if ((mask & ~mode) == 0) return 0; check_capabilities: @@ -226,13 +227,9 @@ int generic_permission(struct inode *inode, int mask, return -EACCES; } -int permission(struct inode *inode, int mask, struct nameidata *nd) +int inode_permission(struct inode *inode, int mask) { - int retval, submask; - struct vfsmount *mnt = NULL; - - if (nd) - mnt = nd->path.mnt; + int retval; if (mask & MAY_WRITE) { umode_t mode = inode->i_mode; @@ -251,19 +248,9 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return -EACCES; } - if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) { - /* - * MAY_EXEC on regular files is denied if the fs is mounted - * with the "noexec" flag. - */ - if (mnt && (mnt->mnt_flags & MNT_NOEXEC)) - return -EACCES; - } - /* Ordinary permission routines do not understand MAY_APPEND. */ - submask = mask & ~MAY_APPEND; if (inode->i_op && inode->i_op->permission) { - retval = inode->i_op->permission(inode, submask, nd); + retval = inode->i_op->permission(inode, mask); if (!retval) { /* * Exec permission on a regular file is denied if none @@ -277,7 +264,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return -EACCES; } } else { - retval = generic_permission(inode, submask, NULL); + retval = generic_permission(inode, mask, NULL); } if (retval) return retval; @@ -286,7 +273,8 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) if (retval) return retval; - return security_inode_permission(inode, mask, nd); + return security_inode_permission(inode, + mask & (MAY_READ|MAY_WRITE|MAY_EXEC|MAY_APPEND)); } /** @@ -301,7 +289,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) */ int vfs_permission(struct nameidata *nd, int mask) { - return permission(nd->path.dentry->d_inode, mask, nd); + return inode_permission(nd->path.dentry->d_inode, mask); } /** @@ -318,7 +306,7 @@ int vfs_permission(struct nameidata *nd, int mask) */ int file_permission(struct file *file, int mask) { - return permission(file->f_path.dentry->d_inode, mask, NULL); + return inode_permission(file->f_path.dentry->d_inode, mask); } /* @@ -459,8 +447,7 @@ static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name, * short-cut DAC fails, then call permission() to do more * complete permission check. */ -static int exec_permission_lite(struct inode *inode, - struct nameidata *nd) +static int exec_permission_lite(struct inode *inode) { umode_t mode = inode->i_mode; @@ -486,7 +473,7 @@ static int exec_permission_lite(struct inode *inode, return -EACCES; ok: - return security_inode_permission(inode, MAY_EXEC, nd); + return security_inode_permission(inode, MAY_EXEC); } /* @@ -519,7 +506,14 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s */ result = d_lookup(parent, name); if (!result) { - struct dentry * dentry = d_alloc(parent, name); + struct dentry *dentry; + + /* Don't create child dentry for a dead directory. */ + result = ERR_PTR(-ENOENT); + if (IS_DEADDIR(dir)) + goto out_unlock; + + dentry = d_alloc(parent, name); result = ERR_PTR(-ENOMEM); if (dentry) { result = dir->i_op->lookup(dir, dentry, nd); @@ -528,6 +522,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s else result = dentry; } +out_unlock: mutex_unlock(&dir->i_mutex); return result; } @@ -545,27 +540,16 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s return result; } -static int __emul_lookup_dentry(const char *, struct nameidata *); - /* SMP-safe */ -static __always_inline int +static __always_inline void walk_init_root(const char *name, struct nameidata *nd) { struct fs_struct *fs = current->fs; read_lock(&fs->lock); - if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { - nd->path = fs->altroot; - path_get(&fs->altroot); - read_unlock(&fs->lock); - if (__emul_lookup_dentry(name,nd)) - return 0; - read_lock(&fs->lock); - } nd->path = fs->root; path_get(&fs->root); read_unlock(&fs->lock); - return 1; } /* @@ -606,12 +590,9 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l if (*link == '/') { path_put(&nd->path); - if (!walk_init_root(link, nd)) - /* weird __emul_prefix() stuff did it */ - goto out; + walk_init_root(link, nd); } res = link_path_walk(link, nd); -out: if (nd->depth || res || nd->last_type!=LAST_NORM) return res; /* @@ -889,7 +870,7 @@ static int __link_path_walk(const char *name, struct nameidata *nd) unsigned int c; nd->flags |= LOOKUP_CONTINUE; - err = exec_permission_lite(inode, nd); + err = exec_permission_lite(inode); if (err == -EAGAIN) err = vfs_permission(nd, MAY_EXEC); if (err) @@ -1060,67 +1041,6 @@ static int path_walk(const char *name, struct nameidata *nd) return link_path_walk(name, nd); } -/* - * SMP-safe: Returns 1 and nd will have valid dentry and mnt, if - * everything is done. Returns 0 and drops input nd, if lookup failed; - */ -static int __emul_lookup_dentry(const char *name, struct nameidata *nd) -{ - if (path_walk(name, nd)) - return 0; /* something went wrong... */ - - if (!nd->path.dentry->d_inode || - S_ISDIR(nd->path.dentry->d_inode->i_mode)) { - struct path old_path = nd->path; - struct qstr last = nd->last; - int last_type = nd->last_type; - struct fs_struct *fs = current->fs; - - /* - * NAME was not found in alternate root or it's a directory. - * Try to find it in the normal root: - */ - nd->last_type = LAST_ROOT; - read_lock(&fs->lock); - nd->path = fs->root; - path_get(&fs->root); - read_unlock(&fs->lock); - if (path_walk(name, nd) == 0) { - if (nd->path.dentry->d_inode) { - path_put(&old_path); - return 1; - } - path_put(&nd->path); - } - nd->path = old_path; - nd->last = last; - nd->last_type = last_type; - } - return 1; -} - -void set_fs_altroot(void) -{ - char *emul = __emul_prefix(); - struct nameidata nd; - struct path path = {}, old_path; - int err; - struct fs_struct *fs = current->fs; - - if (!emul) - goto set_it; - err = path_lookup(emul, LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_NOALT, &nd); - if (!err) - path = nd.path; -set_it: - write_lock(&fs->lock); - old_path = fs->altroot; - fs->altroot = path; - write_unlock(&fs->lock); - if (old_path.dentry) - path_put(&old_path); -} - /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */ static int do_path_lookup(int dfd, const char *name, unsigned int flags, struct nameidata *nd) @@ -1136,14 +1056,6 @@ static int do_path_lookup(int dfd, const char *name, if (*name=='/') { read_lock(&fs->lock); - if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) { - nd->path = fs->altroot; - path_get(&fs->altroot); - read_unlock(&fs->lock); - if (__emul_lookup_dentry(name,nd)) - goto out; /* found in altroot */ - read_lock(&fs->lock); - } nd->path = fs->root; path_get(&fs->root); read_unlock(&fs->lock); @@ -1177,7 +1089,6 @@ static int do_path_lookup(int dfd, const char *name, } retval = path_walk(name, nd); -out: if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && nd->path.dentry->d_inode)) audit_inode(name, nd->path.dentry); @@ -1282,19 +1193,6 @@ static int path_lookup_create(int dfd, const char *name, nd, open_flags, create_mode); } -int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, - struct nameidata *nd, int open_flags) -{ - char *tmp = getname(name); - int err = PTR_ERR(tmp); - - if (!IS_ERR(tmp)) { - err = __path_lookup_intent_open(AT_FDCWD, tmp, lookup_flags, nd, open_flags, 0); - putname(tmp); - } - return err; -} - static struct dentry *__lookup_hash(struct qstr *name, struct dentry *base, struct nameidata *nd) { @@ -1317,7 +1215,14 @@ static struct dentry *__lookup_hash(struct qstr *name, dentry = cached_lookup(base, name, nd); if (!dentry) { - struct dentry *new = d_alloc(base, name); + struct dentry *new; + + /* Don't create child dentry for a dead directory. */ + dentry = ERR_PTR(-ENOENT); + if (IS_DEADDIR(inode)) + goto out; + + new = d_alloc(base, name); dentry = ERR_PTR(-ENOMEM); if (!new) goto out; @@ -1340,7 +1245,7 @@ static struct dentry *lookup_hash(struct nameidata *nd) { int err; - err = permission(nd->path.dentry->d_inode, MAY_EXEC, nd); + err = inode_permission(nd->path.dentry->d_inode, MAY_EXEC); if (err) return ERR_PTR(err); return __lookup_hash(&nd->last, nd->path.dentry, nd); @@ -1388,7 +1293,7 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) if (err) return ERR_PTR(err); - err = permission(base->d_inode, MAY_EXEC, NULL); + err = inode_permission(base->d_inode, MAY_EXEC); if (err) return ERR_PTR(err); return __lookup_hash(&this, base, NULL); @@ -1416,22 +1321,40 @@ struct dentry *lookup_one_noperm(const char *name, struct dentry *base) return __lookup_hash(&this, base, NULL); } -int __user_walk_fd(int dfd, const char __user *name, unsigned flags, - struct nameidata *nd) +int user_path_at(int dfd, const char __user *name, unsigned flags, + struct path *path) { + struct nameidata nd; char *tmp = getname(name); int err = PTR_ERR(tmp); - if (!IS_ERR(tmp)) { - err = do_path_lookup(dfd, tmp, flags, nd); + + BUG_ON(flags & LOOKUP_PARENT); + + err = do_path_lookup(dfd, tmp, flags, &nd); putname(tmp); + if (!err) + *path = nd.path; } return err; } -int __user_walk(const char __user *name, unsigned flags, struct nameidata *nd) +static int user_path_parent(int dfd, const char __user *path, + struct nameidata *nd, char **name) { - return __user_walk_fd(AT_FDCWD, name, flags, nd); + char *s = getname(path); + int error; + + if (IS_ERR(s)) + return PTR_ERR(s); + + error = do_path_lookup(dfd, s, LOOKUP_PARENT, nd); + if (error) + putname(s); + else + *name = s; + + return error; } /* @@ -1478,7 +1401,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) BUG_ON(victim->d_parent->d_inode != dir); audit_inode_child(victim->d_name.name, victim, dir); - error = permission(dir,MAY_WRITE | MAY_EXEC, NULL); + error = inode_permission(dir, MAY_WRITE | MAY_EXEC); if (error) return error; if (IS_APPEND(dir)) @@ -1508,14 +1431,13 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) * 3. We should have write and exec permissions on dir * 4. We can't do it if dir is immutable (done in permission()) */ -static inline int may_create(struct inode *dir, struct dentry *child, - struct nameidata *nd) +static inline int may_create(struct inode *dir, struct dentry *child) { if (child->d_inode) return -EEXIST; if (IS_DEADDIR(dir)) return -ENOENT; - return permission(dir,MAY_WRITE | MAY_EXEC, nd); + return inode_permission(dir, MAY_WRITE | MAY_EXEC); } /* @@ -1581,7 +1503,7 @@ void unlock_rename(struct dentry *p1, struct dentry *p2) int vfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd) { - int error = may_create(dir, dentry, nd); + int error = may_create(dir, dentry); if (error) return error; @@ -1755,7 +1677,7 @@ struct file *do_filp_open(int dfd, const char *pathname, int will_write; int flag = open_to_namei_flags(open_flag); - acc_mode = ACC_MODE(flag); + acc_mode = MAY_OPEN | ACC_MODE(flag); /* O_TRUNC implies we need access checks for write permissions */ if (flag & O_TRUNC) @@ -2025,7 +1947,7 @@ EXPORT_SYMBOL_GPL(lookup_create); int vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) { - int error = may_create(dir, dentry, NULL); + int error = may_create(dir, dentry); if (error) return error; @@ -2071,20 +1993,18 @@ static int may_mknod(mode_t mode) asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode, unsigned dev) { - int error = 0; - char * tmp; - struct dentry * dentry; + int error; + char *tmp; + struct dentry *dentry; struct nameidata nd; if (S_ISDIR(mode)) return -EPERM; - tmp = getname(filename); - if (IS_ERR(tmp)) - return PTR_ERR(tmp); - error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, filename, &nd, &tmp); if (error) - goto out; + return error; + dentry = lookup_create(&nd, 0); if (IS_ERR(dentry)) { error = PTR_ERR(dentry); @@ -2116,7 +2036,6 @@ out_dput: out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(tmp); return error; @@ -2129,7 +2048,7 @@ asmlinkage long sys_mknod(const char __user *filename, int mode, unsigned dev) int vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) { - int error = may_create(dir, dentry, NULL); + int error = may_create(dir, dentry); if (error) return error; @@ -2156,14 +2075,10 @@ asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode) struct dentry *dentry; struct nameidata nd; - tmp = getname(pathname); - error = PTR_ERR(tmp); - if (IS_ERR(tmp)) + error = user_path_parent(dfd, pathname, &nd, &tmp); + if (error) goto out_err; - error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd); - if (error) - goto out; dentry = lookup_create(&nd, 1); error = PTR_ERR(dentry); if (IS_ERR(dentry)) @@ -2181,7 +2096,6 @@ out_dput: out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(tmp); out_err: return error; @@ -2259,13 +2173,9 @@ static long do_rmdir(int dfd, const char __user *pathname) struct dentry *dentry; struct nameidata nd; - name = getname(pathname); - if(IS_ERR(name)) - return PTR_ERR(name); - - error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, pathname, &nd, &name); if (error) - goto exit; + return error; switch(nd.last_type) { case LAST_DOTDOT: @@ -2294,7 +2204,6 @@ exit2: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); exit1: path_put(&nd.path); -exit: putname(name); return error; } @@ -2343,19 +2252,16 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) */ static long do_unlinkat(int dfd, const char __user *pathname) { - int error = 0; - char * name; + int error; + char *name; struct dentry *dentry; struct nameidata nd; struct inode *inode = NULL; - name = getname(pathname); - if(IS_ERR(name)) - return PTR_ERR(name); - - error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd); + error = user_path_parent(dfd, pathname, &nd, &name); if (error) - goto exit; + return error; + error = -EISDIR; if (nd.last_type != LAST_NORM) goto exit1; @@ -2382,7 +2288,6 @@ static long do_unlinkat(int dfd, const char __user *pathname) iput(inode); /* truncate the inode here */ exit1: path_put(&nd.path); -exit: putname(name); return error; @@ -2408,9 +2313,9 @@ asmlinkage long sys_unlink(const char __user *pathname) return do_unlinkat(AT_FDCWD, pathname); } -int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, int mode) +int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) { - int error = may_create(dir, dentry, NULL); + int error = may_create(dir, dentry); if (error) return error; @@ -2432,23 +2337,20 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, i asmlinkage long sys_symlinkat(const char __user *oldname, int newdfd, const char __user *newname) { - int error = 0; - char * from; - char * to; + int error; + char *from; + char *to; struct dentry *dentry; struct nameidata nd; from = getname(oldname); - if(IS_ERR(from)) + if (IS_ERR(from)) return PTR_ERR(from); - to = getname(newname); - error = PTR_ERR(to); - if (IS_ERR(to)) - goto out_putname; - error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); + error = user_path_parent(newdfd, newname, &nd, &to); if (error) - goto out; + goto out_putname; + dentry = lookup_create(&nd, 0); error = PTR_ERR(dentry); if (IS_ERR(dentry)) @@ -2457,14 +2359,13 @@ asmlinkage long sys_symlinkat(const char __user *oldname, error = mnt_want_write(nd.path.mnt); if (error) goto out_dput; - error = vfs_symlink(nd.path.dentry->d_inode, dentry, from, S_IALLUGO); + error = vfs_symlink(nd.path.dentry->d_inode, dentry, from); mnt_drop_write(nd.path.mnt); out_dput: dput(dentry); out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); path_put(&nd.path); -out: putname(to); out_putname: putname(from); @@ -2484,7 +2385,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de if (!inode) return -ENOENT; - error = may_create(dir, new_dentry, NULL); + error = may_create(dir, new_dentry); if (error) return error; @@ -2498,19 +2399,19 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de return -EPERM; if (!dir->i_op || !dir->i_op->link) return -EPERM; - if (S_ISDIR(old_dentry->d_inode->i_mode)) + if (S_ISDIR(inode->i_mode)) return -EPERM; error = security_inode_link(old_dentry, dir, new_dentry); if (error) return error; - mutex_lock(&old_dentry->d_inode->i_mutex); + mutex_lock(&inode->i_mutex); DQUOT_INIT(dir); error = dir->i_op->link(old_dentry, dir, new_dentry); - mutex_unlock(&old_dentry->d_inode->i_mutex); + mutex_unlock(&inode->i_mutex); if (!error) - fsnotify_link(dir, old_dentry->d_inode, new_dentry); + fsnotify_link(dir, inode, new_dentry); return error; } @@ -2528,27 +2429,25 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, int flags) { struct dentry *new_dentry; - struct nameidata nd, old_nd; + struct nameidata nd; + struct path old_path; int error; - char * to; + char *to; if ((flags & ~AT_SYMLINK_FOLLOW) != 0) return -EINVAL; - to = getname(newname); - if (IS_ERR(to)) - return PTR_ERR(to); - - error = __user_walk_fd(olddfd, oldname, - flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0, - &old_nd); + error = user_path_at(olddfd, oldname, + flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0, + &old_path); if (error) - goto exit; - error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd); + return error; + + error = user_path_parent(newdfd, newname, &nd, &to); if (error) goto out; error = -EXDEV; - if (old_nd.path.mnt != nd.path.mnt) + if (old_path.mnt != nd.path.mnt) goto out_release; new_dentry = lookup_create(&nd, 0); error = PTR_ERR(new_dentry); @@ -2557,7 +2456,7 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname, error = mnt_want_write(nd.path.mnt); if (error) goto out_dput; - error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode, new_dentry); + error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry); mnt_drop_write(nd.path.mnt); out_dput: dput(new_dentry); @@ -2565,10 +2464,9 @@ out_unlock: mutex_unlock(&nd.path.dentry->d_inode->i_mutex); out_release: path_put(&nd.path); -out: - path_put(&old_nd.path); -exit: putname(to); +out: + path_put(&old_path); return error; } @@ -2621,7 +2519,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, * we'll need to flip '..'. */ if (new_dir != old_dir) { - error = permission(old_dentry->d_inode, MAY_WRITE, NULL); + error = inode_permission(old_dentry->d_inode, MAY_WRITE); if (error) return error; } @@ -2696,7 +2594,7 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, return error; if (!new_dentry->d_inode) - error = may_create(new_dir, new_dentry, NULL); + error = may_create(new_dir, new_dentry); else error = may_delete(new_dir, new_dentry, is_dir); if (error) @@ -2724,20 +2622,22 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, return error; } -static int do_rename(int olddfd, const char *oldname, - int newdfd, const char *newname) +asmlinkage long sys_renameat(int olddfd, const char __user *oldname, + int newdfd, const char __user *newname) { - int error = 0; - struct dentry * old_dir, * new_dir; - struct dentry * old_dentry, *new_dentry; - struct dentry * trap; + struct dentry *old_dir, *new_dir; + struct dentry *old_dentry, *new_dentry; + struct dentry *trap; struct nameidata oldnd, newnd; + char *from; + char *to; + int error; - error = do_path_lookup(olddfd, oldname, LOOKUP_PARENT, &oldnd); + error = user_path_parent(olddfd, oldname, &oldnd, &from); if (error) goto exit; - error = do_path_lookup(newdfd, newname, LOOKUP_PARENT, &newnd); + error = user_path_parent(newdfd, newname, &newnd, &to); if (error) goto exit1; @@ -2799,29 +2699,11 @@ exit3: unlock_rename(new_dir, old_dir); exit2: path_put(&newnd.path); + putname(to); exit1: path_put(&oldnd.path); -exit: - return error; -} - -asmlinkage long sys_renameat(int olddfd, const char __user *oldname, - int newdfd, const char __user *newname) -{ - int error; - char * from; - char * to; - - from = getname(oldname); - if(IS_ERR(from)) - return PTR_ERR(from); - to = getname(newname); - error = PTR_ERR(to); - if (!IS_ERR(to)) { - error = do_rename(olddfd, from, newdfd, to); - putname(to); - } putname(from); +exit: return error; } @@ -2959,8 +2841,7 @@ const struct inode_operations page_symlink_inode_operations = { .put_link = page_put_link, }; -EXPORT_SYMBOL(__user_walk); -EXPORT_SYMBOL(__user_walk_fd); +EXPORT_SYMBOL(user_path_at); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); EXPORT_SYMBOL(get_write_access); /* binfmt_aout */ @@ -2975,7 +2856,7 @@ EXPORT_SYMBOL(page_symlink); EXPORT_SYMBOL(page_symlink_inode_operations); EXPORT_SYMBOL(path_lookup); EXPORT_SYMBOL(vfs_path_lookup); -EXPORT_SYMBOL(permission); +EXPORT_SYMBOL(inode_permission); EXPORT_SYMBOL(vfs_permission); EXPORT_SYMBOL(file_permission); EXPORT_SYMBOL(unlock_rename); diff --git a/fs/namespace.c b/fs/namespace.c index 4f6f763..6e283c9 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -112,9 +112,13 @@ struct vfsmount *alloc_vfsmnt(const char *name) int err; err = mnt_alloc_id(mnt); - if (err) { - kmem_cache_free(mnt_cache, mnt); - return NULL; + if (err) + goto out_free_cache; + + if (name) { + mnt->mnt_devname = kstrdup(name, GFP_KERNEL); + if (!mnt->mnt_devname) + goto out_free_id; } atomic_set(&mnt->mnt_count, 1); @@ -127,16 +131,14 @@ struct vfsmount *alloc_vfsmnt(const char *name) INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); atomic_set(&mnt->__mnt_writers, 0); - if (name) { - int size = strlen(name) + 1; - char *newname = kmalloc(size, GFP_KERNEL); - if (newname) { - memcpy(newname, name, size); - mnt->mnt_devname = newname; - } - } } return mnt; + +out_free_id: + mnt_free_id(mnt); +out_free_cache: + kmem_cache_free(mnt_cache, mnt); + return NULL; } /* @@ -309,10 +311,9 @@ static void handle_write_count_underflow(struct vfsmount *mnt) */ if ((atomic_read(&mnt->__mnt_writers) < 0) && !(mnt->mnt_flags & MNT_IMBALANCED_WRITE_COUNT)) { - printk(KERN_DEBUG "leak detected on mount(%p) writers " + WARN(1, KERN_DEBUG "leak detected on mount(%p) writers " "count: %d\n", mnt, atomic_read(&mnt->__mnt_writers)); - WARN_ON(1); /* use the flag to keep the dmesg spam down */ mnt->mnt_flags |= MNT_IMBALANCED_WRITE_COUNT; } @@ -1129,27 +1130,27 @@ static int do_umount(struct vfsmount *mnt, int flags) asmlinkage long sys_umount(char __user * name, int flags) { - struct nameidata nd; + struct path path; int retval; - retval = __user_walk(name, LOOKUP_FOLLOW, &nd); + retval = user_path(name, &path); if (retval) goto out; retval = -EINVAL; - if (nd.path.dentry != nd.path.mnt->mnt_root) + if (path.dentry != path.mnt->mnt_root) goto dput_and_out; - if (!check_mnt(nd.path.mnt)) + if (!check_mnt(path.mnt)) goto dput_and_out; retval = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto dput_and_out; - retval = do_umount(nd.path.mnt, flags); + retval = do_umount(path.mnt, flags); dput_and_out: /* we mustn't call path_put() as that would clear mnt_expiry_mark */ - dput(nd.path.dentry); - mntput_no_expire(nd.path.mnt); + dput(path.dentry); + mntput_no_expire(path.mnt); out: return retval; } @@ -1666,31 +1667,31 @@ static noinline int do_new_mount(struct nameidata *nd, char *type, int flags, if (IS_ERR(mnt)) return PTR_ERR(mnt); - return do_add_mount(mnt, nd, mnt_flags, NULL); + return do_add_mount(mnt, &nd->path, mnt_flags, NULL); } /* * add a mount into a namespace's mount tree * - provide the option of adding the new mount to an expiration list */ -int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, +int do_add_mount(struct vfsmount *newmnt, struct path *path, int mnt_flags, struct list_head *fslist) { int err; down_write(&namespace_sem); /* Something was mounted here while we slept */ - while (d_mountpoint(nd->path.dentry) && - follow_down(&nd->path.mnt, &nd->path.dentry)) + while (d_mountpoint(path->dentry) && + follow_down(&path->mnt, &path->dentry)) ; err = -EINVAL; - if (!check_mnt(nd->path.mnt)) + if (!check_mnt(path->mnt)) goto unlock; /* Refuse the same filesystem on the same mount point */ err = -EBUSY; - if (nd->path.mnt->mnt_sb == newmnt->mnt_sb && - nd->path.mnt->mnt_root == nd->path.dentry) + if (path->mnt->mnt_sb == newmnt->mnt_sb && + path->mnt->mnt_root == path->dentry) goto unlock; err = -EINVAL; @@ -1698,7 +1699,7 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, goto unlock; newmnt->mnt_flags = mnt_flags; - if ((err = graft_tree(newmnt, &nd->path))) + if ((err = graft_tree(newmnt, path))) goto unlock; if (fslist) /* add to the specified expiration list */ @@ -1973,7 +1974,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, struct fs_struct *fs) { struct mnt_namespace *new_ns; - struct vfsmount *rootmnt = NULL, *pwdmnt = NULL, *altrootmnt = NULL; + struct vfsmount *rootmnt = NULL, *pwdmnt = NULL; struct vfsmount *p, *q; new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL); @@ -2016,10 +2017,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, pwdmnt = p; fs->pwd.mnt = mntget(q); } - if (p == fs->altroot.mnt) { - altrootmnt = p; - fs->altroot.mnt = mntget(q); - } } p = next_mnt(p, mnt_ns->root); q = next_mnt(q, new_ns->root); @@ -2030,8 +2027,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, mntput(rootmnt); if (pwdmnt) mntput(pwdmnt); - if (altrootmnt) - mntput(altrootmnt); return new_ns; } @@ -2184,28 +2179,26 @@ asmlinkage long sys_pivot_root(const char __user * new_root, const char __user * put_old) { struct vfsmount *tmp; - struct nameidata new_nd, old_nd; - struct path parent_path, root_parent, root; + struct path new, old, parent_path, root_parent, root; int error; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - error = __user_walk(new_root, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, - &new_nd); + error = user_path_dir(new_root, &new); if (error) goto out0; error = -EINVAL; - if (!check_mnt(new_nd.path.mnt)) + if (!check_mnt(new.mnt)) goto out1; - error = __user_walk(put_old, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &old_nd); + error = user_path_dir(put_old, &old); if (error) goto out1; - error = security_sb_pivotroot(&old_nd.path, &new_nd.path); + error = security_sb_pivotroot(&old, &new); if (error) { - path_put(&old_nd.path); + path_put(&old); goto out1; } @@ -2214,69 +2207,69 @@ asmlinkage long sys_pivot_root(const char __user * new_root, path_get(¤t->fs->root); read_unlock(¤t->fs->lock); down_write(&namespace_sem); - mutex_lock(&old_nd.path.dentry->d_inode->i_mutex); + mutex_lock(&old.dentry->d_inode->i_mutex); error = -EINVAL; - if (IS_MNT_SHARED(old_nd.path.mnt) || - IS_MNT_SHARED(new_nd.path.mnt->mnt_parent) || + if (IS_MNT_SHARED(old.mnt) || + IS_MNT_SHARED(new.mnt->mnt_parent) || IS_MNT_SHARED(root.mnt->mnt_parent)) goto out2; if (!check_mnt(root.mnt)) goto out2; error = -ENOENT; - if (IS_DEADDIR(new_nd.path.dentry->d_inode)) + if (IS_DEADDIR(new.dentry->d_inode)) goto out2; - if (d_unhashed(new_nd.path.dentry) && !IS_ROOT(new_nd.path.dentry)) + if (d_unhashed(new.dentry) && !IS_ROOT(new.dentry)) goto out2; - if (d_unhashed(old_nd.path.dentry) && !IS_ROOT(old_nd.path.dentry)) + if (d_unhashed(old.dentry) && !IS_ROOT(old.dentry)) goto out2; error = -EBUSY; - if (new_nd.path.mnt == root.mnt || - old_nd.path.mnt == root.mnt) + if (new.mnt == root.mnt || + old.mnt == root.mnt) goto out2; /* loop, on the same file system */ error = -EINVAL; if (root.mnt->mnt_root != root.dentry) goto out2; /* not a mountpoint */ if (root.mnt->mnt_parent == root.mnt) goto out2; /* not attached */ - if (new_nd.path.mnt->mnt_root != new_nd.path.dentry) + if (new.mnt->mnt_root != new.dentry) goto out2; /* not a mountpoint */ - if (new_nd.path.mnt->mnt_parent == new_nd.path.mnt) + if (new.mnt->mnt_parent == new.mnt) goto out2; /* not attached */ /* make sure we can reach put_old from new_root */ - tmp = old_nd.path.mnt; + tmp = old.mnt; spin_lock(&vfsmount_lock); - if (tmp != new_nd.path.mnt) { + if (tmp != new.mnt) { for (;;) { if (tmp->mnt_parent == tmp) goto out3; /* already mounted on put_old */ - if (tmp->mnt_parent == new_nd.path.mnt) + if (tmp->mnt_parent == new.mnt) break; tmp = tmp->mnt_parent; } - if (!is_subdir(tmp->mnt_mountpoint, new_nd.path.dentry)) + if (!is_subdir(tmp->mnt_mountpoint, new.dentry)) goto out3; - } else if (!is_subdir(old_nd.path.dentry, new_nd.path.dentry)) + } else if (!is_subdir(old.dentry, new.dentry)) goto out3; - detach_mnt(new_nd.path.mnt, &parent_path); + detach_mnt(new.mnt, &parent_path); detach_mnt(root.mnt, &root_parent); /* mount old root on put_old */ - attach_mnt(root.mnt, &old_nd.path); + attach_mnt(root.mnt, &old); /* mount new_root on / */ - attach_mnt(new_nd.path.mnt, &root_parent); + attach_mnt(new.mnt, &root_parent); touch_mnt_namespace(current->nsproxy->mnt_ns); spin_unlock(&vfsmount_lock); - chroot_fs_refs(&root, &new_nd.path); - security_sb_post_pivotroot(&root, &new_nd.path); + chroot_fs_refs(&root, &new); + security_sb_post_pivotroot(&root, &new); error = 0; path_put(&root_parent); path_put(&parent_path); out2: - mutex_unlock(&old_nd.path.dentry->d_inode->i_mutex); + mutex_unlock(&old.dentry->d_inode->i_mutex); up_write(&namespace_sem); path_put(&root); - path_put(&old_nd.path); + path_put(&old); out1: - path_put(&new_nd.path); + path_put(&new); out0: return error; out3: diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 011ef0b..07e9715 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -266,7 +266,7 @@ leave_me:; static int -__ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd) +__ncp_lookup_validate(struct dentry *dentry) { struct ncp_server *server; struct dentry *parent; @@ -340,7 +340,7 @@ ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd) { int res; lock_kernel(); - res = __ncp_lookup_validate(dentry, nd); + res = __ncp_lookup_validate(dentry); unlock_kernel(); return res; } diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 2e5ab12..d642f0e 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -64,7 +64,7 @@ static void ncp_destroy_inode(struct inode *inode) kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 28a238d..74f92b7 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1884,7 +1884,7 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) return status; nfs_access_add_cache(inode, &cache); out: - if ((cache.mask & mask) == mask) + if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) return 0; return -EACCES; } @@ -1907,17 +1907,17 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags) return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags)); } -int nfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int nfs_permission(struct inode *inode, int mask) { struct rpc_cred *cred; int res = 0; nfs_inc_stats(inode, NFSIOS_VFSACCESS); - if (mask == 0) + if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) goto out; /* Is this sys_access() ? */ - if (nd != NULL && (nd->flags & LOOKUP_ACCESS)) + if (mask & MAY_ACCESS) goto force_lookup; switch (inode->i_mode & S_IFMT) { @@ -1926,8 +1926,7 @@ int nfs_permission(struct inode *inode, int mask, struct nameidata *nd) case S_IFREG: /* NFSv4 has atomic_open... */ if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN) - && nd != NULL - && (nd->flags & LOOKUP_OPEN)) + && (mask & MAY_OPEN)) goto out; break; case S_IFDIR: diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index df23f98..52daefa 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1242,7 +1242,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi) #endif } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 2f285ef..66df08d 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -129,7 +129,7 @@ static void * nfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd) goto out_err; mntget(mnt); - err = do_add_mount(mnt, nd, nd->path.mnt->mnt_flags|MNT_SHRINKABLE, + err = do_add_mount(mnt, &nd->path, nd->path.mnt->mnt_flags|MNT_SHRINKABLE, &nfs_automount_list); if (err < 0) { mntput(mnt); diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c index 8478fc2..46763d1c 100644 --- a/fs/nfs/nfsroot.c +++ b/fs/nfs/nfsroot.c @@ -127,7 +127,7 @@ enum { Opt_err }; -static match_table_t __initconst tokens = { +static match_table_t __initdata tokens = { {Opt_port, "port=%u"}, {Opt_rsize, "rsize=%u"}, {Opt_wsize, "wsize=%u"}, diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 1b94e36..e9b2017 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1279,6 +1279,12 @@ static int nfs_parse_mount_options(char *raw, } } + if (errors > 0) { + dfprintk(MOUNT, "NFS: parsing encountered %d error%s\n", + errors, (errors == 1 ? "" : "s")); + if (!sloppy) + return 0; + } return 1; out_nomem: @@ -1718,9 +1724,9 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) * ones were explicitly specified. Fall back to legacy behavior and * just return success. */ - if ((nfsvers == 4 && options4->version == 1) || - (nfsvers <= 3 && options->version >= 1 && - options->version <= 6)) + if ((nfsvers == 4 && (!options4 || options4->version == 1)) || + (nfsvers <= 3 && (!options || (options->version >= 1 && + options->version <= 6)))) return 0; data = kzalloc(sizeof(*data), GFP_KERNEL); diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 3adf8b2..f089e58 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -95,10 +95,11 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) static void nfs_async_unlink_release(void *calldata) { struct nfs_unlinkdata *data = calldata; + struct super_block *sb = data->dir->i_sb; nfs_dec_sillycount(data->dir); - nfs_sb_deactive(NFS_SERVER(data->dir)); nfs_free_unlinkdata(data); + nfs_sb_deactive(NFS_SB(sb)); } static const struct rpc_call_ops nfs_unlink_ops = { diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 33bfcf0..9dc036f 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1023,7 +1023,7 @@ exp_export(struct nfsctl_export *nxp) /* Look up the dentry */ err = path_lookup(nxp->ex_path, 0, &nd); if (err) - goto out_unlock; + goto out_put_clp; err = -EINVAL; exp = exp_get_by_name(clp, nd.path.mnt, nd.path.dentry, NULL); @@ -1090,9 +1090,9 @@ finish: exp_put(exp); if (fsid_key && !IS_ERR(fsid_key)) cache_put(&fsid_key->h, &svc_expkey_cache); - if (clp) - auth_domain_put(clp); path_put(&nd.path); +out_put_clp: + auth_domain_put(clp); out_unlock: exp_writeunlock(); out: diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index b6ed383..54b8b41 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -443,7 +443,7 @@ init_state(struct posix_acl_state *state, int cnt) * enough space for either: */ alloc = sizeof(struct posix_ace_state_array) - + cnt*sizeof(struct posix_ace_state); + + cnt*sizeof(struct posix_user_ace_state); state->users = kzalloc(alloc, GFP_KERNEL); if (!state->users) return -ENOMEM; diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index eef1629..e5b51ff 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -851,7 +851,7 @@ struct nfsd4_operation { static struct nfsd4_operation nfsd4_ops[]; -static inline char *nfsd4_op_name(unsigned opnum); +static const char *nfsd4_op_name(unsigned opnum); /* * COMPOUND call. @@ -867,11 +867,6 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, int slack_bytes; __be32 status; - status = nfserr_resource; - cstate = cstate_alloc(); - if (cstate == NULL) - goto out; - resp->xbuf = &rqstp->rq_res; resp->p = rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len; resp->tagp = resp->p; @@ -890,6 +885,11 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, if (args->minorversion > NFSD_SUPPORTED_MINOR_VERSION) goto out; + status = nfserr_resource; + cstate = cstate_alloc(); + if (cstate == NULL) + goto out; + status = nfs_ok; while (!status && resp->opcnt < args->opcnt) { op = &args->ops[resp->opcnt++]; @@ -957,9 +957,9 @@ encode_op: nfsd4_increment_op_stats(op->opnum); } + cstate_free(cstate); out: nfsd4_release_compoundargs(args); - cstate_free(cstate); dprintk("nfsv4 compound returned %d\n", ntohl(status)); return status; } @@ -1116,8 +1116,7 @@ static struct nfsd4_operation nfsd4_ops[OP_RELEASE_LOCKOWNER+1] = { }, }; -static inline char * -nfsd4_op_name(unsigned opnum) +static const char *nfsd4_op_name(unsigned opnum) { if (opnum < ARRAY_SIZE(nfsd4_ops)) return nfsd4_ops[opnum].op_name; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 1955a27..c53e65f 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -12,6 +12,7 @@ #include <linux/time.h> #include <linux/errno.h> #include <linux/fs.h> +#include <linux/namei.h> #include <linux/fcntl.h> #include <linux/net.h> #include <linux/in.h> diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index f45451e..ea37c96 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -51,7 +51,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) /* make sure parents give x permission to user */ int err; parent = dget_parent(tdentry); - err = permission(parent->d_inode, MAY_EXEC, NULL); + err = inode_permission(parent->d_inode, MAY_EXEC); if (err < 0) { dput(parent); break; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 0f4481e..18060be 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1516,7 +1516,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct dentry *dentry, *dnew; __be32 err, cerr; int host_err; - umode_t mode; err = nfserr_noent; if (!flen || !plen) @@ -1535,11 +1534,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dnew)) goto out_nfserr; - mode = S_IALLUGO; - /* Only the MODE ATTRibute is even vaguely meaningful */ - if (iap && (iap->ia_valid & ATTR_MODE)) - mode = iap->ia_mode & S_IALLUGO; - host_err = mnt_want_write(fhp->fh_export->ex_path.mnt); if (host_err) goto out_nfserr; @@ -1551,11 +1545,11 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced); kfree(path_alloced); } } else - host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path); if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) @@ -1959,12 +1953,12 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, return 0; /* This assumes NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */ - err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL); + err = inode_permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC)); /* Allow read access to binaries even when mode 111 */ if (err == -EACCES && S_ISREG(inode->i_mode) && acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE)) - err = permission(inode, MAY_EXEC, NULL); + err = inode_permission(inode, MAY_EXEC); return err? nfserrno(err) : 0; } diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 00e9ccd..b38f944 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -1194,7 +1194,7 @@ lock_retry_remap: tbh = bhs[i]; if (!tbh) continue; - if (unlikely(test_set_buffer_locked(tbh))) + if (!trylock_buffer(tbh)) BUG(); /* The buffer dirty state is now irrelevant, just clean it. */ clear_buffer_dirty(tbh); diff --git a/fs/ntfs/compress.c b/fs/ntfs/compress.c index 33ff314..9669541 100644 --- a/fs/ntfs/compress.c +++ b/fs/ntfs/compress.c @@ -665,7 +665,7 @@ lock_retry_remap: for (i = 0; i < nr_bhs; i++) { struct buffer_head *tbh = bhs[i]; - if (unlikely(test_set_buffer_locked(tbh))) + if (!trylock_buffer(tbh)) continue; if (unlikely(buffer_uptodate(tbh))) { unlock_buffer(tbh); diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 3c5550c..d020866 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -2118,7 +2118,7 @@ static ssize_t ntfs_file_aio_write_nolock(struct kiocb *iocb, goto out; if (!count) goto out; - err = remove_suid(file->f_path.dentry); + err = file_remove_suid(file); if (err) goto out; file_update_time(file); diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 790defb..17d32ca 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -586,7 +586,7 @@ int ntfs_sync_mft_mirror(ntfs_volume *vol, const unsigned long mft_no, for (i_bhs = 0; i_bhs < nr_bhs; i_bhs++) { struct buffer_head *tbh = bhs[i_bhs]; - if (unlikely(test_set_buffer_locked(tbh))) + if (!trylock_buffer(tbh)) BUG(); BUG_ON(!buffer_uptodate(tbh)); clear_buffer_dirty(tbh); @@ -779,7 +779,7 @@ int write_mft_record_nolock(ntfs_inode *ni, MFT_RECORD *m, int sync) for (i_bhs = 0; i_bhs < nr_bhs; i_bhs++) { struct buffer_head *tbh = bhs[i_bhs]; - if (unlikely(test_set_buffer_locked(tbh))) + if (!trylock_buffer(tbh)) BUG(); BUG_ON(!buffer_uptodate(tbh)); clear_buffer_dirty(tbh); diff --git a/fs/ntfs/namei.c b/fs/ntfs/namei.c index e1781c8..9e8a95b 100644 --- a/fs/ntfs/namei.c +++ b/fs/ntfs/namei.c @@ -174,7 +174,6 @@ static struct dentry *ntfs_lookup(struct inode *dir_ino, struct dentry *dent, // TODO: Consider moving this lot to a separate function! (AIA) handle_name: { - struct dentry *real_dent, *new_dent; MFT_RECORD *m; ntfs_attr_search_ctx *ctx; ntfs_inode *ni = NTFS_I(dent_inode); @@ -255,93 +254,9 @@ handle_name: } nls_name.hash = full_name_hash(nls_name.name, nls_name.len); - /* - * Note: No need for dent->d_lock lock as i_mutex is held on the - * parent inode. - */ - - /* Does a dentry matching the nls_name exist already? */ - real_dent = d_lookup(dent->d_parent, &nls_name); - /* If not, create it now. */ - if (!real_dent) { - real_dent = d_alloc(dent->d_parent, &nls_name); - kfree(nls_name.name); - if (!real_dent) { - err = -ENOMEM; - goto err_out; - } - new_dent = d_splice_alias(dent_inode, real_dent); - if (new_dent) - dput(real_dent); - else - new_dent = real_dent; - ntfs_debug("Done. (Created new dentry.)"); - return new_dent; - } + dent = d_add_ci(dent, dent_inode, &nls_name); kfree(nls_name.name); - /* Matching dentry exists, check if it is negative. */ - if (real_dent->d_inode) { - if (unlikely(real_dent->d_inode != dent_inode)) { - /* This can happen because bad inodes are unhashed. */ - BUG_ON(!is_bad_inode(dent_inode)); - BUG_ON(!is_bad_inode(real_dent->d_inode)); - } - /* - * Already have the inode and the dentry attached, decrement - * the reference count to balance the ntfs_iget() we did - * earlier on. We found the dentry using d_lookup() so it - * cannot be disconnected and thus we do not need to worry - * about any NFS/disconnectedness issues here. - */ - iput(dent_inode); - ntfs_debug("Done. (Already had inode and dentry.)"); - return real_dent; - } - /* - * Negative dentry: instantiate it unless the inode is a directory and - * has a 'disconnected' dentry (i.e. IS_ROOT and DCACHE_DISCONNECTED), - * in which case d_move() that in place of the found dentry. - */ - if (!S_ISDIR(dent_inode->i_mode)) { - /* Not a directory; everything is easy. */ - d_instantiate(real_dent, dent_inode); - ntfs_debug("Done. (Already had negative file dentry.)"); - return real_dent; - } - spin_lock(&dcache_lock); - if (list_empty(&dent_inode->i_dentry)) { - /* - * Directory without a 'disconnected' dentry; we need to do - * d_instantiate() by hand because it takes dcache_lock which - * we already hold. - */ - list_add(&real_dent->d_alias, &dent_inode->i_dentry); - real_dent->d_inode = dent_inode; - spin_unlock(&dcache_lock); - security_d_instantiate(real_dent, dent_inode); - ntfs_debug("Done. (Already had negative directory dentry.)"); - return real_dent; - } - /* - * Directory with a 'disconnected' dentry; get a reference to the - * 'disconnected' dentry. - */ - new_dent = list_entry(dent_inode->i_dentry.next, struct dentry, - d_alias); - dget_locked(new_dent); - spin_unlock(&dcache_lock); - /* Do security vodoo. */ - security_d_instantiate(real_dent, dent_inode); - /* Move new_dent in place of real_dent. */ - d_move(new_dent, real_dent); - /* Balance the ntfs_iget() we did above. */ - iput(dent_inode); - /* Throw away real_dent. */ - dput(real_dent); - /* Use new_dent as the actual dentry. */ - ntfs_debug("Done. (Already had negative, disconnected directory " - "dentry.)"); - return new_dent; + return dent; eio_err_out: ntfs_error(vol->sb, "Illegal file name attribute. Run chkdsk."); diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 3e76f3b..4a46743 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -3080,7 +3080,7 @@ struct kmem_cache *ntfs_inode_cache; struct kmem_cache *ntfs_big_inode_cache; /* Init once constructor for the inode slab cache. */ -static void ntfs_big_inode_init_once(struct kmem_cache *cachep, void *foo) +static void ntfs_big_inode_init_once(void *foo) { ntfs_inode *ni = (ntfs_inode *)foo; diff --git a/fs/ntfs/usnjrnl.h b/fs/ntfs/usnjrnl.h index 3a8af75..4087fbd 100644 --- a/fs/ntfs/usnjrnl.h +++ b/fs/ntfs/usnjrnl.h @@ -113,7 +113,7 @@ typedef struct { * Reason flags (32-bit). Cumulative flags describing the change(s) to the * file since it was last opened. I think the names speak for themselves but * if you disagree check out the descriptions in the Linux NTFS project NTFS - * documentation: http://linux-ntfs.sourceforge.net/ntfs/files/usnjrnl.html + * documentation: http://www.linux-ntfs.org/ */ enum { USN_REASON_DATA_OVERWRITE = const_cpu_to_le32(0x00000001), @@ -145,7 +145,7 @@ typedef le32 USN_REASON_FLAGS; * Source info flags (32-bit). Information about the source of the change(s) * to the file. For detailed descriptions of what these mean, see the Linux * NTFS project NTFS documentation: - * http://linux-ntfs.sourceforge.net/ntfs/files/usnjrnl.html + * http://www.linux-ntfs.org/ */ enum { USN_SOURCE_DATA_MANAGEMENT = const_cpu_to_le32(0x00000001), diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 10bfb46..29ff57e 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -990,15 +990,6 @@ out: } /* - * This is only valid for leaf nodes, which are the only ones that can - * have empty extents anyway. - */ -static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec) -{ - return !rec->e_leaf_clusters; -} - -/* * This function will discard the rightmost extent record. */ static void ocfs2_shift_records_right(struct ocfs2_extent_list *el) diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h index 42ff94b..60cd3d5 100644 --- a/fs/ocfs2/alloc.h +++ b/fs/ocfs2/alloc.h @@ -146,4 +146,13 @@ static inline unsigned int ocfs2_rec_clusters(struct ocfs2_extent_list *el, return le16_to_cpu(rec->e_leaf_clusters); } +/* + * This is only valid for leaf nodes, which are the only ones that can + * have empty extents anyway. + */ +static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec) +{ + return !rec->e_leaf_clusters; +} + #endif /* OCFS2_ALLOC_H */ diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 1db0801..a53da14 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -594,7 +594,7 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock, goto bail; } - if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno) { + if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno && create) { ocfs2_error(inode->i_sb, "Inode %llu has a hole at block %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno, @@ -1073,12 +1073,15 @@ static void ocfs2_write_failure(struct inode *inode, for(i = 0; i < wc->w_num_pages; i++) { tmppage = wc->w_pages[i]; - if (ocfs2_should_order_data(inode)) - walk_page_buffers(wc->w_handle, page_buffers(tmppage), - from, to, NULL, - ocfs2_journal_dirty_data); + if (page_has_buffers(tmppage)) { + if (ocfs2_should_order_data(inode)) + walk_page_buffers(wc->w_handle, + page_buffers(tmppage), + from, to, NULL, + ocfs2_journal_dirty_data); - block_commit_write(tmppage, from, to); + block_commit_write(tmppage, from, to); + } } } @@ -1901,12 +1904,14 @@ int ocfs2_write_end_nolock(struct address_space *mapping, to = PAGE_CACHE_SIZE; } - if (ocfs2_should_order_data(inode)) - walk_page_buffers(wc->w_handle, page_buffers(tmppage), - from, to, NULL, - ocfs2_journal_dirty_data); - - block_commit_write(tmppage, from, to); + if (page_has_buffers(tmppage)) { + if (ocfs2_should_order_data(inode)) + walk_page_buffers(wc->w_handle, + page_buffers(tmppage), + from, to, NULL, + ocfs2_journal_dirty_data); + block_commit_write(tmppage, from, to); + } } out_write_size: diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c index d8bfa0e..52276c0 100644 --- a/fs/ocfs2/cluster/netdebug.c +++ b/fs/ocfs2/cluster/netdebug.c @@ -138,20 +138,20 @@ static int nst_seq_show(struct seq_file *seq, void *v) " message id: %d\n" " message type: %u\n" " message key: 0x%08x\n" - " sock acquiry: %lu.%lu\n" - " send start: %lu.%lu\n" - " wait start: %lu.%lu\n", + " sock acquiry: %lu.%ld\n" + " send start: %lu.%ld\n" + " wait start: %lu.%ld\n", nst, (unsigned long)nst->st_task->pid, (unsigned long)nst->st_task->tgid, nst->st_task->comm, nst->st_node, nst->st_sc, nst->st_id, nst->st_msg_type, nst->st_msg_key, nst->st_sock_time.tv_sec, - (unsigned long)nst->st_sock_time.tv_usec, + (long)nst->st_sock_time.tv_usec, nst->st_send_time.tv_sec, - (unsigned long)nst->st_send_time.tv_usec, + (long)nst->st_send_time.tv_usec, nst->st_status_time.tv_sec, - nst->st_status_time.tv_usec); + (long)nst->st_status_time.tv_usec); } spin_unlock(&o2net_debug_lock); @@ -276,7 +276,7 @@ static void *sc_seq_next(struct seq_file *seq, void *v, loff_t *pos) return sc; /* unused, just needs to be null when done */ } -#define TV_SEC_USEC(TV) TV.tv_sec, (unsigned long)TV.tv_usec +#define TV_SEC_USEC(TV) TV.tv_sec, (long)TV.tv_usec static int sc_seq_show(struct seq_file *seq, void *v) { @@ -309,12 +309,12 @@ static int sc_seq_show(struct seq_file *seq, void *v) " remote node: %s\n" " page off: %zu\n" " handshake ok: %u\n" - " timer: %lu.%lu\n" - " data ready: %lu.%lu\n" - " advance start: %lu.%lu\n" - " advance stop: %lu.%lu\n" - " func start: %lu.%lu\n" - " func stop: %lu.%lu\n" + " timer: %lu.%ld\n" + " data ready: %lu.%ld\n" + " advance start: %lu.%ld\n" + " advance stop: %lu.%ld\n" + " func start: %lu.%ld\n" + " func stop: %lu.%ld\n" " func key: %u\n" " func type: %u\n", sc, diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index a27d615..2bcf706 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -143,8 +143,8 @@ static void o2net_sc_postpone_idle(struct o2net_sock_container *sc); static void o2net_sc_reset_idle_timer(struct o2net_sock_container *sc); #ifdef CONFIG_DEBUG_FS -void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, - u32 msgkey, struct task_struct *task, u8 node) +static void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, + u32 msgkey, struct task_struct *task, u8 node) { INIT_LIST_HEAD(&nst->st_net_debug_item); nst->st_task = task; @@ -153,31 +153,61 @@ void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, nst->st_node = node; } -void o2net_set_nst_sock_time(struct o2net_send_tracking *nst) +static void o2net_set_nst_sock_time(struct o2net_send_tracking *nst) { do_gettimeofday(&nst->st_sock_time); } -void o2net_set_nst_send_time(struct o2net_send_tracking *nst) +static void o2net_set_nst_send_time(struct o2net_send_tracking *nst) { do_gettimeofday(&nst->st_send_time); } -void o2net_set_nst_status_time(struct o2net_send_tracking *nst) +static void o2net_set_nst_status_time(struct o2net_send_tracking *nst) { do_gettimeofday(&nst->st_status_time); } -void o2net_set_nst_sock_container(struct o2net_send_tracking *nst, +static void o2net_set_nst_sock_container(struct o2net_send_tracking *nst, struct o2net_sock_container *sc) { nst->st_sc = sc; } -void o2net_set_nst_msg_id(struct o2net_send_tracking *nst, u32 msg_id) +static void o2net_set_nst_msg_id(struct o2net_send_tracking *nst, u32 msg_id) { nst->st_id = msg_id; } + +#else /* CONFIG_DEBUG_FS */ + +static inline void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, + u32 msgkey, struct task_struct *task, u8 node) +{ +} + +static inline void o2net_set_nst_sock_time(struct o2net_send_tracking *nst) +{ +} + +static inline void o2net_set_nst_send_time(struct o2net_send_tracking *nst) +{ +} + +static inline void o2net_set_nst_status_time(struct o2net_send_tracking *nst) +{ +} + +static inline void o2net_set_nst_sock_container(struct o2net_send_tracking *nst, + struct o2net_sock_container *sc) +{ +} + +static inline void o2net_set_nst_msg_id(struct o2net_send_tracking *nst, + u32 msg_id) +{ +} + #endif /* CONFIG_DEBUG_FS */ static inline int o2net_reconnect_delay(void) diff --git a/fs/ocfs2/cluster/tcp_internal.h b/fs/ocfs2/cluster/tcp_internal.h index 18307ff..8d58cfe 100644 --- a/fs/ocfs2/cluster/tcp_internal.h +++ b/fs/ocfs2/cluster/tcp_internal.h @@ -224,42 +224,10 @@ struct o2net_send_tracking { struct timeval st_send_time; struct timeval st_status_time; }; - -void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, - u32 msgkey, struct task_struct *task, u8 node); -void o2net_set_nst_sock_time(struct o2net_send_tracking *nst); -void o2net_set_nst_send_time(struct o2net_send_tracking *nst); -void o2net_set_nst_status_time(struct o2net_send_tracking *nst); -void o2net_set_nst_sock_container(struct o2net_send_tracking *nst, - struct o2net_sock_container *sc); -void o2net_set_nst_msg_id(struct o2net_send_tracking *nst, u32 msg_id); - #else struct o2net_send_tracking { u32 dummy; }; - -static inline void o2net_init_nst(struct o2net_send_tracking *nst, u32 msgtype, - u32 msgkey, struct task_struct *task, u8 node) -{ -} -static inline void o2net_set_nst_sock_time(struct o2net_send_tracking *nst) -{ -} -static inline void o2net_set_nst_send_time(struct o2net_send_tracking *nst) -{ -} -static inline void o2net_set_nst_status_time(struct o2net_send_tracking *nst) -{ -} -static inline void o2net_set_nst_sock_container(struct o2net_send_tracking *nst, - struct o2net_sock_container *sc) -{ -} -static inline void o2net_set_nst_msg_id(struct o2net_send_tracking *nst, - u32 msg_id) -{ -} #endif /* CONFIG_DEBUG_FS */ #endif /* O2CLUSTER_TCP_INTERNAL_H */ diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 8a18758..9cce563 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -1300,7 +1300,6 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, di->i_size = cpu_to_le64(sb->s_blocksize); di->i_ctime = di->i_mtime = cpu_to_le64(dir->i_ctime.tv_sec); di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(dir->i_ctime.tv_nsec); - dir->i_blocks = ocfs2_inode_sector_count(dir); /* * This should never fail as our extent list is empty and all @@ -1310,9 +1309,15 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, NULL); if (ret) { mlog_errno(ret); - goto out; + goto out_commit; } + /* + * Set i_blocks after the extent insert for the most up to + * date ip_clusters value. + */ + dir->i_blocks = ocfs2_inode_sector_count(dir); + ret = ocfs2_journal_dirty(handle, di_bh); if (ret) { mlog_errno(ret); @@ -1336,7 +1341,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, len, 0, NULL); if (ret) { mlog_errno(ret); - goto out; + goto out_commit; } } diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index e48aba6..533a789 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -267,8 +267,7 @@ static ssize_t dlmfs_file_write(struct file *filp, return writelen; } -static void dlmfs_init_once(struct kmem_cache *cachep, - void *foo) +static void dlmfs_init_once(void *foo) { struct dlmfs_inode_private *ip = (struct dlmfs_inode_private *) foo; diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index c58668a..aed268e 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -25,6 +25,7 @@ #include <linux/fs.h> #include <linux/init.h> #include <linux/types.h> +#include <linux/fiemap.h> #define MLOG_MASK_PREFIX ML_EXTENT_MAP #include <cluster/masklog.h> @@ -32,6 +33,7 @@ #include "ocfs2.h" #include "alloc.h" +#include "dlmglue.h" #include "extent_map.h" #include "inode.h" #include "super.h" @@ -282,6 +284,51 @@ out: kfree(new_emi); } +static int ocfs2_last_eb_is_empty(struct inode *inode, + struct ocfs2_dinode *di) +{ + int ret, next_free; + u64 last_eb_blk = le64_to_cpu(di->i_last_eb_blk); + struct buffer_head *eb_bh = NULL; + struct ocfs2_extent_block *eb; + struct ocfs2_extent_list *el; + + ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), last_eb_blk, + &eb_bh, OCFS2_BH_CACHED, inode); + if (ret) { + mlog_errno(ret); + goto out; + } + + eb = (struct ocfs2_extent_block *) eb_bh->b_data; + el = &eb->h_list; + + if (!OCFS2_IS_VALID_EXTENT_BLOCK(eb)) { + ret = -EROFS; + OCFS2_RO_ON_INVALID_EXTENT_BLOCK(inode->i_sb, eb); + goto out; + } + + if (el->l_tree_depth) { + ocfs2_error(inode->i_sb, + "Inode %lu has non zero tree depth in " + "leaf block %llu\n", inode->i_ino, + (unsigned long long)eb_bh->b_blocknr); + ret = -EROFS; + goto out; + } + + next_free = le16_to_cpu(el->l_next_free_rec); + + if (next_free == 0 || + (next_free == 1 && ocfs2_is_empty_extent(&el->l_recs[0]))) + ret = 1; + +out: + brelse(eb_bh); + return ret; +} + /* * Return the 1st index within el which contains an extent start * larger than v_cluster. @@ -373,42 +420,28 @@ out: return ret; } -int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, - u32 *p_cluster, u32 *num_clusters, - unsigned int *extent_flags) +static int ocfs2_get_clusters_nocache(struct inode *inode, + struct buffer_head *di_bh, + u32 v_cluster, unsigned int *hole_len, + struct ocfs2_extent_rec *ret_rec, + unsigned int *is_last) { - int ret, i; - unsigned int flags = 0; - struct buffer_head *di_bh = NULL; - struct buffer_head *eb_bh = NULL; + int i, ret, tree_height, len; struct ocfs2_dinode *di; - struct ocfs2_extent_block *eb; + struct ocfs2_extent_block *uninitialized_var(eb); struct ocfs2_extent_list *el; struct ocfs2_extent_rec *rec; - u32 coff; - - if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { - ret = -ERANGE; - mlog_errno(ret); - goto out; - } - - ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster, - num_clusters, extent_flags); - if (ret == 0) - goto out; + struct buffer_head *eb_bh = NULL; - ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno, - &di_bh, OCFS2_BH_CACHED, inode); - if (ret) { - mlog_errno(ret); - goto out; - } + memset(ret_rec, 0, sizeof(*ret_rec)); + if (is_last) + *is_last = 0; di = (struct ocfs2_dinode *) di_bh->b_data; el = &di->id2.i_list; + tree_height = le16_to_cpu(el->l_tree_depth); - if (el->l_tree_depth) { + if (tree_height > 0) { ret = ocfs2_find_leaf(inode, el, v_cluster, &eb_bh); if (ret) { mlog_errno(ret); @@ -431,46 +464,143 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, i = ocfs2_search_extent_list(el, v_cluster); if (i == -1) { /* - * A hole was found. Return some canned values that - * callers can key on. If asked for, num_clusters will - * be populated with the size of the hole. + * Holes can be larger than the maximum size of an + * extent, so we return their lengths in a seperate + * field. */ - *p_cluster = 0; - if (num_clusters) { + if (hole_len) { ret = ocfs2_figure_hole_clusters(inode, el, eb_bh, - v_cluster, - num_clusters); + v_cluster, &len); if (ret) { mlog_errno(ret); goto out; } + + *hole_len = len; } - } else { - rec = &el->l_recs[i]; + goto out_hole; + } - BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos)); + rec = &el->l_recs[i]; - if (!rec->e_blkno) { - ocfs2_error(inode->i_sb, "Inode %lu has bad extent " - "record (%u, %u, 0)", inode->i_ino, - le32_to_cpu(rec->e_cpos), - ocfs2_rec_clusters(el, rec)); - ret = -EROFS; - goto out; + BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos)); + + if (!rec->e_blkno) { + ocfs2_error(inode->i_sb, "Inode %lu has bad extent " + "record (%u, %u, 0)", inode->i_ino, + le32_to_cpu(rec->e_cpos), + ocfs2_rec_clusters(el, rec)); + ret = -EROFS; + goto out; + } + + *ret_rec = *rec; + + /* + * Checking for last extent is potentially expensive - we + * might have to look at the next leaf over to see if it's + * empty. + * + * The first two checks are to see whether the caller even + * cares for this information, and if the extent is at least + * the last in it's list. + * + * If those hold true, then the extent is last if any of the + * additional conditions hold true: + * - Extent list is in-inode + * - Extent list is right-most + * - Extent list is 2nd to rightmost, with empty right-most + */ + if (is_last) { + if (i == (le16_to_cpu(el->l_next_free_rec) - 1)) { + if (tree_height == 0) + *is_last = 1; + else if (eb->h_blkno == di->i_last_eb_blk) + *is_last = 1; + else if (eb->h_next_leaf_blk == di->i_last_eb_blk) { + ret = ocfs2_last_eb_is_empty(inode, di); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + if (ret == 1) + *is_last = 1; + } } + } + +out_hole: + ret = 0; +out: + brelse(eb_bh); + return ret; +} + +static void ocfs2_relative_extent_offsets(struct super_block *sb, + u32 v_cluster, + struct ocfs2_extent_rec *rec, + u32 *p_cluster, u32 *num_clusters) + +{ + u32 coff = v_cluster - le32_to_cpu(rec->e_cpos); + + *p_cluster = ocfs2_blocks_to_clusters(sb, le64_to_cpu(rec->e_blkno)); + *p_cluster = *p_cluster + coff; + + if (num_clusters) + *num_clusters = le16_to_cpu(rec->e_leaf_clusters) - coff; +} + +int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, + u32 *p_cluster, u32 *num_clusters, + unsigned int *extent_flags) +{ + int ret; + unsigned int uninitialized_var(hole_len), flags = 0; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; - coff = v_cluster - le32_to_cpu(rec->e_cpos); + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + ret = -ERANGE; + mlog_errno(ret); + goto out; + } - *p_cluster = ocfs2_blocks_to_clusters(inode->i_sb, - le64_to_cpu(rec->e_blkno)); - *p_cluster = *p_cluster + coff; + ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster, + num_clusters, extent_flags); + if (ret == 0) + goto out; - if (num_clusters) - *num_clusters = ocfs2_rec_clusters(el, rec) - coff; + ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno, + &di_bh, OCFS2_BH_CACHED, inode); + if (ret) { + mlog_errno(ret); + goto out; + } - flags = rec->e_flags; + ret = ocfs2_get_clusters_nocache(inode, di_bh, v_cluster, &hole_len, + &rec, NULL); + if (ret) { + mlog_errno(ret); + goto out; + } - ocfs2_extent_map_insert_rec(inode, rec); + if (rec.e_blkno == 0ULL) { + /* + * A hole was found. Return some canned values that + * callers can key on. If asked for, num_clusters will + * be populated with the size of the hole. + */ + *p_cluster = 0; + if (num_clusters) { + *num_clusters = hole_len; + } + } else { + ocfs2_relative_extent_offsets(inode->i_sb, v_cluster, &rec, + p_cluster, num_clusters); + flags = rec.e_flags; + + ocfs2_extent_map_insert_rec(inode, &rec); } if (extent_flags) @@ -478,7 +608,6 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, out: brelse(di_bh); - brelse(eb_bh); return ret; } @@ -521,3 +650,114 @@ int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, out: return ret; } + +static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, + struct fiemap_extent_info *fieinfo, + u64 map_start) +{ + int ret; + unsigned int id_count; + struct ocfs2_dinode *di; + u64 phys; + u32 flags = FIEMAP_EXTENT_DATA_INLINE|FIEMAP_EXTENT_LAST; + struct ocfs2_inode_info *oi = OCFS2_I(inode); + + di = (struct ocfs2_dinode *)di_bh->b_data; + id_count = le16_to_cpu(di->id2.i_data.id_count); + + if (map_start < id_count) { + phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits; + phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data); + + ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count, + flags); + if (ret < 0) + return ret; + } + + return 0; +} + +#define OCFS2_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC) + +int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 map_start, u64 map_len) +{ + int ret, is_last; + u32 mapping_end, cpos; + unsigned int hole_size; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + u64 len_bytes, phys_bytes, virt_bytes; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; + + ret = fiemap_check_flags(fieinfo, OCFS2_FIEMAP_FLAGS); + if (ret) + return ret; + + ret = ocfs2_inode_lock(inode, &di_bh, 0); + if (ret) { + mlog_errno(ret); + goto out; + } + + down_read(&OCFS2_I(inode)->ip_alloc_sem); + + /* + * Handle inline-data separately. + */ + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start); + goto out_unlock; + } + + cpos = map_start >> osb->s_clustersize_bits; + mapping_end = ocfs2_clusters_for_bytes(inode->i_sb, + map_start + map_len); + mapping_end -= cpos; + is_last = 0; + while (cpos < mapping_end && !is_last) { + u32 fe_flags; + + ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, + &hole_size, &rec, &is_last); + if (ret) { + mlog_errno(ret); + goto out; + } + + if (rec.e_blkno == 0ULL) { + cpos += hole_size; + continue; + } + + fe_flags = 0; + if (rec.e_flags & OCFS2_EXT_UNWRITTEN) + fe_flags |= FIEMAP_EXTENT_UNWRITTEN; + if (is_last) + fe_flags |= FIEMAP_EXTENT_LAST; + len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits; + phys_bytes = le64_to_cpu(rec.e_blkno) << osb->sb->s_blocksize_bits; + virt_bytes = (u64)le32_to_cpu(rec.e_cpos) << osb->s_clustersize_bits; + + ret = fiemap_fill_next_extent(fieinfo, virt_bytes, phys_bytes, + len_bytes, fe_flags); + if (ret) + break; + + cpos = le32_to_cpu(rec.e_cpos)+ le16_to_cpu(rec.e_leaf_clusters); + } + + if (ret > 0) + ret = 0; + +out_unlock: + brelse(di_bh); + + up_read(&OCFS2_I(inode)->ip_alloc_sem); + + ocfs2_inode_unlock(inode, 0); +out: + + return ret; +} diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h index de91e3e..1b97490 100644 --- a/fs/ocfs2/extent_map.h +++ b/fs/ocfs2/extent_map.h @@ -50,4 +50,7 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, u32 *p_cluster, int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, u64 *ret_count, unsigned int *extent_flags); +int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 map_start, u64 map_len); + #endif /* _EXTENT_MAP_H */ diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index e8514e8..ed38796 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1176,7 +1176,7 @@ bail: return err; } -int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd) +int ocfs2_permission(struct inode *inode, int mask) { int ret; @@ -1766,8 +1766,8 @@ out_inode_unlock: out_rw_unlock: ocfs2_rw_unlock(inode, 1); - mutex_unlock(&inode->i_mutex); out: + mutex_unlock(&inode->i_mutex); return ret; } @@ -2228,6 +2228,7 @@ const struct inode_operations ocfs2_file_iops = { .getattr = ocfs2_getattr, .permission = ocfs2_permission, .fallocate = ocfs2_fallocate, + .fiemap = ocfs2_fiemap, }; const struct inode_operations ocfs2_special_file_iops = { diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h index 048ddca..1e27b4d 100644 --- a/fs/ocfs2/file.h +++ b/fs/ocfs2/file.h @@ -62,8 +62,7 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di, int ocfs2_setattr(struct dentry *dentry, struct iattr *attr); int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -int ocfs2_permission(struct inode *inode, int mask, - struct nameidata *nd); +int ocfs2_permission(struct inode *inode, int mask); int ocfs2_should_update_atime(struct inode *inode, struct vfsmount *vfsmnt); diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index a8c19cb..c47bc2a 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -57,7 +57,7 @@ static int __ocfs2_recovery_thread(void *arg); static int ocfs2_commit_cache(struct ocfs2_super *osb); static int ocfs2_wait_on_mount(struct ocfs2_super *osb); static int ocfs2_journal_toggle_dirty(struct ocfs2_super *osb, - int dirty); + int dirty, int replayed); static int ocfs2_trylock_journal(struct ocfs2_super *osb, int slot_num); static int ocfs2_recover_orphans(struct ocfs2_super *osb, @@ -562,8 +562,18 @@ done: return status; } +static void ocfs2_bump_recovery_generation(struct ocfs2_dinode *di) +{ + le32_add_cpu(&(di->id1.journal1.ij_recovery_generation), 1); +} + +static u32 ocfs2_get_recovery_generation(struct ocfs2_dinode *di) +{ + return le32_to_cpu(di->id1.journal1.ij_recovery_generation); +} + static int ocfs2_journal_toggle_dirty(struct ocfs2_super *osb, - int dirty) + int dirty, int replayed) { int status; unsigned int flags; @@ -593,6 +603,9 @@ static int ocfs2_journal_toggle_dirty(struct ocfs2_super *osb, flags &= ~OCFS2_JOURNAL_DIRTY_FL; fe->id1.journal1.ij_flags = cpu_to_le32(flags); + if (replayed) + ocfs2_bump_recovery_generation(fe); + status = ocfs2_write_block(osb, bh, journal->j_inode); if (status < 0) mlog_errno(status); @@ -667,7 +680,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb) * Do not toggle if flush was unsuccessful otherwise * will leave dirty metadata in a "clean" journal */ - status = ocfs2_journal_toggle_dirty(osb, 0); + status = ocfs2_journal_toggle_dirty(osb, 0, 0); if (status < 0) mlog_errno(status); } @@ -710,7 +723,7 @@ static void ocfs2_clear_journal_error(struct super_block *sb, } } -int ocfs2_journal_load(struct ocfs2_journal *journal, int local) +int ocfs2_journal_load(struct ocfs2_journal *journal, int local, int replayed) { int status = 0; struct ocfs2_super *osb; @@ -729,7 +742,7 @@ int ocfs2_journal_load(struct ocfs2_journal *journal, int local) ocfs2_clear_journal_error(osb->sb, journal->j_journal, osb->slot_num); - status = ocfs2_journal_toggle_dirty(osb, 1); + status = ocfs2_journal_toggle_dirty(osb, 1, replayed); if (status < 0) { mlog_errno(status); goto done; @@ -771,7 +784,7 @@ int ocfs2_journal_wipe(struct ocfs2_journal *journal, int full) goto bail; } - status = ocfs2_journal_toggle_dirty(journal->j_osb, 0); + status = ocfs2_journal_toggle_dirty(journal->j_osb, 0, 0); if (status < 0) mlog_errno(status); @@ -1034,6 +1047,12 @@ restart: spin_unlock(&osb->osb_lock); mlog(0, "All nodes recovered\n"); + /* Refresh all journal recovery generations from disk */ + status = ocfs2_check_journals_nolocks(osb); + status = (status == -EROFS) ? 0 : status; + if (status < 0) + mlog_errno(status); + ocfs2_super_unlock(osb, 1); /* We always run recovery on our own orphan dir - the dead @@ -1096,6 +1115,42 @@ out: mlog_exit_void(); } +static int ocfs2_read_journal_inode(struct ocfs2_super *osb, + int slot_num, + struct buffer_head **bh, + struct inode **ret_inode) +{ + int status = -EACCES; + struct inode *inode = NULL; + + BUG_ON(slot_num >= osb->max_slots); + + inode = ocfs2_get_system_file_inode(osb, JOURNAL_SYSTEM_INODE, + slot_num); + if (!inode || is_bad_inode(inode)) { + mlog_errno(status); + goto bail; + } + SET_INODE_JOURNAL(inode); + + status = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, bh, 0, inode); + if (status < 0) { + mlog_errno(status); + goto bail; + } + + status = 0; + +bail: + if (inode) { + if (status || !ret_inode) + iput(inode); + else + *ret_inode = inode; + } + return status; +} + /* Does the actual journal replay and marks the journal inode as * clean. Will only replay if the journal inode is marked dirty. */ static int ocfs2_replay_journal(struct ocfs2_super *osb, @@ -1109,22 +1164,36 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, struct ocfs2_dinode *fe; journal_t *journal = NULL; struct buffer_head *bh = NULL; + u32 slot_reco_gen; - inode = ocfs2_get_system_file_inode(osb, JOURNAL_SYSTEM_INODE, - slot_num); - if (inode == NULL) { - status = -EACCES; + status = ocfs2_read_journal_inode(osb, slot_num, &bh, &inode); + if (status) { mlog_errno(status); goto done; } - if (is_bad_inode(inode)) { - status = -EACCES; - iput(inode); - inode = NULL; - mlog_errno(status); + + fe = (struct ocfs2_dinode *)bh->b_data; + slot_reco_gen = ocfs2_get_recovery_generation(fe); + brelse(bh); + bh = NULL; + + /* + * As the fs recovery is asynchronous, there is a small chance that + * another node mounted (and recovered) the slot before the recovery + * thread could get the lock. To handle that, we dirty read the journal + * inode for that slot to get the recovery generation. If it is + * different than what we expected, the slot has been recovered. + * If not, it needs recovery. + */ + if (osb->slot_recovery_generations[slot_num] != slot_reco_gen) { + mlog(0, "Slot %u already recovered (old/new=%u/%u)\n", slot_num, + osb->slot_recovery_generations[slot_num], slot_reco_gen); + osb->slot_recovery_generations[slot_num] = slot_reco_gen; + status = -EBUSY; goto done; } - SET_INODE_JOURNAL(inode); + + /* Continue with recovery as the journal has not yet been recovered */ status = ocfs2_inode_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY); if (status < 0) { @@ -1138,9 +1207,12 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, fe = (struct ocfs2_dinode *) bh->b_data; flags = le32_to_cpu(fe->id1.journal1.ij_flags); + slot_reco_gen = ocfs2_get_recovery_generation(fe); if (!(flags & OCFS2_JOURNAL_DIRTY_FL)) { mlog(0, "No recovery required for node %d\n", node_num); + /* Refresh recovery generation for the slot */ + osb->slot_recovery_generations[slot_num] = slot_reco_gen; goto done; } @@ -1188,6 +1260,11 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, flags &= ~OCFS2_JOURNAL_DIRTY_FL; fe->id1.journal1.ij_flags = cpu_to_le32(flags); + /* Increment recovery generation to indicate successful recovery */ + ocfs2_bump_recovery_generation(fe); + osb->slot_recovery_generations[slot_num] = + ocfs2_get_recovery_generation(fe); + status = ocfs2_write_block(osb, bh, inode); if (status < 0) mlog_errno(status); @@ -1252,6 +1329,13 @@ static int ocfs2_recover_node(struct ocfs2_super *osb, status = ocfs2_replay_journal(osb, node_num, slot_num); if (status < 0) { + if (status == -EBUSY) { + mlog(0, "Skipping recovery for slot %u (node %u) " + "as another node has recovered it\n", slot_num, + node_num); + status = 0; + goto done; + } mlog_errno(status); goto done; } @@ -1334,21 +1418,46 @@ int ocfs2_mark_dead_nodes(struct ocfs2_super *osb) { unsigned int node_num; int status, i; + u32 gen; + struct buffer_head *bh = NULL; + struct ocfs2_dinode *di; /* This is called with the super block cluster lock, so we * know that the slot map can't change underneath us. */ - spin_lock(&osb->osb_lock); for (i = 0; i < osb->max_slots; i++) { - if (i == osb->slot_num) + /* Read journal inode to get the recovery generation */ + status = ocfs2_read_journal_inode(osb, i, &bh, NULL); + if (status) { + mlog_errno(status); + goto bail; + } + di = (struct ocfs2_dinode *)bh->b_data; + gen = ocfs2_get_recovery_generation(di); + brelse(bh); + bh = NULL; + + spin_lock(&osb->osb_lock); + osb->slot_recovery_generations[i] = gen; + + mlog(0, "Slot %u recovery generation is %u\n", i, + osb->slot_recovery_generations[i]); + + if (i == osb->slot_num) { + spin_unlock(&osb->osb_lock); continue; + } status = ocfs2_slot_to_node_num_locked(osb, i, &node_num); - if (status == -ENOENT) + if (status == -ENOENT) { + spin_unlock(&osb->osb_lock); continue; + } - if (__ocfs2_recovery_map_test(osb, node_num)) + if (__ocfs2_recovery_map_test(osb, node_num)) { + spin_unlock(&osb->osb_lock); continue; + } spin_unlock(&osb->osb_lock); /* Ok, we have a slot occupied by another node which @@ -1364,10 +1473,7 @@ int ocfs2_mark_dead_nodes(struct ocfs2_super *osb) mlog_errno(status); goto bail; } - - spin_lock(&osb->osb_lock); } - spin_unlock(&osb->osb_lock); status = 0; bail: @@ -1603,49 +1709,41 @@ static int ocfs2_commit_thread(void *arg) return 0; } -/* Look for a dirty journal without taking any cluster locks. Used for - * hard readonly access to determine whether the file system journals - * require recovery. */ +/* Reads all the journal inodes without taking any cluster locks. Used + * for hard readonly access to determine whether any journal requires + * recovery. Also used to refresh the recovery generation numbers after + * a journal has been recovered by another node. + */ int ocfs2_check_journals_nolocks(struct ocfs2_super *osb) { int ret = 0; unsigned int slot; - struct buffer_head *di_bh; + struct buffer_head *di_bh = NULL; struct ocfs2_dinode *di; - struct inode *journal = NULL; + int journal_dirty = 0; for(slot = 0; slot < osb->max_slots; slot++) { - journal = ocfs2_get_system_file_inode(osb, - JOURNAL_SYSTEM_INODE, - slot); - if (!journal || is_bad_inode(journal)) { - ret = -EACCES; - mlog_errno(ret); - goto out; - } - - di_bh = NULL; - ret = ocfs2_read_block(osb, OCFS2_I(journal)->ip_blkno, &di_bh, - 0, journal); - if (ret < 0) { + ret = ocfs2_read_journal_inode(osb, slot, &di_bh, NULL); + if (ret) { mlog_errno(ret); goto out; } di = (struct ocfs2_dinode *) di_bh->b_data; + osb->slot_recovery_generations[slot] = + ocfs2_get_recovery_generation(di); + if (le32_to_cpu(di->id1.journal1.ij_flags) & OCFS2_JOURNAL_DIRTY_FL) - ret = -EROFS; + journal_dirty = 1; brelse(di_bh); - if (ret) - break; + di_bh = NULL; } out: - if (journal) - iput(journal); - + if (journal_dirty) + ret = -EROFS; return ret; } diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index db82be2..2178ebf 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -161,7 +161,8 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, void ocfs2_journal_shutdown(struct ocfs2_super *osb); int ocfs2_journal_wipe(struct ocfs2_journal *journal, int full); -int ocfs2_journal_load(struct ocfs2_journal *journal, int local); +int ocfs2_journal_load(struct ocfs2_journal *journal, int local, + int replayed); int ocfs2_check_journals_nolocks(struct ocfs2_super *osb); void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num); diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 1cb814b..7f625f2 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -204,6 +204,8 @@ struct ocfs2_super struct ocfs2_slot_info *slot_info; + u32 *slot_recovery_generations; + spinlock_t node_map_lock; u64 root_blkno; diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index 3f19451..4f61985 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -660,7 +660,10 @@ struct ocfs2_dinode { struct { /* Info for journal system inodes */ __le32 ij_flags; /* Mounted, version, etc. */ - __le32 ij_pad; + __le32 ij_recovery_generation; /* Incremented when the + journal is recovered + after an unclean + shutdown */ } journal1; } id1; /* Inode type dependant 1 */ /*C0*/ union { diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c index 10e149a..07f348b 100644 --- a/fs/ocfs2/stackglue.c +++ b/fs/ocfs2/stackglue.c @@ -97,13 +97,14 @@ static int ocfs2_stack_driver_request(const char *stack_name, goto out; } - /* Ok, the stack is pinned */ - p->sp_count++; active_stack = p; - rc = 0; out: + /* If we found it, pin it */ + if (!rc) + active_stack->sp_count++; + spin_unlock(&ocfs2_stack_lock); return rc; } diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index ccecfe5..88255d3f 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1118,7 +1118,7 @@ bail: return status; } -static void ocfs2_inode_init_once(struct kmem_cache *cachep, void *data) +static void ocfs2_inode_init_once(void *data) { struct ocfs2_inode_info *oi = data; @@ -1442,6 +1442,15 @@ static int ocfs2_initialize_super(struct super_block *sb, } mlog(0, "max_slots for this device: %u\n", osb->max_slots); + osb->slot_recovery_generations = + kcalloc(osb->max_slots, sizeof(*osb->slot_recovery_generations), + GFP_KERNEL); + if (!osb->slot_recovery_generations) { + status = -ENOMEM; + mlog_errno(status); + goto bail; + } + init_waitqueue_head(&osb->osb_wipe_event); osb->osb_orphan_wipes = kcalloc(osb->max_slots, sizeof(*osb->osb_orphan_wipes), @@ -1703,7 +1712,7 @@ static int ocfs2_check_volume(struct ocfs2_super *osb) local = ocfs2_mount_local(osb); /* will play back anything left in the journal. */ - status = ocfs2_journal_load(osb->journal, local); + status = ocfs2_journal_load(osb->journal, local, dirty); if (status < 0) { mlog(ML_ERROR, "ocfs2 journal load failed! %d\n", status); goto finally; @@ -1768,6 +1777,7 @@ static void ocfs2_delete_osb(struct ocfs2_super *osb) ocfs2_free_slot_info(osb); kfree(osb->osb_orphan_wipes); + kfree(osb->slot_recovery_generations); /* FIXME * This belongs in journal shutdown, but because we have to * allocate osb->journal at the start of ocfs2_initalize_osb(), diff --git a/fs/omfs/Makefile b/fs/omfs/Makefile new file mode 100644 index 0000000..8b82b63 --- /dev/null +++ b/fs/omfs/Makefile @@ -0,0 +1,4 @@ + +obj-$(CONFIG_OMFS_FS) += omfs.o + +omfs-y := bitmap.o dir.o file.o inode.o diff --git a/fs/omfs/bitmap.c b/fs/omfs/bitmap.c new file mode 100644 index 0000000..e1c0ec0 --- /dev/null +++ b/fs/omfs/bitmap.c @@ -0,0 +1,193 @@ +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/buffer_head.h> +#include <asm/div64.h> +#include "omfs.h" + +unsigned long omfs_count_free(struct super_block *sb) +{ + unsigned int i; + unsigned long sum = 0; + struct omfs_sb_info *sbi = OMFS_SB(sb); + int nbits = sb->s_blocksize * 8; + + for (i = 0; i < sbi->s_imap_size; i++) + sum += nbits - bitmap_weight(sbi->s_imap[i], nbits); + + return sum; +} + +/* + * Counts the run of zero bits starting at bit up to max. + * It handles the case where a run might spill over a buffer. + * Called with bitmap lock. + */ +static int count_run(unsigned long **addr, int nbits, + int addrlen, int bit, int max) +{ + int count = 0; + int x; + + for (; addrlen > 0; addrlen--, addr++) { + x = find_next_bit(*addr, nbits, bit); + count += x - bit; + + if (x < nbits || count > max) + return min(count, max); + + bit = 0; + } + return min(count, max); +} + +/* + * Sets or clears the run of count bits starting with bit. + * Called with bitmap lock. + */ +static int set_run(struct super_block *sb, int map, + int nbits, int bit, int count, int set) +{ + int i; + int err; + struct buffer_head *bh; + struct omfs_sb_info *sbi = OMFS_SB(sb); + + err = -ENOMEM; + bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + + for (i = 0; i < count; i++, bit++) { + if (bit >= nbits) { + bit = 0; + map++; + + mark_buffer_dirty(bh); + brelse(bh); + bh = sb_bread(sb, + clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + } + if (set) { + set_bit(bit, sbi->s_imap[map]); + set_bit(bit, (unsigned long *)bh->b_data); + } else { + clear_bit(bit, sbi->s_imap[map]); + clear_bit(bit, (unsigned long *)bh->b_data); + } + } + mark_buffer_dirty(bh); + brelse(bh); + err = 0; +out: + return err; +} + +/* + * Tries to allocate exactly one block. Returns true if sucessful. + */ +int omfs_allocate_block(struct super_block *sb, u64 block) +{ + struct buffer_head *bh; + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + unsigned int map, bit; + int ret = 0; + u64 tmp; + + tmp = block; + bit = do_div(tmp, bits_per_entry); + map = tmp; + + mutex_lock(&sbi->s_bitmap_lock); + if (map >= sbi->s_imap_size || test_and_set_bit(bit, sbi->s_imap[map])) + goto out; + + if (sbi->s_bitmap_ino > 0) { + bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map); + if (!bh) + goto out; + + set_bit(bit, (unsigned long *)bh->b_data); + mark_buffer_dirty(bh); + brelse(bh); + } + ret = 1; +out: + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} + + +/* + * Tries to allocate a set of blocks. The request size depends on the + * type: for inodes, we must allocate sbi->s_mirrors blocks, and for file + * blocks, we try to allocate sbi->s_clustersize, but can always get away + * with just one block. + */ +int omfs_allocate_range(struct super_block *sb, + int min_request, + int max_request, + u64 *return_block, + int *return_size) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + int ret = 0; + int i, run, bit; + + mutex_lock(&sbi->s_bitmap_lock); + for (i = 0; i < sbi->s_imap_size; i++) { + bit = 0; + while (bit < bits_per_entry) { + bit = find_next_zero_bit(sbi->s_imap[i], bits_per_entry, + bit); + + if (bit == bits_per_entry) + break; + + run = count_run(&sbi->s_imap[i], bits_per_entry, + sbi->s_imap_size-i, bit, max_request); + + if (run >= min_request) + goto found; + bit += run; + } + } + ret = -ENOSPC; + goto out; + +found: + *return_block = i * bits_per_entry + bit; + *return_size = run; + ret = set_run(sb, i, bits_per_entry, bit, run, 1); + +out: + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} + +/* + * Clears count bits starting at a given block. + */ +int omfs_clear_range(struct super_block *sb, u64 block, int count) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int bits_per_entry = 8 * sb->s_blocksize; + u64 tmp; + unsigned int map, bit; + int ret; + + tmp = block; + bit = do_div(tmp, bits_per_entry); + map = tmp; + + if (map >= sbi->s_imap_size) + return 0; + + mutex_lock(&sbi->s_bitmap_lock); + ret = set_run(sb, map, bits_per_entry, bit, count, 0); + mutex_unlock(&sbi->s_bitmap_lock); + return ret; +} diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c new file mode 100644 index 0000000..c0757e9 --- /dev/null +++ b/fs/omfs/dir.c @@ -0,0 +1,504 @@ +/* + * OMFS (as used by RIO Karma) directory operations. + * Copyright (C) 2005 Bob Copeland <me@bobcopeland.com> + * Released under GPL v2. + */ + +#include <linux/fs.h> +#include <linux/ctype.h> +#include <linux/buffer_head.h> +#include "omfs.h" + +static int omfs_hash(const char *name, int namelen, int mod) +{ + int i, hash = 0; + for (i = 0; i < namelen; i++) + hash ^= tolower(name[i]) << (i % 24); + return hash % mod; +} + +/* + * Finds the bucket for a given name and reads the containing block; + * *ofs is set to the offset of the first list entry. + */ +static struct buffer_head *omfs_get_bucket(struct inode *dir, + const char *name, int namelen, int *ofs) +{ + int nbuckets = (dir->i_size - OMFS_DIR_START)/8; + int block = clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino); + int bucket = omfs_hash(name, namelen, nbuckets); + + *ofs = OMFS_DIR_START + bucket * 8; + return sb_bread(dir->i_sb, block); +} + +static struct buffer_head *omfs_scan_list(struct inode *dir, u64 block, + const char *name, int namelen, + u64 *prev_block) +{ + struct buffer_head *bh; + struct omfs_inode *oi; + int err = -ENOENT; + *prev_block = ~0; + + while (block != ~0) { + bh = sb_bread(dir->i_sb, + clus_to_blk(OMFS_SB(dir->i_sb), block)); + if (!bh) { + err = -EIO; + goto err; + } + + oi = (struct omfs_inode *) bh->b_data; + if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, block)) { + brelse(bh); + goto err; + } + + if (strncmp(oi->i_name, name, namelen) == 0) + return bh; + + *prev_block = block; + block = be64_to_cpu(oi->i_sibling); + brelse(bh); + } +err: + return ERR_PTR(err); +} + +static struct buffer_head *omfs_find_entry(struct inode *dir, + const char *name, int namelen) +{ + struct buffer_head *bh; + int ofs; + u64 block, dummy; + + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + return ERR_PTR(-EIO); + + block = be64_to_cpu(*((__be64 *) &bh->b_data[ofs])); + brelse(bh); + + return omfs_scan_list(dir, block, name, namelen, &dummy); +} + +int omfs_make_empty(struct inode *inode, struct super_block *sb) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + int block = clus_to_blk(sbi, inode->i_ino); + struct buffer_head *bh; + struct omfs_inode *oi; + + bh = sb_bread(sb, block); + if (!bh) + return -ENOMEM; + + memset(bh->b_data, 0, sizeof(struct omfs_inode)); + + if (inode->i_mode & S_IFDIR) { + memset(&bh->b_data[OMFS_DIR_START], 0xff, + sbi->s_sys_blocksize - OMFS_DIR_START); + } else + omfs_make_empty_table(bh, OMFS_EXTENT_START); + + oi = (struct omfs_inode *) bh->b_data; + oi->i_head.h_self = cpu_to_be64(inode->i_ino); + oi->i_sibling = ~cpu_to_be64(0ULL); + + mark_buffer_dirty(bh); + brelse(bh); + return 0; +} + +static int omfs_add_link(struct dentry *dentry, struct inode *inode) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct omfs_inode *oi; + struct buffer_head *bh; + u64 block; + __be64 *entry; + int ofs; + + /* just prepend to head of queue in proper bucket */ + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + goto out; + + entry = (__be64 *) &bh->b_data[ofs]; + block = be64_to_cpu(*entry); + *entry = cpu_to_be64(inode->i_ino); + mark_buffer_dirty(bh); + brelse(bh); + + /* now set the sibling and parent pointers on the new inode */ + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), inode->i_ino)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + memcpy(oi->i_name, name, namelen); + memset(oi->i_name + namelen, 0, OMFS_NAMELEN - namelen); + oi->i_sibling = cpu_to_be64(block); + oi->i_parent = cpu_to_be64(dir->i_ino); + mark_buffer_dirty(bh); + brelse(bh); + + dir->i_ctime = CURRENT_TIME_SEC; + + /* mark affected inodes dirty to rebuild checksums */ + mark_inode_dirty(dir); + mark_inode_dirty(inode); + return 0; +out: + return -ENOMEM; +} + +static int omfs_delete_entry(struct dentry *dentry) +{ + struct inode *dir = dentry->d_parent->d_inode; + struct inode *dirty; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct omfs_inode *oi; + struct buffer_head *bh, *bh2; + __be64 *entry, next; + u64 block, prev; + int ofs; + int err = -ENOMEM; + + /* delete the proper node in the bucket's linked list */ + bh = omfs_get_bucket(dir, name, namelen, &ofs); + if (!bh) + goto out; + + entry = (__be64 *) &bh->b_data[ofs]; + block = be64_to_cpu(*entry); + + bh2 = omfs_scan_list(dir, block, name, namelen, &prev); + if (IS_ERR(bh2)) { + err = PTR_ERR(bh2); + goto out_free_bh; + } + + oi = (struct omfs_inode *) bh2->b_data; + next = oi->i_sibling; + brelse(bh2); + + if (prev != ~0) { + /* found in middle of list, get list ptr */ + brelse(bh); + bh = sb_bread(dir->i_sb, + clus_to_blk(OMFS_SB(dir->i_sb), prev)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + entry = &oi->i_sibling; + } + + *entry = next; + mark_buffer_dirty(bh); + + if (prev != ~0) { + dirty = omfs_iget(dir->i_sb, prev); + if (!IS_ERR(dirty)) { + mark_inode_dirty(dirty); + iput(dirty); + } + } + + err = 0; +out_free_bh: + brelse(bh); +out: + return err; +} + +static int omfs_dir_is_empty(struct inode *inode) +{ + int nbuckets = (inode->i_size - OMFS_DIR_START) / 8; + struct buffer_head *bh; + u64 *ptr; + int i; + + bh = sb_bread(inode->i_sb, clus_to_blk(OMFS_SB(inode->i_sb), + inode->i_ino)); + + if (!bh) + return 0; + + ptr = (u64 *) &bh->b_data[OMFS_DIR_START]; + + for (i = 0; i < nbuckets; i++, ptr++) + if (*ptr != ~0) + break; + + brelse(bh); + return *ptr != ~0; +} + +static int omfs_unlink(struct inode *dir, struct dentry *dentry) +{ + int ret; + struct inode *inode = dentry->d_inode; + + ret = omfs_delete_entry(dentry); + if (ret) + goto end_unlink; + + inode_dec_link_count(inode); + mark_inode_dirty(dir); + +end_unlink: + return ret; +} + +static int omfs_rmdir(struct inode *dir, struct dentry *dentry) +{ + int err = -ENOTEMPTY; + struct inode *inode = dentry->d_inode; + + if (omfs_dir_is_empty(inode)) { + err = omfs_unlink(dir, dentry); + if (!err) + inode_dec_link_count(inode); + } + return err; +} + +static int omfs_add_node(struct inode *dir, struct dentry *dentry, int mode) +{ + int err; + struct inode *inode = omfs_new_inode(dir, mode); + + if (IS_ERR(inode)) + return PTR_ERR(inode); + + err = omfs_make_empty(inode, dir->i_sb); + if (err) + goto out_free_inode; + + err = omfs_add_link(dentry, inode); + if (err) + goto out_free_inode; + + d_instantiate(dentry, inode); + return 0; + +out_free_inode: + iput(inode); + return err; +} + +static int omfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +{ + return omfs_add_node(dir, dentry, mode | S_IFDIR); +} + +static int omfs_create(struct inode *dir, struct dentry *dentry, int mode, + struct nameidata *nd) +{ + return omfs_add_node(dir, dentry, mode | S_IFREG); +} + +static struct dentry *omfs_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct buffer_head *bh; + struct inode *inode = NULL; + + if (dentry->d_name.len > OMFS_NAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + bh = omfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len); + if (!IS_ERR(bh)) { + struct omfs_inode *oi = (struct omfs_inode *)bh->b_data; + ino_t ino = be64_to_cpu(oi->i_head.h_self); + brelse(bh); + inode = omfs_iget(dir->i_sb, ino); + if (IS_ERR(inode)) + return ERR_CAST(inode); + } + d_add(dentry, inode); + return NULL; +} + +/* sanity check block's self pointer */ +int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header, + u64 fsblock) +{ + int is_bad; + u64 ino = be64_to_cpu(header->h_self); + is_bad = ((ino != fsblock) || (ino < sbi->s_root_ino) || + (ino > sbi->s_num_blocks)); + + if (is_bad) + printk(KERN_WARNING "omfs: bad hash chain detected\n"); + + return is_bad; +} + +static int omfs_fill_chain(struct file *filp, void *dirent, filldir_t filldir, + u64 fsblock, int hindex) +{ + struct inode *dir = filp->f_dentry->d_inode; + struct buffer_head *bh; + struct omfs_inode *oi; + u64 self; + int res = 0; + unsigned char d_type; + + /* follow chain in this bucket */ + while (fsblock != ~0) { + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), + fsblock)); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, fsblock)) { + brelse(bh); + goto out; + } + + self = fsblock; + fsblock = be64_to_cpu(oi->i_sibling); + + /* skip visited nodes */ + if (hindex) { + hindex--; + brelse(bh); + continue; + } + + d_type = (oi->i_type == OMFS_DIR) ? DT_DIR : DT_REG; + + res = filldir(dirent, oi->i_name, strnlen(oi->i_name, + OMFS_NAMELEN), filp->f_pos, self, d_type); + if (res == 0) + filp->f_pos++; + brelse(bh); + } +out: + return res; +} + +static int omfs_rename(struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry) +{ + struct inode *new_inode = new_dentry->d_inode; + struct inode *old_inode = old_dentry->d_inode; + struct buffer_head *bh; + int is_dir; + int err; + + is_dir = S_ISDIR(old_inode->i_mode); + + if (new_inode) { + /* overwriting existing file/dir */ + err = -ENOTEMPTY; + if (is_dir && !omfs_dir_is_empty(new_inode)) + goto out; + + err = -ENOENT; + bh = omfs_find_entry(new_dir, new_dentry->d_name.name, + new_dentry->d_name.len); + if (IS_ERR(bh)) + goto out; + brelse(bh); + + err = omfs_unlink(new_dir, new_dentry); + if (err) + goto out; + } + + /* since omfs locates files by name, we need to unlink _before_ + * adding the new link or we won't find the old one */ + inode_inc_link_count(old_inode); + err = omfs_unlink(old_dir, old_dentry); + if (err) { + inode_dec_link_count(old_inode); + goto out; + } + + err = omfs_add_link(new_dentry, old_inode); + if (err) + goto out; + + old_inode->i_ctime = CURRENT_TIME_SEC; +out: + return err; +} + +static int omfs_readdir(struct file *filp, void *dirent, filldir_t filldir) +{ + struct inode *dir = filp->f_dentry->d_inode; + struct buffer_head *bh; + loff_t offset, res; + unsigned int hchain, hindex; + int nbuckets; + u64 fsblock; + int ret = -EINVAL; + + if (filp->f_pos >> 32) + goto success; + + switch ((unsigned long) filp->f_pos) { + case 0: + if (filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR) < 0) + goto success; + filp->f_pos++; + /* fall through */ + case 1: + if (filldir(dirent, "..", 2, 1, + parent_ino(filp->f_dentry), DT_DIR) < 0) + goto success; + filp->f_pos = 1 << 20; + /* fall through */ + } + + nbuckets = (dir->i_size - OMFS_DIR_START) / 8; + + /* high 12 bits store bucket + 1 and low 20 bits store hash index */ + hchain = (filp->f_pos >> 20) - 1; + hindex = filp->f_pos & 0xfffff; + + bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino)); + if (!bh) + goto out; + + offset = OMFS_DIR_START + hchain * 8; + + for (; hchain < nbuckets; hchain++, offset += 8) { + fsblock = be64_to_cpu(*((__be64 *) &bh->b_data[offset])); + + res = omfs_fill_chain(filp, dirent, filldir, fsblock, hindex); + hindex = 0; + if (res < 0) + break; + + filp->f_pos = (hchain+2) << 20; + } + brelse(bh); +success: + ret = 0; +out: + return ret; +} + +struct inode_operations omfs_dir_inops = { + .lookup = omfs_lookup, + .mkdir = omfs_mkdir, + .rename = omfs_rename, + .create = omfs_create, + .unlink = omfs_unlink, + .rmdir = omfs_rmdir, +}; + +struct file_operations omfs_dir_operations = { + .read = generic_read_dir, + .readdir = omfs_readdir, +}; diff --git a/fs/omfs/file.c b/fs/omfs/file.c new file mode 100644 index 0000000..834b233 --- /dev/null +++ b/fs/omfs/file.c @@ -0,0 +1,365 @@ +/* + * OMFS (as used by RIO Karma) file operations. + * Copyright (C) 2005 Bob Copeland <me@bobcopeland.com> + * Released under GPL v2. + */ + +#include <linux/version.h> +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/buffer_head.h> +#include <linux/mpage.h> +#include "omfs.h" + +static int omfs_sync_file(struct file *file, struct dentry *dentry, + int datasync) +{ + struct inode *inode = dentry->d_inode; + int err; + + err = sync_mapping_buffers(inode->i_mapping); + if (!(inode->i_state & I_DIRTY)) + return err; + if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) + return err; + err |= omfs_sync_inode(inode); + return err ? -EIO : 0; +} + +static u32 omfs_max_extents(struct omfs_sb_info *sbi, int offset) +{ + return (sbi->s_sys_blocksize - offset - + sizeof(struct omfs_extent)) / + sizeof(struct omfs_extent_entry) + 1; +} + +void omfs_make_empty_table(struct buffer_head *bh, int offset) +{ + struct omfs_extent *oe = (struct omfs_extent *) &bh->b_data[offset]; + + oe->e_next = ~cpu_to_be64(0ULL); + oe->e_extent_count = cpu_to_be32(1), + oe->e_fill = cpu_to_be32(0x22), + oe->e_entry.e_cluster = ~cpu_to_be64(0ULL); + oe->e_entry.e_blocks = ~cpu_to_be64(0ULL); +} + +int omfs_shrink_inode(struct inode *inode) +{ + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + struct omfs_extent *oe; + struct omfs_extent_entry *entry; + struct buffer_head *bh; + u64 next, last; + u32 extent_count; + u32 max_extents; + int ret; + + /* traverse extent table, freeing each entry that is greater + * than inode->i_size; + */ + next = inode->i_ino; + + /* only support truncate -> 0 for now */ + ret = -EIO; + if (inode->i_size != 0) + goto out; + + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + + oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]); + max_extents = omfs_max_extents(sbi, OMFS_EXTENT_START); + + for (;;) { + + if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next)) + goto out_brelse; + + extent_count = be32_to_cpu(oe->e_extent_count); + + if (extent_count > max_extents) + goto out_brelse; + + last = next; + next = be64_to_cpu(oe->e_next); + entry = &oe->e_entry; + + /* ignore last entry as it is the terminator */ + for (; extent_count > 1; extent_count--) { + u64 start, count; + start = be64_to_cpu(entry->e_cluster); + count = be64_to_cpu(entry->e_blocks); + + omfs_clear_range(inode->i_sb, start, (int) count); + entry++; + } + omfs_make_empty_table(bh, (char *) oe - bh->b_data); + mark_buffer_dirty(bh); + brelse(bh); + + if (last != inode->i_ino) + omfs_clear_range(inode->i_sb, last, sbi->s_mirrors); + + if (next == ~0) + break; + + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]); + max_extents = omfs_max_extents(sbi, OMFS_EXTENT_CONT); + } + ret = 0; +out: + return ret; +out_brelse: + brelse(bh); + return ret; +} + +static void omfs_truncate(struct inode *inode) +{ + omfs_shrink_inode(inode); + mark_inode_dirty(inode); +} + +/* + * Add new blocks to the current extent, or create new entries/continuations + * as necessary. + */ +static int omfs_grow_extent(struct inode *inode, struct omfs_extent *oe, + u64 *ret_block) +{ + struct omfs_extent_entry *terminator; + struct omfs_extent_entry *entry = &oe->e_entry; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + u32 extent_count = be32_to_cpu(oe->e_extent_count); + u64 new_block = 0; + u32 max_count; + int new_count; + int ret = 0; + + /* reached the end of the extent table with no blocks mapped. + * there are three possibilities for adding: grow last extent, + * add a new extent to the current extent table, and add a + * continuation inode. in last two cases need an allocator for + * sbi->s_cluster_size + */ + + /* TODO: handle holes */ + + /* should always have a terminator */ + if (extent_count < 1) + return -EIO; + + /* trivially grow current extent, if next block is not taken */ + terminator = entry + extent_count - 1; + if (extent_count > 1) { + entry = terminator-1; + new_block = be64_to_cpu(entry->e_cluster) + + be64_to_cpu(entry->e_blocks); + + if (omfs_allocate_block(inode->i_sb, new_block)) { + entry->e_blocks = + cpu_to_be64(be64_to_cpu(entry->e_blocks) + 1); + terminator->e_blocks = ~(cpu_to_be64( + be64_to_cpu(~terminator->e_blocks) + 1)); + goto out; + } + } + max_count = omfs_max_extents(sbi, OMFS_EXTENT_START); + + /* TODO: add a continuation block here */ + if (be32_to_cpu(oe->e_extent_count) > max_count-1) + return -EIO; + + /* try to allocate a new cluster */ + ret = omfs_allocate_range(inode->i_sb, 1, sbi->s_clustersize, + &new_block, &new_count); + if (ret) + goto out_fail; + + /* copy terminator down an entry */ + entry = terminator; + terminator++; + memcpy(terminator, entry, sizeof(struct omfs_extent_entry)); + + entry->e_cluster = cpu_to_be64(new_block); + entry->e_blocks = cpu_to_be64((u64) new_count); + + terminator->e_blocks = ~(cpu_to_be64( + be64_to_cpu(~terminator->e_blocks) + (u64) new_count)); + + /* write in new entry */ + oe->e_extent_count = cpu_to_be32(1 + be32_to_cpu(oe->e_extent_count)); + +out: + *ret_block = new_block; +out_fail: + return ret; +} + +/* + * Scans across the directory table for a given file block number. + * If block not found, return 0. + */ +static sector_t find_block(struct inode *inode, struct omfs_extent_entry *ent, + sector_t block, int count, int *left) +{ + /* count > 1 because of terminator */ + sector_t searched = 0; + for (; count > 1; count--) { + int numblocks = clus_to_blk(OMFS_SB(inode->i_sb), + be64_to_cpu(ent->e_blocks)); + + if (block >= searched && + block < searched + numblocks) { + /* + * found it at cluster + (block - searched) + * numblocks - (block - searched) is remainder + */ + *left = numblocks - (block - searched); + return clus_to_blk(OMFS_SB(inode->i_sb), + be64_to_cpu(ent->e_cluster)) + + block - searched; + } + searched += numblocks; + ent++; + } + return 0; +} + +static int omfs_get_block(struct inode *inode, sector_t block, + struct buffer_head *bh_result, int create) +{ + struct buffer_head *bh; + sector_t next, offset; + int ret; + u64 new_block; + u32 max_extents; + int extent_count; + struct omfs_extent *oe; + struct omfs_extent_entry *entry; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + int max_blocks = bh_result->b_size >> inode->i_blkbits; + int remain; + + ret = -EIO; + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, inode->i_ino)); + if (!bh) + goto out; + + oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]); + max_extents = omfs_max_extents(sbi, OMFS_EXTENT_START); + next = inode->i_ino; + + for (;;) { + + if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next)) + goto out_brelse; + + extent_count = be32_to_cpu(oe->e_extent_count); + next = be64_to_cpu(oe->e_next); + entry = &oe->e_entry; + + if (extent_count > max_extents) + goto out_brelse; + + offset = find_block(inode, entry, block, extent_count, &remain); + if (offset > 0) { + ret = 0; + map_bh(bh_result, inode->i_sb, offset); + if (remain > max_blocks) + remain = max_blocks; + bh_result->b_size = (remain << inode->i_blkbits); + goto out_brelse; + } + if (next == ~0) + break; + + brelse(bh); + bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next)); + if (!bh) + goto out; + oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]); + max_extents = omfs_max_extents(sbi, OMFS_EXTENT_CONT); + } + if (create) { + ret = omfs_grow_extent(inode, oe, &new_block); + if (ret == 0) { + mark_buffer_dirty(bh); + mark_inode_dirty(inode); + map_bh(bh_result, inode->i_sb, + clus_to_blk(sbi, new_block)); + } + } +out_brelse: + brelse(bh); +out: + return ret; +} + +static int omfs_readpage(struct file *file, struct page *page) +{ + return block_read_full_page(page, omfs_get_block); +} + +static int omfs_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + return mpage_readpages(mapping, pages, nr_pages, omfs_get_block); +} + +static int omfs_writepage(struct page *page, struct writeback_control *wbc) +{ + return block_write_full_page(page, omfs_get_block, wbc); +} + +static int +omfs_writepages(struct address_space *mapping, struct writeback_control *wbc) +{ + return mpage_writepages(mapping, wbc, omfs_get_block); +} + +static int omfs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + *pagep = NULL; + return block_write_begin(file, mapping, pos, len, flags, + pagep, fsdata, omfs_get_block); +} + +static sector_t omfs_bmap(struct address_space *mapping, sector_t block) +{ + return generic_block_bmap(mapping, block, omfs_get_block); +} + +struct file_operations omfs_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = generic_file_aio_read, + .aio_write = generic_file_aio_write, + .mmap = generic_file_mmap, + .fsync = omfs_sync_file, + .splice_read = generic_file_splice_read, +}; + +struct inode_operations omfs_file_inops = { + .truncate = omfs_truncate +}; + +struct address_space_operations omfs_aops = { + .readpage = omfs_readpage, + .readpages = omfs_readpages, + .writepage = omfs_writepage, + .writepages = omfs_writepages, + .sync_page = block_sync_page, + .write_begin = omfs_write_begin, + .write_end = generic_write_end, + .bmap = omfs_bmap, +}; + diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c new file mode 100644 index 0000000..d29047b --- /dev/null +++ b/fs/omfs/inode.c @@ -0,0 +1,553 @@ +/* + * Optimized MPEG FS - inode and super operations. + * Copyright (C) 2006 Bob Copeland <me@bobcopeland.com> + * Released under GPL v2. + */ +#include <linux/version.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/fs.h> +#include <linux/vfs.h> +#include <linux/parser.h> +#include <linux/buffer_head.h> +#include <linux/vmalloc.h> +#include <linux/crc-itu-t.h> +#include "omfs.h" + +MODULE_AUTHOR("Bob Copeland <me@bobcopeland.com>"); +MODULE_DESCRIPTION("OMFS (ReplayTV/Karma) Filesystem for Linux"); +MODULE_LICENSE("GPL"); + +struct inode *omfs_new_inode(struct inode *dir, int mode) +{ + struct inode *inode; + u64 new_block; + int err; + int len; + struct omfs_sb_info *sbi = OMFS_SB(dir->i_sb); + + inode = new_inode(dir->i_sb); + if (!inode) + return ERR_PTR(-ENOMEM); + + err = omfs_allocate_range(dir->i_sb, sbi->s_mirrors, sbi->s_mirrors, + &new_block, &len); + if (err) + goto fail; + + inode->i_ino = new_block; + inode->i_mode = mode; + inode->i_uid = current->fsuid; + inode->i_gid = current->fsgid; + inode->i_blocks = 0; + inode->i_mapping->a_ops = &omfs_aops; + + inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; + switch (mode & S_IFMT) { + case S_IFDIR: + inode->i_op = &omfs_dir_inops; + inode->i_fop = &omfs_dir_operations; + inode->i_size = sbi->s_sys_blocksize; + inc_nlink(inode); + break; + case S_IFREG: + inode->i_op = &omfs_file_inops; + inode->i_fop = &omfs_file_operations; + inode->i_size = 0; + break; + } + + insert_inode_hash(inode); + mark_inode_dirty(inode); + return inode; +fail: + make_bad_inode(inode); + iput(inode); + return ERR_PTR(err); +} + +/* + * Update the header checksums for a dirty inode based on its contents. + * Caller is expected to hold the buffer head underlying oi and mark it + * dirty. + */ +static void omfs_update_checksums(struct omfs_inode *oi) +{ + int xor, i, ofs = 0, count; + u16 crc = 0; + unsigned char *ptr = (unsigned char *) oi; + + count = be32_to_cpu(oi->i_head.h_body_size); + ofs = sizeof(struct omfs_header); + + crc = crc_itu_t(crc, ptr + ofs, count); + oi->i_head.h_crc = cpu_to_be16(crc); + + xor = ptr[0]; + for (i = 1; i < OMFS_XOR_COUNT; i++) + xor ^= ptr[i]; + + oi->i_head.h_check_xor = xor; +} + +static int omfs_write_inode(struct inode *inode, int wait) +{ + struct omfs_inode *oi; + struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb); + struct buffer_head *bh, *bh2; + unsigned int block; + u64 ctime; + int i; + int ret = -EIO; + int sync_failed = 0; + + /* get current inode since we may have written sibling ptrs etc. */ + block = clus_to_blk(sbi, inode->i_ino); + bh = sb_bread(inode->i_sb, block); + if (!bh) + goto out; + + oi = (struct omfs_inode *) bh->b_data; + + oi->i_head.h_self = cpu_to_be64(inode->i_ino); + if (S_ISDIR(inode->i_mode)) + oi->i_type = OMFS_DIR; + else if (S_ISREG(inode->i_mode)) + oi->i_type = OMFS_FILE; + else { + printk(KERN_WARNING "omfs: unknown file type: %d\n", + inode->i_mode); + goto out_brelse; + } + + oi->i_head.h_body_size = cpu_to_be32(sbi->s_sys_blocksize - + sizeof(struct omfs_header)); + oi->i_head.h_version = 1; + oi->i_head.h_type = OMFS_INODE_NORMAL; + oi->i_head.h_magic = OMFS_IMAGIC; + oi->i_size = cpu_to_be64(inode->i_size); + + ctime = inode->i_ctime.tv_sec * 1000LL + + ((inode->i_ctime.tv_nsec + 999)/1000); + oi->i_ctime = cpu_to_be64(ctime); + + omfs_update_checksums(oi); + + mark_buffer_dirty(bh); + if (wait) { + sync_dirty_buffer(bh); + if (buffer_req(bh) && !buffer_uptodate(bh)) + sync_failed = 1; + } + + /* if mirroring writes, copy to next fsblock */ + for (i = 1; i < sbi->s_mirrors; i++) { + bh2 = sb_bread(inode->i_sb, block + i * + (sbi->s_blocksize / sbi->s_sys_blocksize)); + if (!bh2) + goto out_brelse; + + memcpy(bh2->b_data, bh->b_data, bh->b_size); + mark_buffer_dirty(bh2); + if (wait) { + sync_dirty_buffer(bh2); + if (buffer_req(bh2) && !buffer_uptodate(bh2)) + sync_failed = 1; + } + brelse(bh2); + } + ret = (sync_failed) ? -EIO : 0; +out_brelse: + brelse(bh); +out: + return ret; +} + +int omfs_sync_inode(struct inode *inode) +{ + return omfs_write_inode(inode, 1); +} + +/* + * called when an entry is deleted, need to clear the bits in the + * bitmaps. + */ +static void omfs_delete_inode(struct inode *inode) +{ + truncate_inode_pages(&inode->i_data, 0); + + if (S_ISREG(inode->i_mode)) { + inode->i_size = 0; + omfs_shrink_inode(inode); + } + + omfs_clear_range(inode->i_sb, inode->i_ino, 2); + clear_inode(inode); +} + +struct inode *omfs_iget(struct super_block *sb, ino_t ino) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + struct omfs_inode *oi; + struct buffer_head *bh; + unsigned int block; + u64 ctime; + unsigned long nsecs; + struct inode *inode; + + inode = iget_locked(sb, ino); + if (!inode) + return ERR_PTR(-ENOMEM); + if (!(inode->i_state & I_NEW)) + return inode; + + block = clus_to_blk(sbi, ino); + bh = sb_bread(inode->i_sb, block); + if (!bh) + goto iget_failed; + + oi = (struct omfs_inode *)bh->b_data; + + /* check self */ + if (ino != be64_to_cpu(oi->i_head.h_self)) + goto fail_bh; + + inode->i_uid = sbi->s_uid; + inode->i_gid = sbi->s_gid; + + ctime = be64_to_cpu(oi->i_ctime); + nsecs = do_div(ctime, 1000) * 1000L; + + inode->i_atime.tv_sec = ctime; + inode->i_mtime.tv_sec = ctime; + inode->i_ctime.tv_sec = ctime; + inode->i_atime.tv_nsec = nsecs; + inode->i_mtime.tv_nsec = nsecs; + inode->i_ctime.tv_nsec = nsecs; + + inode->i_mapping->a_ops = &omfs_aops; + + switch (oi->i_type) { + case OMFS_DIR: + inode->i_mode = S_IFDIR | (S_IRWXUGO & ~sbi->s_dmask); + inode->i_op = &omfs_dir_inops; + inode->i_fop = &omfs_dir_operations; + inode->i_size = sbi->s_sys_blocksize; + inc_nlink(inode); + break; + case OMFS_FILE: + inode->i_mode = S_IFREG | (S_IRWXUGO & ~sbi->s_fmask); + inode->i_fop = &omfs_file_operations; + inode->i_size = be64_to_cpu(oi->i_size); + break; + } + brelse(bh); + unlock_new_inode(inode); + return inode; +fail_bh: + brelse(bh); +iget_failed: + iget_failed(inode); + return ERR_PTR(-EIO); +} + +static void omfs_put_super(struct super_block *sb) +{ + struct omfs_sb_info *sbi = OMFS_SB(sb); + kfree(sbi->s_imap); + kfree(sbi); + sb->s_fs_info = NULL; +} + +static int omfs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + struct super_block *s = dentry->d_sb; + struct omfs_sb_info *sbi = OMFS_SB(s); + buf->f_type = OMFS_MAGIC; + buf->f_bsize = sbi->s_blocksize; + buf->f_blocks = sbi->s_num_blocks; + buf->f_files = sbi->s_num_blocks; + buf->f_namelen = OMFS_NAMELEN; + + buf->f_bfree = buf->f_bavail = buf->f_ffree = + omfs_count_free(s); + return 0; +} + +static struct super_operations omfs_sops = { + .write_inode = omfs_write_inode, + .delete_inode = omfs_delete_inode, + .put_super = omfs_put_super, + .statfs = omfs_statfs, + .show_options = generic_show_options, +}; + +/* + * For Rio Karma, there is an on-disk free bitmap whose location is + * stored in the root block. For ReplayTV, there is no such free bitmap + * so we have to walk the tree. Both inodes and file data are allocated + * from the same map. This array can be big (300k) so we allocate + * in units of the blocksize. + */ +static int omfs_get_imap(struct super_block *sb) +{ + int bitmap_size; + int array_size; + int count; + struct omfs_sb_info *sbi = OMFS_SB(sb); + struct buffer_head *bh; + unsigned long **ptr; + sector_t block; + + bitmap_size = DIV_ROUND_UP(sbi->s_num_blocks, 8); + array_size = DIV_ROUND_UP(bitmap_size, sb->s_blocksize); + + if (sbi->s_bitmap_ino == ~0ULL) + goto out; + + sbi->s_imap_size = array_size; + sbi->s_imap = kzalloc(array_size * sizeof(unsigned long *), GFP_KERNEL); + if (!sbi->s_imap) + goto nomem; + + block = clus_to_blk(sbi, sbi->s_bitmap_ino); + ptr = sbi->s_imap; + for (count = bitmap_size; count > 0; count -= sb->s_blocksize) { + bh = sb_bread(sb, block++); + if (!bh) + goto nomem_free; + *ptr = kmalloc(sb->s_blocksize, GFP_KERNEL); + if (!*ptr) { + brelse(bh); + goto nomem_free; + } + memcpy(*ptr, bh->b_data, sb->s_blocksize); + if (count < sb->s_blocksize) + memset((void *)*ptr + count, 0xff, + sb->s_blocksize - count); + brelse(bh); + ptr++; + } +out: + return 0; + +nomem_free: + for (count = 0; count < array_size; count++) + kfree(sbi->s_imap[count]); + + kfree(sbi->s_imap); +nomem: + sbi->s_imap = NULL; + sbi->s_imap_size = 0; + return -ENOMEM; +} + +enum { + Opt_uid, Opt_gid, Opt_umask, Opt_dmask, Opt_fmask +}; + +static match_table_t tokens = { + {Opt_uid, "uid=%u"}, + {Opt_gid, "gid=%u"}, + {Opt_umask, "umask=%o"}, + {Opt_dmask, "dmask=%o"}, + {Opt_fmask, "fmask=%o"}, +}; + +static int parse_options(char *options, struct omfs_sb_info *sbi) +{ + char *p; + substring_t args[MAX_OPT_ARGS]; + int option; + + if (!options) + return 1; + + while ((p = strsep(&options, ",")) != NULL) { + int token; + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_uid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_uid = option; + break; + case Opt_gid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_gid = option; + break; + case Opt_umask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_fmask = sbi->s_dmask = option; + break; + case Opt_dmask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_dmask = option; + break; + case Opt_fmask: + if (match_octal(&args[0], &option)) + return 0; + sbi->s_fmask = option; + break; + default: + return 0; + } + } + return 1; +} + +static int omfs_fill_super(struct super_block *sb, void *data, int silent) +{ + struct buffer_head *bh, *bh2; + struct omfs_super_block *omfs_sb; + struct omfs_root_block *omfs_rb; + struct omfs_sb_info *sbi; + struct inode *root; + sector_t start; + int ret = -EINVAL; + + save_mount_options(sb, (char *) data); + + sbi = kzalloc(sizeof(struct omfs_sb_info), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + + sb->s_fs_info = sbi; + + sbi->s_uid = current->uid; + sbi->s_gid = current->gid; + sbi->s_dmask = sbi->s_fmask = current->fs->umask; + + if (!parse_options((char *) data, sbi)) + goto end; + + sb->s_maxbytes = 0xffffffff; + + sb_set_blocksize(sb, 0x200); + + bh = sb_bread(sb, 0); + if (!bh) + goto end; + + omfs_sb = (struct omfs_super_block *)bh->b_data; + + if (omfs_sb->s_magic != cpu_to_be32(OMFS_MAGIC)) { + if (!silent) + printk(KERN_ERR "omfs: Invalid superblock (%x)\n", + omfs_sb->s_magic); + goto out_brelse_bh; + } + sb->s_magic = OMFS_MAGIC; + + sbi->s_num_blocks = be64_to_cpu(omfs_sb->s_num_blocks); + sbi->s_blocksize = be32_to_cpu(omfs_sb->s_blocksize); + sbi->s_mirrors = be32_to_cpu(omfs_sb->s_mirrors); + sbi->s_root_ino = be64_to_cpu(omfs_sb->s_root_block); + sbi->s_sys_blocksize = be32_to_cpu(omfs_sb->s_sys_blocksize); + mutex_init(&sbi->s_bitmap_lock); + + if (sbi->s_sys_blocksize > PAGE_SIZE) { + printk(KERN_ERR "omfs: sysblock size (%d) is out of range\n", + sbi->s_sys_blocksize); + goto out_brelse_bh; + } + + if (sbi->s_blocksize < sbi->s_sys_blocksize || + sbi->s_blocksize > OMFS_MAX_BLOCK_SIZE) { + printk(KERN_ERR "omfs: block size (%d) is out of range\n", + sbi->s_blocksize); + goto out_brelse_bh; + } + + /* + * Use sys_blocksize as the fs block since it is smaller than a + * page while the fs blocksize can be larger. + */ + sb_set_blocksize(sb, sbi->s_sys_blocksize); + + /* + * ...and the difference goes into a shift. sys_blocksize is always + * a power of two factor of blocksize. + */ + sbi->s_block_shift = get_bitmask_order(sbi->s_blocksize) - + get_bitmask_order(sbi->s_sys_blocksize); + + start = clus_to_blk(sbi, be64_to_cpu(omfs_sb->s_root_block)); + bh2 = sb_bread(sb, start); + if (!bh2) + goto out_brelse_bh; + + omfs_rb = (struct omfs_root_block *)bh2->b_data; + + sbi->s_bitmap_ino = be64_to_cpu(omfs_rb->r_bitmap); + sbi->s_clustersize = be32_to_cpu(omfs_rb->r_clustersize); + + if (sbi->s_num_blocks != be64_to_cpu(omfs_rb->r_num_blocks)) { + printk(KERN_ERR "omfs: block count discrepancy between " + "super and root blocks (%llx, %llx)\n", + (unsigned long long)sbi->s_num_blocks, + (unsigned long long)be64_to_cpu(omfs_rb->r_num_blocks)); + goto out_brelse_bh2; + } + + ret = omfs_get_imap(sb); + if (ret) + goto out_brelse_bh2; + + sb->s_op = &omfs_sops; + + root = omfs_iget(sb, be64_to_cpu(omfs_rb->r_root_dir)); + if (IS_ERR(root)) { + ret = PTR_ERR(root); + goto out_brelse_bh2; + } + + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + iput(root); + goto out_brelse_bh2; + } + printk(KERN_DEBUG "omfs: Mounted volume %s\n", omfs_rb->r_name); + + ret = 0; +out_brelse_bh2: + brelse(bh2); +out_brelse_bh: + brelse(bh); +end: + return ret; +} + +static int omfs_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, + void *data, struct vfsmount *m) +{ + return get_sb_bdev(fs_type, flags, dev_name, data, omfs_fill_super, m); +} + +static struct file_system_type omfs_fs_type = { + .owner = THIS_MODULE, + .name = "omfs", + .get_sb = omfs_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +static int __init init_omfs_fs(void) +{ + return register_filesystem(&omfs_fs_type); +} + +static void __exit exit_omfs_fs(void) +{ + unregister_filesystem(&omfs_fs_type); +} + +module_init(init_omfs_fs); +module_exit(exit_omfs_fs); diff --git a/fs/omfs/omfs.h b/fs/omfs/omfs.h new file mode 100644 index 0000000..2bc0f06 --- /dev/null +++ b/fs/omfs/omfs.h @@ -0,0 +1,67 @@ +#ifndef _OMFS_H +#define _OMFS_H + +#include <linux/module.h> +#include <linux/fs.h> + +#include "omfs_fs.h" + +/* In-memory structures */ +struct omfs_sb_info { + u64 s_num_blocks; + u64 s_bitmap_ino; + u64 s_root_ino; + u32 s_blocksize; + u32 s_mirrors; + u32 s_sys_blocksize; + u32 s_clustersize; + int s_block_shift; + unsigned long **s_imap; + int s_imap_size; + struct mutex s_bitmap_lock; + int s_uid; + int s_gid; + int s_dmask; + int s_fmask; +}; + +/* convert a cluster number to a scaled block number */ +static inline sector_t clus_to_blk(struct omfs_sb_info *sbi, sector_t block) +{ + return block << sbi->s_block_shift; +} + +static inline struct omfs_sb_info *OMFS_SB(struct super_block *sb) +{ + return sb->s_fs_info; +} + +/* bitmap.c */ +extern unsigned long omfs_count_free(struct super_block *sb); +extern int omfs_allocate_block(struct super_block *sb, u64 block); +extern int omfs_allocate_range(struct super_block *sb, int min_request, + int max_request, u64 *return_block, int *return_size); +extern int omfs_clear_range(struct super_block *sb, u64 block, int count); + +/* dir.c */ +extern struct file_operations omfs_dir_operations; +extern struct inode_operations omfs_dir_inops; +extern int omfs_make_empty(struct inode *inode, struct super_block *sb); +extern int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header, + u64 fsblock); + +/* file.c */ +extern struct file_operations omfs_file_operations; +extern struct inode_operations omfs_file_inops; +extern struct address_space_operations omfs_aops; +extern void omfs_make_empty_table(struct buffer_head *bh, int offset); +extern int omfs_shrink_inode(struct inode *inode); + +/* inode.c */ +extern struct inode *omfs_iget(struct super_block *sb, ino_t inode); +extern struct inode *omfs_new_inode(struct inode *dir, int mode); +extern int omfs_reserve_block(struct super_block *sb, sector_t block); +extern int omfs_find_empty_block(struct super_block *sb, int mode, ino_t *ino); +extern int omfs_sync_inode(struct inode *inode); + +#endif diff --git a/fs/omfs/omfs_fs.h b/fs/omfs/omfs_fs.h new file mode 100644 index 0000000..12cca24 --- /dev/null +++ b/fs/omfs/omfs_fs.h @@ -0,0 +1,80 @@ +#ifndef _OMFS_FS_H +#define _OMFS_FS_H + +/* OMFS On-disk structures */ + +#define OMFS_MAGIC 0xC2993D87 +#define OMFS_IMAGIC 0xD2 + +#define OMFS_DIR 'D' +#define OMFS_FILE 'F' +#define OMFS_INODE_NORMAL 'e' +#define OMFS_INODE_CONTINUATION 'c' +#define OMFS_INODE_SYSTEM 's' +#define OMFS_NAMELEN 256 +#define OMFS_DIR_START 0x1b8 +#define OMFS_EXTENT_START 0x1d0 +#define OMFS_EXTENT_CONT 0x40 +#define OMFS_XOR_COUNT 19 +#define OMFS_MAX_BLOCK_SIZE 8192 + +struct omfs_super_block { + char s_fill1[256]; + __be64 s_root_block; /* block number of omfs_root_block */ + __be64 s_num_blocks; /* total number of FS blocks */ + __be32 s_magic; /* OMFS_MAGIC */ + __be32 s_blocksize; /* size of a block */ + __be32 s_mirrors; /* # of mirrors of system blocks */ + __be32 s_sys_blocksize; /* size of non-data blocks */ +}; + +struct omfs_header { + __be64 h_self; /* FS block where this is located */ + __be32 h_body_size; /* size of useful data after header */ + __be16 h_crc; /* crc-ccitt of body_size bytes */ + char h_fill1[2]; + u8 h_version; /* version, always 1 */ + char h_type; /* OMFS_INODE_X */ + u8 h_magic; /* OMFS_IMAGIC */ + u8 h_check_xor; /* XOR of header bytes before this */ + __be32 h_fill2; +}; + +struct omfs_root_block { + struct omfs_header r_head; /* header */ + __be64 r_fill1; + __be64 r_num_blocks; /* total number of FS blocks */ + __be64 r_root_dir; /* block # of root directory */ + __be64 r_bitmap; /* block # of free space bitmap */ + __be32 r_blocksize; /* size of a block */ + __be32 r_clustersize; /* size allocated for data blocks */ + __be64 r_mirrors; /* # of mirrors of system blocks */ + char r_name[OMFS_NAMELEN]; /* partition label */ +}; + +struct omfs_inode { + struct omfs_header i_head; /* header */ + __be64 i_parent; /* parent containing this inode */ + __be64 i_sibling; /* next inode in hash bucket */ + __be64 i_ctime; /* ctime, in milliseconds */ + char i_fill1[35]; + char i_type; /* OMFS_[DIR,FILE] */ + __be32 i_fill2; + char i_fill3[64]; + char i_name[OMFS_NAMELEN]; /* filename */ + __be64 i_size; /* size of file, in bytes */ +}; + +struct omfs_extent_entry { + __be64 e_cluster; /* start location of a set of blocks */ + __be64 e_blocks; /* number of blocks after e_cluster */ +}; + +struct omfs_extent { + __be64 e_next; /* next extent table location */ + __be32 e_extent_count; /* total # extents in this table */ + __be32 e_fill; + struct omfs_extent_entry e_entry; /* start of extent entries */ +}; + +#endif @@ -122,37 +122,37 @@ static int vfs_statfs64(struct dentry *dentry, struct statfs64 *buf) return 0; } -asmlinkage long sys_statfs(const char __user * path, struct statfs __user * buf) +asmlinkage long sys_statfs(const char __user *pathname, struct statfs __user * buf) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct statfs tmp; - error = vfs_statfs_native(nd.path.dentry, &tmp); + error = vfs_statfs_native(path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_put(&nd.path); + path_put(&path); } return error; } -asmlinkage long sys_statfs64(const char __user *path, size_t sz, struct statfs64 __user *buf) +asmlinkage long sys_statfs64(const char __user *pathname, size_t sz, struct statfs64 __user *buf) { - struct nameidata nd; + struct path path; long error; if (sz != sizeof(*buf)) return -EINVAL; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (!error) { struct statfs64 tmp; - error = vfs_statfs64(nd.path.dentry, &tmp); + error = vfs_statfs64(path.dentry, &tmp); if (!error && copy_to_user(buf, &tmp, sizeof(tmp))) error = -EFAULT; - path_put(&nd.path); + path_put(&path); } return error; } @@ -223,20 +223,20 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs, return err; } -static long do_sys_truncate(const char __user * path, loff_t length) +static long do_sys_truncate(const char __user *pathname, loff_t length) { - struct nameidata nd; - struct inode * inode; + struct path path; + struct inode *inode; int error; error = -EINVAL; if (length < 0) /* sorry, but loff_t says... */ goto out; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) goto out; - inode = nd.path.dentry->d_inode; + inode = path.dentry->d_inode; /* For directories it's -EISDIR, for other non-regulars - -EINVAL */ error = -EISDIR; @@ -247,16 +247,16 @@ static long do_sys_truncate(const char __user * path, loff_t length) if (!S_ISREG(inode->i_mode)) goto dput_and_out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto dput_and_out; - error = vfs_permission(&nd, MAY_WRITE); + error = inode_permission(inode, MAY_WRITE); if (error) goto mnt_drop_write_and_out; error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + if (IS_APPEND(inode)) goto mnt_drop_write_and_out; error = get_write_access(inode); @@ -274,15 +274,15 @@ static long do_sys_truncate(const char __user * path, loff_t length) error = locks_verify_truncate(inode, NULL, length); if (!error) { DQUOT_INIT(inode); - error = do_truncate(nd.path.dentry, length, 0, NULL); + error = do_truncate(path.dentry, length, 0, NULL); } put_write_and_out: put_write_access(inode); mnt_drop_write_and_out: - mnt_drop_write(nd.path.mnt); + mnt_drop_write(path.mnt); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -425,7 +425,8 @@ out: */ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) { - struct nameidata nd; + struct path path; + struct inode *inode; int old_fsuid, old_fsgid; kernel_cap_t uninitialized_var(old_cap); /* !SECURE_NO_SETUID_FIXUP */ int res; @@ -448,7 +449,7 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) * FIXME: There is a race here against sys_capset. The * capabilities can change yet we will restore the old * value below. We should hold task_capabilities_lock, - * but we cannot because user_path_walk can sleep. + * but we cannot because user_path_at can sleep. */ #endif /* ndef CONFIG_SECURITY_FILE_CAPABILITIES */ if (current->uid) @@ -457,14 +458,25 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) old_cap = cap_set_effective(current->cap_permitted); } - res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd); + res = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path); if (res) goto out; - res = vfs_permission(&nd, mode); + inode = path.dentry->d_inode; + + if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) { + /* + * MAY_EXEC on regular files is denied if the fs is mounted + * with the "noexec" flag. + */ + res = -EACCES; + if (path.mnt->mnt_flags & MNT_NOEXEC) + goto out_path_release; + } + + res = inode_permission(inode, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ - if(res || !(mode & S_IWOTH) || - special_file(nd.path.dentry->d_inode->i_mode)) + if (res || !(mode & S_IWOTH) || special_file(inode->i_mode)) goto out_path_release; /* * This is a rare case where using __mnt_is_readonly() @@ -476,11 +488,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode) * inherently racy and know that the fs may change * state before we even see this result. */ - if (__mnt_is_readonly(nd.path.mnt)) + if (__mnt_is_readonly(path.mnt)) res = -EROFS; out_path_release: - path_put(&nd.path); + path_put(&path); out: current->fsuid = old_fsuid; current->fsgid = old_fsgid; @@ -498,22 +510,21 @@ asmlinkage long sys_access(const char __user *filename, int mode) asmlinkage long sys_chdir(const char __user * filename) { - struct nameidata nd; + struct path path; int error; - error = __user_walk(filename, - LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd); + error = user_path_dir(filename, &path); if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC); + error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; - set_fs_pwd(current->fs, &nd.path); + set_fs_pwd(current->fs, &path); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -535,7 +546,7 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!S_ISDIR(inode->i_mode)) goto out_putf; - error = file_permission(file, MAY_EXEC); + error = inode_permission(inode, MAY_EXEC | MAY_ACCESS); if (!error) set_fs_pwd(current->fs, &file->f_path); out_putf: @@ -546,14 +557,14 @@ out: asmlinkage long sys_chroot(const char __user * filename) { - struct nameidata nd; + struct path path; int error; - error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY | LOOKUP_NOALT, &nd); + error = user_path_dir(filename, &path); if (error) goto out; - error = vfs_permission(&nd, MAY_EXEC); + error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS); if (error) goto dput_and_out; @@ -561,11 +572,10 @@ asmlinkage long sys_chroot(const char __user * filename) if (!capable(CAP_SYS_CHROOT)) goto dput_and_out; - set_fs_root(current->fs, &nd.path); - set_fs_altroot(); + set_fs_root(current->fs, &path); error = 0; dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -590,9 +600,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) err = mnt_want_write(file->f_path.mnt); if (err) goto out_putf; - err = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_drop_write; mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; @@ -600,8 +607,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; err = notify_change(dentry, &newattrs); mutex_unlock(&inode->i_mutex); - -out_drop_write: mnt_drop_write(file->f_path.mnt); out_putf: fput(file); @@ -612,36 +617,29 @@ out: asmlinkage long sys_fchmodat(int dfd, const char __user *filename, mode_t mode) { - struct nameidata nd; - struct inode * inode; + struct path path; + struct inode *inode; int error; struct iattr newattrs; - error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd); + error = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path); if (error) goto out; - inode = nd.path.dentry->d_inode; + inode = path.dentry->d_inode; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto dput_and_out; - - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_drop_write; - mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; - error = notify_change(nd.path.dentry, &newattrs); + error = notify_change(path.dentry, &newattrs); mutex_unlock(&inode->i_mutex); - -out_drop_write: - mnt_drop_write(nd.path.mnt); + mnt_drop_write(path.mnt); dput_and_out: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -653,18 +651,10 @@ asmlinkage long sys_chmod(const char __user *filename, mode_t mode) static int chown_common(struct dentry * dentry, uid_t user, gid_t group) { - struct inode * inode; + struct inode *inode = dentry->d_inode; int error; struct iattr newattrs; - error = -ENOENT; - if (!(inode = dentry->d_inode)) { - printk(KERN_ERR "chown_common: NULL inode\n"); - goto out; - } - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out; newattrs.ia_valid = ATTR_CTIME; if (user != (uid_t) -1) { newattrs.ia_valid |= ATTR_UID; @@ -680,25 +670,25 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group) mutex_lock(&inode->i_mutex); error = notify_change(dentry, &newattrs); mutex_unlock(&inode->i_mutex); -out: + return error; } asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(filename, &nd); + error = user_path(filename, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -706,7 +696,7 @@ out: asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, gid_t group, int flag) { - struct nameidata nd; + struct path path; int error = -EINVAL; int follow; @@ -714,35 +704,35 @@ asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user, goto out; follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW; - error = __user_walk_fd(dfd, filename, follow, &nd); + error = user_path_at(dfd, filename, follow, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(filename, &nd); + error = user_lpath(filename, &path); if (error) goto out; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (error) goto out_release; - error = chown_common(nd.path.dentry, user, group); - mnt_drop_write(nd.path.mnt); + error = chown_common(path.dentry, user, group); + mnt_drop_write(path.mnt); out_release: - path_put(&nd.path); + path_put(&path); out: return error; } @@ -973,71 +963,6 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags) } EXPORT_SYMBOL(dentry_open); -/* - * Find an empty file descriptor entry, and mark it busy. - */ -int get_unused_fd_flags(int flags) -{ - struct files_struct * files = current->files; - int fd, error; - struct fdtable *fdt; - - error = -EMFILE; - spin_lock(&files->file_lock); - -repeat: - fdt = files_fdtable(files); - fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds, - files->next_fd); - - /* - * N.B. For clone tasks sharing a files structure, this test - * will limit the total number of files that can be opened. - */ - if (fd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur) - goto out; - - /* Do we need to expand the fd array or fd set? */ - error = expand_files(files, fd); - if (error < 0) - goto out; - - if (error) { - /* - * If we needed to expand the fs array we - * might have blocked - try again. - */ - error = -EMFILE; - goto repeat; - } - - FD_SET(fd, fdt->open_fds); - if (flags & O_CLOEXEC) - FD_SET(fd, fdt->close_on_exec); - else - FD_CLR(fd, fdt->close_on_exec); - files->next_fd = fd + 1; -#if 1 - /* Sanity check */ - if (fdt->fd[fd] != NULL) { - printk(KERN_WARNING "get_unused_fd: slot %d not NULL!\n", fd); - fdt->fd[fd] = NULL; - } -#endif - error = fd; - -out: - spin_unlock(&files->file_lock); - return error; -} - -int get_unused_fd(void) -{ - return get_unused_fd_flags(0); -} - -EXPORT_SYMBOL(get_unused_fd); - static void __put_unused_fd(struct files_struct *files, unsigned int fd) { struct fdtable *fdt = files_fdtable(files); diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index d17b4fd..9f5b054 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -430,7 +430,7 @@ static struct file_system_type openprom_fs_type = { .kill_sb = kill_anon_super, }; -static void op_inode_init_once(struct kmem_cache * cachep, void *data) +static void op_inode_init_once(void *data) { struct op_inode_info *oi = (struct op_inode_info *) data; diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 7d6b34e..7408227 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -120,22 +120,21 @@ static int (*check_part[])(struct parsed_partitions *, struct block_device *) = * a pointer to that same buffer (for convenience). */ -char *disk_name(struct gendisk *hd, int part, char *buf) +char *disk_name(struct gendisk *hd, int partno, char *buf) { - if (!part) + if (!partno) snprintf(buf, BDEVNAME_SIZE, "%s", hd->disk_name); else if (isdigit(hd->disk_name[strlen(hd->disk_name)-1])) - snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, part); + snprintf(buf, BDEVNAME_SIZE, "%sp%d", hd->disk_name, partno); else - snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, part); + snprintf(buf, BDEVNAME_SIZE, "%s%d", hd->disk_name, partno); return buf; } const char *bdevname(struct block_device *bdev, char *buf) { - int part = MINOR(bdev->bd_dev) - bdev->bd_disk->first_minor; - return disk_name(bdev->bd_disk, part, buf); + return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf); } EXPORT_SYMBOL(bdevname); @@ -169,7 +168,7 @@ check_partition(struct gendisk *hd, struct block_device *bdev) if (isdigit(state->name[strlen(state->name)-1])) sprintf(state->name, "p"); - state->limit = hd->minors; + state->limit = disk_max_parts(hd); i = res = err = 0; while (!res && check_part[i]) { memset(&state->parts, 0, sizeof(state->parts)); @@ -204,21 +203,22 @@ static ssize_t part_start_show(struct device *dev, return sprintf(buf, "%llu\n",(unsigned long long)p->start_sect); } -static ssize_t part_size_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_size_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects); } -static ssize_t part_stat_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_stat_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); + int cpu; - preempt_disable(); - part_round_stats(p); - preempt_enable(); + cpu = part_stat_lock(); + part_round_stats(cpu, p); + part_stat_unlock(); return sprintf(buf, "%8lu %8lu %8llu %8u " "%8lu %8lu %8llu %8u " @@ -238,17 +238,17 @@ static ssize_t part_stat_show(struct device *dev, } #ifdef CONFIG_FAIL_MAKE_REQUEST -static ssize_t part_fail_show(struct device *dev, - struct device_attribute *attr, char *buf) +ssize_t part_fail_show(struct device *dev, + struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); return sprintf(buf, "%d\n", p->make_it_fail); } -static ssize_t part_fail_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) +ssize_t part_fail_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) { struct hd_struct *p = dev_to_part(dev); int i; @@ -300,40 +300,34 @@ struct device_type part_type = { .release = part_release, }; -static inline void partition_sysfs_add_subdir(struct hd_struct *p) -{ - struct kobject *k; - - k = kobject_get(&p->dev.kobj); - p->holder_dir = kobject_create_and_add("holders", k); - kobject_put(k); -} - -static inline void disk_sysfs_add_subdirs(struct gendisk *disk) +static void delete_partition_rcu_cb(struct rcu_head *head) { - struct kobject *k; + struct hd_struct *part = container_of(head, struct hd_struct, rcu_head); - k = kobject_get(&disk->dev.kobj); - disk->holder_dir = kobject_create_and_add("holders", k); - disk->slave_dir = kobject_create_and_add("slaves", k); - kobject_put(k); + part->start_sect = 0; + part->nr_sects = 0; + part_stat_set_all(part, 0); + put_device(part_to_dev(part)); } -void delete_partition(struct gendisk *disk, int part) +void delete_partition(struct gendisk *disk, int partno) { - struct hd_struct *p = disk->part[part-1]; + struct disk_part_tbl *ptbl = disk->part_tbl; + struct hd_struct *part; - if (!p) + if (partno >= ptbl->len) return; - if (!p->nr_sects) + + part = ptbl->part[partno]; + if (!part) return; - disk->part[part-1] = NULL; - p->start_sect = 0; - p->nr_sects = 0; - part_stat_set_all(p, 0); - kobject_put(p->holder_dir); - device_del(&p->dev); - put_device(&p->dev); + + blk_free_devt(part_devt(part)); + rcu_assign_pointer(ptbl->part[partno], NULL); + kobject_put(part->holder_dir); + device_del(part_to_dev(part)); + + call_rcu(&part->rcu_head, delete_partition_rcu_cb); } static ssize_t whole_disk_show(struct device *dev, @@ -344,102 +338,132 @@ static ssize_t whole_disk_show(struct device *dev, static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH, whole_disk_show, NULL); -int add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags) +int add_partition(struct gendisk *disk, int partno, + sector_t start, sector_t len, int flags) { struct hd_struct *p; + dev_t devt = MKDEV(0, 0); + struct device *ddev = disk_to_dev(disk); + struct device *pdev; + struct disk_part_tbl *ptbl; + const char *dname; int err; + err = disk_expand_part_tbl(disk, partno); + if (err) + return err; + ptbl = disk->part_tbl; + + if (ptbl->part[partno]) + return -EBUSY; + p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) return -ENOMEM; if (!init_part_stats(p)) { err = -ENOMEM; - goto out0; + goto out_free; } + pdev = part_to_dev(p); + p->start_sect = start; p->nr_sects = len; - p->partno = part; - p->policy = disk->policy; + p->partno = partno; + p->policy = get_disk_ro(disk); - if (isdigit(disk->dev.bus_id[strlen(disk->dev.bus_id)-1])) - snprintf(p->dev.bus_id, BUS_ID_SIZE, - "%sp%d", disk->dev.bus_id, part); + dname = dev_name(ddev); + if (isdigit(dname[strlen(dname) - 1])) + snprintf(pdev->bus_id, BUS_ID_SIZE, "%sp%d", dname, partno); else - snprintf(p->dev.bus_id, BUS_ID_SIZE, - "%s%d", disk->dev.bus_id, part); + snprintf(pdev->bus_id, BUS_ID_SIZE, "%s%d", dname, partno); + + device_initialize(pdev); + pdev->class = &block_class; + pdev->type = &part_type; + pdev->parent = ddev; - device_initialize(&p->dev); - p->dev.devt = MKDEV(disk->major, disk->first_minor + part); - p->dev.class = &block_class; - p->dev.type = &part_type; - p->dev.parent = &disk->dev; - disk->part[part-1] = p; + err = blk_alloc_devt(p, &devt); + if (err) + goto out_free; + pdev->devt = devt; /* delay uevent until 'holders' subdir is created */ - p->dev.uevent_suppress = 1; - err = device_add(&p->dev); + pdev->uevent_suppress = 1; + err = device_add(pdev); if (err) - goto out1; - partition_sysfs_add_subdir(p); - p->dev.uevent_suppress = 0; + goto out_put; + + err = -ENOMEM; + p->holder_dir = kobject_create_and_add("holders", &pdev->kobj); + if (!p->holder_dir) + goto out_del; + + pdev->uevent_suppress = 0; if (flags & ADDPART_FLAG_WHOLEDISK) { - err = device_create_file(&p->dev, &dev_attr_whole_disk); + err = device_create_file(pdev, &dev_attr_whole_disk); if (err) - goto out2; + goto out_del; } + /* everything is up and running, commence */ + INIT_RCU_HEAD(&p->rcu_head); + rcu_assign_pointer(ptbl->part[partno], p); + /* suppress uevent if the disk supresses it */ - if (!disk->dev.uevent_suppress) - kobject_uevent(&p->dev.kobj, KOBJ_ADD); + if (!ddev->uevent_suppress) + kobject_uevent(&pdev->kobj, KOBJ_ADD); return 0; -out2: - device_del(&p->dev); -out1: - put_device(&p->dev); - free_part_stats(p); -out0: +out_free: kfree(p); return err; +out_del: + kobject_put(p->holder_dir); + device_del(pdev); +out_put: + put_device(pdev); + blk_free_devt(devt); + return err; } /* Not exported, helper to add_disk(). */ void register_disk(struct gendisk *disk) { + struct device *ddev = disk_to_dev(disk); struct block_device *bdev; + struct disk_part_iter piter; + struct hd_struct *part; char *s; - int i; - struct hd_struct *p; int err; - disk->dev.parent = disk->driverfs_dev; - disk->dev.devt = MKDEV(disk->major, disk->first_minor); + ddev->parent = disk->driverfs_dev; - strlcpy(disk->dev.bus_id, disk->disk_name, BUS_ID_SIZE); + strlcpy(ddev->bus_id, disk->disk_name, BUS_ID_SIZE); /* ewww... some of these buggers have / in the name... */ - s = strchr(disk->dev.bus_id, '/'); + s = strchr(ddev->bus_id, '/'); if (s) *s = '!'; /* delay uevents, until we scanned partition table */ - disk->dev.uevent_suppress = 1; + ddev->uevent_suppress = 1; - if (device_add(&disk->dev)) + if (device_add(ddev)) return; #ifndef CONFIG_SYSFS_DEPRECATED - err = sysfs_create_link(block_depr, &disk->dev.kobj, - kobject_name(&disk->dev.kobj)); + err = sysfs_create_link(block_depr, &ddev->kobj, + kobject_name(&ddev->kobj)); if (err) { - device_del(&disk->dev); + device_del(ddev); return; } #endif - disk_sysfs_add_subdirs(disk); + disk->part0.holder_dir = kobject_create_and_add("holders", &ddev->kobj); + disk->slave_dir = kobject_create_and_add("slaves", &ddev->kobj); /* No minors to use for partitions */ - if (disk->minors == 1) + if (!disk_partitionable(disk)) goto exit; /* No such device (e.g., media were just removed) */ @@ -458,50 +482,66 @@ void register_disk(struct gendisk *disk) exit: /* announce disk after possible partitions are created */ - disk->dev.uevent_suppress = 0; - kobject_uevent(&disk->dev.kobj, KOBJ_ADD); + ddev->uevent_suppress = 0; + kobject_uevent(&ddev->kobj, KOBJ_ADD); /* announce possible partitions */ - for (i = 1; i < disk->minors; i++) { - p = disk->part[i-1]; - if (!p || !p->nr_sects) - continue; - kobject_uevent(&p->dev.kobj, KOBJ_ADD); - } + disk_part_iter_init(&piter, disk, 0); + while ((part = disk_part_iter_next(&piter))) + kobject_uevent(&part_to_dev(part)->kobj, KOBJ_ADD); + disk_part_iter_exit(&piter); } int rescan_partitions(struct gendisk *disk, struct block_device *bdev) { + struct disk_part_iter piter; + struct hd_struct *part; struct parsed_partitions *state; - int p, res; + int p, highest, res; if (bdev->bd_part_count) return -EBUSY; res = invalidate_partition(disk, 0); if (res) return res; - bdev->bd_invalidated = 0; - for (p = 1; p < disk->minors; p++) - delete_partition(disk, p); + + disk_part_iter_init(&piter, disk, DISK_PITER_INCL_EMPTY); + while ((part = disk_part_iter_next(&piter))) + delete_partition(disk, part->partno); + disk_part_iter_exit(&piter); + if (disk->fops->revalidate_disk) disk->fops->revalidate_disk(disk); + check_disk_size_change(disk, bdev); + bdev->bd_invalidated = 0; if (!get_capacity(disk) || !(state = check_partition(disk, bdev))) return 0; if (IS_ERR(state)) /* I/O error reading the partition table */ return -EIO; /* tell userspace that the media / partition table may have changed */ - kobject_uevent(&disk->dev.kobj, KOBJ_CHANGE); + kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE); + /* Detect the highest partition number and preallocate + * disk->part_tbl. This is an optimization and not strictly + * necessary. + */ + for (p = 1, highest = 0; p < state->limit; p++) + if (state->parts[p].size) + highest = p; + + disk_expand_part_tbl(disk, highest); + + /* add partitions */ for (p = 1; p < state->limit; p++) { sector_t size = state->parts[p].size; sector_t from = state->parts[p].from; if (!size) continue; if (from + size > get_capacity(disk)) { - printk(KERN_ERR " %s: p%d exceeds device capacity\n", + printk(KERN_WARNING + "%s: p%d exceeds device capacity\n", disk->disk_name, p); - continue; } res = add_partition(disk, p, from, size, state->parts[p].flags); if (res) { @@ -541,25 +581,31 @@ EXPORT_SYMBOL(read_dev_sector); void del_gendisk(struct gendisk *disk) { - int p; + struct disk_part_iter piter; + struct hd_struct *part; /* invalidate stuff */ - for (p = disk->minors - 1; p > 0; p--) { - invalidate_partition(disk, p); - delete_partition(disk, p); + disk_part_iter_init(&piter, disk, + DISK_PITER_INCL_EMPTY | DISK_PITER_REVERSE); + while ((part = disk_part_iter_next(&piter))) { + invalidate_partition(disk, part->partno); + delete_partition(disk, part->partno); } + disk_part_iter_exit(&piter); + invalidate_partition(disk, 0); - disk->capacity = 0; + blk_free_devt(disk_to_dev(disk)->devt); + set_capacity(disk, 0); disk->flags &= ~GENHD_FL_UP; unlink_gendisk(disk); - disk_stat_set_all(disk, 0); - disk->stamp = 0; + part_stat_set_all(&disk->part0, 0); + disk->part0.stamp = 0; - kobject_put(disk->holder_dir); + kobject_put(disk->part0.holder_dir); kobject_put(disk->slave_dir); disk->driverfs_dev = NULL; #ifndef CONFIG_SYSFS_DEPRECATED - sysfs_remove_link(block_depr, disk->dev.bus_id); + sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk))); #endif - device_del(&disk->dev); + device_del(disk_to_dev(disk)); } diff --git a/fs/partitions/check.h b/fs/partitions/check.h index 17ae8ec..98dbe1a 100644 --- a/fs/partitions/check.h +++ b/fs/partitions/check.h @@ -5,15 +5,13 @@ * add_gd_partition adds a partitions details to the devices partition * description. */ -enum { MAX_PART = 256 }; - struct parsed_partitions { char name[BDEVNAME_SIZE]; struct { sector_t from; sector_t size; int flags; - } parts[MAX_PART]; + } parts[DISK_MAX_PARTS]; int next; int limit; }; @@ -777,45 +777,10 @@ pipe_rdwr_open(struct inode *inode, struct file *filp) /* * The file_operations structs are not static because they * are also used in linux/fs/fifo.c to do operations on FIFOs. + * + * Pipes reuse fifos' file_operations structs. */ -const struct file_operations read_fifo_fops = { - .llseek = no_llseek, - .read = do_sync_read, - .aio_read = pipe_read, - .write = bad_pipe_w, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_read_open, - .release = pipe_read_release, - .fasync = pipe_read_fasync, -}; - -const struct file_operations write_fifo_fops = { - .llseek = no_llseek, - .read = bad_pipe_r, - .write = do_sync_write, - .aio_write = pipe_write, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_write_open, - .release = pipe_write_release, - .fasync = pipe_write_fasync, -}; - -const struct file_operations rdwr_fifo_fops = { - .llseek = no_llseek, - .read = do_sync_read, - .aio_read = pipe_read, - .write = do_sync_write, - .aio_write = pipe_write, - .poll = pipe_poll, - .unlocked_ioctl = pipe_ioctl, - .open = pipe_rdwr_open, - .release = pipe_rdwr_release, - .fasync = pipe_rdwr_fasync, -}; - -static const struct file_operations read_pipe_fops = { +const struct file_operations read_pipefifo_fops = { .llseek = no_llseek, .read = do_sync_read, .aio_read = pipe_read, @@ -827,7 +792,7 @@ static const struct file_operations read_pipe_fops = { .fasync = pipe_read_fasync, }; -static const struct file_operations write_pipe_fops = { +const struct file_operations write_pipefifo_fops = { .llseek = no_llseek, .read = bad_pipe_r, .write = do_sync_write, @@ -839,7 +804,7 @@ static const struct file_operations write_pipe_fops = { .fasync = pipe_write_fasync, }; -static const struct file_operations rdwr_pipe_fops = { +const struct file_operations rdwr_pipefifo_fops = { .llseek = no_llseek, .read = do_sync_read, .aio_read = pipe_read, @@ -927,7 +892,7 @@ static struct inode * get_pipe_inode(void) inode->i_pipe = pipe; pipe->readers = pipe->writers = 1; - inode->i_fop = &rdwr_pipe_fops; + inode->i_fop = &rdwr_pipefifo_fops; /* * Mark the inode dirty from the very beginning, @@ -978,7 +943,7 @@ struct file *create_write_pipe(int flags) d_instantiate(dentry, inode); err = -ENFILE; - f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipe_fops); + f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipefifo_fops); if (!f) goto err_dentry; f->f_mapping = inode->i_mapping; @@ -1020,7 +985,7 @@ struct file *create_read_pipe(struct file *wrf, int flags) f->f_pos = 0; f->f_flags = O_RDONLY | (flags & O_NONBLOCK); - f->f_op = &read_pipe_fops; + f->f_op = &read_pipefifo_fops; f->f_mode = FMODE_READ; f->f_version = 0; diff --git a/fs/proc/array.c b/fs/proc/array.c index 797d775..71c9be5 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -80,6 +80,7 @@ #include <linux/delayacct.h> #include <linux/seq_file.h> #include <linux/pid_namespace.h> +#include <linux/tracehook.h> #include <asm/pgtable.h> #include <asm/processor.h> @@ -168,8 +169,12 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns, rcu_read_lock(); ppid = pid_alive(p) ? task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0; - tpid = pid_alive(p) && p->ptrace ? - task_pid_nr_ns(rcu_dereference(p->parent), ns) : 0; + tpid = 0; + if (pid_alive(p)) { + struct task_struct *tracer = tracehook_tracer_task(p); + if (tracer) + tpid = task_pid_nr_ns(tracer, ns); + } seq_printf(m, "State:\t%s\n" "Tgid:\t%d\n" @@ -332,65 +337,6 @@ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns, return 0; } -/* - * Use precise platform statistics if available: - */ -#ifdef CONFIG_VIRT_CPU_ACCOUNTING -static cputime_t task_utime(struct task_struct *p) -{ - return p->utime; -} - -static cputime_t task_stime(struct task_struct *p) -{ - return p->stime; -} -#else -static cputime_t task_utime(struct task_struct *p) -{ - clock_t utime = cputime_to_clock_t(p->utime), - total = utime + cputime_to_clock_t(p->stime); - u64 temp; - - /* - * Use CFS's precise accounting: - */ - temp = (u64)nsec_to_clock_t(p->se.sum_exec_runtime); - - if (total) { - temp *= utime; - do_div(temp, total); - } - utime = (clock_t)temp; - - p->prev_utime = max(p->prev_utime, clock_t_to_cputime(utime)); - return p->prev_utime; -} - -static cputime_t task_stime(struct task_struct *p) -{ - clock_t stime; - - /* - * Use CFS's precise accounting. (we subtract utime from - * the total, to make sure the total observed by userspace - * grows monotonically - apps rely on that): - */ - stime = nsec_to_clock_t(p->se.sum_exec_runtime) - - cputime_to_clock_t(task_utime(p)); - - if (stime >= 0) - p->prev_stime = max(p->prev_stime, clock_t_to_cputime(stime)); - - return p->prev_stime; -} -#endif - -static cputime_t task_gtime(struct task_struct *p) -{ - return p->gtime; -} - static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task, int whole) { diff --git a/fs/proc/base.c b/fs/proc/base.c index a891fe4..a28840b 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -53,6 +53,7 @@ #include <linux/time.h> #include <linux/proc_fs.h> #include <linux/stat.h> +#include <linux/task_io_accounting_ops.h> #include <linux/init.h> #include <linux/capability.h> #include <linux/file.h> @@ -69,6 +70,7 @@ #include <linux/mount.h> #include <linux/security.h> #include <linux/ptrace.h> +#include <linux/tracehook.h> #include <linux/cgroup.h> #include <linux/cpuset.h> #include <linux/audit.h> @@ -231,10 +233,14 @@ static int check_mem_permission(struct task_struct *task) * If current is actively ptrace'ing, and would also be * permitted to freshly attach with ptrace now, permit it. */ - if (task->parent == current && (task->ptrace & PT_PTRACED) && - task_is_stopped_or_traced(task) && - ptrace_may_access(task, PTRACE_MODE_ATTACH)) - return 0; + if (task_is_stopped_or_traced(task)) { + int match; + rcu_read_lock(); + match = (tracehook_tracer_task(task) == current); + rcu_read_unlock(); + if (match && ptrace_may_access(task, PTRACE_MODE_ATTACH)) + return 0; + } /* * Noone else is allowed. @@ -504,6 +510,26 @@ static int proc_pid_limits(struct task_struct *task, char *buffer) return count; } +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK +static int proc_pid_syscall(struct task_struct *task, char *buffer) +{ + long nr; + unsigned long args[6], sp, pc; + + if (task_current_syscall(task, &nr, args, 6, &sp, &pc)) + return sprintf(buffer, "running\n"); + + if (nr < 0) + return sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc); + + return sprintf(buffer, + "%ld 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n", + nr, + args[0], args[1], args[2], args[3], args[4], args[5], + sp, pc); +} +#endif /* CONFIG_HAVE_ARCH_TRACEHOOK */ + /************************************************************************/ /* Here the fs part begins */ /************************************************************************/ @@ -1834,8 +1860,7 @@ static const struct file_operations proc_fd_operations = { * /proc/pid/fd needs a special permission handler so that a process can still * access /proc/self/fd after it has executed a setuid(). */ -static int proc_fd_permission(struct inode *inode, int mask, - struct nameidata *nd) +static int proc_fd_permission(struct inode *inode, int mask) { int rv; @@ -2378,53 +2403,18 @@ static int proc_base_fill_cache(struct file *filp, void *dirent, #ifdef CONFIG_TASK_IO_ACCOUNTING static int do_io_accounting(struct task_struct *task, char *buffer, int whole) { - u64 rchar, wchar, syscr, syscw; - struct task_io_accounting ioac; - - if (!whole) { - rchar = task->rchar; - wchar = task->wchar; - syscr = task->syscr; - syscw = task->syscw; - memcpy(&ioac, &task->ioac, sizeof(ioac)); - } else { - unsigned long flags; - struct task_struct *t = task; - rchar = wchar = syscr = syscw = 0; - memset(&ioac, 0, sizeof(ioac)); - - rcu_read_lock(); - do { - rchar += t->rchar; - wchar += t->wchar; - syscr += t->syscr; - syscw += t->syscw; - - ioac.read_bytes += t->ioac.read_bytes; - ioac.write_bytes += t->ioac.write_bytes; - ioac.cancelled_write_bytes += - t->ioac.cancelled_write_bytes; - t = next_thread(t); - } while (t != task); - rcu_read_unlock(); - - if (lock_task_sighand(task, &flags)) { - struct signal_struct *sig = task->signal; + struct task_io_accounting acct = task->ioac; + unsigned long flags; - rchar += sig->rchar; - wchar += sig->wchar; - syscr += sig->syscr; - syscw += sig->syscw; + if (whole && lock_task_sighand(task, &flags)) { + struct task_struct *t = task; - ioac.read_bytes += sig->ioac.read_bytes; - ioac.write_bytes += sig->ioac.write_bytes; - ioac.cancelled_write_bytes += - sig->ioac.cancelled_write_bytes; + task_io_accounting_add(&acct, &task->signal->ioac); + while_each_thread(task, t) + task_io_accounting_add(&acct, &t->ioac); - unlock_task_sighand(task, &flags); - } + unlock_task_sighand(task, &flags); } - return sprintf(buffer, "rchar: %llu\n" "wchar: %llu\n" @@ -2433,13 +2423,13 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) "read_bytes: %llu\n" "write_bytes: %llu\n" "cancelled_write_bytes: %llu\n", - (unsigned long long)rchar, - (unsigned long long)wchar, - (unsigned long long)syscr, - (unsigned long long)syscw, - (unsigned long long)ioac.read_bytes, - (unsigned long long)ioac.write_bytes, - (unsigned long long)ioac.cancelled_write_bytes); + (unsigned long long)acct.rchar, + (unsigned long long)acct.wchar, + (unsigned long long)acct.syscr, + (unsigned long long)acct.syscw, + (unsigned long long)acct.read_bytes, + (unsigned long long)acct.write_bytes, + (unsigned long long)acct.cancelled_write_bytes); } static int proc_tid_io_accounting(struct task_struct *task, char *buffer) @@ -2473,6 +2463,9 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), #endif +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUSR, pid_syscall), +#endif INF("cmdline", S_IRUGO, pid_cmdline), ONE("stat", S_IRUGO, tgid_stat), ONE("statm", S_IRUGO, pid_statm), @@ -2805,6 +2798,9 @@ static const struct pid_entry tid_base_stuff[] = { #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, pid_sched), #endif +#ifdef CONFIG_HAVE_ARCH_TRACEHOOK + INF("syscall", S_IRUSR, pid_syscall), +#endif INF("cmdline", S_IRUGO, pid_cmdline), ONE("stat", S_IRUGO, tid_stat), ONE("statm", S_IRUGO, pid_statm), diff --git a/fs/proc/generic.c b/fs/proc/generic.c index bc0a0dd..7821589 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -300,10 +300,10 @@ out: return rtn; } -static DEFINE_IDR(proc_inum_idr); +static DEFINE_IDA(proc_inum_ida); static DEFINE_SPINLOCK(proc_inum_lock); /* protects the above */ -#define PROC_DYNAMIC_FIRST 0xF0000000UL +#define PROC_DYNAMIC_FIRST 0xF0000000U /* * Return an inode number between PROC_DYNAMIC_FIRST and @@ -311,36 +311,34 @@ static DEFINE_SPINLOCK(proc_inum_lock); /* protects the above */ */ static unsigned int get_inode_number(void) { - int i, inum = 0; + unsigned int i; int error; retry: - if (idr_pre_get(&proc_inum_idr, GFP_KERNEL) == 0) + if (ida_pre_get(&proc_inum_ida, GFP_KERNEL) == 0) return 0; spin_lock(&proc_inum_lock); - error = idr_get_new(&proc_inum_idr, NULL, &i); + error = ida_get_new(&proc_inum_ida, &i); spin_unlock(&proc_inum_lock); if (error == -EAGAIN) goto retry; else if (error) return 0; - inum = (i & MAX_ID_MASK) + PROC_DYNAMIC_FIRST; - - /* inum will never be more than 0xf0ffffff, so no check - * for overflow. - */ - - return inum; + if (i > UINT_MAX - PROC_DYNAMIC_FIRST) { + spin_lock(&proc_inum_lock); + ida_remove(&proc_inum_ida, i); + spin_unlock(&proc_inum_lock); + return 0; + } + return PROC_DYNAMIC_FIRST + i; } static void release_inode_number(unsigned int inum) { - int id = (inum - PROC_DYNAMIC_FIRST) | ~MAX_ID_MASK; - spin_lock(&proc_inum_lock); - idr_remove(&proc_inum_idr, id); + ida_remove(&proc_inum_ida, inum - PROC_DYNAMIC_FIRST); spin_unlock(&proc_inum_lock); } @@ -549,8 +547,8 @@ static int proc_register(struct proc_dir_entry * dir, struct proc_dir_entry * dp for (tmp = dir->subdir; tmp; tmp = tmp->next) if (strcmp(tmp->name, dp->name) == 0) { - printk(KERN_WARNING "proc_dir_entry '%s' already " - "registered\n", dp->name); + printk(KERN_WARNING "proc_dir_entry '%s/%s' already registered\n", + dir->name, dp->name); dump_stack(); break; } @@ -806,12 +804,9 @@ continue_removing: if (S_ISDIR(de->mode)) parent->nlink--; de->nlink = 0; - if (de->subdir) { - printk(KERN_WARNING "%s: removing non-empty directory " + WARN(de->subdir, KERN_WARNING "%s: removing non-empty directory " "'%s/%s', leaking at least '%s'\n", __func__, de->parent->name, de->name, de->subdir->name); - WARN_ON(1); - } if (atomic_dec_and_test(&de->count)) free_proc_entry(de); } diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 02eca2e..8bb03f0 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -17,6 +17,7 @@ #include <linux/init.h> #include <linux/module.h> #include <linux/smp_lock.h> +#include <linux/sysctl.h> #include <asm/system.h> #include <asm/uaccess.h> @@ -65,6 +66,8 @@ static void proc_delete_inode(struct inode *inode) module_put(de->owner); de_put(de); } + if (PROC_I(inode)->sysctl) + sysctl_head_put(PROC_I(inode)->sysctl); clear_inode(inode); } @@ -84,6 +87,8 @@ static struct inode *proc_alloc_inode(struct super_block *sb) ei->fd = 0; ei->op.proc_get_link = NULL; ei->pde = NULL; + ei->sysctl = NULL; + ei->sysctl_entry = NULL; inode = &ei->vfs_inode; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; return inode; @@ -94,7 +99,7 @@ static void proc_destroy_inode(struct inode *inode) kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c index 79ecd28..3f87d26 100644 --- a/fs/proc/nommu.c +++ b/fs/proc/nommu.c @@ -52,14 +52,14 @@ int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma) } seq_printf(m, - "%08lx-%08lx %c%c%c%c %08lx %02x:%02x %lu %n", + "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", vma->vm_start, vma->vm_end, flags & VM_READ ? 'r' : '-', flags & VM_WRITE ? 'w' : '-', flags & VM_EXEC ? 'x' : '-', flags & VM_MAYSHARE ? flags & VM_SHARED ? 'S' : 's' : 'p', - vma->vm_pgoff << PAGE_SHIFT, + ((loff_t)vma->vm_pgoff) << PAGE_SHIFT, MAJOR(dev), MINOR(dev), ino, &len); if (file) { diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index ded9698..29e20c6 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -24,6 +24,7 @@ #include <linux/tty.h> #include <linux/string.h> #include <linux/mman.h> +#include <linux/quicklist.h> #include <linux/proc_fs.h> #include <linux/ioport.h> #include <linux/mm.h> @@ -182,6 +183,9 @@ static int meminfo_read_proc(char *page, char **start, off_t off, "SReclaimable: %8lu kB\n" "SUnreclaim: %8lu kB\n" "PageTables: %8lu kB\n" +#ifdef CONFIG_QUICKLIST + "Quicklists: %8lu kB\n" +#endif "NFS_Unstable: %8lu kB\n" "Bounce: %8lu kB\n" "WritebackTmp: %8lu kB\n" @@ -214,6 +218,9 @@ static int meminfo_read_proc(char *page, char **start, off_t off, K(global_page_state(NR_SLAB_RECLAIMABLE)), K(global_page_state(NR_SLAB_UNRECLAIMABLE)), K(global_page_state(NR_PAGETABLE)), +#ifdef CONFIG_QUICKLIST + K(quicklist_total_size()), +#endif K(global_page_state(NR_UNSTABLE_NFS)), K(global_page_state(NR_BOUNCE)), K(global_page_state(NR_WRITEBACK_TEMP)), diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 5acc001..f9a8b89 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -10,149 +10,110 @@ static struct dentry_operations proc_sys_dentry_operations; static const struct file_operations proc_sys_file_operations; static const struct inode_operations proc_sys_inode_operations; +static const struct file_operations proc_sys_dir_file_operations; +static const struct inode_operations proc_sys_dir_operations; -static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table) -{ - /* Refresh the cached information bits in the inode */ - if (table) { - inode->i_uid = 0; - inode->i_gid = 0; - inode->i_mode = table->mode; - if (table->proc_handler) { - inode->i_mode |= S_IFREG; - inode->i_nlink = 1; - } else { - inode->i_mode |= S_IFDIR; - inode->i_nlink = 0; /* It is too hard to figure out */ - } - } -} - -static struct inode *proc_sys_make_inode(struct inode *dir, struct ctl_table *table) +static struct inode *proc_sys_make_inode(struct super_block *sb, + struct ctl_table_header *head, struct ctl_table *table) { struct inode *inode; - struct proc_inode *dir_ei, *ei; - int depth; + struct proc_inode *ei; - inode = new_inode(dir->i_sb); + inode = new_inode(sb); if (!inode) goto out; - /* A directory is always one deeper than it's parent */ - dir_ei = PROC_I(dir); - depth = dir_ei->fd + 1; - + sysctl_head_get(head); ei = PROC_I(inode); - ei->fd = depth; + ei->sysctl = head; + ei->sysctl_entry = table; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; - inode->i_op = &proc_sys_inode_operations; - inode->i_fop = &proc_sys_file_operations; inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */ - proc_sys_refresh_inode(inode, table); + inode->i_mode = table->mode; + if (!table->child) { + inode->i_mode |= S_IFREG; + inode->i_op = &proc_sys_inode_operations; + inode->i_fop = &proc_sys_file_operations; + } else { + inode->i_mode |= S_IFDIR; + inode->i_nlink = 0; + inode->i_op = &proc_sys_dir_operations; + inode->i_fop = &proc_sys_dir_file_operations; + } out: return inode; } -static struct dentry *proc_sys_ancestor(struct dentry *dentry, int depth) -{ - for (;;) { - struct proc_inode *ei; - - ei = PROC_I(dentry->d_inode); - if (ei->fd == depth) - break; /* found */ - - dentry = dentry->d_parent; - } - return dentry; -} - -static struct ctl_table *proc_sys_lookup_table_one(struct ctl_table *table, - struct qstr *name) +static struct ctl_table *find_in_table(struct ctl_table *p, struct qstr *name) { int len; - for ( ; table->ctl_name || table->procname; table++) { + for ( ; p->ctl_name || p->procname; p++) { - if (!table->procname) + if (!p->procname) continue; - len = strlen(table->procname); + len = strlen(p->procname); if (len != name->len) continue; - if (memcmp(table->procname, name->name, len) != 0) + if (memcmp(p->procname, name->name, len) != 0) continue; /* I have a match */ - return table; + return p; } return NULL; } -static struct ctl_table *proc_sys_lookup_table(struct dentry *dentry, - struct ctl_table *table) +struct ctl_table_header *grab_header(struct inode *inode) { - struct dentry *ancestor; - struct proc_inode *ei; - int depth, i; + if (PROC_I(inode)->sysctl) + return sysctl_head_grab(PROC_I(inode)->sysctl); + else + return sysctl_head_next(NULL); +} - ei = PROC_I(dentry->d_inode); - depth = ei->fd; +static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct ctl_table_header *head = grab_header(dir); + struct ctl_table *table = PROC_I(dir)->sysctl_entry; + struct ctl_table_header *h = NULL; + struct qstr *name = &dentry->d_name; + struct ctl_table *p; + struct inode *inode; + struct dentry *err = ERR_PTR(-ENOENT); - if (depth == 0) - return table; + if (IS_ERR(head)) + return ERR_CAST(head); - for (i = 1; table && (i <= depth); i++) { - ancestor = proc_sys_ancestor(dentry, i); - table = proc_sys_lookup_table_one(table, &ancestor->d_name); - if (table) - table = table->child; + if (table && !table->child) { + WARN_ON(1); + goto out; } - return table; - -} -static struct ctl_table *proc_sys_lookup_entry(struct dentry *dparent, - struct qstr *name, - struct ctl_table *table) -{ - table = proc_sys_lookup_table(dparent, table); - if (table) - table = proc_sys_lookup_table_one(table, name); - return table; -} -static struct ctl_table *do_proc_sys_lookup(struct dentry *parent, - struct qstr *name, - struct ctl_table_header **ptr) -{ - struct ctl_table_header *head; - struct ctl_table *table = NULL; + table = table ? table->child : head->ctl_table; - for (head = sysctl_head_next(NULL); head; - head = sysctl_head_next(head)) { - table = proc_sys_lookup_entry(parent, name, head->ctl_table); - if (table) - break; + p = find_in_table(table, name); + if (!p) { + for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) { + if (h->attached_to != table) + continue; + p = find_in_table(h->attached_by, name); + if (p) + break; + } } - *ptr = head; - return table; -} - -static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) -{ - struct ctl_table_header *head; - struct inode *inode; - struct dentry *err; - struct ctl_table *table; - err = ERR_PTR(-ENOENT); - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - if (!table) + if (!p) goto out; err = ERR_PTR(-ENOMEM); - inode = proc_sys_make_inode(dir, table); + inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p); + if (h) + sysctl_head_finish(h); + if (!inode) goto out; @@ -168,22 +129,14 @@ out: static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf, size_t count, loff_t *ppos, int write) { - struct dentry *dentry = filp->f_dentry; - struct ctl_table_header *head; - struct ctl_table *table; + struct inode *inode = filp->f_path.dentry->d_inode; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; ssize_t error; size_t res; - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - /* Has the sysctl entry disappeared on us? */ - error = -ENOENT; - if (!table) - goto out; - - /* Has the sysctl entry been replaced by a directory? */ - error = -EISDIR; - if (!table->proc_handler) - goto out; + if (IS_ERR(head)) + return PTR_ERR(head); /* * At this point we know that the sysctl was not unregistered @@ -193,6 +146,11 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf, if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ)) goto out; + /* if that can happen at all, it should be -EINVAL, not -EISDIR */ + error = -EINVAL; + if (!table->proc_handler) + goto out; + /* careful: calling conventions are nasty here */ res = count; error = table->proc_handler(table, write, filp, buf, &res, ppos); @@ -218,82 +176,86 @@ static ssize_t proc_sys_write(struct file *filp, const char __user *buf, static int proc_sys_fill_cache(struct file *filp, void *dirent, - filldir_t filldir, struct ctl_table *table) + filldir_t filldir, + struct ctl_table_header *head, + struct ctl_table *table) { - struct ctl_table_header *head; - struct ctl_table *child_table = NULL; struct dentry *child, *dir = filp->f_path.dentry; struct inode *inode; struct qstr qname; ino_t ino = 0; unsigned type = DT_UNKNOWN; - int ret; qname.name = table->procname; qname.len = strlen(table->procname); qname.hash = full_name_hash(qname.name, qname.len); - /* Suppress duplicates. - * Only fill a directory entry if it is the value that - * an ordinary lookup of that name returns. Hide all - * others. - * - * If we ever cache this translation in the dcache - * I should do a dcache lookup first. But for now - * it is just simpler not to. - */ - ret = 0; - child_table = do_proc_sys_lookup(dir, &qname, &head); - sysctl_head_finish(head); - if (child_table != table) - return 0; - child = d_lookup(dir, &qname); if (!child) { - struct dentry *new; - new = d_alloc(dir, &qname); - if (new) { - inode = proc_sys_make_inode(dir->d_inode, table); - if (!inode) - child = ERR_PTR(-ENOMEM); - else { - new->d_op = &proc_sys_dentry_operations; - d_add(new, inode); + child = d_alloc(dir, &qname); + if (child) { + inode = proc_sys_make_inode(dir->d_sb, head, table); + if (!inode) { + dput(child); + return -ENOMEM; + } else { + child->d_op = &proc_sys_dentry_operations; + d_add(child, inode); } - if (child) - dput(new); - else - child = new; + } else { + return -ENOMEM; } } - if (!child || IS_ERR(child) || !child->d_inode) - goto end_instantiate; inode = child->d_inode; - if (inode) { - ino = inode->i_ino; - type = inode->i_mode >> 12; - } + ino = inode->i_ino; + type = inode->i_mode >> 12; dput(child); -end_instantiate: - if (!ino) - ino= find_inode_number(dir, &qname); - if (!ino) - ino = 1; - return filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type); + return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type); +} + +static int scan(struct ctl_table_header *head, ctl_table *table, + unsigned long *pos, struct file *file, + void *dirent, filldir_t filldir) +{ + + for (; table->ctl_name || table->procname; table++, (*pos)++) { + int res; + + /* Can't do anything without a proc name */ + if (!table->procname) + continue; + + if (*pos < file->f_pos) + continue; + + res = proc_sys_fill_cache(file, dirent, filldir, head, table); + if (res) + return res; + + file->f_pos = *pos + 1; + } + return 0; } static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir) { - struct dentry *dentry = filp->f_dentry; + struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; - struct ctl_table_header *head = NULL; - struct ctl_table *table; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + struct ctl_table_header *h = NULL; unsigned long pos; - int ret; + int ret = -EINVAL; + + if (IS_ERR(head)) + return PTR_ERR(head); - ret = -ENOTDIR; - if (!S_ISDIR(inode->i_mode)) + if (table && !table->child) { + WARN_ON(1); goto out; + } + + table = table ? table->child : head->ctl_table; ret = 0; /* Avoid a switch here: arm builds fail with missing __cmpdi2 */ @@ -311,30 +273,17 @@ static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir) } pos = 2; - /* - Find each instance of the directory - * - Read all entries in each instance - * - Before returning an entry to user space lookup the entry - * by name and if I find a different entry don't return - * this one because it means it is a buried dup. - * For sysctl this should only happen for directory entries. - */ - for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) { - table = proc_sys_lookup_table(dentry, head->ctl_table); + ret = scan(head, table, &pos, filp, dirent, filldir); + if (ret) + goto out; - if (!table) + for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) { + if (h->attached_to != table) continue; - - for (; table->ctl_name || table->procname; table++, pos++) { - /* Can't do anything without a proc name */ - if (!table->procname) - continue; - - if (pos < filp->f_pos) - continue; - - if (proc_sys_fill_cache(filp, dirent, filldir, table) < 0) - goto out; - filp->f_pos = pos + 1; + ret = scan(h, h->attached_by, &pos, filp, dirent, filldir); + if (ret) { + sysctl_head_finish(h); + break; } } ret = 1; @@ -343,53 +292,24 @@ out: return ret; } -static int proc_sys_permission(struct inode *inode, int mask, struct nameidata *nd) +static int proc_sys_permission(struct inode *inode, int mask) { /* * sysctl entries that are not writeable, * are _NOT_ writeable, capabilities or not. */ - struct ctl_table_header *head; - struct ctl_table *table; - struct dentry *dentry; - int mode; - int depth; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; int error; - head = NULL; - depth = PROC_I(inode)->fd; - - /* First check the cached permissions, in case we don't have - * enough information to lookup the sysctl table entry. - */ - error = -EACCES; - mode = inode->i_mode; - - if (current->euid == 0) - mode >>= 6; - else if (in_group_p(0)) - mode >>= 3; - - if ((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask) - error = 0; - - /* If we can't get a sysctl table entry the permission - * checks on the cached mode will have to be enough. - */ - if (!nd || !depth) - goto out; + if (IS_ERR(head)) + return PTR_ERR(head); - dentry = nd->path.dentry; - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); + if (!table) /* global root - r-xr-xr-x */ + error = mask & MAY_WRITE ? -EACCES : 0; + else /* Use the permissions on the sysctl table entry */ + error = sysctl_perm(head->root, table, mask); - /* If the entry does not exist deny permission */ - error = -EACCES; - if (!table) - goto out; - - /* Use the permissions on the sysctl table entry */ - error = sysctl_perm(head->root, table, mask); -out: sysctl_head_finish(head); return error; } @@ -409,33 +329,70 @@ static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr) return error; } -/* I'm lazy and don't distinguish between files and directories, - * until access time. - */ +static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct ctl_table_header *head = grab_header(inode); + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + + if (IS_ERR(head)) + return PTR_ERR(head); + + generic_fillattr(inode, stat); + if (table) + stat->mode = (stat->mode & S_IFMT) | table->mode; + + sysctl_head_finish(head); + return 0; +} + static const struct file_operations proc_sys_file_operations = { .read = proc_sys_read, .write = proc_sys_write, +}; + +static const struct file_operations proc_sys_dir_file_operations = { .readdir = proc_sys_readdir, }; static const struct inode_operations proc_sys_inode_operations = { + .permission = proc_sys_permission, + .setattr = proc_sys_setattr, + .getattr = proc_sys_getattr, +}; + +static const struct inode_operations proc_sys_dir_operations = { .lookup = proc_sys_lookup, .permission = proc_sys_permission, .setattr = proc_sys_setattr, + .getattr = proc_sys_getattr, }; static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct ctl_table_header *head; - struct ctl_table *table; - table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head); - proc_sys_refresh_inode(dentry->d_inode, table); - sysctl_head_finish(head); - return !!table; + return !PROC_I(dentry->d_inode)->sysctl->unregistering; +} + +static int proc_sys_delete(struct dentry *dentry) +{ + return !!PROC_I(dentry->d_inode)->sysctl->unregistering; +} + +static int proc_sys_compare(struct dentry *dir, struct qstr *qstr, + struct qstr *name) +{ + struct dentry *dentry = container_of(qstr, struct dentry, d_name); + if (qstr->len != name->len) + return 1; + if (memcmp(qstr->name, name->name, name->len)) + return 1; + return !sysctl_is_seen(PROC_I(dentry->d_inode)->sysctl); } static struct dentry_operations proc_sys_dentry_operations = { .d_revalidate = proc_sys_revalidate, + .d_delete = proc_sys_delete, + .d_compare = proc_sys_compare, }; static struct proc_dir_entry *proc_sys_root; @@ -443,8 +400,8 @@ static struct proc_dir_entry *proc_sys_root; int proc_sys_init(void) { proc_sys_root = proc_mkdir("sys", NULL); - proc_sys_root->proc_iops = &proc_sys_inode_operations; - proc_sys_root->proc_fops = &proc_sys_file_operations; + proc_sys_root->proc_iops = &proc_sys_dir_operations; + proc_sys_root->proc_fops = &proc_sys_dir_file_operations; proc_sys_root->nlink = 0; return 0; } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 7546a91..73d1891 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -219,14 +219,14 @@ static int show_map(struct seq_file *m, void *v) ino = inode->i_ino; } - seq_printf(m, "%08lx-%08lx %c%c%c%c %08lx %02x:%02x %lu %n", + seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", vma->vm_start, vma->vm_end, flags & VM_READ ? 'r' : '-', flags & VM_WRITE ? 'w' : '-', flags & VM_EXEC ? 'x' : '-', flags & VM_MAYSHARE ? 's' : 'p', - vma->vm_pgoff << PAGE_SHIFT, + ((loff_t)vma->vm_pgoff) << PAGE_SHIFT, MAJOR(dev), MINOR(dev), ino, &len); /* diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index b31ab78..2aad104 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -553,7 +553,7 @@ static void qnx4_destroy_inode(struct inode *inode) kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 52312ec..5145cb9 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -58,7 +58,7 @@ const struct inode_operations ramfs_file_inode_operations = { * size 0 on the assumption that it's going to be used for an mmap of shared * memory */ -static int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) +int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) { struct pagevec lru_pvec; unsigned long npages, xpages, loop, limit; diff --git a/fs/readdir.c b/fs/readdir.c index 4e026e5..93a7559 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -80,8 +80,10 @@ static int fillonedir(void * __buf, const char * name, int namlen, loff_t offset if (buf->result) return -EINVAL; d_ino = ino; - if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) + if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) { + buf->result = -EOVERFLOW; return -EOVERFLOW; + } buf->result++; dirent = buf->dirent; if (!access_ok(VERIFY_WRITE, dirent, @@ -155,8 +157,10 @@ static int filldir(void * __buf, const char * name, int namlen, loff_t offset, if (reclen > buf->count) return -EINVAL; d_ino = ino; - if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) + if (sizeof(d_ino) < sizeof(ino) && d_ino != ino) { + buf->error = -EOVERFLOW; return -EOVERFLOW; + } dirent = buf->previous; if (dirent) { if (__put_user(offset, &dirent->d_off)) diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 1922696..5699171 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2435,7 +2435,7 @@ static int reiserfs_write_full_page(struct page *page, if (wbc->sync_mode != WB_SYNC_NONE || !wbc->nonblocking) { lock_buffer(bh); } else { - if (test_set_buffer_locked(bh)) { + if (!trylock_buffer(bh)) { redirty_page_for_writepage(wbc, page); continue; } diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index c8f60ee..c21df71 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -627,7 +627,7 @@ static int journal_list_still_alive(struct super_block *s, static void release_buffer_page(struct buffer_head *bh) { struct page *page = bh->b_page; - if (!page->mapping && !TestSetPageLocked(page)) { + if (!page->mapping && trylock_page(page)) { page_cache_get(page); put_bh(bh); if (!page->mapping) @@ -855,7 +855,7 @@ static int write_ordered_buffers(spinlock_t * lock, jh = JH_ENTRY(list->next); bh = jh->bh; get_bh(bh); - if (test_set_buffer_locked(bh)) { + if (!trylock_buffer(bh)) { if (!buffer_dirty(bh)) { list_move(&jh->list, &tmp); goto loop_next; @@ -3871,7 +3871,7 @@ int reiserfs_prepare_for_journal(struct super_block *p_s_sb, { PROC_INFO_INC(p_s_sb, journal.prepare); - if (test_set_buffer_locked(bh)) { + if (!trylock_buffer(bh)) { if (!wait) return 0; lock_buffer(bh); diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 2ec748b..d318c7e 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -27,7 +27,6 @@ #include <linux/mnt_namespace.h> #include <linux/mount.h> #include <linux/namei.h> -#include <linux/quotaops.h> struct file_system_type reiserfs_fs_type; @@ -521,7 +520,7 @@ static void reiserfs_destroy_inode(struct inode *inode) kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; @@ -2076,8 +2075,8 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, return err; /* Quotafile not on the same filesystem? */ if (nd.path.mnt->mnt_sb != sb) { - path_put(&nd.path); - return -EXDEV; + err = -EXDEV; + goto out; } inode = nd.path.dentry->d_inode; /* We must not pack tails for quota files on reiserfs for quota IO to work */ @@ -2087,8 +2086,8 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, reiserfs_warning(sb, "reiserfs: Unpacking tail of quota file failed" " (%d). Cannot turn on quotas.", err); - path_put(&nd.path); - return -EINVAL; + err = -EINVAL; + goto out; } mark_inode_dirty(inode); } @@ -2109,13 +2108,15 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id, /* Just start temporary transaction and finish it */ err = journal_begin(&th, sb, 1); if (err) - return err; + goto out; err = journal_end_sync(&th, sb, 1); if (err) - return err; + goto out; } + err = vfs_quota_on_path(sb, type, format_id, &nd.path); +out: path_put(&nd.path); - return vfs_quota_on(sb, type, format_id, path, 0); + return err; } /* Read data from quotafile - avoid pagecache and such because we cannot afford diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index d7c4935..bb3cb5b 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -1250,7 +1250,7 @@ static int reiserfs_check_acl(struct inode *inode, int mask) return error; } -int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd) +int reiserfs_permission(struct inode *inode, int mask) { /* * We don't do permission checks on the internal objects. diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index 3f13d49..60d2f82 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -418,7 +418,8 @@ static int romfs_readpage(struct file *file, struct page * page) { struct inode *inode = page->mapping->host; - loff_t offset, avail, readlen; + loff_t offset, size; + unsigned long filled; void *buf; int result = -EIO; @@ -430,21 +431,29 @@ romfs_readpage(struct file *file, struct page * page) /* 32 bit warning -- but not for us :) */ offset = page_offset(page); - if (offset < i_size_read(inode)) { - avail = inode->i_size-offset; - readlen = min_t(unsigned long, avail, PAGE_SIZE); - if (romfs_copyfrom(inode, buf, ROMFS_I(inode)->i_dataoffset+offset, readlen) == readlen) { - if (readlen < PAGE_SIZE) { - memset(buf + readlen,0,PAGE_SIZE-readlen); - } - SetPageUptodate(page); - result = 0; + size = i_size_read(inode); + filled = 0; + result = 0; + if (offset < size) { + unsigned long readlen; + + size -= offset; + readlen = size > PAGE_SIZE ? PAGE_SIZE : size; + + filled = romfs_copyfrom(inode, buf, ROMFS_I(inode)->i_dataoffset+offset, readlen); + + if (filled != readlen) { + SetPageError(page); + filled = 0; + result = -EIO; } } - if (result) { - memset(buf, 0, PAGE_SIZE); - SetPageError(page); - } + + if (filled < PAGE_SIZE) + memset(buf + filled, 0, PAGE_SIZE-filled); + + if (!result) + SetPageUptodate(page); flush_dcache_page(page); unlock_page(page); @@ -577,7 +586,7 @@ static void romfs_destroy_inode(struct inode *inode) kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct romfs_inode_info *ei = foo; diff --git a/fs/seq_file.c b/fs/seq_file.c index 3f54dbd..bd20f7f 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -108,9 +108,9 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) goto Done; } /* we need at least one record in buffer */ + pos = m->index; + p = m->op->start(m, &pos); while (1) { - pos = m->index; - p = m->op->start(m, &pos); err = PTR_ERR(p); if (!p || IS_ERR(p)) break; @@ -119,6 +119,11 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) break; if (unlikely(err)) m->count = 0; + if (unlikely(!m->count)) { + p = m->op->next(m, p, &pos); + m->index = pos; + continue; + } if (m->count < m->size) goto Fill; m->op->stop(m, p); @@ -128,6 +133,8 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) goto Enomem; m->count = 0; m->version = 0; + pos = m->index; + p = m->op->start(m, &pos); } m->op->stop(m, p); m->count = 0; @@ -443,6 +450,20 @@ int seq_dentry(struct seq_file *m, struct dentry *dentry, char *esc) return -1; } +int seq_bitmap(struct seq_file *m, unsigned long *bits, unsigned int nr_bits) +{ + size_t len = bitmap_scnprintf_len(nr_bits); + + if (m->count + len < m->size) { + bitmap_scnprintf(m->buf + m->count, m->size - m->count, + bits, nr_bits); + m->count += len; + return 0; + } + m->count = m->size; + return -1; +} + static void *single_start(struct seq_file *p, loff_t *pos) { return NULL + (*pos == 0); diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c index 2294783..e4f8d51 100644 --- a/fs/smbfs/file.c +++ b/fs/smbfs/file.c @@ -408,7 +408,7 @@ smb_file_release(struct inode *inode, struct file * file) * privileges, so we need our own check for this. */ static int -smb_file_permission(struct inode *inode, int mask, struct nameidata *nd) +smb_file_permission(struct inode *inode, int mask) { int mode = inode->i_mode; int error = 0; @@ -417,7 +417,7 @@ smb_file_permission(struct inode *inode, int mask, struct nameidata *nd) /* Look at user permissions */ mode >>= 6; - if ((mode & 7 & mask) != mask) + if (mask & ~mode & (MAY_READ | MAY_WRITE | MAY_EXEC)) error = -EACCES; return error; } diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 376ef3e..3528f40 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -67,7 +67,7 @@ static void smb_destroy_inode(struct inode *inode) kmem_cache_free(smb_inode_cachep, SMB_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct smb_inode_info *ei = (struct smb_inode_info *) foo; diff --git a/fs/splice.c b/fs/splice.c index 3994421..a1e701c 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -371,7 +371,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * for an in-flight io page */ if (flags & SPLICE_F_NONBLOCK) { - if (TestSetPageLocked(page)) { + if (!trylock_page(page)) { error = -EAGAIN; break; } @@ -772,7 +772,7 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; int err; - err = remove_suid(out->f_path.dentry); + err = file_remove_suid(out); if (unlikely(err)) return err; @@ -830,7 +830,7 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, ssize_t ret; inode_double_lock(inode, pipe->inode); - ret = remove_suid(out->f_path.dentry); + ret = file_remove_suid(out); if (likely(!ret)) ret = __splice_from_pipe(pipe, &sd, pipe_to_file); inode_double_unlock(inode, pipe->inode); @@ -898,6 +898,9 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, if (unlikely(!(out->f_mode & FMODE_WRITE))) return -EBADF; + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + ret = rw_verify_area(WRITE, out, ppos, len); if (unlikely(ret < 0)) return ret; @@ -1161,36 +1164,6 @@ static long do_splice(struct file *in, loff_t __user *off_in, } /* - * Do a copy-from-user while holding the mmap_semaphore for reading, in a - * manner safe from deadlocking with simultaneous mmap() (grabbing mmap_sem - * for writing) and page faulting on the user memory pointed to by src. - * This assumes that we will very rarely hit the partial != 0 path, or this - * will not be a win. - */ -static int copy_from_user_mmap_sem(void *dst, const void __user *src, size_t n) -{ - int partial; - - if (!access_ok(VERIFY_READ, src, n)) - return -EFAULT; - - pagefault_disable(); - partial = __copy_from_user_inatomic(dst, src, n); - pagefault_enable(); - - /* - * Didn't copy everything, drop the mmap_sem and do a faulting copy - */ - if (unlikely(partial)) { - up_read(¤t->mm->mmap_sem); - partial = copy_from_user(dst, src, n); - down_read(¤t->mm->mmap_sem); - } - - return partial; -} - -/* * Map an iov into an array of pages and offset/length tupples. With the * partial_page structure, we can map several non-contiguous ranges into * our ones pages[] map instead of splitting that operation into pieces. @@ -1203,8 +1176,6 @@ static int get_iovec_page_array(const struct iovec __user *iov, { int buffers = 0, error = 0; - down_read(¤t->mm->mmap_sem); - while (nr_vecs) { unsigned long off, npages; struct iovec entry; @@ -1213,7 +1184,7 @@ static int get_iovec_page_array(const struct iovec __user *iov, int i; error = -EFAULT; - if (copy_from_user_mmap_sem(&entry, iov, sizeof(entry))) + if (copy_from_user(&entry, iov, sizeof(entry))) break; base = entry.iov_base; @@ -1247,9 +1218,8 @@ static int get_iovec_page_array(const struct iovec __user *iov, if (npages > PIPE_BUFFERS - buffers) npages = PIPE_BUFFERS - buffers; - error = get_user_pages(current, current->mm, - (unsigned long) base, npages, 0, 0, - &pages[buffers], NULL); + error = get_user_pages_fast((unsigned long)base, npages, + 0, &pages[buffers]); if (unlikely(error <= 0)) break; @@ -1288,8 +1258,6 @@ static int get_iovec_page_array(const struct iovec __user *iov, iov++; } - up_read(¤t->mm->mmap_sem); - if (buffers) return buffers; @@ -57,13 +57,13 @@ EXPORT_SYMBOL(vfs_getattr); int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat) { - struct nameidata nd; + struct path path; int error; - error = __user_walk_fd(dfd, name, LOOKUP_FOLLOW, &nd); + error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path); if (!error) { - error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); - path_put(&nd.path); + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); } return error; } @@ -77,13 +77,13 @@ EXPORT_SYMBOL(vfs_stat); int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat) { - struct nameidata nd; + struct path path; int error; - error = __user_walk_fd(dfd, name, 0, &nd); + error = user_path_at(dfd, name, 0, &path); if (!error) { - error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat); - path_put(&nd.path); + error = vfs_getattr(path.mnt, path.dentry, stat); + path_put(&path); } return error; } @@ -291,29 +291,29 @@ asmlinkage long sys_newfstat(unsigned int fd, struct stat __user *statbuf) return error; } -asmlinkage long sys_readlinkat(int dfd, const char __user *path, +asmlinkage long sys_readlinkat(int dfd, const char __user *pathname, char __user *buf, int bufsiz) { - struct nameidata nd; + struct path path; int error; if (bufsiz <= 0) return -EINVAL; - error = __user_walk_fd(dfd, path, 0, &nd); + error = user_path_at(dfd, pathname, 0, &path); if (!error) { - struct inode *inode = nd.path.dentry->d_inode; + struct inode *inode = path.dentry->d_inode; error = -EINVAL; if (inode->i_op && inode->i_op->readlink) { - error = security_inode_readlink(nd.path.dentry); + error = security_inode_readlink(path.dentry); if (!error) { - touch_atime(nd.path.mnt, nd.path.dentry); - error = inode->i_op->readlink(nd.path.dentry, + touch_atime(path.mnt, path.dentry); + error = inode->i_op->readlink(path.dentry, buf, bufsiz); } } - path_put(&nd.path); + path_put(&path); } return error; } diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index c1a7efb..aedaeba 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -459,11 +459,8 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) int ret; ret = __sysfs_add_one(acxt, sd); - if (ret == -EEXIST) { - printk(KERN_WARNING "sysfs: duplicate filename '%s' " + WARN(ret == -EEXIST, KERN_WARNING "sysfs: duplicate filename '%s' " "can not be created\n", sd->s_name); - WARN_ON(1); - } return ret; } diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 3f07893..c9e4e50 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -337,9 +337,8 @@ static int sysfs_open_file(struct inode *inode, struct file *file) if (kobj->ktype && kobj->ktype->sysfs_ops) ops = kobj->ktype->sysfs_ops; else { - printk(KERN_ERR "missing sysfs attribute operations for " + WARN(1, KERN_ERR "missing sysfs attribute operations for " "kobject: %s\n", kobject_name(kobj)); - WARN_ON(1); goto err_out; } diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index eeba384..fe61194 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -134,9 +134,8 @@ void sysfs_remove_group(struct kobject * kobj, if (grp->name) { sd = sysfs_get_dirent(dir_sd, grp->name); if (!sd) { - printk(KERN_WARNING "sysfs group %p not found for " + WARN(!sd, KERN_WARNING "sysfs group %p not found for " "kobject '%s'\n", grp, kobject_name(kobj)); - WARN_ON(!sd); return; } } else diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index c5d60de..df0d435 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -326,7 +326,7 @@ static void sysv_destroy_inode(struct inode *inode) kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *p) +static void init_once(void *p) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index d81fb9e..73db464 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -263,8 +263,8 @@ int ubifs_calc_min_idx_lebs(struct ubifs_info *c) idx_size = c->old_idx_sz + c->budg_idx_growth + c->budg_uncommitted_idx; - /* And make sure we have twice the index size of space reserved */ - idx_size <<= 1; + /* And make sure we have thrice the index size of space reserved */ + idx_size = idx_size + (idx_size << 1); /* * We do not maintain 'old_idx_size' as 'old_idx_lebs'/'old_idx_bytes' @@ -302,18 +302,6 @@ long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs) int subtract_lebs; long long available; - /* - * Force the amount available to the total size reported if the used - * space is zero. - */ - if (c->lst.total_used <= UBIFS_INO_NODE_SZ && - c->budg_data_growth + c->budg_dd_growth == 0) { - /* Do the same calculation as for c->block_cnt */ - available = c->main_lebs - 2; - available *= c->leb_size - c->dark_wm; - return available; - } - available = c->main_bytes - c->lst.total_used; /* @@ -388,11 +376,11 @@ static int can_use_rp(struct ubifs_info *c) * This function makes sure UBIFS has enough free eraseblocks for index growth * and data. * - * When budgeting index space, UBIFS reserves twice as more LEBs as the index + * When budgeting index space, UBIFS reserves thrice as many LEBs as the index * would take if it was consolidated and written to the flash. This guarantees * that the "in-the-gaps" commit method always succeeds and UBIFS will always * be able to commit dirty index. So this function basically adds amount of - * budgeted index space to the size of the current index, multiplies this by 2, + * 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 eraseblocks. * * Notes about @c->min_idx_lebs and @c->lst.idx_lebs variables: @@ -543,8 +531,16 @@ int ubifs_budget_space(struct ubifs_info *c, struct ubifs_budget_req *req) int err, idx_growth, data_growth, dd_growth; struct retries_info ri; + ubifs_assert(req->new_page <= 1); + ubifs_assert(req->dirtied_page <= 1); + ubifs_assert(req->new_dent <= 1); + ubifs_assert(req->mod_dent <= 1); + ubifs_assert(req->new_ino <= 1); + ubifs_assert(req->new_ino_d <= UBIFS_MAX_INO_DATA); ubifs_assert(req->dirtied_ino <= 4); ubifs_assert(req->dirtied_ino_d <= UBIFS_MAX_INO_DATA * 4); + ubifs_assert(!(req->new_ino_d & 7)); + ubifs_assert(!(req->dirtied_ino_d & 7)); data_growth = calc_data_growth(c, req); dd_growth = calc_dd_growth(c, req); @@ -618,8 +614,16 @@ again: */ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) { + ubifs_assert(req->new_page <= 1); + ubifs_assert(req->dirtied_page <= 1); + ubifs_assert(req->new_dent <= 1); + ubifs_assert(req->mod_dent <= 1); + ubifs_assert(req->new_ino <= 1); + ubifs_assert(req->new_ino_d <= UBIFS_MAX_INO_DATA); ubifs_assert(req->dirtied_ino <= 4); ubifs_assert(req->dirtied_ino_d <= UBIFS_MAX_INO_DATA * 4); + ubifs_assert(!(req->new_ino_d & 7)); + ubifs_assert(!(req->dirtied_ino_d & 7)); if (!req->recalculate) { ubifs_assert(req->idx_growth >= 0); ubifs_assert(req->data_growth >= 0); @@ -647,7 +651,11 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) 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)); spin_unlock(&c->space_lock); } @@ -686,41 +694,114 @@ void ubifs_convert_page_budget(struct ubifs_info *c) void ubifs_release_dirty_inode_budget(struct ubifs_info *c, struct ubifs_inode *ui) { - struct ubifs_budget_req req = {.dd_growth = c->inode_budget, - .dirtied_ino_d = ui->data_len}; + struct ubifs_budget_req req; + memset(&req, 0, sizeof(struct ubifs_budget_req)); + req.dd_growth = c->inode_budget + ALIGN(ui->data_len, 8); ubifs_release_budget(c, &req); } /** - * ubifs_budg_get_free_space - return amount of free space. + * ubifs_reported_space - calculate reported free space. + * @c: the UBIFS file-system description object + * @free: amount of free space + * + * This function calculates amount of free space which will be reported to + * user-space. User-space application tend to expect that if the file-system + * (e.g., via the 'statfs()' call) reports that it has N bytes available, they + * are able to write a file of size N. UBIFS attaches node headers to each data + * node and it has to write indexind nodes as well. This introduces additional + * overhead, and UBIFS it has to report sligtly less free space to meet the + * above expectetion. + * + * This function assumes free space is made up of uncompressed data nodes and + * full index nodes (one per data node, tripled because we always allow enough + * space to write the index thrice). + * + * Note, the calculation is pessimistic, which means that most of the time + * UBIFS reports less space than it actually has. + */ +long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free) +{ + int divisor, factor, f; + + /* + * Reported space size is @free * X, where X is UBIFS block size + * divided by UBIFS block size + all overhead one data block + * introduces. The overhead is the node header + indexing overhead. + * + * Indexing overhead calculations are based on the following formula: + * I = N/(f - 1) + 1, where I - number of indexing nodes, N - number + * of data nodes, f - fanout. Because effective UBIFS fanout is twice + * as less than maximum fanout, we assume that each data node + * introduces 3 * @c->max_idx_node_sz / (@c->fanout/2 - 1) bytes. + * Note, the multiplier 3 is because UBIFS reseves thrice as more space + * for the index. + */ + f = c->fanout > 3 ? c->fanout >> 1 : 2; + factor = UBIFS_BLOCK_SIZE; + divisor = UBIFS_MAX_DATA_NODE_SZ; + divisor += (c->max_idx_node_sz * 3) / (f - 1); + free *= factor; + do_div(free, divisor); + return free; +} + +/** + * ubifs_get_free_space - return amount of free space. * @c: UBIFS file-system description object * - * This function returns amount of free space on the file-system. + * This function calculates amount of free space to report to user-space. + * + * Because UBIFS may introduce substantial overhead (the index, node headers, + * alighment, wastage at the end of eraseblocks, etc), it cannot report real + * amount of free flash space it has (well, because not all dirty space is + * reclamable, UBIFS does not actually know the real amount). If UBIFS did so, + * it would bread user expectetion about what free space is. Users seem to + * accustomed to assume that if the file-system reports N bytes of free space, + * they would be able to fit a file of N bytes to the FS. This almost works for + * traditional file-systems, because they have way less overhead than UBIFS. + * So, to keep users happy, UBIFS tries to take the overhead into account. */ -long long ubifs_budg_get_free_space(struct ubifs_info *c) +long long ubifs_get_free_space(struct ubifs_info *c) { - int min_idx_lebs, rsvd_idx_lebs; + int min_idx_lebs, rsvd_idx_lebs, lebs; long long available, outstanding, free; - /* Do exactly the same calculations as in 'do_budget_space()' */ spin_lock(&c->space_lock); min_idx_lebs = ubifs_calc_min_idx_lebs(c); + outstanding = c->budg_data_growth + c->budg_dd_growth; - if (min_idx_lebs > c->lst.idx_lebs) - rsvd_idx_lebs = min_idx_lebs - c->lst.idx_lebs; - else - rsvd_idx_lebs = 0; - - if (rsvd_idx_lebs > c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - - c->lst.taken_empty_lebs) { + /* + * Force the amount available to the total size reported if the used + * space is zero. + */ + if (c->lst.total_used <= UBIFS_INO_NODE_SZ && !outstanding) { spin_unlock(&c->space_lock); - return 0; + return (long long)c->block_cnt << UBIFS_BLOCK_SHIFT; } available = ubifs_calc_available(c, min_idx_lebs); - outstanding = c->budg_data_growth + c->budg_dd_growth; - c->min_idx_lebs = min_idx_lebs; + + /* + * When reporting free space to user-space, UBIFS guarantees that it is + * possible to write a file of free space size. This means that for + * empty LEBs we may use more precise calculations than + * 'ubifs_calc_available()' is using. Namely, we know that in empty + * LEBs we would waste only @c->leb_overhead bytes, not @c->dark_wm. + * Thus, amend the available space. + * + * Note, the calculations below are similar to what we have in + * 'do_budget_space()', so refer there for comments. + */ + if (min_idx_lebs > c->lst.idx_lebs) + rsvd_idx_lebs = min_idx_lebs - c->lst.idx_lebs; + else + rsvd_idx_lebs = 0; + lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - + c->lst.taken_empty_lebs; + lebs -= rsvd_idx_lebs; + available += lebs * (c->dark_wm - c->leb_overhead); spin_unlock(&c->space_lock); if (available > outstanding) diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c index 3b51631..0a6aa2c 100644 --- a/fs/ubifs/commit.c +++ b/fs/ubifs/commit.c @@ -74,6 +74,7 @@ static int do_commit(struct ubifs_info *c) goto out_up; } + c->cmt_no += 1; err = ubifs_gc_start_commit(c); if (err) goto out_up; @@ -115,7 +116,7 @@ static int do_commit(struct ubifs_info *c) goto out; mutex_lock(&c->mst_mutex); - c->mst_node->cmt_no = cpu_to_le64(++c->cmt_no); + c->mst_node->cmt_no = cpu_to_le64(c->cmt_no); c->mst_node->log_lnum = cpu_to_le32(new_ltail_lnum); c->mst_node->root_lnum = cpu_to_le32(zroot.lnum); c->mst_node->root_offs = cpu_to_le32(zroot.offs); diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index 4e3aaeb..d7f7645 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -538,7 +538,7 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node) printk(KERN_DEBUG "\t%d orphan inode numbers:\n", n); for (i = 0; i < n; i++) printk(KERN_DEBUG "\t ino %llu\n", - le64_to_cpu(orph->inos[i])); + (unsigned long long)le64_to_cpu(orph->inos[i])); break; } default: @@ -568,8 +568,8 @@ void dbg_dump_budget_req(const struct ubifs_budget_req *req) void dbg_dump_lstats(const struct ubifs_lp_stats *lst) { spin_lock(&dbg_lock); - printk(KERN_DEBUG "Lprops statistics: empty_lebs %d, idx_lebs %d\n", - lst->empty_lebs, lst->idx_lebs); + printk(KERN_DEBUG "(pid %d) Lprops statistics: empty_lebs %d, " + "idx_lebs %d\n", current->pid, lst->empty_lebs, lst->idx_lebs); printk(KERN_DEBUG "\ttaken_empty_lebs %d, total_free %lld, " "total_dirty %lld\n", lst->taken_empty_lebs, lst->total_free, lst->total_dirty); @@ -587,8 +587,8 @@ void dbg_dump_budg(struct ubifs_info *c) struct ubifs_gced_idx_leb *idx_gc; spin_lock(&dbg_lock); - printk(KERN_DEBUG "Budgeting info: budg_data_growth %lld, " - "budg_dd_growth %lld, budg_idx_growth %lld\n", + 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, @@ -634,7 +634,7 @@ void dbg_dump_lprops(struct ubifs_info *c) struct ubifs_lprops lp; struct ubifs_lp_stats lst; - printk(KERN_DEBUG "Dumping LEB properties\n"); + printk(KERN_DEBUG "(pid %d) Dumping LEB properties\n", current->pid); ubifs_get_lp_stats(c, &lst); dbg_dump_lstats(&lst); @@ -655,7 +655,7 @@ void dbg_dump_leb(const struct ubifs_info *c, int lnum) if (dbg_failure_mode) return; - printk(KERN_DEBUG "Dumping LEB %d\n", lnum); + printk(KERN_DEBUG "(pid %d) Dumping LEB %d\n", current->pid, lnum); sleb = ubifs_scan(c, lnum, 0, c->dbg_buf); if (IS_ERR(sleb)) { @@ -720,8 +720,8 @@ void dbg_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat) { int i; - printk(KERN_DEBUG "Dumping heap cat %d (%d elements)\n", - cat, heap->cnt); + printk(KERN_DEBUG "(pid %d) Dumping heap cat %d (%d elements)\n", + current->pid, cat, heap->cnt); for (i = 0; i < heap->cnt; i++) { struct ubifs_lprops *lprops = heap->arr[i]; @@ -736,7 +736,7 @@ void dbg_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode, { int i; - printk(KERN_DEBUG "Dumping pnode:\n"); + printk(KERN_DEBUG "(pid %d) Dumping pnode:\n", current->pid); printk(KERN_DEBUG "\taddress %zx parent %zx cnext %zx\n", (size_t)pnode, (size_t)parent, (size_t)pnode->cnext); printk(KERN_DEBUG "\tflags %lu iip %d level %d num %d\n", @@ -755,7 +755,7 @@ void dbg_dump_tnc(struct ubifs_info *c) int level; printk(KERN_DEBUG "\n"); - printk(KERN_DEBUG "Dumping the TNC tree\n"); + printk(KERN_DEBUG "(pid %d) Dumping the TNC tree\n", current->pid); znode = ubifs_tnc_levelorder_next(c->zroot.znode, NULL); level = znode->level; printk(KERN_DEBUG "== Level %d ==\n", level); @@ -2208,16 +2208,17 @@ int dbg_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset, int dbg_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf, int offset, int len, int dtype) { - int err; + int err, failing; if (in_failure_mode(desc)) return -EIO; - if (do_fail(desc, lnum, 1)) + failing = do_fail(desc, lnum, 1); + if (failing) cut_data(buf, len); err = ubi_leb_write(desc, lnum, buf, offset, len, dtype); if (err) return err; - if (in_failure_mode(desc)) + if (failing) return -EIO; return 0; } diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index 3c4f1e9..50315fc 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -27,7 +27,7 @@ #define UBIFS_DBG(op) op -#define ubifs_assert(expr) do { \ +#define ubifs_assert(expr) do { \ if (unlikely(!(expr))) { \ printk(KERN_CRIT "UBIFS assert failed in %s at %u (pid %d)\n", \ __func__, __LINE__, current->pid); \ @@ -73,50 +73,50 @@ const char *dbg_key_str1(const struct ubifs_info *c, const union ubifs_key *key); /* - * DBGKEY macros require dbg_lock to be held, which it is in the dbg message + * DBGKEY macros require @dbg_lock to be held, which it is in the dbg message * macros. */ #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 dbg_gen(fmt, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_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, ...) dbg_do_msg(UBIFS_MSG_SCAN, fmt, ##__VA_ARGS__) /* Additional recovery messages */ -#define dbg_rcvry(fmt, ...) dbg_do_msg(UBIFS_MSG_RCVRY, fmt, ##__VA_ARGS__) +#define dbg_rcvry(fmt, ...) dbg_do_msg(UBIFS_MSG_RCVRY, fmt, ##__VA_ARGS__) /* * Debugging message type flags (must match msg_type_names in debug.c). @@ -239,34 +239,23 @@ typedef int (*dbg_leaf_callback)(struct ubifs_info *c, struct ubifs_zbranch *zbr, void *priv); typedef int (*dbg_znode_callback)(struct ubifs_info *c, struct ubifs_znode *znode, void *priv); - int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb, dbg_znode_callback znode_cb, void *priv); /* Checking functions */ int dbg_check_lprops(struct ubifs_info *c); - int dbg_old_index_check_init(struct ubifs_info *c, struct ubifs_zbranch *zroot); int dbg_check_old_index(struct ubifs_info *c, struct ubifs_zbranch *zroot); - int dbg_check_cats(struct ubifs_info *c); - int dbg_check_ltab(struct ubifs_info *c); - int dbg_check_synced_i_size(struct inode *inode); - int dbg_check_dir_size(struct ubifs_info *c, const struct inode *dir); - int dbg_check_tnc(struct ubifs_info *c, int extra); - int dbg_check_idx_size(struct ubifs_info *c, long long idx_size); - int dbg_check_filesystem(struct ubifs_info *c); - void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat, int add_pos); - int dbg_check_lprops(struct ubifs_info *c); int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode, int row, int col); @@ -329,71 +318,77 @@ static inline int dbg_change(struct ubi_volume_desc *desc, int lnum, #else /* !CONFIG_UBIFS_FS_DEBUG */ #define UBIFS_DBG(op) -#define ubifs_assert(expr) ({}) -#define ubifs_assert_cmt_locked(c) + +/* Use "if (0)" to make compiler check arguments even if debugging is off */ +#define ubifs_assert(expr) do { \ + if (0 && (expr)) \ + printk(KERN_CRIT "UBIFS assert failed in %s at %u (pid %d)\n", \ + __func__, __LINE__, current->pid); \ +} while (0) + +#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__); \ +} while (0) + #define dbg_dump_stack() -#define dbg_err(fmt, ...) ({}) -#define dbg_msg(fmt, ...) ({}) -#define dbg_key(c, key, fmt, ...) ({}) - -#define dbg_gen(fmt, ...) ({}) -#define dbg_jnl(fmt, ...) ({}) -#define dbg_tnc(fmt, ...) ({}) -#define dbg_lp(fmt, ...) ({}) -#define dbg_find(fmt, ...) ({}) -#define dbg_mnt(fmt, ...) ({}) -#define dbg_io(fmt, ...) ({}) -#define dbg_cmt(fmt, ...) ({}) -#define dbg_budg(fmt, ...) ({}) -#define dbg_log(fmt, ...) ({}) -#define dbg_gc(fmt, ...) ({}) -#define dbg_scan(fmt, ...) ({}) -#define dbg_rcvry(fmt, ...) ({}) - -#define dbg_ntype(type) "" -#define dbg_cstate(cmt_state) "" -#define dbg_get_key_dump(c, key) ({}) -#define dbg_dump_inode(c, inode) ({}) -#define dbg_dump_node(c, node) ({}) -#define dbg_dump_budget_req(req) ({}) -#define dbg_dump_lstats(lst) ({}) -#define dbg_dump_budg(c) ({}) -#define dbg_dump_lprop(c, lp) ({}) -#define dbg_dump_lprops(c) ({}) -#define dbg_dump_leb(c, lnum) ({}) -#define dbg_dump_znode(c, znode) ({}) -#define dbg_dump_heap(c, heap, cat) ({}) -#define dbg_dump_pnode(c, pnode, parent, iip) ({}) -#define dbg_dump_tnc(c) ({}) -#define dbg_dump_index(c) ({}) +#define ubifs_assert_cmt_locked(c) -#define dbg_walk_index(c, leaf_cb, znode_cb, priv) 0 +#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 DBGKEY(key) ((char *)(key)) +#define DBGKEY1(key) ((char *)(key)) + +#define dbg_ntype(type) "" +#define dbg_cstate(cmt_state) "" +#define dbg_get_key_dump(c, key) ({}) +#define dbg_dump_inode(c, inode) ({}) +#define dbg_dump_node(c, node) ({}) +#define dbg_dump_budget_req(req) ({}) +#define dbg_dump_lstats(lst) ({}) +#define dbg_dump_budg(c) ({}) +#define dbg_dump_lprop(c, lp) ({}) +#define dbg_dump_lprops(c) ({}) +#define dbg_dump_leb(c, lnum) ({}) +#define dbg_dump_znode(c, znode) ({}) +#define dbg_dump_heap(c, heap, cat) ({}) +#define dbg_dump_pnode(c, pnode, parent, iip) ({}) +#define dbg_dump_tnc(c) ({}) +#define dbg_dump_index(c) ({}) +#define dbg_walk_index(c, leaf_cb, znode_cb, priv) 0 #define dbg_old_index_check_init(c, zroot) 0 #define dbg_check_old_index(c, zroot) 0 - #define dbg_check_cats(c) 0 - #define dbg_check_ltab(c) 0 - #define dbg_check_synced_i_size(inode) 0 - #define dbg_check_dir_size(c, dir) 0 - #define dbg_check_tnc(c, x) 0 - #define dbg_check_idx_size(c, idx_size) 0 - #define dbg_check_filesystem(c) 0 - #define dbg_check_heap(c, heap, cat, add_pos) ({}) - #define dbg_check_lprops(c) 0 #define dbg_check_lpt_nodes(c, cnode, row, col) 0 - #define dbg_force_in_the_gaps_enabled 0 #define dbg_force_in_the_gaps() 0 - #define dbg_failure_mode 0 #define dbg_failure_mode_registration(c) ({}) #define dbg_failure_mode_deregistration(c) ({}) diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index e90374b..526c01e 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -165,7 +165,6 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir, } inode->i_ino = ++c->highest_inum; - inode->i_generation = ++c->vfs_gen; /* * The creation sequence number remains with this inode for its * lifetime. All nodes for this inode have a greater sequence number, @@ -220,15 +219,7 @@ static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, err = ubifs_tnc_lookup_nm(c, &key, dent, &dentry->d_name); if (err) { - /* - * Do not hash the direntry if parent 'i_nlink' is zero, because - * this has side-effects - '->delete_inode()' call will not be - * called for the parent orphan inode, because 'd_count' of its - * direntry will stay 1 (it'll be negative direntry I guess) - * and prevent 'iput_final()' until the dentry is destroyed due - * to unmount or memory pressure. - */ - if (err == -ENOENT && dir->i_nlink != 0) { + if (err == -ENOENT) { dbg_gen("not found"); goto done; } @@ -435,7 +426,7 @@ static int ubifs_readdir(struct file *file, void *dirent, filldir_t filldir) while (1) { dbg_gen("feed '%s', ino %llu, new f_pos %#x", - dent->name, le64_to_cpu(dent->inum), + dent->name, (unsigned long long)le64_to_cpu(dent->inum), key_hash_flash(c, &dent->key)); ubifs_assert(dent->ch.sqnum > ubifs_inode(dir)->creat_sqnum); @@ -525,7 +516,7 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir, struct ubifs_inode *dir_ui = ubifs_inode(dir); int err, sz_change = CALC_DENT_SIZE(dentry->d_name.len); struct ubifs_budget_req req = { .new_dent = 1, .dirtied_ino = 2, - .dirtied_ino_d = ui->data_len }; + .dirtied_ino_d = ALIGN(ui->data_len, 8) }; /* * Budget request settings: new direntry, changing the target inode, @@ -596,7 +587,6 @@ static int ubifs_unlink(struct inode *dir, struct dentry *dentry) if (err) { if (err != -ENOSPC) return err; - err = 0; budgeted = 0; } @@ -727,8 +717,7 @@ static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, int mode) struct ubifs_inode *dir_ui = ubifs_inode(dir); struct ubifs_info *c = dir->i_sb->s_fs_info; int err, sz_change = CALC_DENT_SIZE(dentry->d_name.len); - struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1, - .dirtied_ino_d = 1 }; + struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1 }; /* * Budget request settings: new inode, new direntry and changing parent @@ -789,7 +778,8 @@ static int ubifs_mknod(struct inode *dir, struct dentry *dentry, int sz_change = CALC_DENT_SIZE(dentry->d_name.len); int err, devlen = 0; struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1, - .new_ino_d = devlen, .dirtied_ino = 1 }; + .new_ino_d = ALIGN(devlen, 8), + .dirtied_ino = 1 }; /* * Budget request settings: new inode, new direntry and changing parent @@ -863,7 +853,8 @@ static int ubifs_symlink(struct inode *dir, struct dentry *dentry, int err, len = strlen(symname); int sz_change = CALC_DENT_SIZE(dentry->d_name.len); struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1, - .new_ino_d = len, .dirtied_ino = 1 }; + .new_ino_d = ALIGN(len, 8), + .dirtied_ino = 1 }; /* * Budget request settings: new inode, new direntry and changing parent @@ -1012,7 +1003,7 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry, struct ubifs_budget_req req = { .new_dent = 1, .mod_dent = 1, .dirtied_ino = 3 }; struct ubifs_budget_req ino_req = { .dirtied_ino = 1, - .dirtied_ino_d = old_inode_ui->data_len }; + .dirtied_ino_d = ALIGN(old_inode_ui->data_len, 8) }; struct timespec time; /* diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 005a3b8..3d698e2 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -53,6 +53,7 @@ #include "ubifs.h" #include <linux/mount.h> +#include <linux/namei.h> static int read_block(struct inode *inode, void *addr, unsigned int block, struct ubifs_data_node *dn) @@ -792,7 +793,7 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, int err; struct ubifs_budget_req req; loff_t old_size = inode->i_size, new_size = attr->ia_size; - int offset = new_size & (UBIFS_BLOCK_SIZE - 1); + int offset = new_size & (UBIFS_BLOCK_SIZE - 1), budgeted = 1; struct ubifs_inode *ui = ubifs_inode(inode); dbg_gen("ino %lu, size %lld -> %lld", inode->i_ino, old_size, new_size); @@ -810,8 +811,15 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, /* A funny way to budget for truncation node */ req.dirtied_ino_d = UBIFS_TRUN_NODE_SZ; err = ubifs_budget_space(c, &req); - if (err) - return err; + if (err) { + /* + * Treat truncations to zero as deletion and always allow them, + * just like we do for '->unlink()'. + */ + if (new_size || err != -ENOSPC) + return err; + budgeted = 0; + } err = vmtruncate(inode, new_size); if (err) @@ -868,7 +876,12 @@ static int do_truncation(struct ubifs_info *c, struct inode *inode, err = ubifs_jnl_truncate(c, inode, old_size, new_size); mutex_unlock(&ui->ui_mutex); out_budg: - ubifs_release_budget(c, &req); + if (budgeted) + ubifs_release_budget(c, &req); + else { + c->nospace = c->nospace_rp = 0; + smp_wmb(); + } return err; } @@ -889,7 +902,7 @@ static int do_setattr(struct ubifs_info *c, struct inode *inode, loff_t new_size = attr->ia_size; struct ubifs_inode *ui = ubifs_inode(inode); struct ubifs_budget_req req = { .dirtied_ino = 1, - .dirtied_ino_d = ui->data_len }; + .dirtied_ino_d = ALIGN(ui->data_len, 8) }; err = ubifs_budget_space(c, &req); if (err) @@ -940,7 +953,8 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr) struct inode *inode = dentry->d_inode; struct ubifs_info *c = inode->i_sb->s_fs_info; - dbg_gen("ino %lu, ia_valid %#x", inode->i_ino, attr->ia_valid); + dbg_gen("ino %lu, mode %#x, ia_valid %#x", + inode->i_ino, inode->i_mode, attr->ia_valid); err = inode_change_ok(inode, attr); if (err) return err; @@ -1050,7 +1064,7 @@ static int update_mctime(struct ubifs_info *c, struct inode *inode) if (mctime_update_needed(inode, &now)) { int err, release; struct ubifs_budget_req req = { .dirtied_ino = 1, - .dirtied_ino_d = ui->data_len }; + .dirtied_ino_d = ALIGN(ui->data_len, 8) }; err = ubifs_budget_space(c, &req); if (err) @@ -1269,6 +1283,7 @@ struct file_operations ubifs_file_operations = { .fsync = ubifs_fsync, .unlocked_ioctl = ubifs_ioctl, .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, #ifdef CONFIG_COMPAT .compat_ioctl = ubifs_compat_ioctl, #endif diff --git a/fs/ubifs/find.c b/fs/ubifs/find.c index 10394c5..47814cd 100644 --- a/fs/ubifs/find.c +++ b/fs/ubifs/find.c @@ -211,14 +211,8 @@ static const struct ubifs_lprops *scan_for_dirty(struct ubifs_info *c, * dirty index heap, and it falls-back to LPT scanning if the heaps are empty * or do not have an LEB which satisfies the @min_space criteria. * - * Note: - * o LEBs which have less than dead watermark of dirty space are never picked - * by this function; - * - * Returns zero and the LEB properties of - * found dirty LEB in case of success, %-ENOSPC if no dirty LEB was found and a - * negative error code in case of other failures. The returned LEB is marked as - * "taken". + * Note, LEBs which have less than dead watermark of free + dirty space are + * never picked by this function. * * The additional @pick_free argument controls if this function has to return a * free or freeable LEB if one is present. For example, GC must to set it to %1, @@ -231,6 +225,10 @@ static const struct ubifs_lprops *scan_for_dirty(struct ubifs_info *c, * * In addition @pick_free is set to %2 by the recovery process in order to * recover gc_lnum in which case an index LEB must not be returned. + * + * This function returns zero and the LEB properties of found dirty LEB in case + * of success, %-ENOSPC if no dirty LEB was found and a negative error code in + * case of other failures. The returned LEB is marked as "taken". */ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, int min_space, int pick_free) @@ -245,7 +243,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, int lebs, rsvd_idx_lebs = 0; spin_lock(&c->space_lock); - lebs = c->lst.empty_lebs; + lebs = c->lst.empty_lebs + c->idx_gc_cnt; lebs += c->freeable_cnt - c->lst.taken_empty_lebs; /* @@ -290,9 +288,14 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, idx_lp = idx_heap->arr[0]; sum = idx_lp->free + idx_lp->dirty; /* - * Since we reserve twice as more space for the index than it + * Since we reserve thrice as much space for the index than it * actually takes, it does not make sense to pick indexing LEBs - * with less than half LEB of dirty space. + * with less than, say, half LEB of dirty space. May be half is + * not the optimal boundary - this should be tested and + * checked. This boundary should determine how much we use + * in-the-gaps to consolidate the index comparing to how much + * we use garbage collector to consolidate it. The "half" + * criteria just feels to be fine. */ if (sum < min_space || sum < c->half_leb_size) idx_lp = NULL; @@ -312,7 +315,7 @@ int ubifs_find_dirty_leb(struct ubifs_info *c, struct ubifs_lprops *ret_lp, lp = idx_lp; if (lp) { - ubifs_assert(lp->dirty >= c->dead_wm); + ubifs_assert(lp->free + lp->dirty >= c->dead_wm); goto found; } @@ -504,7 +507,6 @@ int ubifs_find_free_space(struct ubifs_info *c, int min_space, int *free, rsvd_idx_lebs = 0; lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - c->lst.taken_empty_lebs; - ubifs_assert(lebs + c->lst.idx_lebs >= c->min_idx_lebs); if (rsvd_idx_lebs < lebs) /* * OK to allocate an empty LEB, but we still don't want to go diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index d0f3dac..02aba36 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -334,15 +334,21 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp) err = move_nodes(c, sleb); if (err) - goto out; + goto out_inc_seq; err = gc_sync_wbufs(c); if (err) - goto out; + goto out_inc_seq; err = ubifs_change_one_lp(c, lnum, c->leb_size, 0, 0, 0, 0); if (err) - goto out; + goto out_inc_seq; + + /* Allow for races with TNC */ + c->gced_lnum = lnum; + smp_wmb(); + c->gc_seq += 1; + smp_wmb(); if (c->gc_lnum == -1) { c->gc_lnum = lnum; @@ -363,6 +369,14 @@ int ubifs_garbage_collect_leb(struct ubifs_info *c, struct ubifs_lprops *lp) out: ubifs_scan_destroy(sleb); return err; + +out_inc_seq: + /* We may have moved at least some nodes so allow for races with TNC */ + c->gced_lnum = lnum; + smp_wmb(); + c->gc_seq += 1; + smp_wmb(); + goto out; } /** diff --git a/fs/ubifs/io.c b/fs/ubifs/io.c index 3374f91..054363f 100644 --- a/fs/ubifs/io.c +++ b/fs/ubifs/io.c @@ -54,6 +54,20 @@ #include "ubifs.h" /** + * ubifs_ro_mode - switch UBIFS to read read-only mode. + * @c: UBIFS file-system description object + * @err: error code which is the reason of switching to R/O mode + */ +void ubifs_ro_mode(struct ubifs_info *c, int err) +{ + if (!c->ro_media) { + c->ro_media = 1; + ubifs_warn("switched to read-only mode, error %d", err); + dbg_dump_stack(); + } +} + +/** * ubifs_check_node - check node. * @c: UBIFS file-system description object * @buf: node to check diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index 283155a..22993f8 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -447,13 +447,11 @@ static int get_dent_type(int mode) * @ino: buffer in which to pack inode node * @inode: inode to pack * @last: indicates the last node of the group - * @last_reference: non-zero if this is a deletion inode */ static void pack_inode(struct ubifs_info *c, struct ubifs_ino_node *ino, - const struct inode *inode, int last, - int last_reference) + const struct inode *inode, int last) { - int data_len = 0; + int data_len = 0, last_reference = !inode->i_nlink; struct ubifs_inode *ui = ubifs_inode(inode); ino->ch.node_type = UBIFS_INO_NODE; @@ -596,9 +594,9 @@ int ubifs_jnl_update(struct ubifs_info *c, const struct inode *dir, ubifs_prep_grp_node(c, dent, dlen, 0); ino = (void *)dent + aligned_dlen; - pack_inode(c, ino, inode, 0, last_reference); + pack_inode(c, ino, inode, 0); ino = (void *)ino + aligned_ilen; - pack_inode(c, ino, dir, 1, 0); + pack_inode(c, ino, dir, 1); if (last_reference) { err = ubifs_add_orphan(c, inode->i_ino); @@ -606,6 +604,7 @@ int ubifs_jnl_update(struct ubifs_info *c, const struct inode *dir, release_head(c, BASEHD); goto out_finish; } + ui->del_cmtno = c->cmt_no; } err = write_head(c, BASEHD, dent, len, &lnum, &dent_offs, sync); @@ -750,30 +749,25 @@ out_free: * ubifs_jnl_write_inode - flush inode to the journal. * @c: UBIFS file-system description object * @inode: inode to flush - * @deletion: inode has been deleted * * This function writes inode @inode to the journal. If the inode is * synchronous, it also synchronizes the write-buffer. Returns zero in case of * success and a negative error code in case of failure. */ -int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode, - int deletion) +int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode) { - int err, len, lnum, offs, sync = 0; + int err, lnum, offs; struct ubifs_ino_node *ino; struct ubifs_inode *ui = ubifs_inode(inode); + int sync = 0, len = UBIFS_INO_NODE_SZ, last_reference = !inode->i_nlink; - dbg_jnl("ino %lu%s", inode->i_ino, - deletion ? " (last reference)" : ""); - if (deletion) - ubifs_assert(inode->i_nlink == 0); + dbg_jnl("ino %lu, nlink %u", inode->i_ino, inode->i_nlink); - len = UBIFS_INO_NODE_SZ; /* * If the inode is being deleted, do not write the attached data. No * need to synchronize the write-buffer either. */ - if (!deletion) { + if (!last_reference) { len += ui->data_len; sync = IS_SYNC(inode); } @@ -786,7 +780,7 @@ int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode, if (err) goto out_free; - pack_inode(c, ino, inode, 1, deletion); + pack_inode(c, ino, inode, 1); err = write_head(c, BASEHD, ino, len, &lnum, &offs, sync); if (err) goto out_release; @@ -795,7 +789,7 @@ int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode, inode->i_ino); release_head(c, BASEHD); - if (deletion) { + if (last_reference) { err = ubifs_tnc_remove_ino(c, inode->i_ino); if (err) goto out_ro; @@ -828,6 +822,65 @@ out_free: } /** + * ubifs_jnl_delete_inode - delete an inode. + * @c: UBIFS file-system description object + * @inode: inode to delete + * + * This function deletes inode @inode which includes removing it from orphans, + * deleting it from TNC and, in some cases, writing a deletion inode to the + * journal. + * + * When regular file inodes are unlinked or a directory inode is removed, the + * 'ubifs_jnl_update()' function writes a corresponding deletion inode and + * direntry to the media, and adds the inode to orphans. After this, when the + * last reference to this inode has been dropped, this function is called. In + * general, it has to write one more deletion inode to the media, because if + * a commit happened between 'ubifs_jnl_update()' and + * 'ubifs_jnl_delete_inode()', the deletion inode is not in the journal + * anymore, and in fact it might not be on the flash anymore, because it might + * have been garbage-collected already. And for optimization reasons UBIFS does + * not read the orphan area if it has been unmounted cleanly, so it would have + * no indication in the journal that there is a deleted inode which has to be + * removed from TNC. + * + * However, if there was no commit between 'ubifs_jnl_update()' and + * 'ubifs_jnl_delete_inode()', then there is no need to write the deletion + * inode to the media for the second time. And this is quite a typical case. + * + * This function returns zero in case of success and a negative error code in + * case of failure. + */ +int ubifs_jnl_delete_inode(struct ubifs_info *c, const struct inode *inode) +{ + int err; + struct ubifs_inode *ui = ubifs_inode(inode); + + ubifs_assert(inode->i_nlink == 0); + + if (ui->del_cmtno != c->cmt_no) + /* A commit happened for sure */ + return ubifs_jnl_write_inode(c, inode); + + down_read(&c->commit_sem); + /* + * Check commit number again, because the first test has been done + * without @c->commit_sem, so a commit might have happened. + */ + if (ui->del_cmtno != c->cmt_no) { + up_read(&c->commit_sem); + return ubifs_jnl_write_inode(c, inode); + } + + err = ubifs_tnc_remove_ino(c, inode->i_ino); + if (err) + ubifs_ro_mode(c, err); + else + ubifs_delete_orphan(c, inode->i_ino); + up_read(&c->commit_sem); + return err; +} + +/** * ubifs_jnl_rename - rename a directory entry. * @c: UBIFS file-system description object * @old_dir: parent inode of directory entry to rename @@ -917,16 +970,16 @@ int ubifs_jnl_rename(struct ubifs_info *c, const struct inode *old_dir, p = (void *)dent2 + aligned_dlen2; if (new_inode) { - pack_inode(c, p, new_inode, 0, last_reference); + pack_inode(c, p, new_inode, 0); p += ALIGN(ilen, 8); } if (!move) - pack_inode(c, p, old_dir, 1, 0); + pack_inode(c, p, old_dir, 1); else { - pack_inode(c, p, old_dir, 0, 0); + pack_inode(c, p, old_dir, 0); p += ALIGN(plen, 8); - pack_inode(c, p, new_dir, 1, 0); + pack_inode(c, p, new_dir, 1); } if (last_reference) { @@ -935,6 +988,7 @@ int ubifs_jnl_rename(struct ubifs_info *c, const struct inode *old_dir, release_head(c, BASEHD); goto out_finish; } + new_ui->del_cmtno = c->cmt_no; } err = write_head(c, BASEHD, dent, len, &lnum, &offs, sync); @@ -1131,7 +1185,7 @@ int ubifs_jnl_truncate(struct ubifs_info *c, const struct inode *inode, if (err) goto out_free; - pack_inode(c, ino, inode, 0, 0); + pack_inode(c, ino, inode, 0); ubifs_prep_grp_node(c, trun, UBIFS_TRUN_NODE_SZ, dlen ? 0 : 1); if (dlen) ubifs_prep_grp_node(c, dn, dlen, 1); @@ -1251,9 +1305,9 @@ int ubifs_jnl_delete_xattr(struct ubifs_info *c, const struct inode *host, ubifs_prep_grp_node(c, xent, xlen, 0); ino = (void *)xent + aligned_xlen; - pack_inode(c, ino, inode, 0, 1); + pack_inode(c, ino, inode, 0); ino = (void *)ino + UBIFS_INO_NODE_SZ; - pack_inode(c, ino, host, 1, 0); + pack_inode(c, ino, host, 1); err = write_head(c, BASEHD, xent, len, &lnum, &xent_offs, sync); if (!sync && !err) @@ -1320,7 +1374,7 @@ int ubifs_jnl_change_xattr(struct ubifs_info *c, const struct inode *inode, const struct inode *host) { int err, len1, len2, aligned_len, aligned_len1, lnum, offs; - struct ubifs_inode *host_ui = ubifs_inode(inode); + struct ubifs_inode *host_ui = ubifs_inode(host); struct ubifs_ino_node *ino; union ubifs_key key; int sync = IS_DIRSYNC(host); @@ -1344,8 +1398,8 @@ int ubifs_jnl_change_xattr(struct ubifs_info *c, const struct inode *inode, if (err) goto out_free; - pack_inode(c, ino, host, 0, 0); - pack_inode(c, (void *)ino + aligned_len1, inode, 1, 0); + pack_inode(c, ino, host, 0); + pack_inode(c, (void *)ino + aligned_len1, inode, 1); err = write_head(c, BASEHD, ino, aligned_len, &lnum, &offs, 0); if (!sync && !err) { diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c index 36857b9..3e0aa73 100644 --- a/fs/ubifs/log.c +++ b/fs/ubifs/log.c @@ -317,6 +317,8 @@ int ubifs_add_bud_to_log(struct ubifs_info *c, int jhead, int lnum, int offs) return 0; out_unlock: + if (err != -EAGAIN) + ubifs_ro_mode(c, err); mutex_unlock(&c->log_mutex); kfree(ref); kfree(bud); @@ -410,7 +412,7 @@ int ubifs_log_start_commit(struct ubifs_info *c, int *ltail_lnum) return -ENOMEM; cs->ch.node_type = UBIFS_CS_NODE; - cs->cmt_no = cpu_to_le64(c->cmt_no + 1); + cs->cmt_no = cpu_to_le64(c->cmt_no); ubifs_prepare_node(c, cs, UBIFS_CS_NODE_SZ, 0); /* diff --git a/fs/ubifs/misc.h b/fs/ubifs/misc.h index 4beccfc..4c12a92 100644 --- a/fs/ubifs/misc.h +++ b/fs/ubifs/misc.h @@ -80,20 +80,6 @@ static inline struct ubifs_inode *ubifs_inode(const struct inode *inode) } /** - * ubifs_ro_mode - switch UBIFS to read read-only mode. - * @c: UBIFS file-system description object - * @err: error code which is the reason of switching to R/O mode - */ -static inline void ubifs_ro_mode(struct ubifs_info *c, int err) -{ - if (!c->ro_media) { - c->ro_media = 1; - ubifs_warn("switched to read-only mode, error %d", err); - dbg_dump_stack(); - } -} - -/** * ubifs_compr_present - check if compressor was compiled in. * @compr_type: compressor type to check * @@ -298,38 +284,6 @@ static inline void *ubifs_idx_key(const struct ubifs_info *c, } /** - * ubifs_reported_space - calculate reported free space. - * @c: the UBIFS file-system description object - * @free: amount of free space - * - * This function calculates amount of free space which will be reported to - * user-space. User-space application tend to expect that if the file-system - * (e.g., via the 'statfs()' call) reports that it has N bytes available, they - * are able to write a file of size N. UBIFS attaches node headers to each data - * node and it has to write indexind nodes as well. This introduces additional - * overhead, and UBIFS it has to report sligtly less free space to meet the - * above expectetion. - * - * This function assumes free space is made up of uncompressed data nodes and - * full index nodes (one per data node, doubled because we always allow enough - * space to write the index twice). - * - * Note, the calculation is pessimistic, which means that most of the time - * UBIFS reports less space than it actually has. - */ -static inline long long ubifs_reported_space(const struct ubifs_info *c, - uint64_t free) -{ - int divisor, factor; - - divisor = UBIFS_MAX_DATA_NODE_SZ + (c->max_idx_node_sz << 1); - factor = UBIFS_MAX_DATA_NODE_SZ - UBIFS_DATA_NODE_SZ; - do_div(free, divisor); - - return free * factor; -} - -/** * ubifs_current_time - round current time to time granularity. * @inode: inode */ @@ -339,4 +293,21 @@ static inline struct timespec ubifs_current_time(struct inode *inode) current_fs_time(inode->i_sb) : CURRENT_TIME_SEC; } +/** + * ubifs_tnc_lookup - look up a file-system node. + * @c: UBIFS file-system description object + * @key: node key to lookup + * @node: the node is returned here + * + * This function look up and reads node with key @key. The caller has to make + * sure the @node buffer is large enough to fit the node. Returns zero in case + * of success, %-ENOENT if the node was not found, and a negative error code in + * case of failure. + */ +static inline int ubifs_tnc_lookup(struct ubifs_info *c, + const union ubifs_key *key, void *node) +{ + return ubifs_tnc_locate(c, key, node, NULL, NULL); +} + #endif /* __UBIFS_MISC_H__ */ diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c index 3afeb92..02d3462 100644 --- a/fs/ubifs/orphan.c +++ b/fs/ubifs/orphan.c @@ -310,10 +310,10 @@ static int write_orph_node(struct ubifs_info *c, int atomic) c->cmt_orphans -= cnt; spin_unlock(&c->orphan_lock); if (c->cmt_orphans) - orph->cmt_no = cpu_to_le64(c->cmt_no + 1); + orph->cmt_no = cpu_to_le64(c->cmt_no); else /* Mark the last node of the commit */ - orph->cmt_no = cpu_to_le64((c->cmt_no + 1) | (1ULL << 63)); + orph->cmt_no = cpu_to_le64((c->cmt_no) | (1ULL << 63)); ubifs_assert(c->ohead_offs + len <= c->leb_size); ubifs_assert(c->ohead_lnum >= c->orph_first); ubifs_assert(c->ohead_lnum <= c->orph_last); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 00eb9c6..3f49020 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -30,7 +30,6 @@ #include <linux/slab.h> #include <linux/module.h> #include <linux/ctype.h> -#include <linux/random.h> #include <linux/kthread.h> #include <linux/parser.h> #include <linux/seq_file.h> @@ -149,7 +148,7 @@ struct inode *ubifs_iget(struct super_block *sb, unsigned long inum) if (err) goto out_invalid; - /* Disable readahead */ + /* Disable read-ahead */ inode->i_mapping->backing_dev_info = &c->bdi; switch (inode->i_mode & S_IFMT) { @@ -278,7 +277,7 @@ static void ubifs_destroy_inode(struct inode *inode) */ static int ubifs_write_inode(struct inode *inode, int wait) { - int err; + int err = 0; struct ubifs_info *c = inode->i_sb->s_fs_info; struct ubifs_inode *ui = ubifs_inode(inode); @@ -299,10 +298,18 @@ static int ubifs_write_inode(struct inode *inode, int wait) return 0; } - dbg_gen("inode %lu", inode->i_ino); - err = ubifs_jnl_write_inode(c, inode, 0); - if (err) - ubifs_err("can't write inode %lu, error %d", inode->i_ino, err); + /* + * As an optimization, do not write orphan inodes to the media just + * because this is not needed. + */ + dbg_gen("inode %lu, mode %#x, nlink %u", + inode->i_ino, (int)inode->i_mode, inode->i_nlink); + if (inode->i_nlink) { + err = ubifs_jnl_write_inode(c, inode); + if (err) + ubifs_err("can't write inode %lu, error %d", + inode->i_ino, err); + } ui->dirty = 0; mutex_unlock(&ui->ui_mutex); @@ -314,8 +321,9 @@ static void ubifs_delete_inode(struct inode *inode) { int err; struct ubifs_info *c = inode->i_sb->s_fs_info; + struct ubifs_inode *ui = ubifs_inode(inode); - if (ubifs_inode(inode)->xattr) + if (ui->xattr) /* * Extended attribute inode deletions are fully handled in * 'ubifs_removexattr()'. These inodes are special and have @@ -323,7 +331,7 @@ static void ubifs_delete_inode(struct inode *inode) */ goto out; - dbg_gen("inode %lu", inode->i_ino); + dbg_gen("inode %lu, mode %#x", inode->i_ino, (int)inode->i_mode); ubifs_assert(!atomic_read(&inode->i_count)); ubifs_assert(inode->i_nlink == 0); @@ -331,15 +339,19 @@ static void ubifs_delete_inode(struct inode *inode) if (is_bad_inode(inode)) goto out; - ubifs_inode(inode)->ui_size = inode->i_size = 0; - err = ubifs_jnl_write_inode(c, inode, 1); + ui->ui_size = inode->i_size = 0; + err = ubifs_jnl_delete_inode(c, inode); if (err) /* * Worst case we have a lost orphan inode wasting space, so a - * simple error message is ok here. + * simple error message is OK here. */ - ubifs_err("can't write inode %lu, error %d", inode->i_ino, err); + ubifs_err("can't delete inode %lu, error %d", + inode->i_ino, err); + out: + if (ui->dirty) + ubifs_release_dirty_inode_budget(c, ui); clear_inode(inode); } @@ -358,8 +370,9 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct ubifs_info *c = dentry->d_sb->s_fs_info; unsigned long long free; + __le32 *uuid = (__le32 *)c->uuid; - free = ubifs_budg_get_free_space(c); + free = ubifs_get_free_space(c); dbg_gen("free space %lld bytes (%lld blocks)", free, free >> UBIFS_BLOCK_SHIFT); @@ -374,7 +387,8 @@ static int ubifs_statfs(struct dentry *dentry, struct kstatfs *buf) buf->f_files = 0; buf->f_ffree = 0; buf->f_namelen = UBIFS_MAX_NLEN; - + buf->f_fsid.val[0] = le32_to_cpu(uuid[0]) ^ le32_to_cpu(uuid[2]); + buf->f_fsid.val[1] = le32_to_cpu(uuid[1]) ^ le32_to_cpu(uuid[3]); return 0; } @@ -518,6 +532,12 @@ static int init_constants_early(struct ubifs_info *c) c->dead_wm = ALIGN(MIN_WRITE_SZ, c->min_io_size); c->dark_wm = ALIGN(UBIFS_MAX_NODE_SZ, c->min_io_size); + /* + * Calculate how many bytes would be wasted at the end of LEB if it was + * fully filled with data nodes of maximum size. This is used in + * calculations when reporting free space. + */ + c->leb_overhead = c->leb_size % UBIFS_MAX_DATA_NODE_SZ; return 0; } @@ -635,13 +655,11 @@ static int init_constants_late(struct ubifs_info *c) * internally because it does not make much sense for UBIFS, but it is * necessary to report something for the 'statfs()' call. * - * Subtract the LEB reserved for GC and the LEB which is reserved for - * deletions. - * - * Review 'ubifs_calc_available()' if changing this calculation. + * Subtract the LEB reserved for GC, the LEB which is reserved for + * deletions, and assume only one journal head is available. */ - tmp64 = c->main_lebs - 2; - tmp64 *= (uint64_t)c->leb_size - c->dark_wm; + tmp64 = c->main_lebs - 2 - c->jhead_cnt + 1; + tmp64 *= (uint64_t)c->leb_size - c->leb_overhead; tmp64 = ubifs_reported_space(c, tmp64); c->block_cnt = tmp64 >> UBIFS_BLOCK_SHIFT; @@ -1006,14 +1024,13 @@ static int mount_ubifs(struct ubifs_info *c) goto out_dereg; } + sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id); if (!mounted_read_only) { err = alloc_wbufs(c); if (err) goto out_cbuf; /* Create background thread */ - sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, - c->vi.vol_id); c->bgt = kthread_create(ubifs_bg_thread, c, c->bgt_name); if (!c->bgt) c->bgt = ERR_PTR(-EINVAL); @@ -1122,8 +1139,8 @@ static int mount_ubifs(struct ubifs_info *c) if (err) goto out_infos; - ubifs_msg("mounted UBI device %d, volume %d", c->vi.ubi_num, - c->vi.vol_id); + ubifs_msg("mounted UBI device %d, volume %d, name \"%s\"", + c->vi.ubi_num, c->vi.vol_id, c->vi.name); if (mounted_read_only) ubifs_msg("mounted read-only"); x = (long long)c->main_lebs * c->leb_size; @@ -1469,6 +1486,7 @@ static void ubifs_put_super(struct super_block *sb) */ 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); /* @@ -1657,7 +1675,6 @@ static int ubifs_fill_super(struct super_block *sb, void *data, int silent) INIT_LIST_HEAD(&c->orph_new); c->highest_inum = UBIFS_FIRST_INO; - get_random_bytes(&c->vfs_gen, sizeof(int)); c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM; ubi_get_volume_info(ubi, &c->vi); @@ -1671,10 +1688,10 @@ static int ubifs_fill_super(struct super_block *sb, void *data, int silent) } /* - * UBIFS provids 'backing_dev_info' in order to disable readahead. For + * UBIFS provides 'backing_dev_info' in order to disable read-ahead. For * UBIFS, I/O is not deferred, it is done immediately in readpage, * which means the user would have to wait not just for their own I/O - * but the readahead I/O as well i.e. completely pointless. + * but the read-ahead I/O as well i.e. completely pointless. * * Read-ahead will be disabled because @c->bdi.ra_pages is 0. */ @@ -1841,7 +1858,7 @@ static struct file_system_type ubifs_fs_type = { /* * Inode slab cache constructor. */ -static void inode_slab_ctor(struct kmem_cache *cachep, void *obj) +static void inode_slab_ctor(void *obj) { struct ubifs_inode *ui = obj; inode_init_once(&ui->vfs_inode); diff --git a/fs/ubifs/tnc.c b/fs/ubifs/tnc.c index e909f4a..7634c59 100644 --- a/fs/ubifs/tnc.c +++ b/fs/ubifs/tnc.c @@ -506,7 +506,7 @@ static int fallible_read_node(struct ubifs_info *c, const union ubifs_key *key, if (keys_cmp(c, key, &node_key) != 0) ret = 0; } - if (ret == 0) + if (ret == 0 && c->replaying) dbg_mnt("dangling branch LEB %d:%d len %d, key %s", zbr->lnum, zbr->offs, zbr->len, DBGKEY(key)); return ret; @@ -1382,50 +1382,39 @@ static int lookup_level0_dirty(struct ubifs_info *c, const union ubifs_key *key, } /** - * ubifs_tnc_lookup - look up a file-system node. + * maybe_leb_gced - determine if a LEB may have been garbage collected. * @c: UBIFS file-system description object - * @key: node key to lookup - * @node: the node is returned here + * @lnum: LEB number + * @gc_seq1: garbage collection sequence number * - * This function look up and reads node with key @key. The caller has to make - * sure the @node buffer is large enough to fit the node. Returns zero in case - * of success, %-ENOENT if the node was not found, and a negative error code in - * case of failure. + * This function determines if @lnum may have been garbage collected since + * sequence number @gc_seq1. If it may have been then %1 is returned, otherwise + * %0 is returned. */ -int ubifs_tnc_lookup(struct ubifs_info *c, const union ubifs_key *key, - void *node) +static int maybe_leb_gced(struct ubifs_info *c, int lnum, int gc_seq1) { - int found, n, err; - struct ubifs_znode *znode; - struct ubifs_zbranch zbr, *zt; + int gc_seq2, gced_lnum; - mutex_lock(&c->tnc_mutex); - found = ubifs_lookup_level0(c, key, &znode, &n); - if (!found) { - err = -ENOENT; - goto out; - } else if (found < 0) { - err = found; - goto out; - } - zt = &znode->zbranch[n]; - if (is_hash_key(c, key)) { - /* - * In this case the leaf node cache gets used, so we pass the - * address of the zbranch and keep the mutex locked - */ - err = tnc_read_node_nm(c, zt, node); - goto out; - } - zbr = znode->zbranch[n]; - mutex_unlock(&c->tnc_mutex); - - err = ubifs_tnc_read_node(c, &zbr, node); - return err; - -out: - mutex_unlock(&c->tnc_mutex); - return err; + gced_lnum = c->gced_lnum; + smp_rmb(); + gc_seq2 = c->gc_seq; + /* Same seq means no GC */ + if (gc_seq1 == gc_seq2) + return 0; + /* Different by more than 1 means we don't know */ + if (gc_seq1 + 1 != gc_seq2) + return 1; + /* + * We have seen the sequence number has increased by 1. Now we need to + * be sure we read the right LEB number, so read it again. + */ + smp_rmb(); + if (gced_lnum != c->gced_lnum) + return 1; + /* Finally we can check lnum */ + if (gced_lnum == lnum) + return 1; + return 0; } /** @@ -1436,16 +1425,19 @@ out: * @lnum: LEB number is returned here * @offs: offset is returned here * - * This function is the same as 'ubifs_tnc_lookup()' but it returns the node - * location also. See 'ubifs_tnc_lookup()'. + * This function look up and reads node with key @key. The caller has to make + * sure the @node buffer is large enough to fit the node. Returns zero in case + * of success, %-ENOENT if the node was not found, and a negative error code in + * case of failure. The node location can be returned in @lnum and @offs. */ int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, void *node, int *lnum, int *offs) { - int found, n, err; + int found, n, err, safely = 0, gc_seq1; struct ubifs_znode *znode; struct ubifs_zbranch zbr, *zt; +again: mutex_lock(&c->tnc_mutex); found = ubifs_lookup_level0(c, key, &znode, &n); if (!found) { @@ -1456,24 +1448,43 @@ int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, goto out; } zt = &znode->zbranch[n]; + if (lnum) { + *lnum = zt->lnum; + *offs = zt->offs; + } if (is_hash_key(c, key)) { /* * In this case the leaf node cache gets used, so we pass the * address of the zbranch and keep the mutex locked */ - *lnum = zt->lnum; - *offs = zt->offs; err = tnc_read_node_nm(c, zt, node); goto out; } + if (safely) { + err = ubifs_tnc_read_node(c, zt, node); + goto out; + } + /* Drop the TNC mutex prematurely and race with garbage collection */ zbr = znode->zbranch[n]; + gc_seq1 = c->gc_seq; mutex_unlock(&c->tnc_mutex); - *lnum = zbr.lnum; - *offs = zbr.offs; + if (ubifs_get_wbuf(c, zbr.lnum)) { + /* We do not GC journal heads */ + err = ubifs_tnc_read_node(c, &zbr, node); + return err; + } - err = ubifs_tnc_read_node(c, &zbr, node); - return err; + err = fallible_read_node(c, key, &zbr, node); + if (err <= 0 || maybe_leb_gced(c, zbr.lnum, gc_seq1)) { + /* + * The node may have been GC'ed out from under us so try again + * while keeping the TNC mutex locked. + */ + safely = 1; + goto again; + } + return 0; out: mutex_unlock(&c->tnc_mutex); @@ -1498,7 +1509,6 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, { int found, n, err; struct ubifs_znode *znode; - struct ubifs_zbranch zbr; dbg_tnc("name '%.*s' key %s", nm->len, nm->name, DBGKEY(key)); mutex_lock(&c->tnc_mutex); @@ -1522,11 +1532,7 @@ static int do_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, goto out_unlock; } - zbr = znode->zbranch[n]; - mutex_unlock(&c->tnc_mutex); - - err = tnc_read_node_nm(c, &zbr, node); - return err; + err = tnc_read_node_nm(c, &znode->zbranch[n], node); out_unlock: mutex_unlock(&c->tnc_mutex); diff --git a/fs/ubifs/tnc_commit.c b/fs/ubifs/tnc_commit.c index 8117e65..8ac76b1 100644 --- a/fs/ubifs/tnc_commit.c +++ b/fs/ubifs/tnc_commit.c @@ -372,26 +372,25 @@ static int layout_in_gaps(struct ubifs_info *c, int cnt) written = layout_leb_in_gaps(c, p); if (written < 0) { err = written; - if (err == -ENOSPC) { - 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); - dbg_dump_lprops(c); - } - /* Try to commit anyway */ - err = 0; - break; + if (err != -ENOSPC) { + kfree(c->gap_lebs); + c->gap_lebs = NULL; + return err; } - kfree(c->gap_lebs); - c->gap_lebs = NULL; - return err; + 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); + dbg_dump_lprops(c); + } + /* Try to commit anyway */ + err = 0; + break; } p++; cnt -= written; diff --git a/fs/ubifs/ubifs-media.h b/fs/ubifs/ubifs-media.h index 0cc7da9..a9ecbd9 100644 --- a/fs/ubifs/ubifs-media.h +++ b/fs/ubifs/ubifs-media.h @@ -87,7 +87,7 @@ #define UBIFS_SK_LEN 8 /* Minimum index tree fanout */ -#define UBIFS_MIN_FANOUT 2 +#define UBIFS_MIN_FANOUT 3 /* Maximum number of levels in UBIFS indexing B-tree */ #define UBIFS_MAX_LEVELS 512 @@ -228,10 +228,10 @@ enum { /* Minimum number of orphan area logical eraseblocks */ #define UBIFS_MIN_ORPH_LEBS 1 /* - * Minimum number of main area logical eraseblocks (buds, 2 for the index, 1 + * Minimum number of main area logical eraseblocks (buds, 3 for the index, 1 * for GC, 1 for deletions, and at least 1 for committed data). */ -#define UBIFS_MIN_MAIN_LEBS (UBIFS_MIN_BUD_LEBS + 5) +#define UBIFS_MIN_MAIN_LEBS (UBIFS_MIN_BUD_LEBS + 6) /* Minimum number of logical eraseblocks */ #define UBIFS_MIN_LEB_CNT (UBIFS_SB_LEBS + UBIFS_MST_LEBS + \ diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index e4f89f2..17c620b 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -20,8 +20,6 @@ * Adrian Hunter */ -/* Implementation version 0.7 */ - #ifndef __UBIFS_H__ #define __UBIFS_H__ @@ -322,6 +320,8 @@ struct ubifs_gced_idx_leb { * struct ubifs_inode - UBIFS in-memory inode description. * @vfs_inode: VFS inode description object * @creat_sqnum: sequence number at time of creation + * @del_cmtno: commit number corresponding to the time the inode was deleted, + * protected by @c->commit_sem; * @xattr_size: summarized size of all extended attributes in bytes * @xattr_cnt: count of extended attributes this inode has * @xattr_names: sum of lengths of all extended attribute names belonging to @@ -373,6 +373,7 @@ struct ubifs_gced_idx_leb { struct ubifs_inode { struct inode vfs_inode; unsigned long long creat_sqnum; + unsigned long long del_cmtno; unsigned int xattr_size; unsigned int xattr_cnt; unsigned int xattr_names; @@ -779,7 +780,7 @@ struct ubifs_compressor { /** * struct ubifs_budget_req - budget requirements of an operation. * - * @fast: non-zero if the budgeting should try to aquire budget quickly and + * @fast: non-zero if the budgeting should try to acquire budget quickly and * should not try to call write-back * @recalculate: non-zero if @idx_growth, @data_growth, and @dd_growth fields * have to be re-calculated @@ -805,21 +806,31 @@ struct ubifs_compressor { * An inode may contain 4KiB of data at max., thus the widths of @new_ino_d * is 13 bits, and @dirtied_ino_d - 15, because up to 4 inodes may be made * dirty by the re-name operation. + * + * Note, UBIFS aligns node lengths to 8-bytes boundary, so the requester has to + * make sure the amount of inode data which contribute to @new_ino_d and + * @dirtied_ino_d fields are aligned. */ struct ubifs_budget_req { unsigned int fast:1; unsigned int recalculate:1; +#ifndef UBIFS_DEBUG unsigned int new_page:1; unsigned int dirtied_page:1; unsigned int new_dent:1; unsigned int mod_dent:1; unsigned int new_ino:1; unsigned int new_ino_d:13; -#ifndef UBIFS_DEBUG unsigned int dirtied_ino:4; unsigned int dirtied_ino_d:15; #else /* Not bit-fields to check for overflows */ + unsigned int new_page; + unsigned int dirtied_page; + unsigned int new_dent; + unsigned int mod_dent; + unsigned int new_ino; + unsigned int new_ino_d; unsigned int dirtied_ino; unsigned int dirtied_ino_d; #endif @@ -860,13 +871,13 @@ struct ubifs_mount_opts { * struct ubifs_info - UBIFS file-system description data structure * (per-superblock). * @vfs_sb: VFS @struct super_block object - * @bdi: backing device info object to make VFS happy and disable readahead + * @bdi: backing device info object to make VFS happy and disable read-ahead * * @highest_inum: highest used inode number - * @vfs_gen: VFS inode generation counter * @max_sqnum: current global sequence number - * @cmt_no: commit number (last successfully completed commit) - * @cnt_lock: protects @highest_inum, @vfs_gen, and @max_sqnum counters + * @cmt_no: commit number of the last successfully completed commit, protected + * by @commit_sem + * @cnt_lock: protects @highest_inum and @max_sqnum counters * @fmt_version: UBIFS on-flash format version * @uuid: UUID from super block * @@ -984,6 +995,9 @@ struct ubifs_mount_opts { * @max_idx_node_sz: maximum indexing node aligned on 8-bytes boundary * @max_inode_sz: maximum possible inode size in bytes * @max_znode_sz: size of znode in bytes + * + * @leb_overhead: how many bytes are wasted in an LEB when it is filled with + * data nodes of maximum size - used in free space reporting * @dead_wm: LEB dead space watermark * @dark_wm: LEB dark space watermark * @block_cnt: count of 4KiB blocks on the FS @@ -1017,6 +1031,8 @@ struct ubifs_mount_opts { * @sbuf: a buffer of LEB size used by GC and replay for scanning * @idx_gc: list of index LEBs that have been garbage collected * @idx_gc_cnt: number of elements on the idx_gc list + * @gc_seq: incremented for every non-index LEB garbage collected + * @gced_lnum: last non-index LEB that was garbage collected * * @infos_list: links all 'ubifs_info' objects * @umount_mutex: serializes shrinker and un-mount @@ -1103,7 +1119,6 @@ struct ubifs_info { struct backing_dev_info bdi; ino_t highest_inum; - unsigned int vfs_gen; unsigned long long max_sqnum; unsigned long long cmt_no; spinlock_t cnt_lock; @@ -1214,6 +1229,8 @@ struct ubifs_info { int max_idx_node_sz; long long max_inode_sz; int max_znode_sz; + + int leb_overhead; int dead_wm; int dark_wm; int block_cnt; @@ -1247,6 +1264,8 @@ struct ubifs_info { void *sbuf; struct list_head idx_gc; int idx_gc_cnt; + volatile int gc_seq; + volatile int gced_lnum; struct list_head infos_list; struct mutex umount_mutex; @@ -1346,6 +1365,7 @@ extern struct backing_dev_info ubifs_backing_dev_info; extern struct ubifs_compressor *ubifs_compressors[UBIFS_COMPR_TYPES_CNT]; /* io.c */ +void ubifs_ro_mode(struct ubifs_info *c, int err); int ubifs_wbuf_write_nolock(struct ubifs_wbuf *wbuf, void *buf, int len); int ubifs_wbuf_seek_nolock(struct ubifs_wbuf *wbuf, int lnum, int offs, int dtype); @@ -1399,8 +1419,8 @@ int ubifs_jnl_update(struct ubifs_info *c, const struct inode *dir, int deletion, int xent); int ubifs_jnl_write_data(struct ubifs_info *c, const struct inode *inode, const union ubifs_key *key, const void *buf, int len); -int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode, - int last_reference); +int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode); +int ubifs_jnl_delete_inode(struct ubifs_info *c, const struct inode *inode); int ubifs_jnl_rename(struct ubifs_info *c, const struct inode *old_dir, const struct dentry *old_dentry, const struct inode *new_dir, @@ -1423,9 +1443,10 @@ void ubifs_release_ino_dirty(struct ubifs_info *c, struct inode *inode, struct ubifs_budget_req *req); void ubifs_cancel_ino_op(struct ubifs_info *c, struct inode *inode, struct ubifs_budget_req *req); -long long ubifs_budg_get_free_space(struct ubifs_info *c); +long long ubifs_get_free_space(struct ubifs_info *c); int ubifs_calc_min_idx_lebs(struct ubifs_info *c); void ubifs_convert_page_budget(struct ubifs_info *c); +long long ubifs_reported_space(const struct ubifs_info *c, uint64_t free); long long ubifs_calc_available(const struct ubifs_info *c, int min_idx_lebs); /* find.c */ @@ -1440,8 +1461,6 @@ int ubifs_save_dirty_idx_lnums(struct ubifs_info *c); /* tnc.c */ int ubifs_lookup_level0(struct ubifs_info *c, const union ubifs_key *key, struct ubifs_znode **zn, int *n); -int ubifs_tnc_lookup(struct ubifs_info *c, const union ubifs_key *key, - void *node); int ubifs_tnc_lookup_nm(struct ubifs_info *c, const union ubifs_key *key, void *node, const struct qstr *nm); int ubifs_tnc_locate(struct ubifs_info *c, const union ubifs_key *key, diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index 1388a07..649bec7 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -61,7 +61,7 @@ /* * Limit the number of extended attributes per inode so that the total size - * (xattr_size) is guaranteeded to fit in an 'unsigned int'. + * (@xattr_size) is guaranteeded to fit in an 'unsigned int'. */ #define MAX_XATTRS_PER_INODE 65535 @@ -103,14 +103,14 @@ static int create_xattr(struct ubifs_info *c, struct inode *host, struct inode *inode; struct ubifs_inode *ui, *host_ui = ubifs_inode(host); struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1, - .new_ino_d = size, .dirtied_ino = 1, - .dirtied_ino_d = host_ui->data_len}; + .new_ino_d = ALIGN(size, 8), .dirtied_ino = 1, + .dirtied_ino_d = ALIGN(host_ui->data_len, 8) }; if (host_ui->xattr_cnt >= MAX_XATTRS_PER_INODE) return -ENOSPC; /* * Linux limits the maximum size of the extended attribute names list - * to %XATTR_LIST_MAX. This means we should not allow creating more* + * to %XATTR_LIST_MAX. This means we should not allow creating more * extended attributes if the name list becomes larger. This limitation * is artificial for UBIFS, though. */ @@ -128,7 +128,6 @@ static int create_xattr(struct ubifs_info *c, struct inode *host, goto out_budg; } - mutex_lock(&host_ui->ui_mutex); /* Re-define all operations to be "nothing" */ inode->i_mapping->a_ops = &none_address_operations; inode->i_op = &none_inode_operations; @@ -141,23 +140,19 @@ static int create_xattr(struct ubifs_info *c, struct inode *host, ui->data = kmalloc(size, GFP_NOFS); if (!ui->data) { err = -ENOMEM; - goto out_unlock; + goto out_free; } - memcpy(ui->data, value, size); + inode->i_size = ui->ui_size = size; + ui->data_len = size; + + mutex_lock(&host_ui->ui_mutex); host->i_ctime = ubifs_current_time(host); host_ui->xattr_cnt += 1; host_ui->xattr_size += CALC_DENT_SIZE(nm->len); host_ui->xattr_size += CALC_XATTR_BYTES(size); host_ui->xattr_names += nm->len; - /* - * We do not use i_size_write() because nobody can race with us as we - * are holding host @host->i_mutex - every xattr operation for this - * inode is serialized by it. - */ - inode->i_size = ui->ui_size = size; - ui->data_len = size; err = ubifs_jnl_update(c, host, nm, inode, 0, 1); if (err) goto out_cancel; @@ -172,8 +167,8 @@ out_cancel: host_ui->xattr_cnt -= 1; host_ui->xattr_size -= CALC_DENT_SIZE(nm->len); host_ui->xattr_size -= CALC_XATTR_BYTES(size); -out_unlock: mutex_unlock(&host_ui->ui_mutex); +out_free: make_bad_inode(inode); iput(inode); out_budg: @@ -200,29 +195,28 @@ static int change_xattr(struct ubifs_info *c, struct inode *host, struct ubifs_inode *host_ui = ubifs_inode(host); struct ubifs_inode *ui = ubifs_inode(inode); struct ubifs_budget_req req = { .dirtied_ino = 2, - .dirtied_ino_d = size + host_ui->data_len }; + .dirtied_ino_d = ALIGN(size, 8) + ALIGN(host_ui->data_len, 8) }; ubifs_assert(ui->data_len == inode->i_size); err = ubifs_budget_space(c, &req); if (err) return err; - mutex_lock(&host_ui->ui_mutex); - host->i_ctime = ubifs_current_time(host); - host_ui->xattr_size -= CALC_XATTR_BYTES(ui->data_len); - host_ui->xattr_size += CALC_XATTR_BYTES(size); - kfree(ui->data); ui->data = kmalloc(size, GFP_NOFS); if (!ui->data) { err = -ENOMEM; - goto out_unlock; + goto out_free; } - memcpy(ui->data, value, size); inode->i_size = ui->ui_size = size; ui->data_len = size; + mutex_lock(&host_ui->ui_mutex); + host->i_ctime = ubifs_current_time(host); + host_ui->xattr_size -= CALC_XATTR_BYTES(ui->data_len); + host_ui->xattr_size += CALC_XATTR_BYTES(size); + /* * It is important to write the host inode after the xattr inode * because if the host inode gets synchronized (via 'fsync()'), then @@ -240,9 +234,9 @@ static int change_xattr(struct ubifs_info *c, struct inode *host, out_cancel: host_ui->xattr_size -= CALC_XATTR_BYTES(size); host_ui->xattr_size += CALC_XATTR_BYTES(ui->data_len); - make_bad_inode(inode); -out_unlock: mutex_unlock(&host_ui->ui_mutex); + make_bad_inode(inode); +out_free: ubifs_release_budget(c, &req); return err; } @@ -312,6 +306,7 @@ int ubifs_setxattr(struct dentry *dentry, const char *name, dbg_gen("xattr '%s', host ino %lu ('%.*s'), size %zd", name, host->i_ino, dentry->d_name.len, dentry->d_name.name, size); + ubifs_assert(mutex_is_locked(&host->i_mutex)); if (size > UBIFS_MAX_INO_DATA) return -ERANGE; @@ -384,7 +379,6 @@ ssize_t ubifs_getxattr(struct dentry *dentry, const char *name, void *buf, if (!xent) return -ENOMEM; - mutex_lock(&host->i_mutex); xent_key_init(c, &key, host->i_ino, &nm); err = ubifs_tnc_lookup_nm(c, &key, xent, &nm); if (err) { @@ -419,7 +413,6 @@ ssize_t ubifs_getxattr(struct dentry *dentry, const char *name, void *buf, out_iput: iput(inode); out_unlock: - mutex_unlock(&host->i_mutex); kfree(xent); return err; } @@ -449,8 +442,6 @@ ssize_t ubifs_listxattr(struct dentry *dentry, char *buffer, size_t size) return -ERANGE; lowest_xent_key(c, &key, host->i_ino); - - mutex_lock(&host->i_mutex); while (1) { int type; @@ -479,7 +470,6 @@ ssize_t ubifs_listxattr(struct dentry *dentry, char *buffer, size_t size) pxent = xent; key_read(c, &xent->key, &key); } - mutex_unlock(&host->i_mutex); kfree(pxent); if (err != -ENOENT) { @@ -497,8 +487,8 @@ static int remove_xattr(struct ubifs_info *c, struct inode *host, int err; struct ubifs_inode *host_ui = ubifs_inode(host); struct ubifs_inode *ui = ubifs_inode(inode); - struct ubifs_budget_req req = { .dirtied_ino = 1, .mod_dent = 1, - .dirtied_ino_d = host_ui->data_len }; + struct ubifs_budget_req req = { .dirtied_ino = 2, .mod_dent = 1, + .dirtied_ino_d = ALIGN(host_ui->data_len, 8) }; ubifs_assert(ui->data_len == inode->i_size); diff --git a/fs/udf/file.c b/fs/udf/file.c index 0ed6e14..eb91f3b 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -211,6 +211,7 @@ const struct file_operations udf_file_operations = { .release = udf_release_file, .fsync = udf_fsync_file, .splice_read = generic_file_splice_read, + .llseek = generic_file_llseek, }; const struct inode_operations udf_file_inode_operations = { diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index eb9cfa2..a4f2b3c 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -76,11 +76,24 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) *err = -ENOSPC; iinfo = UDF_I(inode); - iinfo->i_unique = 0; - iinfo->i_lenExtents = 0; - iinfo->i_next_alloc_block = 0; - iinfo->i_next_alloc_goal = 0; - iinfo->i_strat4096 = 0; + if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { + iinfo->i_efe = 1; + if (UDF_VERS_USE_EXTENDED_FE > sbi->s_udfrev) + sbi->s_udfrev = UDF_VERS_USE_EXTENDED_FE; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct extendedFileEntry), + GFP_KERNEL); + } else { + iinfo->i_efe = 0; + iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - + sizeof(struct fileEntry), + GFP_KERNEL); + } + if (!iinfo->i_ext.i_data) { + iput(inode); + *err = -ENOMEM; + return NULL; + } block = udf_new_block(dir->i_sb, NULL, dinfo->i_location.partitionReferenceNum, @@ -111,6 +124,7 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) lvhd->uniqueID = cpu_to_le64(uniqueID); mark_buffer_dirty(sbi->s_lvid_bh); } + mutex_unlock(&sbi->s_alloc_mutex); inode->i_mode = mode; inode->i_uid = current->fsuid; if (dir->i_mode & S_ISGID) { @@ -129,25 +143,6 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) iinfo->i_lenEAttr = 0; iinfo->i_lenAlloc = 0; iinfo->i_use = 0; - if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_EXTENDED_FE)) { - iinfo->i_efe = 1; - if (UDF_VERS_USE_EXTENDED_FE > sbi->s_udfrev) - sbi->s_udfrev = UDF_VERS_USE_EXTENDED_FE; - iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - - sizeof(struct extendedFileEntry), - GFP_KERNEL); - } else { - iinfo->i_efe = 0; - iinfo->i_ext.i_data = kzalloc(inode->i_sb->s_blocksize - - sizeof(struct fileEntry), - GFP_KERNEL); - } - if (!iinfo->i_ext.i_data) { - iput(inode); - *err = -ENOMEM; - mutex_unlock(&sbi->s_alloc_mutex); - return NULL; - } if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_AD_IN_ICB)) iinfo->i_alloc_type = ICBTAG_FLAG_AD_IN_ICB; else if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_SHORT_AD)) @@ -158,7 +153,6 @@ struct inode *udf_new_inode(struct inode *dir, int mode, int *err) iinfo->i_crtime = current_fs_time(inode->i_sb); insert_inode_hash(inode); mark_inode_dirty(inode); - mutex_unlock(&sbi->s_alloc_mutex); if (DQUOT_ALLOC_INODE(inode)) { DQUOT_DROP(inode); diff --git a/fs/udf/super.c b/fs/udf/super.c index 44cc702..5698bbf 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -148,7 +148,7 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(struct kmem_cache *cachep, void *foo) +static void init_once(void *foo) { struct udf_inode_info *ei = (struct udf_inode_info *)foo; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 227c9d7..3141969 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1233,7 +1233,7 @@ static int ufs_show_options(struct seq_file *seq, struct vfsmount *vfs) { struct ufs_sb_info *sbi = UFS_SB(vfs->mnt_sb); unsigned mval = sbi->s_mount_opt & UFS_MOUNT_UFSTYPE; - const struct match_token *tp = tokens; + struct match_token *tp = tokens; while (tp->token != Opt_onerror_panic && tp->token != mval) ++tp; @@ -1302,7 +1302,7 @@ static void ufs_destroy_inode(struct inode *inode) kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } -static void init_once(struct kmem_cache * cachep, void *foo) +static void init_once(void *foo) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/fs/utimes.c b/fs/utimes.c index b6b664e..6929e3e 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -48,66 +48,22 @@ static bool nsec_valid(long nsec) return nsec >= 0 && nsec <= 999999999; } -/* If times==NULL, set access and modification to current time, - * must be owner or have write permission. - * Else, update from *times, must be owner or super user. - */ -long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags) +static int utimes_common(struct path *path, struct timespec *times) { int error; - struct nameidata nd; - struct dentry *dentry; - struct inode *inode; struct iattr newattrs; - struct file *f = NULL; - struct vfsmount *mnt; - - error = -EINVAL; - if (times && (!nsec_valid(times[0].tv_nsec) || - !nsec_valid(times[1].tv_nsec))) { - goto out; - } - - if (flags & ~AT_SYMLINK_NOFOLLOW) - goto out; - - if (filename == NULL && dfd != AT_FDCWD) { - error = -EINVAL; - if (flags & AT_SYMLINK_NOFOLLOW) - goto out; + struct inode *inode = path->dentry->d_inode; - error = -EBADF; - f = fget(dfd); - if (!f) - goto out; - dentry = f->f_path.dentry; - mnt = f->f_path.mnt; - } else { - error = __user_walk_fd(dfd, filename, (flags & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW, &nd); - if (error) - goto out; - - dentry = nd.path.dentry; - mnt = nd.path.mnt; - } - - inode = dentry->d_inode; - - error = mnt_want_write(mnt); + error = mnt_want_write(path->mnt); if (error) - goto dput_and_out; + goto out; if (times && times[0].tv_nsec == UTIME_NOW && times[1].tv_nsec == UTIME_NOW) times = NULL; - /* In most cases, the checks are done in inode_change_ok() */ newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME; if (times) { - error = -EPERM; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - goto mnt_drop_write_and_out; - if (times[0].tv_nsec == UTIME_OMIT) newattrs.ia_valid &= ~ATTR_ATIME; else if (times[0].tv_nsec != UTIME_NOW) { @@ -123,21 +79,13 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags newattrs.ia_mtime.tv_nsec = times[1].tv_nsec; newattrs.ia_valid |= ATTR_MTIME_SET; } - /* - * For the UTIME_OMIT/UTIME_NOW and UTIME_NOW/UTIME_OMIT - * cases, we need to make an extra check that is not done by - * inode_change_ok(). + * Tell inode_change_ok(), that this is an explicit time + * update, even if neither ATTR_ATIME_SET nor ATTR_MTIME_SET + * were used. */ - if (((times[0].tv_nsec == UTIME_NOW && - times[1].tv_nsec == UTIME_OMIT) - || - (times[0].tv_nsec == UTIME_OMIT && - times[1].tv_nsec == UTIME_NOW)) - && !is_owner_or_cap(inode)) - goto mnt_drop_write_and_out; + newattrs.ia_valid |= ATTR_TIMES_SET; } else { - /* * If times is NULL (or both times are UTIME_NOW), * then we need to check permissions, because @@ -148,21 +96,76 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags goto mnt_drop_write_and_out; if (!is_owner_or_cap(inode)) { - error = permission(inode, MAY_WRITE, NULL); + error = inode_permission(inode, MAY_WRITE); if (error) goto mnt_drop_write_and_out; } } mutex_lock(&inode->i_mutex); - error = notify_change(dentry, &newattrs); + error = notify_change(path->dentry, &newattrs); mutex_unlock(&inode->i_mutex); + mnt_drop_write_and_out: - mnt_drop_write(mnt); -dput_and_out: - if (f) - fput(f); - else - path_put(&nd.path); + mnt_drop_write(path->mnt); +out: + return error; +} + +/* + * do_utimes - change times on filename or file descriptor + * @dfd: open file descriptor, -1 or AT_FDCWD + * @filename: path name or NULL + * @times: new times or NULL + * @flags: zero or more flags (only AT_SYMLINK_NOFOLLOW for the moment) + * + * If filename is NULL and dfd refers to an open file, then operate on + * the file. Otherwise look up filename, possibly using dfd as a + * starting point. + * + * If times==NULL, set access and modification to current time, + * must be owner or have write permission. + * Else, update from *times, must be owner or super user. + */ +long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags) +{ + int error = -EINVAL; + + if (times && (!nsec_valid(times[0].tv_nsec) || + !nsec_valid(times[1].tv_nsec))) { + goto out; + } + + if (flags & ~AT_SYMLINK_NOFOLLOW) + goto out; + + if (filename == NULL && dfd != AT_FDCWD) { + struct file *file; + + if (flags & AT_SYMLINK_NOFOLLOW) + goto out; + + file = fget(dfd); + error = -EBADF; + if (!file) + goto out; + + error = utimes_common(&file->f_path, times); + fput(file); + } else { + struct path path; + int lookup_flags = 0; + + if (!(flags & AT_SYMLINK_NOFOLLOW)) + lookup_flags |= LOOKUP_FOLLOW; + + error = user_path_at(dfd, filename, lookup_flags, &path); + if (error) + goto out; + + error = utimes_common(&path, times); + path_put(&path); + } + out: return error; } @@ -63,7 +63,7 @@ xattr_permission(struct inode *inode, const char *name, int mask) return -EPERM; } - return permission(inode, mask, NULL); + return inode_permission(inode, mask); } int @@ -252,40 +252,40 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, } asmlinkage long -sys_setxattr(const char __user *path, const char __user *name, +sys_setxattr(const char __user *pathname, const char __user *name, const void __user *value, size_t size, int flags) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = setxattr(nd.path.dentry, name, value, size, flags); - mnt_drop_write(nd.path.mnt); + error = setxattr(path.dentry, name, value, size, flags); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } asmlinkage long -sys_lsetxattr(const char __user *path, const char __user *name, +sys_lsetxattr(const char __user *pathname, const char __user *name, const void __user *value, size_t size, int flags) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = setxattr(nd.path.dentry, name, value, size, flags); - mnt_drop_write(nd.path.mnt); + error = setxattr(path.dentry, name, value, size, flags); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } @@ -350,32 +350,32 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, } asmlinkage ssize_t -sys_getxattr(const char __user *path, const char __user *name, +sys_getxattr(const char __user *pathname, const char __user *name, void __user *value, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = getxattr(nd.path.dentry, name, value, size); - path_put(&nd.path); + error = getxattr(path.dentry, name, value, size); + path_put(&path); return error; } asmlinkage ssize_t -sys_lgetxattr(const char __user *path, const char __user *name, void __user *value, +sys_lgetxattr(const char __user *pathname, const char __user *name, void __user *value, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = getxattr(nd.path.dentry, name, value, size); - path_put(&nd.path); + error = getxattr(path.dentry, name, value, size); + path_put(&path); return error; } @@ -425,30 +425,30 @@ listxattr(struct dentry *d, char __user *list, size_t size) } asmlinkage ssize_t -sys_listxattr(const char __user *path, char __user *list, size_t size) +sys_listxattr(const char __user *pathname, char __user *list, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = listxattr(nd.path.dentry, list, size); - path_put(&nd.path); + error = listxattr(path.dentry, list, size); + path_put(&path); return error; } asmlinkage ssize_t -sys_llistxattr(const char __user *path, char __user *list, size_t size) +sys_llistxattr(const char __user *pathname, char __user *list, size_t size) { - struct nameidata nd; + struct path path; ssize_t error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = listxattr(nd.path.dentry, list, size); - path_put(&nd.path); + error = listxattr(path.dentry, list, size); + path_put(&path); return error; } @@ -486,38 +486,38 @@ removexattr(struct dentry *d, const char __user *name) } asmlinkage long -sys_removexattr(const char __user *path, const char __user *name) +sys_removexattr(const char __user *pathname, const char __user *name) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk(path, &nd); + error = user_path(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = removexattr(nd.path.dentry, name); - mnt_drop_write(nd.path.mnt); + error = removexattr(path.dentry, name); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } asmlinkage long -sys_lremovexattr(const char __user *path, const char __user *name) +sys_lremovexattr(const char __user *pathname, const char __user *name) { - struct nameidata nd; + struct path path; int error; - error = user_path_walk_link(path, &nd); + error = user_lpath(pathname, &path); if (error) return error; - error = mnt_want_write(nd.path.mnt); + error = mnt_want_write(path.mnt); if (!error) { - error = removexattr(nd.path.dentry, name); - mnt_drop_write(nd.path.mnt); + error = removexattr(path.dentry, name); + mnt_drop_write(path.mnt); } - path_put(&nd.path); + path_put(&path); return error; } diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index 36ec614..737c9a4 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -106,7 +106,8 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \ xfs_iops.o \ xfs_lrw.o \ xfs_super.o \ - xfs_vnode.o) + xfs_vnode.o \ + xfs_xattr.o) # Objects in support/ xfs-y += $(addprefix support/, \ diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index 9b1bb17..1cd3b55 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -90,7 +90,7 @@ kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize, } void -kmem_free(void *ptr, size_t size) +kmem_free(const void *ptr) { if (!is_vmalloc_addr(ptr)) { kfree(ptr); @@ -100,7 +100,7 @@ kmem_free(void *ptr, size_t size) } void * -kmem_realloc(void *ptr, size_t newsize, size_t oldsize, +kmem_realloc(const void *ptr, size_t newsize, size_t oldsize, unsigned int __nocast flags) { void *new; @@ -110,7 +110,7 @@ kmem_realloc(void *ptr, size_t newsize, size_t oldsize, if (new) memcpy(new, ptr, ((oldsize < newsize) ? oldsize : newsize)); - kmem_free(ptr, oldsize); + kmem_free(ptr); } return new; } diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h index 5e95649..af6843c 100644 --- a/fs/xfs/linux-2.6/kmem.h +++ b/fs/xfs/linux-2.6/kmem.h @@ -57,8 +57,8 @@ kmem_flags_convert(unsigned int __nocast flags) extern void *kmem_alloc(size_t, unsigned int __nocast); extern void *kmem_zalloc(size_t, unsigned int __nocast); extern void *kmem_zalloc_greedy(size_t *, size_t, size_t, unsigned int __nocast); -extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast); -extern void kmem_free(void *, size_t); +extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast); +extern void kmem_free(const void *); /* * Zone interfaces @@ -79,7 +79,7 @@ kmem_zone_init(int size, char *zone_name) static inline kmem_zone_t * kmem_zone_init_flags(int size, char *zone_name, unsigned long flags, - void (*construct)(kmem_zone_t *, void *)) + void (*construct)(void *)) { return kmem_cache_create(zone_name, size, 0, flags, construct); } diff --git a/fs/xfs/linux-2.6/sema.h b/fs/xfs/linux-2.6/sema.h deleted file mode 100644 index 3abe7e9..0000000 --- a/fs/xfs/linux-2.6/sema.h +++ /dev/null @@ -1,52 +0,0 @@ -/* - * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. - * 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 as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would 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 the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_SEMA_H__ -#define __XFS_SUPPORT_SEMA_H__ - -#include <linux/time.h> -#include <linux/wait.h> -#include <linux/semaphore.h> -#include <asm/atomic.h> - -/* - * sema_t structure just maps to struct semaphore in Linux kernel. - */ - -typedef struct semaphore sema_t; - -#define initnsema(sp, val, name) sema_init(sp, val) -#define psema(sp, b) down(sp) -#define vsema(sp) up(sp) -#define freesema(sema) do { } while (0) - -static inline int issemalocked(sema_t *sp) -{ - return down_trylock(sp) || (up(sp), 0); -} - -/* - * Map cpsema (try to get the sema) to down_trylock. We need to switch - * the return values since cpsema returns 1 (acquired) 0 (failed) and - * down_trylock returns the reverse 0 (acquired) 1 (failed). - */ -static inline int cpsema(sema_t *sp) -{ - return down_trylock(sp) ? 0 : 1; -} - -#endif /* __XFS_SUPPORT_SEMA_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index a55c3b2..a44d68e 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -73,7 +73,6 @@ xfs_page_trace( unsigned long pgoff) { xfs_inode_t *ip; - bhv_vnode_t *vp = vn_from_inode(inode); loff_t isize = i_size_read(inode); loff_t offset = page_offset(page); int delalloc = -1, unmapped = -1, unwritten = -1; @@ -81,7 +80,7 @@ xfs_page_trace( if (page_has_buffers(page)) xfs_count_page_state(page, &delalloc, &unmapped, &unwritten); - ip = xfs_vtoi(vp); + ip = XFS_I(inode); if (!ip->i_rwtrace) return; @@ -409,7 +408,6 @@ xfs_start_buffer_writeback( STATIC void xfs_start_page_writeback( struct page *page, - struct writeback_control *wbc, int clear_dirty, int buffers) { @@ -676,7 +674,7 @@ xfs_probe_cluster( } else pg_offset = PAGE_CACHE_SIZE; - if (page->index == tindex && !TestSetPageLocked(page)) { + if (page->index == tindex && trylock_page(page)) { pg_len = xfs_probe_page(page, pg_offset, mapped); unlock_page(page); } @@ -760,7 +758,7 @@ xfs_convert_page( if (page->index != tindex) goto fail; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto fail; if (PageWriteback(page)) goto fail_unlock_page; @@ -858,7 +856,7 @@ xfs_convert_page( done = 1; } } - xfs_start_page_writeback(page, wbc, !page_dirty, count); + xfs_start_page_writeback(page, !page_dirty, count); } return done; @@ -1105,7 +1103,7 @@ xfs_page_state_convert( * that we are writing into for the first time. */ type = IOMAP_NEW; - if (!test_and_set_bit(BH_Lock, &bh->b_state)) { + if (trylock_buffer(bh)) { ASSERT(buffer_mapped(bh)); if (iomap_valid) all_bh = 1; @@ -1130,7 +1128,7 @@ xfs_page_state_convert( SetPageUptodate(page); if (startio) - xfs_start_page_writeback(page, wbc, 1, count); + xfs_start_page_writeback(page, 1, count); if (ioend && iomap_valid) { offset = (iomap.iomap_offset + iomap.iomap_bsize - 1) >> @@ -1340,6 +1338,10 @@ __xfs_get_blocks( offset = (xfs_off_t)iblock << inode->i_blkbits; ASSERT(bh_result->b_size >= (1 << inode->i_blkbits)); size = bh_result->b_size; + + if (!create && direct && offset >= i_size_read(inode)) + return 0; + error = xfs_iomap(XFS_I(inode), offset, size, create ? flags : BMAPI_READ, &iomap, &niomap); if (error) diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 98e0e86..36d5fcd 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -58,7 +58,7 @@ xfs_buf_trace( bp, id, (void *)(unsigned long)bp->b_flags, (void *)(unsigned long)bp->b_hold.counter, - (void *)(unsigned long)bp->b_sema.count.counter, + (void *)(unsigned long)bp->b_sema.count, (void *)current, data, ra, (void *)(unsigned long)((bp->b_file_offset>>32) & 0xffffffff), @@ -253,7 +253,7 @@ _xfs_buf_initialize( memset(bp, 0, sizeof(xfs_buf_t)); atomic_set(&bp->b_hold, 1); - init_MUTEX_LOCKED(&bp->b_iodonesema); + init_completion(&bp->b_iowait); INIT_LIST_HEAD(&bp->b_list); INIT_LIST_HEAD(&bp->b_hash_list); init_MUTEX_LOCKED(&bp->b_sema); /* held, no waiters */ @@ -310,8 +310,7 @@ _xfs_buf_free_pages( xfs_buf_t *bp) { if (bp->b_pages != bp->b_page_array) { - kmem_free(bp->b_pages, - bp->b_page_count * sizeof(struct page *)); + kmem_free(bp->b_pages); } } @@ -839,6 +838,7 @@ xfs_buf_rele( return; } + ASSERT(atomic_read(&bp->b_hold) > 0); if (atomic_dec_and_lock(&bp->b_hold, &hash->bh_lock)) { if (bp->b_relse) { atomic_inc(&bp->b_hold); @@ -852,11 +852,6 @@ xfs_buf_rele( spin_unlock(&hash->bh_lock); xfs_buf_free(bp); } - } else { - /* - * Catch reference count leaks - */ - ASSERT(atomic_read(&bp->b_hold) >= 0); } } @@ -1006,12 +1001,13 @@ xfs_buf_iodone_work( * We can get an EOPNOTSUPP to ordered writes. Here we clear the * ordered flag and reissue them. Because we can't tell the higher * layers directly that they should not issue ordered I/O anymore, they - * need to check if the ordered flag was cleared during I/O completion. + * need to check if the _XFS_BARRIER_FAILED flag was set during I/O completion. */ if ((bp->b_error == EOPNOTSUPP) && (bp->b_flags & (XBF_ORDERED|XBF_ASYNC)) == (XBF_ORDERED|XBF_ASYNC)) { XB_TRACE(bp, "ordered_retry", bp->b_iodone); bp->b_flags &= ~XBF_ORDERED; + bp->b_flags |= _XFS_BARRIER_FAILED; xfs_buf_iorequest(bp); } else if (bp->b_iodone) (*(bp->b_iodone))(bp); @@ -1038,7 +1034,7 @@ xfs_buf_ioend( xfs_buf_iodone_work(&bp->b_iodone_work); } } else { - up(&bp->b_iodonesema); + complete(&bp->b_iowait); } } @@ -1276,7 +1272,7 @@ xfs_buf_iowait( XB_TRACE(bp, "iowait", 0); if (atomic_read(&bp->b_io_remaining)) blk_run_address_space(bp->b_target->bt_mapping); - down(&bp->b_iodonesema); + wait_for_completion(&bp->b_iowait); XB_TRACE(bp, "iowaited", (long)bp->b_error); return bp->b_error; } @@ -1398,7 +1394,7 @@ STATIC void xfs_free_bufhash( xfs_buftarg_t *btp) { - kmem_free(btp->bt_hash, (1<<btp->bt_hashshift) * sizeof(xfs_bufhash_t)); + kmem_free(btp->bt_hash); btp->bt_hash = NULL; } @@ -1428,13 +1424,10 @@ xfs_unregister_buftarg( void xfs_free_buftarg( - xfs_buftarg_t *btp, - int external) + xfs_buftarg_t *btp) { xfs_flush_buftarg(btp, 1); xfs_blkdev_issue_flush(btp); - if (external) - xfs_blkdev_put(btp->bt_bdev); xfs_free_bufhash(btp); iput(btp->bt_mapping->host); @@ -1444,7 +1437,7 @@ xfs_free_buftarg( xfs_unregister_buftarg(btp); kthread_stop(btp->bt_task); - kmem_free(btp, sizeof(*btp)); + kmem_free(btp); } STATIC int @@ -1575,7 +1568,7 @@ xfs_alloc_buftarg( return btp; error: - kmem_free(btp, sizeof(*btp)); + kmem_free(btp); return NULL; } @@ -1803,7 +1796,7 @@ int __init xfs_buf_init(void) { #ifdef XFS_BUF_TRACE - xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_SLEEP); + xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_NOFS); #endif xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf", diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index f948ec7..456519a 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -85,6 +85,14 @@ typedef enum { * modifications being lost. */ _XBF_PAGE_LOCKED = (1 << 22), + + /* + * If we try a barrier write, but it fails we have to communicate + * this to the upper layers. Unfortunately b_error gets overwritten + * when the buffer is re-issued so we have to add another flag to + * keep this information. + */ + _XFS_BARRIER_FAILED = (1 << 23), } xfs_buf_flags_t; typedef enum { @@ -157,7 +165,7 @@ typedef struct xfs_buf { xfs_buf_iodone_t b_iodone; /* I/O completion function */ xfs_buf_relse_t b_relse; /* releasing function */ xfs_buf_bdstrat_t b_strat; /* pre-write function */ - struct semaphore b_iodonesema; /* Semaphore for I/O waiters */ + struct completion b_iowait; /* queue for I/O waiters */ void *b_fspriv; void *b_fspriv2; void *b_fspriv3; @@ -352,7 +360,7 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *); #define XFS_BUF_CPSEMA(bp) (xfs_buf_cond_lock(bp) == 0) #define XFS_BUF_VSEMA(bp) xfs_buf_unlock(bp) #define XFS_BUF_PSEMA(bp,x) xfs_buf_lock(bp) -#define XFS_BUF_V_IODONESEMA(bp) up(&bp->b_iodonesema); +#define XFS_BUF_FINISH_IOWAIT(bp) complete(&bp->b_iowait); #define XFS_BUF_SET_TARGET(bp, target) ((bp)->b_target = (target)) #define XFS_BUF_TARGET(bp) ((bp)->b_target) @@ -429,7 +437,7 @@ static inline void xfs_bdwrite(void *mp, xfs_buf_t *bp) * Handling of buftargs. */ extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int); -extern void xfs_free_buftarg(xfs_buftarg_t *, int); +extern void xfs_free_buftarg(xfs_buftarg_t *); extern void xfs_wait_buftarg(xfs_buftarg_t *); extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int); extern int xfs_flush_buftarg(xfs_buftarg_t *, int); diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index c672b32..24fd598 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -139,7 +139,7 @@ xfs_nfs_get_inode( } xfs_iunlock(ip, XFS_ILOCK_SHARED); - return ip->i_vnode; + return VFS_I(ip); } STATIC struct dentry * @@ -167,7 +167,7 @@ xfs_fs_fh_to_dentry(struct super_block *sb, struct fid *fid, if (!inode) return NULL; if (IS_ERR(inode)) - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); result = d_alloc_anon(inode); if (!result) { iput(inode); @@ -198,7 +198,7 @@ xfs_fs_fh_to_parent(struct super_block *sb, struct fid *fid, if (!inode) return NULL; if (IS_ERR(inode)) - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); result = d_alloc_anon(inode); if (!result) { iput(inode); @@ -215,13 +215,13 @@ xfs_fs_get_parent( struct xfs_inode *cip; struct dentry *parent; - error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip); + error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL); if (unlikely(error)) return ERR_PTR(-error); - parent = d_alloc_anon(cip->i_vnode); + parent = d_alloc_anon(VFS_I(cip)); if (unlikely(!parent)) { - iput(cip->i_vnode); + iput(VFS_I(cip)); return ERR_PTR(-ENOMEM); } return parent; diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 5f60363..5311c1a 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -475,6 +475,7 @@ const struct file_operations xfs_invis_file_operations = { const struct file_operations xfs_dir_file_operations = { .read = generic_read_dir, .readdir = xfs_file_readdir, + .llseek = generic_file_llseek, .unlocked_ioctl = xfs_file_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = xfs_file_compat_ioctl, diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c index 1eefe61..36caa6d 100644 --- a/fs/xfs/linux-2.6/xfs_fs_subr.c +++ b/fs/xfs/linux-2.6/xfs_fs_subr.c @@ -31,7 +31,7 @@ xfs_tosspages( xfs_off_t last, int fiopt) { - struct address_space *mapping = ip->i_vnode->i_mapping; + struct address_space *mapping = VFS_I(ip)->i_mapping; if (mapping->nrpages) truncate_inode_pages(mapping, first); @@ -44,7 +44,7 @@ xfs_flushinval_pages( xfs_off_t last, int fiopt) { - struct address_space *mapping = ip->i_vnode->i_mapping; + struct address_space *mapping = VFS_I(ip)->i_mapping; int ret = 0; if (mapping->nrpages) { @@ -64,7 +64,7 @@ xfs_flush_pages( uint64_t flags, int fiopt) { - struct address_space *mapping = ip->i_vnode->i_mapping; + struct address_space *mapping = VFS_I(ip)->i_mapping; int ret = 0; int ret2; diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index a42ba9d..48799ba 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -48,6 +48,8 @@ #include "xfs_dfrag.h" #include "xfs_fsops.h" #include "xfs_vnodeops.h" +#include "xfs_quota.h" +#include "xfs_inode_item.h" #include <linux/capability.h> #include <linux/dcache.h> @@ -84,17 +86,15 @@ xfs_find_handle( switch (cmd) { case XFS_IOC_PATH_TO_FSHANDLE: case XFS_IOC_PATH_TO_HANDLE: { - struct nameidata nd; - int error; - - error = user_path_walk_link((const char __user *)hreq.path, &nd); + struct path path; + int error = user_lpath((const char __user *)hreq.path, &path); if (error) return error; - ASSERT(nd.path.dentry); - ASSERT(nd.path.dentry->d_inode); - inode = igrab(nd.path.dentry->d_inode); - path_put(&nd.path); + ASSERT(path.dentry); + ASSERT(path.dentry->d_inode); + inode = igrab(path.dentry->d_inode); + path_put(&path); break; } @@ -245,7 +245,7 @@ xfs_vget_fsop_handlereq( xfs_iunlock(ip, XFS_ILOCK_SHARED); - *inode = XFS_ITOV(ip); + *inode = VFS_I(ip); return 0; } @@ -470,6 +470,12 @@ xfs_attrlist_by_handle( if (al_hreq.buflen > XATTR_LIST_MAX) return -XFS_ERROR(EINVAL); + /* + * Reject flags, only allow namespaces. + */ + if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) + return -XFS_ERROR(EINVAL); + error = xfs_vget_fsop_handlereq(mp, parinode, &al_hreq.hreq, &inode); if (error) goto out; @@ -589,7 +595,7 @@ xfs_attrmulti_by_handle( goto out; error = E2BIG; - size = am_hreq.opcount * sizeof(attr_multiop_t); + size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); if (!size || size > 16 * PAGE_SIZE) goto out_vn_rele; @@ -682,9 +688,9 @@ xfs_ioc_space( return -XFS_ERROR(EFAULT); if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - attr_flags |= ATTR_NONBLOCK; + attr_flags |= XFS_ATTR_NONBLOCK; if (ioflags & IO_INVIS) - attr_flags |= ATTR_DMI; + attr_flags |= XFS_ATTR_DMI; error = xfs_change_file_space(ip, cmd, &bf, filp->f_pos, NULL, attr_flags); @@ -875,6 +881,322 @@ xfs_ioc_fsgetxattr( return 0; } +STATIC void +xfs_set_diflags( + struct xfs_inode *ip, + unsigned int xflags) +{ + unsigned int di_flags; + + /* can't set PREALLOC this way, just preserve it */ + di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); + if (xflags & XFS_XFLAG_IMMUTABLE) + di_flags |= XFS_DIFLAG_IMMUTABLE; + if (xflags & XFS_XFLAG_APPEND) + di_flags |= XFS_DIFLAG_APPEND; + if (xflags & XFS_XFLAG_SYNC) + di_flags |= XFS_DIFLAG_SYNC; + if (xflags & XFS_XFLAG_NOATIME) + di_flags |= XFS_DIFLAG_NOATIME; + if (xflags & XFS_XFLAG_NODUMP) + di_flags |= XFS_DIFLAG_NODUMP; + if (xflags & XFS_XFLAG_PROJINHERIT) + di_flags |= XFS_DIFLAG_PROJINHERIT; + if (xflags & XFS_XFLAG_NODEFRAG) + di_flags |= XFS_DIFLAG_NODEFRAG; + if (xflags & XFS_XFLAG_FILESTREAM) + di_flags |= XFS_DIFLAG_FILESTREAM; + if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) { + if (xflags & XFS_XFLAG_RTINHERIT) + di_flags |= XFS_DIFLAG_RTINHERIT; + if (xflags & XFS_XFLAG_NOSYMLINKS) + di_flags |= XFS_DIFLAG_NOSYMLINKS; + if (xflags & XFS_XFLAG_EXTSZINHERIT) + di_flags |= XFS_DIFLAG_EXTSZINHERIT; + } else if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) { + if (xflags & XFS_XFLAG_REALTIME) + di_flags |= XFS_DIFLAG_REALTIME; + if (xflags & XFS_XFLAG_EXTSIZE) + di_flags |= XFS_DIFLAG_EXTSIZE; + } + + ip->i_d.di_flags = di_flags; +} + +STATIC void +xfs_diflags_to_linux( + struct xfs_inode *ip) +{ + struct inode *inode = VFS_I(ip); + unsigned int xflags = xfs_ip2xflags(ip); + + if (xflags & XFS_XFLAG_IMMUTABLE) + inode->i_flags |= S_IMMUTABLE; + else + inode->i_flags &= ~S_IMMUTABLE; + if (xflags & XFS_XFLAG_APPEND) + inode->i_flags |= S_APPEND; + else + inode->i_flags &= ~S_APPEND; + if (xflags & XFS_XFLAG_SYNC) + inode->i_flags |= S_SYNC; + else + inode->i_flags &= ~S_SYNC; + if (xflags & XFS_XFLAG_NOATIME) + inode->i_flags |= S_NOATIME; + else + inode->i_flags &= ~S_NOATIME; +} + +#define FSX_PROJID 1 +#define FSX_EXTSIZE 2 +#define FSX_XFLAGS 4 +#define FSX_NONBLOCK 8 + +STATIC int +xfs_ioctl_setattr( + xfs_inode_t *ip, + struct fsxattr *fa, + int mask) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + unsigned int lock_flags = 0; + struct xfs_dquot *udqp = NULL, *gdqp = NULL; + struct xfs_dquot *olddquot = NULL; + int code; + + xfs_itrace_entry(ip); + + if (mp->m_flags & XFS_MOUNT_RDONLY) + return XFS_ERROR(EROFS); + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + /* + * If disk quotas is on, we make sure that the dquots do exist on disk, + * before we start any other transactions. Trying to do this later + * is messy. We don't care to take a readlock to look at the ids + * in inode here, because we can't hold it across the trans_reserve. + * If the IDs do change before we take the ilock, we're covered + * because the i_*dquot fields will get updated anyway. + */ + if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) { + code = XFS_QM_DQVOPALLOC(mp, ip, ip->i_d.di_uid, + ip->i_d.di_gid, fa->fsx_projid, + XFS_QMOPT_PQUOTA, &udqp, &gdqp); + if (code) + return code; + } + + /* + * For the other attributes, we acquire the inode lock and + * first do an error checking pass. + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); + code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + if (code) + goto error_return; + + lock_flags = XFS_ILOCK_EXCL; + xfs_ilock(ip, lock_flags); + + /* + * CAP_FOWNER overrides the following restrictions: + * + * The user ID of the calling process must be equal + * to the file owner ID, except in cases where the + * CAP_FSETID capability is applicable. + */ + if (current->fsuid != ip->i_d.di_uid && !capable(CAP_FOWNER)) { + code = XFS_ERROR(EPERM); + goto error_return; + } + + /* + * Do a quota reservation only if projid is actually going to change. + */ + if (mask & FSX_PROJID) { + if (XFS_IS_PQUOTA_ON(mp) && + ip->i_d.di_projid != fa->fsx_projid) { + ASSERT(tp); + code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp, + capable(CAP_FOWNER) ? + XFS_QMOPT_FORCE_RES : 0); + if (code) /* out of quota */ + goto error_return; + } + } + + if (mask & FSX_EXTSIZE) { + /* + * Can't change extent size if any extents are allocated. + */ + if (ip->i_d.di_nextents && + ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != + fa->fsx_extsize)) { + code = XFS_ERROR(EINVAL); /* EFBIG? */ + goto error_return; + } + + /* + * Extent size must be a multiple of the appropriate block + * size, if set at all. + */ + if (fa->fsx_extsize != 0) { + xfs_extlen_t size; + + if (XFS_IS_REALTIME_INODE(ip) || + ((mask & FSX_XFLAGS) && + (fa->fsx_xflags & XFS_XFLAG_REALTIME))) { + size = mp->m_sb.sb_rextsize << + mp->m_sb.sb_blocklog; + } else { + size = mp->m_sb.sb_blocksize; + } + + if (fa->fsx_extsize % size) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + } + } + + + if (mask & FSX_XFLAGS) { + /* + * Can't change realtime flag if any extents are allocated. + */ + if ((ip->i_d.di_nextents || ip->i_delayed_blks) && + (XFS_IS_REALTIME_INODE(ip)) != + (fa->fsx_xflags & XFS_XFLAG_REALTIME)) { + code = XFS_ERROR(EINVAL); /* EFBIG? */ + goto error_return; + } + + /* + * If realtime flag is set then must have realtime data. + */ + if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) { + if ((mp->m_sb.sb_rblocks == 0) || + (mp->m_sb.sb_rextsize == 0) || + (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + } + + /* + * Can't modify an immutable/append-only file unless + * we have appropriate permission. + */ + if ((ip->i_d.di_flags & + (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) || + (fa->fsx_xflags & + (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && + !capable(CAP_LINUX_IMMUTABLE)) { + code = XFS_ERROR(EPERM); + goto error_return; + } + } + + xfs_trans_ijoin(tp, ip, lock_flags); + xfs_trans_ihold(tp, ip); + + /* + * Change file ownership. Must be the owner or privileged. + * If the system was configured with the "restricted_chown" + * option, the owner is not permitted to give away the file, + * and can change the group id only to a group of which he + * or she is a member. + */ + if (mask & FSX_PROJID) { + /* + * CAP_FSETID overrides the following restrictions: + * + * The set-user-ID and set-group-ID bits of a file will be + * cleared upon successful return from chown() + */ + if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && + !capable(CAP_FSETID)) + ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); + + /* + * Change the ownerships and register quota modifications + * in the transaction. + */ + if (ip->i_d.di_projid != fa->fsx_projid) { + if (XFS_IS_PQUOTA_ON(mp)) { + olddquot = XFS_QM_DQVOPCHOWN(mp, tp, ip, + &ip->i_gdquot, gdqp); + } + ip->i_d.di_projid = fa->fsx_projid; + + /* + * We may have to rev the inode as well as + * the superblock version number since projids didn't + * exist before DINODE_VERSION_2 and SB_VERSION_NLINK. + */ + if (ip->i_d.di_version == XFS_DINODE_VERSION_1) + xfs_bump_ino_vers2(tp, ip); + } + + } + + if (mask & FSX_EXTSIZE) + ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog; + if (mask & FSX_XFLAGS) { + xfs_set_diflags(ip, fa->fsx_xflags); + xfs_diflags_to_linux(ip); + } + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + xfs_ichgtime(ip, XFS_ICHGTIME_CHG); + + XFS_STATS_INC(xs_ig_attrchg); + + /* + * If this is a synchronous mount, make sure that the + * transaction goes to disk before returning to the user. + * This is slightly sub-optimal in that truncates require + * two sync transactions instead of one for wsync filesystems. + * One for the truncate and one for the timestamps since we + * don't want to change the timestamps unless we're sure the + * truncate worked. Truncates are less than 1% of the laddis + * mix so this probably isn't worth the trouble to optimize. + */ + if (mp->m_flags & XFS_MOUNT_WSYNC) + xfs_trans_set_sync(tp); + code = xfs_trans_commit(tp, 0); + xfs_iunlock(ip, lock_flags); + + /* + * Release any dquot(s) the inode had kept before chown. + */ + XFS_QM_DQRELE(mp, olddquot); + XFS_QM_DQRELE(mp, udqp); + XFS_QM_DQRELE(mp, gdqp); + + if (code) + return code; + + if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE)) { + XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL, + NULL, DM_RIGHT_NULL, NULL, NULL, 0, 0, + (mask & FSX_NONBLOCK) ? DM_FLAGS_NDELAY : 0); + } + + return 0; + + error_return: + XFS_QM_DQRELE(mp, udqp); + XFS_QM_DQRELE(mp, gdqp); + xfs_trans_cancel(tp, 0); + if (lock_flags) + xfs_iunlock(ip, lock_flags); + return code; +} + STATIC int xfs_ioc_fssetxattr( xfs_inode_t *ip, @@ -882,31 +1204,16 @@ xfs_ioc_fssetxattr( void __user *arg) { struct fsxattr fa; - struct bhv_vattr *vattr; - int error; - int attr_flags; + unsigned int mask; if (copy_from_user(&fa, arg, sizeof(fa))) return -EFAULT; - vattr = kmalloc(sizeof(*vattr), GFP_KERNEL); - if (unlikely(!vattr)) - return -ENOMEM; - - attr_flags = 0; + mask = FSX_XFLAGS | FSX_EXTSIZE | FSX_PROJID; if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - attr_flags |= ATTR_NONBLOCK; - - vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | XFS_AT_PROJID; - vattr->va_xflags = fa.fsx_xflags; - vattr->va_extsize = fa.fsx_extsize; - vattr->va_projid = fa.fsx_projid; + mask |= FSX_NONBLOCK; - error = -xfs_setattr(ip, vattr, attr_flags, NULL); - if (!error) - vn_revalidate(XFS_ITOV(ip)); /* update flags */ - kfree(vattr); - return 0; + return -xfs_ioctl_setattr(ip, &fa, mask); } STATIC int @@ -928,10 +1235,9 @@ xfs_ioc_setxflags( struct file *filp, void __user *arg) { - struct bhv_vattr *vattr; + struct fsxattr fa; unsigned int flags; - int attr_flags; - int error; + unsigned int mask; if (copy_from_user(&flags, arg, sizeof(flags))) return -EFAULT; @@ -941,22 +1247,12 @@ xfs_ioc_setxflags( FS_SYNC_FL)) return -EOPNOTSUPP; - vattr = kmalloc(sizeof(*vattr), GFP_KERNEL); - if (unlikely(!vattr)) - return -ENOMEM; - - attr_flags = 0; + mask = FSX_XFLAGS; if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - attr_flags |= ATTR_NONBLOCK; - - vattr->va_mask = XFS_AT_XFLAGS; - vattr->va_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip)); + mask |= FSX_NONBLOCK; + fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip)); - error = -xfs_setattr(ip, vattr, attr_flags, NULL); - if (likely(!error)) - vn_revalidate(XFS_ITOV(ip)); /* update flags */ - kfree(vattr); - return error; + return -xfs_ioctl_setattr(ip, &fa, mask); } STATIC int diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 2bf287e..095d271 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -62,7 +62,7 @@ void xfs_synchronize_atime( xfs_inode_t *ip) { - struct inode *inode = ip->i_vnode; + struct inode *inode = VFS_I(ip); if (inode) { ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec; @@ -79,7 +79,7 @@ void xfs_mark_inode_dirty_sync( xfs_inode_t *ip) { - struct inode *inode = ip->i_vnode; + struct inode *inode = VFS_I(ip); if (inode) mark_inode_dirty_sync(inode); @@ -89,36 +89,31 @@ xfs_mark_inode_dirty_sync( * Change the requested timestamp in the given inode. * We don't lock across timestamp updates, and we don't log them but * we do record the fact that there is dirty information in core. - * - * NOTE -- callers MUST combine XFS_ICHGTIME_MOD or XFS_ICHGTIME_CHG - * with XFS_ICHGTIME_ACC to be sure that access time - * update will take. Calling first with XFS_ICHGTIME_ACC - * and then XFS_ICHGTIME_MOD may fail to modify the access - * timestamp if the filesystem is mounted noacctm. */ void xfs_ichgtime( xfs_inode_t *ip, int flags) { - struct inode *inode = vn_to_inode(XFS_ITOV(ip)); + struct inode *inode = VFS_I(ip); timespec_t tv; + int sync_it = 0; + + tv = current_fs_time(inode->i_sb); - nanotime(&tv); - if (flags & XFS_ICHGTIME_MOD) { + if ((flags & XFS_ICHGTIME_MOD) && + !timespec_equal(&inode->i_mtime, &tv)) { inode->i_mtime = tv; ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec; ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec; + sync_it = 1; } - if (flags & XFS_ICHGTIME_ACC) { - inode->i_atime = tv; - ip->i_d.di_atime.t_sec = (__int32_t)tv.tv_sec; - ip->i_d.di_atime.t_nsec = (__int32_t)tv.tv_nsec; - } - if (flags & XFS_ICHGTIME_CHG) { + if ((flags & XFS_ICHGTIME_CHG) && + !timespec_equal(&inode->i_ctime, &tv)) { inode->i_ctime = tv; ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec; ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec; + sync_it = 1; } /* @@ -130,72 +125,11 @@ xfs_ichgtime( * ensure that the compiler does not reorder the update * of i_update_core above the timestamp updates above. */ - SYNCHRONIZE(); - ip->i_update_core = 1; - if (!(inode->i_state & I_NEW)) + if (sync_it) { + SYNCHRONIZE(); + ip->i_update_core = 1; mark_inode_dirty_sync(inode); -} - -/* - * Variant on the above which avoids querying the system clock - * in situations where we know the Linux inode timestamps have - * just been updated (and so we can update our inode cheaply). - */ -void -xfs_ichgtime_fast( - xfs_inode_t *ip, - struct inode *inode, - int flags) -{ - timespec_t *tvp; - - /* - * Atime updates for read() & friends are handled lazily now, and - * explicit updates must go through xfs_ichgtime() - */ - ASSERT((flags & XFS_ICHGTIME_ACC) == 0); - - if (flags & XFS_ICHGTIME_MOD) { - tvp = &inode->i_mtime; - ip->i_d.di_mtime.t_sec = (__int32_t)tvp->tv_sec; - ip->i_d.di_mtime.t_nsec = (__int32_t)tvp->tv_nsec; - } - if (flags & XFS_ICHGTIME_CHG) { - tvp = &inode->i_ctime; - ip->i_d.di_ctime.t_sec = (__int32_t)tvp->tv_sec; - ip->i_d.di_ctime.t_nsec = (__int32_t)tvp->tv_nsec; } - - /* - * We update the i_update_core field _after_ changing - * the timestamps in order to coordinate properly with - * xfs_iflush() so that we don't lose timestamp updates. - * This keeps us from having to hold the inode lock - * while doing this. We use the SYNCHRONIZE macro to - * ensure that the compiler does not reorder the update - * of i_update_core above the timestamp updates above. - */ - SYNCHRONIZE(); - ip->i_update_core = 1; - if (!(inode->i_state & I_NEW)) - mark_inode_dirty_sync(inode); -} - - -/* - * Pull the link count and size up from the xfs inode to the linux inode - */ -STATIC void -xfs_validate_fields( - struct inode *inode) -{ - struct xfs_inode *ip = XFS_I(inode); - loff_t size; - - /* we're under i_sem so i_size can't change under us */ - size = XFS_ISIZE(ip); - if (i_size_read(inode) != size) - i_size_write(inode, size); } /* @@ -245,8 +179,7 @@ STATIC void xfs_cleanup_inode( struct inode *dir, struct inode *inode, - struct dentry *dentry, - int mode) + struct dentry *dentry) { struct xfs_name teardown; @@ -257,10 +190,7 @@ xfs_cleanup_inode( */ xfs_dentry_to_name(&teardown, dentry); - if (S_ISDIR(mode)) - xfs_rmdir(XFS_I(dir), &teardown, XFS_I(inode)); - else - xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); + xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); iput(inode); } @@ -275,7 +205,7 @@ xfs_vn_mknod( struct xfs_inode *ip = NULL; xfs_acl_t *default_acl = NULL; struct xfs_name name; - attrexists_t test_default_acl = _ACL_DEFAULT_EXISTS; + int (*test_default_acl)(struct inode *) = _ACL_DEFAULT_EXISTS; int error; /* @@ -320,7 +250,7 @@ xfs_vn_mknod( if (unlikely(error)) goto out_free_acl; - inode = ip->i_vnode; + inode = VFS_I(ip); error = xfs_init_security(inode, dir); if (unlikely(error)) @@ -335,14 +265,11 @@ xfs_vn_mknod( } - if (S_ISDIR(mode)) - xfs_validate_fields(inode); d_instantiate(dentry, inode); - xfs_validate_fields(dir); return -error; out_cleanup_inode: - xfs_cleanup_inode(dir, inode, dentry, mode); + xfs_cleanup_inode(dir, inode, dentry); out_free_acl: if (default_acl) _ACL_FREE(default_acl); @@ -382,7 +309,7 @@ xfs_vn_lookup( return ERR_PTR(-ENAMETOOLONG); xfs_dentry_to_name(&name, dentry); - error = xfs_lookup(XFS_I(dir), &name, &cip); + error = xfs_lookup(XFS_I(dir), &name, &cip, NULL); if (unlikely(error)) { if (unlikely(error != ENOENT)) return ERR_PTR(-error); @@ -390,7 +317,47 @@ xfs_vn_lookup( return NULL; } - return d_splice_alias(cip->i_vnode, dentry); + return d_splice_alias(VFS_I(cip), dentry); +} + +STATIC struct dentry * +xfs_vn_ci_lookup( + struct inode *dir, + struct dentry *dentry, + struct nameidata *nd) +{ + struct xfs_inode *ip; + struct xfs_name xname; + struct xfs_name ci_name; + struct qstr dname; + int error; + + if (dentry->d_name.len >= MAXNAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + xfs_dentry_to_name(&xname, dentry); + error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name); + if (unlikely(error)) { + if (unlikely(error != ENOENT)) + return ERR_PTR(-error); + /* + * call d_add(dentry, NULL) here when d_drop_negative_children + * is called in xfs_vn_mknod (ie. allow negative dentries + * with CI filesystems). + */ + return NULL; + } + + /* if exact match, just splice and exit */ + if (!ci_name.name) + return d_splice_alias(VFS_I(ip), dentry); + + /* else case-insensitive match... */ + dname.name = ci_name.name; + dname.len = ci_name.len; + dentry = d_add_ci(dentry, VFS_I(ip), &dname); + kmem_free(ci_name.name); + return dentry; } STATIC int @@ -414,7 +381,6 @@ xfs_vn_link( } xfs_iflags_set(XFS_I(dir), XFS_IMODIFIED); - xfs_validate_fields(inode); d_instantiate(dentry, inode); return 0; } @@ -424,19 +390,23 @@ xfs_vn_unlink( struct inode *dir, struct dentry *dentry) { - struct inode *inode; struct xfs_name name; int error; - inode = dentry->d_inode; xfs_dentry_to_name(&name, dentry); - error = xfs_remove(XFS_I(dir), &name, XFS_I(inode)); - if (likely(!error)) { - xfs_validate_fields(dir); /* size needs update */ - xfs_validate_fields(inode); - } - return -error; + error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode)); + if (error) + return error; + + /* + * With unlink, the VFS makes the dentry "negative": no inode, + * but still hashed. This is incompatible with case-insensitive + * mode, so invalidate (unhash) the dentry in CI-mode. + */ + if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb)) + d_invalidate(dentry); + return 0; } STATIC int @@ -459,43 +429,22 @@ xfs_vn_symlink( if (unlikely(error)) goto out; - inode = cip->i_vnode; + inode = VFS_I(cip); error = xfs_init_security(inode, dir); if (unlikely(error)) goto out_cleanup_inode; d_instantiate(dentry, inode); - xfs_validate_fields(dir); - xfs_validate_fields(inode); return 0; out_cleanup_inode: - xfs_cleanup_inode(dir, inode, dentry, 0); + xfs_cleanup_inode(dir, inode, dentry); out: return -error; } STATIC int -xfs_vn_rmdir( - struct inode *dir, - struct dentry *dentry) -{ - struct inode *inode = dentry->d_inode; - struct xfs_name name; - int error; - - xfs_dentry_to_name(&name, dentry); - - error = xfs_rmdir(XFS_I(dir), &name, XFS_I(inode)); - if (likely(!error)) { - xfs_validate_fields(inode); - xfs_validate_fields(dir); - } - return -error; -} - -STATIC int xfs_vn_rename( struct inode *odir, struct dentry *odentry, @@ -505,22 +454,13 @@ xfs_vn_rename( struct inode *new_inode = ndentry->d_inode; struct xfs_name oname; struct xfs_name nname; - int error; xfs_dentry_to_name(&oname, odentry); xfs_dentry_to_name(&nname, ndentry); - error = xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), + return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), XFS_I(ndir), &nname, new_inode ? XFS_I(new_inode) : NULL); - if (likely(!error)) { - if (new_inode) - xfs_validate_fields(new_inode); - xfs_validate_fields(odir); - if (ndir != odir) - xfs_validate_fields(ndir); - } - return -error; } /* @@ -589,8 +529,7 @@ xfs_check_acl( STATIC int xfs_vn_permission( struct inode *inode, - int mask, - struct nameidata *nd) + int mask) { return generic_permission(inode, mask, xfs_check_acl); } @@ -660,57 +599,9 @@ xfs_vn_getattr( STATIC int xfs_vn_setattr( struct dentry *dentry, - struct iattr *attr) + struct iattr *iattr) { - struct inode *inode = dentry->d_inode; - unsigned int ia_valid = attr->ia_valid; - bhv_vattr_t vattr = { 0 }; - int flags = 0; - int error; - - if (ia_valid & ATTR_UID) { - vattr.va_mask |= XFS_AT_UID; - vattr.va_uid = attr->ia_uid; - } - if (ia_valid & ATTR_GID) { - vattr.va_mask |= XFS_AT_GID; - vattr.va_gid = attr->ia_gid; - } - if (ia_valid & ATTR_SIZE) { - vattr.va_mask |= XFS_AT_SIZE; - vattr.va_size = attr->ia_size; - } - if (ia_valid & ATTR_ATIME) { - vattr.va_mask |= XFS_AT_ATIME; - vattr.va_atime = attr->ia_atime; - inode->i_atime = attr->ia_atime; - } - if (ia_valid & ATTR_MTIME) { - vattr.va_mask |= XFS_AT_MTIME; - vattr.va_mtime = attr->ia_mtime; - } - if (ia_valid & ATTR_CTIME) { - vattr.va_mask |= XFS_AT_CTIME; - vattr.va_ctime = attr->ia_ctime; - } - if (ia_valid & ATTR_MODE) { - vattr.va_mask |= XFS_AT_MODE; - vattr.va_mode = attr->ia_mode; - if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) - inode->i_mode &= ~S_ISGID; - } - - if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) - flags |= ATTR_UTIME; -#ifdef ATTR_NO_BLOCK - if ((ia_valid & ATTR_NO_BLOCK)) - flags |= ATTR_NONBLOCK; -#endif - - error = xfs_setattr(XFS_I(inode), &vattr, flags, NULL); - if (likely(!error)) - vn_revalidate(vn_from_inode(inode)); - return -error; + return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0, NULL); } /* @@ -728,109 +619,6 @@ xfs_vn_truncate( WARN_ON(error); } -STATIC int -xfs_vn_setxattr( - struct dentry *dentry, - const char *name, - const void *data, - size_t size, - int flags) -{ - bhv_vnode_t *vp = vn_from_inode(dentry->d_inode); - char *attr = (char *)name; - attrnames_t *namesp; - int xflags = 0; - int error; - - namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT); - if (!namesp) - return -EOPNOTSUPP; - attr += namesp->attr_namelen; - error = namesp->attr_capable(vp, NULL); - if (error) - return error; - - /* Convert Linux syscall to XFS internal ATTR flags */ - if (flags & XATTR_CREATE) - xflags |= ATTR_CREATE; - if (flags & XATTR_REPLACE) - xflags |= ATTR_REPLACE; - xflags |= namesp->attr_flag; - return namesp->attr_set(vp, attr, (void *)data, size, xflags); -} - -STATIC ssize_t -xfs_vn_getxattr( - struct dentry *dentry, - const char *name, - void *data, - size_t size) -{ - bhv_vnode_t *vp = vn_from_inode(dentry->d_inode); - char *attr = (char *)name; - attrnames_t *namesp; - int xflags = 0; - ssize_t error; - - namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT); - if (!namesp) - return -EOPNOTSUPP; - attr += namesp->attr_namelen; - error = namesp->attr_capable(vp, NULL); - if (error) - return error; - - /* Convert Linux syscall to XFS internal ATTR flags */ - if (!size) { - xflags |= ATTR_KERNOVAL; - data = NULL; - } - xflags |= namesp->attr_flag; - return namesp->attr_get(vp, attr, (void *)data, size, xflags); -} - -STATIC ssize_t -xfs_vn_listxattr( - struct dentry *dentry, - char *data, - size_t size) -{ - bhv_vnode_t *vp = vn_from_inode(dentry->d_inode); - int error, xflags = ATTR_KERNAMELS; - ssize_t result; - - if (!size) - xflags |= ATTR_KERNOVAL; - xflags |= capable(CAP_SYS_ADMIN) ? ATTR_KERNFULLS : ATTR_KERNORMALS; - - error = attr_generic_list(vp, data, size, xflags, &result); - if (error < 0) - return error; - return result; -} - -STATIC int -xfs_vn_removexattr( - struct dentry *dentry, - const char *name) -{ - bhv_vnode_t *vp = vn_from_inode(dentry->d_inode); - char *attr = (char *)name; - attrnames_t *namesp; - int xflags = 0; - int error; - - namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT); - if (!namesp) - return -EOPNOTSUPP; - attr += namesp->attr_namelen; - error = namesp->attr_capable(vp, NULL); - if (error) - return error; - xflags |= namesp->attr_flag; - return namesp->attr_remove(vp, attr, xflags); -} - STATIC long xfs_vn_fallocate( struct inode *inode, @@ -854,18 +642,18 @@ xfs_vn_fallocate( xfs_ilock(ip, XFS_IOLOCK_EXCL); error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf, - 0, NULL, ATTR_NOLOCK); + 0, NULL, XFS_ATTR_NOLOCK); if (!error && !(mode & FALLOC_FL_KEEP_SIZE) && offset + len > i_size_read(inode)) new_size = offset + len; /* Change file size if needed */ if (new_size) { - bhv_vattr_t va; + struct iattr iattr; - va.va_mask = XFS_AT_SIZE; - va.va_size = new_size; - error = xfs_setattr(ip, &va, ATTR_NOLOCK, NULL); + iattr.ia_valid = ATTR_SIZE; + iattr.ia_size = new_size; + error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK, NULL); } xfs_iunlock(ip, XFS_IOLOCK_EXCL); @@ -873,46 +661,172 @@ out_error: return error; } -const struct inode_operations xfs_inode_operations = { +static const struct inode_operations xfs_inode_operations = { .permission = xfs_vn_permission, .truncate = xfs_vn_truncate, .getattr = xfs_vn_getattr, .setattr = xfs_vn_setattr, - .setxattr = xfs_vn_setxattr, - .getxattr = xfs_vn_getxattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, .listxattr = xfs_vn_listxattr, - .removexattr = xfs_vn_removexattr, .fallocate = xfs_vn_fallocate, }; -const struct inode_operations xfs_dir_inode_operations = { +static const struct inode_operations xfs_dir_inode_operations = { .create = xfs_vn_create, .lookup = xfs_vn_lookup, .link = xfs_vn_link, .unlink = xfs_vn_unlink, .symlink = xfs_vn_symlink, .mkdir = xfs_vn_mkdir, - .rmdir = xfs_vn_rmdir, + /* + * Yes, XFS uses the same method for rmdir and unlink. + * + * There are some subtile differences deeper in the code, + * but we use S_ISDIR to check for those. + */ + .rmdir = xfs_vn_unlink, .mknod = xfs_vn_mknod, .rename = xfs_vn_rename, .permission = xfs_vn_permission, .getattr = xfs_vn_getattr, .setattr = xfs_vn_setattr, - .setxattr = xfs_vn_setxattr, - .getxattr = xfs_vn_getxattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, .listxattr = xfs_vn_listxattr, - .removexattr = xfs_vn_removexattr, }; -const struct inode_operations xfs_symlink_inode_operations = { +static const struct inode_operations xfs_dir_ci_inode_operations = { + .create = xfs_vn_create, + .lookup = xfs_vn_ci_lookup, + .link = xfs_vn_link, + .unlink = xfs_vn_unlink, + .symlink = xfs_vn_symlink, + .mkdir = xfs_vn_mkdir, + /* + * Yes, XFS uses the same method for rmdir and unlink. + * + * There are some subtile differences deeper in the code, + * but we use S_ISDIR to check for those. + */ + .rmdir = xfs_vn_unlink, + .mknod = xfs_vn_mknod, + .rename = xfs_vn_rename, + .permission = xfs_vn_permission, + .getattr = xfs_vn_getattr, + .setattr = xfs_vn_setattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, + .listxattr = xfs_vn_listxattr, +}; + +static const struct inode_operations xfs_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = xfs_vn_follow_link, .put_link = xfs_vn_put_link, .permission = xfs_vn_permission, .getattr = xfs_vn_getattr, .setattr = xfs_vn_setattr, - .setxattr = xfs_vn_setxattr, - .getxattr = xfs_vn_getxattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, .listxattr = xfs_vn_listxattr, - .removexattr = xfs_vn_removexattr, }; + +STATIC void +xfs_diflags_to_iflags( + struct inode *inode, + struct xfs_inode *ip) +{ + if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) + inode->i_flags |= S_IMMUTABLE; + else + inode->i_flags &= ~S_IMMUTABLE; + if (ip->i_d.di_flags & XFS_DIFLAG_APPEND) + inode->i_flags |= S_APPEND; + else + inode->i_flags &= ~S_APPEND; + if (ip->i_d.di_flags & XFS_DIFLAG_SYNC) + inode->i_flags |= S_SYNC; + else + inode->i_flags &= ~S_SYNC; + if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME) + inode->i_flags |= S_NOATIME; + else + inode->i_flags &= ~S_NOATIME; +} + +/* + * Initialize the Linux inode, set up the operation vectors and + * unlock the inode. + * + * When reading existing inodes from disk this is called directly + * from xfs_iget, when creating a new inode it is called from + * xfs_ialloc after setting up the inode. + */ +void +xfs_setup_inode( + struct xfs_inode *ip) +{ + struct inode *inode = ip->i_vnode; + + inode->i_mode = ip->i_d.di_mode; + inode->i_nlink = ip->i_d.di_nlink; + inode->i_uid = ip->i_d.di_uid; + inode->i_gid = ip->i_d.di_gid; + + switch (inode->i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + inode->i_rdev = + MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, + sysv_minor(ip->i_df.if_u2.if_rdev)); + break; + default: + inode->i_rdev = 0; + break; + } + + inode->i_generation = ip->i_d.di_gen; + i_size_write(inode, ip->i_d.di_size); + inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec; + inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec; + inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; + inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; + inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; + inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; + xfs_diflags_to_iflags(inode, ip); + xfs_iflags_clear(ip, XFS_IMODIFIED); + + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + inode->i_op = &xfs_inode_operations; + inode->i_fop = &xfs_file_operations; + inode->i_mapping->a_ops = &xfs_address_space_operations; + break; + case S_IFDIR: + if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb)) + inode->i_op = &xfs_dir_ci_inode_operations; + else + inode->i_op = &xfs_dir_inode_operations; + inode->i_fop = &xfs_dir_file_operations; + break; + case S_IFLNK: + inode->i_op = &xfs_symlink_inode_operations; + if (!(ip->i_df.if_flags & XFS_IFINLINE)) + inode->i_mapping->a_ops = &xfs_address_space_operations; + break; + default: + inode->i_op = &xfs_inode_operations; + init_special_inode(inode, inode->i_mode, inode->i_rdev); + break; + } + + xfs_iflags_clear(ip, XFS_INEW); + barrier(); + + unlock_new_inode(inode); +} diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h index 14d0deb..8b1a1e3 100644 --- a/fs/xfs/linux-2.6/xfs_iops.h +++ b/fs/xfs/linux-2.6/xfs_iops.h @@ -18,23 +18,14 @@ #ifndef __XFS_IOPS_H__ #define __XFS_IOPS_H__ -extern const struct inode_operations xfs_inode_operations; -extern const struct inode_operations xfs_dir_inode_operations; -extern const struct inode_operations xfs_symlink_inode_operations; +struct xfs_inode; extern const struct file_operations xfs_file_operations; extern const struct file_operations xfs_dir_file_operations; extern const struct file_operations xfs_invis_file_operations; +extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); -struct xfs_inode; -extern void xfs_ichgtime(struct xfs_inode *, int); -extern void xfs_ichgtime_fast(struct xfs_inode *, struct inode *, int); - -#define xfs_vtoi(vp) \ - ((struct xfs_inode *)vn_to_inode(vp)->i_private) - -#define XFS_I(inode) \ - ((struct xfs_inode *)(inode)->i_private) +extern void xfs_setup_inode(struct xfs_inode *); #endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 4edc469..cc0f7b3 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -45,13 +45,13 @@ #include <mrlock.h> #include <sv.h> #include <mutex.h> -#include <sema.h> #include <time.h> #include <support/ktrace.h> #include <support/debug.h> #include <support/uuid.h> +#include <linux/semaphore.h> #include <linux/mm.h> #include <linux/kernel.h> #include <linux/blkdev.h> @@ -76,6 +76,7 @@ #include <linux/log2.h> #include <linux/spinlock.h> #include <linux/random.h> +#include <linux/ctype.h> #include <asm/page.h> #include <asm/div64.h> @@ -125,8 +126,6 @@ #define current_cpu() (raw_smp_processor_id()) #define current_pid() (current->pid) -#define current_fsuid(cred) (current->fsuid) -#define current_fsgid(cred) (current->fsgid) #define current_test_flags(f) (current->flags & (f)) #define current_set_flags_nested(sp, f) \ (*(sp) = current->flags, current->flags |= (f)) @@ -179,7 +178,7 @@ #define xfs_sort(a,n,s,fn) sort(a,n,s,fn,NULL) #define xfs_stack_trace() dump_stack() #define xfs_itruncate_data(ip, off) \ - (-vmtruncate(vn_to_inode(XFS_ITOV(ip)), (off))) + (-vmtruncate(VFS_I(ip), (off))) /* Move the kernel do_div definition off to one side */ @@ -299,4 +298,11 @@ static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y) return x; } +/* ARM old ABI has some weird alignment/padding */ +#if defined(__arm__) && !defined(__ARM_EABI__) +#define __arch_pack __attribute__((packed)) +#else +#define __arch_pack +#endif + #endif /* __XFS_LINUX__ */ diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 5e3b575..1957e535 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -137,7 +137,7 @@ xfs_iozero( struct address_space *mapping; int status; - mapping = ip->i_vnode->i_mapping; + mapping = VFS_I(ip)->i_mapping; do { unsigned offset, bytes; void *fsdata; @@ -674,9 +674,7 @@ start: */ if (likely(!(ioflags & IO_INVIS) && !mnt_want_write(file->f_path.mnt))) { - file_update_time(file); - xfs_ichgtime_fast(xip, inode, - XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); mnt_drop_write(file->f_path.mnt); } @@ -711,7 +709,7 @@ start: !capable(CAP_FSETID)) { error = xfs_write_clear_setuid(xip); if (likely(!error)) - error = -remove_suid(file->f_path.dentry); + error = -file_remove_suid(file); if (unlikely(error)) { goto out_unlock_internal; } diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c index e480b61..3d5b67c 100644 --- a/fs/xfs/linux-2.6/xfs_stats.c +++ b/fs/xfs/linux-2.6/xfs_stats.c @@ -98,12 +98,21 @@ xfs_read_xfsstats( return len; } -void +int xfs_init_procfs(void) { if (!proc_mkdir("fs/xfs", NULL)) - return; - create_proc_read_entry("fs/xfs/stat", 0, NULL, xfs_read_xfsstats, NULL); + goto out; + + if (!create_proc_read_entry("fs/xfs/stat", 0, NULL, + xfs_read_xfsstats, NULL)) + goto out_remove_entry; + return 0; + + out_remove_entry: + remove_proc_entry("fs/xfs", NULL); + out: + return -ENOMEM; } void diff --git a/fs/xfs/linux-2.6/xfs_stats.h b/fs/xfs/linux-2.6/xfs_stats.h index afd0b0d..e83820f 100644 --- a/fs/xfs/linux-2.6/xfs_stats.h +++ b/fs/xfs/linux-2.6/xfs_stats.h @@ -134,7 +134,7 @@ DECLARE_PER_CPU(struct xfsstats, xfsstats); #define XFS_STATS_DEC(v) (per_cpu(xfsstats, current_cpu()).v--) #define XFS_STATS_ADD(v, inc) (per_cpu(xfsstats, current_cpu()).v += (inc)) -extern void xfs_init_procfs(void); +extern int xfs_init_procfs(void); extern void xfs_cleanup_procfs(void); @@ -144,8 +144,14 @@ extern void xfs_cleanup_procfs(void); # define XFS_STATS_DEC(count) # define XFS_STATS_ADD(count, inc) -static inline void xfs_init_procfs(void) { }; -static inline void xfs_cleanup_procfs(void) { }; +static inline int xfs_init_procfs(void) +{ + return 0; +} + +static inline void xfs_cleanup_procfs(void) +{ +} #endif /* !CONFIG_PROC_FS */ diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 742b2c7..18d3c84 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -52,6 +52,12 @@ #include "xfs_version.h" #include "xfs_log_priv.h" #include "xfs_trans_priv.h" +#include "xfs_filestream.h" +#include "xfs_da_btree.h" +#include "xfs_dir2_trace.h" +#include "xfs_extfree_item.h" +#include "xfs_mru_cache.h" +#include "xfs_inode_item.h" #include <linux/namei.h> #include <linux/init.h> @@ -60,6 +66,7 @@ #include <linux/writeback.h> #include <linux/kthread.h> #include <linux/freezer.h> +#include <linux/parser.h> static struct quotactl_ops xfs_quotactl_operations; static struct super_operations xfs_super_operations; @@ -74,7 +81,10 @@ xfs_args_allocate( { struct xfs_mount_args *args; - args = kmem_zalloc(sizeof(struct xfs_mount_args), KM_SLEEP); + args = kzalloc(sizeof(struct xfs_mount_args), GFP_KERNEL); + if (!args) + return NULL; + args->logbufs = args->logbufsize = -1; strncpy(args->fsname, sb->s_id, MAXNAMELEN); @@ -138,6 +148,23 @@ xfs_args_allocate( #define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */ #define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */ +/* + * Table driven mount option parser. + * + * Currently only used for remount, but it will be used for mount + * in the future, too. + */ +enum { + Opt_barrier, Opt_nobarrier, Opt_err +}; + +static match_table_t tokens = { + {Opt_barrier, "barrier"}, + {Opt_nobarrier, "nobarrier"}, + {Opt_err, NULL} +}; + + STATIC unsigned long suffix_strtoul(char *s, char **endp, unsigned int base) { @@ -314,6 +341,7 @@ xfs_parseargs( args->flags |= XFSMNT_ATTR2; } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { args->flags &= ~XFSMNT_ATTR2; + args->flags |= XFSMNT_NOATTR2; } else if (!strcmp(this_char, MNTOPT_FILESTREAM)) { args->flags2 |= XFSMNT2_FILESTREAMS; } else if (!strcmp(this_char, MNTOPT_NOQUOTA)) { @@ -553,115 +581,6 @@ xfs_max_file_offset( return (((__uint64_t)pagefactor) << bitshift) - 1; } -STATIC_INLINE void -xfs_set_inodeops( - struct inode *inode) -{ - switch (inode->i_mode & S_IFMT) { - case S_IFREG: - inode->i_op = &xfs_inode_operations; - inode->i_fop = &xfs_file_operations; - inode->i_mapping->a_ops = &xfs_address_space_operations; - break; - case S_IFDIR: - inode->i_op = &xfs_dir_inode_operations; - inode->i_fop = &xfs_dir_file_operations; - break; - case S_IFLNK: - inode->i_op = &xfs_symlink_inode_operations; - if (!(XFS_I(inode)->i_df.if_flags & XFS_IFINLINE)) - inode->i_mapping->a_ops = &xfs_address_space_operations; - break; - default: - inode->i_op = &xfs_inode_operations; - init_special_inode(inode, inode->i_mode, inode->i_rdev); - break; - } -} - -STATIC_INLINE void -xfs_revalidate_inode( - xfs_mount_t *mp, - bhv_vnode_t *vp, - xfs_inode_t *ip) -{ - struct inode *inode = vn_to_inode(vp); - - inode->i_mode = ip->i_d.di_mode; - inode->i_nlink = ip->i_d.di_nlink; - inode->i_uid = ip->i_d.di_uid; - inode->i_gid = ip->i_d.di_gid; - - switch (inode->i_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - inode->i_rdev = - MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, - sysv_minor(ip->i_df.if_u2.if_rdev)); - break; - default: - inode->i_rdev = 0; - break; - } - - inode->i_generation = ip->i_d.di_gen; - i_size_write(inode, ip->i_d.di_size); - inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec; - inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec; - inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; - inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; - inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; - inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; - if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - if (ip->i_d.di_flags & XFS_DIFLAG_APPEND) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; - if (ip->i_d.di_flags & XFS_DIFLAG_SYNC) - inode->i_flags |= S_SYNC; - else - inode->i_flags &= ~S_SYNC; - if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME) - inode->i_flags |= S_NOATIME; - else - inode->i_flags &= ~S_NOATIME; - xfs_iflags_clear(ip, XFS_IMODIFIED); -} - -void -xfs_initialize_vnode( - struct xfs_mount *mp, - bhv_vnode_t *vp, - struct xfs_inode *ip) -{ - struct inode *inode = vn_to_inode(vp); - - if (!ip->i_vnode) { - ip->i_vnode = vp; - inode->i_private = ip; - } - - /* - * We need to set the ops vectors, and unlock the inode, but if - * we have been called during the new inode create process, it is - * too early to fill in the Linux inode. We will get called a - * second time once the inode is properly set up, and then we can - * finish our work. - */ - if (ip->i_d.di_mode != 0 && (inode->i_state & I_NEW)) { - xfs_revalidate_inode(mp, vp, ip); - xfs_set_inodeops(inode); - - xfs_iflags_clear(ip, XFS_INEW); - barrier(); - - unlock_new_inode(inode); - } -} - int xfs_blkdev_get( xfs_mount_t *mp, @@ -733,14 +652,6 @@ xfs_mountfs_check_barriers(xfs_mount_t *mp) return; } - if (mp->m_ddev_targp->bt_bdev->bd_disk->queue->ordered == - QUEUE_ORDERED_NONE) { - xfs_fs_cmn_err(CE_NOTE, mp, - "Disabling barriers, not supported by the underlying device"); - mp->m_flags &= ~XFS_MOUNT_BARRIER; - return; - } - if (xfs_readonly_buftarg(mp->m_ddev_targp)) { xfs_fs_cmn_err(CE_NOTE, mp, "Disabling barriers, underlying device is readonly"); @@ -764,6 +675,139 @@ xfs_blkdev_issue_flush( blkdev_issue_flush(buftarg->bt_bdev, NULL); } +STATIC void +xfs_close_devices( + struct xfs_mount *mp) +{ + if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { + struct block_device *logdev = mp->m_logdev_targp->bt_bdev; + xfs_free_buftarg(mp->m_logdev_targp); + xfs_blkdev_put(logdev); + } + if (mp->m_rtdev_targp) { + struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev; + xfs_free_buftarg(mp->m_rtdev_targp); + xfs_blkdev_put(rtdev); + } + xfs_free_buftarg(mp->m_ddev_targp); +} + +/* + * The file system configurations are: + * (1) device (partition) with data and internal log + * (2) logical volume with data and log subvolumes. + * (3) logical volume with data, log, and realtime subvolumes. + * + * We only have to handle opening the log and realtime volumes here if + * they are present. The data subvolume has already been opened by + * get_sb_bdev() and is stored in sb->s_bdev. + */ +STATIC int +xfs_open_devices( + struct xfs_mount *mp, + struct xfs_mount_args *args) +{ + struct block_device *ddev = mp->m_super->s_bdev; + struct block_device *logdev = NULL, *rtdev = NULL; + int error; + + /* + * Open real time and log devices - order is important. + */ + if (args->logname[0]) { + error = xfs_blkdev_get(mp, args->logname, &logdev); + if (error) + goto out; + } + + if (args->rtname[0]) { + error = xfs_blkdev_get(mp, args->rtname, &rtdev); + if (error) + goto out_close_logdev; + + if (rtdev == ddev || rtdev == logdev) { + cmn_err(CE_WARN, + "XFS: Cannot mount filesystem with identical rtdev and ddev/logdev."); + error = EINVAL; + goto out_close_rtdev; + } + } + + /* + * Setup xfs_mount buffer target pointers + */ + error = ENOMEM; + mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0); + if (!mp->m_ddev_targp) + goto out_close_rtdev; + + if (rtdev) { + mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1); + if (!mp->m_rtdev_targp) + goto out_free_ddev_targ; + } + + if (logdev && logdev != ddev) { + mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1); + if (!mp->m_logdev_targp) + goto out_free_rtdev_targ; + } else { + mp->m_logdev_targp = mp->m_ddev_targp; + } + + return 0; + + out_free_rtdev_targ: + if (mp->m_rtdev_targp) + xfs_free_buftarg(mp->m_rtdev_targp); + out_free_ddev_targ: + xfs_free_buftarg(mp->m_ddev_targp); + out_close_rtdev: + if (rtdev) + xfs_blkdev_put(rtdev); + out_close_logdev: + if (logdev && logdev != ddev) + xfs_blkdev_put(logdev); + out: + return error; +} + +/* + * Setup xfs_mount buffer target pointers based on superblock + */ +STATIC int +xfs_setup_devices( + struct xfs_mount *mp) +{ + int error; + + error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize, + mp->m_sb.sb_sectsize); + if (error) + return error; + + if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { + unsigned int log_sector_size = BBSIZE; + + if (xfs_sb_version_hassector(&mp->m_sb)) + log_sector_size = mp->m_sb.sb_logsectsize; + error = xfs_setsize_buftarg(mp->m_logdev_targp, + mp->m_sb.sb_blocksize, + log_sector_size); + if (error) + return error; + } + if (mp->m_rtdev_targp) { + error = xfs_setsize_buftarg(mp->m_rtdev_targp, + mp->m_sb.sb_blocksize, + mp->m_sb.sb_sectsize); + if (error) + return error; + } + + return 0; +} + /* * XFS AIL push thread support */ @@ -826,63 +870,21 @@ STATIC struct inode * xfs_fs_alloc_inode( struct super_block *sb) { - bhv_vnode_t *vp; - - vp = kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP); - if (unlikely(!vp)) - return NULL; - return vn_to_inode(vp); + return kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP); } STATIC void xfs_fs_destroy_inode( struct inode *inode) { - kmem_zone_free(xfs_vnode_zone, vn_from_inode(inode)); + kmem_zone_free(xfs_vnode_zone, inode); } STATIC void xfs_fs_inode_init_once( - kmem_zone_t *zonep, void *vnode) { - inode_init_once(vn_to_inode((bhv_vnode_t *)vnode)); -} - -STATIC int __init -xfs_init_zones(void) -{ - xfs_vnode_zone = kmem_zone_init_flags(sizeof(bhv_vnode_t), "xfs_vnode", - KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | - KM_ZONE_SPREAD, - xfs_fs_inode_init_once); - if (!xfs_vnode_zone) - goto out; - - xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend"); - if (!xfs_ioend_zone) - goto out_destroy_vnode_zone; - - xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE, - xfs_ioend_zone); - if (!xfs_ioend_pool) - goto out_free_ioend_zone; - return 0; - - out_free_ioend_zone: - kmem_zone_destroy(xfs_ioend_zone); - out_destroy_vnode_zone: - kmem_zone_destroy(xfs_vnode_zone); - out: - return -ENOMEM; -} - -STATIC void -xfs_destroy_zones(void) -{ - mempool_destroy(xfs_ioend_pool); - kmem_zone_destroy(xfs_vnode_zone); - kmem_zone_destroy(xfs_ioend_zone); + inode_init_once((struct inode *)vnode); } /* @@ -987,7 +989,7 @@ void xfs_flush_inode( xfs_inode_t *ip) { - struct inode *inode = ip->i_vnode; + struct inode *inode = VFS_I(ip); igrab(inode); xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work); @@ -1012,7 +1014,7 @@ void xfs_flush_device( xfs_inode_t *ip) { - struct inode *inode = vn_to_inode(XFS_ITOV(ip)); + struct inode *inode = VFS_I(ip); igrab(inode); xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work); @@ -1074,7 +1076,7 @@ xfssyncd( list_del(&work->w_list); if (work == &mp->m_sync_work) continue; - kmem_free(work, sizeof(struct bhv_vfs_sync_work)); + kmem_free(work); } } @@ -1082,18 +1084,76 @@ xfssyncd( } STATIC void +xfs_free_fsname( + struct xfs_mount *mp) +{ + kfree(mp->m_fsname); + kfree(mp->m_rtname); + kfree(mp->m_logname); +} + +STATIC void xfs_fs_put_super( struct super_block *sb) { struct xfs_mount *mp = XFS_M(sb); + struct xfs_inode *rip = mp->m_rootip; + int unmount_event_flags = 0; int error; kthread_stop(mp->m_sync_task); xfs_sync(mp, SYNC_ATTR | SYNC_DELWRI); - error = xfs_unmount(mp, 0, NULL); - if (error) - printk("XFS: unmount got error=%d\n", error); + +#ifdef HAVE_DMAPI + if (mp->m_flags & XFS_MOUNT_DMAPI) { + unmount_event_flags = + (mp->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ? + 0 : DM_FLAGS_UNWANTED; + /* + * Ignore error from dmapi here, first unmount is not allowed + * to fail anyway, and second we wouldn't want to fail a + * unmount because of dmapi. + */ + XFS_SEND_PREUNMOUNT(mp, rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL, + NULL, NULL, 0, 0, unmount_event_flags); + } +#endif + + /* + * Blow away any referenced inode in the filestreams cache. + * This can and will cause log traffic as inodes go inactive + * here. + */ + xfs_filestream_unmount(mp); + + XFS_bflush(mp->m_ddev_targp); + error = xfs_unmount_flush(mp, 0); + WARN_ON(error); + + /* + * If we're forcing a shutdown, typically because of a media error, + * we want to make sure we invalidate dirty pages that belong to + * referenced vnodes as well. + */ + if (XFS_FORCED_SHUTDOWN(mp)) { + error = xfs_sync(mp, SYNC_WAIT | SYNC_CLOSE); + ASSERT(error != EFSCORRUPTED); + } + + if (mp->m_flags & XFS_MOUNT_DMAPI) { + XFS_SEND_UNMOUNT(mp, rip, DM_RIGHT_NULL, 0, 0, + unmount_event_flags); + } + + xfs_unmountfs(mp); + xfs_freesb(mp); + xfs_icsb_destroy_counters(mp); + xfs_close_devices(mp); + xfs_qmops_put(mp); + xfs_dmops_put(mp); + xfs_free_fsname(mp); + kfree(mp); } STATIC void @@ -1216,14 +1276,74 @@ xfs_fs_remount( char *options) { struct xfs_mount *mp = XFS_M(sb); - struct xfs_mount_args *args = xfs_args_allocate(sb, 0); - int error; + substring_t args[MAX_OPT_ARGS]; + char *p; - error = xfs_parseargs(mp, options, args, 1); - if (!error) - error = xfs_mntupdate(mp, flags, args); - kmem_free(args, sizeof(*args)); - return -error; + while ((p = strsep(&options, ",")) != NULL) { + int token; + + if (!*p) + continue; + + token = match_token(p, tokens, args); + 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; + break; + default: + /* + * Logically we would return an error here to prevent + * users from believing they might have changed + * mount options using remount which can't be changed. + * + * But unfortunately mount(8) adds all options from + * mtab and fstab to the mount arguments in some cases + * so we can't blindly reject options, but have to + * check for each specified option if it actually + * differs from the currently set option and only + * reject it if that's the case. + * + * Until that is implemented we return success for + * every remount request, and silently ignore all + * options that we can't actually change. + */ +#if 0 + printk(KERN_INFO + "XFS: mount option \"%s\" not supported for remount\n", p); + return -EINVAL; +#else + return 0; +#endif + } + } + + /* rw/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); + } + + /* rw -> ro */ + if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) { + xfs_filestream_flush(mp); + xfs_sync(mp, SYNC_DATA_QUIESCE); + xfs_attr_quiesce(mp); + mp->m_flags |= XFS_MOUNT_RDONLY; + } + + return 0; } /* @@ -1300,6 +1420,245 @@ xfs_fs_setxquota( Q_XSETPQLIM), id, (caddr_t)fdq); } +/* + * This function fills in xfs_mount_t fields based on mount args. + * Note: the superblock has _not_ yet been read in. + */ +STATIC int +xfs_start_flags( + struct xfs_mount_args *ap, + struct xfs_mount *mp) +{ + int error; + + /* Values are in BBs */ + if ((ap->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) { + /* + * At this point the superblock has not been read + * in, therefore we do not know the block size. + * Before the mount call ends we will convert + * these to FSBs. + */ + mp->m_dalign = ap->sunit; + mp->m_swidth = ap->swidth; + } + + if (ap->logbufs != -1 && + ap->logbufs != 0 && + (ap->logbufs < XLOG_MIN_ICLOGS || + ap->logbufs > XLOG_MAX_ICLOGS)) { + cmn_err(CE_WARN, + "XFS: invalid logbufs value: %d [not %d-%d]", + ap->logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS); + return XFS_ERROR(EINVAL); + } + mp->m_logbufs = ap->logbufs; + if (ap->logbufsize != -1 && + ap->logbufsize != 0 && + (ap->logbufsize < XLOG_MIN_RECORD_BSIZE || + ap->logbufsize > XLOG_MAX_RECORD_BSIZE || + !is_power_of_2(ap->logbufsize))) { + cmn_err(CE_WARN, + "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]", + ap->logbufsize); + return XFS_ERROR(EINVAL); + } + + error = ENOMEM; + + mp->m_logbsize = ap->logbufsize; + mp->m_fsname_len = strlen(ap->fsname) + 1; + + mp->m_fsname = kstrdup(ap->fsname, GFP_KERNEL); + if (!mp->m_fsname) + goto out; + + if (ap->rtname[0]) { + mp->m_rtname = kstrdup(ap->rtname, GFP_KERNEL); + if (!mp->m_rtname) + goto out_free_fsname; + + } + + if (ap->logname[0]) { + mp->m_logname = kstrdup(ap->logname, GFP_KERNEL); + if (!mp->m_logname) + goto out_free_rtname; + } + + if (ap->flags & XFSMNT_WSYNC) + mp->m_flags |= XFS_MOUNT_WSYNC; +#if XFS_BIG_INUMS + if (ap->flags & XFSMNT_INO64) { + mp->m_flags |= XFS_MOUNT_INO64; + mp->m_inoadd = XFS_INO64_OFFSET; + } +#endif + if (ap->flags & XFSMNT_RETERR) + mp->m_flags |= XFS_MOUNT_RETERR; + if (ap->flags & XFSMNT_NOALIGN) + mp->m_flags |= XFS_MOUNT_NOALIGN; + if (ap->flags & XFSMNT_SWALLOC) + mp->m_flags |= XFS_MOUNT_SWALLOC; + if (ap->flags & XFSMNT_OSYNCISOSYNC) + mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC; + if (ap->flags & XFSMNT_32BITINODES) + mp->m_flags |= XFS_MOUNT_32BITINODES; + + if (ap->flags & XFSMNT_IOSIZE) { + if (ap->iosizelog > XFS_MAX_IO_LOG || + ap->iosizelog < XFS_MIN_IO_LOG) { + cmn_err(CE_WARN, + "XFS: invalid log iosize: %d [not %d-%d]", + ap->iosizelog, XFS_MIN_IO_LOG, + XFS_MAX_IO_LOG); + return XFS_ERROR(EINVAL); + } + + mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE; + mp->m_readio_log = mp->m_writeio_log = ap->iosizelog; + } + + if (ap->flags & XFSMNT_IKEEP) + mp->m_flags |= XFS_MOUNT_IKEEP; + if (ap->flags & XFSMNT_DIRSYNC) + mp->m_flags |= XFS_MOUNT_DIRSYNC; + if (ap->flags & XFSMNT_ATTR2) + mp->m_flags |= XFS_MOUNT_ATTR2; + if (ap->flags & XFSMNT_NOATTR2) + mp->m_flags |= XFS_MOUNT_NOATTR2; + + if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE) + mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; + + /* + * no recovery flag requires a read-only mount + */ + if (ap->flags & XFSMNT_NORECOVERY) { + if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { + cmn_err(CE_WARN, + "XFS: tried to mount a FS read-write without recovery!"); + return XFS_ERROR(EINVAL); + } + mp->m_flags |= XFS_MOUNT_NORECOVERY; + } + + if (ap->flags & XFSMNT_NOUUID) + mp->m_flags |= XFS_MOUNT_NOUUID; + if (ap->flags & XFSMNT_BARRIER) + mp->m_flags |= XFS_MOUNT_BARRIER; + else + mp->m_flags &= ~XFS_MOUNT_BARRIER; + + if (ap->flags2 & XFSMNT2_FILESTREAMS) + mp->m_flags |= XFS_MOUNT_FILESTREAMS; + + if (ap->flags & XFSMNT_DMAPI) + mp->m_flags |= XFS_MOUNT_DMAPI; + return 0; + + + out_free_rtname: + kfree(mp->m_rtname); + out_free_fsname: + kfree(mp->m_fsname); + out: + return error; +} + +/* + * This function fills in xfs_mount_t fields based on mount args. + * Note: the superblock _has_ now been read in. + */ +STATIC int +xfs_finish_flags( + struct xfs_mount_args *ap, + struct xfs_mount *mp) +{ + int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); + + /* Fail a mount where the logbuf is smaller then the log stripe */ + if (xfs_sb_version_haslogv2(&mp->m_sb)) { + if ((ap->logbufsize <= 0) && + (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) { + mp->m_logbsize = mp->m_sb.sb_logsunit; + } else if (ap->logbufsize > 0 && + ap->logbufsize < mp->m_sb.sb_logsunit) { + cmn_err(CE_WARN, + "XFS: logbuf size must be greater than or equal to log stripe size"); + return XFS_ERROR(EINVAL); + } + } else { + /* Fail a mount if the logbuf is larger than 32K */ + if (ap->logbufsize > XLOG_BIG_RECORD_BSIZE) { + cmn_err(CE_WARN, + "XFS: logbuf size for version 1 logs must be 16K or 32K"); + return XFS_ERROR(EINVAL); + } + } + + /* + * mkfs'ed attr2 will turn on attr2 mount unless explicitly + * told by noattr2 to turn it off + */ + if (xfs_sb_version_hasattr2(&mp->m_sb) && + !(ap->flags & XFSMNT_NOATTR2)) + mp->m_flags |= XFS_MOUNT_ATTR2; + + /* + * prohibit r/w mounts of read-only filesystems + */ + if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) { + cmn_err(CE_WARN, + "XFS: cannot mount a read-only filesystem as read-write"); + return XFS_ERROR(EROFS); + } + + /* + * check for shared mount. + */ + if (ap->flags & XFSMNT_SHARED) { + if (!xfs_sb_version_hasshared(&mp->m_sb)) + return XFS_ERROR(EINVAL); + + /* + * For IRIX 6.5, shared mounts must have the shared + * version bit set, have the persistent readonly + * field set, must be version 0 and can only be mounted + * read-only. + */ + if (!ronly || !(mp->m_sb.sb_flags & XFS_SBF_READONLY) || + (mp->m_sb.sb_shared_vn != 0)) + return XFS_ERROR(EINVAL); + + mp->m_flags |= XFS_MOUNT_SHARED; + + /* + * Shared XFS V0 can't deal with DMI. Return EINVAL. + */ + if (mp->m_sb.sb_shared_vn == 0 && (ap->flags & XFSMNT_DMAPI)) + return XFS_ERROR(EINVAL); + } + + if (ap->flags & XFSMNT_UQUOTA) { + mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE); + if (ap->flags & XFSMNT_UQUOTAENF) + mp->m_qflags |= XFS_UQUOTA_ENFD; + } + + if (ap->flags & XFSMNT_GQUOTA) { + mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE); + if (ap->flags & XFSMNT_GQUOTAENF) + mp->m_qflags |= XFS_OQUOTA_ENFD; + } else if (ap->flags & XFSMNT_PQUOTA) { + mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE); + if (ap->flags & XFSMNT_PQUOTAENF) + mp->m_qflags |= XFS_OQUOTA_ENFD; + } + + return 0; +} + STATIC int xfs_fs_fill_super( struct super_block *sb, @@ -1308,11 +1667,21 @@ xfs_fs_fill_super( { struct inode *root; struct xfs_mount *mp = NULL; - struct xfs_mount_args *args = xfs_args_allocate(sb, silent); - int error; + struct xfs_mount_args *args; + int flags = 0, error = ENOMEM; - mp = xfs_mount_init(); + args = xfs_args_allocate(sb, silent); + if (!args) + return -ENOMEM; + mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL); + if (!mp) + goto out_free_args; + + spin_lock_init(&mp->m_sb_lock); + mutex_init(&mp->m_ilock); + mutex_init(&mp->m_growlock); + atomic_set(&mp->m_active_trans, 0); INIT_LIST_HEAD(&mp->m_sync_list); spin_lock_init(&mp->m_sync_lock); init_waitqueue_head(&mp->m_wait_single_sync_task); @@ -1325,16 +1694,60 @@ xfs_fs_fill_super( error = xfs_parseargs(mp, (char *)data, args, 0); if (error) - goto fail_vfsop; + goto out_free_mp; sb_min_blocksize(sb, BBSIZE); + sb->s_xattr = xfs_xattr_handlers; sb->s_export_op = &xfs_export_operations; sb->s_qcop = &xfs_quotactl_operations; sb->s_op = &xfs_super_operations; - error = xfs_mount(mp, args, NULL); + error = xfs_dmops_get(mp, args); + if (error) + goto out_free_mp; + error = xfs_qmops_get(mp, args); + if (error) + goto out_put_dmops; + + if (args->flags & XFSMNT_QUIET) + flags |= XFS_MFSI_QUIET; + + error = xfs_open_devices(mp, args); + if (error) + goto out_put_qmops; + + if (xfs_icsb_init_counters(mp)) + mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB; + + /* + * Setup flags based on mount(2) options and then the superblock + */ + error = xfs_start_flags(args, mp); + if (error) + goto out_free_fsname; + error = xfs_readsb(mp, flags); + if (error) + goto out_free_fsname; + error = xfs_finish_flags(args, mp); if (error) - goto fail_vfsop; + goto out_free_sb; + + error = xfs_setup_devices(mp); + 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; + + error = xfs_mountfs(mp); + if (error) + goto out_filestream_unmount; + + XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, args->mtpt, args->fsname); sb->s_dirt = 1; sb->s_magic = XFS_SB_MAGIC; @@ -1344,7 +1757,7 @@ xfs_fs_fill_super( sb->s_time_gran = 1; set_posix_acl_flag(sb); - root = igrab(mp->m_rootip->i_vnode); + root = igrab(VFS_I(mp->m_rootip)); if (!root) { error = ENOENT; goto fail_unmount; @@ -1369,10 +1782,28 @@ xfs_fs_fill_super( xfs_itrace_exit(XFS_I(sb->s_root->d_inode)); - kmem_free(args, sizeof(*args)); + kfree(args); return 0; -fail_vnrele: + out_filestream_unmount: + xfs_filestream_unmount(mp); + out_free_sb: + xfs_freesb(mp); + out_free_fsname: + xfs_free_fsname(mp); + xfs_icsb_destroy_counters(mp); + xfs_close_devices(mp); + out_put_qmops: + xfs_qmops_put(mp); + out_put_dmops: + xfs_dmops_put(mp); + out_free_mp: + kfree(mp); + out_free_args: + kfree(args); + return -error; + + fail_vnrele: if (sb->s_root) { dput(sb->s_root); sb->s_root = NULL; @@ -1380,12 +1811,20 @@ fail_vnrele: iput(root); } -fail_unmount: - xfs_unmount(mp, 0, NULL); + fail_unmount: + /* + * Blow away any referenced inode in the filestreams cache. + * This can and will cause log traffic as inodes go inactive + * here. + */ + xfs_filestream_unmount(mp); -fail_vfsop: - kmem_free(args, sizeof(*args)); - return -error; + XFS_bflush(mp->m_ddev_targp); + error = xfs_unmount_flush(mp, 0); + WARN_ON(error); + + xfs_unmountfs(mp); + goto out_free_sb; } STATIC int @@ -1430,9 +1869,235 @@ static struct file_system_type xfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; +STATIC int __init +xfs_alloc_trace_bufs(void) +{ +#ifdef XFS_ALLOC_TRACE + xfs_alloc_trace_buf = ktrace_alloc(XFS_ALLOC_TRACE_SIZE, KM_MAYFAIL); + if (!xfs_alloc_trace_buf) + goto out; +#endif +#ifdef XFS_BMAP_TRACE + xfs_bmap_trace_buf = ktrace_alloc(XFS_BMAP_TRACE_SIZE, KM_MAYFAIL); + if (!xfs_bmap_trace_buf) + goto out_free_alloc_trace; +#endif +#ifdef XFS_BMBT_TRACE + xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_MAYFAIL); + if (!xfs_bmbt_trace_buf) + goto out_free_bmap_trace; +#endif +#ifdef XFS_ATTR_TRACE + xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_MAYFAIL); + if (!xfs_attr_trace_buf) + goto out_free_bmbt_trace; +#endif +#ifdef XFS_DIR2_TRACE + xfs_dir2_trace_buf = ktrace_alloc(XFS_DIR2_GTRACE_SIZE, KM_MAYFAIL); + if (!xfs_dir2_trace_buf) + goto out_free_attr_trace; +#endif + + return 0; + +#ifdef XFS_DIR2_TRACE + out_free_attr_trace: +#endif +#ifdef XFS_ATTR_TRACE + ktrace_free(xfs_attr_trace_buf); + out_free_bmbt_trace: +#endif +#ifdef XFS_BMBT_TRACE + ktrace_free(xfs_bmbt_trace_buf); + out_free_bmap_trace: +#endif +#ifdef XFS_BMAP_TRACE + ktrace_free(xfs_bmap_trace_buf); + out_free_alloc_trace: +#endif +#ifdef XFS_ALLOC_TRACE + ktrace_free(xfs_alloc_trace_buf); + out: +#endif + return -ENOMEM; +} + +STATIC void +xfs_free_trace_bufs(void) +{ +#ifdef XFS_DIR2_TRACE + ktrace_free(xfs_dir2_trace_buf); +#endif +#ifdef XFS_ATTR_TRACE + ktrace_free(xfs_attr_trace_buf); +#endif +#ifdef XFS_BMBT_TRACE + ktrace_free(xfs_bmbt_trace_buf); +#endif +#ifdef XFS_BMAP_TRACE + ktrace_free(xfs_bmap_trace_buf); +#endif +#ifdef XFS_ALLOC_TRACE + ktrace_free(xfs_alloc_trace_buf); +#endif +} + +STATIC int __init +xfs_init_zones(void) +{ + xfs_vnode_zone = kmem_zone_init_flags(sizeof(struct inode), "xfs_vnode", + KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | + KM_ZONE_SPREAD, + xfs_fs_inode_init_once); + if (!xfs_vnode_zone) + goto out; + + xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend"); + if (!xfs_ioend_zone) + goto out_destroy_vnode_zone; + + xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE, + xfs_ioend_zone); + if (!xfs_ioend_pool) + goto out_destroy_ioend_zone; + + xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t), + "xfs_log_ticket"); + if (!xfs_log_ticket_zone) + goto out_destroy_ioend_pool; + + xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t), + "xfs_bmap_free_item"); + if (!xfs_bmap_free_item_zone) + goto out_destroy_log_ticket_zone; + xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t), + "xfs_btree_cur"); + if (!xfs_btree_cur_zone) + goto out_destroy_bmap_free_item_zone; + + xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t), + "xfs_da_state"); + if (!xfs_da_state_zone) + goto out_destroy_btree_cur_zone; + + xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf"); + if (!xfs_dabuf_zone) + goto out_destroy_da_state_zone; + + xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork"); + if (!xfs_ifork_zone) + goto out_destroy_dabuf_zone; + + xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans"); + if (!xfs_trans_zone) + goto out_destroy_ifork_zone; + + /* + * The size of the zone allocated buf log item is the maximum + * size possible under XFS. This wastes a little bit of memory, + * but it is much faster. + */ + xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) + + (((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) / + NBWORD) * sizeof(int))), "xfs_buf_item"); + if (!xfs_buf_item_zone) + goto out_destroy_trans_zone; + + xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) + + ((XFS_EFD_MAX_FAST_EXTENTS - 1) * + sizeof(xfs_extent_t))), "xfs_efd_item"); + if (!xfs_efd_zone) + goto out_destroy_buf_item_zone; + + xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) + + ((XFS_EFI_MAX_FAST_EXTENTS - 1) * + sizeof(xfs_extent_t))), "xfs_efi_item"); + if (!xfs_efi_zone) + goto out_destroy_efd_zone; + + xfs_inode_zone = + kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode", + KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | + KM_ZONE_SPREAD, NULL); + if (!xfs_inode_zone) + goto out_destroy_efi_zone; + + xfs_ili_zone = + kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili", + KM_ZONE_SPREAD, NULL); + if (!xfs_ili_zone) + goto out_destroy_inode_zone; + +#ifdef CONFIG_XFS_POSIX_ACL + xfs_acl_zone = kmem_zone_init(sizeof(xfs_acl_t), "xfs_acl"); + if (!xfs_acl_zone) + goto out_destroy_ili_zone; +#endif + + return 0; + +#ifdef CONFIG_XFS_POSIX_ACL + out_destroy_ili_zone: +#endif + kmem_zone_destroy(xfs_ili_zone); + out_destroy_inode_zone: + kmem_zone_destroy(xfs_inode_zone); + out_destroy_efi_zone: + kmem_zone_destroy(xfs_efi_zone); + out_destroy_efd_zone: + kmem_zone_destroy(xfs_efd_zone); + out_destroy_buf_item_zone: + kmem_zone_destroy(xfs_buf_item_zone); + out_destroy_trans_zone: + kmem_zone_destroy(xfs_trans_zone); + out_destroy_ifork_zone: + kmem_zone_destroy(xfs_ifork_zone); + out_destroy_dabuf_zone: + kmem_zone_destroy(xfs_dabuf_zone); + out_destroy_da_state_zone: + kmem_zone_destroy(xfs_da_state_zone); + out_destroy_btree_cur_zone: + kmem_zone_destroy(xfs_btree_cur_zone); + out_destroy_bmap_free_item_zone: + kmem_zone_destroy(xfs_bmap_free_item_zone); + out_destroy_log_ticket_zone: + kmem_zone_destroy(xfs_log_ticket_zone); + out_destroy_ioend_pool: + mempool_destroy(xfs_ioend_pool); + out_destroy_ioend_zone: + kmem_zone_destroy(xfs_ioend_zone); + out_destroy_vnode_zone: + kmem_zone_destroy(xfs_vnode_zone); + out: + return -ENOMEM; +} + +STATIC void +xfs_destroy_zones(void) +{ +#ifdef CONFIG_XFS_POSIX_ACL + kmem_zone_destroy(xfs_acl_zone); +#endif + kmem_zone_destroy(xfs_ili_zone); + kmem_zone_destroy(xfs_inode_zone); + kmem_zone_destroy(xfs_efi_zone); + kmem_zone_destroy(xfs_efd_zone); + kmem_zone_destroy(xfs_buf_item_zone); + kmem_zone_destroy(xfs_trans_zone); + kmem_zone_destroy(xfs_ifork_zone); + kmem_zone_destroy(xfs_dabuf_zone); + kmem_zone_destroy(xfs_da_state_zone); + kmem_zone_destroy(xfs_btree_cur_zone); + kmem_zone_destroy(xfs_bmap_free_item_zone); + kmem_zone_destroy(xfs_log_ticket_zone); + mempool_destroy(xfs_ioend_pool); + kmem_zone_destroy(xfs_ioend_zone); + kmem_zone_destroy(xfs_vnode_zone); + +} STATIC int __init -init_xfs_fs( void ) +init_xfs_fs(void) { int error; static char message[] __initdata = KERN_INFO \ @@ -1441,42 +2106,73 @@ init_xfs_fs( void ) printk(message); ktrace_init(64); + vn_init(); + xfs_dir_startup(); error = xfs_init_zones(); - if (error < 0) - goto undo_zones; + if (error) + goto out; + + error = xfs_alloc_trace_bufs(); + if (error) + goto out_destroy_zones; + + error = xfs_mru_cache_init(); + if (error) + goto out_free_trace_buffers; + + error = xfs_filestream_init(); + if (error) + goto out_mru_cache_uninit; error = xfs_buf_init(); - if (error < 0) - goto undo_buffers; + if (error) + goto out_filestream_uninit; + + error = xfs_init_procfs(); + if (error) + goto out_buf_terminate; + + error = xfs_sysctl_register(); + if (error) + goto out_cleanup_procfs; - vn_init(); - xfs_init(); - uuid_init(); vfs_initquota(); error = register_filesystem(&xfs_fs_type); if (error) - goto undo_register; + goto out_sysctl_unregister; return 0; -undo_register: + out_sysctl_unregister: + xfs_sysctl_unregister(); + out_cleanup_procfs: + xfs_cleanup_procfs(); + out_buf_terminate: xfs_buf_terminate(); - -undo_buffers: + out_filestream_uninit: + xfs_filestream_uninit(); + out_mru_cache_uninit: + xfs_mru_cache_uninit(); + out_free_trace_buffers: + xfs_free_trace_bufs(); + out_destroy_zones: xfs_destroy_zones(); - -undo_zones: + out: return error; } STATIC void __exit -exit_xfs_fs( void ) +exit_xfs_fs(void) { vfs_exitquota(); unregister_filesystem(&xfs_fs_type); - xfs_cleanup(); + xfs_sysctl_unregister(); + xfs_cleanup_procfs(); xfs_buf_terminate(); + xfs_filestream_uninit(); + xfs_mru_cache_uninit(); + xfs_free_trace_bufs(); xfs_destroy_zones(); ktrace_uninit(); } diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h index 3efb7c6..fe2ef4e 100644 --- a/fs/xfs/linux-2.6/xfs_super.h +++ b/fs/xfs/linux-2.6/xfs_super.h @@ -101,18 +101,13 @@ struct block_device; extern __uint64_t xfs_max_file_offset(unsigned int); -extern void xfs_initialize_vnode(struct xfs_mount *mp, bhv_vnode_t *vp, - struct xfs_inode *ip); - extern void xfs_flush_inode(struct xfs_inode *); extern void xfs_flush_device(struct xfs_inode *); -extern int xfs_blkdev_get(struct xfs_mount *, const char *, - struct block_device **); -extern void xfs_blkdev_put(struct block_device *); extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); extern const struct export_operations xfs_export_operations; +extern struct xattr_handler *xfs_xattr_handlers[]; #define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info)) diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c index bb997d7..7dacb5b 100644 --- a/fs/xfs/linux-2.6/xfs_sysctl.c +++ b/fs/xfs/linux-2.6/xfs_sysctl.c @@ -259,15 +259,17 @@ static ctl_table xfs_root_table[] = { {} }; -void +int xfs_sysctl_register(void) { xfs_table_header = register_sysctl_table(xfs_root_table); + if (!xfs_table_header) + return -ENOMEM; + return 0; } void xfs_sysctl_unregister(void) { - if (xfs_table_header) - unregister_sysctl_table(xfs_table_header); + unregister_sysctl_table(xfs_table_header); } diff --git a/fs/xfs/linux-2.6/xfs_sysctl.h b/fs/xfs/linux-2.6/xfs_sysctl.h index 98b97e3..4aadb80 100644 --- a/fs/xfs/linux-2.6/xfs_sysctl.h +++ b/fs/xfs/linux-2.6/xfs_sysctl.h @@ -93,10 +93,10 @@ enum { extern xfs_param_t xfs_params; #ifdef CONFIG_SYSCTL -extern void xfs_sysctl_register(void); +extern int xfs_sysctl_register(void); extern void xfs_sysctl_unregister(void); #else -# define xfs_sysctl_register() do { } while (0) +# define xfs_sysctl_register() (0) # define xfs_sysctl_unregister() do { } while (0) #endif /* CONFIG_SYSCTL */ diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c index bc7afe0..b52528b 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.c +++ b/fs/xfs/linux-2.6/xfs_vnode.c @@ -33,7 +33,7 @@ /* - * Dedicated vnode inactive/reclaim sync semaphores. + * Dedicated vnode inactive/reclaim sync wait queues. * Prime number of hash buckets since address is used as the key. */ #define NVSYNC 37 @@ -82,74 +82,6 @@ vn_ioerror( xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, f, l); } -/* - * Revalidate the Linux inode from the XFS inode. - * Note: i_size _not_ updated; we must hold the inode - * semaphore when doing that - callers responsibility. - */ -int -vn_revalidate( - bhv_vnode_t *vp) -{ - struct inode *inode = vn_to_inode(vp); - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - unsigned long xflags; - - xfs_itrace_entry(ip); - - if (XFS_FORCED_SHUTDOWN(mp)) - return -EIO; - - xfs_ilock(ip, XFS_ILOCK_SHARED); - inode->i_mode = ip->i_d.di_mode; - inode->i_uid = ip->i_d.di_uid; - inode->i_gid = ip->i_d.di_gid; - inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; - inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; - inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; - inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; - - xflags = xfs_ip2xflags(ip); - if (xflags & XFS_XFLAG_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; - if (xflags & XFS_XFLAG_SYNC) - inode->i_flags |= S_SYNC; - else - inode->i_flags &= ~S_SYNC; - if (xflags & XFS_XFLAG_NOATIME) - inode->i_flags |= S_NOATIME; - else - inode->i_flags &= ~S_NOATIME; - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - xfs_iflags_clear(ip, XFS_IMODIFIED); - return 0; -} - -/* - * Add a reference to a referenced vnode. - */ -bhv_vnode_t * -vn_hold( - bhv_vnode_t *vp) -{ - struct inode *inode; - - XFS_STATS_INC(vn_hold); - - inode = igrab(vn_to_inode(vp)); - ASSERT(inode); - - return vp; -} - #ifdef XFS_INODE_TRACE /* @@ -158,7 +90,7 @@ vn_hold( */ static inline int xfs_icount(struct xfs_inode *ip) { - bhv_vnode_t *vp = XFS_ITOV_NULL(ip); + struct inode *vp = VFS_I(ip); if (vp) return vn_count(vp); diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h index 25eb2a9..683ce16 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ b/fs/xfs/linux-2.6/xfs_vnode.h @@ -19,24 +19,9 @@ #define __XFS_VNODE_H__ struct file; -struct bhv_vattr; struct xfs_iomap; struct attrlist_cursor_kern; -typedef struct inode bhv_vnode_t; - -/* - * Vnode to Linux inode mapping. - */ -static inline bhv_vnode_t *vn_from_inode(struct inode *inode) -{ - return inode; -} -static inline struct inode *vn_to_inode(bhv_vnode_t *vnode) -{ - return vnode; -} - /* * Return values for xfs_inactive. A return value of * VN_INACTIVE_NOCACHE implies that the file system behavior @@ -66,87 +51,8 @@ static inline struct inode *vn_to_inode(bhv_vnode_t *vnode) Prevent VM access to the pages until the operation completes. */ -/* - * Vnode attributes. va_mask indicates those attributes the caller - * wants to set or extract. - */ -typedef struct bhv_vattr { - int va_mask; /* bit-mask of attributes present */ - mode_t va_mode; /* file access mode and type */ - xfs_nlink_t va_nlink; /* number of references to file */ - uid_t va_uid; /* owner user id */ - gid_t va_gid; /* owner group id */ - xfs_ino_t va_nodeid; /* file id */ - xfs_off_t va_size; /* file size in bytes */ - u_long va_blocksize; /* blocksize preferred for i/o */ - struct timespec va_atime; /* time of last access */ - struct timespec va_mtime; /* time of last modification */ - struct timespec va_ctime; /* time file changed */ - u_int va_gen; /* generation number of file */ - xfs_dev_t va_rdev; /* device the special file represents */ - __int64_t va_nblocks; /* number of blocks allocated */ - u_long va_xflags; /* random extended file flags */ - u_long va_extsize; /* file extent size */ - u_long va_nextents; /* number of extents in file */ - u_long va_anextents; /* number of attr extents in file */ - prid_t va_projid; /* project id */ -} bhv_vattr_t; - -/* - * setattr or getattr attributes - */ -#define XFS_AT_TYPE 0x00000001 -#define XFS_AT_MODE 0x00000002 -#define XFS_AT_UID 0x00000004 -#define XFS_AT_GID 0x00000008 -#define XFS_AT_FSID 0x00000010 -#define XFS_AT_NODEID 0x00000020 -#define XFS_AT_NLINK 0x00000040 -#define XFS_AT_SIZE 0x00000080 -#define XFS_AT_ATIME 0x00000100 -#define XFS_AT_MTIME 0x00000200 -#define XFS_AT_CTIME 0x00000400 -#define XFS_AT_RDEV 0x00000800 -#define XFS_AT_BLKSIZE 0x00001000 -#define XFS_AT_NBLOCKS 0x00002000 -#define XFS_AT_VCODE 0x00004000 -#define XFS_AT_MAC 0x00008000 -#define XFS_AT_UPDATIME 0x00010000 -#define XFS_AT_UPDMTIME 0x00020000 -#define XFS_AT_UPDCTIME 0x00040000 -#define XFS_AT_ACL 0x00080000 -#define XFS_AT_CAP 0x00100000 -#define XFS_AT_INF 0x00200000 -#define XFS_AT_XFLAGS 0x00400000 -#define XFS_AT_EXTSIZE 0x00800000 -#define XFS_AT_NEXTENTS 0x01000000 -#define XFS_AT_ANEXTENTS 0x02000000 -#define XFS_AT_PROJID 0x04000000 -#define XFS_AT_SIZE_NOPERM 0x08000000 -#define XFS_AT_GENCOUNT 0x10000000 - -#define XFS_AT_ALL (XFS_AT_TYPE|XFS_AT_MODE|XFS_AT_UID|XFS_AT_GID|\ - XFS_AT_FSID|XFS_AT_NODEID|XFS_AT_NLINK|XFS_AT_SIZE|\ - XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME|XFS_AT_RDEV|\ - XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_VCODE|XFS_AT_MAC|\ - XFS_AT_ACL|XFS_AT_CAP|XFS_AT_INF|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|\ - XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|XFS_AT_PROJID|XFS_AT_GENCOUNT) - -#define XFS_AT_STAT (XFS_AT_TYPE|XFS_AT_MODE|XFS_AT_UID|XFS_AT_GID|\ - XFS_AT_FSID|XFS_AT_NODEID|XFS_AT_NLINK|XFS_AT_SIZE|\ - XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME|XFS_AT_RDEV|\ - XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_PROJID) - -#define XFS_AT_TIMES (XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME) - -#define XFS_AT_UPDTIMES (XFS_AT_UPDATIME|XFS_AT_UPDMTIME|XFS_AT_UPDCTIME) - -#define XFS_AT_NOSET (XFS_AT_NLINK|XFS_AT_RDEV|XFS_AT_FSID|XFS_AT_NODEID|\ - XFS_AT_TYPE|XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_VCODE|\ - XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|XFS_AT_GENCOUNT) extern void vn_init(void); -extern int vn_revalidate(bhv_vnode_t *); /* * Yeah, these don't take vnode anymore at all, all this should be @@ -156,57 +62,52 @@ extern void vn_iowait(struct xfs_inode *ip); extern void vn_iowake(struct xfs_inode *ip); extern void vn_ioerror(struct xfs_inode *ip, int error, char *f, int l); -static inline int vn_count(bhv_vnode_t *vp) +static inline int vn_count(struct inode *vp) { - return atomic_read(&vn_to_inode(vp)->i_count); + return atomic_read(&vp->i_count); } -/* - * Vnode reference counting functions (and macros for compatibility). - */ -extern bhv_vnode_t *vn_hold(bhv_vnode_t *); +#define IHOLD(ip) \ +do { \ + ASSERT(atomic_read(&VFS_I(ip)->i_count) > 0) ; \ + atomic_inc(&(VFS_I(ip)->i_count)); \ + xfs_itrace_hold((ip), __FILE__, __LINE__, (inst_t *)__return_address); \ +} while (0) -#if defined(XFS_INODE_TRACE) -#define VN_HOLD(vp) \ - ((void)vn_hold(vp), \ - xfs_itrace_hold(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address)) -#define VN_RELE(vp) \ - (xfs_itrace_rele(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address), \ - iput(vn_to_inode(vp))) -#else -#define VN_HOLD(vp) ((void)vn_hold(vp)) -#define VN_RELE(vp) (iput(vn_to_inode(vp))) -#endif +#define IRELE(ip) \ +do { \ + xfs_itrace_rele((ip), __FILE__, __LINE__, (inst_t *)__return_address); \ + iput(VFS_I(ip)); \ +} while (0) -static inline bhv_vnode_t *vn_grab(bhv_vnode_t *vp) +static inline struct inode *vn_grab(struct inode *vp) { - struct inode *inode = igrab(vn_to_inode(vp)); - return inode ? vn_from_inode(inode) : NULL; + return igrab(vp); } /* * Dealing with bad inodes */ -static inline int VN_BAD(bhv_vnode_t *vp) +static inline int VN_BAD(struct inode *vp) { - return is_bad_inode(vn_to_inode(vp)); + return is_bad_inode(vp); } /* * Extracting atime values in various formats */ -static inline void vn_atime_to_bstime(bhv_vnode_t *vp, xfs_bstime_t *bs_atime) +static inline void vn_atime_to_bstime(struct inode *vp, xfs_bstime_t *bs_atime) { bs_atime->tv_sec = vp->i_atime.tv_sec; bs_atime->tv_nsec = vp->i_atime.tv_nsec; } -static inline void vn_atime_to_timespec(bhv_vnode_t *vp, struct timespec *ts) +static inline void vn_atime_to_timespec(struct inode *vp, struct timespec *ts) { *ts = vp->i_atime; } -static inline void vn_atime_to_time_t(bhv_vnode_t *vp, time_t *tt) +static inline void vn_atime_to_time_t(struct inode *vp, time_t *tt) { *tt = vp->i_atime.tv_sec; } @@ -214,20 +115,11 @@ static inline void vn_atime_to_time_t(bhv_vnode_t *vp, time_t *tt) /* * Some useful predicates. */ -#define VN_MAPPED(vp) mapping_mapped(vn_to_inode(vp)->i_mapping) -#define VN_CACHED(vp) (vn_to_inode(vp)->i_mapping->nrpages) -#define VN_DIRTY(vp) mapping_tagged(vn_to_inode(vp)->i_mapping, \ +#define VN_MAPPED(vp) mapping_mapped(vp->i_mapping) +#define VN_CACHED(vp) (vp->i_mapping->nrpages) +#define VN_DIRTY(vp) mapping_tagged(vp->i_mapping, \ PAGECACHE_TAG_DIRTY) -/* - * Flags to vop_setattr/getattr. - */ -#define ATTR_UTIME 0x01 /* non-default utime(2) request */ -#define ATTR_DMI 0x08 /* invocation from a DMI function */ -#define ATTR_LAZY 0x80 /* set/get attributes lazily */ -#define ATTR_NONBLOCK 0x100 /* return EAGAIN if operation would block */ -#define ATTR_NOLOCK 0x200 /* Don't grab any conflicting locks */ -#define ATTR_NOSIZETOK 0x400 /* Don't get the SIZE token */ /* * Tracking vnode activity. diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c new file mode 100644 index 0000000..964621f --- /dev/null +++ b/fs/xfs/linux-2.6/xfs_xattr.c @@ -0,0 +1,330 @@ +/* + * Copyright (C) 2008 Christoph Hellwig. + * Portions Copyright (C) 2000-2008 Silicon Graphics, Inc. + * + * 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. + * + * This program is distributed in the hope that it would 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 the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "xfs.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_acl.h" +#include "xfs_vnodeops.h" + +#include <linux/posix_acl_xattr.h> +#include <linux/xattr.h> + + +/* + * ACL handling. Should eventually be moved into xfs_acl.c + */ + +static int +xfs_decode_acl(const char *name) +{ + if (strcmp(name, "posix_acl_access") == 0) + return _ACL_TYPE_ACCESS; + else if (strcmp(name, "posix_acl_default") == 0) + return _ACL_TYPE_DEFAULT; + return -EINVAL; +} + +/* + * Get system extended attributes which at the moment only + * includes Posix ACLs. + */ +static int +xfs_xattr_system_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + int acl; + + acl = xfs_decode_acl(name); + if (acl < 0) + return acl; + + return xfs_acl_vget(inode, buffer, size, acl); +} + +static int +xfs_xattr_system_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + int acl; + + acl = xfs_decode_acl(name); + if (acl < 0) + return acl; + if (flags & XATTR_CREATE) + return -EINVAL; + + if (!value) + return xfs_acl_vremove(inode, acl); + + return xfs_acl_vset(inode, (void *)value, size, acl); +} + +static struct xattr_handler xfs_xattr_system_handler = { + .prefix = XATTR_SYSTEM_PREFIX, + .get = xfs_xattr_system_get, + .set = xfs_xattr_system_set, +}; + + +/* + * Real xattr handling. The only difference between the namespaces is + * a flag passed to the low-level attr code. + */ + +static int +__xfs_xattr_get(struct inode *inode, const char *name, + void *value, size_t size, int xflags) +{ + struct xfs_inode *ip = XFS_I(inode); + int error, asize = size; + + if (strcmp(name, "") == 0) + return -EINVAL; + + /* Convert Linux syscall to XFS internal ATTR flags */ + if (!size) { + xflags |= ATTR_KERNOVAL; + value = NULL; + } + + error = -xfs_attr_get(ip, name, value, &asize, xflags); + if (error) + return error; + return asize; +} + +static int +__xfs_xattr_set(struct inode *inode, const char *name, const void *value, + size_t size, int flags, int xflags) +{ + struct xfs_inode *ip = XFS_I(inode); + + if (strcmp(name, "") == 0) + return -EINVAL; + + /* Convert Linux syscall to XFS internal ATTR flags */ + if (flags & XATTR_CREATE) + xflags |= ATTR_CREATE; + if (flags & XATTR_REPLACE) + xflags |= ATTR_REPLACE; + + if (!value) + return -xfs_attr_remove(ip, name, xflags); + return -xfs_attr_set(ip, name, (void *)value, size, xflags); +} + +static int +xfs_xattr_user_get(struct inode *inode, const char *name, + void *value, size_t size) +{ + return __xfs_xattr_get(inode, name, value, size, 0); +} + +static int +xfs_xattr_user_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + return __xfs_xattr_set(inode, name, value, size, flags, 0); +} + +static struct xattr_handler xfs_xattr_user_handler = { + .prefix = XATTR_USER_PREFIX, + .get = xfs_xattr_user_get, + .set = xfs_xattr_user_set, +}; + + +static int +xfs_xattr_trusted_get(struct inode *inode, const char *name, + void *value, size_t size) +{ + return __xfs_xattr_get(inode, name, value, size, ATTR_ROOT); +} + +static int +xfs_xattr_trusted_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + return __xfs_xattr_set(inode, name, value, size, flags, ATTR_ROOT); +} + +static struct xattr_handler xfs_xattr_trusted_handler = { + .prefix = XATTR_TRUSTED_PREFIX, + .get = xfs_xattr_trusted_get, + .set = xfs_xattr_trusted_set, +}; + + +static int +xfs_xattr_secure_get(struct inode *inode, const char *name, + void *value, size_t size) +{ + return __xfs_xattr_get(inode, name, value, size, ATTR_SECURE); +} + +static int +xfs_xattr_secure_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + return __xfs_xattr_set(inode, name, value, size, flags, ATTR_SECURE); +} + +static struct xattr_handler xfs_xattr_security_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .get = xfs_xattr_secure_get, + .set = xfs_xattr_secure_set, +}; + + +struct xattr_handler *xfs_xattr_handlers[] = { + &xfs_xattr_user_handler, + &xfs_xattr_trusted_handler, + &xfs_xattr_security_handler, + &xfs_xattr_system_handler, + NULL +}; + +static unsigned int xfs_xattr_prefix_len(int flags) +{ + if (flags & XFS_ATTR_SECURE) + return sizeof("security"); + else if (flags & XFS_ATTR_ROOT) + return sizeof("trusted"); + else + return sizeof("user"); +} + +static const char *xfs_xattr_prefix(int flags) +{ + if (flags & XFS_ATTR_SECURE) + return xfs_xattr_security_handler.prefix; + else if (flags & XFS_ATTR_ROOT) + return xfs_xattr_trusted_handler.prefix; + else + return xfs_xattr_user_handler.prefix; +} + +static int +xfs_xattr_put_listent(struct xfs_attr_list_context *context, int flags, + char *name, int namelen, int valuelen, char *value) +{ + unsigned int prefix_len = xfs_xattr_prefix_len(flags); + char *offset; + int arraytop; + + ASSERT(context->count >= 0); + + /* + * Only show root namespace entries if we are actually allowed to + * see them. + */ + if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN)) + return 0; + + arraytop = context->count + prefix_len + namelen + 1; + if (arraytop > context->firstu) { + context->count = -1; /* insufficient space */ + return 1; + } + offset = (char *)context->alist + context->count; + strncpy(offset, xfs_xattr_prefix(flags), prefix_len); + offset += prefix_len; + strncpy(offset, name, namelen); /* real name */ + offset += namelen; + *offset = '\0'; + context->count += prefix_len + namelen + 1; + return 0; +} + +static int +xfs_xattr_put_listent_sizes(struct xfs_attr_list_context *context, int flags, + char *name, int namelen, int valuelen, char *value) +{ + context->count += xfs_xattr_prefix_len(flags) + namelen + 1; + return 0; +} + +static int +list_one_attr(const char *name, const size_t len, void *data, + size_t size, ssize_t *result) +{ + char *p = data + *result; + + *result += len; + if (!size) + return 0; + if (*result > size) + return -ERANGE; + + strcpy(p, name); + return 0; +} + +ssize_t +xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size) +{ + struct xfs_attr_list_context context; + struct attrlist_cursor_kern cursor = { 0 }; + struct inode *inode = dentry->d_inode; + int error; + + /* + * First read the regular on-disk attributes. + */ + memset(&context, 0, sizeof(context)); + context.dp = XFS_I(inode); + context.cursor = &cursor; + context.resynch = 1; + context.alist = data; + context.bufsize = size; + context.firstu = context.bufsize; + + if (size) + context.put_listent = xfs_xattr_put_listent; + else + context.put_listent = xfs_xattr_put_listent_sizes; + + xfs_attr_list_int(&context); + if (context.count < 0) + return -ERANGE; + + /* + * Then add the two synthetic ACL attributes. + */ + if (xfs_acl_vhasacl_access(inode)) { + error = list_one_attr(POSIX_ACL_XATTR_ACCESS, + strlen(POSIX_ACL_XATTR_ACCESS) + 1, + data, size, &context.count); + if (error) + return error; + } + + if (xfs_acl_vhasacl_default(inode)) { + error = list_one_attr(POSIX_ACL_XATTR_DEFAULT, + strlen(POSIX_ACL_XATTR_DEFAULT) + 1, + data, size, &context.count); + if (error) + return error; + } + + return context.count; +} diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index 85df328..f2705f2 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -101,11 +101,18 @@ xfs_qm_dqinit( if (brandnewdquot) { dqp->dq_flnext = dqp->dq_flprev = dqp; mutex_init(&dqp->q_qlock); - initnsema(&dqp->q_flock, 1, "fdq"); sv_init(&dqp->q_pinwait, SV_DEFAULT, "pdq"); + /* + * Because we want to use a counting completion, complete + * the flush completion once to allow a single access to + * the flush completion without blocking. + */ + init_completion(&dqp->q_flush); + complete(&dqp->q_flush); + #ifdef XFS_DQUOT_TRACE - dqp->q_trace = ktrace_alloc(DQUOT_TRACE_SIZE, KM_SLEEP); + dqp->q_trace = ktrace_alloc(DQUOT_TRACE_SIZE, KM_NOFS); xfs_dqtrace_entry(dqp, "DQINIT"); #endif } else { @@ -150,7 +157,6 @@ xfs_qm_dqdestroy( ASSERT(! XFS_DQ_IS_ON_FREELIST(dqp)); mutex_destroy(&dqp->q_qlock); - freesema(&dqp->q_flock); sv_destroy(&dqp->q_pinwait); #ifdef XFS_DQUOT_TRACE @@ -431,7 +437,7 @@ xfs_qm_dqalloc( * when it unlocks the inode. Since we want to keep the quota * inode around, we bump the vnode ref count now. */ - VN_HOLD(XFS_ITOV(quotip)); + IHOLD(quotip); xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL); nmaps = 1; @@ -1211,7 +1217,7 @@ xfs_qm_dqflush( int error; ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(XFS_DQ_IS_FLUSH_LOCKED(dqp)); + ASSERT(!completion_done(&dqp->q_flush)); xfs_dqtrace_entry(dqp, "DQFLUSH"); /* @@ -1348,34 +1354,18 @@ xfs_qm_dqflush_done( xfs_dqfunlock(dqp); } - -int -xfs_qm_dqflock_nowait( - xfs_dquot_t *dqp) -{ - int locked; - - locked = cpsema(&((dqp)->q_flock)); - - /* XXX ifdef these out */ - if (locked) - (dqp)->dq_flags |= XFS_DQ_FLOCKED; - return (locked); -} - - int xfs_qm_dqlock_nowait( xfs_dquot_t *dqp) { - return (mutex_trylock(&((dqp)->q_qlock))); + return mutex_trylock(&dqp->q_qlock); } void xfs_dqlock( xfs_dquot_t *dqp) { - mutex_lock(&(dqp->q_qlock)); + mutex_lock(&dqp->q_qlock); } void @@ -1435,8 +1425,7 @@ xfs_dqlock2( /* ARGSUSED */ int xfs_qm_dqpurge( - xfs_dquot_t *dqp, - uint flags) + xfs_dquot_t *dqp) { xfs_dqhash_t *thishash; xfs_mount_t *mp = dqp->q_mount; @@ -1469,7 +1458,7 @@ xfs_qm_dqpurge( * if we're turning off quotas. Basically, we need this flush * lock, and are willing to block on it. */ - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { /* * Block on the flush lock after nudging dquot buffer, * if it is incore. diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h index 5c371a9..8958d0f 100644 --- a/fs/xfs/quota/xfs_dquot.h +++ b/fs/xfs/quota/xfs_dquot.h @@ -82,7 +82,7 @@ typedef struct xfs_dquot { xfs_qcnt_t q_res_icount; /* total inos allocd+reserved */ xfs_qcnt_t q_res_rtbcount;/* total realtime blks used+reserved */ mutex_t q_qlock; /* quota lock */ - sema_t q_flock; /* flush lock */ + struct completion q_flush; /* flush completion queue */ uint q_pincount; /* pin count for this dquot */ sv_t q_pinwait; /* sync var for pinning */ #ifdef XFS_DQUOT_TRACE @@ -113,17 +113,25 @@ XFS_DQ_IS_LOCKED(xfs_dquot_t *dqp) /* - * The following three routines simply manage the q_flock - * semaphore embedded in the dquot. This semaphore synchronizes - * processes attempting to flush the in-core dquot back to disk. + * Manage the q_flush completion queue embedded in the dquot. This completion + * queue synchronizes processes attempting to flush the in-core dquot back to + * disk. */ -#define xfs_dqflock(dqp) { psema(&((dqp)->q_flock), PINOD | PRECALC);\ - (dqp)->dq_flags |= XFS_DQ_FLOCKED; } -#define xfs_dqfunlock(dqp) { ASSERT(issemalocked(&((dqp)->q_flock))); \ - vsema(&((dqp)->q_flock)); \ - (dqp)->dq_flags &= ~(XFS_DQ_FLOCKED); } +static inline void xfs_dqflock(xfs_dquot_t *dqp) +{ + wait_for_completion(&dqp->q_flush); +} + +static inline int xfs_dqflock_nowait(xfs_dquot_t *dqp) +{ + return try_wait_for_completion(&dqp->q_flush); +} + +static inline void xfs_dqfunlock(xfs_dquot_t *dqp) +{ + complete(&dqp->q_flush); +} -#define XFS_DQ_IS_FLUSH_LOCKED(dqp) (issemalocked(&((dqp)->q_flock))) #define XFS_DQ_IS_ON_FREELIST(dqp) ((dqp)->dq_flnext != (dqp)) #define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY) #define XFS_QM_ISUDQ(dqp) ((dqp)->dq_flags & XFS_DQ_USER) @@ -164,10 +172,9 @@ extern void xfs_qm_dqprint(xfs_dquot_t *); extern void xfs_qm_dqdestroy(xfs_dquot_t *); extern int xfs_qm_dqflush(xfs_dquot_t *, uint); -extern int xfs_qm_dqpurge(xfs_dquot_t *, uint); +extern int xfs_qm_dqpurge(xfs_dquot_t *); extern void xfs_qm_dqunpin_wait(xfs_dquot_t *); extern int xfs_qm_dqlock_nowait(xfs_dquot_t *); -extern int xfs_qm_dqflock_nowait(xfs_dquot_t *); extern void xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp); extern void xfs_qm_adjust_dqtimers(xfs_mount_t *, xfs_disk_dquot_t *); diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c index 36e05ca..f028644 100644 --- a/fs/xfs/quota/xfs_dquot_item.c +++ b/fs/xfs/quota/xfs_dquot_item.c @@ -151,7 +151,7 @@ xfs_qm_dquot_logitem_push( dqp = logitem->qli_dquot; ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(XFS_DQ_IS_FLUSH_LOCKED(dqp)); + ASSERT(!completion_done(&dqp->q_flush)); /* * Since we were able to lock the dquot's flush lock and @@ -245,7 +245,7 @@ xfs_qm_dquot_logitem_pushbuf( * inode flush completed and the inode was taken off the AIL. * So, just get out. */ - if (!issemalocked(&(dqp->q_flock)) || + if (completion_done(&dqp->q_flush) || ((qip->qli_item.li_flags & XFS_LI_IN_AIL) == 0)) { qip->qli_pushbuf_flag = 0; xfs_dqunlock(dqp); @@ -258,7 +258,7 @@ xfs_qm_dquot_logitem_pushbuf( if (bp != NULL) { if (XFS_BUF_ISDELAYWRITE(bp)) { dopush = ((qip->qli_item.li_flags & XFS_LI_IN_AIL) && - issemalocked(&(dqp->q_flock))); + !completion_done(&dqp->q_flush)); qip->qli_pushbuf_flag = 0; xfs_dqunlock(dqp); @@ -317,7 +317,7 @@ xfs_qm_dquot_logitem_trylock( return (XFS_ITEM_LOCKED); retval = XFS_ITEM_SUCCESS; - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { /* * The dquot is already being flushed. It may have been * flushed delayed write, however, and we don't want to @@ -576,8 +576,8 @@ xfs_qm_qoffend_logitem_committed( * xfs_trans_delete_ail() drops the AIL lock. */ xfs_trans_delete_ail(qfs->qql_item.li_mountp, (xfs_log_item_t *)qfs); - kmem_free(qfs, sizeof(xfs_qoff_logitem_t)); - kmem_free(qfe, sizeof(xfs_qoff_logitem_t)); + kmem_free(qfs); + kmem_free(qfe); return (xfs_lsn_t)-1; } diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index d31cce1..df0ffef 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -192,8 +192,8 @@ xfs_qm_destroy( xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i])); xfs_qm_list_destroy(&(xqm->qm_grp_dqhtable[i])); } - kmem_free(xqm->qm_usr_dqhtable, hsize * sizeof(xfs_dqhash_t)); - kmem_free(xqm->qm_grp_dqhtable, hsize * sizeof(xfs_dqhash_t)); + kmem_free(xqm->qm_usr_dqhtable); + kmem_free(xqm->qm_grp_dqhtable); xqm->qm_usr_dqhtable = NULL; xqm->qm_grp_dqhtable = NULL; xqm->qm_dqhashmask = 0; @@ -201,7 +201,7 @@ xfs_qm_destroy( #ifdef DEBUG mutex_destroy(&qcheck_lock); #endif - kmem_free(xqm, sizeof(xfs_qm_t)); + kmem_free(xqm); } /* @@ -310,8 +310,7 @@ xfs_qm_unmount_quotadestroy( */ void xfs_qm_mount_quotas( - xfs_mount_t *mp, - int mfsi_flags) + xfs_mount_t *mp) { int error = 0; uint sbf; @@ -346,8 +345,7 @@ xfs_qm_mount_quotas( /* * If any of the quotas are not consistent, do a quotacheck. */ - if (XFS_QM_NEED_QUOTACHECK(mp) && - !(mfsi_flags & XFS_MFSI_NO_QUOTACHECK)) { + if (XFS_QM_NEED_QUOTACHECK(mp)) { error = xfs_qm_quotacheck(mp); if (error) { /* Quotacheck failed and disabled quotas. */ @@ -445,11 +443,11 @@ xfs_qm_unmount_quotas( } } if (uqp) { - XFS_PURGE_INODE(uqp); + IRELE(uqp); mp->m_quotainfo->qi_uquotaip = NULL; } if (gqp) { - XFS_PURGE_INODE(gqp); + IRELE(gqp); mp->m_quotainfo->qi_gquotaip = NULL; } out: @@ -484,7 +482,7 @@ again: xfs_dqtrace_entry(dqp, "FLUSHALL: DQDIRTY"); /* XXX a sentinel would be better */ recl = XFS_QI_MPLRECLAIMS(mp); - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { /* * If we can't grab the flush lock then check * to see if the dquot has been flushed delayed @@ -631,7 +629,7 @@ xfs_qm_dqpurge_int( * freelist in INACTIVE state. */ nextdqp = dqp->MPL_NEXT; - nmisses += xfs_qm_dqpurge(dqp, flags); + nmisses += xfs_qm_dqpurge(dqp); dqp = nextdqp; } xfs_qm_mplist_unlock(mp); @@ -1062,7 +1060,7 @@ xfs_qm_sync( /* XXX a sentinel would be better */ recl = XFS_QI_MPLRECLAIMS(mp); - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { if (nowait) { xfs_dqunlock(dqp); continue; @@ -1134,7 +1132,7 @@ xfs_qm_init_quotainfo( * and change the superblock accordingly. */ if ((error = xfs_qm_init_quotainos(mp))) { - kmem_free(qinf, sizeof(xfs_quotainfo_t)); + kmem_free(qinf); mp->m_quotainfo = NULL; return error; } @@ -1240,15 +1238,15 @@ xfs_qm_destroy_quotainfo( xfs_qm_list_destroy(&qi->qi_dqlist); if (qi->qi_uquotaip) { - XFS_PURGE_INODE(qi->qi_uquotaip); + IRELE(qi->qi_uquotaip); qi->qi_uquotaip = NULL; /* paranoia */ } if (qi->qi_gquotaip) { - XFS_PURGE_INODE(qi->qi_gquotaip); + IRELE(qi->qi_gquotaip); qi->qi_gquotaip = NULL; } mutex_destroy(&qi->qi_quotaofflock); - kmem_free(qi, sizeof(xfs_quotainfo_t)); + kmem_free(qi); mp->m_quotainfo = NULL; } @@ -1394,7 +1392,7 @@ xfs_qm_qino_alloc( * locked exclusively and joined to the transaction already. */ ASSERT(xfs_isilocked(*ip, XFS_ILOCK_EXCL)); - VN_HOLD(XFS_ITOV((*ip))); + IHOLD(*ip); /* * Make the changes in the superblock, and log those too. @@ -1623,7 +1621,7 @@ xfs_qm_dqiterate( break; } while (nmaps > 0); - kmem_free(map, XFS_DQITER_MAP_SIZE * sizeof(*map)); + kmem_free(map); return error; } @@ -2079,7 +2077,7 @@ xfs_qm_shake_freelist( * Try to grab the flush lock. If this dquot is in the process of * getting flushed to disk, we don't want to reclaim it. */ - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { xfs_dqunlock(dqp); dqp = dqp->dq_flnext; continue; @@ -2257,7 +2255,7 @@ xfs_qm_dqreclaim_one(void) * Try to grab the flush lock. If this dquot is in the process of * getting flushed to disk, we don't want to reclaim it. */ - if (! xfs_qm_dqflock_nowait(dqp)) { + if (!xfs_dqflock_nowait(dqp)) { xfs_dqunlock(dqp); continue; } diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h index cd2300e..44f2534 100644 --- a/fs/xfs/quota/xfs_qm.h +++ b/fs/xfs/quota/xfs_qm.h @@ -165,7 +165,7 @@ typedef struct xfs_dquot_acct { #define XFS_QM_RELE(xqm) ((xqm)->qm_nrefs--) extern void xfs_qm_destroy_quotainfo(xfs_mount_t *); -extern void xfs_qm_mount_quotas(xfs_mount_t *, int); +extern void xfs_qm_mount_quotas(xfs_mount_t *); extern int xfs_qm_quotacheck(xfs_mount_t *); extern void xfs_qm_unmount_quotadestroy(xfs_mount_t *); extern int xfs_qm_unmount_quotas(xfs_mount_t *); diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index f4f6c4c..eea2e60 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -162,7 +162,7 @@ xfs_qm_newmount( * mounting, and get on with the boring life * without disk quotas. */ - xfs_qm_mount_quotas(mp, 0); + xfs_qm_mount_quotas(mp); } else { /* * Clear the quota flags, but remember them. This @@ -184,13 +184,12 @@ STATIC int xfs_qm_endmount( xfs_mount_t *mp, uint needquotamount, - uint quotaflags, - int mfsi_flags) + uint quotaflags) { if (needquotamount) { ASSERT(mp->m_qflags == 0); mp->m_qflags = quotaflags; - xfs_qm_mount_quotas(mp, mfsi_flags); + xfs_qm_mount_quotas(mp); } #if defined(DEBUG) && defined(XFS_LOUD_RECOVERY) diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 768a3b2..1a3b803 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -362,11 +362,11 @@ xfs_qm_scall_quotaoff( * if we don't need them anymore. */ if ((dqtype & XFS_QMOPT_UQUOTA) && XFS_QI_UQIP(mp)) { - XFS_PURGE_INODE(XFS_QI_UQIP(mp)); + IRELE(XFS_QI_UQIP(mp)); XFS_QI_UQIP(mp) = NULL; } if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && XFS_QI_GQIP(mp)) { - XFS_PURGE_INODE(XFS_QI_GQIP(mp)); + IRELE(XFS_QI_GQIP(mp)); XFS_QI_GQIP(mp) = NULL; } out_error: @@ -1034,7 +1034,7 @@ xfs_qm_dqrele_all_inodes( { xfs_inode_t *ip, *topino; uint ireclaims; - bhv_vnode_t *vp; + struct inode *vp; boolean_t vnode_refd; ASSERT(mp->m_quotainfo); @@ -1059,7 +1059,7 @@ again: ip = ip->i_mnext; continue; } - vp = XFS_ITOV_NULL(ip); + vp = VFS_I(ip); if (!vp) { ASSERT(ip->i_udquot == NULL); ASSERT(ip->i_gdquot == NULL); @@ -1449,14 +1449,14 @@ xfs_qm_internalqcheck( for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) { xfs_dqtest_cmp(d); e = (xfs_dqtest_t *) d->HL_NEXT; - kmem_free(d, sizeof(xfs_dqtest_t)); + kmem_free(d); d = e; } h1 = &qmtest_gdqtab[i]; for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) { xfs_dqtest_cmp(d); e = (xfs_dqtest_t *) d->HL_NEXT; - kmem_free(d, sizeof(xfs_dqtest_t)); + kmem_free(d); d = e; } } @@ -1467,8 +1467,8 @@ xfs_qm_internalqcheck( } else { cmn_err(CE_DEBUG, "******** quotacheck successful! ********"); } - kmem_free(qmtest_udqtab, qmtest_hashmask * sizeof(xfs_dqhash_t)); - kmem_free(qmtest_gdqtab, qmtest_hashmask * sizeof(xfs_dqhash_t)); + kmem_free(qmtest_udqtab); + kmem_free(qmtest_gdqtab); mutex_unlock(&qcheck_lock); return (qmtest_nfails); } diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h index 5e4a40b..c4fcea6 100644 --- a/fs/xfs/quota/xfs_quota_priv.h +++ b/fs/xfs/quota/xfs_quota_priv.h @@ -158,9 +158,6 @@ for ((dqp) = (qlist)->qh_next; (dqp) != (xfs_dquot_t *)(qlist); \ #define XFS_IS_SUSER_DQUOT(dqp) \ (!((dqp)->q_core.d_id)) -#define XFS_PURGE_INODE(ip) \ - IRELE(ip); - #define DQFLAGTO_TYPESTR(d) (((d)->dq_flags & XFS_DQ_USER) ? "USR" : \ (((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \ (((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???"))) diff --git a/fs/xfs/support/ktrace.c b/fs/xfs/support/ktrace.c index 0b75d30..a34ef05 100644 --- a/fs/xfs/support/ktrace.c +++ b/fs/xfs/support/ktrace.c @@ -89,7 +89,7 @@ ktrace_alloc(int nentries, unsigned int __nocast sleep) if (sleep & KM_SLEEP) panic("ktrace_alloc: NULL memory on KM_SLEEP request!"); - kmem_free(ktp, sizeof(*ktp)); + kmem_free(ktp); return NULL; } @@ -126,7 +126,7 @@ ktrace_free(ktrace_t *ktp) } else { entries_size = (int)(ktp->kt_nentries * sizeof(ktrace_entry_t)); - kmem_free(ktp->kt_entries, entries_size); + kmem_free(ktp->kt_entries); } kmem_zone_free(ktrace_hdr_zone, ktp); diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c index 493a6ec..5830c04 100644 --- a/fs/xfs/support/uuid.c +++ b/fs/xfs/support/uuid.c @@ -17,7 +17,7 @@ */ #include <xfs.h> -static mutex_t uuid_monitor; +static DEFINE_MUTEX(uuid_monitor); static int uuid_table_size; static uuid_t *uuid_table; @@ -132,9 +132,3 @@ uuid_table_remove(uuid_t *uuid) ASSERT(i < uuid_table_size); mutex_unlock(&uuid_monitor); } - -void __init -uuid_init(void) -{ - mutex_init(&uuid_monitor); -} diff --git a/fs/xfs/support/uuid.h b/fs/xfs/support/uuid.h index b6f5922..cff5b60 100644 --- a/fs/xfs/support/uuid.h +++ b/fs/xfs/support/uuid.h @@ -22,7 +22,6 @@ typedef struct { unsigned char __u_bits[16]; } uuid_t; -extern void uuid_init(void); extern void uuid_create_nil(uuid_t *uuid); extern int uuid_is_nil(uuid_t *uuid); extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2); diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index ebee3a4..b2f639a 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -37,15 +37,15 @@ #include <linux/capability.h> #include <linux/posix_acl_xattr.h> -STATIC int xfs_acl_setmode(bhv_vnode_t *, xfs_acl_t *, int *); +STATIC int xfs_acl_setmode(struct inode *, xfs_acl_t *, int *); STATIC void xfs_acl_filter_mode(mode_t, xfs_acl_t *); STATIC void xfs_acl_get_endian(xfs_acl_t *); STATIC int xfs_acl_access(uid_t, gid_t, xfs_acl_t *, mode_t, cred_t *); STATIC int xfs_acl_invalid(xfs_acl_t *); STATIC void xfs_acl_sync_mode(mode_t, xfs_acl_t *); -STATIC void xfs_acl_get_attr(bhv_vnode_t *, xfs_acl_t *, int, int, int *); -STATIC void xfs_acl_set_attr(bhv_vnode_t *, xfs_acl_t *, int, int *); -STATIC int xfs_acl_allow_set(bhv_vnode_t *, int); +STATIC void xfs_acl_get_attr(struct inode *, xfs_acl_t *, int, int, int *); +STATIC void xfs_acl_set_attr(struct inode *, xfs_acl_t *, int, int *); +STATIC int xfs_acl_allow_set(struct inode *, int); kmem_zone_t *xfs_acl_zone; @@ -55,7 +55,7 @@ kmem_zone_t *xfs_acl_zone; */ int xfs_acl_vhasacl_access( - bhv_vnode_t *vp) + struct inode *vp) { int error; @@ -68,7 +68,7 @@ xfs_acl_vhasacl_access( */ int xfs_acl_vhasacl_default( - bhv_vnode_t *vp) + struct inode *vp) { int error; @@ -207,7 +207,7 @@ posix_acl_xfs_to_xattr( int xfs_acl_vget( - bhv_vnode_t *vp, + struct inode *vp, void *acl, size_t size, int kind) @@ -217,7 +217,6 @@ xfs_acl_vget( posix_acl_xattr_header *ext_acl = acl; int flags = 0; - VN_HOLD(vp); if(size) { if (!(_ACL_ALLOC(xfs_acl))) { error = ENOMEM; @@ -239,11 +238,10 @@ xfs_acl_vget( goto out; } if (kind == _ACL_TYPE_ACCESS) - xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, xfs_acl); + xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, xfs_acl); error = -posix_acl_xfs_to_xattr(xfs_acl, ext_acl, size); } out: - VN_RELE(vp); if(xfs_acl) _ACL_FREE(xfs_acl); return -error; @@ -251,28 +249,26 @@ out: int xfs_acl_vremove( - bhv_vnode_t *vp, + struct inode *vp, int kind) { int error; - VN_HOLD(vp); error = xfs_acl_allow_set(vp, kind); if (!error) { - error = xfs_attr_remove(xfs_vtoi(vp), + error = xfs_attr_remove(XFS_I(vp), kind == _ACL_TYPE_DEFAULT? SGI_ACL_DEFAULT: SGI_ACL_FILE, ATTR_ROOT); if (error == ENOATTR) error = 0; /* 'scool */ } - VN_RELE(vp); return -error; } int xfs_acl_vset( - bhv_vnode_t *vp, + struct inode *vp, void *acl, size_t size, int kind) @@ -298,7 +294,6 @@ xfs_acl_vset( return 0; } - VN_HOLD(vp); error = xfs_acl_allow_set(vp, kind); /* Incoming ACL exists, set file mode based on its value */ @@ -321,7 +316,6 @@ xfs_acl_vset( } out: - VN_RELE(vp); _ACL_FREE(xfs_acl); return -error; } @@ -341,8 +335,7 @@ xfs_acl_iaccess( /* If the file has no ACL return -1. */ rval = sizeof(xfs_acl_t); - if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval, - ATTR_ROOT | ATTR_KERNACCESS)) { + if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval, ATTR_ROOT)) { _ACL_FREE(acl); return -1; } @@ -364,7 +357,7 @@ xfs_acl_iaccess( STATIC int xfs_acl_allow_set( - bhv_vnode_t *vp, + struct inode *vp, int kind) { if (vp->i_flags & (S_IMMUTABLE|S_APPEND)) @@ -373,7 +366,7 @@ xfs_acl_allow_set( return ENOTDIR; if (vp->i_sb->s_flags & MS_RDONLY) return EROFS; - if (xfs_vtoi(vp)->i_d.di_uid != current->fsuid && !capable(CAP_FOWNER)) + if (XFS_I(vp)->i_d.di_uid != current->fsuid && !capable(CAP_FOWNER)) return EPERM; return 0; } @@ -567,7 +560,7 @@ xfs_acl_get_endian( */ STATIC void xfs_acl_get_attr( - bhv_vnode_t *vp, + struct inode *vp, xfs_acl_t *aclp, int kind, int flags, @@ -577,7 +570,7 @@ xfs_acl_get_attr( ASSERT((flags & ATTR_KERNOVAL) ? (aclp == NULL) : 1); flags |= ATTR_ROOT; - *error = xfs_attr_get(xfs_vtoi(vp), + *error = xfs_attr_get(XFS_I(vp), kind == _ACL_TYPE_ACCESS ? SGI_ACL_FILE : SGI_ACL_DEFAULT, (char *)aclp, &len, flags); @@ -591,7 +584,7 @@ xfs_acl_get_attr( */ STATIC void xfs_acl_set_attr( - bhv_vnode_t *vp, + struct inode *vp, xfs_acl_t *aclp, int kind, int *error) @@ -616,7 +609,7 @@ xfs_acl_set_attr( INT_SET(newace->ae_perm, ARCH_CONVERT, ace->ae_perm); } INT_SET(newacl->acl_cnt, ARCH_CONVERT, aclp->acl_cnt); - *error = xfs_attr_set(xfs_vtoi(vp), + *error = xfs_attr_set(XFS_I(vp), kind == _ACL_TYPE_ACCESS ? SGI_ACL_FILE: SGI_ACL_DEFAULT, (char *)newacl, len, ATTR_ROOT); @@ -625,7 +618,7 @@ xfs_acl_set_attr( int xfs_acl_vtoacl( - bhv_vnode_t *vp, + struct inode *vp, xfs_acl_t *access_acl, xfs_acl_t *default_acl) { @@ -640,7 +633,7 @@ xfs_acl_vtoacl( if (error) access_acl->acl_cnt = XFS_ACL_NOT_PRESENT; else /* We have a good ACL and the file mode, synchronize. */ - xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, access_acl); + xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, access_acl); } if (default_acl) { @@ -657,7 +650,7 @@ xfs_acl_vtoacl( */ int xfs_acl_inherit( - bhv_vnode_t *vp, + struct inode *vp, mode_t mode, xfs_acl_t *pdaclp) { @@ -716,11 +709,11 @@ out_error: */ STATIC int xfs_acl_setmode( - bhv_vnode_t *vp, + struct inode *vp, xfs_acl_t *acl, int *basicperms) { - bhv_vattr_t va; + struct iattr iattr; xfs_acl_entry_t *ap; xfs_acl_entry_t *gap = NULL; int i, nomask = 1; @@ -734,25 +727,25 @@ xfs_acl_setmode( * Copy the u::, g::, o::, and m:: bits from the ACL into the * mode. The m:: bits take precedence over the g:: bits. */ - va.va_mask = XFS_AT_MODE; - va.va_mode = xfs_vtoi(vp)->i_d.di_mode; - va.va_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO); + iattr.ia_valid = ATTR_MODE; + iattr.ia_mode = XFS_I(vp)->i_d.di_mode; + iattr.ia_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO); ap = acl->acl_entry; for (i = 0; i < acl->acl_cnt; ++i) { switch (ap->ae_tag) { case ACL_USER_OBJ: - va.va_mode |= ap->ae_perm << 6; + iattr.ia_mode |= ap->ae_perm << 6; break; case ACL_GROUP_OBJ: gap = ap; break; case ACL_MASK: /* more than just standard modes */ nomask = 0; - va.va_mode |= ap->ae_perm << 3; + iattr.ia_mode |= ap->ae_perm << 3; *basicperms = 0; break; case ACL_OTHER: - va.va_mode |= ap->ae_perm; + iattr.ia_mode |= ap->ae_perm; break; default: /* more than just standard modes */ *basicperms = 0; @@ -763,9 +756,9 @@ xfs_acl_setmode( /* Set the group bits from ACL_GROUP_OBJ if there's no ACL_MASK */ if (gap && nomask) - va.va_mode |= gap->ae_perm << 3; + iattr.ia_mode |= gap->ae_perm << 3; - return xfs_setattr(xfs_vtoi(vp), &va, 0, sys_cred); + return xfs_setattr(XFS_I(vp), &iattr, 0, sys_cred); } /* diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 332a772..a4e293b 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -46,6 +46,8 @@ typedef struct xfs_acl { #define SGI_ACL_FILE_SIZE (sizeof(SGI_ACL_FILE)-1) #define SGI_ACL_DEFAULT_SIZE (sizeof(SGI_ACL_DEFAULT)-1) +#define _ACL_TYPE_ACCESS 1 +#define _ACL_TYPE_DEFAULT 2 #ifdef CONFIG_XFS_POSIX_ACL @@ -57,17 +59,15 @@ extern struct kmem_zone *xfs_acl_zone; (zone) = kmem_zone_init(sizeof(xfs_acl_t), (name)) #define xfs_acl_zone_destroy(zone) kmem_zone_destroy(zone) -extern int xfs_acl_inherit(bhv_vnode_t *, mode_t mode, xfs_acl_t *); +extern int xfs_acl_inherit(struct inode *, mode_t mode, xfs_acl_t *); extern int xfs_acl_iaccess(struct xfs_inode *, mode_t, cred_t *); -extern int xfs_acl_vtoacl(bhv_vnode_t *, xfs_acl_t *, xfs_acl_t *); -extern int xfs_acl_vhasacl_access(bhv_vnode_t *); -extern int xfs_acl_vhasacl_default(bhv_vnode_t *); -extern int xfs_acl_vset(bhv_vnode_t *, void *, size_t, int); -extern int xfs_acl_vget(bhv_vnode_t *, void *, size_t, int); -extern int xfs_acl_vremove(bhv_vnode_t *, int); +extern int xfs_acl_vtoacl(struct inode *, xfs_acl_t *, xfs_acl_t *); +extern int xfs_acl_vhasacl_access(struct inode *); +extern int xfs_acl_vhasacl_default(struct inode *); +extern int xfs_acl_vset(struct inode *, void *, size_t, int); +extern int xfs_acl_vget(struct inode *, void *, size_t, int); +extern int xfs_acl_vremove(struct inode *, int); -#define _ACL_TYPE_ACCESS 1 -#define _ACL_TYPE_DEFAULT 2 #define _ACL_PERM_INVALID(perm) ((perm) & ~(ACL_READ|ACL_WRITE|ACL_EXECUTE)) #define _ACL_INHERIT(c,m,d) (xfs_acl_inherit(c,m,d)) diff --git a/fs/xfs/xfs_arch.h b/fs/xfs/xfs_arch.h index f9472a2..0b3b5ef 100644 --- a/fs/xfs/xfs_arch.h +++ b/fs/xfs/xfs_arch.h @@ -92,16 +92,6 @@ ((__u8*)(pointer))[1] = (((value) ) & 0xff); \ } -/* define generic INT_ macros */ - -#define INT_GET(reference,arch) \ - (((arch) == ARCH_NOCONVERT) \ - ? \ - (reference) \ - : \ - INT_SWAP((reference),(reference)) \ - ) - /* does not return a value */ #define INT_SET(reference,arch,valueref) \ (__builtin_constant_p(valueref) ? \ @@ -112,64 +102,6 @@ ) \ ) -/* does not return a value */ -#define INT_MOD_EXPR(reference,arch,code) \ - (((arch) == ARCH_NOCONVERT) \ - ? \ - (void)((reference) code) \ - : \ - (void)( \ - (reference) = INT_GET((reference),arch) , \ - ((reference) code), \ - INT_SET(reference, arch, reference) \ - ) \ - ) - -/* does not return a value */ -#define INT_MOD(reference,arch,delta) \ - (void)( \ - INT_MOD_EXPR(reference,arch,+=(delta)) \ - ) - -/* - * INT_COPY - copy a value between two locations with the - * _same architecture_ but _potentially different sizes_ - * - * if the types of the two parameters are equal or they are - * in native architecture, a simple copy is done - * - * otherwise, architecture conversions are done - * - */ - -/* does not return a value */ -#define INT_COPY(dst,src,arch) \ - ( \ - ((sizeof(dst) == sizeof(src)) || ((arch) == ARCH_NOCONVERT)) \ - ? \ - (void)((dst) = (src)) \ - : \ - INT_SET(dst, arch, INT_GET(src, arch)) \ - ) - -/* - * INT_XLATE - copy a value in either direction between two locations - * with different architectures - * - * dir < 0 - copy from memory to buffer (native to arch) - * dir > 0 - copy from buffer to memory (arch to native) - */ - -/* does not return a value */ -#define INT_XLATE(buf,mem,dir,arch) {\ - ASSERT(dir); \ - if (dir>0) { \ - (mem)=INT_GET(buf, arch); \ - } else { \ - INT_SET(buf, arch, mem); \ - } \ -} - /* * In directories inode numbers are stored as unaligned arrays of unsigned * 8bit integers on disk. diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index df151a8..f7cdc28 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -16,8 +16,6 @@ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -#include <linux/capability.h> - #include "xfs.h" #include "xfs_fs.h" #include "xfs_types.h" @@ -57,11 +55,6 @@ * Provide the external interfaces to manage attribute lists. */ -#define ATTR_SYSCOUNT 2 -static struct attrnames posix_acl_access; -static struct attrnames posix_acl_default; -static struct attrnames *attr_system_names[ATTR_SYSCOUNT]; - /*======================================================================== * Function prototypes for the kernel. *========================================================================*/ @@ -116,6 +109,17 @@ xfs_attr_name_to_xname( return 0; } +STATIC int +xfs_inode_hasattr( + struct xfs_inode *ip) +{ + if (!XFS_IFORK_Q(ip) || + (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && + ip->i_d.di_anextents == 0)) + return 0; + return 1; +} + /*======================================================================== * Overall external interface routines. *========================================================================*/ @@ -127,10 +131,8 @@ xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name, xfs_da_args_t args; int error; - if ((XFS_IFORK_Q(ip) == 0) || - (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - ip->i_d.di_anextents == 0)) - return(ENOATTR); + if (!xfs_inode_hasattr(ip)) + return ENOATTR; /* * Fill in the arg structure for this request. @@ -148,11 +150,7 @@ xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name, /* * Decide on what work routines to call based on the inode size. */ - if (XFS_IFORK_Q(ip) == 0 || - (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - ip->i_d.di_anextents == 0)) { - error = XFS_ERROR(ENOATTR); - } else if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { + if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { error = xfs_attr_shortform_getvalue(&args); } else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK)) { error = xfs_attr_leaf_get(&args); @@ -196,6 +194,46 @@ xfs_attr_get( return(error); } +/* + * Calculate how many blocks we need for the new attribute, + */ +int +xfs_attr_calc_size( + struct xfs_inode *ip, + int namelen, + int valuelen, + int *local) +{ + struct xfs_mount *mp = ip->i_mount; + int size; + int nblks; + + /* + * Determine space new attribute will use, and if it would be + * "local" or "remote" (note: local != inline). + */ + size = xfs_attr_leaf_newentsize(namelen, valuelen, + mp->m_sb.sb_blocksize, local); + + nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); + if (*local) { + if (size > (mp->m_sb.sb_blocksize >> 1)) { + /* Double split possible */ + nblks *= 2; + } + } else { + /* + * Out of line attribute, cannot double split, but + * make room for the attribute value itself. + */ + uint dblocks = XFS_B_TO_FSB(mp, valuelen); + nblks += dblocks; + nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK); + } + + return nblks; +} + STATIC int xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, char *value, int valuelen, int flags) @@ -204,10 +242,9 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, xfs_fsblock_t firstblock; xfs_bmap_free_t flist; int error, err2, committed; - int local, size; - uint nblks; xfs_mount_t *mp = dp->i_mount; int rsvd = (flags & ATTR_ROOT) != 0; + int local; /* * Attach the dquots to the inode. @@ -241,33 +278,10 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, args.firstblock = &firstblock; args.flist = &flist; args.whichfork = XFS_ATTR_FORK; - args.addname = 1; - args.oknoent = 1; - - /* - * Determine space new attribute will use, and if it would be - * "local" or "remote" (note: local != inline). - */ - size = xfs_attr_leaf_newentsize(name->len, valuelen, - mp->m_sb.sb_blocksize, &local); - - nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); - if (local) { - if (size > (mp->m_sb.sb_blocksize >> 1)) { - /* Double split possible */ - nblks <<= 1; - } - } else { - uint dblocks = XFS_B_TO_FSB(mp, valuelen); - /* Out of line attribute, cannot double split, but make - * room for the attribute value itself. - */ - nblks += dblocks; - nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK); - } + args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; /* Size is now blocks for attribute data */ - args.total = nblks; + args.total = xfs_attr_calc_size(dp, name->len, valuelen, &local); /* * Start our first transaction of the day. @@ -289,18 +303,17 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, if (rsvd) args.trans->t_flags |= XFS_TRANS_RESERVE; - if ((error = xfs_trans_reserve(args.trans, (uint) nblks, - XFS_ATTRSET_LOG_RES(mp, nblks), - 0, XFS_TRANS_PERM_LOG_RES, - XFS_ATTRSET_LOG_COUNT))) { + if ((error = xfs_trans_reserve(args.trans, args.total, + XFS_ATTRSET_LOG_RES(mp, args.total), 0, + XFS_TRANS_PERM_LOG_RES, XFS_ATTRSET_LOG_COUNT))) { xfs_trans_cancel(args.trans, 0); return(error); } xfs_ilock(dp, XFS_ILOCK_EXCL); - error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, nblks, 0, - rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES : - XFS_QMOPT_RES_REGBLKS); + error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, args.total, 0, + rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES : + XFS_QMOPT_RES_REGBLKS); if (error) { xfs_iunlock(dp, XFS_ILOCK_EXCL); xfs_trans_cancel(args.trans, XFS_TRANS_RELEASE_LOG_RES); @@ -387,7 +400,9 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name, * Commit the leaf transformation. We'll need another (linked) * transaction to add the new attribute to the leaf. */ - if ((error = xfs_attr_rolltrans(&args.trans, dp))) + + error = xfs_trans_roll(&args.trans, dp); + if (error) goto out; } @@ -529,9 +544,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) /* * Decide on what work routines to call based on the inode size. */ - if (XFS_IFORK_Q(dp) == 0 || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { + if (!xfs_inode_hasattr(dp)) { error = XFS_ERROR(ENOATTR); goto out; } @@ -601,29 +614,33 @@ xfs_attr_remove( return error; xfs_ilock(dp, XFS_ILOCK_SHARED); - if (XFS_IFORK_Q(dp) == 0 || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { + if (!xfs_inode_hasattr(dp)) { xfs_iunlock(dp, XFS_ILOCK_SHARED); - return(XFS_ERROR(ENOATTR)); + return XFS_ERROR(ENOATTR); } xfs_iunlock(dp, XFS_ILOCK_SHARED); return xfs_attr_remove_int(dp, &xname, flags); } -STATIC int +int xfs_attr_list_int(xfs_attr_list_context_t *context) { int error; xfs_inode_t *dp = context->dp; + XFS_STATS_INC(xs_attr_list); + + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) + return EIO; + + xfs_ilock(dp, XFS_ILOCK_SHARED); + xfs_attr_trace_l_c("syscall start", context); + /* * Decide on what work routines to call based on the inode size. */ - if (XFS_IFORK_Q(dp) == 0 || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { + if (!xfs_inode_hasattr(dp)) { error = 0; } else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { error = xfs_attr_shortform_list(context); @@ -632,6 +649,10 @@ xfs_attr_list_int(xfs_attr_list_context_t *context) } else { error = xfs_attr_node_list(context); } + + xfs_iunlock(dp, XFS_ILOCK_SHARED); + xfs_attr_trace_l_c("syscall end", context); + return error; } @@ -648,74 +669,50 @@ xfs_attr_list_int(xfs_attr_list_context_t *context) */ /*ARGSUSED*/ STATIC int -xfs_attr_put_listent(xfs_attr_list_context_t *context, attrnames_t *namesp, +xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags, char *name, int namelen, int valuelen, char *value) { + struct attrlist *alist = (struct attrlist *)context->alist; attrlist_ent_t *aep; int arraytop; ASSERT(!(context->flags & ATTR_KERNOVAL)); ASSERT(context->count >= 0); ASSERT(context->count < (ATTR_MAX_VALUELEN/8)); - ASSERT(context->firstu >= sizeof(*context->alist)); + ASSERT(context->firstu >= sizeof(*alist)); ASSERT(context->firstu <= context->bufsize); - arraytop = sizeof(*context->alist) + - context->count * sizeof(context->alist->al_offset[0]); + /* + * Only list entries in the right namespace. + */ + if (((context->flags & ATTR_SECURE) == 0) != + ((flags & XFS_ATTR_SECURE) == 0)) + return 0; + if (((context->flags & ATTR_ROOT) == 0) != + ((flags & XFS_ATTR_ROOT) == 0)) + return 0; + + arraytop = sizeof(*alist) + + context->count * sizeof(alist->al_offset[0]); context->firstu -= ATTR_ENTSIZE(namelen); if (context->firstu < arraytop) { xfs_attr_trace_l_c("buffer full", context); - context->alist->al_more = 1; + alist->al_more = 1; context->seen_enough = 1; return 1; } - aep = (attrlist_ent_t *)&(((char *)context->alist)[ context->firstu ]); + aep = (attrlist_ent_t *)&context->alist[context->firstu]; aep->a_valuelen = valuelen; memcpy(aep->a_name, name, namelen); - aep->a_name[ namelen ] = 0; - context->alist->al_offset[ context->count++ ] = context->firstu; - context->alist->al_count = context->count; + aep->a_name[namelen] = 0; + alist->al_offset[context->count++] = context->firstu; + alist->al_count = context->count; xfs_attr_trace_l_c("add", context); return 0; } -STATIC int -xfs_attr_kern_list(xfs_attr_list_context_t *context, attrnames_t *namesp, - char *name, int namelen, - int valuelen, char *value) -{ - char *offset; - int arraytop; - - ASSERT(context->count >= 0); - - arraytop = context->count + namesp->attr_namelen + namelen + 1; - if (arraytop > context->firstu) { - context->count = -1; /* insufficient space */ - return 1; - } - offset = (char *)context->alist + context->count; - strncpy(offset, namesp->attr_name, namesp->attr_namelen); - offset += namesp->attr_namelen; - strncpy(offset, name, namelen); /* real name */ - offset += namelen; - *offset = '\0'; - context->count += namesp->attr_namelen + namelen + 1; - return 0; -} - -/*ARGSUSED*/ -STATIC int -xfs_attr_kern_list_sizes(xfs_attr_list_context_t *context, attrnames_t *namesp, - char *name, int namelen, - int valuelen, char *value) -{ - context->count += namesp->attr_namelen + namelen + 1; - return 0; -} - /* * Generate a list of extended attribute names and optionally * also value lengths. Positive return value follows the XFS @@ -732,10 +729,9 @@ xfs_attr_list( attrlist_cursor_kern_t *cursor) { xfs_attr_list_context_t context; + struct attrlist *alist; int error; - XFS_STATS_INC(xs_attr_list); - /* * Validate the cursor. */ @@ -756,52 +752,23 @@ xfs_attr_list( /* * Initialize the output buffer. */ + memset(&context, 0, sizeof(context)); context.dp = dp; context.cursor = cursor; - context.count = 0; - context.dupcnt = 0; context.resynch = 1; context.flags = flags; - context.seen_enough = 0; - context.alist = (attrlist_t *)buffer; - context.put_value = 0; - - if (flags & ATTR_KERNAMELS) { - context.bufsize = bufsize; - context.firstu = context.bufsize; - if (flags & ATTR_KERNOVAL) - context.put_listent = xfs_attr_kern_list_sizes; - else - context.put_listent = xfs_attr_kern_list; - } else { - context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */ - context.firstu = context.bufsize; - context.alist->al_count = 0; - context.alist->al_more = 0; - context.alist->al_offset[0] = context.bufsize; - context.put_listent = xfs_attr_put_listent; - } - - if (XFS_FORCED_SHUTDOWN(dp->i_mount)) - return EIO; + context.alist = buffer; + context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */ + context.firstu = context.bufsize; + context.put_listent = xfs_attr_put_listent; - xfs_ilock(dp, XFS_ILOCK_SHARED); - xfs_attr_trace_l_c("syscall start", &context); + alist = (struct attrlist *)context.alist; + alist->al_count = 0; + alist->al_more = 0; + alist->al_offset[0] = context.bufsize; error = xfs_attr_list_int(&context); - - xfs_iunlock(dp, XFS_ILOCK_SHARED); - xfs_attr_trace_l_c("syscall end", &context); - - if (context.flags & (ATTR_KERNOVAL|ATTR_KERNAMELS)) { - /* must return negated buffer size or the error */ - if (context.count < 0) - error = XFS_ERROR(ERANGE); - else - error = -context.count; - } else - ASSERT(error >= 0); - + ASSERT(error >= 0); return error; } @@ -816,12 +783,10 @@ xfs_attr_inactive(xfs_inode_t *dp) ASSERT(! XFS_NOT_DQATTACHED(mp, dp)); xfs_ilock(dp, XFS_ILOCK_SHARED); - if ((XFS_IFORK_Q(dp) == 0) || - (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { + if (!xfs_inode_hasattr(dp) || + dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { xfs_iunlock(dp, XFS_ILOCK_SHARED); - return(0); + return 0; } xfs_iunlock(dp, XFS_ILOCK_SHARED); @@ -854,10 +819,8 @@ xfs_attr_inactive(xfs_inode_t *dp) /* * Decide on what work routines to call based on the inode size. */ - if ((XFS_IFORK_Q(dp) == 0) || - (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { + if (!xfs_inode_hasattr(dp) || + dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) { error = 0; goto out; } @@ -974,7 +937,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) xfs_da_brelse(args->trans, bp); return(retval); } - args->rename = 1; /* an atomic rename */ + args->op_flags |= XFS_DA_OP_RENAME; /* an atomic rename */ args->blkno2 = args->blkno; /* set 2nd entry info*/ args->index2 = args->index; args->rmtblkno2 = args->rmtblkno; @@ -1019,7 +982,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) * Commit the current trans (including the inode) and start * a new one. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) return (error); /* @@ -1033,7 +997,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) * Commit the transaction that added the attr name so that * later routines can manage their own transactions. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) return (error); /* @@ -1054,7 +1019,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) * so that one disappears and one appears atomically. Then we * must remove the "old" attribute/value pair. */ - if (args->rename) { + if (args->op_flags & XFS_DA_OP_RENAME) { /* * In a separate transaction, set the incomplete flag on the * "old" attr and clear the incomplete flag on the "new" attr. @@ -1122,7 +1087,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) /* * Commit the remove and start the next trans in series. */ - error = xfs_attr_rolltrans(&args->trans, dp); + error = xfs_trans_roll(&args->trans, dp); } else if (args->rmtblkno > 0) { /* @@ -1307,7 +1272,7 @@ restart: } else if (retval == EEXIST) { if (args->flags & ATTR_CREATE) goto out; - args->rename = 1; /* atomic rename op */ + args->op_flags |= XFS_DA_OP_RENAME; /* atomic rename op */ args->blkno2 = args->blkno; /* set 2nd entry info*/ args->index2 = args->index; args->rmtblkno2 = args->rmtblkno; @@ -1353,7 +1318,8 @@ restart: * Commit the node conversion and start the next * trans in the chain. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) goto out; goto restart; @@ -1404,7 +1370,8 @@ restart: * Commit the leaf addition or btree split and start the next * trans in the chain. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) goto out; /* @@ -1425,7 +1392,7 @@ restart: * so that one disappears and one appears atomically. Then we * must remove the "old" attribute/value pair. */ - if (args->rename) { + if (args->op_flags & XFS_DA_OP_RENAME) { /* * In a separate transaction, set the incomplete flag on the * "old" attr and clear the incomplete flag on the "new" attr. @@ -1504,7 +1471,8 @@ restart: /* * Commit and start the next trans in the chain. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) goto out; } else if (args->rmtblkno > 0) { @@ -1636,7 +1604,8 @@ xfs_attr_node_removename(xfs_da_args_t *args) /* * Commit the Btree join operation and start a new trans. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) goto out; } @@ -2137,7 +2106,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args) /* * Start the next trans in the chain. */ - if ((error = xfs_attr_rolltrans(&args->trans, dp))) + error = xfs_trans_roll(&args->trans, dp); + if (error) return (error); } @@ -2287,7 +2257,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) /* * Close out trans and start the next one in the chain. */ - if ((error = xfs_attr_rolltrans(&args->trans, args->dp))) + error = xfs_trans_roll(&args->trans, args->dp); + if (error) return (error); } return(0); @@ -2300,23 +2271,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args) void xfs_attr_trace_l_c(char *where, struct xfs_attr_list_context *context) { - xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where, - (__psunsigned_t)context->dp, - (__psunsigned_t)context->cursor->hashval, - (__psunsigned_t)context->cursor->blkno, - (__psunsigned_t)context->cursor->offset, - (__psunsigned_t)context->alist, - (__psunsigned_t)context->bufsize, - (__psunsigned_t)context->count, - (__psunsigned_t)context->firstu, - (__psunsigned_t) - ((context->count > 0) && - !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL))) - ? (ATTR_ENTRY(context->alist, - context->count-1)->a_valuelen) - : 0, - (__psunsigned_t)context->dupcnt, - (__psunsigned_t)context->flags, + xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where, context, (__psunsigned_t)NULL, (__psunsigned_t)NULL, (__psunsigned_t)NULL); @@ -2329,23 +2284,7 @@ void xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context, struct xfs_da_intnode *node) { - xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where, - (__psunsigned_t)context->dp, - (__psunsigned_t)context->cursor->hashval, - (__psunsigned_t)context->cursor->blkno, - (__psunsigned_t)context->cursor->offset, - (__psunsigned_t)context->alist, - (__psunsigned_t)context->bufsize, - (__psunsigned_t)context->count, - (__psunsigned_t)context->firstu, - (__psunsigned_t) - ((context->count > 0) && - !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL))) - ? (ATTR_ENTRY(context->alist, - context->count-1)->a_valuelen) - : 0, - (__psunsigned_t)context->dupcnt, - (__psunsigned_t)context->flags, + xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where, context, (__psunsigned_t)be16_to_cpu(node->hdr.count), (__psunsigned_t)be32_to_cpu(node->btree[0].hashval), (__psunsigned_t)be32_to_cpu(node->btree[ @@ -2359,23 +2298,7 @@ void xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context, struct xfs_da_node_entry *btree) { - xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where, - (__psunsigned_t)context->dp, - (__psunsigned_t)context->cursor->hashval, - (__psunsigned_t)context->cursor->blkno, - (__psunsigned_t)context->cursor->offset, - (__psunsigned_t)context->alist, - (__psunsigned_t)context->bufsize, - (__psunsigned_t)context->count, - (__psunsigned_t)context->firstu, - (__psunsigned_t) - ((context->count > 0) && - !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL))) - ? (ATTR_ENTRY(context->alist, - context->count-1)->a_valuelen) - : 0, - (__psunsigned_t)context->dupcnt, - (__psunsigned_t)context->flags, + xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where, context, (__psunsigned_t)be32_to_cpu(btree->hashval), (__psunsigned_t)be32_to_cpu(btree->before), (__psunsigned_t)NULL); @@ -2388,23 +2311,7 @@ void xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context, struct xfs_attr_leafblock *leaf) { - xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where, - (__psunsigned_t)context->dp, - (__psunsigned_t)context->cursor->hashval, - (__psunsigned_t)context->cursor->blkno, - (__psunsigned_t)context->cursor->offset, - (__psunsigned_t)context->alist, - (__psunsigned_t)context->bufsize, - (__psunsigned_t)context->count, - (__psunsigned_t)context->firstu, - (__psunsigned_t) - ((context->count > 0) && - !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL))) - ? (ATTR_ENTRY(context->alist, - context->count-1)->a_valuelen) - : 0, - (__psunsigned_t)context->dupcnt, - (__psunsigned_t)context->flags, + xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where, context, (__psunsigned_t)be16_to_cpu(leaf->hdr.count), (__psunsigned_t)be32_to_cpu(leaf->entries[0].hashval), (__psunsigned_t)be32_to_cpu(leaf->entries[ @@ -2417,329 +2324,24 @@ xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context, */ void xfs_attr_trace_enter(int type, char *where, - __psunsigned_t a2, __psunsigned_t a3, - __psunsigned_t a4, __psunsigned_t a5, - __psunsigned_t a6, __psunsigned_t a7, - __psunsigned_t a8, __psunsigned_t a9, - __psunsigned_t a10, __psunsigned_t a11, - __psunsigned_t a12, __psunsigned_t a13, - __psunsigned_t a14, __psunsigned_t a15) + struct xfs_attr_list_context *context, + __psunsigned_t a13, __psunsigned_t a14, + __psunsigned_t a15) { ASSERT(xfs_attr_trace_buf); ktrace_enter(xfs_attr_trace_buf, (void *)((__psunsigned_t)type), - (void *)where, - (void *)a2, (void *)a3, (void *)a4, - (void *)a5, (void *)a6, (void *)a7, - (void *)a8, (void *)a9, (void *)a10, - (void *)a11, (void *)a12, (void *)a13, - (void *)a14, (void *)a15); + (void *)((__psunsigned_t)where), + (void *)((__psunsigned_t)context->dp), + (void *)((__psunsigned_t)context->cursor->hashval), + (void *)((__psunsigned_t)context->cursor->blkno), + (void *)((__psunsigned_t)context->cursor->offset), + (void *)((__psunsigned_t)context->alist), + (void *)((__psunsigned_t)context->bufsize), + (void *)((__psunsigned_t)context->count), + (void *)((__psunsigned_t)context->firstu), + NULL, + (void *)((__psunsigned_t)context->dupcnt), + (void *)((__psunsigned_t)context->flags), + (void *)a13, (void *)a14, (void *)a15); } #endif /* XFS_ATTR_TRACE */ - - -/*======================================================================== - * System (pseudo) namespace attribute interface routines. - *========================================================================*/ - -STATIC int -posix_acl_access_set( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - return xfs_acl_vset(vp, data, size, _ACL_TYPE_ACCESS); -} - -STATIC int -posix_acl_access_remove( - bhv_vnode_t *vp, char *name, int xflags) -{ - return xfs_acl_vremove(vp, _ACL_TYPE_ACCESS); -} - -STATIC int -posix_acl_access_get( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - return xfs_acl_vget(vp, data, size, _ACL_TYPE_ACCESS); -} - -STATIC int -posix_acl_access_exists( - bhv_vnode_t *vp) -{ - return xfs_acl_vhasacl_access(vp); -} - -STATIC int -posix_acl_default_set( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - return xfs_acl_vset(vp, data, size, _ACL_TYPE_DEFAULT); -} - -STATIC int -posix_acl_default_get( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - return xfs_acl_vget(vp, data, size, _ACL_TYPE_DEFAULT); -} - -STATIC int -posix_acl_default_remove( - bhv_vnode_t *vp, char *name, int xflags) -{ - return xfs_acl_vremove(vp, _ACL_TYPE_DEFAULT); -} - -STATIC int -posix_acl_default_exists( - bhv_vnode_t *vp) -{ - return xfs_acl_vhasacl_default(vp); -} - -static struct attrnames posix_acl_access = { - .attr_name = "posix_acl_access", - .attr_namelen = sizeof("posix_acl_access") - 1, - .attr_get = posix_acl_access_get, - .attr_set = posix_acl_access_set, - .attr_remove = posix_acl_access_remove, - .attr_exists = posix_acl_access_exists, -}; - -static struct attrnames posix_acl_default = { - .attr_name = "posix_acl_default", - .attr_namelen = sizeof("posix_acl_default") - 1, - .attr_get = posix_acl_default_get, - .attr_set = posix_acl_default_set, - .attr_remove = posix_acl_default_remove, - .attr_exists = posix_acl_default_exists, -}; - -static struct attrnames *attr_system_names[] = - { &posix_acl_access, &posix_acl_default }; - - -/*======================================================================== - * Namespace-prefix-style attribute name interface routines. - *========================================================================*/ - -STATIC int -attr_generic_set( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - return -xfs_attr_set(xfs_vtoi(vp), name, data, size, xflags); -} - -STATIC int -attr_generic_get( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - int error, asize = size; - - error = xfs_attr_get(xfs_vtoi(vp), name, data, &asize, xflags); - if (!error) - return asize; - return -error; -} - -STATIC int -attr_generic_remove( - bhv_vnode_t *vp, char *name, int xflags) -{ - return -xfs_attr_remove(xfs_vtoi(vp), name, xflags); -} - -STATIC int -attr_generic_listadd( - attrnames_t *prefix, - attrnames_t *namesp, - void *data, - size_t size, - ssize_t *result) -{ - char *p = data + *result; - - *result += prefix->attr_namelen; - *result += namesp->attr_namelen + 1; - if (!size) - return 0; - if (*result > size) - return -ERANGE; - strcpy(p, prefix->attr_name); - p += prefix->attr_namelen; - strcpy(p, namesp->attr_name); - p += namesp->attr_namelen + 1; - return 0; -} - -STATIC int -attr_system_list( - bhv_vnode_t *vp, - void *data, - size_t size, - ssize_t *result) -{ - attrnames_t *namesp; - int i, error = 0; - - for (i = 0; i < ATTR_SYSCOUNT; i++) { - namesp = attr_system_names[i]; - if (!namesp->attr_exists || !namesp->attr_exists(vp)) - continue; - error = attr_generic_listadd(&attr_system, namesp, - data, size, result); - if (error) - break; - } - return error; -} - -int -attr_generic_list( - bhv_vnode_t *vp, void *data, size_t size, int xflags, ssize_t *result) -{ - attrlist_cursor_kern_t cursor = { 0 }; - int error; - - error = xfs_attr_list(xfs_vtoi(vp), data, size, xflags, &cursor); - if (error > 0) - return -error; - *result = -error; - return attr_system_list(vp, data, size, result); -} - -attrnames_t * -attr_lookup_namespace( - char *name, - struct attrnames **names, - int nnames) -{ - int i; - - for (i = 0; i < nnames; i++) - if (!strncmp(name, names[i]->attr_name, names[i]->attr_namelen)) - return names[i]; - return NULL; -} - -/* - * Some checks to prevent people abusing EAs to get over quota: - * - Don't allow modifying user EAs on devices/symlinks; - * - Don't allow modifying user EAs if sticky bit set; - */ -STATIC int -attr_user_capable( - bhv_vnode_t *vp, - cred_t *cred) -{ - struct inode *inode = vn_to_inode(vp); - - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return -EPERM; - if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode) && - !capable(CAP_SYS_ADMIN)) - return -EPERM; - if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) && - (current_fsuid(cred) != inode->i_uid) && !capable(CAP_FOWNER)) - return -EPERM; - return 0; -} - -STATIC int -attr_trusted_capable( - bhv_vnode_t *vp, - cred_t *cred) -{ - struct inode *inode = vn_to_inode(vp); - - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return -EPERM; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - return 0; -} - -STATIC int -attr_system_set( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - attrnames_t *namesp; - int error; - - if (xflags & ATTR_CREATE) - return -EINVAL; - - namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT); - if (!namesp) - return -EOPNOTSUPP; - error = namesp->attr_set(vp, name, data, size, xflags); - if (!error) - error = vn_revalidate(vp); - return error; -} - -STATIC int -attr_system_get( - bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags) -{ - attrnames_t *namesp; - - namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT); - if (!namesp) - return -EOPNOTSUPP; - return namesp->attr_get(vp, name, data, size, xflags); -} - -STATIC int -attr_system_remove( - bhv_vnode_t *vp, char *name, int xflags) -{ - attrnames_t *namesp; - - namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT); - if (!namesp) - return -EOPNOTSUPP; - return namesp->attr_remove(vp, name, xflags); -} - -struct attrnames attr_system = { - .attr_name = "system.", - .attr_namelen = sizeof("system.") - 1, - .attr_flag = ATTR_SYSTEM, - .attr_get = attr_system_get, - .attr_set = attr_system_set, - .attr_remove = attr_system_remove, - .attr_capable = (attrcapable_t)fs_noerr, -}; - -struct attrnames attr_trusted = { - .attr_name = "trusted.", - .attr_namelen = sizeof("trusted.") - 1, - .attr_flag = ATTR_ROOT, - .attr_get = attr_generic_get, - .attr_set = attr_generic_set, - .attr_remove = attr_generic_remove, - .attr_capable = attr_trusted_capable, -}; - -struct attrnames attr_secure = { - .attr_name = "security.", - .attr_namelen = sizeof("security.") - 1, - .attr_flag = ATTR_SECURE, - .attr_get = attr_generic_get, - .attr_set = attr_generic_set, - .attr_remove = attr_generic_remove, - .attr_capable = (attrcapable_t)fs_noerr, -}; - -struct attrnames attr_user = { - .attr_name = "user.", - .attr_namelen = sizeof("user.") - 1, - .attr_get = attr_generic_get, - .attr_set = attr_generic_set, - .attr_remove = attr_generic_remove, - .attr_capable = attr_user_capable, -}; - -struct attrnames *attr_namespaces[] = - { &attr_system, &attr_trusted, &attr_secure, &attr_user }; diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h index 6cfc938..fb3b2a6 100644 --- a/fs/xfs/xfs_attr.h +++ b/fs/xfs/xfs_attr.h @@ -18,9 +18,11 @@ #ifndef __XFS_ATTR_H__ #define __XFS_ATTR_H__ +struct xfs_inode; +struct xfs_da_args; +struct xfs_attr_list_context; + /* - * xfs_attr.h - * * Large attribute lists are structured around Btrees where all the data * elements are in the leaf nodes. Attribute names are hashed into an int, * then that int is used as the index into the Btree. Since the hashval @@ -35,35 +37,6 @@ * External interfaces *========================================================================*/ -struct cred; -struct xfs_attr_list_context; - -typedef int (*attrset_t)(bhv_vnode_t *, char *, void *, size_t, int); -typedef int (*attrget_t)(bhv_vnode_t *, char *, void *, size_t, int); -typedef int (*attrremove_t)(bhv_vnode_t *, char *, int); -typedef int (*attrexists_t)(bhv_vnode_t *); -typedef int (*attrcapable_t)(bhv_vnode_t *, struct cred *); - -typedef struct attrnames { - char * attr_name; - unsigned int attr_namelen; - unsigned int attr_flag; - attrget_t attr_get; - attrset_t attr_set; - attrremove_t attr_remove; - attrexists_t attr_exists; - attrcapable_t attr_capable; -} attrnames_t; - -#define ATTR_NAMECOUNT 4 -extern struct attrnames attr_user; -extern struct attrnames attr_secure; -extern struct attrnames attr_system; -extern struct attrnames attr_trusted; -extern struct attrnames *attr_namespaces[ATTR_NAMECOUNT]; - -extern attrnames_t *attr_lookup_namespace(char *, attrnames_t **, int); -extern int attr_generic_list(bhv_vnode_t *, void *, size_t, int, ssize_t *); #define ATTR_DONTFOLLOW 0x0001 /* -- unused, from IRIX -- */ #define ATTR_ROOT 0x0002 /* use attrs in root (trusted) namespace */ @@ -71,16 +44,9 @@ extern int attr_generic_list(bhv_vnode_t *, void *, size_t, int, ssize_t *); #define ATTR_SECURE 0x0008 /* use attrs in security namespace */ #define ATTR_CREATE 0x0010 /* pure create: fail if attr already exists */ #define ATTR_REPLACE 0x0020 /* pure set: fail if attr does not exist */ -#define ATTR_SYSTEM 0x0100 /* use attrs in system (pseudo) namespace */ -#define ATTR_KERNACCESS 0x0400 /* [kernel] iaccess, inode held io-locked */ #define ATTR_KERNOTIME 0x1000 /* [kernel] don't update inode timestamps */ #define ATTR_KERNOVAL 0x2000 /* [kernel] get attr size only, not value */ -#define ATTR_KERNAMELS 0x4000 /* [kernel] list attr names (simple list) */ - -#define ATTR_KERNORMALS 0x0800 /* [kernel] normal attr list: user+secure */ -#define ATTR_KERNROOTLS 0x8000 /* [kernel] include root in the attr list */ -#define ATTR_KERNFULLS (ATTR_KERNORMALS|ATTR_KERNROOTLS) /* * The maximum size (into the kernel or returned from the kernel) of an @@ -119,22 +85,6 @@ typedef struct attrlist_ent { /* data from attr_list() */ &((char *)buffer)[ ((attrlist_t *)(buffer))->al_offset[index] ]) /* - * Multi-attribute operation vector. - */ -typedef struct attr_multiop { - int am_opcode; /* operation to perform (ATTR_OP_GET, etc.) */ - int am_error; /* [out arg] result of this sub-op (an errno) */ - char *am_attrname; /* attribute name to work with */ - char *am_attrvalue; /* [in/out arg] attribute value (raw bytes) */ - int am_length; /* [in/out arg] length of value */ - int am_flags; /* bitwise OR of attr API flags defined above */ -} attr_multiop_t; - -#define ATTR_OP_GET 1 /* return the indicated attr's value */ -#define ATTR_OP_SET 2 /* set/create the indicated attr/value pair */ -#define ATTR_OP_REMOVE 3 /* remove the indicated attr */ - -/* * Kernel-internal version of the attrlist cursor. */ typedef struct attrlist_cursor_kern { @@ -148,20 +98,41 @@ typedef struct attrlist_cursor_kern { /*======================================================================== - * Function prototypes for the kernel. + * Structure used to pass context around among the routines. *========================================================================*/ -struct xfs_inode; -struct attrlist_cursor_kern; -struct xfs_da_args; + +typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, int, + char *, int, int, char *); + +typedef struct xfs_attr_list_context { + struct xfs_inode *dp; /* inode */ + struct attrlist_cursor_kern *cursor; /* position in list */ + char *alist; /* output buffer */ + int seen_enough; /* T/F: seen enough of list? */ + ssize_t count; /* num used entries */ + int dupcnt; /* count dup hashvals seen */ + int bufsize; /* total buffer size */ + int firstu; /* first used byte in buffer */ + int flags; /* from VOP call */ + int resynch; /* T/F: resynch with cursor */ + int put_value; /* T/F: need value for listent */ + put_listent_func_t put_listent; /* list output fmt function */ + int index; /* index into output buffer */ +} xfs_attr_list_context_t; + + +/*======================================================================== + * Function prototypes for the kernel. + *========================================================================*/ /* * Overall external interface routines. */ +int xfs_attr_calc_size(struct xfs_inode *, int, int, int *); int xfs_attr_inactive(struct xfs_inode *dp); - -int xfs_attr_shortform_getvalue(struct xfs_da_args *); int xfs_attr_fetch(struct xfs_inode *, struct xfs_name *, char *, int *, int); int xfs_attr_rmtval_get(struct xfs_da_args *args); +int xfs_attr_list_int(struct xfs_attr_list_context *); #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 303d41e..79da6b2 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -94,13 +94,6 @@ STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index); * Namespace helper routines *========================================================================*/ -STATIC_INLINE attrnames_t * -xfs_attr_flags_namesp(int flags) -{ - return ((flags & XFS_ATTR_SECURE) ? &attr_secure: - ((flags & XFS_ATTR_ROOT) ? &attr_trusted : &attr_user)); -} - /* * If namespace bits don't match return 0. * If all match then return 1. @@ -111,25 +104,6 @@ xfs_attr_namesp_match(int arg_flags, int ondisk_flags) return XFS_ATTR_NSP_ONDISK(ondisk_flags) == XFS_ATTR_NSP_ARGS_TO_ONDISK(arg_flags); } -/* - * If namespace bits don't match and we don't have an override for it - * then return 0. - * If all match or are overridable then return 1. - */ -STATIC_INLINE int -xfs_attr_namesp_match_overrides(int arg_flags, int ondisk_flags) -{ - if (((arg_flags & ATTR_SECURE) == 0) != - ((ondisk_flags & XFS_ATTR_SECURE) == 0) && - !(arg_flags & ATTR_KERNORMALS)) - return 0; - if (((arg_flags & ATTR_ROOT) == 0) != - ((ondisk_flags & XFS_ATTR_ROOT) == 0) && - !(arg_flags & ATTR_KERNROOTLS)) - return 0; - return 1; -} - /*======================================================================== * External routines when attribute fork size < XFS_LITINO(mp). @@ -369,9 +343,10 @@ xfs_attr_shortform_remove(xfs_da_args_t *args) * Fix up the start offset of the attribute fork */ totsize -= size; - if (totsize == sizeof(xfs_attr_sf_hdr_t) && !args->addname && - (mp->m_flags & XFS_MOUNT_ATTR2) && - (dp->i_d.di_format != XFS_DINODE_FMT_BTREE)) { + if (totsize == sizeof(xfs_attr_sf_hdr_t) && + !(args->op_flags & XFS_DA_OP_ADDNAME) && + (mp->m_flags & XFS_MOUNT_ATTR2) && + (dp->i_d.di_format != XFS_DINODE_FMT_BTREE)) { /* * Last attribute now removed, revert to original * inode format making all literal area available @@ -389,9 +364,10 @@ xfs_attr_shortform_remove(xfs_da_args_t *args) xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); dp->i_d.di_forkoff = xfs_attr_shortform_bytesfit(dp, totsize); ASSERT(dp->i_d.di_forkoff); - ASSERT(totsize > sizeof(xfs_attr_sf_hdr_t) || args->addname || - !(mp->m_flags & XFS_MOUNT_ATTR2) || - dp->i_d.di_format == XFS_DINODE_FMT_BTREE); + ASSERT(totsize > sizeof(xfs_attr_sf_hdr_t) || + (args->op_flags & XFS_DA_OP_ADDNAME) || + !(mp->m_flags & XFS_MOUNT_ATTR2) || + dp->i_d.di_format == XFS_DINODE_FMT_BTREE); dp->i_afp->if_ext_max = XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); dp->i_df.if_ext_max = @@ -531,7 +507,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args) nargs.total = args->total; nargs.whichfork = XFS_ATTR_FORK; nargs.trans = args->trans; - nargs.oknoent = 1; + nargs.op_flags = XFS_DA_OP_OKNOENT; sfe = &sf->list[0]; for (i = 0; i < sf->hdr.count; i++) { @@ -555,7 +531,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args) out: if(bp) xfs_da_buf_done(bp); - kmem_free(tmpbuffer, size); + kmem_free(tmpbuffer); return(error); } @@ -624,15 +600,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) (XFS_ISRESET_CURSOR(cursor) && (dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) { for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) { - attrnames_t *namesp; - - if (!xfs_attr_namesp_match_overrides(context->flags, sfe->flags)) { - sfe = XFS_ATTR_SF_NEXTENTRY(sfe); - continue; - } - namesp = xfs_attr_flags_namesp(sfe->flags); error = context->put_listent(context, - namesp, + sfe->flags, (char *)sfe->nameval, (int)sfe->namelen, (int)sfe->valuelen, @@ -676,13 +645,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) XFS_ERRLEVEL_LOW, context->dp->i_mount, sfe); xfs_attr_trace_l_c("sf corrupted", context); - kmem_free(sbuf, sbsize); + kmem_free(sbuf); return XFS_ERROR(EFSCORRUPTED); } - if (!xfs_attr_namesp_match_overrides(context->flags, sfe->flags)) { - sfe = XFS_ATTR_SF_NEXTENTRY(sfe); - continue; - } + sbp->entno = i; sbp->hash = xfs_da_hashname((char *)sfe->nameval, sfe->namelen); sbp->name = (char *)sfe->nameval; @@ -717,7 +683,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) } } if (i == nsbuf) { - kmem_free(sbuf, sbsize); + kmem_free(sbuf); xfs_attr_trace_l_c("blk end", context); return(0); } @@ -726,16 +692,12 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) * Loop putting entries into the user buffer. */ for ( ; i < nsbuf; i++, sbp++) { - attrnames_t *namesp; - - namesp = xfs_attr_flags_namesp(sbp->flags); - if (cursor->hashval != sbp->hash) { cursor->hashval = sbp->hash; cursor->offset = 0; } error = context->put_listent(context, - namesp, + sbp->flags, sbp->name, sbp->namelen, sbp->valuelen, @@ -747,7 +709,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) cursor->offset++; } - kmem_free(sbuf, sbsize); + kmem_free(sbuf); xfs_attr_trace_l_c("sf E-O-F", context); return(0); } @@ -853,7 +815,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff) nargs.total = args->total; nargs.whichfork = XFS_ATTR_FORK; nargs.trans = args->trans; - nargs.oknoent = 1; + nargs.op_flags = XFS_DA_OP_OKNOENT; entry = &leaf->entries[0]; for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) { if (entry->flags & XFS_ATTR_INCOMPLETE) @@ -873,7 +835,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff) error = 0; out: - kmem_free(tmpbuffer, XFS_LBSIZE(dp->i_mount)); + kmem_free(tmpbuffer); return(error); } @@ -1155,7 +1117,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex) entry->hashval = cpu_to_be32(args->hashval); entry->flags = tmp ? XFS_ATTR_LOCAL : 0; entry->flags |= XFS_ATTR_NSP_ARGS_TO_ONDISK(args->flags); - if (args->rename) { + if (args->op_flags & XFS_DA_OP_RENAME) { entry->flags |= XFS_ATTR_INCOMPLETE; if ((args->blkno2 == args->blkno) && (args->index2 <= args->index)) { @@ -1271,7 +1233,7 @@ xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *bp) be16_to_cpu(hdr_s->count), mp); xfs_da_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1); - kmem_free(tmpbuffer, XFS_LBSIZE(mp)); + kmem_free(tmpbuffer); } /* @@ -1921,7 +1883,7 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk, be16_to_cpu(drop_hdr->count), mp); } memcpy((char *)save_leaf, (char *)tmp_leaf, state->blocksize); - kmem_free(tmpbuffer, state->blocksize); + kmem_free(tmpbuffer); } xfs_da_log_buf(state->args->trans, save_blk->bp, 0, @@ -2400,8 +2362,6 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context) */ retval = 0; for ( ; (i < be16_to_cpu(leaf->hdr.count)); entry++, i++) { - attrnames_t *namesp; - if (be32_to_cpu(entry->hashval) != cursor->hashval) { cursor->hashval = be32_to_cpu(entry->hashval); cursor->offset = 0; @@ -2409,17 +2369,13 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context) if (entry->flags & XFS_ATTR_INCOMPLETE) continue; /* skip incomplete entries */ - if (!xfs_attr_namesp_match_overrides(context->flags, entry->flags)) - continue; - - namesp = xfs_attr_flags_namesp(entry->flags); if (entry->flags & XFS_ATTR_LOCAL) { xfs_attr_leaf_name_local_t *name_loc = XFS_ATTR_LEAF_NAME_LOCAL(leaf, i); retval = context->put_listent(context, - namesp, + entry->flags, (char *)name_loc->nameval, (int)name_loc->namelen, be16_to_cpu(name_loc->valuelen), @@ -2446,16 +2402,15 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context) if (retval) return retval; retval = context->put_listent(context, - namesp, + entry->flags, (char *)name_rmt->name, (int)name_rmt->namelen, valuelen, (char*)args.value); - kmem_free(args.value, valuelen); - } - else { + kmem_free(args.value); + } else { retval = context->put_listent(context, - namesp, + entry->flags, (char *)name_rmt->name, (int)name_rmt->namelen, valuelen, @@ -2543,9 +2498,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args) /* * Commit the flag value change and start the next trans in series. */ - error = xfs_attr_rolltrans(&args->trans, args->dp); - - return(error); + return xfs_trans_roll(&args->trans, args->dp); } /* @@ -2592,9 +2545,7 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args) /* * Commit the flag value change and start the next trans in series. */ - error = xfs_attr_rolltrans(&args->trans, args->dp); - - return(error); + return xfs_trans_roll(&args->trans, args->dp); } /* @@ -2710,7 +2661,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args) /* * Commit the flag value change and start the next trans in series. */ - error = xfs_attr_rolltrans(&args->trans, args->dp); + error = xfs_trans_roll(&args->trans, args->dp); return(error); } @@ -2768,7 +2719,7 @@ xfs_attr_root_inactive(xfs_trans_t **trans, xfs_inode_t *dp) /* * Commit the invalidate and start the next transaction. */ - error = xfs_attr_rolltrans(trans, dp); + error = xfs_trans_roll(trans, dp); return (error); } @@ -2870,7 +2821,8 @@ xfs_attr_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp, /* * Atomically commit the whole invalidate stuff. */ - if ((error = xfs_attr_rolltrans(trans, dp))) + error = xfs_trans_roll(trans, dp); + if (error) return (error); } @@ -2954,7 +2906,7 @@ xfs_attr_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp) error = tmp; /* save only the 1st errno */ } - kmem_free((xfs_caddr_t)list, size); + kmem_free((xfs_caddr_t)list); return(error); } @@ -3009,7 +2961,8 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp, /* * Roll to next transaction. */ - if ((error = xfs_attr_rolltrans(trans, dp))) + error = xfs_trans_roll(trans, dp); + if (error) return (error); } @@ -3019,60 +2972,3 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp, return(0); } - - -/* - * Roll from one trans in the sequence of PERMANENT transactions to the next. - */ -int -xfs_attr_rolltrans(xfs_trans_t **transp, xfs_inode_t *dp) -{ - xfs_trans_t *trans; - unsigned int logres, count; - int error; - - /* - * Ensure that the inode is always logged. - */ - trans = *transp; - xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE); - - /* - * Copy the critical parameters from one trans to the next. - */ - logres = trans->t_log_res; - count = trans->t_log_count; - *transp = xfs_trans_dup(trans); - - /* - * Commit the current transaction. - * If this commit failed, then it'd just unlock those items that - * are not marked ihold. That also means that a filesystem shutdown - * is in progress. The caller takes the responsibility to cancel - * the duplicate transaction that gets returned. - */ - if ((error = xfs_trans_commit(trans, 0))) - return (error); - - trans = *transp; - - /* - * Reserve space in the log for th next transaction. - * This also pushes items in the "AIL", the list of logged items, - * out to disk if they are taking up space at the tail of the log - * that we want to use. This requires that either nothing be locked - * across this call, or that anything that is locked be logged in - * the prior and the next transactions. - */ - error = xfs_trans_reserve(trans, 0, logres, 0, - XFS_TRANS_PERM_LOG_RES, count); - /* - * Ensure that the inode is in the new transaction and locked. - */ - if (!error) { - xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL); - xfs_trans_ihold(trans, dp); - } - return (error); - -} diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h index 040f732..83e9af4 100644 --- a/fs/xfs/xfs_attr_leaf.h +++ b/fs/xfs/xfs_attr_leaf.h @@ -30,7 +30,7 @@ struct attrlist; struct attrlist_cursor_kern; -struct attrnames; +struct xfs_attr_list_context; struct xfs_dabuf; struct xfs_da_args; struct xfs_da_state; @@ -204,33 +204,6 @@ static inline int xfs_attr_leaf_entsize_local_max(int bsize) return (((bsize) >> 1) + ((bsize) >> 2)); } - -/*======================================================================== - * Structure used to pass context around among the routines. - *========================================================================*/ - - -struct xfs_attr_list_context; - -typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, struct attrnames *, - char *, int, int, char *); - -typedef struct xfs_attr_list_context { - struct xfs_inode *dp; /* inode */ - struct attrlist_cursor_kern *cursor; /* position in list */ - struct attrlist *alist; /* output buffer */ - int seen_enough; /* T/F: seen enough of list? */ - int count; /* num used entries */ - int dupcnt; /* count dup hashvals seen */ - int bufsize; /* total buffer size */ - int firstu; /* first used byte in buffer */ - int flags; /* from VOP call */ - int resynch; /* T/F: resynch with cursor */ - int put_value; /* T/F: need value for listent */ - put_listent_func_t put_listent; /* list output fmt function */ - int index; /* index into output buffer */ -} xfs_attr_list_context_t; - /* * Used to keep a list of "remote value" extents when unlinking an inode. */ @@ -301,6 +274,4 @@ int xfs_attr_leaf_order(struct xfs_dabuf *leaf1_bp, struct xfs_dabuf *leaf2_bp); int xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local); -int xfs_attr_rolltrans(struct xfs_trans **transp, struct xfs_inode *dp); - #endif /* __XFS_ATTR_LEAF_H__ */ diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h index f67f917..ea22839 100644 --- a/fs/xfs/xfs_attr_sf.h +++ b/fs/xfs/xfs_attr_sf.h @@ -97,13 +97,9 @@ void xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context, void xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context, struct xfs_attr_leafblock *leaf); void xfs_attr_trace_enter(int type, char *where, - __psunsigned_t a2, __psunsigned_t a3, - __psunsigned_t a4, __psunsigned_t a5, - __psunsigned_t a6, __psunsigned_t a7, - __psunsigned_t a8, __psunsigned_t a9, - __psunsigned_t a10, __psunsigned_t a11, - __psunsigned_t a12, __psunsigned_t a13, - __psunsigned_t a14, __psunsigned_t a15); + struct xfs_attr_list_context *context, + __psunsigned_t a13, __psunsigned_t a14, + __psunsigned_t a15); #else #define xfs_attr_trace_l_c(w,c) #define xfs_attr_trace_l_cn(w,c,n) diff --git a/fs/xfs/xfs_bit.c b/fs/xfs/xfs_bit.c index fab0b6d..4822884 100644 --- a/fs/xfs/xfs_bit.c +++ b/fs/xfs/xfs_bit.c @@ -25,109 +25,6 @@ * XFS bit manipulation routines, used in non-realtime code. */ -#ifndef HAVE_ARCH_HIGHBIT -/* - * Index of high bit number in byte, -1 for none set, 0..7 otherwise. - */ -static const char xfs_highbit[256] = { - -1, 0, 1, 1, 2, 2, 2, 2, /* 00 .. 07 */ - 3, 3, 3, 3, 3, 3, 3, 3, /* 08 .. 0f */ - 4, 4, 4, 4, 4, 4, 4, 4, /* 10 .. 17 */ - 4, 4, 4, 4, 4, 4, 4, 4, /* 18 .. 1f */ - 5, 5, 5, 5, 5, 5, 5, 5, /* 20 .. 27 */ - 5, 5, 5, 5, 5, 5, 5, 5, /* 28 .. 2f */ - 5, 5, 5, 5, 5, 5, 5, 5, /* 30 .. 37 */ - 5, 5, 5, 5, 5, 5, 5, 5, /* 38 .. 3f */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 40 .. 47 */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 48 .. 4f */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 50 .. 57 */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 58 .. 5f */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 60 .. 67 */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 68 .. 6f */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 70 .. 77 */ - 6, 6, 6, 6, 6, 6, 6, 6, /* 78 .. 7f */ - 7, 7, 7, 7, 7, 7, 7, 7, /* 80 .. 87 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* 88 .. 8f */ - 7, 7, 7, 7, 7, 7, 7, 7, /* 90 .. 97 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* 98 .. 9f */ - 7, 7, 7, 7, 7, 7, 7, 7, /* a0 .. a7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* a8 .. af */ - 7, 7, 7, 7, 7, 7, 7, 7, /* b0 .. b7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* b8 .. bf */ - 7, 7, 7, 7, 7, 7, 7, 7, /* c0 .. c7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* c8 .. cf */ - 7, 7, 7, 7, 7, 7, 7, 7, /* d0 .. d7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* d8 .. df */ - 7, 7, 7, 7, 7, 7, 7, 7, /* e0 .. e7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* e8 .. ef */ - 7, 7, 7, 7, 7, 7, 7, 7, /* f0 .. f7 */ - 7, 7, 7, 7, 7, 7, 7, 7, /* f8 .. ff */ -}; -#endif - -/* - * xfs_highbit32: get high bit set out of 32-bit argument, -1 if none set. - */ -inline int -xfs_highbit32( - __uint32_t v) -{ -#ifdef HAVE_ARCH_HIGHBIT - return highbit32(v); -#else - int i; - - if (v & 0xffff0000) - if (v & 0xff000000) - i = 24; - else - i = 16; - else if (v & 0x0000ffff) - if (v & 0x0000ff00) - i = 8; - else - i = 0; - else - return -1; - return i + xfs_highbit[(v >> i) & 0xff]; -#endif -} - -/* - * xfs_lowbit64: get low bit set out of 64-bit argument, -1 if none set. - */ -int -xfs_lowbit64( - __uint64_t v) -{ - __uint32_t w = (__uint32_t)v; - int n = 0; - - if (w) { /* lower bits */ - n = ffs(w); - } else { /* upper bits */ - w = (__uint32_t)(v >> 32); - if (w && (n = ffs(w))) - n += 32; - } - return n - 1; -} - -/* - * xfs_highbit64: get high bit set out of 64-bit argument, -1 if none set. - */ -int -xfs_highbit64( - __uint64_t v) -{ - __uint32_t h = (__uint32_t)(v >> 32); - - if (h) - return xfs_highbit32(h) + 32; - return xfs_highbit32((__uint32_t)v); -} - - /* * Return whether bitmap is empty. * Size is number of words in the bitmap, which is padded to word boundary diff --git a/fs/xfs/xfs_bit.h b/fs/xfs/xfs_bit.h index 082641a..8e0e463 100644 --- a/fs/xfs/xfs_bit.h +++ b/fs/xfs/xfs_bit.h @@ -47,13 +47,39 @@ static inline __uint64_t xfs_mask64lo(int n) } /* Get high bit set out of 32-bit argument, -1 if none set */ -extern int xfs_highbit32(__uint32_t v); +static inline int xfs_highbit32(__uint32_t v) +{ + return fls(v) - 1; +} + +/* Get high bit set out of 64-bit argument, -1 if none set */ +static inline int xfs_highbit64(__uint64_t v) +{ + return fls64(v) - 1; +} + +/* Get low bit set out of 32-bit argument, -1 if none set */ +static inline int xfs_lowbit32(__uint32_t v) +{ + unsigned long t = v; + return (v) ? find_first_bit(&t, 32) : -1; +} /* Get low bit set out of 64-bit argument, -1 if none set */ -extern int xfs_lowbit64(__uint64_t v); +static inline int xfs_lowbit64(__uint64_t v) +{ + __uint32_t w = (__uint32_t)v; + int n = 0; -/* Get high bit set out of 64-bit argument, -1 if none set */ -extern int xfs_highbit64(__uint64_t); + if (w) { /* lower bits */ + n = ffs(w); + } else { /* upper bits */ + w = (__uint32_t)(v >> 32); + if (w && (n = ffs(w))) + n += 32; + } + return n - 1; +} /* Return whether bitmap is empty (1 == empty) */ extern int xfs_bitmap_empty(uint *map, uint size); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 53c259f..a1aab92 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -384,14 +384,14 @@ xfs_bmap_count_tree( int levelin, int *count); -STATIC int +STATIC void xfs_bmap_count_leaves( xfs_ifork_t *ifp, xfs_extnum_t idx, int numrecs, int *count); -STATIC int +STATIC void xfs_bmap_disk_count_leaves( xfs_extnum_t idx, xfs_bmbt_block_t *block, @@ -428,7 +428,8 @@ xfs_bmap_add_attrfork_btree( cur->bc_private.b.firstblock = *firstblock; if ((error = xfs_bmbt_lookup_ge(cur, 0, 0, 0, &stat))) goto error0; - ASSERT(stat == 1); /* must be at least one entry */ + /* must be at least one entry */ + XFS_WANT_CORRUPTED_GOTO(stat == 1, error0); if ((error = xfs_bmbt_newroot(cur, flags, &stat))) goto error0; if (stat == 0) { @@ -816,13 +817,13 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_startblock, RIGHT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, LEFT.br_startoff, LEFT.br_startblock, LEFT.br_blockcount + @@ -860,7 +861,7 @@ xfs_bmap_add_extent_delay_real( LEFT.br_startblock, LEFT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, LEFT.br_startoff, LEFT.br_startblock, LEFT.br_blockcount + @@ -895,7 +896,7 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_startblock, RIGHT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, PREV.br_startoff, new->br_startblock, PREV.br_blockcount + @@ -928,11 +929,11 @@ xfs_bmap_add_extent_delay_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = XFS_EXT_NORM; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } *dnew = 0; /* DELTA: The in-core extent described by new changed type. */ @@ -963,7 +964,7 @@ xfs_bmap_add_extent_delay_real( LEFT.br_startblock, LEFT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, LEFT.br_startoff, LEFT.br_startblock, LEFT.br_blockcount + @@ -1004,11 +1005,11 @@ xfs_bmap_add_extent_delay_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = XFS_EXT_NORM; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && ip->i_d.di_nextents > ip->i_df.if_ext_max) { @@ -1054,7 +1055,7 @@ xfs_bmap_add_extent_delay_real( RIGHT.br_startblock, RIGHT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, new->br_startoff, new->br_startblock, new->br_blockcount + @@ -1094,11 +1095,11 @@ xfs_bmap_add_extent_delay_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = XFS_EXT_NORM; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && ip->i_d.di_nextents > ip->i_df.if_ext_max) { @@ -1149,11 +1150,11 @@ xfs_bmap_add_extent_delay_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = XFS_EXT_NORM; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS && ip->i_d.di_nextents > ip->i_df.if_ext_max) { @@ -1377,19 +1378,19 @@ xfs_bmap_add_extent_unwritten_real( RIGHT.br_startblock, RIGHT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, LEFT.br_startoff, LEFT.br_startblock, LEFT.br_blockcount + PREV.br_blockcount + @@ -1426,13 +1427,13 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, LEFT.br_startoff, LEFT.br_startblock, LEFT.br_blockcount + PREV.br_blockcount, @@ -1469,13 +1470,13 @@ xfs_bmap_add_extent_unwritten_real( RIGHT.br_startblock, RIGHT.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, new->br_startoff, new->br_startblock, new->br_blockcount + RIGHT.br_blockcount, @@ -1508,7 +1509,7 @@ xfs_bmap_add_extent_unwritten_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, new->br_startoff, new->br_startblock, new->br_blockcount, newext))) @@ -1549,7 +1550,7 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, PREV.br_startoff + new->br_blockcount, PREV.br_startblock + new->br_blockcount, @@ -1596,7 +1597,7 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, PREV.br_startoff + new->br_blockcount, PREV.br_startblock + new->br_blockcount, @@ -1606,7 +1607,7 @@ xfs_bmap_add_extent_unwritten_real( cur->bc_rec.b = *new; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } /* DELTA: One in-core extent is split in two. */ temp = PREV.br_startoff; @@ -1640,7 +1641,7 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, PREV.br_startoff, PREV.br_startblock, PREV.br_blockcount - new->br_blockcount, @@ -1682,7 +1683,7 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, PREV.br_startoff, PREV.br_startblock, PREV.br_blockcount - new->br_blockcount, @@ -1692,11 +1693,11 @@ xfs_bmap_add_extent_unwritten_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = XFS_EXT_NORM; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } /* DELTA: One in-core extent is split in two. */ temp = PREV.br_startoff; @@ -1732,27 +1733,34 @@ xfs_bmap_add_extent_unwritten_real( PREV.br_startblock, PREV.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); /* new right extent - oldext */ if ((error = xfs_bmbt_update(cur, r[1].br_startoff, r[1].br_startblock, r[1].br_blockcount, r[1].br_state))) goto done; /* new left extent - oldext */ - PREV.br_blockcount = - new->br_startoff - PREV.br_startoff; cur->bc_rec.b = PREV; + cur->bc_rec.b.br_blockcount = + new->br_startoff - PREV.br_startoff; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); - if ((error = xfs_bmbt_increment(cur, 0, &i))) + XFS_WANT_CORRUPTED_GOTO(i == 1, done); + /* + * Reset the cursor to the position of the new extent + * we are about to insert as we can't trust it after + * the previous insert. + */ + if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff, + new->br_startblock, new->br_blockcount, + &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); /* new middle extent - newext */ - cur->bc_rec.b = *new; + cur->bc_rec.b.br_state = new->br_state; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } /* DELTA: One in-core extent is split in three. */ temp = PREV.br_startoff; @@ -2097,13 +2105,13 @@ xfs_bmap_add_extent_hole_real( right.br_startblock, right.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_decrement(cur, 0, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, left.br_startoff, left.br_startblock, left.br_blockcount + @@ -2139,7 +2147,7 @@ xfs_bmap_add_extent_hole_real( left.br_startblock, left.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, left.br_startoff, left.br_startblock, left.br_blockcount + @@ -2174,7 +2182,7 @@ xfs_bmap_add_extent_hole_real( right.br_startblock, right.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); if ((error = xfs_bmbt_update(cur, new->br_startoff, new->br_startblock, new->br_blockcount + @@ -2208,11 +2216,11 @@ xfs_bmap_add_extent_hole_real( new->br_startblock, new->br_blockcount, &i))) goto done; - ASSERT(i == 0); + XFS_WANT_CORRUPTED_GOTO(i == 0, done); cur->bc_rec.b.br_state = new->br_state; if ((error = xfs_bmbt_insert(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } /* DELTA: A new extent was added in a hole. */ temp = new->br_startoff; @@ -3131,7 +3139,7 @@ xfs_bmap_del_extent( got.br_startblock, got.br_blockcount, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } da_old = da_new = 0; } else { @@ -3164,7 +3172,7 @@ xfs_bmap_del_extent( } if ((error = xfs_bmbt_delete(cur, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); break; case 2: @@ -3268,7 +3276,7 @@ xfs_bmap_del_extent( got.br_startblock, temp, &i))) goto done; - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); /* * Update the btree record back * to the original value. @@ -3289,7 +3297,7 @@ xfs_bmap_del_extent( error = XFS_ERROR(ENOSPC); goto done; } - ASSERT(i == 1); + XFS_WANT_CORRUPTED_GOTO(i == 1, done); } else flags |= XFS_ILOG_FEXT(whichfork); XFS_IFORK_NEXT_SET(ip, whichfork, @@ -3992,7 +4000,7 @@ xfs_bmap_add_attrfork( ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS; } ASSERT(ip->i_d.di_anextents == 0); - VN_HOLD(XFS_ITOV(ip)); + IHOLD(ip); xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); switch (ip->i_d.di_format) { @@ -5970,7 +5978,7 @@ unlock_and_return: xfs_iunlock_map_shared(ip, lock); xfs_iunlock(ip, XFS_IOLOCK_SHARED); - kmem_free(map, subnex * sizeof(*map)); + kmem_free(map); return error; } @@ -6088,7 +6096,7 @@ xfs_bmap_get_bp( tp = cur->bc_tp; licp = &tp->t_items; while (!bp && licp != NULL) { - if (XFS_LIC_ARE_ALL_FREE(licp)) { + if (xfs_lic_are_all_free(licp)) { licp = licp->lic_next; continue; } @@ -6098,11 +6106,11 @@ xfs_bmap_get_bp( xfs_buf_log_item_t *bip; xfs_buf_t *lbp; - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } - lidp = XFS_LIC_SLOT(licp, i); + lidp = xfs_lic_slot(licp, i); lip = lidp->lid_item; if (lip->li_type != XFS_LI_BUF) continue; @@ -6359,13 +6367,9 @@ xfs_bmap_count_blocks( mp = ip->i_mount; ifp = XFS_IFORK_PTR(ip, whichfork); if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) { - if (unlikely(xfs_bmap_count_leaves(ifp, 0, + xfs_bmap_count_leaves(ifp, 0, ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t), - count) < 0)) { - XFS_ERROR_REPORT("xfs_bmap_count_blocks(1)", - XFS_ERRLEVEL_LOW, mp); - return XFS_ERROR(EFSCORRUPTED); - } + count); return 0; } @@ -6446,13 +6450,7 @@ xfs_bmap_count_tree( for (;;) { nextbno = be64_to_cpu(block->bb_rightsib); numrecs = be16_to_cpu(block->bb_numrecs); - if (unlikely(xfs_bmap_disk_count_leaves(0, - block, numrecs, count) < 0)) { - xfs_trans_brelse(tp, bp); - XFS_ERROR_REPORT("xfs_bmap_count_tree(2)", - XFS_ERRLEVEL_LOW, mp); - return XFS_ERROR(EFSCORRUPTED); - } + xfs_bmap_disk_count_leaves(0, block, numrecs, count); xfs_trans_brelse(tp, bp); if (nextbno == NULLFSBLOCK) break; @@ -6470,7 +6468,7 @@ xfs_bmap_count_tree( /* * Count leaf blocks given a range of extent records. */ -STATIC int +STATIC void xfs_bmap_count_leaves( xfs_ifork_t *ifp, xfs_extnum_t idx, @@ -6483,14 +6481,13 @@ xfs_bmap_count_leaves( xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b); *count += xfs_bmbt_get_blockcount(frp); } - return 0; } /* * Count leaf blocks given a range of extent records originally * in btree format. */ -STATIC int +STATIC void xfs_bmap_disk_count_leaves( xfs_extnum_t idx, xfs_bmbt_block_t *block, @@ -6504,5 +6501,4 @@ xfs_bmap_disk_count_leaves( frp = XFS_BTREE_REC_ADDR(xfs_bmbt, block, idx + b); *count += xfs_bmbt_disk_get_blockcount(frp); } - return 0; } diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 6ff70cd..9f3e3a8 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -54,12 +54,23 @@ typedef struct xfs_bmap_free_item /* * Header for free extent list. + * + * xbf_low is used by the allocator to activate the lowspace algorithm - + * when free space is running low the extent allocator may choose to + * allocate an extent from an AG without leaving sufficient space for + * a btree split when inserting the new extent. In this case the allocator + * will enable the lowspace algorithm which is supposed to allow further + * allocations (such as btree splits and newroots) to allocate from + * sequential AGs. In order to avoid locking AGs out of order the lowspace + * algorithm will start searching for free space from AG 0. If the correct + * transaction reservations have been made then this algorithm will eventually + * find all the space it needs. */ typedef struct xfs_bmap_free { xfs_bmap_free_item_t *xbf_first; /* list of to-be-free extents */ int xbf_count; /* count of items on list */ - int xbf_low; /* kludge: alloc in low mode */ + int xbf_low; /* alloc in low mode */ } xfs_bmap_free_t; #define XFS_BMAP_MAX_NMAP 4 diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 4f0e849..23efad2 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -1493,12 +1493,27 @@ xfs_bmbt_split( left = XFS_BUF_TO_BMBT_BLOCK(lbp); args.fsbno = cur->bc_private.b.firstblock; args.firstblock = args.fsbno; + args.minleft = 0; if (args.fsbno == NULLFSBLOCK) { args.fsbno = lbno; args.type = XFS_ALLOCTYPE_START_BNO; - } else + /* + * Make sure there is sufficient room left in the AG to + * complete a full tree split for an extent insert. If + * we are converting the middle part of an extent then + * we may need space for two tree splits. + * + * We are relying on the caller to make the correct block + * reservation for this operation to succeed. If the + * reservation amount is insufficient then we may fail a + * block allocation here and corrupt the filesystem. + */ + args.minleft = xfs_trans_get_block_res(args.tp); + } else if (cur->bc_private.b.flist->xbf_low) + args.type = XFS_ALLOCTYPE_START_BNO; + else args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.mod = args.minleft = args.alignment = args.total = args.isfl = + args.mod = args.alignment = args.total = args.isfl = args.userdata = args.minalignslop = 0; args.minlen = args.maxlen = args.prod = 1; args.wasdel = cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL; @@ -1510,6 +1525,21 @@ xfs_bmbt_split( XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } + if (args.fsbno == NULLFSBLOCK && args.minleft) { + /* + * Could not find an AG with enough free space to satisfy + * a full btree split. Try again without minleft and if + * successful activate the lowspace algorithm. + */ + args.fsbno = 0; + args.type = XFS_ALLOCTYPE_FIRST_AG; + args.minleft = 0; + if ((error = xfs_alloc_vextent(&args))) { + XFS_BMBT_TRACE_CURSOR(cur, ERROR); + return error; + } + cur->bc_private.b.flist->xbf_low = 1; + } if (args.fsbno == NULLFSBLOCK) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; @@ -2029,22 +2059,8 @@ xfs_bmbt_increment( * Insert the current record at the point referenced by cur. * * A multi-level split of the tree on insert will invalidate the original - * cursor. It appears, however, that some callers assume that the cursor is - * always valid. Hence if we do a multi-level split we need to revalidate the - * cursor. - * - * When a split occurs, we will see a new cursor returned. Use that as a - * trigger to determine if we need to revalidate the original cursor. If we get - * a split, then use the original irec to lookup up the path of the record we - * just inserted. - * - * Note that the fact that the btree root is in the inode means that we can - * have the level of the tree change without a "split" occurring at the root - * level. What happens is that the root is migrated to an allocated block and - * the inode root is pointed to it. This means a single split can change the - * level of the tree (level 2 -> level 3) and invalidate the old cursor. Hence - * the level change should be accounted as a split so as to correctly trigger a - * revalidation of the old cursor. + * cursor. All callers of this function should assume that the cursor is + * no longer valid and revalidate it. */ int /* error */ xfs_bmbt_insert( @@ -2057,14 +2073,11 @@ xfs_bmbt_insert( xfs_fsblock_t nbno; xfs_btree_cur_t *ncur; xfs_bmbt_rec_t nrec; - xfs_bmbt_irec_t oirec; /* original irec */ xfs_btree_cur_t *pcur; - int splits = 0; XFS_BMBT_TRACE_CURSOR(cur, ENTRY); level = 0; nbno = NULLFSBLOCK; - oirec = cur->bc_rec.b; xfs_bmbt_disk_set_all(&nrec, &cur->bc_rec.b); ncur = NULL; pcur = cur; @@ -2073,13 +2086,11 @@ xfs_bmbt_insert( &i))) { if (pcur != cur) xfs_btree_del_cursor(pcur, XFS_BTREE_ERROR); - goto error0; + XFS_BMBT_TRACE_CURSOR(cur, ERROR); + return error; } XFS_WANT_CORRUPTED_GOTO(i == 1, error0); if (pcur != cur && (ncur || nbno == NULLFSBLOCK)) { - /* allocating a new root is effectively a split */ - if (cur->bc_nlevels != pcur->bc_nlevels) - splits++; cur->bc_nlevels = pcur->bc_nlevels; cur->bc_private.b.allocated += pcur->bc_private.b.allocated; @@ -2093,21 +2104,10 @@ xfs_bmbt_insert( xfs_btree_del_cursor(pcur, XFS_BTREE_NOERROR); } if (ncur) { - splits++; pcur = ncur; ncur = NULL; } } while (nbno != NULLFSBLOCK); - - if (splits > 1) { - /* revalidate the old cursor as we had a multi-level split */ - error = xfs_bmbt_lookup_eq(cur, oirec.br_startoff, - oirec.br_startblock, oirec.br_blockcount, &i); - if (error) - goto error0; - ASSERT(i == 1); - } - XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = i; return 0; @@ -2254,7 +2254,9 @@ xfs_bmbt_newroot( #endif args.fsbno = be64_to_cpu(*pp); args.type = XFS_ALLOCTYPE_START_BNO; - } else + } else if (cur->bc_private.b.flist->xbf_low) + args.type = XFS_ALLOCTYPE_START_BNO; + else args.type = XFS_ALLOCTYPE_NEAR_BNO; if ((error = xfs_alloc_vextent(&args))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index aeb87ca..cc593a8 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -46,38 +46,11 @@ kmem_zone_t *xfs_btree_cur_zone; /* * Btree magic numbers. */ -const __uint32_t xfs_magics[XFS_BTNUM_MAX] = -{ +const __uint32_t xfs_magics[XFS_BTNUM_MAX] = { XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC }; /* - * Prototypes for internal routines. - */ - -/* - * Checking routine: return maxrecs for the block. - */ -STATIC int /* number of records fitting in block */ -xfs_btree_maxrecs( - xfs_btree_cur_t *cur, /* btree cursor */ - xfs_btree_block_t *block);/* generic btree block pointer */ - -/* - * Internal routines. - */ - -/* - * Retrieve the block pointer from the cursor at the given level. - * This may be a bmap btree root or from a buffer. - */ -STATIC xfs_btree_block_t * /* generic btree block pointer */ -xfs_btree_get_block( - xfs_btree_cur_t *cur, /* btree cursor */ - int level, /* level in btree */ - struct xfs_buf **bpp); /* buffer containing the block */ - -/* * Checking routine: return maxrecs for the block. */ STATIC int /* number of records fitting in block */ @@ -457,35 +430,6 @@ xfs_btree_dup_cursor( } /* - * Change the cursor to point to the first record at the given level. - * Other levels are unaffected. - */ -int /* success=1, failure=0 */ -xfs_btree_firstrec( - xfs_btree_cur_t *cur, /* btree cursor */ - int level) /* level to change */ -{ - xfs_btree_block_t *block; /* generic btree block pointer */ - xfs_buf_t *bp; /* buffer containing block */ - - /* - * Get the block pointer for this level. - */ - block = xfs_btree_get_block(cur, level, &bp); - xfs_btree_check_block(cur, block, level, bp); - /* - * It's empty, there is no such record. - */ - if (!block->bb_h.bb_numrecs) - return 0; - /* - * Set the ptr value to 1, that's the first record/key. - */ - cur->bc_ptrs[level] = 1; - return 1; -} - -/* * Retrieve the block pointer from the cursor at the given level. * This may be a bmap btree root or from a buffer. */ @@ -626,6 +570,13 @@ xfs_btree_init_cursor( cur->bc_private.a.agbp = agbp; cur->bc_private.a.agno = agno; break; + case XFS_BTNUM_INO: + /* + * Inode allocation btree fields. + */ + cur->bc_private.a.agbp = agbp; + cur->bc_private.a.agno = agno; + break; case XFS_BTNUM_BMAP: /* * Bmap btree fields. @@ -638,13 +589,6 @@ xfs_btree_init_cursor( cur->bc_private.b.flags = 0; cur->bc_private.b.whichfork = whichfork; break; - case XFS_BTNUM_INO: - /* - * Inode allocation btree fields. - */ - cur->bc_private.i.agbp = agbp; - cur->bc_private.i.agno = agno; - break; default: ASSERT(0); } @@ -671,6 +615,35 @@ xfs_btree_islastblock( } /* + * Change the cursor to point to the first record at the given level. + * Other levels are unaffected. + */ +int /* success=1, failure=0 */ +xfs_btree_firstrec( + xfs_btree_cur_t *cur, /* btree cursor */ + int level) /* level to change */ +{ + xfs_btree_block_t *block; /* generic btree block pointer */ + xfs_buf_t *bp; /* buffer containing block */ + + /* + * Get the block pointer for this level. + */ + block = xfs_btree_get_block(cur, level, &bp); + xfs_btree_check_block(cur, block, level, bp); + /* + * It's empty, there is no such record. + */ + if (!block->bb_h.bb_numrecs) + return 0; + /* + * Set the ptr value to 1, that's the first record/key. + */ + cur->bc_ptrs[level] = 1; + return 1; +} + +/* * Change the cursor to point to the last record in the current block * at the given level. Other levels are unaffected. */ @@ -890,12 +863,12 @@ xfs_btree_readahead_core( case XFS_BTNUM_INO: i = XFS_BUF_TO_INOBT_BLOCK(cur->bc_bufs[lev]); if ((lr & XFS_BTCUR_LEFTRA) && be32_to_cpu(i->bb_leftsib) != NULLAGBLOCK) { - xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno, + xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, be32_to_cpu(i->bb_leftsib), 1); rval++; } if ((lr & XFS_BTCUR_RIGHTRA) && be32_to_cpu(i->bb_rightsib) != NULLAGBLOCK) { - xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno, + xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, be32_to_cpu(i->bb_rightsib), 1); rval++; } diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 7440b78..1f528a2 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -158,8 +158,8 @@ typedef struct xfs_btree_cur __uint8_t bc_blocklog; /* log2(blocksize) of btree blocks */ xfs_btnum_t bc_btnum; /* identifies which btree type */ union { - struct { /* needed for BNO, CNT */ - struct xfs_buf *agbp; /* agf buffer pointer */ + struct { /* needed for BNO, CNT, INO */ + struct xfs_buf *agbp; /* agf/agi buffer pointer */ xfs_agnumber_t agno; /* ag number */ } a; struct { /* needed for BMAP */ @@ -172,10 +172,6 @@ typedef struct xfs_btree_cur char flags; /* flags */ #define XFS_BTCUR_BPRV_WASDEL 1 /* was delayed */ } b; - struct { /* needed for INO */ - struct xfs_buf *agbp; /* agi buffer pointer */ - xfs_agnumber_t agno; /* ag number */ - } i; } bc_private; /* per-btree type data */ } xfs_btree_cur_t; diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 53a71c6..002fc26 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -732,12 +732,13 @@ xfs_buf_item_init( bip->bli_item.li_ops = &xfs_buf_item_ops; bip->bli_item.li_mountp = mp; bip->bli_buf = bp; + xfs_buf_hold(bp); bip->bli_format.blf_type = XFS_LI_BUF; bip->bli_format.blf_blkno = (__int64_t)XFS_BUF_ADDR(bp); bip->bli_format.blf_len = (ushort)BTOBB(XFS_BUF_COUNT(bp)); bip->bli_format.blf_map_size = map_size; #ifdef XFS_BLI_TRACE - bip->bli_trace = ktrace_alloc(XFS_BLI_TRACE_SIZE, KM_SLEEP); + bip->bli_trace = ktrace_alloc(XFS_BLI_TRACE_SIZE, KM_NOFS); #endif #ifdef XFS_TRANS_DEBUG @@ -867,6 +868,21 @@ xfs_buf_item_dirty( return (bip->bli_flags & XFS_BLI_DIRTY); } +STATIC void +xfs_buf_item_free( + xfs_buf_log_item_t *bip) +{ +#ifdef XFS_TRANS_DEBUG + kmem_free(bip->bli_orig); + kmem_free(bip->bli_logged); +#endif /* XFS_TRANS_DEBUG */ + +#ifdef XFS_BLI_TRACE + ktrace_free(bip->bli_trace); +#endif + kmem_zone_free(xfs_buf_item_zone, bip); +} + /* * This is called when the buf log item is no longer needed. It should * free the buf log item associated with the given buffer and clear @@ -887,18 +903,8 @@ xfs_buf_item_relse( (XFS_BUF_IODONE_FUNC(bp) != NULL)) { XFS_BUF_CLR_IODONE_FUNC(bp); } - -#ifdef XFS_TRANS_DEBUG - kmem_free(bip->bli_orig, XFS_BUF_COUNT(bp)); - bip->bli_orig = NULL; - kmem_free(bip->bli_logged, XFS_BUF_COUNT(bp) / NBBY); - bip->bli_logged = NULL; -#endif /* XFS_TRANS_DEBUG */ - -#ifdef XFS_BLI_TRACE - ktrace_free(bip->bli_trace); -#endif - kmem_zone_free(xfs_buf_item_zone, bip); + xfs_buf_rele(bp); + xfs_buf_item_free(bip); } @@ -1056,7 +1062,7 @@ xfs_buf_iodone_callbacks( anyway. */ XFS_BUF_SET_BRELSE_FUNC(bp,xfs_buf_error_relse); XFS_BUF_DONE(bp); - XFS_BUF_V_IODONESEMA(bp); + XFS_BUF_FINISH_IOWAIT(bp); } return; } @@ -1120,6 +1126,7 @@ xfs_buf_iodone( ASSERT(bip->bli_buf == bp); + xfs_buf_rele(bp); mp = bip->bli_item.li_mountp; /* @@ -1136,18 +1143,7 @@ xfs_buf_iodone( * xfs_trans_delete_ail() drops the AIL lock. */ xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip); - -#ifdef XFS_TRANS_DEBUG - kmem_free(bip->bli_orig, XFS_BUF_COUNT(bp)); - bip->bli_orig = NULL; - kmem_free(bip->bli_logged, XFS_BUF_COUNT(bp) / NBBY); - bip->bli_logged = NULL; -#endif /* XFS_TRANS_DEBUG */ - -#ifdef XFS_BLI_TRACE - ktrace_free(bip->bli_trace); -#endif - kmem_zone_free(xfs_buf_item_zone, bip); + xfs_buf_item_free(bip); } #if defined(XFS_BLI_TRACE) diff --git a/fs/xfs/xfs_clnt.h b/fs/xfs/xfs_clnt.h index d5d1e60..d2ce5dd 100644 --- a/fs/xfs/xfs_clnt.h +++ b/fs/xfs/xfs_clnt.h @@ -78,6 +78,7 @@ struct xfs_mount_args { #define XFSMNT_IOSIZE 0x00002000 /* optimize for I/O size */ #define XFSMNT_OSYNCISOSYNC 0x00004000 /* o_sync is REALLY o_sync */ /* (osyncisdsync is default) */ +#define XFSMNT_NOATTR2 0x00008000 /* turn off ATTR2 EA format */ #define XFSMNT_32BITINODES 0x00200000 /* restrict inodes to 32 * bits of address space */ #define XFSMNT_GQUOTA 0x00400000 /* group quota accounting */ diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 021a8f7..9e561a9 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -1431,7 +1431,7 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path, } if (level < 0) { *result = XFS_ERROR(ENOENT); /* we're out of our tree */ - ASSERT(args->oknoent); + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); return(0); } @@ -1530,6 +1530,28 @@ xfs_da_hashname(const uchar_t *name, int namelen) } } +enum xfs_dacmp +xfs_da_compname( + struct xfs_da_args *args, + const char *name, + int len) +{ + return (args->namelen == len && memcmp(args->name, name, len) == 0) ? + XFS_CMP_EXACT : XFS_CMP_DIFFERENT; +} + +static xfs_dahash_t +xfs_default_hashname( + struct xfs_name *name) +{ + return xfs_da_hashname(name->name, name->len); +} + +const struct xfs_nameops xfs_default_nameops = { + .hashname = xfs_default_hashname, + .compname = xfs_da_compname +}; + /* * Add a block to the btree ahead of the file. * Return the new block number to the caller. @@ -1598,7 +1620,7 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno) args->firstblock, args->total, &mapp[mapi], &nmap, args->flist, NULL))) { - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); return error; } if (nmap < 1) @@ -1620,11 +1642,11 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno) mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount != bno + count) { if (mapp != &map) - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); return XFS_ERROR(ENOSPC); } if (mapp != &map) - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); *new_blkno = (xfs_dablk_t)bno; return 0; } @@ -2090,10 +2112,10 @@ xfs_da_do_buf( } } if (bplist) { - kmem_free(bplist, sizeof(*bplist) * nmap); + kmem_free(bplist); } if (mapp != &map) { - kmem_free(mapp, sizeof(*mapp) * nfsb); + kmem_free(mapp); } if (bpp) *bpp = rbp; @@ -2102,11 +2124,11 @@ exit1: if (bplist) { for (i = 0; i < nbplist; i++) xfs_trans_brelse(trans, bplist[i]); - kmem_free(bplist, sizeof(*bplist) * nmap); + kmem_free(bplist); } exit0: if (mapp != &map) - kmem_free(mapp, sizeof(*mapp) * nfsb); + kmem_free(mapp); if (bpp) *bpp = NULL; return error; @@ -2218,7 +2240,7 @@ xfs_da_state_free(xfs_da_state_t *state) #ifdef XFS_DABUF_DEBUG xfs_dabuf_t *xfs_dabuf_global_list; -spinlock_t xfs_dabuf_global_lock; +static DEFINE_SPINLOCK(xfs_dabuf_global_lock); #endif /* @@ -2315,7 +2337,7 @@ xfs_da_buf_done(xfs_dabuf_t *dabuf) if (dabuf->dirty) xfs_da_buf_clean(dabuf); if (dabuf->nbuf > 1) - kmem_free(dabuf->data, BBTOB(dabuf->bbcount)); + kmem_free(dabuf->data); #ifdef XFS_DABUF_DEBUG { spin_lock(&xfs_dabuf_global_lock); @@ -2332,7 +2354,7 @@ xfs_da_buf_done(xfs_dabuf_t *dabuf) if (dabuf->nbuf == 1) kmem_zone_free(xfs_dabuf_zone, dabuf); else - kmem_free(dabuf, XFS_DA_BUF_SIZE(dabuf->nbuf)); + kmem_free(dabuf); } /* @@ -2403,7 +2425,7 @@ xfs_da_brelse(xfs_trans_t *tp, xfs_dabuf_t *dabuf) for (i = 0; i < nbuf; i++) xfs_trans_brelse(tp, bplist[i]); if (bplist != &bp) - kmem_free(bplist, nbuf * sizeof(*bplist)); + kmem_free(bplist); } /* @@ -2429,7 +2451,7 @@ xfs_da_binval(xfs_trans_t *tp, xfs_dabuf_t *dabuf) for (i = 0; i < nbuf; i++) xfs_trans_binval(tp, bplist[i]); if (bplist != &bp) - kmem_free(bplist, nbuf * sizeof(*bplist)); + kmem_free(bplist); } /* diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h index 7facf86..8be0b00 100644 --- a/fs/xfs/xfs_da_btree.h +++ b/fs/xfs/xfs_da_btree.h @@ -99,6 +99,15 @@ typedef struct xfs_da_node_entry xfs_da_node_entry_t; *========================================================================*/ /* + * Search comparison results + */ +enum xfs_dacmp { + XFS_CMP_DIFFERENT, /* names are completely different */ + XFS_CMP_EXACT, /* names are exactly the same */ + XFS_CMP_CASE /* names are same but differ in case */ +}; + +/* * Structure to ease passing around component names. */ typedef struct xfs_da_args { @@ -123,13 +132,20 @@ typedef struct xfs_da_args { int index2; /* index of 2nd attr in blk */ xfs_dablk_t rmtblkno2; /* remote attr value starting blkno */ int rmtblkcnt2; /* remote attr value block count */ - unsigned char justcheck; /* T/F: check for ok with no space */ - unsigned char rename; /* T/F: this is an atomic rename op */ - unsigned char addname; /* T/F: this is an add operation */ - unsigned char oknoent; /* T/F: ok to return ENOENT, else die */ + int op_flags; /* operation flags */ + enum xfs_dacmp cmpresult; /* name compare result for lookups */ } xfs_da_args_t; /* + * Operation flags: + */ +#define XFS_DA_OP_JUSTCHECK 0x0001 /* check for ok with no space */ +#define XFS_DA_OP_RENAME 0x0002 /* this is an atomic rename op */ +#define XFS_DA_OP_ADDNAME 0x0004 /* this is an add operation */ +#define XFS_DA_OP_OKNOENT 0x0008 /* lookup/add op, ENOENT ok, else die */ +#define XFS_DA_OP_CILOOKUP 0x0010 /* lookup to return CI name if found */ + +/* * Structure to describe buffer(s) for a block. * This is needed in the directory version 2 format case, when * multiple non-contiguous fsblocks might be needed to cover one @@ -201,6 +217,14 @@ typedef struct xfs_da_state { (uint)(XFS_DA_LOGOFF(BASE, ADDR)), \ (uint)(XFS_DA_LOGOFF(BASE, ADDR)+(SIZE)-1) +/* + * Name ops for directory and/or attr name operations + */ +struct xfs_nameops { + xfs_dahash_t (*hashname)(struct xfs_name *); + enum xfs_dacmp (*compname)(struct xfs_da_args *, const char *, int); +}; + #ifdef __KERNEL__ /*======================================================================== @@ -249,6 +273,10 @@ int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno, xfs_dabuf_t *dead_buf); uint xfs_da_hashname(const uchar_t *name_string, int name_length); +enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args, + const char *name, int len); + + xfs_da_state_t *xfs_da_state_alloc(void); void xfs_da_state_free(xfs_da_state_t *state); diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 5f3647c..75b0cd4 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -116,7 +116,7 @@ xfs_swapext( out_put_file: fput(file); out_free_sxp: - kmem_free(sxp, sizeof(xfs_swapext_t)); + kmem_free(sxp); out: return error; } @@ -128,10 +128,8 @@ xfs_swap_extents( xfs_swapext_t *sxp) { xfs_mount_t *mp; - xfs_inode_t *ips[2]; xfs_trans_t *tp; xfs_bstat_t *sbp = &sxp->sx_stat; - bhv_vnode_t *vp, *tvp; xfs_ifork_t *tempifp, *ifp, *tifp; int ilf_fields, tilf_fields; static uint lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL; @@ -150,19 +148,15 @@ xfs_swap_extents( } sbp = &sxp->sx_stat; - vp = XFS_ITOV(ip); - tvp = XFS_ITOV(tip); - - /* Lock in i_ino order */ - if (ip->i_ino < tip->i_ino) { - ips[0] = ip; - ips[1] = tip; - } else { - ips[0] = tip; - ips[1] = ip; - } - xfs_lock_inodes(ips, 2, lock_flags); + /* + * 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 + * report false positives when we drop the ILOCK and regain them + * below. + */ + xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL); + xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); locked = 1; /* Verify that both files have the same format */ @@ -184,7 +178,7 @@ xfs_swap_extents( goto error0; } - if (VN_CACHED(tvp) != 0) { + if (VN_CACHED(VFS_I(tip)) != 0) { xfs_inval_cached_trace(tip, 0, -1, 0, -1); error = xfs_flushinval_pages(tip, 0, -1, FI_REMAPF_LOCKED); @@ -193,7 +187,7 @@ xfs_swap_extents( } /* Verify O_DIRECT for ftmp */ - if (VN_CACHED(tvp) != 0) { + if (VN_CACHED(VFS_I(tip)) != 0) { error = XFS_ERROR(EINVAL); goto error0; } @@ -237,7 +231,7 @@ xfs_swap_extents( * vop_read (or write in the case of autogrow) they block on the iolock * until we have switched the extents. */ - if (VN_MAPPED(vp)) { + if (VN_MAPPED(VFS_I(ip))) { error = XFS_ERROR(EBUSY); goto error0; } @@ -265,7 +259,7 @@ xfs_swap_extents( locked = 0; goto error0; } - xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); + xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL); /* * Count the number of extended attribute blocks @@ -350,15 +344,11 @@ xfs_swap_extents( break; } - /* - * Increment vnode ref counts since xfs_trans_commit & - * xfs_trans_cancel will both unlock the inodes and - * decrement the associated ref counts. - */ - VN_HOLD(vp); - VN_HOLD(tvp); + IHOLD(ip); xfs_trans_ijoin(tp, ip, lock_flags); + + IHOLD(tip); xfs_trans_ijoin(tp, tip, lock_flags); xfs_trans_log_inode(tp, ip, ilf_fields); @@ -381,6 +371,6 @@ xfs_swap_extents( xfs_iunlock(tip, lock_flags); } if (tempifp != NULL) - kmem_free(tempifp, sizeof(xfs_ifork_t)); + kmem_free(tempifp); return error; } diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index 7cb2652..80e0dc5 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -46,6 +46,54 @@ struct xfs_name xfs_name_dotdot = {"..", 2}; +extern const struct xfs_nameops xfs_default_nameops; + +/* + * ASCII case-insensitive (ie. A-Z) support for directories that was + * used in IRIX. + */ +STATIC xfs_dahash_t +xfs_ascii_ci_hashname( + struct xfs_name *name) +{ + xfs_dahash_t hash; + int i; + + for (i = 0, hash = 0; i < name->len; i++) + hash = tolower(name->name[i]) ^ rol32(hash, 7); + + return hash; +} + +STATIC enum xfs_dacmp +xfs_ascii_ci_compname( + struct xfs_da_args *args, + const char *name, + int len) +{ + enum xfs_dacmp result; + int i; + + if (args->namelen != len) + return XFS_CMP_DIFFERENT; + + result = XFS_CMP_EXACT; + for (i = 0; i < len; i++) { + if (args->name[i] == name[i]) + continue; + if (tolower(args->name[i]) != tolower(name[i])) + return XFS_CMP_DIFFERENT; + result = XFS_CMP_CASE; + } + + return result; +} + +static struct xfs_nameops xfs_ascii_ci_nameops = { + .hashname = xfs_ascii_ci_hashname, + .compname = xfs_ascii_ci_compname, +}; + void xfs_dir_mount( xfs_mount_t *mp) @@ -65,6 +113,10 @@ xfs_dir_mount( (mp->m_dirblksize - (uint)sizeof(xfs_da_node_hdr_t)) / (uint)sizeof(xfs_da_node_entry_t); mp->m_dir_magicpct = (mp->m_dirblksize * 37) / 100; + if (xfs_sb_version_hasasciici(&mp->m_sb)) + mp->m_dirnameops = &xfs_ascii_ci_nameops; + else + mp->m_dirnameops = &xfs_default_nameops; } /* @@ -162,9 +214,10 @@ xfs_dir_createname( return rval; XFS_STATS_INC(xs_dir_create); + memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; - args.hashval = xfs_da_hashname(name->name, name->len); + args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = inum; args.dp = dp; args.firstblock = first; @@ -172,8 +225,7 @@ xfs_dir_createname( args.total = total; args.whichfork = XFS_DATA_FORK; args.trans = tp; - args.justcheck = 0; - args.addname = args.oknoent = 1; + args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_addname(&args); @@ -191,14 +243,43 @@ xfs_dir_createname( } /* + * If doing a CI lookup and case-insensitive match, dup actual name into + * args.value. Return EEXIST for success (ie. name found) or an error. + */ +int +xfs_dir_cilookup_result( + struct xfs_da_args *args, + const char *name, + int len) +{ + if (args->cmpresult == XFS_CMP_DIFFERENT) + return ENOENT; + if (args->cmpresult != XFS_CMP_CASE || + !(args->op_flags & XFS_DA_OP_CILOOKUP)) + return EEXIST; + + args->value = kmem_alloc(len, KM_MAYFAIL); + if (!args->value) + return ENOMEM; + + memcpy(args->value, name, len); + args->valuelen = len; + return EEXIST; +} + +/* * Lookup a name in a directory, give back the inode number. + * If ci_name is not NULL, returns the actual name in ci_name if it differs + * to name, or ci_name->name is set to NULL for an exact match. */ + int xfs_dir_lookup( xfs_trans_t *tp, xfs_inode_t *dp, struct xfs_name *name, - xfs_ino_t *inum) /* out: inode number */ + xfs_ino_t *inum, /* out: inode number */ + struct xfs_name *ci_name) /* out: actual name if CI match */ { xfs_da_args_t args; int rval; @@ -206,15 +287,17 @@ xfs_dir_lookup( ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR); XFS_STATS_INC(xs_dir_lookup); - memset(&args, 0, sizeof(xfs_da_args_t)); + memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; - args.hashval = xfs_da_hashname(name->name, name->len); + args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.dp = dp; args.whichfork = XFS_DATA_FORK; args.trans = tp; - args.oknoent = 1; + args.op_flags = XFS_DA_OP_OKNOENT; + if (ci_name) + args.op_flags |= XFS_DA_OP_CILOOKUP; if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_lookup(&args); @@ -230,8 +313,13 @@ xfs_dir_lookup( rval = xfs_dir2_node_lookup(&args); if (rval == EEXIST) rval = 0; - if (rval == 0) + if (!rval) { *inum = args.inumber; + if (ci_name) { + ci_name->name = args.value; + ci_name->len = args.valuelen; + } + } return rval; } @@ -255,9 +343,10 @@ xfs_dir_removename( ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR); XFS_STATS_INC(xs_dir_remove); + memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; - args.hashval = xfs_da_hashname(name->name, name->len); + args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = ino; args.dp = dp; args.firstblock = first; @@ -265,7 +354,6 @@ xfs_dir_removename( args.total = total; args.whichfork = XFS_DATA_FORK; args.trans = tp; - args.justcheck = args.addname = args.oknoent = 0; if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_removename(&args); @@ -338,9 +426,10 @@ xfs_dir_replace( if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum))) return rval; + memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; - args.hashval = xfs_da_hashname(name->name, name->len); + args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.inumber = inum; args.dp = dp; args.firstblock = first; @@ -348,7 +437,6 @@ xfs_dir_replace( args.total = total; args.whichfork = XFS_DATA_FORK; args.trans = tp; - args.justcheck = args.addname = args.oknoent = 0; if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_replace(&args); @@ -384,15 +472,16 @@ xfs_dir_canenter( return 0; ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR); - memset(&args, 0, sizeof(xfs_da_args_t)); + memset(&args, 0, sizeof(xfs_da_args_t)); args.name = name->name; args.namelen = name->len; - args.hashval = xfs_da_hashname(name->name, name->len); + args.hashval = dp->i_mount->m_dirnameops->hashname(name); args.dp = dp; args.whichfork = XFS_DATA_FORK; args.trans = tp; - args.justcheck = args.addname = args.oknoent = 1; + args.op_flags = XFS_DA_OP_JUSTCHECK | XFS_DA_OP_ADDNAME | + XFS_DA_OP_OKNOENT; if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) rval = xfs_dir2_sf_addname(&args); @@ -493,7 +582,7 @@ xfs_dir2_grow_inode( args->firstblock, args->total, &mapp[mapi], &nmap, args->flist, NULL))) { - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); return error; } if (nmap < 1) @@ -525,14 +614,14 @@ xfs_dir2_grow_inode( mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount != bno + count) { if (mapp != &map) - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); return XFS_ERROR(ENOSPC); } /* * Done with the temporary mapping table. */ if (mapp != &map) - kmem_free(mapp, sizeof(*mapp) * count); + kmem_free(mapp); *dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno); /* * Update file's size if this is the data space and it grew. diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h index 6392f93..1d9ef96 100644 --- a/fs/xfs/xfs_dir2.h +++ b/fs/xfs/xfs_dir2.h @@ -74,7 +74,8 @@ extern int xfs_dir_createname(struct xfs_trans *tp, struct xfs_inode *dp, xfs_fsblock_t *first, struct xfs_bmap_free *flist, xfs_extlen_t tot); extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp, - struct xfs_name *name, xfs_ino_t *inum); + struct xfs_name *name, xfs_ino_t *inum, + struct xfs_name *ci_name); extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_name *name, xfs_ino_t ino, xfs_fsblock_t *first, @@ -99,4 +100,7 @@ extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db, struct xfs_dabuf *bp); +extern int xfs_dir_cilookup_result(struct xfs_da_args *args, const char *name, + int len); + #endif /* __XFS_DIR2_H__ */ diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index fb5a556..e2fa0a1 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -215,7 +215,7 @@ xfs_dir2_block_addname( /* * If this isn't a real add, we're done with the buffer. */ - if (args->justcheck) + if (args->op_flags & XFS_DA_OP_JUSTCHECK) xfs_da_brelse(tp, bp); /* * If we don't have space for the new entry & leaf ... @@ -225,7 +225,7 @@ xfs_dir2_block_addname( * Not trying to actually do anything, or don't have * a space reservation: return no-space. */ - if (args->justcheck || args->total == 0) + if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0) return XFS_ERROR(ENOSPC); /* * Convert to the next larger format. @@ -240,7 +240,7 @@ xfs_dir2_block_addname( /* * Just checking, and it would work, so say so. */ - if (args->justcheck) + if (args->op_flags & XFS_DA_OP_JUSTCHECK) return 0; needlog = needscan = 0; /* @@ -610,14 +610,15 @@ xfs_dir2_block_lookup( /* * Get the offset from the leaf entry, to point to the data. */ - dep = (xfs_dir2_data_entry_t *) - ((char *)block + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address))); + dep = (xfs_dir2_data_entry_t *)((char *)block + + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address))); /* - * Fill in inode number, release the block. + * Fill in inode number, CI name if appropriate, release the block. */ args->inumber = be64_to_cpu(dep->inumber); + error = xfs_dir_cilookup_result(args, dep->name, dep->namelen); xfs_da_brelse(args->trans, bp); - return XFS_ERROR(EEXIST); + return XFS_ERROR(error); } /* @@ -643,6 +644,7 @@ xfs_dir2_block_lookup_int( int mid; /* binary search current idx */ xfs_mount_t *mp; /* filesystem mount point */ xfs_trans_t *tp; /* transaction pointer */ + enum xfs_dacmp cmp; /* comparison result */ dp = args->dp; tp = args->trans; @@ -673,7 +675,7 @@ xfs_dir2_block_lookup_int( else high = mid - 1; if (low > high) { - ASSERT(args->oknoent); + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); xfs_da_brelse(tp, bp); return XFS_ERROR(ENOENT); } @@ -697,20 +699,31 @@ xfs_dir2_block_lookup_int( dep = (xfs_dir2_data_entry_t *) ((char *)block + xfs_dir2_dataptr_to_off(mp, addr)); /* - * Compare, if it's right give back buffer & entry number. + * Compare name and if it's an exact match, return the index + * and buffer. If it's the first case-insensitive match, store + * the index and buffer and continue looking for an exact match. */ - if (dep->namelen == args->namelen && - dep->name[0] == args->name[0] && - memcmp(dep->name, args->name, args->namelen) == 0) { + cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen); + if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) { + args->cmpresult = cmp; *bpp = bp; *entno = mid; - return 0; + if (cmp == XFS_CMP_EXACT) + return 0; } - } while (++mid < be32_to_cpu(btp->count) && be32_to_cpu(blp[mid].hashval) == hash); + } while (++mid < be32_to_cpu(btp->count) && + be32_to_cpu(blp[mid].hashval) == hash); + + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); + /* + * Here, we can only be doing a lookup (not a rename or replace). + * If a case-insensitive match was found earlier, return success. + */ + if (args->cmpresult == XFS_CMP_CASE) + return 0; /* * No match, release the buffer and return ENOENT. */ - ASSERT(args->oknoent); xfs_da_brelse(tp, bp); return XFS_ERROR(ENOENT); } @@ -1033,6 +1046,7 @@ xfs_dir2_sf_to_block( xfs_dir2_sf_t *sfp; /* shortform structure */ __be16 *tagp; /* end of data entry */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_name name; xfs_dir2_trace_args("sf_to_block", args); dp = args->dp; @@ -1071,7 +1085,7 @@ xfs_dir2_sf_to_block( */ error = xfs_dir2_grow_inode(args, XFS_DIR2_DATA_SPACE, &blkno); if (error) { - kmem_free(buf, buf_len); + kmem_free(buf); return error; } /* @@ -1079,7 +1093,7 @@ xfs_dir2_sf_to_block( */ error = xfs_dir2_data_init(args, blkno, &bp); if (error) { - kmem_free(buf, buf_len); + kmem_free(buf); return error; } block = bp->data; @@ -1187,8 +1201,10 @@ xfs_dir2_sf_to_block( tagp = xfs_dir2_data_entry_tag_p(dep); *tagp = cpu_to_be16((char *)dep - (char *)block); xfs_dir2_data_log_entry(tp, bp, dep); - blp[2 + i].hashval = cpu_to_be32(xfs_da_hashname( - (char *)sfep->name, sfep->namelen)); + name.name = sfep->name; + name.len = sfep->namelen; + blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops-> + hashname(&name)); blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp, (char *)dep - (char *)block)); offset = (int)((char *)(tagp + 1) - (char *)block); @@ -1198,7 +1214,7 @@ xfs_dir2_sf_to_block( sfep = xfs_dir2_sf_nextentry(sfp, sfep); } /* Done with the temporary buffer */ - kmem_free(buf, buf_len); + kmem_free(buf); /* * Sort the leaf entries by hash value. */ diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c index fb8c9e0..498f8d6 100644 --- a/fs/xfs/xfs_dir2_data.c +++ b/fs/xfs/xfs_dir2_data.c @@ -65,6 +65,7 @@ xfs_dir2_data_check( xfs_mount_t *mp; /* filesystem mount point */ char *p; /* current data position */ int stale; /* count of stale leaves */ + struct xfs_name name; mp = dp->i_mount; d = bp->data; @@ -140,7 +141,9 @@ xfs_dir2_data_check( addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk, (xfs_dir2_data_aoff_t) ((char *)dep - (char *)d)); - hash = xfs_da_hashname((char *)dep->name, dep->namelen); + name.name = dep->name; + name.len = dep->namelen; + hash = mp->m_dirnameops->hashname(&name); for (i = 0; i < be32_to_cpu(btp->count); i++) { if (be32_to_cpu(lep[i].address) == addr && be32_to_cpu(lep[i].hashval) == hash) diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index bc52b80..9353599 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -263,20 +263,21 @@ xfs_dir2_leaf_addname( * If we don't have enough free bytes but we can make enough * by compacting out stale entries, we'll do that. */ - if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] < needbytes && - be16_to_cpu(leaf->hdr.stale) > 1) { + if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] < + needbytes && be16_to_cpu(leaf->hdr.stale) > 1) { compact = 1; } /* * Otherwise if we don't have enough free bytes we need to * convert to node form. */ - else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] < - needbytes) { + else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu( + leaf->hdr.count)] < needbytes) { /* * Just checking or no space reservation, give up. */ - if (args->justcheck || args->total == 0) { + if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || + args->total == 0) { xfs_da_brelse(tp, lbp); return XFS_ERROR(ENOSPC); } @@ -301,7 +302,7 @@ xfs_dir2_leaf_addname( * If just checking, then it will fit unless we needed to allocate * a new data block. */ - if (args->justcheck) { + if (args->op_flags & XFS_DA_OP_JUSTCHECK) { xfs_da_brelse(tp, lbp); return use_block == -1 ? XFS_ERROR(ENOSPC) : 0; } @@ -1110,7 +1111,7 @@ xfs_dir2_leaf_getdents( *offset = XFS_DIR2_MAX_DATAPTR; else *offset = xfs_dir2_byte_to_dataptr(mp, curoff); - kmem_free(map, map_size * sizeof(*map)); + kmem_free(map); if (bp) xfs_da_brelse(NULL, bp); return error; @@ -1298,12 +1299,13 @@ xfs_dir2_leaf_lookup( ((char *)dbp->data + xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address))); /* - * Return the found inode number. + * Return the found inode number & CI name if appropriate */ args->inumber = be64_to_cpu(dep->inumber); + error = xfs_dir_cilookup_result(args, dep->name, dep->namelen); xfs_da_brelse(tp, dbp); xfs_da_brelse(tp, lbp); - return XFS_ERROR(EEXIST); + return XFS_ERROR(error); } /* @@ -1319,8 +1321,8 @@ xfs_dir2_leaf_lookup_int( int *indexp, /* out: index in leaf block */ xfs_dabuf_t **dbpp) /* out: data buffer */ { - xfs_dir2_db_t curdb; /* current data block number */ - xfs_dabuf_t *dbp; /* data buffer */ + xfs_dir2_db_t curdb = -1; /* current data block number */ + xfs_dabuf_t *dbp = NULL; /* data buffer */ xfs_dir2_data_entry_t *dep; /* data entry */ xfs_inode_t *dp; /* incore directory inode */ int error; /* error return code */ @@ -1331,6 +1333,8 @@ xfs_dir2_leaf_lookup_int( xfs_mount_t *mp; /* filesystem mount point */ xfs_dir2_db_t newdb; /* new data block number */ xfs_trans_t *tp; /* transaction pointer */ + xfs_dir2_db_t cidb = -1; /* case match data block no. */ + enum xfs_dacmp cmp; /* name compare result */ dp = args->dp; tp = args->trans; @@ -1338,11 +1342,10 @@ xfs_dir2_leaf_lookup_int( /* * Read the leaf block into the buffer. */ - if ((error = - xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp, - XFS_DATA_FORK))) { + error = xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp, + XFS_DATA_FORK); + if (error) return error; - } *lbpp = lbp; leaf = lbp->data; xfs_dir2_leaf_check(dp, lbp); @@ -1354,9 +1357,9 @@ xfs_dir2_leaf_lookup_int( * Loop over all the entries with the right hash value * looking to match the name. */ - for (lep = &leaf->ents[index], dbp = NULL, curdb = -1; - index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval; - lep++, index++) { + for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) && + be32_to_cpu(lep->hashval) == args->hashval; + lep++, index++) { /* * Skip over stale leaf entries. */ @@ -1373,10 +1376,10 @@ xfs_dir2_leaf_lookup_int( if (newdb != curdb) { if (dbp) xfs_da_brelse(tp, dbp); - if ((error = - xfs_da_read_buf(tp, dp, - xfs_dir2_db_to_da(mp, newdb), -1, &dbp, - XFS_DATA_FORK))) { + error = xfs_da_read_buf(tp, dp, + xfs_dir2_db_to_da(mp, newdb), + -1, &dbp, XFS_DATA_FORK); + if (error) { xfs_da_brelse(tp, lbp); return error; } @@ -1386,24 +1389,50 @@ xfs_dir2_leaf_lookup_int( /* * Point to the data entry. */ - dep = (xfs_dir2_data_entry_t *) - ((char *)dbp->data + - xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address))); + dep = (xfs_dir2_data_entry_t *)((char *)dbp->data + + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address))); /* - * If it matches then return it. + * Compare name and if it's an exact match, return the index + * and buffer. If it's the first case-insensitive match, store + * the index and buffer and continue looking for an exact match. */ - if (dep->namelen == args->namelen && - dep->name[0] == args->name[0] && - memcmp(dep->name, args->name, args->namelen) == 0) { - *dbpp = dbp; + cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen); + if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) { + args->cmpresult = cmp; *indexp = index; - return 0; + /* case exact match: return the current buffer. */ + if (cmp == XFS_CMP_EXACT) { + *dbpp = dbp; + return 0; + } + cidb = curdb; } } + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); + /* + * Here, we can only be doing a lookup (not a rename or remove). + * If a case-insensitive match was found earlier, re-read the + * appropriate data block if required and return it. + */ + if (args->cmpresult == XFS_CMP_CASE) { + ASSERT(cidb != -1); + if (cidb != curdb) { + xfs_da_brelse(tp, dbp); + error = xfs_da_read_buf(tp, dp, + xfs_dir2_db_to_da(mp, cidb), + -1, &dbp, XFS_DATA_FORK); + if (error) { + xfs_da_brelse(tp, lbp); + return error; + } + } + *dbpp = dbp; + return 0; + } /* * No match found, return ENOENT. */ - ASSERT(args->oknoent); + ASSERT(cidb == -1); if (dbp) xfs_da_brelse(tp, dbp); xfs_da_brelse(tp, lbp); diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index 8dade71..fa6c3a5 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -226,7 +226,7 @@ xfs_dir2_leafn_add( ASSERT(index == be16_to_cpu(leaf->hdr.count) || be32_to_cpu(leaf->ents[index].hashval) >= args->hashval); - if (args->justcheck) + if (args->op_flags & XFS_DA_OP_JUSTCHECK) return 0; /* @@ -387,28 +387,26 @@ xfs_dir2_leafn_lasthash( } /* - * Look up a leaf entry in a node-format leaf block. - * If this is an addname then the extrablk in state is a freespace block, - * otherwise it's a data block. + * Look up a leaf entry for space to add a name in a node-format leaf block. + * The extrablk in state is a freespace block. */ -int -xfs_dir2_leafn_lookup_int( +STATIC int +xfs_dir2_leafn_lookup_for_addname( xfs_dabuf_t *bp, /* leaf buffer */ xfs_da_args_t *args, /* operation arguments */ int *indexp, /* out: leaf entry index */ xfs_da_state_t *state) /* state to fill in */ { - xfs_dabuf_t *curbp; /* current data/free buffer */ - xfs_dir2_db_t curdb; /* current data block number */ - xfs_dir2_db_t curfdb; /* current free block number */ - xfs_dir2_data_entry_t *dep; /* data block entry */ + xfs_dabuf_t *curbp = NULL; /* current data/free buffer */ + xfs_dir2_db_t curdb = -1; /* current data block number */ + xfs_dir2_db_t curfdb = -1; /* current free block number */ xfs_inode_t *dp; /* incore directory inode */ int error; /* error return value */ int fi; /* free entry index */ - xfs_dir2_free_t *free=NULL; /* free block structure */ + xfs_dir2_free_t *free = NULL; /* free block structure */ int index; /* leaf entry index */ xfs_dir2_leaf_t *leaf; /* leaf structure */ - int length=0; /* length of new data entry */ + int length; /* length of new data entry */ xfs_dir2_leaf_entry_t *lep; /* leaf entry */ xfs_mount_t *mp; /* filesystem mount point */ xfs_dir2_db_t newdb; /* new data block number */ @@ -431,33 +429,20 @@ xfs_dir2_leafn_lookup_int( /* * Do we have a buffer coming in? */ - if (state->extravalid) + if (state->extravalid) { + /* If so, it's a free block buffer, get the block number. */ curbp = state->extrablk.bp; - else - curbp = NULL; - /* - * For addname, it's a free block buffer, get the block number. - */ - if (args->addname) { - curfdb = curbp ? state->extrablk.blkno : -1; - curdb = -1; - length = xfs_dir2_data_entsize(args->namelen); - if ((free = (curbp ? curbp->data : NULL))) - ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC); - } - /* - * For others, it's a data block buffer, get the block number. - */ - else { - curfdb = -1; - curdb = curbp ? state->extrablk.blkno : -1; + curfdb = state->extrablk.blkno; + free = curbp->data; + ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC); } + length = xfs_dir2_data_entsize(args->namelen); /* * Loop over leaf entries with the right hash value. */ - for (lep = &leaf->ents[index]; - index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval; - lep++, index++) { + for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) && + be32_to_cpu(lep->hashval) == args->hashval; + lep++, index++) { /* * Skip stale leaf entries. */ @@ -471,161 +456,244 @@ xfs_dir2_leafn_lookup_int( * For addname, we're looking for a place to put the new entry. * We want to use a data block with an entry of equal * hash value to ours if there is one with room. + * + * If this block isn't the data block we already have + * in hand, take a look at it. */ - if (args->addname) { + if (newdb != curdb) { + curdb = newdb; /* - * If this block isn't the data block we already have - * in hand, take a look at it. + * Convert the data block to the free block + * holding its freespace information. */ - if (newdb != curdb) { - curdb = newdb; - /* - * Convert the data block to the free block - * holding its freespace information. - */ - newfdb = xfs_dir2_db_to_fdb(mp, newdb); - /* - * If it's not the one we have in hand, - * read it in. - */ - if (newfdb != curfdb) { - /* - * If we had one before, drop it. - */ - if (curbp) - xfs_da_brelse(tp, curbp); - /* - * Read the free block. - */ - if ((error = xfs_da_read_buf(tp, dp, - xfs_dir2_db_to_da(mp, - newfdb), - -1, &curbp, - XFS_DATA_FORK))) { - return error; - } - free = curbp->data; - ASSERT(be32_to_cpu(free->hdr.magic) == - XFS_DIR2_FREE_MAGIC); - ASSERT((be32_to_cpu(free->hdr.firstdb) % - XFS_DIR2_MAX_FREE_BESTS(mp)) == - 0); - ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb); - ASSERT(curdb < - be32_to_cpu(free->hdr.firstdb) + - be32_to_cpu(free->hdr.nvalid)); - } - /* - * Get the index for our entry. - */ - fi = xfs_dir2_db_to_fdindex(mp, curdb); - /* - * If it has room, return it. - */ - if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) { - XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int", - XFS_ERRLEVEL_LOW, mp); - if (curfdb != newfdb) - xfs_da_brelse(tp, curbp); - return XFS_ERROR(EFSCORRUPTED); - } - curfdb = newfdb; - if (be16_to_cpu(free->bests[fi]) >= length) { - *indexp = index; - state->extravalid = 1; - state->extrablk.bp = curbp; - state->extrablk.blkno = curfdb; - state->extrablk.index = fi; - state->extrablk.magic = - XFS_DIR2_FREE_MAGIC; - ASSERT(args->oknoent); - return XFS_ERROR(ENOENT); - } - } - } - /* - * Not adding a new entry, so we really want to find - * the name given to us. - */ - else { + newfdb = xfs_dir2_db_to_fdb(mp, newdb); /* - * If it's a different data block, go get it. + * If it's not the one we have in hand, read it in. */ - if (newdb != curdb) { + if (newfdb != curfdb) { /* - * If we had a block before, drop it. + * If we had one before, drop it. */ if (curbp) xfs_da_brelse(tp, curbp); /* - * Read the data block. + * Read the free block. */ - if ((error = - xfs_da_read_buf(tp, dp, - xfs_dir2_db_to_da(mp, newdb), -1, - &curbp, XFS_DATA_FORK))) { + error = xfs_da_read_buf(tp, dp, + xfs_dir2_db_to_da(mp, newfdb), + -1, &curbp, XFS_DATA_FORK); + if (error) return error; - } - xfs_dir2_data_check(dp, curbp); - curdb = newdb; + free = curbp->data; + ASSERT(be32_to_cpu(free->hdr.magic) == + XFS_DIR2_FREE_MAGIC); + ASSERT((be32_to_cpu(free->hdr.firstdb) % + XFS_DIR2_MAX_FREE_BESTS(mp)) == 0); + ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb); + ASSERT(curdb < be32_to_cpu(free->hdr.firstdb) + + be32_to_cpu(free->hdr.nvalid)); } /* - * Point to the data entry. + * Get the index for our entry. */ - dep = (xfs_dir2_data_entry_t *) - ((char *)curbp->data + - xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address))); + fi = xfs_dir2_db_to_fdindex(mp, curdb); /* - * Compare the entry, return it if it matches. + * If it has room, return it. */ - if (dep->namelen == args->namelen && - dep->name[0] == args->name[0] && - memcmp(dep->name, args->name, args->namelen) == 0) { - args->inumber = be64_to_cpu(dep->inumber); - *indexp = index; - state->extravalid = 1; - state->extrablk.bp = curbp; - state->extrablk.blkno = curdb; - state->extrablk.index = - (int)((char *)dep - - (char *)curbp->data); - state->extrablk.magic = XFS_DIR2_DATA_MAGIC; - return XFS_ERROR(EEXIST); + if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) { + XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int", + XFS_ERRLEVEL_LOW, mp); + if (curfdb != newfdb) + xfs_da_brelse(tp, curbp); + return XFS_ERROR(EFSCORRUPTED); } + curfdb = newfdb; + if (be16_to_cpu(free->bests[fi]) >= length) + goto out; } } + /* Didn't find any space */ + fi = -1; +out: + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); + if (curbp) { + /* Giving back a free block. */ + state->extravalid = 1; + state->extrablk.bp = curbp; + state->extrablk.index = fi; + state->extrablk.blkno = curfdb; + state->extrablk.magic = XFS_DIR2_FREE_MAGIC; + } else { + state->extravalid = 0; + } /* - * Didn't find a match. - * If we are holding a buffer, give it back in case our caller - * finds it useful. + * Return the index, that will be the insertion point. */ - if ((state->extravalid = (curbp != NULL))) { - state->extrablk.bp = curbp; - state->extrablk.index = -1; + *indexp = index; + return XFS_ERROR(ENOENT); +} + +/* + * Look up a leaf entry in a node-format leaf block. + * The extrablk in state a data block. + */ +STATIC int +xfs_dir2_leafn_lookup_for_entry( + xfs_dabuf_t *bp, /* leaf buffer */ + xfs_da_args_t *args, /* operation arguments */ + int *indexp, /* out: leaf entry index */ + xfs_da_state_t *state) /* state to fill in */ +{ + xfs_dabuf_t *curbp = NULL; /* current data/free buffer */ + xfs_dir2_db_t curdb = -1; /* current data block number */ + xfs_dir2_data_entry_t *dep; /* data block entry */ + xfs_inode_t *dp; /* incore directory inode */ + int error; /* error return value */ + int index; /* leaf entry index */ + xfs_dir2_leaf_t *leaf; /* leaf structure */ + xfs_dir2_leaf_entry_t *lep; /* leaf entry */ + xfs_mount_t *mp; /* filesystem mount point */ + xfs_dir2_db_t newdb; /* new data block number */ + xfs_trans_t *tp; /* transaction pointer */ + enum xfs_dacmp cmp; /* comparison result */ + + dp = args->dp; + tp = args->trans; + mp = dp->i_mount; + leaf = bp->data; + ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC); +#ifdef __KERNEL__ + ASSERT(be16_to_cpu(leaf->hdr.count) > 0); +#endif + xfs_dir2_leafn_check(dp, bp); + /* + * Look up the hash value in the leaf entries. + */ + index = xfs_dir2_leaf_search_hash(args, bp); + /* + * Do we have a buffer coming in? + */ + if (state->extravalid) { + curbp = state->extrablk.bp; + curdb = state->extrablk.blkno; + } + /* + * Loop over leaf entries with the right hash value. + */ + for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) && + be32_to_cpu(lep->hashval) == args->hashval; + lep++, index++) { /* - * For addname, giving back a free block. + * Skip stale leaf entries. */ - if (args->addname) { - state->extrablk.blkno = curfdb; - state->extrablk.magic = XFS_DIR2_FREE_MAGIC; + if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR) + continue; + /* + * Pull the data block number from the entry. + */ + newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address)); + /* + * Not adding a new entry, so we really want to find + * the name given to us. + * + * If it's a different data block, go get it. + */ + if (newdb != curdb) { + /* + * If we had a block before that we aren't saving + * for a CI name, drop it + */ + if (curbp && (args->cmpresult == XFS_CMP_DIFFERENT || + curdb != state->extrablk.blkno)) + xfs_da_brelse(tp, curbp); + /* + * If needing the block that is saved with a CI match, + * use it otherwise read in the new data block. + */ + if (args->cmpresult != XFS_CMP_DIFFERENT && + newdb == state->extrablk.blkno) { + ASSERT(state->extravalid); + curbp = state->extrablk.bp; + } else { + error = xfs_da_read_buf(tp, dp, + xfs_dir2_db_to_da(mp, newdb), + -1, &curbp, XFS_DATA_FORK); + if (error) + return error; + } + xfs_dir2_data_check(dp, curbp); + curdb = newdb; } /* - * For other callers, giving back a data block. + * Point to the data entry. */ - else { + dep = (xfs_dir2_data_entry_t *)((char *)curbp->data + + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address))); + /* + * Compare the entry and if it's an exact match, return + * EEXIST immediately. If it's the first case-insensitive + * match, store the block & inode number and continue looking. + */ + cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen); + if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) { + /* If there is a CI match block, drop it */ + if (args->cmpresult != XFS_CMP_DIFFERENT && + curdb != state->extrablk.blkno) + xfs_da_brelse(tp, state->extrablk.bp); + args->cmpresult = cmp; + args->inumber = be64_to_cpu(dep->inumber); + *indexp = index; + state->extravalid = 1; + state->extrablk.bp = curbp; state->extrablk.blkno = curdb; + state->extrablk.index = (int)((char *)dep - + (char *)curbp->data); state->extrablk.magic = XFS_DIR2_DATA_MAGIC; + if (cmp == XFS_CMP_EXACT) + return XFS_ERROR(EEXIST); } } - /* - * Return the final index, that will be the insertion point. - */ + ASSERT(index == be16_to_cpu(leaf->hdr.count) || + (args->op_flags & XFS_DA_OP_OKNOENT)); + if (curbp) { + if (args->cmpresult == XFS_CMP_DIFFERENT) { + /* Giving back last used data block. */ + state->extravalid = 1; + state->extrablk.bp = curbp; + state->extrablk.index = -1; + state->extrablk.blkno = curdb; + state->extrablk.magic = XFS_DIR2_DATA_MAGIC; + } else { + /* If the curbp is not the CI match block, drop it */ + if (state->extrablk.bp != curbp) + xfs_da_brelse(tp, curbp); + } + } else { + state->extravalid = 0; + } *indexp = index; - ASSERT(index == be16_to_cpu(leaf->hdr.count) || args->oknoent); return XFS_ERROR(ENOENT); } /* + * Look up a leaf entry in a node-format leaf block. + * If this is an addname then the extrablk in state is a freespace block, + * otherwise it's a data block. + */ +int +xfs_dir2_leafn_lookup_int( + xfs_dabuf_t *bp, /* leaf buffer */ + xfs_da_args_t *args, /* operation arguments */ + int *indexp, /* out: leaf entry index */ + xfs_da_state_t *state) /* state to fill in */ +{ + if (args->op_flags & XFS_DA_OP_ADDNAME) + return xfs_dir2_leafn_lookup_for_addname(bp, args, indexp, + state); + return xfs_dir2_leafn_lookup_for_entry(bp, args, indexp, state); +} + +/* * Move count leaf entries from source to destination leaf. * Log entries and headers. Stale entries are preserved. */ @@ -823,9 +891,10 @@ xfs_dir2_leafn_rebalance( */ if (!state->inleaf) blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count); - - /* - * Finally sanity check just to make sure we are not returning a negative index + + /* + * Finally sanity check just to make sure we are not returning a + * negative index */ if(blk2->index < 0) { state->inleaf = 1; @@ -1332,7 +1401,7 @@ xfs_dir2_node_addname( /* * It worked, fix the hash values up the btree. */ - if (!args->justcheck) + if (!(args->op_flags & XFS_DA_OP_JUSTCHECK)) xfs_da_fixhashpath(state, &state->path); } else { /* @@ -1515,7 +1584,8 @@ xfs_dir2_node_addname_int( /* * Not allowed to allocate, return failure. */ - if (args->justcheck || args->total == 0) { + if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || + args->total == 0) { /* * Drop the freespace buffer unless it came from our * caller. @@ -1661,7 +1731,7 @@ xfs_dir2_node_addname_int( /* * If just checking, we succeeded. */ - if (args->justcheck) { + if (args->op_flags & XFS_DA_OP_JUSTCHECK) { if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) xfs_da_buf_done(fbp); return 0; @@ -1767,6 +1837,14 @@ xfs_dir2_node_lookup( error = xfs_da_node_lookup_int(state, &rval); if (error) rval = error; + else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) { + /* If a CI match, dup the actual name and return EEXIST */ + xfs_dir2_data_entry_t *dep; + + dep = (xfs_dir2_data_entry_t *)((char *)state->extrablk.bp-> + data + state->extrablk.index); + rval = xfs_dir_cilookup_result(args, dep->name, dep->namelen); + } /* * Release the btree blocks and leaf block. */ @@ -1810,9 +1888,8 @@ xfs_dir2_node_removename( * Look up the entry we're deleting, set up the cursor. */ error = xfs_da_node_lookup_int(state, &rval); - if (error) { + if (error) rval = error; - } /* * Didn't find it, upper layer screwed up. */ @@ -1829,9 +1906,8 @@ xfs_dir2_node_removename( */ error = xfs_dir2_leafn_remove(args, blk->bp, blk->index, &state->extrablk, &rval); - if (error) { + if (error) return error; - } /* * Fix the hash values up the btree. */ diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c index 919d275a..b46af00 100644 --- a/fs/xfs/xfs_dir2_sf.c +++ b/fs/xfs/xfs_dir2_sf.c @@ -255,7 +255,7 @@ xfs_dir2_block_to_sf( xfs_dir2_sf_check(args); out: xfs_trans_log_inode(args->trans, dp, logflags); - kmem_free(block, mp->m_dirblksize); + kmem_free(block); return error; } @@ -332,7 +332,7 @@ xfs_dir2_sf_addname( /* * Just checking or no space reservation, it doesn't fit. */ - if (args->justcheck || args->total == 0) + if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0) return XFS_ERROR(ENOSPC); /* * Convert to block form then add the name. @@ -345,7 +345,7 @@ xfs_dir2_sf_addname( /* * Just checking, it fits. */ - if (args->justcheck) + if (args->op_flags & XFS_DA_OP_JUSTCHECK) return 0; /* * Do it the easy way - just add it at the end. @@ -512,7 +512,7 @@ xfs_dir2_sf_addname_hard( sfep = xfs_dir2_sf_nextentry(sfp, sfep); memcpy(sfep, oldsfep, old_isize - nbytes); } - kmem_free(buf, old_isize); + kmem_free(buf); dp->i_d.di_size = new_isize; xfs_dir2_sf_check(args); } @@ -812,8 +812,11 @@ xfs_dir2_sf_lookup( { xfs_inode_t *dp; /* incore directory inode */ int i; /* entry index */ + int error; xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */ xfs_dir2_sf_t *sfp; /* shortform structure */ + enum xfs_dacmp cmp; /* comparison result */ + xfs_dir2_sf_entry_t *ci_sfep; /* case-insens. entry */ xfs_dir2_trace_args("sf_lookup", args); xfs_dir2_sf_check(args); @@ -836,6 +839,7 @@ xfs_dir2_sf_lookup( */ if (args->namelen == 1 && args->name[0] == '.') { args->inumber = dp->i_ino; + args->cmpresult = XFS_CMP_EXACT; return XFS_ERROR(EEXIST); } /* @@ -844,28 +848,41 @@ xfs_dir2_sf_lookup( if (args->namelen == 2 && args->name[0] == '.' && args->name[1] == '.') { args->inumber = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent); + args->cmpresult = XFS_CMP_EXACT; return XFS_ERROR(EEXIST); } /* * Loop over all the entries trying to match ours. */ - for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); - i < sfp->hdr.count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { - if (sfep->namelen == args->namelen && - sfep->name[0] == args->name[0] && - memcmp(args->name, sfep->name, args->namelen) == 0) { - args->inumber = - xfs_dir2_sf_get_inumber(sfp, - xfs_dir2_sf_inumberp(sfep)); - return XFS_ERROR(EEXIST); + ci_sfep = NULL; + for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count; + i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + /* + * Compare name and if it's an exact match, return the inode + * number. If it's the first case-insensitive match, store the + * inode number and continue looking for an exact match. + */ + cmp = dp->i_mount->m_dirnameops->compname(args, sfep->name, + sfep->namelen); + if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) { + args->cmpresult = cmp; + args->inumber = xfs_dir2_sf_get_inumber(sfp, + xfs_dir2_sf_inumberp(sfep)); + if (cmp == XFS_CMP_EXACT) + return XFS_ERROR(EEXIST); + ci_sfep = sfep; } } + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); /* - * Didn't find it. + * Here, we can only be doing a lookup (not a rename or replace). + * If a case-insensitive match was not found, return ENOENT. */ - ASSERT(args->oknoent); - return XFS_ERROR(ENOENT); + if (!ci_sfep) + return XFS_ERROR(ENOENT); + /* otherwise process the CI match as required by the caller */ + error = xfs_dir_cilookup_result(args, ci_sfep->name, ci_sfep->namelen); + return XFS_ERROR(error); } /* @@ -904,24 +921,21 @@ xfs_dir2_sf_removename( * Loop over the old directory entries. * Find the one we're deleting. */ - for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); - i < sfp->hdr.count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { - if (sfep->namelen == args->namelen && - sfep->name[0] == args->name[0] && - memcmp(sfep->name, args->name, args->namelen) == 0) { + for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count; + i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + if (xfs_da_compname(args, sfep->name, sfep->namelen) == + XFS_CMP_EXACT) { ASSERT(xfs_dir2_sf_get_inumber(sfp, - xfs_dir2_sf_inumberp(sfep)) == - args->inumber); + xfs_dir2_sf_inumberp(sfep)) == + args->inumber); break; } } /* * Didn't find it. */ - if (i == sfp->hdr.count) { + if (i == sfp->hdr.count) return XFS_ERROR(ENOENT); - } /* * Calculate sizes. */ @@ -1042,11 +1056,10 @@ xfs_dir2_sf_replace( */ else { for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); - i < sfp->hdr.count; - i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { - if (sfep->namelen == args->namelen && - sfep->name[0] == args->name[0] && - memcmp(args->name, sfep->name, args->namelen) == 0) { + i < sfp->hdr.count; + i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) { + if (xfs_da_compname(args, sfep->name, sfep->namelen) == + XFS_CMP_EXACT) { #if XFS_BIG_INUMS || defined(DEBUG) ino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep)); @@ -1061,7 +1074,7 @@ xfs_dir2_sf_replace( * Didn't find it. */ if (i == sfp->hdr.count) { - ASSERT(args->oknoent); + ASSERT(args->op_flags & XFS_DA_OP_OKNOENT); #if XFS_BIG_INUMS if (i8elevated) xfs_dir2_sf_toino4(args); @@ -1174,7 +1187,7 @@ xfs_dir2_sf_toino4( /* * Clean up the inode. */ - kmem_free(buf, oldsize); + kmem_free(buf); dp->i_d.di_size = newsize; xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA); } @@ -1251,7 +1264,7 @@ xfs_dir2_sf_toino8( /* * Clean up the inode. */ - kmem_free(buf, oldsize); + kmem_free(buf); dp->i_d.di_size = newsize; xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA); } diff --git a/fs/xfs/xfs_dir2_sf.h b/fs/xfs/xfs_dir2_sf.h index 005629d..deecc9d 100644 --- a/fs/xfs/xfs_dir2_sf.h +++ b/fs/xfs/xfs_dir2_sf.h @@ -62,7 +62,7 @@ typedef union { * Normalized offset (in a data block) of the entry, really xfs_dir2_data_off_t. * Only need 16 bits, this is the byte offset into the single block form. */ -typedef struct { __uint8_t i[2]; } xfs_dir2_sf_off_t; +typedef struct { __uint8_t i[2]; } __arch_pack xfs_dir2_sf_off_t; /* * The parent directory has a dedicated field, and the self-pointer must @@ -76,14 +76,14 @@ typedef struct xfs_dir2_sf_hdr { __uint8_t count; /* count of entries */ __uint8_t i8count; /* count of 8-byte inode #s */ xfs_dir2_inou_t parent; /* parent dir inode number */ -} xfs_dir2_sf_hdr_t; +} __arch_pack xfs_dir2_sf_hdr_t; typedef struct xfs_dir2_sf_entry { __uint8_t namelen; /* actual name length */ xfs_dir2_sf_off_t offset; /* saved offset */ __uint8_t name[1]; /* name, variable size */ xfs_dir2_inou_t inumber; /* inode number, var. offset */ -} xfs_dir2_sf_entry_t; +} __arch_pack xfs_dir2_sf_entry_t; typedef struct xfs_dir2_sf { xfs_dir2_sf_hdr_t hdr; /* shortform header */ diff --git a/fs/xfs/xfs_dir2_trace.c b/fs/xfs/xfs_dir2_trace.c index f3fb2ff..6cc7c0c 100644 --- a/fs/xfs/xfs_dir2_trace.c +++ b/fs/xfs/xfs_dir2_trace.c @@ -85,7 +85,8 @@ xfs_dir2_trace_args( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, NULL, NULL); + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), + NULL, NULL); } void @@ -100,7 +101,7 @@ xfs_dir2_trace_args_b( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), (void *)(bp ? bp->bps[0] : NULL), NULL); } @@ -117,7 +118,7 @@ xfs_dir2_trace_args_bb( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), (void *)(lbp ? lbp->bps[0] : NULL), (void *)(dbp ? dbp->bps[0] : NULL)); } @@ -157,8 +158,8 @@ xfs_dir2_trace_args_db( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, (void *)(long)db, - (void *)dbp); + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), + (void *)(long)db, (void *)dbp); } void @@ -173,7 +174,7 @@ xfs_dir2_trace_args_i( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), (void *)((unsigned long)(i >> 32)), (void *)((unsigned long)(i & 0xFFFFFFFF))); } @@ -190,7 +191,8 @@ xfs_dir2_trace_args_s( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, (void *)(long)s, NULL); + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), + (void *)(long)s, NULL); } void @@ -208,7 +210,7 @@ xfs_dir2_trace_args_sb( (void *)((unsigned long)(args->inumber >> 32)), (void *)((unsigned long)(args->inumber & 0xFFFFFFFF)), (void *)args->dp, (void *)args->trans, - (void *)(unsigned long)args->justcheck, (void *)(long)s, - (void *)dbp); + (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK), + (void *)(long)s, (void *)dbp); } #endif /* XFS_DIR2_TRACE */ diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h index f71784a..2813cdd 100644 --- a/fs/xfs/xfs_dmapi.h +++ b/fs/xfs/xfs_dmapi.h @@ -18,7 +18,6 @@ #ifndef __XFS_DMAPI_H__ #define __XFS_DMAPI_H__ -#include <linux/version.h> /* Values used to define the on-disk version of dm_attrname_t. All * on-disk attribute names start with the 8-byte string "SGI_DMI_". * @@ -166,6 +165,6 @@ typedef enum { #define FILP_DELAY_FLAG(filp) ((filp->f_flags&(O_NDELAY|O_NONBLOCK)) ? \ DM_FLAGS_NDELAY : 0) -#define AT_DELAY_FLAG(f) ((f&ATTR_NONBLOCK) ? DM_FLAGS_NDELAY : 0) +#define AT_DELAY_FLAG(f) ((f & XFS_ATTR_NONBLOCK) ? DM_FLAGS_NDELAY : 0) #endif /* __XFS_DMAPI_H__ */ diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index 05e5365..f227ecd 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -58,22 +58,11 @@ xfs_error_trap(int e) } return e; } -#endif - -#if (defined(DEBUG) || defined(INDUCE_IO_ERROR)) int xfs_etest[XFS_NUM_INJECT_ERROR]; int64_t xfs_etest_fsid[XFS_NUM_INJECT_ERROR]; char * xfs_etest_fsname[XFS_NUM_INJECT_ERROR]; -void -xfs_error_test_init(void) -{ - memset(xfs_etest, 0, sizeof(xfs_etest)); - memset(xfs_etest_fsid, 0, sizeof(xfs_etest_fsid)); - memset(xfs_etest_fsname, 0, sizeof(xfs_etest_fsname)); -} - int xfs_error_test(int error_tag, int *fsidp, char *expression, int line, char *file, unsigned long randfactor) @@ -150,8 +139,7 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud) xfs_etest[i]); xfs_etest[i] = 0; xfs_etest_fsid[i] = 0LL; - kmem_free(xfs_etest_fsname[i], - strlen(xfs_etest_fsname[i]) + 1); + kmem_free(xfs_etest_fsname[i]); xfs_etest_fsname[i] = NULL; } } @@ -163,7 +151,7 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud) return 0; } -#endif /* DEBUG || INDUCE_IO_ERROR */ +#endif /* DEBUG */ static void xfs_fs_vcmn_err(int level, xfs_mount_t *mp, char *fmt, va_list ap) @@ -175,7 +163,7 @@ xfs_fs_vcmn_err(int level, xfs_mount_t *mp, char *fmt, va_list ap) newfmt = kmem_alloc(len, KM_SLEEP); sprintf(newfmt, "Filesystem \"%s\": %s", mp->m_fsname, fmt); icmn_err(level, newfmt, ap); - kmem_free(newfmt, len); + kmem_free(newfmt); } else { icmn_err(level, fmt, ap); } diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h index 6490d2a..11543f1 100644 --- a/fs/xfs/xfs_error.h +++ b/fs/xfs/xfs_error.h @@ -125,23 +125,14 @@ extern void xfs_corruption_error(char *tag, int level, struct xfs_mount *mp, #define XFS_RANDOM_DIOWRITE_IOERR (XFS_RANDOM_DEFAULT/10) #define XFS_RANDOM_BMAPIFORMAT XFS_RANDOM_DEFAULT -#if (defined(DEBUG) || defined(INDUCE_IO_ERROR)) +#ifdef DEBUG extern int xfs_error_test(int, int *, char *, int, char *, unsigned long); -extern void xfs_error_test_init(void); #define XFS_NUM_INJECT_ERROR 10 - -#ifdef __ANSI_CPP__ -#define XFS_TEST_ERROR(expr, mp, tag, rf) \ - ((expr) || \ - xfs_error_test((tag), (mp)->m_fixedfsid, #expr, __LINE__, __FILE__, \ - (rf))) -#else #define XFS_TEST_ERROR(expr, mp, tag, rf) \ ((expr) || \ xfs_error_test((tag), (mp)->m_fixedfsid, "expr", __LINE__, __FILE__, \ (rf))) -#endif /* __ANSI_CPP__ */ extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp); extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud); @@ -149,7 +140,7 @@ extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud); #define XFS_TEST_ERROR(expr, mp, tag, rf) (expr) #define xfs_errortag_add(tag, mp) (ENOSYS) #define xfs_errortag_clearall(mp, loud) (ENOSYS) -#endif /* (DEBUG || INDUCE_IO_ERROR) */ +#endif /* DEBUG */ /* * XFS panic tags -- allow a call to xfs_cmn_err() be turned into diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index 132bd07..8aa28f7 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -41,8 +41,7 @@ xfs_efi_item_free(xfs_efi_log_item_t *efip) int nexts = efip->efi_format.efi_nextents; if (nexts > XFS_EFI_MAX_FAST_EXTENTS) { - kmem_free(efip, sizeof(xfs_efi_log_item_t) + - (nexts - 1) * sizeof(xfs_extent_t)); + kmem_free(efip); } else { kmem_zone_free(xfs_efi_zone, efip); } @@ -374,8 +373,7 @@ xfs_efd_item_free(xfs_efd_log_item_t *efdp) int nexts = efdp->efd_format.efd_nextents; if (nexts > XFS_EFD_MAX_FAST_EXTENTS) { - kmem_free(efdp, sizeof(xfs_efd_log_item_t) + - (nexts - 1) * sizeof(xfs_extent_t)); + kmem_free(efdp); } else { kmem_zone_free(xfs_efd_zone, efdp); } diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 3f3785b..f3bb75d 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -397,10 +397,12 @@ int xfs_filestream_init(void) { item_zone = kmem_zone_init(sizeof(fstrm_item_t), "fstrm_item"); + if (!item_zone) + return -ENOMEM; #ifdef XFS_FILESTREAMS_TRACE - xfs_filestreams_trace_buf = ktrace_alloc(XFS_FSTRM_KTRACE_SIZE, KM_SLEEP); + xfs_filestreams_trace_buf = ktrace_alloc(XFS_FSTRM_KTRACE_SIZE, KM_NOFS); #endif - return item_zone ? 0 : -ENOMEM; + return 0; } /* diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 3bed643..01c0cc8 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -239,6 +239,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_LOGV2 0x0100 /* log format version 2 */ #define XFS_FSOP_GEOM_FLAGS_SECTOR 0x0200 /* sector sizes >1BB */ #define XFS_FSOP_GEOM_FLAGS_ATTR2 0x0400 /* inline attributes rework */ +#define XFS_FSOP_GEOM_FLAGS_DIRV2CI 0x1000 /* ASCII only CI names */ #define XFS_FSOP_GEOM_FLAGS_LAZYSB 0x4000 /* lazy superblock counters */ @@ -371,6 +372,9 @@ typedef struct xfs_fsop_attrlist_handlereq { typedef struct xfs_attr_multiop { __u32 am_opcode; +#define ATTR_OP_GET 1 /* return the indicated attr's value */ +#define ATTR_OP_SET 2 /* set/create the indicated attr/value pair */ +#define ATTR_OP_REMOVE 3 /* remove the indicated attr */ __s32 am_error; void __user *am_attrname; void __user *am_attrvalue; diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 381ebda..84583cf7 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -95,6 +95,8 @@ xfs_fs_geometry( XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) | (xfs_sb_version_hassector(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SECTOR : 0) | + (xfs_sb_version_hasasciici(&mp->m_sb) ? + XFS_FSOP_GEOM_FLAGS_DIRV2CI : 0) | (xfs_sb_version_haslazysbcount(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) | (xfs_sb_version_hasattr2(&mp->m_sb) ? @@ -625,7 +627,7 @@ xfs_fs_goingdown( xfs_force_shutdown(mp, SHUTDOWN_FORCE_UMOUNT); thaw_bdev(sb->s_bdev, sb); } - + break; } case XFS_FSOP_GOING_FLAGS_LOGFLUSH: diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c index e5310c90..83502f3 100644 --- a/fs/xfs/xfs_ialloc_btree.c +++ b/fs/xfs/xfs_ialloc_btree.c @@ -181,7 +181,7 @@ xfs_inobt_delrec( * then we can get rid of this level. */ if (numrecs == 1 && level > 0) { - agbp = cur->bc_private.i.agbp; + agbp = cur->bc_private.a.agbp; agi = XFS_BUF_TO_AGI(agbp); /* * pp is still set to the first pointer in the block. @@ -194,7 +194,7 @@ xfs_inobt_delrec( * Free the block. */ if ((error = xfs_free_extent(cur->bc_tp, - XFS_AGB_TO_FSB(mp, cur->bc_private.i.agno, bno), 1))) + XFS_AGB_TO_FSB(mp, cur->bc_private.a.agno, bno), 1))) return error; xfs_trans_binval(cur->bc_tp, bp); xfs_ialloc_log_agi(cur->bc_tp, agbp, @@ -379,7 +379,7 @@ xfs_inobt_delrec( rrecs = be16_to_cpu(right->bb_numrecs); rbp = bp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, - cur->bc_private.i.agno, lbno, 0, &lbp, + cur->bc_private.a.agno, lbno, 0, &lbp, XFS_INO_BTREE_REF))) return error; left = XFS_BUF_TO_INOBT_BLOCK(lbp); @@ -401,7 +401,7 @@ xfs_inobt_delrec( lrecs = be16_to_cpu(left->bb_numrecs); lbp = bp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, - cur->bc_private.i.agno, rbno, 0, &rbp, + cur->bc_private.a.agno, rbno, 0, &rbp, XFS_INO_BTREE_REF))) return error; right = XFS_BUF_TO_INOBT_BLOCK(rbp); @@ -484,7 +484,7 @@ xfs_inobt_delrec( xfs_buf_t *rrbp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, - cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib), 0, + cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib), 0, &rrbp, XFS_INO_BTREE_REF))) return error; rrblock = XFS_BUF_TO_INOBT_BLOCK(rrbp); @@ -497,7 +497,7 @@ xfs_inobt_delrec( * Free the deleting block. */ if ((error = xfs_free_extent(cur->bc_tp, XFS_AGB_TO_FSB(mp, - cur->bc_private.i.agno, rbno), 1))) + cur->bc_private.a.agno, rbno), 1))) return error; xfs_trans_binval(cur->bc_tp, rbp); /* @@ -854,7 +854,7 @@ xfs_inobt_lookup( { xfs_agi_t *agi; /* a.g. inode header */ - agi = XFS_BUF_TO_AGI(cur->bc_private.i.agbp); + agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp); agno = be32_to_cpu(agi->agi_seqno); agbno = be32_to_cpu(agi->agi_root); } @@ -1089,7 +1089,7 @@ xfs_inobt_lshift( * Set up the left neighbor as "left". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, be32_to_cpu(right->bb_leftsib), + cur->bc_private.a.agno, be32_to_cpu(right->bb_leftsib), 0, &lbp, XFS_INO_BTREE_REF))) return error; left = XFS_BUF_TO_INOBT_BLOCK(lbp); @@ -1207,10 +1207,10 @@ xfs_inobt_newroot( /* * Get a block & a buffer. */ - agi = XFS_BUF_TO_AGI(cur->bc_private.i.agbp); + agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp); args.tp = cur->bc_tp; args.mp = cur->bc_mp; - args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.i.agno, + args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, be32_to_cpu(agi->agi_root)); args.mod = args.minleft = args.alignment = args.total = args.wasdel = args.isfl = args.userdata = args.minalignslop = 0; @@ -1233,7 +1233,7 @@ xfs_inobt_newroot( */ agi->agi_root = cpu_to_be32(args.agbno); be32_add_cpu(&agi->agi_level, 1); - xfs_ialloc_log_agi(args.tp, cur->bc_private.i.agbp, + xfs_ialloc_log_agi(args.tp, cur->bc_private.a.agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL); /* * At the previous root level there are now two blocks: the old @@ -1376,7 +1376,7 @@ xfs_inobt_rshift( * Set up the right neighbor as "right". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib), + cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib), 0, &rbp, XFS_INO_BTREE_REF))) return error; right = XFS_BUF_TO_INOBT_BLOCK(rbp); @@ -1492,7 +1492,7 @@ xfs_inobt_split( * Allocate the new block. * If we can't do it, we're toast. Give up. */ - args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.i.agno, lbno); + args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, lbno); args.mod = args.minleft = args.alignment = args.total = args.wasdel = args.isfl = args.userdata = args.minalignslop = 0; args.minlen = args.maxlen = args.prod = 1; @@ -1725,7 +1725,7 @@ xfs_inobt_decrement( agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, agbno, 0, &bp, + cur->bc_private.a.agno, agbno, 0, &bp, XFS_INO_BTREE_REF))) return error; lev--; @@ -1897,7 +1897,7 @@ xfs_inobt_increment( agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, agbno, 0, &bp, + cur->bc_private.a.agno, agbno, 0, &bp, XFS_INO_BTREE_REF))) return error; lev--; diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index b07604b..e229e9e 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -216,7 +216,14 @@ finish_inode: mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino); init_waitqueue_head(&ip->i_ipin_wait); atomic_set(&ip->i_pincount, 0); - initnsema(&ip->i_flock, 1, "xfsfino"); + + /* + * Because we want to use a counting completion, complete + * the flush completion once to allow a single access to + * the flush completion without blocking. + */ + init_completion(&ip->i_flush); + complete(&ip->i_flush); if (lock_flags) xfs_ilock(ip, lock_flags); @@ -288,10 +295,17 @@ finish_inode: *ipp = ip; /* + * Set up the Linux with the Linux inode. + */ + ip->i_vnode = inode; + inode->i_private = ip; + + /* * If we have a real type for an on-disk inode, we can set ops(&unlock) * now. If it's a new inode being created, xfs_ialloc will handle it. */ - xfs_initialize_vnode(mp, inode, ip); + if (ip->i_d.di_mode != 0) + xfs_setup_inode(ip); return 0; } @@ -411,10 +425,11 @@ xfs_iput(xfs_inode_t *ip, * Special iput for brand-new inodes that are still locked */ void -xfs_iput_new(xfs_inode_t *ip, - uint lock_flags) +xfs_iput_new( + xfs_inode_t *ip, + uint lock_flags) { - struct inode *inode = ip->i_vnode; + struct inode *inode = VFS_I(ip); xfs_itrace_entry(ip); @@ -775,26 +790,3 @@ xfs_isilocked( } #endif -/* - * The following three routines simply manage the i_flock - * semaphore embedded in the inode. This semaphore synchronizes - * processes attempting to flush the in-core inode back to disk. - */ -void -xfs_iflock(xfs_inode_t *ip) -{ - psema(&(ip->i_flock), PINOD|PLTWAIT); -} - -int -xfs_iflock_nowait(xfs_inode_t *ip) -{ - return (cpsema(&(ip->i_flock))); -} - -void -xfs_ifunlock(xfs_inode_t *ip) -{ - ASSERT(issemalocked(&(ip->i_flock))); - vsema(&(ip->i_flock)); -} diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index e569bf5..dbd9cef 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -580,8 +580,8 @@ xfs_iformat_extents( xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip)); for (i = 0; i < nex; i++, dp++) { xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i); - ep->l0 = be64_to_cpu(get_unaligned(&dp->l0)); - ep->l1 = be64_to_cpu(get_unaligned(&dp->l1)); + ep->l0 = get_unaligned_be64(&dp->l0); + ep->l1 = get_unaligned_be64(&dp->l1); } XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork); if (whichfork != XFS_DATA_FORK || @@ -835,22 +835,22 @@ xfs_iread( * Do this before xfs_iformat in case it adds entries. */ #ifdef XFS_INODE_TRACE - ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_SLEEP); + ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_NOFS); #endif #ifdef XFS_BMAP_TRACE - ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_SLEEP); + ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_NOFS); #endif #ifdef XFS_BMBT_TRACE - ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_SLEEP); + ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_NOFS); #endif #ifdef XFS_RW_TRACE - ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_SLEEP); + ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_NOFS); #endif #ifdef XFS_ILOCK_TRACE - ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_SLEEP); + ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_NOFS); #endif #ifdef XFS_DIR2_TRACE - ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_SLEEP); + ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_NOFS); #endif /* @@ -1046,9 +1046,9 @@ xfs_ialloc( { xfs_ino_t ino; xfs_inode_t *ip; - bhv_vnode_t *vp; uint flags; int error; + timespec_t tv; /* * Call the space management code to pick @@ -1077,13 +1077,12 @@ xfs_ialloc( } ASSERT(ip != NULL); - vp = XFS_ITOV(ip); ip->i_d.di_mode = (__uint16_t)mode; ip->i_d.di_onlink = 0; ip->i_d.di_nlink = nlink; ASSERT(ip->i_d.di_nlink == nlink); - ip->i_d.di_uid = current_fsuid(cr); - ip->i_d.di_gid = current_fsgid(cr); + ip->i_d.di_uid = current_fsuid(); + ip->i_d.di_gid = current_fsgid(); ip->i_d.di_projid = prid; memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); @@ -1130,7 +1129,13 @@ xfs_ialloc( ip->i_size = 0; ip->i_d.di_nextents = 0; ASSERT(ip->i_d.di_nblocks == 0); - xfs_ichgtime(ip, XFS_ICHGTIME_CHG|XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD); + + nanotime(&tv); + ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec; + ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec; + ip->i_d.di_atime = ip->i_d.di_mtime; + ip->i_d.di_ctime = ip->i_d.di_mtime; + /* * di_gen will have been taken care of in xfs_iread. */ @@ -1220,7 +1225,7 @@ xfs_ialloc( xfs_trans_log_inode(tp, ip, flags); /* now that we have an i_mode we can setup inode ops and unlock */ - xfs_initialize_vnode(tp->t_mountp, vp, ip); + xfs_setup_inode(ip); *ipp = ip; return 0; @@ -1399,7 +1404,6 @@ xfs_itruncate_start( xfs_fsize_t last_byte; xfs_off_t toss_start; xfs_mount_t *mp; - bhv_vnode_t *vp; int error = 0; ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL)); @@ -1408,7 +1412,6 @@ xfs_itruncate_start( (flags == XFS_ITRUNC_MAYBE)); mp = ip->i_mount; - vp = XFS_ITOV(ip); /* wait for the completion of any pending DIOs */ if (new_size < ip->i_size) @@ -1457,7 +1460,7 @@ xfs_itruncate_start( #ifdef DEBUG if (new_size == 0) { - ASSERT(VN_CACHED(vp) == 0); + ASSERT(VN_CACHED(VFS_I(ip)) == 0); } #endif return error; @@ -1763,67 +1766,6 @@ xfs_itruncate_finish( return 0; } - -/* - * xfs_igrow_start - * - * Do the first part of growing a file: zero any data in the last - * block that is beyond the old EOF. We need to do this before - * the inode is joined to the transaction to modify the i_size. - * That way we can drop the inode lock and call into the buffer - * cache to get the buffer mapping the EOF. - */ -int -xfs_igrow_start( - xfs_inode_t *ip, - xfs_fsize_t new_size, - cred_t *credp) -{ - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - ASSERT(new_size > ip->i_size); - - /* - * Zero any pages that may have been created by - * xfs_write_file() beyond the end of the file - * and any blocks between the old and new file sizes. - */ - return xfs_zero_eof(ip, new_size, ip->i_size); -} - -/* - * xfs_igrow_finish - * - * This routine is called to extend the size of a file. - * The inode must have both the iolock and the ilock locked - * for update and it must be a part of the current transaction. - * The xfs_igrow_start() function must have been called previously. - * If the change_flag is not zero, the inode change timestamp will - * be updated. - */ -void -xfs_igrow_finish( - xfs_trans_t *tp, - xfs_inode_t *ip, - xfs_fsize_t new_size, - int change_flag) -{ - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - ASSERT(ip->i_transp == tp); - ASSERT(new_size > ip->i_size); - - /* - * Update the file size. Update the inode change timestamp - * if change_flag set. - */ - ip->i_d.di_size = new_size; - ip->i_size = new_size; - if (change_flag) - xfs_ichgtime(ip, XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - -} - - /* * This is called when the inode's link count goes to 0. * We place the on-disk inode on a list in the AGI. It @@ -2258,7 +2200,7 @@ xfs_ifree_cluster( xfs_trans_binval(tp, bp); } - kmem_free(ip_found, ninodes * sizeof(xfs_inode_t *)); + kmem_free(ip_found); xfs_put_perag(mp, pag); } @@ -2470,7 +2412,7 @@ xfs_iroot_realloc( (int)new_size); memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t)); } - kmem_free(ifp->if_broot, ifp->if_broot_bytes); + kmem_free(ifp->if_broot); ifp->if_broot = new_broot; ifp->if_broot_bytes = (int)new_size; ASSERT(ifp->if_broot_bytes <= @@ -2514,7 +2456,7 @@ xfs_idata_realloc( if (new_size == 0) { if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) { - kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes); + kmem_free(ifp->if_u1.if_data); } ifp->if_u1.if_data = NULL; real_size = 0; @@ -2529,7 +2471,7 @@ xfs_idata_realloc( ASSERT(ifp->if_real_bytes != 0); memcpy(ifp->if_u2.if_inline_data, ifp->if_u1.if_data, new_size); - kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes); + kmem_free(ifp->if_u1.if_data); ifp->if_u1.if_data = ifp->if_u2.if_inline_data; } real_size = 0; @@ -2636,7 +2578,7 @@ xfs_idestroy_fork( ifp = XFS_IFORK_PTR(ip, whichfork); if (ifp->if_broot != NULL) { - kmem_free(ifp->if_broot, ifp->if_broot_bytes); + kmem_free(ifp->if_broot); ifp->if_broot = NULL; } @@ -2650,7 +2592,7 @@ xfs_idestroy_fork( if ((ifp->if_u1.if_data != ifp->if_u2.if_inline_data) && (ifp->if_u1.if_data != NULL)) { ASSERT(ifp->if_real_bytes != 0); - kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes); + kmem_free(ifp->if_u1.if_data); ifp->if_u1.if_data = NULL; ifp->if_real_bytes = 0; } @@ -2691,7 +2633,6 @@ xfs_idestroy( xfs_idestroy_fork(ip, XFS_ATTR_FORK); mrfree(&ip->i_lock); mrfree(&ip->i_iolock); - freesema(&ip->i_flock); #ifdef XFS_INODE_TRACE ktrace_free(ip->i_trace); @@ -3058,7 +2999,7 @@ xfs_iflush_cluster( out_free: read_unlock(&pag->pag_ici_lock); - kmem_free(ilist, ilist_size); + kmem_free(ilist); return 0; @@ -3102,17 +3043,17 @@ cluster_corrupt_out: * Unlocks the flush lock */ xfs_iflush_abort(iq); - kmem_free(ilist, ilist_size); + kmem_free(ilist); return XFS_ERROR(EFSCORRUPTED); } /* * xfs_iflush() will write a modified inode's changes out to the * inode's on disk home. The caller must have the inode lock held - * in at least shared mode and the inode flush semaphore must be - * held as well. The inode lock will still be held upon return from + * in at least shared mode and the inode flush completion must be + * active as well. The inode lock will still be held upon return from * the call and the caller is free to unlock it. - * The inode flush lock will be unlocked when the inode reaches the disk. + * The inode flush will be completed when the inode reaches the disk. * The flags indicate how the inode's buffer should be written out. */ int @@ -3131,7 +3072,7 @@ xfs_iflush( XFS_STATS_INC(xs_iflush_count); ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - ASSERT(issemalocked(&(ip->i_flock))); + ASSERT(!completion_done(&ip->i_flush)); ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || ip->i_d.di_nextents > ip->i_df.if_ext_max); @@ -3143,8 +3084,6 @@ xfs_iflush( * flush lock and do nothing. */ if (xfs_inode_clean(ip)) { - ASSERT((iip != NULL) ? - !(iip->ili_item.li_flags & XFS_LI_IN_AIL) : 1); xfs_ifunlock(ip); return 0; } @@ -3296,7 +3235,7 @@ xfs_iflush_int( #endif ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - ASSERT(issemalocked(&(ip->i_flock))); + ASSERT(!completion_done(&ip->i_flush)); ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || ip->i_d.di_nextents > ip->i_df.if_ext_max); @@ -3528,7 +3467,6 @@ xfs_iflush_all( xfs_mount_t *mp) { xfs_inode_t *ip; - bhv_vnode_t *vp; again: XFS_MOUNT_ILOCK(mp); @@ -3543,14 +3481,13 @@ xfs_iflush_all( continue; } - vp = XFS_ITOV_NULL(ip); - if (!vp) { + if (!VFS_I(ip)) { XFS_MOUNT_IUNLOCK(mp); xfs_finish_reclaim(ip, 0, XFS_IFLUSH_ASYNC); goto again; } - ASSERT(vn_count(vp) == 0); + ASSERT(vn_count(VFS_I(ip)) == 0); ip = ip->i_mnext; } while (ip != mp->m_inodes); @@ -3770,7 +3707,7 @@ xfs_iext_add_indirect_multi( * (all extents past */ if (nex2) { byte_diff = nex2 * sizeof(xfs_bmbt_rec_t); - nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_SLEEP); + nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_NOFS); memmove(nex2_ep, &erp->er_extbuf[idx], byte_diff); erp->er_extcount -= nex2; xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -nex2); @@ -3836,7 +3773,7 @@ xfs_iext_add_indirect_multi( erp = xfs_iext_irec_new(ifp, erp_idx); } memmove(&erp->er_extbuf[i], nex2_ep, byte_diff); - kmem_free(nex2_ep, byte_diff); + kmem_free(nex2_ep); erp->er_extcount += nex2; xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, nex2); } @@ -4070,8 +4007,7 @@ xfs_iext_realloc_direct( ifp->if_u1.if_extents = kmem_realloc(ifp->if_u1.if_extents, rnew_size, - ifp->if_real_bytes, - KM_SLEEP); + ifp->if_real_bytes, KM_NOFS); } if (rnew_size > ifp->if_real_bytes) { memset(&ifp->if_u1.if_extents[ifp->if_bytes / @@ -4112,7 +4048,7 @@ xfs_iext_direct_to_inline( */ memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents, nextents * sizeof(xfs_bmbt_rec_t)); - kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes); + kmem_free(ifp->if_u1.if_extents); ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext; ifp->if_real_bytes = 0; } @@ -4130,7 +4066,7 @@ xfs_iext_inline_to_direct( xfs_ifork_t *ifp, /* inode fork pointer */ int new_size) /* number of extents in file */ { - ifp->if_u1.if_extents = kmem_alloc(new_size, KM_SLEEP); + ifp->if_u1.if_extents = kmem_alloc(new_size, KM_NOFS); memset(ifp->if_u1.if_extents, 0, new_size); if (ifp->if_bytes) { memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext, @@ -4162,7 +4098,7 @@ xfs_iext_realloc_indirect( } else { ifp->if_u1.if_ext_irec = (xfs_ext_irec_t *) kmem_realloc(ifp->if_u1.if_ext_irec, - new_size, size, KM_SLEEP); + new_size, size, KM_NOFS); } } @@ -4182,11 +4118,11 @@ xfs_iext_indirect_to_direct( ASSERT(nextents <= XFS_LINEAR_EXTS); size = nextents * sizeof(xfs_bmbt_rec_t); - xfs_iext_irec_compact_full(ifp); + xfs_iext_irec_compact_pages(ifp); ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ); ep = ifp->if_u1.if_ext_irec->er_extbuf; - kmem_free(ifp->if_u1.if_ext_irec, sizeof(xfs_ext_irec_t)); + kmem_free(ifp->if_u1.if_ext_irec); ifp->if_flags &= ~XFS_IFEXTIREC; ifp->if_u1.if_extents = ep; ifp->if_bytes = size; @@ -4212,7 +4148,7 @@ xfs_iext_destroy( } ifp->if_flags &= ~XFS_IFEXTIREC; } else if (ifp->if_real_bytes) { - kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes); + kmem_free(ifp->if_u1.if_extents); } else if (ifp->if_bytes) { memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS * sizeof(xfs_bmbt_rec_t)); @@ -4404,11 +4340,10 @@ xfs_iext_irec_init( nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); ASSERT(nextents <= XFS_LINEAR_EXTS); - erp = (xfs_ext_irec_t *) - kmem_alloc(sizeof(xfs_ext_irec_t), KM_SLEEP); + erp = kmem_alloc(sizeof(xfs_ext_irec_t), KM_NOFS); if (nextents == 0) { - ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP); + ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); } else if (!ifp->if_real_bytes) { xfs_iext_inline_to_direct(ifp, XFS_IEXT_BUFSZ); } else if (ifp->if_real_bytes < XFS_IEXT_BUFSZ) { @@ -4456,7 +4391,7 @@ xfs_iext_irec_new( /* Initialize new extent record */ erp = ifp->if_u1.if_ext_irec; - erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP); + erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS); ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; memset(erp[erp_idx].er_extbuf, 0, XFS_IEXT_BUFSZ); erp[erp_idx].er_extcount = 0; @@ -4483,7 +4418,7 @@ xfs_iext_irec_remove( if (erp->er_extbuf) { xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -erp->er_extcount); - kmem_free(erp->er_extbuf, XFS_IEXT_BUFSZ); + kmem_free(erp->er_extbuf); } /* Compact extent records */ erp = ifp->if_u1.if_ext_irec; @@ -4501,8 +4436,7 @@ xfs_iext_irec_remove( xfs_iext_realloc_indirect(ifp, nlists * sizeof(xfs_ext_irec_t)); } else { - kmem_free(ifp->if_u1.if_ext_irec, - sizeof(xfs_ext_irec_t)); + kmem_free(ifp->if_u1.if_ext_irec); } ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ; } @@ -4515,8 +4449,7 @@ xfs_iext_irec_remove( * compaction policy is as follows: * * Full Compaction: Extents fit into a single page (or inline buffer) - * Full Compaction: Extents occupy less than 10% of allocated space - * Partial Compaction: Extents occupy > 10% and < 50% of allocated space + * Partial Compaction: Extents occupy less than 50% of allocated space * No Compaction: Extents occupy at least 50% of allocated space */ void @@ -4537,8 +4470,6 @@ xfs_iext_irec_compact( xfs_iext_direct_to_inline(ifp, nextents); } else if (nextents <= XFS_LINEAR_EXTS) { xfs_iext_indirect_to_direct(ifp); - } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 3) { - xfs_iext_irec_compact_full(ifp); } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) { xfs_iext_irec_compact_pages(ifp); } @@ -4562,7 +4493,7 @@ xfs_iext_irec_compact_pages( erp_next = erp + 1; if (erp_next->er_extcount <= (XFS_LINEAR_EXTS - erp->er_extcount)) { - memmove(&erp->er_extbuf[erp->er_extcount], + memcpy(&erp->er_extbuf[erp->er_extcount], erp_next->er_extbuf, erp_next->er_extcount * sizeof(xfs_bmbt_rec_t)); erp->er_extcount += erp_next->er_extcount; @@ -4571,75 +4502,13 @@ xfs_iext_irec_compact_pages( * so er_extoffs don't get modified in * xfs_iext_irec_remove. */ - kmem_free(erp_next->er_extbuf, XFS_IEXT_BUFSZ); - erp_next->er_extbuf = NULL; - xfs_iext_irec_remove(ifp, erp_idx + 1); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - } else { - erp_idx++; - } - } -} - -/* - * Fully compact the extent records managed by the indirection array. - */ -void -xfs_iext_irec_compact_full( - xfs_ifork_t *ifp) /* inode fork pointer */ -{ - xfs_bmbt_rec_host_t *ep, *ep_next; /* extent record pointers */ - xfs_ext_irec_t *erp, *erp_next; /* extent irec pointers */ - int erp_idx = 0; /* extent irec index */ - int ext_avail; /* empty entries in ex list */ - int ext_diff; /* number of exts to add */ - int nlists; /* number of irec's (ex lists) */ - - ASSERT(ifp->if_flags & XFS_IFEXTIREC); - nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - erp = ifp->if_u1.if_ext_irec; - ep = &erp->er_extbuf[erp->er_extcount]; - erp_next = erp + 1; - ep_next = erp_next->er_extbuf; - while (erp_idx < nlists - 1) { - ext_avail = XFS_LINEAR_EXTS - erp->er_extcount; - ext_diff = MIN(ext_avail, erp_next->er_extcount); - memcpy(ep, ep_next, ext_diff * sizeof(xfs_bmbt_rec_t)); - erp->er_extcount += ext_diff; - erp_next->er_extcount -= ext_diff; - /* Remove next page */ - if (erp_next->er_extcount == 0) { - /* - * Free page before removing extent record - * so er_extoffs don't get modified in - * xfs_iext_irec_remove. - */ - kmem_free(erp_next->er_extbuf, - erp_next->er_extcount * sizeof(xfs_bmbt_rec_t)); + kmem_free(erp_next->er_extbuf); erp_next->er_extbuf = NULL; xfs_iext_irec_remove(ifp, erp_idx + 1); - erp = &ifp->if_u1.if_ext_irec[erp_idx]; nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ; - /* Update next page */ } else { - /* Move rest of page up to become next new page */ - memmove(erp_next->er_extbuf, ep_next, - erp_next->er_extcount * sizeof(xfs_bmbt_rec_t)); - ep_next = erp_next->er_extbuf; - memset(&ep_next[erp_next->er_extcount], 0, - (XFS_LINEAR_EXTS - erp_next->er_extcount) * - sizeof(xfs_bmbt_rec_t)); - } - if (erp->er_extcount == XFS_LINEAR_EXTS) { erp_idx++; - if (erp_idx < nlists) - erp = &ifp->if_u1.if_ext_irec[erp_idx]; - else - break; } - ep = &erp->er_extbuf[erp->er_extcount]; - erp_next = erp + 1; - ep_next = erp_next->er_extbuf; } } diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 0a999fe..1420c49 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -87,8 +87,7 @@ typedef struct xfs_ifork { * Flags for xfs_ichgtime(). */ #define XFS_ICHGTIME_MOD 0x1 /* data fork modification timestamp */ -#define XFS_ICHGTIME_ACC 0x2 /* data fork access timestamp */ -#define XFS_ICHGTIME_CHG 0x4 /* inode field change timestamp */ +#define XFS_ICHGTIME_CHG 0x2 /* inode field change timestamp */ /* * Per-fork incore inode flags. @@ -204,7 +203,7 @@ typedef struct xfs_inode { struct xfs_inode *i_mprev; /* ptr to prev inode */ struct xfs_mount *i_mount; /* fs mount struct ptr */ struct list_head i_reclaim; /* reclaim list */ - bhv_vnode_t *i_vnode; /* vnode backpointer */ + struct inode *i_vnode; /* vnode backpointer */ struct xfs_dquot *i_udquot; /* user dquot */ struct xfs_dquot *i_gdquot; /* group dquot */ @@ -223,7 +222,7 @@ typedef struct xfs_inode { struct xfs_inode_log_item *i_itemp; /* logging information */ mrlock_t i_lock; /* inode lock */ mrlock_t i_iolock; /* inode IO lock */ - sema_t i_flock; /* inode flush lock */ + struct completion i_flush; /* inode flush completion q */ atomic_t i_pincount; /* inode pin count */ wait_queue_head_t i_ipin_wait; /* inode pinning wait queue */ spinlock_t i_flags_lock; /* inode i_flags lock */ @@ -263,6 +262,18 @@ typedef struct xfs_inode { #define XFS_ISIZE(ip) (((ip)->i_d.di_mode & S_IFMT) == S_IFREG) ? \ (ip)->i_size : (ip)->i_d.di_size; +/* Convert from vfs inode to xfs inode */ +static inline struct xfs_inode *XFS_I(struct inode *inode) +{ + return (struct xfs_inode *)inode->i_private; +} + +/* convert from xfs inode to vfs inode */ +static inline struct inode *VFS_I(struct xfs_inode *ip) +{ + return (struct inode *)ip->i_vnode; +} + /* * i_flags helper functions */ @@ -439,9 +450,6 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) #define XFS_ITRUNC_DEFINITE 0x1 #define XFS_ITRUNC_MAYBE 0x2 -#define XFS_ITOV(ip) ((ip)->i_vnode) -#define XFS_ITOV_NULL(ip) ((ip)->i_vnode) - /* * For multiple groups support: if S_ISGID bit is set in the parent * directory, group of new file is set to that of the parent, and @@ -473,11 +481,8 @@ int xfs_ilock_nowait(xfs_inode_t *, uint); void xfs_iunlock(xfs_inode_t *, uint); void xfs_ilock_demote(xfs_inode_t *, uint); int xfs_isilocked(xfs_inode_t *, uint); -void xfs_iflock(xfs_inode_t *); -int xfs_iflock_nowait(xfs_inode_t *); uint xfs_ilock_map_shared(xfs_inode_t *); void xfs_iunlock_map_shared(xfs_inode_t *, uint); -void xfs_ifunlock(xfs_inode_t *); void xfs_ireclaim(xfs_inode_t *); int xfs_finish_reclaim(xfs_inode_t *, int, int); int xfs_finish_reclaim_all(struct xfs_mount *, int); @@ -507,9 +512,6 @@ int xfs_itruncate_start(xfs_inode_t *, uint, xfs_fsize_t); int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *, xfs_fsize_t, int, int); int xfs_iunlink(struct xfs_trans *, xfs_inode_t *); -int xfs_igrow_start(xfs_inode_t *, xfs_fsize_t, struct cred *); -void xfs_igrow_finish(struct xfs_trans *, xfs_inode_t *, - xfs_fsize_t, int); void xfs_idestroy_fork(xfs_inode_t *, int); void xfs_idestroy(xfs_inode_t *); @@ -525,6 +527,7 @@ void xfs_iflush_all(struct xfs_mount *); void xfs_ichgtime(xfs_inode_t *, int); xfs_fsize_t xfs_file_last_byte(xfs_inode_t *); void xfs_lock_inodes(xfs_inode_t **, int, uint); +void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint); void xfs_synchronize_atime(xfs_inode_t *); void xfs_mark_inode_dirty_sync(xfs_inode_t *); @@ -573,6 +576,26 @@ extern struct kmem_zone *xfs_ifork_zone; extern struct kmem_zone *xfs_inode_zone; extern struct kmem_zone *xfs_ili_zone; +/* + * Manage the i_flush queue embedded in the inode. This completion + * queue synchronizes processes attempting to flush the in-core + * inode back to disk. + */ +static inline void xfs_iflock(xfs_inode_t *ip) +{ + wait_for_completion(&ip->i_flush); +} + +static inline int xfs_iflock_nowait(xfs_inode_t *ip) +{ + return try_wait_for_completion(&ip->i_flush); +} + +static inline void xfs_ifunlock(xfs_inode_t *ip) +{ + complete(&ip->i_flush); +} + #endif /* __KERNEL__ */ #endif /* __XFS_INODE_H__ */ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 167b33f..97c7452 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -686,7 +686,7 @@ xfs_inode_item_unlock( ASSERT(ip->i_d.di_nextents > 0); ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_DEXT); ASSERT(ip->i_df.if_bytes > 0); - kmem_free(iip->ili_extents_buf, ip->i_df.if_bytes); + kmem_free(iip->ili_extents_buf); iip->ili_extents_buf = NULL; } if (iip->ili_aextents_buf != NULL) { @@ -694,7 +694,7 @@ xfs_inode_item_unlock( ASSERT(ip->i_d.di_anextents > 0); ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_AEXT); ASSERT(ip->i_afp->if_bytes > 0); - kmem_free(iip->ili_aextents_buf, ip->i_afp->if_bytes); + kmem_free(iip->ili_aextents_buf); iip->ili_aextents_buf = NULL; } @@ -779,11 +779,10 @@ xfs_inode_item_pushbuf( ASSERT(iip->ili_push_owner == current_pid()); /* - * If flushlock isn't locked anymore, chances are that the - * inode flush completed and the inode was taken off the AIL. - * So, just get out. + * If a flush is not in progress anymore, chances are that the + * inode was taken off the AIL. So, just get out. */ - if (!issemalocked(&(ip->i_flock)) || + if (completion_done(&ip->i_flush) || ((iip->ili_item.li_flags & XFS_LI_IN_AIL) == 0)) { iip->ili_pushbuf_flag = 0; xfs_iunlock(ip, XFS_ILOCK_SHARED); @@ -805,7 +804,7 @@ xfs_inode_item_pushbuf( * If not, we can flush it async. */ dopush = ((iip->ili_item.li_flags & XFS_LI_IN_AIL) && - issemalocked(&(ip->i_flock))); + !completion_done(&ip->i_flush)); iip->ili_pushbuf_flag = 0; xfs_iunlock(ip, XFS_ILOCK_SHARED); xfs_buftrace("INODE ITEM PUSH", bp); @@ -858,7 +857,7 @@ xfs_inode_item_push( ip = iip->ili_inode; ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED)); - ASSERT(issemalocked(&(ip->i_flock))); + ASSERT(!completion_done(&ip->i_flush)); /* * Since we were able to lock the inode's flush lock and * we found it on the AIL, the inode must be dirty. This @@ -957,8 +956,7 @@ xfs_inode_item_destroy( { #ifdef XFS_TRANS_DEBUG if (ip->i_itemp->ili_root_size != 0) { - kmem_free(ip->i_itemp->ili_orig_root, - ip->i_itemp->ili_root_size); + kmem_free(ip->i_itemp->ili_orig_root); } #endif kmem_zone_free(xfs_ili_zone, ip->i_itemp); diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 7edcde6..67f22b2 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -889,6 +889,16 @@ xfs_iomap_write_unwritten( count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count); count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb); + /* + * Reserve enough blocks in this transaction for two complete extent + * btree splits. We may be converting the middle part of an unwritten + * extent and in this case we will insert two new extents in the btree + * each of which could cause a full split. + * + * This reservation amount will be used in the first call to + * xfs_bmbt_split() to select an AG with enough space to satisfy the + * rest of the operation. + */ resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1; do { diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 419de15..cf6754a 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -59,7 +59,6 @@ xfs_bulkstat_one_iget( { xfs_icdinode_t *dic; /* dinode core info pointer */ xfs_inode_t *ip; /* incore inode pointer */ - bhv_vnode_t *vp; int error; error = xfs_iget(mp, NULL, ino, @@ -72,7 +71,6 @@ xfs_bulkstat_one_iget( ASSERT(ip != NULL); ASSERT(ip->i_blkno != (xfs_daddr_t)0); - vp = XFS_ITOV(ip); dic = &ip->i_d; /* xfs_iget returns the following without needing @@ -85,7 +83,7 @@ xfs_bulkstat_one_iget( buf->bs_uid = dic->di_uid; buf->bs_gid = dic->di_gid; buf->bs_size = dic->di_size; - vn_atime_to_bstime(vp, &buf->bs_atime); + vn_atime_to_bstime(VFS_I(ip), &buf->bs_atime); buf->bs_mtime.tv_sec = dic->di_mtime.t_sec; buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec; buf->bs_ctime.tv_sec = dic->di_ctime.t_sec; @@ -257,7 +255,7 @@ xfs_bulkstat_one( *ubused = error; out_free: - kmem_free(buf, sizeof(*buf)); + kmem_free(buf); return error; } @@ -708,7 +706,7 @@ xfs_bulkstat( /* * Done, we're either out of filesystem or space to put the data. */ - kmem_free(irbuf, irbsize); + kmem_free(irbuf); *ubcountp = ubelem; /* * Found some inodes, return them now and return the error next time. @@ -914,7 +912,7 @@ xfs_inumbers( } *lastino = XFS_AGINO_TO_INO(mp, agno, agino); } - kmem_free(buffer, bcount * sizeof(*buffer)); + kmem_free(buffer); if (cur) xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR)); diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index ad3d26d..0b02c64 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -124,16 +124,27 @@ STATIC void xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog, STATIC int xlog_iclogs_empty(xlog_t *log); #if defined(XFS_LOG_TRACE) + +#define XLOG_TRACE_LOGGRANT_SIZE 2048 +#define XLOG_TRACE_ICLOG_SIZE 256 + +void +xlog_trace_loggrant_alloc(xlog_t *log) +{ + log->l_grant_trace = ktrace_alloc(XLOG_TRACE_LOGGRANT_SIZE, KM_NOFS); +} + +void +xlog_trace_loggrant_dealloc(xlog_t *log) +{ + ktrace_free(log->l_grant_trace); +} + void xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) { unsigned long cnts; - if (!log->l_grant_trace) { - log->l_grant_trace = ktrace_alloc(2048, KM_NOSLEEP); - if (!log->l_grant_trace) - return; - } /* ticket counts are 1 byte each */ cnts = ((unsigned long)tic->t_ocnt) | ((unsigned long)tic->t_cnt) << 8; @@ -157,10 +168,20 @@ xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) } void +xlog_trace_iclog_alloc(xlog_in_core_t *iclog) +{ + iclog->ic_trace = ktrace_alloc(XLOG_TRACE_ICLOG_SIZE, KM_NOFS); +} + +void +xlog_trace_iclog_dealloc(xlog_in_core_t *iclog) +{ + ktrace_free(iclog->ic_trace); +} + +void xlog_trace_iclog(xlog_in_core_t *iclog, uint state) { - if (!iclog->ic_trace) - iclog->ic_trace = ktrace_alloc(256, KM_SLEEP); ktrace_enter(iclog->ic_trace, (void *)((unsigned long)state), (void *)((unsigned long)current_pid()), @@ -170,8 +191,15 @@ xlog_trace_iclog(xlog_in_core_t *iclog, uint state) (void *)NULL, (void *)NULL); } #else + +#define xlog_trace_loggrant_alloc(log) +#define xlog_trace_loggrant_dealloc(log) #define xlog_trace_loggrant(log,tic,string) + +#define xlog_trace_iclog_alloc(iclog) +#define xlog_trace_iclog_dealloc(iclog) #define xlog_trace_iclog(iclog,state) + #endif /* XFS_LOG_TRACE */ @@ -226,20 +254,24 @@ xlog_grant_sub_space(struct log *log, int bytes) static void xlog_grant_add_space_write(struct log *log, int bytes) { - log->l_grant_write_bytes += bytes; - if (log->l_grant_write_bytes > log->l_logsize) { - log->l_grant_write_bytes -= log->l_logsize; + int tmp = log->l_logsize - log->l_grant_write_bytes; + if (tmp > bytes) + log->l_grant_write_bytes += bytes; + else { log->l_grant_write_cycle++; + log->l_grant_write_bytes = bytes - tmp; } } static void xlog_grant_add_space_reserve(struct log *log, int bytes) { - log->l_grant_reserve_bytes += bytes; - if (log->l_grant_reserve_bytes > log->l_logsize) { - log->l_grant_reserve_bytes -= log->l_logsize; + int tmp = log->l_logsize - log->l_grant_reserve_bytes; + if (tmp > bytes) + log->l_grant_reserve_bytes += bytes; + else { log->l_grant_reserve_cycle++; + log->l_grant_reserve_bytes = bytes - tmp; } } @@ -332,15 +364,12 @@ xfs_log_done(xfs_mount_t *mp, } else { xlog_trace_loggrant(log, ticket, "xfs_log_done: (permanent)"); xlog_regrant_reserve_log_space(log, ticket); - } - - /* If this ticket was a permanent reservation and we aren't - * trying to release it, reset the inited flags; so next time - * we write, a start record will be written out. - */ - if ((ticket->t_flags & XLOG_TIC_PERM_RESERV) && - (flags & XFS_LOG_REL_PERM_RESERV) == 0) + /* If this ticket was a permanent reservation and we aren't + * trying to release it, reset the inited flags; so next time + * we write, a start record will be written out. + */ ticket->t_flags |= XLOG_TIC_INITED; + } return lsn; } /* xfs_log_done */ @@ -353,11 +382,11 @@ xfs_log_done(xfs_mount_t *mp, * Asynchronous forces are implemented by setting the WANT_SYNC * bit in the appropriate in-core log and then returning. * - * Synchronous forces are implemented with a semaphore. All callers - * to force a given lsn to disk will wait on a semaphore attached to the + * Synchronous forces are implemented with a signal variable. All callers + * to force a given lsn to disk will wait on a the sv attached to the * specific in-core log. When given in-core log finally completes its * write to disk, that thread will wake up all threads waiting on the - * semaphore. + * sv. */ int _xfs_log_force( @@ -584,12 +613,12 @@ error: * mp - ubiquitous xfs mount point structure */ int -xfs_log_mount_finish(xfs_mount_t *mp, int mfsi_flags) +xfs_log_mount_finish(xfs_mount_t *mp) { int error; if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) - error = xlog_recover_finish(mp->m_log, mfsi_flags); + error = xlog_recover_finish(mp->m_log); else { error = 0; ASSERT(mp->m_flags & XFS_MOUNT_RDONLY); @@ -703,7 +732,7 @@ xfs_log_unmount_write(xfs_mount_t *mp) if (!(iclog->ic_state == XLOG_STATE_ACTIVE || iclog->ic_state == XLOG_STATE_DIRTY)) { if (!XLOG_FORCED_SHUTDOWN(log)) { - sv_wait(&iclog->ic_forcesema, PMEM, + sv_wait(&iclog->ic_force_wait, PMEM, &log->l_icloglock, s); } else { spin_unlock(&log->l_icloglock); @@ -744,7 +773,7 @@ xfs_log_unmount_write(xfs_mount_t *mp) || iclog->ic_state == XLOG_STATE_DIRTY || iclog->ic_state == XLOG_STATE_IOERROR) ) { - sv_wait(&iclog->ic_forcesema, PMEM, + sv_wait(&iclog->ic_force_wait, PMEM, &log->l_icloglock, s); } else { spin_unlock(&log->l_icloglock); @@ -834,7 +863,7 @@ xfs_log_move_tail(xfs_mount_t *mp, break; tail_lsn = 0; free_bytes -= tic->t_unit_res; - sv_signal(&tic->t_sema); + sv_signal(&tic->t_wait); tic = tic->t_next; } while (tic != log->l_write_headq); } @@ -855,7 +884,7 @@ xfs_log_move_tail(xfs_mount_t *mp, break; tail_lsn = 0; free_bytes -= need_bytes; - sv_signal(&tic->t_sema); + sv_signal(&tic->t_wait); tic = tic->t_next; } while (tic != log->l_reserve_headq); } @@ -1004,11 +1033,12 @@ xlog_iodone(xfs_buf_t *bp) l = iclog->ic_log; /* - * If the ordered flag has been removed by a lower - * layer, it means the underlyin device no longer supports + * If the _XFS_BARRIER_FAILED flag was set by a lower + * layer, it means the underlying device no longer supports * barrier I/O. Warn loudly and turn off barriers. */ - if ((l->l_mp->m_flags & XFS_MOUNT_BARRIER) && !XFS_BUF_ORDERED(bp)) { + if (bp->b_flags & _XFS_BARRIER_FAILED) { + bp->b_flags &= ~_XFS_BARRIER_FAILED; l->l_mp->m_flags &= ~XFS_MOUNT_BARRIER; xfs_fs_cmn_err(CE_WARN, l->l_mp, "xlog_iodone: Barriers are no longer supported" @@ -1228,8 +1258,9 @@ xlog_alloc_log(xfs_mount_t *mp, spin_lock_init(&log->l_icloglock); spin_lock_init(&log->l_grant_lock); - initnsema(&log->l_flushsema, 0, "ic-flush"); + sv_init(&log->l_flush_wait, 0, "flush_wait"); + xlog_trace_loggrant_alloc(log); /* log record size must be multiple of BBSIZE; see xlog_rec_header_t */ ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0); @@ -1281,8 +1312,10 @@ xlog_alloc_log(xfs_mount_t *mp, ASSERT(XFS_BUF_ISBUSY(iclog->ic_bp)); ASSERT(XFS_BUF_VALUSEMA(iclog->ic_bp) <= 0); - sv_init(&iclog->ic_forcesema, SV_DEFAULT, "iclog-force"); - sv_init(&iclog->ic_writesema, SV_DEFAULT, "iclog-write"); + sv_init(&iclog->ic_force_wait, SV_DEFAULT, "iclog-force"); + sv_init(&iclog->ic_write_wait, SV_DEFAULT, "iclog-write"); + + xlog_trace_iclog_alloc(iclog); iclogp = &iclog->ic_next; } @@ -1561,33 +1594,21 @@ xlog_dealloc_log(xlog_t *log) iclog = log->l_iclog; for (i=0; i<log->l_iclog_bufs; i++) { - sv_destroy(&iclog->ic_forcesema); - sv_destroy(&iclog->ic_writesema); + sv_destroy(&iclog->ic_force_wait); + sv_destroy(&iclog->ic_write_wait); xfs_buf_free(iclog->ic_bp); -#ifdef XFS_LOG_TRACE - if (iclog->ic_trace != NULL) { - ktrace_free(iclog->ic_trace); - } -#endif + xlog_trace_iclog_dealloc(iclog); next_iclog = iclog->ic_next; - kmem_free(iclog, sizeof(xlog_in_core_t)); + kmem_free(iclog); iclog = next_iclog; } - freesema(&log->l_flushsema); spinlock_destroy(&log->l_icloglock); spinlock_destroy(&log->l_grant_lock); xfs_buf_free(log->l_xbuf); -#ifdef XFS_LOG_TRACE - if (log->l_trace != NULL) { - ktrace_free(log->l_trace); - } - if (log->l_grant_trace != NULL) { - ktrace_free(log->l_grant_trace); - } -#endif + xlog_trace_loggrant_dealloc(log); log->l_mp->m_log = NULL; - kmem_free(log, sizeof(xlog_t)); + kmem_free(log); } /* xlog_dealloc_log */ /* @@ -1973,7 +1994,7 @@ xlog_write(xfs_mount_t * mp, /* Clean iclogs starting from the head. This ordering must be * maintained, so an iclog doesn't become ACTIVE beyond one that * is SYNCING. This is also required to maintain the notion that we use - * a counting semaphore to hold off would be writers to the log when every + * a ordered wait queue to hold off would be writers to the log when every * iclog is trying to sync to disk. * * State Change: DIRTY -> ACTIVE @@ -2097,6 +2118,7 @@ xlog_state_do_callback( int funcdidcallbacks; /* flag: function did callbacks */ int repeats; /* for issuing console warnings if * looping too many times */ + int wake = 0; spin_lock(&log->l_icloglock); first_iclog = iclog = log->l_iclog; @@ -2236,7 +2258,7 @@ xlog_state_do_callback( xlog_state_clean_log(log); /* wake up threads waiting in xfs_log_force() */ - sv_broadcast(&iclog->ic_forcesema); + sv_broadcast(&iclog->ic_force_wait); iclog = iclog->ic_next; } while (first_iclog != iclog); @@ -2278,15 +2300,13 @@ xlog_state_do_callback( } #endif - flushcnt = 0; - if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR)) { - flushcnt = log->l_flushcnt; - log->l_flushcnt = 0; - } + if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR)) + wake = 1; spin_unlock(&log->l_icloglock); - while (flushcnt--) - vsema(&log->l_flushsema); -} /* xlog_state_do_callback */ + + if (wake) + sv_broadcast(&log->l_flush_wait); +} /* @@ -2300,8 +2320,7 @@ xlog_state_do_callback( * the second completion goes through. * * Callbacks could take time, so they are done outside the scope of the - * global state machine log lock. Assume that the calls to cvsema won't - * take a long time. At least we know it won't sleep. + * global state machine log lock. */ STATIC void xlog_state_done_syncing( @@ -2337,7 +2356,7 @@ xlog_state_done_syncing( * iclog buffer, we wake them all, one will get to do the * I/O, the others get to wait for the result. */ - sv_broadcast(&iclog->ic_writesema); + sv_broadcast(&iclog->ic_write_wait); spin_unlock(&log->l_icloglock); xlog_state_do_callback(log, aborted, iclog); /* also cleans log */ } /* xlog_state_done_syncing */ @@ -2345,11 +2364,9 @@ xlog_state_done_syncing( /* * If the head of the in-core log ring is not (ACTIVE or DIRTY), then we must - * sleep. The flush semaphore is set to the number of in-core buffers and - * decremented around disk syncing. Therefore, if all buffers are syncing, - * this semaphore will cause new writes to sleep until a sync completes. - * Otherwise, this code just does p() followed by v(). This approximates - * a sleep/wakeup except we can't race. + * sleep. We wait on the flush queue on the head iclog as that should be + * the first iclog to complete flushing. Hence if all iclogs are syncing, + * we will wait here and all new writes will sleep until a sync completes. * * The in-core logs are used in a circular fashion. They are not used * out-of-order even when an iclog past the head is free. @@ -2384,16 +2401,15 @@ restart: } iclog = log->l_iclog; - if (! (iclog->ic_state == XLOG_STATE_ACTIVE)) { - log->l_flushcnt++; - spin_unlock(&log->l_icloglock); + if (iclog->ic_state != XLOG_STATE_ACTIVE) { xlog_trace_iclog(iclog, XLOG_TRACE_SLEEP_FLUSH); XFS_STATS_INC(xs_log_noiclogs); - /* Ensure that log writes happen */ - psema(&log->l_flushsema, PINOD); + + /* Wait for log writes to have flushed */ + sv_wait(&log->l_flush_wait, 0, &log->l_icloglock, 0); goto restart; } - ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE); + head = &iclog->ic_header; atomic_inc(&iclog->ic_refcnt); /* prevents sync */ @@ -2507,7 +2523,7 @@ xlog_grant_log_space(xlog_t *log, goto error_return; XFS_STATS_INC(xs_sleep_logspace); - sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s); + sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); /* * If we got an error, and the filesystem is shutting down, * we'll catch it down below. So just continue... @@ -2533,7 +2549,7 @@ redo: xlog_trace_loggrant(log, tic, "xlog_grant_log_space: sleep 2"); XFS_STATS_INC(xs_sleep_logspace); - sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s); + sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); if (XLOG_FORCED_SHUTDOWN(log)) { spin_lock(&log->l_grant_lock); @@ -2632,7 +2648,7 @@ xlog_regrant_write_log_space(xlog_t *log, if (free_bytes < ntic->t_unit_res) break; free_bytes -= ntic->t_unit_res; - sv_signal(&ntic->t_sema); + sv_signal(&ntic->t_wait); ntic = ntic->t_next; } while (ntic != log->l_write_headq); @@ -2643,7 +2659,7 @@ xlog_regrant_write_log_space(xlog_t *log, xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: sleep 1"); XFS_STATS_INC(xs_sleep_logspace); - sv_wait(&tic->t_sema, PINOD|PLTWAIT, + sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); /* If we're shutting down, this tic is already @@ -2672,7 +2688,7 @@ redo: if ((tic->t_flags & XLOG_TIC_IN_Q) == 0) xlog_ins_ticketq(&log->l_write_headq, tic); XFS_STATS_INC(xs_sleep_logspace); - sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s); + sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s); /* If we're shutting down, this tic is already off the queue */ if (XLOG_FORCED_SHUTDOWN(log)) { @@ -2915,7 +2931,7 @@ xlog_state_switch_iclogs(xlog_t *log, * 2. the current iclog is drity, and the previous iclog is in the * active or dirty state. * - * We may sleep (call psema) if: + * We may sleep if: * * 1. the current iclog is not in the active nor dirty state. * 2. the current iclog dirty, and the previous iclog is not in the @@ -3012,7 +3028,7 @@ maybe_sleep: return XFS_ERROR(EIO); } XFS_STATS_INC(xs_log_force_sleep); - sv_wait(&iclog->ic_forcesema, PINOD, &log->l_icloglock, s); + sv_wait(&iclog->ic_force_wait, PINOD, &log->l_icloglock, s); /* * No need to grab the log lock here since we're * only deciding whether or not to return EIO @@ -3095,7 +3111,7 @@ try_again: XLOG_STATE_SYNCING))) { ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR)); XFS_STATS_INC(xs_log_force_sleep); - sv_wait(&iclog->ic_prev->ic_writesema, PSWP, + sv_wait(&iclog->ic_prev->ic_write_wait, PSWP, &log->l_icloglock, s); *log_flushed = 1; already_slept = 1; @@ -3115,7 +3131,7 @@ try_again: !(iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))) { /* - * Don't wait on the forcesema if we know that we've + * Don't wait on completion if we know that we've * gotten a log write error. */ if (iclog->ic_state & XLOG_STATE_IOERROR) { @@ -3123,7 +3139,7 @@ try_again: return XFS_ERROR(EIO); } XFS_STATS_INC(xs_log_force_sleep); - sv_wait(&iclog->ic_forcesema, PSWP, &log->l_icloglock, s); + sv_wait(&iclog->ic_force_wait, PSWP, &log->l_icloglock, s); /* * No need to grab the log lock here since we're * only deciding whether or not to return EIO @@ -3179,7 +3195,7 @@ STATIC void xlog_ticket_put(xlog_t *log, xlog_ticket_t *ticket) { - sv_destroy(&ticket->t_sema); + sv_destroy(&ticket->t_wait); kmem_zone_free(xfs_log_ticket_zone, ticket); } /* xlog_ticket_put */ @@ -3269,7 +3285,7 @@ xlog_ticket_get(xlog_t *log, tic->t_trans_type = 0; if (xflags & XFS_LOG_PERM_RESERV) tic->t_flags |= XLOG_TIC_PERM_RESERV; - sv_init(&(tic->t_sema), SV_DEFAULT, "logtick"); + sv_init(&(tic->t_wait), SV_DEFAULT, "logtick"); xlog_tic_reset_res(tic); @@ -3556,14 +3572,14 @@ xfs_log_force_umount( */ if ((tic = log->l_reserve_headq)) { do { - sv_signal(&tic->t_sema); + sv_signal(&tic->t_wait); tic = tic->t_next; } while (tic != log->l_reserve_headq); } if ((tic = log->l_write_headq)) { do { - sv_signal(&tic->t_sema); + sv_signal(&tic->t_wait); tic = tic->t_next; } while (tic != log->l_write_headq); } diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index d1d678e..d47b91f 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -149,7 +149,7 @@ int xfs_log_mount(struct xfs_mount *mp, struct xfs_buftarg *log_target, xfs_daddr_t start_block, int num_bblocks); -int xfs_log_mount_finish(struct xfs_mount *mp, int); +int xfs_log_mount_finish(struct xfs_mount *mp); void xfs_log_move_tail(struct xfs_mount *mp, xfs_lsn_t tail_lsn); int xfs_log_notify(struct xfs_mount *mp, diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 8952a39..e7d8f84 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -241,7 +241,7 @@ typedef struct xlog_res { } xlog_res_t; typedef struct xlog_ticket { - sv_t t_sema; /* sleep on this semaphore : 20 */ + sv_t t_wait; /* ticket wait queue : 20 */ struct xlog_ticket *t_next; /* :4|8 */ struct xlog_ticket *t_prev; /* :4|8 */ xlog_tid_t t_tid; /* transaction identifier : 4 */ @@ -314,7 +314,7 @@ typedef struct xlog_rec_ext_header { * xlog_rec_header_t into the reserved space. * - ic_data follows, so a write to disk can start at the beginning of * the iclog. - * - ic_forcesema is used to implement synchronous forcing of the iclog to disk. + * - ic_forcewait is used to implement synchronous forcing of the iclog to disk. * - ic_next is the pointer to the next iclog in the ring. * - ic_bp is a pointer to the buffer used to write this incore log to disk. * - ic_log is a pointer back to the global log structure. @@ -339,8 +339,8 @@ typedef struct xlog_rec_ext_header { * and move everything else out to subsequent cachelines. */ typedef struct xlog_iclog_fields { - sv_t ic_forcesema; - sv_t ic_writesema; + sv_t ic_force_wait; + sv_t ic_write_wait; struct xlog_in_core *ic_next; struct xlog_in_core *ic_prev; struct xfs_buf *ic_bp; @@ -377,8 +377,8 @@ typedef struct xlog_in_core { /* * Defines to save our code from this glop. */ -#define ic_forcesema hic_fields.ic_forcesema -#define ic_writesema hic_fields.ic_writesema +#define ic_force_wait hic_fields.ic_force_wait +#define ic_write_wait hic_fields.ic_write_wait #define ic_next hic_fields.ic_next #define ic_prev hic_fields.ic_prev #define ic_bp hic_fields.ic_bp @@ -423,10 +423,8 @@ typedef struct log { int l_logBBsize; /* size of log in BB chunks */ /* The following block of fields are changed while holding icloglock */ - sema_t l_flushsema ____cacheline_aligned_in_smp; - /* iclog flushing semaphore */ - int l_flushcnt; /* # of procs waiting on this - * sema */ + sv_t l_flush_wait ____cacheline_aligned_in_smp; + /* waiting for iclog flush */ int l_covered_state;/* state of "covering disk * log entries" */ xlog_in_core_t *l_iclog; /* head log queue */ @@ -450,7 +448,6 @@ typedef struct log { int l_grant_write_bytes; #ifdef XFS_LOG_TRACE - struct ktrace *l_trace; struct ktrace *l_grant_trace; #endif @@ -470,7 +467,7 @@ extern int xlog_find_tail(xlog_t *log, xfs_daddr_t *head_blk, xfs_daddr_t *tail_blk); extern int xlog_recover(xlog_t *log); -extern int xlog_recover_finish(xlog_t *log, int mfsi_flags); +extern int xlog_recover_finish(xlog_t *log); extern void xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int); extern void xlog_recover_process_iunlinks(xlog_t *log); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index e65ab4a..82d46ce 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1715,8 +1715,7 @@ xlog_check_buffer_cancelled( } else { prevp->bc_next = bcp->bc_next; } - kmem_free(bcp, - sizeof(xfs_buf_cancel_t)); + kmem_free(bcp); } } return 1; @@ -2519,7 +2518,7 @@ write_inode_buffer: error: if (need_free) - kmem_free(in_f, sizeof(*in_f)); + kmem_free(in_f); return XFS_ERROR(error); } @@ -2830,16 +2829,14 @@ xlog_recover_free_trans( item = item->ri_next; /* Free the regions in the item. */ for (i = 0; i < free_item->ri_cnt; i++) { - kmem_free(free_item->ri_buf[i].i_addr, - free_item->ri_buf[i].i_len); + kmem_free(free_item->ri_buf[i].i_addr); } /* Free the item itself */ - kmem_free(free_item->ri_buf, - (free_item->ri_total * sizeof(xfs_log_iovec_t))); - kmem_free(free_item, sizeof(xlog_recover_item_t)); + kmem_free(free_item->ri_buf); + kmem_free(free_item); } while (first_item != item); /* Free the transaction recover structure */ - kmem_free(trans, sizeof(xlog_recover_t)); + kmem_free(trans); } STATIC int @@ -3786,8 +3783,7 @@ xlog_do_log_recovery( error = xlog_do_recovery_pass(log, head_blk, tail_blk, XLOG_RECOVER_PASS1); if (error != 0) { - kmem_free(log->l_buf_cancel_table, - XLOG_BC_TABLE_SIZE * sizeof(xfs_buf_cancel_t*)); + kmem_free(log->l_buf_cancel_table); log->l_buf_cancel_table = NULL; return error; } @@ -3806,8 +3802,7 @@ xlog_do_log_recovery( } #endif /* DEBUG */ - kmem_free(log->l_buf_cancel_table, - XLOG_BC_TABLE_SIZE * sizeof(xfs_buf_cancel_t*)); + kmem_free(log->l_buf_cancel_table); log->l_buf_cancel_table = NULL; return error; @@ -3945,8 +3940,7 @@ xlog_recover( */ int xlog_recover_finish( - xlog_t *log, - int mfsi_flags) + xlog_t *log) { /* * Now we're ready to do the transactions needed for the @@ -3974,9 +3968,7 @@ xlog_recover_finish( xfs_log_force(log->l_mp, (xfs_lsn_t)0, (XFS_LOG_FORCE | XFS_LOG_SYNC)); - if ( (mfsi_flags & XFS_MFSI_NOUNLINK) == 0 ) { - xlog_recover_process_iunlinks(log); - } + xlog_recover_process_iunlinks(log); xlog_recover_check_summary(log); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index da39884..a4503f5 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -47,12 +47,10 @@ STATIC int xfs_mount_log_sb(xfs_mount_t *, __int64_t); STATIC int xfs_uuid_mount(xfs_mount_t *); -STATIC void xfs_uuid_unmount(xfs_mount_t *mp); STATIC void xfs_unmountfs_wait(xfs_mount_t *); #ifdef HAVE_PERCPU_SB -STATIC void xfs_icsb_destroy_counters(xfs_mount_t *); STATIC void xfs_icsb_balance_counter(xfs_mount_t *, xfs_sb_field_t, int); STATIC void xfs_icsb_balance_counter_locked(xfs_mount_t *, xfs_sb_field_t, @@ -63,7 +61,6 @@ STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t); #else -#define xfs_icsb_destroy_counters(mp) do { } while (0) #define xfs_icsb_balance_counter(mp, a, b) do { } while (0) #define xfs_icsb_balance_counter_locked(mp, a, b) do { } while (0) #define xfs_icsb_modify_counters(mp, a, b, c) do { } while (0) @@ -126,34 +123,12 @@ static const struct { }; /* - * Return a pointer to an initialized xfs_mount structure. - */ -xfs_mount_t * -xfs_mount_init(void) -{ - xfs_mount_t *mp; - - mp = kmem_zalloc(sizeof(xfs_mount_t), KM_SLEEP); - - if (xfs_icsb_init_counters(mp)) { - mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB; - } - - spin_lock_init(&mp->m_sb_lock); - mutex_init(&mp->m_ilock); - mutex_init(&mp->m_growlock); - atomic_set(&mp->m_active_trans, 0); - - return mp; -} - -/* * Free up the resources associated with a mount structure. Assume that * the structure was initially zeroed, so we can tell which fields got * initialized. */ -void -xfs_mount_free( +STATIC void +xfs_free_perag( xfs_mount_t *mp) { if (mp->m_perag) { @@ -161,28 +136,9 @@ xfs_mount_free( for (agno = 0; agno < mp->m_maxagi; agno++) if (mp->m_perag[agno].pagb_list) - kmem_free(mp->m_perag[agno].pagb_list, - sizeof(xfs_perag_busy_t) * - XFS_PAGB_NUM_SLOTS); - kmem_free(mp->m_perag, - sizeof(xfs_perag_t) * mp->m_sb.sb_agcount); + kmem_free(mp->m_perag[agno].pagb_list); + kmem_free(mp->m_perag); } - - spinlock_destroy(&mp->m_ail_lock); - spinlock_destroy(&mp->m_sb_lock); - mutex_destroy(&mp->m_ilock); - mutex_destroy(&mp->m_growlock); - if (mp->m_quotainfo) - XFS_QM_DONE(mp); - - if (mp->m_fsname != NULL) - kmem_free(mp->m_fsname, mp->m_fsname_len); - if (mp->m_rtname != NULL) - kmem_free(mp->m_rtname, strlen(mp->m_rtname) + 1); - if (mp->m_logname != NULL) - kmem_free(mp->m_logname, strlen(mp->m_logname) + 1); - - xfs_icsb_destroy_counters(mp); } /* @@ -288,6 +244,19 @@ xfs_mount_validate_sb( return XFS_ERROR(EFSCORRUPTED); } + /* + * Until this is fixed only page-sized or smaller data blocks work. + */ + if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) { + xfs_fs_mount_cmn_err(flags, + "file system with blocksize %d bytes", + sbp->sb_blocksize); + xfs_fs_mount_cmn_err(flags, + "only pagesize (%ld) or less will currently work.", + PAGE_SIZE); + return XFS_ERROR(ENOSYS); + } + if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) || xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) { xfs_fs_mount_cmn_err(flags, @@ -309,19 +278,6 @@ xfs_mount_validate_sb( return XFS_ERROR(ENOSYS); } - /* - * Until this is fixed only page-sized or smaller data blocks work. - */ - if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) { - xfs_fs_mount_cmn_err(flags, - "file system with blocksize %d bytes", - sbp->sb_blocksize); - xfs_fs_mount_cmn_err(flags, - "only pagesize (%ld) or less will currently work.", - PAGE_SIZE); - return XFS_ERROR(ENOSYS); - } - return 0; } @@ -734,11 +690,11 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount) * Update alignment values based on mount options and sb values */ STATIC int -xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) +xfs_update_alignment(xfs_mount_t *mp, __uint64_t *update_flags) { xfs_sb_t *sbp = &(mp->m_sb); - if (mp->m_dalign && !(mfsi_flags & XFS_MFSI_SECOND)) { + if (mp->m_dalign) { /* * If stripe unit and stripe width are not multiples * of the fs blocksize turn off alignment. @@ -894,7 +850,7 @@ xfs_set_inoalignment(xfs_mount_t *mp) * Check that the data (and log if separate) are an ok size. */ STATIC int -xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags) +xfs_check_sizes(xfs_mount_t *mp) { xfs_buf_t *bp; xfs_daddr_t d; @@ -917,8 +873,7 @@ xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags) return error; } - if (((mfsi_flags & XFS_MFSI_CLIENT) == 0) && - mp->m_logdev_targp != mp->m_ddev_targp) { + if (mp->m_logdev_targp != mp->m_ddev_targp) { d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks); if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) { cmn_err(CE_WARN, "XFS: size check 3 failed"); @@ -953,15 +908,13 @@ xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags) */ int xfs_mountfs( - xfs_mount_t *mp, - int mfsi_flags) + xfs_mount_t *mp) { xfs_sb_t *sbp = &(mp->m_sb); xfs_inode_t *rip; __uint64_t resblks; __int64_t update_flags = 0LL; uint quotamount, quotaflags; - int agno; int uuid_mounted = 0; int error = 0; @@ -994,9 +947,19 @@ xfs_mountfs( * Re-check for ATTR2 in case it was found in bad_features2 * slot. */ - if (xfs_sb_version_hasattr2(&mp->m_sb)) + if (xfs_sb_version_hasattr2(&mp->m_sb) && + !(mp->m_flags & XFS_MOUNT_NOATTR2)) mp->m_flags |= XFS_MOUNT_ATTR2; + } + + if (xfs_sb_version_hasattr2(&mp->m_sb) && + (mp->m_flags & XFS_MOUNT_NOATTR2)) { + xfs_sb_version_removeattr2(&mp->m_sb); + update_flags |= XFS_SB_FEATURES2; + /* update sb_versionnum for the clearing of the morebits */ + if (!sbp->sb_features2) + update_flags |= XFS_SB_VERSIONNUM; } /* @@ -1005,7 +968,7 @@ xfs_mountfs( * allocator alignment is within an ag, therefore ag has * to be aligned at stripe boundary. */ - error = xfs_update_alignment(mp, mfsi_flags, &update_flags); + error = xfs_update_alignment(mp, &update_flags); if (error) goto error1; @@ -1024,8 +987,7 @@ xfs_mountfs( * since a single partition filesystem is identical to a single * partition volume/filesystem. */ - if ((mfsi_flags & XFS_MFSI_SECOND) == 0 && - (mp->m_flags & XFS_MOUNT_NOUUID) == 0) { + if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0) { if (xfs_uuid_mount(mp)) { error = XFS_ERROR(EINVAL); goto error1; @@ -1053,7 +1015,7 @@ xfs_mountfs( /* * Check that the data (and log if separate) are an ok size. */ - error = xfs_check_sizes(mp, mfsi_flags); + error = xfs_check_sizes(mp); if (error) goto error1; @@ -1067,13 +1029,6 @@ xfs_mountfs( } /* - * For client case we are done now - */ - if (mfsi_flags & XFS_MFSI_CLIENT) { - return 0; - } - - /* * Copies the low order bits of the timestamp and the randomly * set "sequence" number out of a UUID. */ @@ -1097,8 +1052,10 @@ xfs_mountfs( * Allocate and initialize the per-ag data. */ init_rwsem(&mp->m_peraglock); - mp->m_perag = - kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t), KM_SLEEP); + mp->m_perag = kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t), + KM_MAYFAIL); + if (!mp->m_perag) + goto error1; mp->m_maxagi = xfs_initialize_perag(mp, sbp->sb_agcount); @@ -1210,7 +1167,7 @@ xfs_mountfs( * delayed until after the root and real-time bitmap inodes * were consistently read in. */ - error = xfs_log_mount_finish(mp, mfsi_flags); + error = xfs_log_mount_finish(mp); if (error) { cmn_err(CE_WARN, "XFS: log mount finish failed"); goto error4; @@ -1219,7 +1176,7 @@ xfs_mountfs( /* * Complete the quota initialisation, post-log-replay component. */ - error = XFS_QM_MOUNT(mp, quotamount, quotaflags, mfsi_flags); + error = XFS_QM_MOUNT(mp, quotamount, quotaflags); if (error) goto error4; @@ -1253,31 +1210,25 @@ xfs_mountfs( error3: xfs_log_unmount_dealloc(mp); error2: - for (agno = 0; agno < sbp->sb_agcount; agno++) - if (mp->m_perag[agno].pagb_list) - kmem_free(mp->m_perag[agno].pagb_list, - sizeof(xfs_perag_busy_t) * XFS_PAGB_NUM_SLOTS); - kmem_free(mp->m_perag, sbp->sb_agcount * sizeof(xfs_perag_t)); - mp->m_perag = NULL; - /* FALLTHROUGH */ + xfs_free_perag(mp); error1: if (uuid_mounted) - xfs_uuid_unmount(mp); - xfs_freesb(mp); + uuid_table_remove(&mp->m_sb.sb_uuid); return error; } /* - * xfs_unmountfs - * * This flushes out the inodes,dquots and the superblock, unmounts the * log and makes sure that incore structures are freed. */ -int -xfs_unmountfs(xfs_mount_t *mp, struct cred *cr) +void +xfs_unmountfs( + struct xfs_mount *mp) { - __uint64_t resblks; - int error = 0; + __uint64_t resblks; + int error; + + IRELE(mp->m_rootip); /* * We can potentially deadlock here if we have an inode cluster @@ -1334,32 +1285,20 @@ xfs_unmountfs(xfs_mount_t *mp, struct cred *cr) xfs_unmountfs_wait(mp); /* wait for async bufs */ xfs_log_unmount(mp); /* Done! No more fs ops. */ - xfs_freesb(mp); - /* * All inodes from this mount point should be freed. */ ASSERT(mp->m_inodes == NULL); - xfs_unmountfs_close(mp, cr); if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0) - xfs_uuid_unmount(mp); + uuid_table_remove(&mp->m_sb.sb_uuid); -#if defined(DEBUG) || defined(INDUCE_IO_ERROR) +#if defined(DEBUG) xfs_errortag_clearall(mp, 0); #endif - xfs_mount_free(mp); - return 0; -} - -void -xfs_unmountfs_close(xfs_mount_t *mp, struct cred *cr) -{ - if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) - xfs_free_buftarg(mp->m_logdev_targp, 1); - if (mp->m_rtdev_targp) - xfs_free_buftarg(mp->m_rtdev_targp, 1); - xfs_free_buftarg(mp->m_ddev_targp, 0); + xfs_free_perag(mp); + if (mp->m_quotainfo) + XFS_QM_DONE(mp); } STATIC void @@ -1905,16 +1844,6 @@ xfs_uuid_mount( } /* - * Remove filesystem from the UUID table. - */ -STATIC void -xfs_uuid_unmount( - xfs_mount_t *mp) -{ - uuid_table_remove(&mp->m_sb.sb_uuid); -} - -/* * Used to log changes to the superblock unit and width fields which could * be altered by the mount options, as well as any potential sb_features2 * fixup. Only the first superblock is updated. @@ -1928,7 +1857,8 @@ xfs_mount_log_sb( int error; ASSERT(fields & (XFS_SB_UNIT | XFS_SB_WIDTH | XFS_SB_UUID | - XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2)); + XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2 | + XFS_SB_VERSIONNUM)); tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT); error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0, @@ -2109,7 +2039,7 @@ xfs_icsb_reinit_counters( xfs_icsb_unlock(mp); } -STATIC void +void xfs_icsb_destroy_counters( xfs_mount_t *mp) { diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 63e0693..f3c1024 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -61,6 +61,7 @@ struct xfs_bmap_free; struct xfs_extdelta; struct xfs_swapext; struct xfs_mru_cache; +struct xfs_nameops; /* * Prototypes and functions for the Data Migration subsystem. @@ -113,7 +114,7 @@ struct xfs_dqtrxops; struct xfs_quotainfo; typedef int (*xfs_qminit_t)(struct xfs_mount *, uint *, uint *); -typedef int (*xfs_qmmount_t)(struct xfs_mount *, uint, uint, int); +typedef int (*xfs_qmmount_t)(struct xfs_mount *, uint, uint); typedef int (*xfs_qmunmount_t)(struct xfs_mount *); typedef void (*xfs_qmdone_t)(struct xfs_mount *); typedef void (*xfs_dqrele_t)(struct xfs_dquot *); @@ -157,8 +158,8 @@ typedef struct xfs_qmops { #define XFS_QM_INIT(mp, mnt, fl) \ (*(mp)->m_qm_ops->xfs_qminit)(mp, mnt, fl) -#define XFS_QM_MOUNT(mp, mnt, fl, mfsi_flags) \ - (*(mp)->m_qm_ops->xfs_qmmount)(mp, mnt, fl, mfsi_flags) +#define XFS_QM_MOUNT(mp, mnt, fl) \ + (*(mp)->m_qm_ops->xfs_qmmount)(mp, mnt, fl) #define XFS_QM_UNMOUNT(mp) \ (*(mp)->m_qm_ops->xfs_qmunmount)(mp) #define XFS_QM_DONE(mp) \ @@ -210,12 +211,14 @@ typedef struct xfs_icsb_cnts { extern int xfs_icsb_init_counters(struct xfs_mount *); extern void xfs_icsb_reinit_counters(struct xfs_mount *); +extern void xfs_icsb_destroy_counters(struct xfs_mount *); extern void xfs_icsb_sync_counters(struct xfs_mount *, int); extern void xfs_icsb_sync_counters_locked(struct xfs_mount *, int); #else -#define xfs_icsb_init_counters(mp) (0) -#define xfs_icsb_reinit_counters(mp) do { } while (0) +#define xfs_icsb_init_counters(mp) (0) +#define xfs_icsb_destroy_counters(mp) do { } while (0) +#define xfs_icsb_reinit_counters(mp) do { } while (0) #define xfs_icsb_sync_counters(mp, flags) do { } while (0) #define xfs_icsb_sync_counters_locked(mp, flags) do { } while (0) #endif @@ -313,6 +316,7 @@ typedef struct xfs_mount { __uint8_t m_inode_quiesce;/* call quiesce on new inodes. field governed by m_ilock */ __uint8_t m_sectbb_log; /* sectlog - BBSHIFT */ + const struct xfs_nameops *m_dirnameops; /* vector of dir name ops */ int m_dirblksize; /* directory block sz--bytes */ int m_dirblkfsbs; /* directory block sz--fsbs */ xfs_dablk_t m_dirdatablk; /* blockno of dir data v2 */ @@ -378,6 +382,7 @@ typedef struct xfs_mount { counters */ #define XFS_MOUNT_FILESTREAMS (1ULL << 24) /* enable the filestreams allocator */ +#define XFS_MOUNT_NOATTR2 (1ULL << 25) /* disable use of attr2 format */ /* @@ -437,13 +442,6 @@ void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname, /* * Flags for xfs_mountfs */ -#define XFS_MFSI_SECOND 0x01 /* Secondary mount -- skip stuff */ -#define XFS_MFSI_CLIENT 0x02 /* Is a client -- skip lots of stuff */ -/* XFS_MFSI_RRINODES */ -#define XFS_MFSI_NOUNLINK 0x08 /* Skip unlinked inode processing in */ - /* log recovery */ -#define XFS_MFSI_NO_QUOTACHECK 0x10 /* Skip quotacheck processing */ -/* XFS_MFSI_CONVERT_SUNIT */ #define XFS_MFSI_QUIET 0x40 /* Be silent if mount errors found */ #define XFS_DADDR_TO_AGNO(mp,d) xfs_daddr_to_agno(mp,d) @@ -510,15 +508,12 @@ typedef struct xfs_mod_sb { #define XFS_MOUNT_ILOCK(mp) mutex_lock(&((mp)->m_ilock)) #define XFS_MOUNT_IUNLOCK(mp) mutex_unlock(&((mp)->m_ilock)) -extern xfs_mount_t *xfs_mount_init(void); extern void xfs_mod_sb(xfs_trans_t *, __int64_t); extern int xfs_log_sbcount(xfs_mount_t *, uint); -extern void xfs_mount_free(xfs_mount_t *mp); -extern int xfs_mountfs(xfs_mount_t *mp, int); +extern int xfs_mountfs(xfs_mount_t *mp); extern void xfs_mountfs_check_barriers(xfs_mount_t *mp); -extern int xfs_unmountfs(xfs_mount_t *, struct cred *); -extern void xfs_unmountfs_close(xfs_mount_t *, struct cred *); +extern void xfs_unmountfs(xfs_mount_t *); extern int xfs_unmountfs_writesb(xfs_mount_t *); extern int xfs_unmount_flush(xfs_mount_t *, int); extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int); @@ -544,9 +539,6 @@ extern void xfs_qmops_put(struct xfs_mount *); extern struct xfs_dmops xfs_dmcore_xfs; -extern int xfs_init(void); -extern void xfs_cleanup(void); - #endif /* __KERNEL__ */ #endif /* __XFS_MOUNT_H__ */ diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index a0b2c0a..afee7eb 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -307,15 +307,18 @@ xfs_mru_cache_init(void) xfs_mru_elem_zone = kmem_zone_init(sizeof(xfs_mru_cache_elem_t), "xfs_mru_cache_elem"); if (!xfs_mru_elem_zone) - return ENOMEM; + goto out; xfs_mru_reap_wq = create_singlethread_workqueue("xfs_mru_cache"); - if (!xfs_mru_reap_wq) { - kmem_zone_destroy(xfs_mru_elem_zone); - return ENOMEM; - } + if (!xfs_mru_reap_wq) + goto out_destroy_mru_elem_zone; return 0; + + out_destroy_mru_elem_zone: + kmem_zone_destroy(xfs_mru_elem_zone); + out: + return -ENOMEM; } void @@ -382,9 +385,9 @@ xfs_mru_cache_create( exit: if (err && mru && mru->lists) - kmem_free(mru->lists, mru->grp_count * sizeof(*mru->lists)); + kmem_free(mru->lists); if (err && mru) - kmem_free(mru, sizeof(*mru)); + kmem_free(mru); return err; } @@ -424,8 +427,8 @@ xfs_mru_cache_destroy( xfs_mru_cache_flush(mru); - kmem_free(mru->lists, mru->grp_count * sizeof(*mru->lists)); - kmem_free(mru, sizeof(*mru)); + kmem_free(mru->lists); + kmem_free(mru); } /* diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index d8063e1..d700dac 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -336,22 +336,18 @@ xfs_rename( ASSERT(error != EEXIST); if (error) goto abort_return; - xfs_ichgtime(src_ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - - } else { - /* - * We always want to hit the ctime on the source inode. - * We do it in the if clause above for the 'new_parent && - * src_is_directory' case, and here we get all the other - * cases. This isn't strictly required by the standards - * since the source inode isn't really being changed, - * but old unix file systems did it and some incremental - * backup programs won't work without it. - */ - xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG); } /* + * We always want to hit the ctime on the source inode. + * + * This isn't strictly required by the standards since the source + * inode isn't really being changed, but old unix file systems did + * it and some incremental backup programs won't work without it. + */ + xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG); + + /* * Adjust the link count on src_dp. This is necessary when * renaming a directory, either within one parent when * the target existed, or across two parent directories. diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index a0dc6e5..e2f68de 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -74,18 +74,6 @@ STATIC int xfs_rtmodify_summary(xfs_mount_t *, xfs_trans_t *, int, */ /* - * xfs_lowbit32: get low bit set out of 32-bit argument, -1 if none set. - */ -STATIC int -xfs_lowbit32( - __uint32_t v) -{ - if (v) - return ffs(v) - 1; - return -1; -} - -/* * Allocate space to the bitmap or summary file, and zero it, for growfs. */ STATIC int /* error */ @@ -450,6 +438,7 @@ xfs_rtallocate_extent_near( } bbno = XFS_BITTOBLOCK(mp, bno); i = 0; + ASSERT(minlen != 0); log2len = xfs_highbit32(minlen); /* * Loop over all bitmap blocks (bbno + i is current block). @@ -618,6 +607,8 @@ xfs_rtallocate_extent_size( xfs_suminfo_t sum; /* summary information for extents */ ASSERT(minlen % prod == 0 && maxlen % prod == 0); + ASSERT(maxlen != 0); + /* * Loop over all the levels starting with maxlen. * At each level, look at all the bitmap blocks, to see if there @@ -675,6 +666,9 @@ xfs_rtallocate_extent_size( *rtblock = NULLRTBLOCK; return 0; } + ASSERT(minlen != 0); + ASSERT(maxlen != 0); + /* * Loop over sizes, from maxlen down to minlen. * This time, when we do the allocations, allow smaller ones @@ -1961,6 +1955,7 @@ xfs_growfs_rt( nsbp->sb_blocksize * nsbp->sb_rextsize); nsbp->sb_rextents = nsbp->sb_rblocks; do_div(nsbp->sb_rextents, nsbp->sb_rextsize); + ASSERT(nsbp->sb_rextents != 0); nsbp->sb_rextslog = xfs_highbit32(nsbp->sb_rextents); nrsumlevels = nmp->m_rsumlevels = nsbp->sb_rextslog + 1; nrsumsize = @@ -2062,7 +2057,7 @@ xfs_growfs_rt( /* * Free the fake mp structure. */ - kmem_free(nmp, sizeof(*nmp)); + kmem_free(nmp); return error; } diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c index b0f31c0..3a82576 100644 --- a/fs/xfs/xfs_rw.c +++ b/fs/xfs/xfs_rw.c @@ -314,7 +314,7 @@ xfs_bioerror_relse( * ASYNC buffers. */ XFS_BUF_ERROR(bp, EIO); - XFS_BUF_V_IODONESEMA(bp); + XFS_BUF_FINISH_IOWAIT(bp); } else { xfs_buf_relse(bp); } diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index d904efe..3f8cf15 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -46,10 +46,12 @@ struct xfs_mount; #define XFS_SB_VERSION_SECTORBIT 0x0800 #define XFS_SB_VERSION_EXTFLGBIT 0x1000 #define XFS_SB_VERSION_DIRV2BIT 0x2000 +#define XFS_SB_VERSION_BORGBIT 0x4000 /* ASCII only case-insens. */ #define XFS_SB_VERSION_MOREBITSBIT 0x8000 #define XFS_SB_VERSION_OKSASHFBITS \ (XFS_SB_VERSION_EXTFLGBIT | \ - XFS_SB_VERSION_DIRV2BIT) + XFS_SB_VERSION_DIRV2BIT | \ + XFS_SB_VERSION_BORGBIT) #define XFS_SB_VERSION_OKREALFBITS \ (XFS_SB_VERSION_ATTRBIT | \ XFS_SB_VERSION_NLINKBIT | \ @@ -437,6 +439,12 @@ static inline int xfs_sb_version_hassector(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT); } +static inline int xfs_sb_version_hasasciici(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + (sbp->sb_versionnum & XFS_SB_VERSION_BORGBIT); +} + static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ @@ -473,6 +481,13 @@ static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp) ((sbp)->sb_features2 | XFS_SB_VERSION2_ATTR2BIT))); } +static inline void xfs_sb_version_removeattr2(xfs_sb_t *sbp) +{ + sbp->sb_features2 &= ~XFS_SB_VERSION2_ATTR2BIT; + if (!sbp->sb_features2) + sbp->sb_versionnum &= ~XFS_SB_VERSION_MOREBITSBIT; +} + /* * end of superblock version macros */ diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 1403864..4e1c22a 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -43,6 +43,7 @@ #include "xfs_quota.h" #include "xfs_trans_priv.h" #include "xfs_trans_space.h" +#include "xfs_inode_item.h" STATIC void xfs_trans_apply_sb_deltas(xfs_trans_t *); @@ -253,7 +254,7 @@ _xfs_trans_alloc( tp->t_mountp = mp; tp->t_items_free = XFS_LIC_NUM_SLOTS; tp->t_busy_free = XFS_LBC_NUM_SLOTS; - XFS_LIC_INIT(&(tp->t_items)); + xfs_lic_init(&(tp->t_items)); XFS_LBC_INIT(&(tp->t_busy)); return tp; } @@ -282,7 +283,7 @@ xfs_trans_dup( ntp->t_mountp = tp->t_mountp; ntp->t_items_free = XFS_LIC_NUM_SLOTS; ntp->t_busy_free = XFS_LBC_NUM_SLOTS; - XFS_LIC_INIT(&(ntp->t_items)); + xfs_lic_init(&(ntp->t_items)); XFS_LBC_INIT(&(ntp->t_busy)); ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); @@ -889,7 +890,7 @@ shut_us_down: tp->t_commit_lsn = commit_lsn; if (nvec > XFS_TRANS_LOGVEC_COUNT) { - kmem_free(log_vector, nvec * sizeof(xfs_log_iovec_t)); + kmem_free(log_vector); } /* @@ -1169,7 +1170,7 @@ xfs_trans_cancel( while (licp != NULL) { lidp = licp->lic_descs; for (i = 0; i < licp->lic_unused; i++, lidp++) { - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } @@ -1216,6 +1217,68 @@ xfs_trans_free( kmem_zone_free(xfs_trans_zone, tp); } +/* + * Roll from one trans in the sequence of PERMANENT transactions to + * the next: permanent transactions are only flushed out when + * committed with XFS_TRANS_RELEASE_LOG_RES, but we still want as soon + * as possible to let chunks of it go to the log. So we commit the + * chunk we've been working on and get a new transaction to continue. + */ +int +xfs_trans_roll( + struct xfs_trans **tpp, + struct xfs_inode *dp) +{ + struct xfs_trans *trans; + unsigned int logres, count; + int error; + + /* + * Ensure that the inode is always logged. + */ + trans = *tpp; + xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE); + + /* + * Copy the critical parameters from one trans to the next. + */ + logres = trans->t_log_res; + count = trans->t_log_count; + *tpp = xfs_trans_dup(trans); + + /* + * Commit the current transaction. + * If this commit failed, then it'd just unlock those items that + * are not marked ihold. That also means that a filesystem shutdown + * is in progress. The caller takes the responsibility to cancel + * the duplicate transaction that gets returned. + */ + error = xfs_trans_commit(trans, 0); + if (error) + return (error); + + trans = *tpp; + + /* + * Reserve space in the log for th next transaction. + * This also pushes items in the "AIL", the list of logged items, + * out to disk if they are taking up space at the tail of the log + * that we want to use. This requires that either nothing be locked + * across this call, or that anything that is locked be logged in + * the prior and the next transactions. + */ + error = xfs_trans_reserve(trans, 0, logres, 0, + XFS_TRANS_PERM_LOG_RES, count); + /* + * Ensure that the inode is in the new transaction and locked. + */ + if (error) + return error; + + xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL); + xfs_trans_ihold(trans, dp); + return 0; +} /* * THIS SHOULD BE REWRITTEN TO USE xfs_trans_next_item(). @@ -1253,7 +1316,7 @@ xfs_trans_committed( * Special case the chunk embedded in the transaction. */ licp = &(tp->t_items); - if (!(XFS_LIC_ARE_ALL_FREE(licp))) { + if (!(xfs_lic_are_all_free(licp))) { xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag); } @@ -1262,10 +1325,10 @@ xfs_trans_committed( */ licp = licp->lic_next; while (licp != NULL) { - ASSERT(!XFS_LIC_ARE_ALL_FREE(licp)); + ASSERT(!xfs_lic_are_all_free(licp)); xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag); next_licp = licp->lic_next; - kmem_free(licp, sizeof(xfs_log_item_chunk_t)); + kmem_free(licp); licp = next_licp; } @@ -1325,7 +1388,7 @@ xfs_trans_chunk_committed( lidp = licp->lic_descs; for (i = 0; i < licp->lic_unused; i++, lidp++) { - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 0804207..74c80bd 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -210,62 +210,52 @@ typedef struct xfs_log_item_chunk { * lic_unused to the right value (0 matches all free). The * lic_descs.lid_index values are set up as each desc is allocated. */ -#define XFS_LIC_INIT(cp) xfs_lic_init(cp) static inline void xfs_lic_init(xfs_log_item_chunk_t *cp) { cp->lic_free = XFS_LIC_FREEMASK; } -#define XFS_LIC_INIT_SLOT(cp,slot) xfs_lic_init_slot(cp, slot) static inline void xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot) { cp->lic_descs[slot].lid_index = (unsigned char)(slot); } -#define XFS_LIC_VACANCY(cp) xfs_lic_vacancy(cp) static inline int xfs_lic_vacancy(xfs_log_item_chunk_t *cp) { return cp->lic_free & XFS_LIC_FREEMASK; } -#define XFS_LIC_ALL_FREE(cp) xfs_lic_all_free(cp) static inline void xfs_lic_all_free(xfs_log_item_chunk_t *cp) { cp->lic_free = XFS_LIC_FREEMASK; } -#define XFS_LIC_ARE_ALL_FREE(cp) xfs_lic_are_all_free(cp) static inline int xfs_lic_are_all_free(xfs_log_item_chunk_t *cp) { return ((cp->lic_free & XFS_LIC_FREEMASK) == XFS_LIC_FREEMASK); } -#define XFS_LIC_ISFREE(cp,slot) xfs_lic_isfree(cp,slot) static inline int xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot) { return (cp->lic_free & (1 << slot)); } -#define XFS_LIC_CLAIM(cp,slot) xfs_lic_claim(cp,slot) static inline void xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot) { cp->lic_free &= ~(1 << slot); } -#define XFS_LIC_RELSE(cp,slot) xfs_lic_relse(cp,slot) static inline void xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot) { cp->lic_free |= 1 << slot; } -#define XFS_LIC_SLOT(cp,slot) xfs_lic_slot(cp,slot) static inline xfs_log_item_desc_t * xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot) { return &(cp->lic_descs[slot]); } -#define XFS_LIC_DESC_TO_SLOT(dp) xfs_lic_desc_to_slot(dp) static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp) { return (uint)dp->lid_index; @@ -278,7 +268,6 @@ static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp) * All of this yields the address of the chunk, which is * cast to a chunk pointer. */ -#define XFS_LIC_DESC_TO_CHUNK(dp) xfs_lic_desc_to_chunk(dp) static inline xfs_log_item_chunk_t * xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp) { @@ -986,6 +975,7 @@ int _xfs_trans_commit(xfs_trans_t *, int *); #define xfs_trans_commit(tp, flags) _xfs_trans_commit(tp, flags, NULL) void xfs_trans_cancel(xfs_trans_t *, int); +int xfs_trans_roll(struct xfs_trans **, struct xfs_inode *); int xfs_trans_ail_init(struct xfs_mount *); void xfs_trans_ail_destroy(struct xfs_mount *); void xfs_trans_push_ail(struct xfs_mount *, xfs_lsn_t); diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index cb0c583..4e855b5 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -1021,16 +1021,16 @@ xfs_trans_buf_item_match( bp = NULL; len = BBTOB(len); licp = &tp->t_items; - if (!XFS_LIC_ARE_ALL_FREE(licp)) { + if (!xfs_lic_are_all_free(licp)) { for (i = 0; i < licp->lic_unused; i++) { /* * Skip unoccupied slots. */ - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } - lidp = XFS_LIC_SLOT(licp, i); + lidp = xfs_lic_slot(licp, i); blip = (xfs_buf_log_item_t *)lidp->lid_item; if (blip->bli_item.li_type != XFS_LI_BUF) { continue; @@ -1074,7 +1074,7 @@ xfs_trans_buf_item_match_all( bp = NULL; len = BBTOB(len); for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) { - if (XFS_LIC_ARE_ALL_FREE(licp)) { + if (xfs_lic_are_all_free(licp)) { ASSERT(licp == &tp->t_items); ASSERT(licp->lic_next == NULL); return NULL; @@ -1083,11 +1083,11 @@ xfs_trans_buf_item_match_all( /* * Skip unoccupied slots. */ - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } - lidp = XFS_LIC_SLOT(licp, i); + lidp = xfs_lic_slot(licp, i); blip = (xfs_buf_log_item_t *)lidp->lid_item; if (blip->bli_item.li_type != XFS_LI_BUF) { continue; diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c index 4c70bf5..2a1c0f0 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/xfs_trans_inode.c @@ -291,7 +291,7 @@ xfs_trans_inode_broot_debug( iip = ip->i_itemp; if (iip->ili_root_size != 0) { ASSERT(iip->ili_orig_root != NULL); - kmem_free(iip->ili_orig_root, iip->ili_root_size); + kmem_free(iip->ili_orig_root); iip->ili_root_size = 0; iip->ili_orig_root = NULL; } diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c index 66a09f0..3c666e8 100644 --- a/fs/xfs/xfs_trans_item.c +++ b/fs/xfs/xfs_trans_item.c @@ -53,11 +53,11 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip) * Initialize the chunk, and then * claim the first slot in the newly allocated chunk. */ - XFS_LIC_INIT(licp); - XFS_LIC_CLAIM(licp, 0); + xfs_lic_init(licp); + xfs_lic_claim(licp, 0); licp->lic_unused = 1; - XFS_LIC_INIT_SLOT(licp, 0); - lidp = XFS_LIC_SLOT(licp, 0); + xfs_lic_init_slot(licp, 0); + lidp = xfs_lic_slot(licp, 0); /* * Link in the new chunk and update the free count. @@ -88,14 +88,14 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip) */ licp = &tp->t_items; while (licp != NULL) { - if (XFS_LIC_VACANCY(licp)) { + if (xfs_lic_vacancy(licp)) { if (licp->lic_unused <= XFS_LIC_MAX_SLOT) { i = licp->lic_unused; - ASSERT(XFS_LIC_ISFREE(licp, i)); + ASSERT(xfs_lic_isfree(licp, i)); break; } for (i = 0; i <= XFS_LIC_MAX_SLOT; i++) { - if (XFS_LIC_ISFREE(licp, i)) + if (xfs_lic_isfree(licp, i)) break; } ASSERT(i <= XFS_LIC_MAX_SLOT); @@ -108,12 +108,12 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip) * If we find a free descriptor, claim it, * initialize it, and return it. */ - XFS_LIC_CLAIM(licp, i); + xfs_lic_claim(licp, i); if (licp->lic_unused <= i) { licp->lic_unused = i + 1; - XFS_LIC_INIT_SLOT(licp, i); + xfs_lic_init_slot(licp, i); } - lidp = XFS_LIC_SLOT(licp, i); + lidp = xfs_lic_slot(licp, i); tp->t_items_free--; lidp->lid_item = lip; lidp->lid_flags = 0; @@ -136,9 +136,9 @@ xfs_trans_free_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp) xfs_log_item_chunk_t *licp; xfs_log_item_chunk_t **licpp; - slot = XFS_LIC_DESC_TO_SLOT(lidp); - licp = XFS_LIC_DESC_TO_CHUNK(lidp); - XFS_LIC_RELSE(licp, slot); + slot = xfs_lic_desc_to_slot(lidp); + licp = xfs_lic_desc_to_chunk(lidp); + xfs_lic_relse(licp, slot); lidp->lid_item->li_desc = NULL; tp->t_items_free++; @@ -154,14 +154,14 @@ xfs_trans_free_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp) * Also decrement the transaction structure's count of free items * by the number in a chunk since we are freeing an empty chunk. */ - if (XFS_LIC_ARE_ALL_FREE(licp) && (licp != &(tp->t_items))) { + if (xfs_lic_are_all_free(licp) && (licp != &(tp->t_items))) { licpp = &(tp->t_items.lic_next); while (*licpp != licp) { ASSERT(*licpp != NULL); licpp = &((*licpp)->lic_next); } *licpp = licp->lic_next; - kmem_free(licp, sizeof(xfs_log_item_chunk_t)); + kmem_free(licp); tp->t_items_free -= XFS_LIC_NUM_SLOTS; } } @@ -207,20 +207,20 @@ xfs_trans_first_item(xfs_trans_t *tp) /* * If it's not in the first chunk, skip to the second. */ - if (XFS_LIC_ARE_ALL_FREE(licp)) { + if (xfs_lic_are_all_free(licp)) { licp = licp->lic_next; } /* * Return the first non-free descriptor in the chunk. */ - ASSERT(!XFS_LIC_ARE_ALL_FREE(licp)); + ASSERT(!xfs_lic_are_all_free(licp)); for (i = 0; i < licp->lic_unused; i++) { - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } - return XFS_LIC_SLOT(licp, i); + return xfs_lic_slot(licp, i); } cmn_err(CE_WARN, "xfs_trans_first_item() -- no first item"); return NULL; @@ -242,18 +242,18 @@ xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp) xfs_log_item_chunk_t *licp; int i; - licp = XFS_LIC_DESC_TO_CHUNK(lidp); + licp = xfs_lic_desc_to_chunk(lidp); /* * First search the rest of the chunk. The for loop keeps us * from referencing things beyond the end of the chunk. */ - for (i = (int)XFS_LIC_DESC_TO_SLOT(lidp) + 1; i < licp->lic_unused; i++) { - if (XFS_LIC_ISFREE(licp, i)) { + for (i = (int)xfs_lic_desc_to_slot(lidp) + 1; i < licp->lic_unused; i++) { + if (xfs_lic_isfree(licp, i)) { continue; } - return XFS_LIC_SLOT(licp, i); + return xfs_lic_slot(licp, i); } /* @@ -266,13 +266,13 @@ xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp) } licp = licp->lic_next; - ASSERT(!XFS_LIC_ARE_ALL_FREE(licp)); + ASSERT(!xfs_lic_are_all_free(licp)); for (i = 0; i < licp->lic_unused; i++) { - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } - return XFS_LIC_SLOT(licp, i); + return xfs_lic_slot(licp, i); } ASSERT(0); /* NOTREACHED */ @@ -300,9 +300,9 @@ xfs_trans_free_items( /* * Special case the embedded chunk so we don't free it below. */ - if (!XFS_LIC_ARE_ALL_FREE(licp)) { + if (!xfs_lic_are_all_free(licp)) { (void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN); - XFS_LIC_ALL_FREE(licp); + xfs_lic_all_free(licp); licp->lic_unused = 0; } licp = licp->lic_next; @@ -311,10 +311,10 @@ xfs_trans_free_items( * Unlock each item in each chunk and free the chunks. */ while (licp != NULL) { - ASSERT(!XFS_LIC_ARE_ALL_FREE(licp)); + ASSERT(!xfs_lic_are_all_free(licp)); (void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN); next_licp = licp->lic_next; - kmem_free(licp, sizeof(xfs_log_item_chunk_t)); + kmem_free(licp); licp = next_licp; } @@ -347,7 +347,7 @@ xfs_trans_unlock_items(xfs_trans_t *tp, xfs_lsn_t commit_lsn) /* * Special case the embedded chunk so we don't free. */ - if (!XFS_LIC_ARE_ALL_FREE(licp)) { + if (!xfs_lic_are_all_free(licp)) { freed = xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn); } licpp = &(tp->t_items.lic_next); @@ -358,12 +358,12 @@ xfs_trans_unlock_items(xfs_trans_t *tp, xfs_lsn_t commit_lsn) * and free empty chunks. */ while (licp != NULL) { - ASSERT(!XFS_LIC_ARE_ALL_FREE(licp)); + ASSERT(!xfs_lic_are_all_free(licp)); freed += xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn); next_licp = licp->lic_next; - if (XFS_LIC_ARE_ALL_FREE(licp)) { + if (xfs_lic_are_all_free(licp)) { *licpp = next_licp; - kmem_free(licp, sizeof(xfs_log_item_chunk_t)); + kmem_free(licp); freed -= XFS_LIC_NUM_SLOTS; } else { licpp = &(licp->lic_next); @@ -402,7 +402,7 @@ xfs_trans_unlock_chunk( freed = 0; lidp = licp->lic_descs; for (i = 0; i < licp->lic_unused; i++, lidp++) { - if (XFS_LIC_ISFREE(licp, i)) { + if (xfs_lic_isfree(licp, i)) { continue; } lip = lidp->lid_item; @@ -421,7 +421,7 @@ xfs_trans_unlock_chunk( */ if (!(freeing_chunk) && (!(lidp->lid_flags & XFS_LID_DIRTY) || abort)) { - XFS_LIC_RELSE(licp, i); + xfs_lic_relse(licp, i); freed++; } } @@ -530,7 +530,7 @@ xfs_trans_free_busy(xfs_trans_t *tp) lbcp = tp->t_busy.lbc_next; while (lbcp != NULL) { lbcq = lbcp->lbc_next; - kmem_free(lbcp, sizeof(xfs_log_busy_chunk_t)); + kmem_free(lbcp); lbcp = lbcq; } diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 98e5f11..35d4d41 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -237,7 +237,7 @@ xfs_droplink( ASSERT (ip->i_d.di_nlink > 0); ip->i_d.di_nlink--; - drop_nlink(ip->i_vnode); + drop_nlink(VFS_I(ip)); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = 0; @@ -301,7 +301,7 @@ xfs_bumplink( ASSERT(ip->i_d.di_nlink > 0); ip->i_d.di_nlink++; - inc_nlink(ip->i_vnode); + inc_nlink(VFS_I(ip)); if ((ip->i_d.di_version == XFS_DINODE_VERSION_1) && (ip->i_d.di_nlink > XFS_MAXLINK_1)) { /* diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index f316cb8..ef32122 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -18,9 +18,6 @@ #ifndef __XFS_UTILS_H__ #define __XFS_UTILS_H__ -#define IRELE(ip) VN_RELE(XFS_ITOV(ip)) -#define IHOLD(ip) VN_HOLD(XFS_ITOV(ip)) - extern int xfs_truncate_file(xfs_mount_t *, xfs_inode_t *); extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t, xfs_dev_t, cred_t *, prid_t, int, diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 30bacd8..439dd39 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -58,586 +58,6 @@ #include "xfs_utils.h" -int __init -xfs_init(void) -{ -#ifdef XFS_DABUF_DEBUG - extern spinlock_t xfs_dabuf_global_lock; - spin_lock_init(&xfs_dabuf_global_lock); -#endif - - /* - * Initialize all of the zone allocators we use. - */ - xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t), - "xfs_log_ticket"); - xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t), - "xfs_bmap_free_item"); - xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t), - "xfs_btree_cur"); - xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t), - "xfs_da_state"); - xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf"); - xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork"); - xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans"); - xfs_acl_zone_init(xfs_acl_zone, "xfs_acl"); - xfs_mru_cache_init(); - xfs_filestream_init(); - - /* - * The size of the zone allocated buf log item is the maximum - * size possible under XFS. This wastes a little bit of memory, - * but it is much faster. - */ - xfs_buf_item_zone = - kmem_zone_init((sizeof(xfs_buf_log_item_t) + - (((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) / - NBWORD) * sizeof(int))), - "xfs_buf_item"); - xfs_efd_zone = - kmem_zone_init((sizeof(xfs_efd_log_item_t) + - ((XFS_EFD_MAX_FAST_EXTENTS - 1) * - sizeof(xfs_extent_t))), - "xfs_efd_item"); - xfs_efi_zone = - kmem_zone_init((sizeof(xfs_efi_log_item_t) + - ((XFS_EFI_MAX_FAST_EXTENTS - 1) * - sizeof(xfs_extent_t))), - "xfs_efi_item"); - - /* - * These zones warrant special memory allocator hints - */ - xfs_inode_zone = - kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode", - KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | - KM_ZONE_SPREAD, NULL); - xfs_ili_zone = - kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili", - KM_ZONE_SPREAD, NULL); - - /* - * Allocate global trace buffers. - */ -#ifdef XFS_ALLOC_TRACE - xfs_alloc_trace_buf = ktrace_alloc(XFS_ALLOC_TRACE_SIZE, KM_SLEEP); -#endif -#ifdef XFS_BMAP_TRACE - xfs_bmap_trace_buf = ktrace_alloc(XFS_BMAP_TRACE_SIZE, KM_SLEEP); -#endif -#ifdef XFS_BMBT_TRACE - xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_SLEEP); -#endif -#ifdef XFS_ATTR_TRACE - xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_SLEEP); -#endif -#ifdef XFS_DIR2_TRACE - xfs_dir2_trace_buf = ktrace_alloc(XFS_DIR2_GTRACE_SIZE, KM_SLEEP); -#endif - - xfs_dir_startup(); - -#if (defined(DEBUG) || defined(INDUCE_IO_ERROR)) - xfs_error_test_init(); -#endif /* DEBUG || INDUCE_IO_ERROR */ - - xfs_init_procfs(); - xfs_sysctl_register(); - return 0; -} - -void __exit -xfs_cleanup(void) -{ - extern kmem_zone_t *xfs_inode_zone; - extern kmem_zone_t *xfs_efd_zone; - extern kmem_zone_t *xfs_efi_zone; - - xfs_cleanup_procfs(); - xfs_sysctl_unregister(); - xfs_filestream_uninit(); - xfs_mru_cache_uninit(); - xfs_acl_zone_destroy(xfs_acl_zone); - -#ifdef XFS_DIR2_TRACE - ktrace_free(xfs_dir2_trace_buf); -#endif -#ifdef XFS_ATTR_TRACE - ktrace_free(xfs_attr_trace_buf); -#endif -#ifdef XFS_BMBT_TRACE - ktrace_free(xfs_bmbt_trace_buf); -#endif -#ifdef XFS_BMAP_TRACE - ktrace_free(xfs_bmap_trace_buf); -#endif -#ifdef XFS_ALLOC_TRACE - ktrace_free(xfs_alloc_trace_buf); -#endif - - kmem_zone_destroy(xfs_bmap_free_item_zone); - kmem_zone_destroy(xfs_btree_cur_zone); - kmem_zone_destroy(xfs_inode_zone); - kmem_zone_destroy(xfs_trans_zone); - kmem_zone_destroy(xfs_da_state_zone); - kmem_zone_destroy(xfs_dabuf_zone); - kmem_zone_destroy(xfs_buf_item_zone); - kmem_zone_destroy(xfs_efd_zone); - kmem_zone_destroy(xfs_efi_zone); - kmem_zone_destroy(xfs_ifork_zone); - kmem_zone_destroy(xfs_ili_zone); - kmem_zone_destroy(xfs_log_ticket_zone); -} - -/* - * xfs_start_flags - * - * This function fills in xfs_mount_t fields based on mount args. - * Note: the superblock has _not_ yet been read in. - */ -STATIC int -xfs_start_flags( - struct xfs_mount_args *ap, - struct xfs_mount *mp) -{ - /* Values are in BBs */ - if ((ap->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) { - /* - * At this point the superblock has not been read - * in, therefore we do not know the block size. - * Before the mount call ends we will convert - * these to FSBs. - */ - mp->m_dalign = ap->sunit; - mp->m_swidth = ap->swidth; - } - - if (ap->logbufs != -1 && - ap->logbufs != 0 && - (ap->logbufs < XLOG_MIN_ICLOGS || - ap->logbufs > XLOG_MAX_ICLOGS)) { - cmn_err(CE_WARN, - "XFS: invalid logbufs value: %d [not %d-%d]", - ap->logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS); - return XFS_ERROR(EINVAL); - } - mp->m_logbufs = ap->logbufs; - if (ap->logbufsize != -1 && - ap->logbufsize != 0 && - (ap->logbufsize < XLOG_MIN_RECORD_BSIZE || - ap->logbufsize > XLOG_MAX_RECORD_BSIZE || - !is_power_of_2(ap->logbufsize))) { - cmn_err(CE_WARN, - "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]", - ap->logbufsize); - return XFS_ERROR(EINVAL); - } - mp->m_logbsize = ap->logbufsize; - mp->m_fsname_len = strlen(ap->fsname) + 1; - mp->m_fsname = kmem_alloc(mp->m_fsname_len, KM_SLEEP); - strcpy(mp->m_fsname, ap->fsname); - if (ap->rtname[0]) { - mp->m_rtname = kmem_alloc(strlen(ap->rtname) + 1, KM_SLEEP); - strcpy(mp->m_rtname, ap->rtname); - } - if (ap->logname[0]) { - mp->m_logname = kmem_alloc(strlen(ap->logname) + 1, KM_SLEEP); - strcpy(mp->m_logname, ap->logname); - } - - if (ap->flags & XFSMNT_WSYNC) - mp->m_flags |= XFS_MOUNT_WSYNC; -#if XFS_BIG_INUMS - if (ap->flags & XFSMNT_INO64) { - mp->m_flags |= XFS_MOUNT_INO64; - mp->m_inoadd = XFS_INO64_OFFSET; - } -#endif - if (ap->flags & XFSMNT_RETERR) - mp->m_flags |= XFS_MOUNT_RETERR; - if (ap->flags & XFSMNT_NOALIGN) - mp->m_flags |= XFS_MOUNT_NOALIGN; - if (ap->flags & XFSMNT_SWALLOC) - mp->m_flags |= XFS_MOUNT_SWALLOC; - if (ap->flags & XFSMNT_OSYNCISOSYNC) - mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC; - if (ap->flags & XFSMNT_32BITINODES) - mp->m_flags |= XFS_MOUNT_32BITINODES; - - if (ap->flags & XFSMNT_IOSIZE) { - if (ap->iosizelog > XFS_MAX_IO_LOG || - ap->iosizelog < XFS_MIN_IO_LOG) { - cmn_err(CE_WARN, - "XFS: invalid log iosize: %d [not %d-%d]", - ap->iosizelog, XFS_MIN_IO_LOG, - XFS_MAX_IO_LOG); - return XFS_ERROR(EINVAL); - } - - mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE; - mp->m_readio_log = mp->m_writeio_log = ap->iosizelog; - } - - if (ap->flags & XFSMNT_IKEEP) - mp->m_flags |= XFS_MOUNT_IKEEP; - if (ap->flags & XFSMNT_DIRSYNC) - mp->m_flags |= XFS_MOUNT_DIRSYNC; - if (ap->flags & XFSMNT_ATTR2) - mp->m_flags |= XFS_MOUNT_ATTR2; - - if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE) - mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; - - /* - * no recovery flag requires a read-only mount - */ - if (ap->flags & XFSMNT_NORECOVERY) { - if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { - cmn_err(CE_WARN, - "XFS: tried to mount a FS read-write without recovery!"); - return XFS_ERROR(EINVAL); - } - mp->m_flags |= XFS_MOUNT_NORECOVERY; - } - - if (ap->flags & XFSMNT_NOUUID) - mp->m_flags |= XFS_MOUNT_NOUUID; - if (ap->flags & XFSMNT_BARRIER) - mp->m_flags |= XFS_MOUNT_BARRIER; - else - mp->m_flags &= ~XFS_MOUNT_BARRIER; - - if (ap->flags2 & XFSMNT2_FILESTREAMS) - mp->m_flags |= XFS_MOUNT_FILESTREAMS; - - if (ap->flags & XFSMNT_DMAPI) - mp->m_flags |= XFS_MOUNT_DMAPI; - return 0; -} - -/* - * This function fills in xfs_mount_t fields based on mount args. - * Note: the superblock _has_ now been read in. - */ -STATIC int -xfs_finish_flags( - struct xfs_mount_args *ap, - struct xfs_mount *mp) -{ - int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); - - /* Fail a mount where the logbuf is smaller then the log stripe */ - if (xfs_sb_version_haslogv2(&mp->m_sb)) { - if ((ap->logbufsize <= 0) && - (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) { - mp->m_logbsize = mp->m_sb.sb_logsunit; - } else if (ap->logbufsize > 0 && - ap->logbufsize < mp->m_sb.sb_logsunit) { - cmn_err(CE_WARN, - "XFS: logbuf size must be greater than or equal to log stripe size"); - return XFS_ERROR(EINVAL); - } - } else { - /* Fail a mount if the logbuf is larger than 32K */ - if (ap->logbufsize > XLOG_BIG_RECORD_BSIZE) { - cmn_err(CE_WARN, - "XFS: logbuf size for version 1 logs must be 16K or 32K"); - return XFS_ERROR(EINVAL); - } - } - - if (xfs_sb_version_hasattr2(&mp->m_sb)) - mp->m_flags |= XFS_MOUNT_ATTR2; - - /* - * prohibit r/w mounts of read-only filesystems - */ - if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) { - cmn_err(CE_WARN, - "XFS: cannot mount a read-only filesystem as read-write"); - return XFS_ERROR(EROFS); - } - - /* - * check for shared mount. - */ - if (ap->flags & XFSMNT_SHARED) { - if (!xfs_sb_version_hasshared(&mp->m_sb)) - return XFS_ERROR(EINVAL); - - /* - * For IRIX 6.5, shared mounts must have the shared - * version bit set, have the persistent readonly - * field set, must be version 0 and can only be mounted - * read-only. - */ - if (!ronly || !(mp->m_sb.sb_flags & XFS_SBF_READONLY) || - (mp->m_sb.sb_shared_vn != 0)) - return XFS_ERROR(EINVAL); - - mp->m_flags |= XFS_MOUNT_SHARED; - - /* - * Shared XFS V0 can't deal with DMI. Return EINVAL. - */ - if (mp->m_sb.sb_shared_vn == 0 && (ap->flags & XFSMNT_DMAPI)) - return XFS_ERROR(EINVAL); - } - - if (ap->flags & XFSMNT_UQUOTA) { - mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE); - if (ap->flags & XFSMNT_UQUOTAENF) - mp->m_qflags |= XFS_UQUOTA_ENFD; - } - - if (ap->flags & XFSMNT_GQUOTA) { - mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE); - if (ap->flags & XFSMNT_GQUOTAENF) - mp->m_qflags |= XFS_OQUOTA_ENFD; - } else if (ap->flags & XFSMNT_PQUOTA) { - mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE); - if (ap->flags & XFSMNT_PQUOTAENF) - mp->m_qflags |= XFS_OQUOTA_ENFD; - } - - return 0; -} - -/* - * xfs_mount - * - * The file system configurations are: - * (1) device (partition) with data and internal log - * (2) logical volume with data and log subvolumes. - * (3) logical volume with data, log, and realtime subvolumes. - * - * We only have to handle opening the log and realtime volumes here if - * they are present. The data subvolume has already been opened by - * get_sb_bdev() and is stored in vfsp->vfs_super->s_bdev. - */ -int -xfs_mount( - struct xfs_mount *mp, - struct xfs_mount_args *args, - cred_t *credp) -{ - struct block_device *ddev, *logdev, *rtdev; - int flags = 0, error; - - ddev = mp->m_super->s_bdev; - logdev = rtdev = NULL; - - error = xfs_dmops_get(mp, args); - if (error) - return error; - error = xfs_qmops_get(mp, args); - if (error) - return error; - - if (args->flags & XFSMNT_QUIET) - flags |= XFS_MFSI_QUIET; - - /* - * Open real time and log devices - order is important. - */ - if (args->logname[0]) { - error = xfs_blkdev_get(mp, args->logname, &logdev); - if (error) - return error; - } - if (args->rtname[0]) { - error = xfs_blkdev_get(mp, args->rtname, &rtdev); - if (error) { - xfs_blkdev_put(logdev); - return error; - } - - if (rtdev == ddev || rtdev == logdev) { - cmn_err(CE_WARN, - "XFS: Cannot mount filesystem with identical rtdev and ddev/logdev."); - xfs_blkdev_put(logdev); - xfs_blkdev_put(rtdev); - return EINVAL; - } - } - - /* - * Setup xfs_mount buffer target pointers - */ - error = ENOMEM; - mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0); - if (!mp->m_ddev_targp) { - xfs_blkdev_put(logdev); - xfs_blkdev_put(rtdev); - return error; - } - if (rtdev) { - mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1); - if (!mp->m_rtdev_targp) { - xfs_blkdev_put(logdev); - xfs_blkdev_put(rtdev); - goto error0; - } - } - mp->m_logdev_targp = (logdev && logdev != ddev) ? - xfs_alloc_buftarg(logdev, 1) : mp->m_ddev_targp; - if (!mp->m_logdev_targp) { - xfs_blkdev_put(logdev); - xfs_blkdev_put(rtdev); - goto error0; - } - - /* - * Setup flags based on mount(2) options and then the superblock - */ - error = xfs_start_flags(args, mp); - if (error) - goto error1; - error = xfs_readsb(mp, flags); - if (error) - goto error1; - error = xfs_finish_flags(args, mp); - if (error) - goto error2; - - /* - * Setup xfs_mount buffer target pointers based on superblock - */ - error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize, - mp->m_sb.sb_sectsize); - if (!error && logdev && logdev != ddev) { - unsigned int log_sector_size = BBSIZE; - - if (xfs_sb_version_hassector(&mp->m_sb)) - log_sector_size = mp->m_sb.sb_logsectsize; - error = xfs_setsize_buftarg(mp->m_logdev_targp, - mp->m_sb.sb_blocksize, - log_sector_size); - } - if (!error && rtdev) - error = xfs_setsize_buftarg(mp->m_rtdev_targp, - mp->m_sb.sb_blocksize, - mp->m_sb.sb_sectsize); - if (error) - goto error2; - - if (mp->m_flags & XFS_MOUNT_BARRIER) - xfs_mountfs_check_barriers(mp); - - if ((error = xfs_filestream_mount(mp))) - goto error2; - - error = xfs_mountfs(mp, flags); - if (error) - goto error2; - - XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, args->mtpt, args->fsname); - - return 0; - -error2: - if (mp->m_sb_bp) - xfs_freesb(mp); -error1: - xfs_binval(mp->m_ddev_targp); - if (logdev && logdev != ddev) - xfs_binval(mp->m_logdev_targp); - if (rtdev) - xfs_binval(mp->m_rtdev_targp); -error0: - xfs_unmountfs_close(mp, credp); - xfs_qmops_put(mp); - xfs_dmops_put(mp); - return error; -} - -int -xfs_unmount( - xfs_mount_t *mp, - int flags, - cred_t *credp) -{ - xfs_inode_t *rip; - bhv_vnode_t *rvp; - int unmount_event_wanted = 0; - int unmount_event_flags = 0; - int xfs_unmountfs_needed = 0; - int error; - - rip = mp->m_rootip; - rvp = XFS_ITOV(rip); - -#ifdef HAVE_DMAPI - if (mp->m_flags & XFS_MOUNT_DMAPI) { - error = XFS_SEND_PREUNMOUNT(mp, - rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL, - NULL, NULL, 0, 0, - (mp->m_dmevmask & (1<<DM_EVENT_PREUNMOUNT))? - 0:DM_FLAGS_UNWANTED); - if (error) - return XFS_ERROR(error); - unmount_event_wanted = 1; - unmount_event_flags = (mp->m_dmevmask & (1<<DM_EVENT_UNMOUNT))? - 0 : DM_FLAGS_UNWANTED; - } -#endif - - /* - * Blow away any referenced inode in the filestreams cache. - * This can and will cause log traffic as inodes go inactive - * here. - */ - xfs_filestream_unmount(mp); - - XFS_bflush(mp->m_ddev_targp); - error = xfs_unmount_flush(mp, 0); - if (error) - goto out; - - ASSERT(vn_count(rvp) == 1); - - /* - * Drop the reference count - */ - IRELE(rip); - - /* - * If we're forcing a shutdown, typically because of a media error, - * we want to make sure we invalidate dirty pages that belong to - * referenced vnodes as well. - */ - if (XFS_FORCED_SHUTDOWN(mp)) { - error = xfs_sync(mp, SYNC_WAIT | SYNC_CLOSE); - ASSERT(error != EFSCORRUPTED); - } - xfs_unmountfs_needed = 1; - -out: - /* Send DMAPI event, if required. - * Then do xfs_unmountfs() if needed. - * Then return error (or zero). - */ - if (unmount_event_wanted) { - /* Note: mp structure must still exist for - * XFS_SEND_UNMOUNT() call. - */ - XFS_SEND_UNMOUNT(mp, error == 0 ? rip : NULL, - DM_RIGHT_NULL, 0, error, unmount_event_flags); - } - if (xfs_unmountfs_needed) { - /* - * Call common unmount function to flush to disk - * and free the super block buffer & mount structures. - */ - xfs_unmountfs(mp, credp); - xfs_qmops_put(mp); - xfs_dmops_put(mp); - kmem_free(mp, sizeof(xfs_mount_t)); - } - - return XFS_ERROR(error); -} - STATIC void xfs_quiesce_fs( xfs_mount_t *mp) @@ -694,30 +114,6 @@ xfs_attr_quiesce( xfs_unmountfs_writesb(mp); } -int -xfs_mntupdate( - struct xfs_mount *mp, - int *flags, - struct xfs_mount_args *args) -{ - if (!(*flags & MS_RDONLY)) { /* rw/ro -> rw */ - if (mp->m_flags & XFS_MOUNT_RDONLY) - mp->m_flags &= ~XFS_MOUNT_RDONLY; - if (args->flags & XFSMNT_BARRIER) { - mp->m_flags |= XFS_MOUNT_BARRIER; - xfs_mountfs_check_barriers(mp); - } else { - mp->m_flags &= ~XFS_MOUNT_BARRIER; - } - } else if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { /* rw -> ro */ - xfs_filestream_flush(mp); - xfs_sync(mp, SYNC_DATA_QUIESCE); - xfs_attr_quiesce(mp); - mp->m_flags |= XFS_MOUNT_RDONLY; - } - return 0; -} - /* * xfs_unmount_flush implements a set of flush operation on special * inodes, which are needed as a separate set of operations so that @@ -732,7 +128,6 @@ xfs_unmount_flush( xfs_inode_t *rip = mp->m_rootip; xfs_inode_t *rbmip; xfs_inode_t *rsumip = NULL; - bhv_vnode_t *rvp = XFS_ITOV(rip); int error; xfs_ilock(rip, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); @@ -750,7 +145,7 @@ xfs_unmount_flush( if (error == EFSCORRUPTED) goto fscorrupt_out; - ASSERT(vn_count(XFS_ITOV(rbmip)) == 1); + ASSERT(vn_count(VFS_I(rbmip)) == 1); rsumip = mp->m_rsumip; xfs_ilock(rsumip, XFS_ILOCK_EXCL); @@ -761,7 +156,7 @@ xfs_unmount_flush( if (error == EFSCORRUPTED) goto fscorrupt_out; - ASSERT(vn_count(XFS_ITOV(rsumip)) == 1); + ASSERT(vn_count(VFS_I(rsumip)) == 1); } /* @@ -771,7 +166,7 @@ xfs_unmount_flush( if (error == EFSCORRUPTED) goto fscorrupt_out2; - if (vn_count(rvp) != 1 && !relocation) { + if (vn_count(VFS_I(rip)) != 1 && !relocation) { xfs_iunlock(rip, XFS_ILOCK_EXCL); return XFS_ERROR(EBUSY); } @@ -888,7 +283,7 @@ xfs_sync_inodes( int *bypassed) { xfs_inode_t *ip = NULL; - bhv_vnode_t *vp = NULL; + struct inode *vp = NULL; int error; int last_error; uint64_t fflag; @@ -1008,7 +403,7 @@ xfs_sync_inodes( continue; } - vp = XFS_ITOV_NULL(ip); + vp = VFS_I(ip); /* * If the vnode is gone then this is being torn down, @@ -1048,7 +443,7 @@ xfs_sync_inodes( if (XFS_FORCED_SHUTDOWN(mp) && !(flags & SYNC_CLOSE)) { XFS_MOUNT_IUNLOCK(mp); - kmem_free(ipointer, sizeof(xfs_iptr_t)); + kmem_free(ipointer); return 0; } @@ -1083,7 +478,7 @@ xfs_sync_inodes( IPOINTER_INSERT(ip, mp); xfs_ilock(ip, lock_flags); - ASSERT(vp == XFS_ITOV(ip)); + ASSERT(vp == VFS_I(ip)); ASSERT(ip->i_mount == mp); vnode_refed = B_TRUE; @@ -1194,7 +589,7 @@ xfs_sync_inodes( } XFS_MOUNT_IUNLOCK(mp); ASSERT(ipointer_in == B_FALSE); - kmem_free(ipointer, sizeof(xfs_iptr_t)); + kmem_free(ipointer); return XFS_ERROR(error); } @@ -1224,7 +619,7 @@ xfs_sync_inodes( ASSERT(ipointer_in == B_FALSE); - kmem_free(ipointer, sizeof(xfs_iptr_t)); + kmem_free(ipointer); return XFS_ERROR(last_error); } diff --git a/fs/xfs/xfs_vfsops.h b/fs/xfs/xfs_vfsops.h index 1688817..a74b050 100644 --- a/fs/xfs/xfs_vfsops.h +++ b/fs/xfs/xfs_vfsops.h @@ -8,11 +8,6 @@ struct kstatfs; struct xfs_mount; struct xfs_mount_args; -int xfs_mount(struct xfs_mount *mp, struct xfs_mount_args *args, - struct cred *credp); -int xfs_unmount(struct xfs_mount *mp, int flags, struct cred *credp); -int xfs_mntupdate(struct xfs_mount *mp, int *flags, - struct xfs_mount_args *args); int xfs_sync(struct xfs_mount *mp, int flags); void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname, int lnnum); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index e475e37..8b6812f 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -75,26 +75,23 @@ xfs_open( return 0; } -/* - * xfs_setattr - */ int xfs_setattr( - xfs_inode_t *ip, - bhv_vattr_t *vap, + struct xfs_inode *ip, + struct iattr *iattr, int flags, cred_t *credp) { xfs_mount_t *mp = ip->i_mount; + struct inode *inode = VFS_I(ip); + int mask = iattr->ia_valid; xfs_trans_t *tp; - int mask; int code; uint lock_flags; uint commit_flags=0; uid_t uid=0, iuid=0; gid_t gid=0, igid=0; int timeflags = 0; - xfs_prid_t projid=0, iprojid=0; struct xfs_dquot *udqp, *gdqp, *olddquot1, *olddquot2; int file_owner; int need_iolock = 1; @@ -104,30 +101,9 @@ xfs_setattr( if (mp->m_flags & XFS_MOUNT_RDONLY) return XFS_ERROR(EROFS); - /* - * Cannot set certain attributes. - */ - mask = vap->va_mask; - if (mask & XFS_AT_NOSET) { - return XFS_ERROR(EINVAL); - } - if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); - /* - * Timestamps do not need to be logged and hence do not - * need to be done within a transaction. - */ - if (mask & XFS_AT_UPDTIMES) { - ASSERT((mask & ~XFS_AT_UPDTIMES) == 0); - timeflags = ((mask & XFS_AT_UPDATIME) ? XFS_ICHGTIME_ACC : 0) | - ((mask & XFS_AT_UPDCTIME) ? XFS_ICHGTIME_CHG : 0) | - ((mask & XFS_AT_UPDMTIME) ? XFS_ICHGTIME_MOD : 0); - xfs_ichgtime(ip, timeflags); - return 0; - } - olddquot1 = olddquot2 = NULL; udqp = gdqp = NULL; @@ -139,28 +115,22 @@ xfs_setattr( * If the IDs do change before we take the ilock, we're covered * because the i_*dquot fields will get updated anyway. */ - if (XFS_IS_QUOTA_ON(mp) && - (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID))) { + if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) { uint qflags = 0; - if ((mask & XFS_AT_UID) && XFS_IS_UQUOTA_ON(mp)) { - uid = vap->va_uid; + if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) { + uid = iattr->ia_uid; qflags |= XFS_QMOPT_UQUOTA; } else { uid = ip->i_d.di_uid; } - if ((mask & XFS_AT_GID) && XFS_IS_GQUOTA_ON(mp)) { - gid = vap->va_gid; + if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) { + gid = iattr->ia_gid; qflags |= XFS_QMOPT_GQUOTA; } else { gid = ip->i_d.di_gid; } - if ((mask & XFS_AT_PROJID) && XFS_IS_PQUOTA_ON(mp)) { - projid = vap->va_projid; - qflags |= XFS_QMOPT_PQUOTA; - } else { - projid = ip->i_d.di_projid; - } + /* * We take a reference when we initialize udqp and gdqp, * so it is important that we never blindly double trip on @@ -168,8 +138,8 @@ xfs_setattr( */ ASSERT(udqp == NULL); ASSERT(gdqp == NULL); - code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, projid, qflags, - &udqp, &gdqp); + code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, ip->i_d.di_projid, + qflags, &udqp, &gdqp); if (code) return code; } @@ -180,10 +150,10 @@ xfs_setattr( */ tp = NULL; lock_flags = XFS_ILOCK_EXCL; - if (flags & ATTR_NOLOCK) + if (flags & XFS_ATTR_NOLOCK) need_iolock = 0; - if (!(mask & XFS_AT_SIZE)) { - if ((mask != (XFS_AT_CTIME|XFS_AT_ATIME|XFS_AT_MTIME)) || + if (!(mask & ATTR_SIZE)) { + if ((mask != (ATTR_CTIME|ATTR_ATIME|ATTR_MTIME)) || (mp->m_flags & XFS_MOUNT_WSYNC)) { tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); commit_flags = 0; @@ -196,10 +166,10 @@ xfs_setattr( } } else { if (DM_EVENT_ENABLED(ip, DM_EVENT_TRUNCATE) && - !(flags & ATTR_DMI)) { + !(flags & XFS_ATTR_DMI)) { int dmflags = AT_DELAY_FLAG(flags) | DM_SEM_FLAG_WR; code = XFS_SEND_DATA(mp, DM_EVENT_TRUNCATE, ip, - vap->va_size, 0, dmflags, NULL); + iattr->ia_size, 0, dmflags, NULL); if (code) { lock_flags = 0; goto error_return; @@ -212,16 +182,14 @@ xfs_setattr( xfs_ilock(ip, lock_flags); /* boolean: are we the file owner? */ - file_owner = (current_fsuid(credp) == ip->i_d.di_uid); + file_owner = (current_fsuid() == ip->i_d.di_uid); /* * Change various properties of a file. * Only the owner or users with CAP_FOWNER * capability may do these things. */ - if (mask & - (XFS_AT_MODE|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_UID| - XFS_AT_GID|XFS_AT_PROJID)) { + if (mask & (ATTR_MODE|ATTR_UID|ATTR_GID)) { /* * CAP_FOWNER overrides the following restrictions: * @@ -245,21 +213,21 @@ xfs_setattr( * IDs of the calling process shall match the group owner of * the file when setting the set-group-ID bit on that file */ - if (mask & XFS_AT_MODE) { + if (mask & ATTR_MODE) { mode_t m = 0; - if ((vap->va_mode & S_ISUID) && !file_owner) + if ((iattr->ia_mode & S_ISUID) && !file_owner) m |= S_ISUID; - if ((vap->va_mode & S_ISGID) && + if ((iattr->ia_mode & S_ISGID) && !in_group_p((gid_t)ip->i_d.di_gid)) m |= S_ISGID; #if 0 /* Linux allows this, Irix doesn't. */ - if ((vap->va_mode & S_ISVTX) && !S_ISDIR(ip->i_d.di_mode)) + if ((iattr->ia_mode & S_ISVTX) && !S_ISDIR(ip->i_d.di_mode)) m |= S_ISVTX; #endif if (m && !capable(CAP_FSETID)) - vap->va_mode &= ~m; + iattr->ia_mode &= ~m; } } @@ -270,7 +238,7 @@ xfs_setattr( * and can change the group id only to a group of which he * or she is a member. */ - if (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID)) { + if (mask & (ATTR_UID|ATTR_GID)) { /* * These IDs could have changed since we last looked at them. * But, we're assured that if the ownership did change @@ -278,12 +246,9 @@ xfs_setattr( * would have changed also. */ iuid = ip->i_d.di_uid; - iprojid = ip->i_d.di_projid; igid = ip->i_d.di_gid; - gid = (mask & XFS_AT_GID) ? vap->va_gid : igid; - uid = (mask & XFS_AT_UID) ? vap->va_uid : iuid; - projid = (mask & XFS_AT_PROJID) ? (xfs_prid_t)vap->va_projid : - iprojid; + gid = (mask & ATTR_GID) ? iattr->ia_gid : igid; + uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid; /* * CAP_CHOWN overrides the following restrictions: @@ -303,11 +268,10 @@ xfs_setattr( goto error_return; } /* - * Do a quota reservation only if uid/projid/gid is actually + * Do a quota reservation only if uid/gid is actually * going to change. */ if ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) || - (XFS_IS_PQUOTA_ON(mp) && iprojid != projid) || (XFS_IS_GQUOTA_ON(mp) && igid != gid)) { ASSERT(tp); code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp, @@ -321,13 +285,13 @@ xfs_setattr( /* * Truncate file. Must have write permission and not be a directory. */ - if (mask & XFS_AT_SIZE) { + if (mask & ATTR_SIZE) { /* Short circuit the truncate case for zero length files */ - if ((vap->va_size == 0) && - (ip->i_size == 0) && (ip->i_d.di_nextents == 0)) { + if (iattr->ia_size == 0 && + ip->i_size == 0 && ip->i_d.di_nextents == 0) { xfs_iunlock(ip, XFS_ILOCK_EXCL); lock_flags &= ~XFS_ILOCK_EXCL; - if (mask & XFS_AT_CTIME) + if (mask & ATTR_CTIME) xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); code = 0; goto error_return; @@ -350,9 +314,9 @@ xfs_setattr( /* * Change file access or modified times. */ - if (mask & (XFS_AT_ATIME|XFS_AT_MTIME)) { + if (mask & (ATTR_ATIME|ATTR_MTIME)) { if (!file_owner) { - if ((flags & ATTR_UTIME) && + if ((mask & (ATTR_MTIME_SET|ATTR_ATIME_SET)) && !capable(CAP_FOWNER)) { code = XFS_ERROR(EPERM); goto error_return; @@ -361,90 +325,23 @@ xfs_setattr( } /* - * Change extent size or realtime flag. - */ - if (mask & (XFS_AT_EXTSIZE|XFS_AT_XFLAGS)) { - /* - * Can't change extent size if any extents are allocated. - */ - if (ip->i_d.di_nextents && (mask & XFS_AT_EXTSIZE) && - ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != - vap->va_extsize) ) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } - - /* - * Can't change realtime flag if any extents are allocated. - */ - if ((ip->i_d.di_nextents || ip->i_delayed_blks) && - (mask & XFS_AT_XFLAGS) && - (XFS_IS_REALTIME_INODE(ip)) != - (vap->va_xflags & XFS_XFLAG_REALTIME)) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } - /* - * Extent size must be a multiple of the appropriate block - * size, if set at all. - */ - if ((mask & XFS_AT_EXTSIZE) && vap->va_extsize != 0) { - xfs_extlen_t size; - - if (XFS_IS_REALTIME_INODE(ip) || - ((mask & XFS_AT_XFLAGS) && - (vap->va_xflags & XFS_XFLAG_REALTIME))) { - size = mp->m_sb.sb_rextsize << - mp->m_sb.sb_blocklog; - } else { - size = mp->m_sb.sb_blocksize; - } - if (vap->va_extsize % size) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - /* - * If realtime flag is set then must have realtime data. - */ - if ((mask & XFS_AT_XFLAGS) && - (vap->va_xflags & XFS_XFLAG_REALTIME)) { - if ((mp->m_sb.sb_rblocks == 0) || - (mp->m_sb.sb_rextsize == 0) || - (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - - /* - * Can't modify an immutable/append-only file unless - * we have appropriate permission. - */ - if ((mask & XFS_AT_XFLAGS) && - (ip->i_d.di_flags & - (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) || - (vap->va_xflags & - (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && - !capable(CAP_LINUX_IMMUTABLE)) { - code = XFS_ERROR(EPERM); - goto error_return; - } - } - - /* * Now we can make the changes. Before we join the inode - * to the transaction, if XFS_AT_SIZE is set then take care of + * to the transaction, if ATTR_SIZE is set then take care of * the part of the truncation that must be done without the * inode lock. This needs to be done before joining the inode * to the transaction, because the inode cannot be unlocked * once it is a part of the transaction. */ - if (mask & XFS_AT_SIZE) { + if (mask & ATTR_SIZE) { code = 0; - if ((vap->va_size > ip->i_size) && - (flags & ATTR_NOSIZETOK) == 0) { - code = xfs_igrow_start(ip, vap->va_size, credp); + if (iattr->ia_size > ip->i_size) { + /* + * Do the first part of growing a file: zero any data + * in the last block that is beyond the old EOF. We + * need to do this before the inode is joined to the + * transaction to modify the i_size. + */ + code = xfs_zero_eof(ip, iattr->ia_size, ip->i_size); } xfs_iunlock(ip, XFS_ILOCK_EXCL); @@ -461,10 +358,10 @@ xfs_setattr( * not within the range we care about here. */ if (!code && - (ip->i_size != ip->i_d.di_size) && - (vap->va_size > ip->i_d.di_size)) { + ip->i_size != ip->i_d.di_size && + iattr->ia_size > ip->i_d.di_size) { code = xfs_flush_pages(ip, - ip->i_d.di_size, vap->va_size, + ip->i_d.di_size, iattr->ia_size, XFS_B_ASYNC, FI_NONE); } @@ -472,7 +369,7 @@ xfs_setattr( vn_iowait(ip); if (!code) - code = xfs_itruncate_data(ip, vap->va_size); + code = xfs_itruncate_data(ip, iattr->ia_size); if (code) { ASSERT(tp == NULL); lock_flags &= ~XFS_ILOCK_EXCL; @@ -501,28 +398,30 @@ xfs_setattr( /* * Truncate file. Must have write permission and not be a directory. */ - if (mask & XFS_AT_SIZE) { + if (mask & ATTR_SIZE) { /* * Only change the c/mtime if we are changing the size * or we are explicitly asked to change it. This handles * the semantic difference between truncate() and ftruncate() * as implemented in the VFS. */ - if (vap->va_size != ip->i_size || (mask & XFS_AT_CTIME)) + if (iattr->ia_size != ip->i_size || (mask & ATTR_CTIME)) timeflags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG; - if (vap->va_size > ip->i_size) { - xfs_igrow_finish(tp, ip, vap->va_size, - !(flags & ATTR_DMI)); - } else if ((vap->va_size <= ip->i_size) || - ((vap->va_size == 0) && ip->i_d.di_nextents)) { + if (iattr->ia_size > ip->i_size) { + ip->i_d.di_size = iattr->ia_size; + ip->i_size = iattr->ia_size; + if (!(flags & XFS_ATTR_DMI)) + xfs_ichgtime(ip, XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + } else if (iattr->ia_size <= ip->i_size || + (iattr->ia_size == 0 && ip->i_d.di_nextents)) { /* * signal a sync transaction unless * we're truncating an already unlinked * file on a wsync filesystem */ - code = xfs_itruncate_finish(&tp, ip, - (xfs_fsize_t)vap->va_size, + code = xfs_itruncate_finish(&tp, ip, iattr->ia_size, XFS_DATA_FORK, ((ip->i_d.di_nlink != 0 || !(mp->m_flags & XFS_MOUNT_WSYNC)) @@ -544,9 +443,12 @@ xfs_setattr( /* * Change file access modes. */ - if (mask & XFS_AT_MODE) { + if (mask & ATTR_MODE) { ip->i_d.di_mode &= S_IFMT; - ip->i_d.di_mode |= vap->va_mode & ~S_IFMT; + ip->i_d.di_mode |= iattr->ia_mode & ~S_IFMT; + + inode->i_mode &= S_IFMT; + inode->i_mode |= iattr->ia_mode & ~S_IFMT; xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE); timeflags |= XFS_ICHGTIME_CHG; @@ -559,7 +461,7 @@ xfs_setattr( * and can change the group id only to a group of which he * or she is a member. */ - if (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID)) { + if (mask & (ATTR_UID|ATTR_GID)) { /* * CAP_FSETID overrides the following restrictions: * @@ -577,39 +479,24 @@ xfs_setattr( */ if (iuid != uid) { if (XFS_IS_UQUOTA_ON(mp)) { - ASSERT(mask & XFS_AT_UID); + ASSERT(mask & ATTR_UID); ASSERT(udqp); olddquot1 = XFS_QM_DQVOPCHOWN(mp, tp, ip, &ip->i_udquot, udqp); } ip->i_d.di_uid = uid; + inode->i_uid = uid; } if (igid != gid) { if (XFS_IS_GQUOTA_ON(mp)) { ASSERT(!XFS_IS_PQUOTA_ON(mp)); - ASSERT(mask & XFS_AT_GID); + ASSERT(mask & ATTR_GID); ASSERT(gdqp); olddquot2 = XFS_QM_DQVOPCHOWN(mp, tp, ip, &ip->i_gdquot, gdqp); } ip->i_d.di_gid = gid; - } - if (iprojid != projid) { - if (XFS_IS_PQUOTA_ON(mp)) { - ASSERT(!XFS_IS_GQUOTA_ON(mp)); - ASSERT(mask & XFS_AT_PROJID); - ASSERT(gdqp); - olddquot2 = XFS_QM_DQVOPCHOWN(mp, tp, ip, - &ip->i_gdquot, gdqp); - } - ip->i_d.di_projid = projid; - /* - * We may have to rev the inode as well as - * the superblock version number since projids didn't - * exist before DINODE_VERSION_2 and SB_VERSION_NLINK. - */ - if (ip->i_d.di_version == XFS_DINODE_VERSION_1) - xfs_bump_ino_vers2(tp, ip); + inode->i_gid = gid; } xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE); @@ -620,82 +507,33 @@ xfs_setattr( /* * Change file access or modified times. */ - if (mask & (XFS_AT_ATIME|XFS_AT_MTIME)) { - if (mask & XFS_AT_ATIME) { - ip->i_d.di_atime.t_sec = vap->va_atime.tv_sec; - ip->i_d.di_atime.t_nsec = vap->va_atime.tv_nsec; + if (mask & (ATTR_ATIME|ATTR_MTIME)) { + if (mask & ATTR_ATIME) { + inode->i_atime = iattr->ia_atime; + ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec; + ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec; ip->i_update_core = 1; - timeflags &= ~XFS_ICHGTIME_ACC; } - if (mask & XFS_AT_MTIME) { - ip->i_d.di_mtime.t_sec = vap->va_mtime.tv_sec; - ip->i_d.di_mtime.t_nsec = vap->va_mtime.tv_nsec; + if (mask & ATTR_MTIME) { + inode->i_mtime = iattr->ia_mtime; + ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec; + ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec; timeflags &= ~XFS_ICHGTIME_MOD; timeflags |= XFS_ICHGTIME_CHG; } - if (tp && (flags & ATTR_UTIME)) + if (tp && (mask & (ATTR_MTIME_SET|ATTR_ATIME_SET))) xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE); } /* - * Change XFS-added attributes. - */ - if (mask & (XFS_AT_EXTSIZE|XFS_AT_XFLAGS)) { - if (mask & XFS_AT_EXTSIZE) { - /* - * Converting bytes to fs blocks. - */ - ip->i_d.di_extsize = vap->va_extsize >> - mp->m_sb.sb_blocklog; - } - if (mask & XFS_AT_XFLAGS) { - uint di_flags; - - /* can't set PREALLOC this way, just preserve it */ - di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); - if (vap->va_xflags & XFS_XFLAG_IMMUTABLE) - di_flags |= XFS_DIFLAG_IMMUTABLE; - if (vap->va_xflags & XFS_XFLAG_APPEND) - di_flags |= XFS_DIFLAG_APPEND; - if (vap->va_xflags & XFS_XFLAG_SYNC) - di_flags |= XFS_DIFLAG_SYNC; - if (vap->va_xflags & XFS_XFLAG_NOATIME) - di_flags |= XFS_DIFLAG_NOATIME; - if (vap->va_xflags & XFS_XFLAG_NODUMP) - di_flags |= XFS_DIFLAG_NODUMP; - if (vap->va_xflags & XFS_XFLAG_PROJINHERIT) - di_flags |= XFS_DIFLAG_PROJINHERIT; - if (vap->va_xflags & XFS_XFLAG_NODEFRAG) - di_flags |= XFS_DIFLAG_NODEFRAG; - if (vap->va_xflags & XFS_XFLAG_FILESTREAM) - di_flags |= XFS_DIFLAG_FILESTREAM; - if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) { - if (vap->va_xflags & XFS_XFLAG_RTINHERIT) - di_flags |= XFS_DIFLAG_RTINHERIT; - if (vap->va_xflags & XFS_XFLAG_NOSYMLINKS) - di_flags |= XFS_DIFLAG_NOSYMLINKS; - if (vap->va_xflags & XFS_XFLAG_EXTSZINHERIT) - di_flags |= XFS_DIFLAG_EXTSZINHERIT; - } else if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) { - if (vap->va_xflags & XFS_XFLAG_REALTIME) - di_flags |= XFS_DIFLAG_REALTIME; - if (vap->va_xflags & XFS_XFLAG_EXTSIZE) - di_flags |= XFS_DIFLAG_EXTSIZE; - } - ip->i_d.di_flags = di_flags; - } - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - timeflags |= XFS_ICHGTIME_CHG; - } - - /* - * Change file inode change time only if XFS_AT_CTIME set + * Change file inode change time only if ATTR_CTIME set * AND we have been called by a DMI function. */ - if ( (flags & ATTR_DMI) && (mask & XFS_AT_CTIME) ) { - ip->i_d.di_ctime.t_sec = vap->va_ctime.tv_sec; - ip->i_d.di_ctime.t_nsec = vap->va_ctime.tv_nsec; + if ((flags & XFS_ATTR_DMI) && (mask & ATTR_CTIME)) { + inode->i_ctime = iattr->ia_ctime; + ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; + ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec; ip->i_update_core = 1; timeflags &= ~XFS_ICHGTIME_CHG; } @@ -704,7 +542,7 @@ xfs_setattr( * Send out timestamp changes that need to be set to the * current time. Not done when called by a DMI function. */ - if (timeflags && !(flags & ATTR_DMI)) + if (timeflags && !(flags & XFS_ATTR_DMI)) xfs_ichgtime(ip, timeflags); XFS_STATS_INC(xs_ig_attrchg); @@ -742,7 +580,7 @@ xfs_setattr( } if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE) && - !(flags & ATTR_DMI)) { + !(flags & XFS_ATTR_DMI)) { (void) XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL, NULL, DM_RIGHT_NULL, NULL, NULL, 0, 0, AT_DELAY_FLAG(flags)); @@ -875,7 +713,7 @@ xfs_fsync( return XFS_ERROR(EIO); /* capture size updates in I/O completion before writing the inode. */ - error = filemap_fdatawait(vn_to_inode(XFS_ITOV(ip))->i_mapping); + error = filemap_fdatawait(VFS_I(ip)->i_mapping); if (error) return XFS_ERROR(error); @@ -1321,7 +1159,6 @@ int xfs_release( xfs_inode_t *ip) { - bhv_vnode_t *vp = XFS_ITOV(ip); xfs_mount_t *mp = ip->i_mount; int error; @@ -1356,13 +1193,13 @@ xfs_release( * be exposed to that problem. */ truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED); - if (truncated && VN_DIRTY(vp) && ip->i_delayed_blks > 0) + if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) xfs_flush_pages(ip, 0, -1, XFS_B_ASYNC, FI_NONE); } if (ip->i_d.di_nlink != 0) { if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) && - ((ip->i_size > 0) || (VN_CACHED(vp) > 0 || + ((ip->i_size > 0) || (VN_CACHED(VFS_I(ip)) > 0 || ip->i_delayed_blks > 0)) && (ip->i_df.if_flags & XFS_IFEXTENTS)) && (!(ip->i_d.di_flags & @@ -1388,7 +1225,6 @@ int xfs_inactive( xfs_inode_t *ip) { - bhv_vnode_t *vp = XFS_ITOV(ip); xfs_bmap_free_t free_list; xfs_fsblock_t first_block; int committed; @@ -1403,7 +1239,7 @@ xfs_inactive( * If the inode is already free, then there can be nothing * to clean up here. */ - if (ip->i_d.di_mode == 0 || VN_BAD(vp)) { + if (ip->i_d.di_mode == 0 || VN_BAD(VFS_I(ip))) { ASSERT(ip->i_df.if_real_bytes == 0); ASSERT(ip->i_df.if_broot_bytes == 0); return VN_INACTIVE_CACHE; @@ -1433,7 +1269,7 @@ xfs_inactive( if (ip->i_d.di_nlink != 0) { if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) && - ((ip->i_size > 0) || (VN_CACHED(vp) > 0 || + ((ip->i_size > 0) || (VN_CACHED(VFS_I(ip)) > 0 || ip->i_delayed_blks > 0)) && (ip->i_df.if_flags & XFS_IFEXTENTS) && (!(ip->i_d.di_flags & @@ -1601,12 +1437,18 @@ xfs_inactive( return VN_INACTIVE_CACHE; } - +/* + * Lookups up an inode from "name". If ci_name is not NULL, then a CI match + * is allowed, otherwise it has to be an exact match. If a CI match is found, + * ci_name->name will point to a the actual name (caller must free) or + * will be set to NULL if an exact match is found. + */ int xfs_lookup( xfs_inode_t *dp, struct xfs_name *name, - xfs_inode_t **ipp) + xfs_inode_t **ipp, + struct xfs_name *ci_name) { xfs_ino_t inum; int error; @@ -1618,7 +1460,7 @@ xfs_lookup( return XFS_ERROR(EIO); lock_mode = xfs_ilock_map_shared(dp); - error = xfs_dir_lookup(NULL, dp, name, &inum); + error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name); xfs_iunlock_map_shared(dp, lock_mode); if (error) @@ -1626,12 +1468,15 @@ xfs_lookup( error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp, 0); if (error) - goto out; + goto out_free_name; xfs_itrace_ref(*ipp); return 0; - out: +out_free_name: + if (ci_name) + kmem_free(ci_name->name); +out: *ipp = NULL; return error; } @@ -1688,7 +1533,7 @@ xfs_create( * Make sure that we have allocated dquot(s) on disk. */ error = XFS_QM_DQVOPALLOC(mp, dp, - current_fsuid(credp), current_fsgid(credp), prid, + current_fsuid(), current_fsgid(), prid, XFS_QMOPT_QUOTALL|XFS_QMOPT_INHERIT, &udqp, &gdqp); if (error) goto std_return; @@ -1860,111 +1705,6 @@ std_return: } #ifdef DEBUG -/* - * Some counters to see if (and how often) we are hitting some deadlock - * prevention code paths. - */ - -int xfs_rm_locks; -int xfs_rm_lock_delays; -int xfs_rm_attempts; -#endif - -/* - * The following routine will lock the inodes associated with the - * directory and the named entry in the directory. The locks are - * acquired in increasing inode number. - * - * If the entry is "..", then only the directory is locked. The - * vnode ref count will still include that from the .. entry in - * this case. - * - * There is a deadlock we need to worry about. If the locked directory is - * in the AIL, it might be blocking up the log. The next inode we lock - * could be already locked by another thread waiting for log space (e.g - * a permanent log reservation with a long running transaction (see - * xfs_itruncate_finish)). To solve this, we must check if the directory - * is in the ail and use lock_nowait. If we can't lock, we need to - * drop the inode lock on the directory and try again. xfs_iunlock will - * potentially push the tail if we were holding up the log. - */ -STATIC int -xfs_lock_dir_and_entry( - xfs_inode_t *dp, - xfs_inode_t *ip) /* inode of entry 'name' */ -{ - int attempts; - xfs_ino_t e_inum; - xfs_inode_t *ips[2]; - xfs_log_item_t *lp; - -#ifdef DEBUG - xfs_rm_locks++; -#endif - attempts = 0; - -again: - xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); - - e_inum = ip->i_ino; - - xfs_itrace_ref(ip); - - /* - * We want to lock in increasing inum. Since we've already - * acquired the lock on the directory, we may need to release - * if if the inum of the entry turns out to be less. - */ - if (e_inum > dp->i_ino) { - /* - * We are already in the right order, so just - * lock on the inode of the entry. - * We need to use nowait if dp is in the AIL. - */ - - lp = (xfs_log_item_t *)dp->i_itemp; - if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { - if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) { - attempts++; -#ifdef DEBUG - xfs_rm_attempts++; -#endif - - /* - * Unlock dp and try again. - * xfs_iunlock will try to push the tail - * if the inode is in the AIL. - */ - - xfs_iunlock(dp, XFS_ILOCK_EXCL); - - if ((attempts % 5) == 0) { - delay(1); /* Don't just spin the CPU */ -#ifdef DEBUG - xfs_rm_lock_delays++; -#endif - } - goto again; - } - } else { - xfs_ilock(ip, XFS_ILOCK_EXCL); - } - } else if (e_inum < dp->i_ino) { - xfs_iunlock(dp, XFS_ILOCK_EXCL); - - ips[0] = ip; - ips[1] = dp; - xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); - } - /* else e_inum == dp->i_ino */ - /* This can happen if we're asked to lock /x/.. - * the entry is "..", which is also the parent directory. - */ - - return 0; -} - -#ifdef DEBUG int xfs_locked_n; int xfs_small_retries; int xfs_middle_retries; @@ -2098,12 +1838,52 @@ again: #endif } -#ifdef DEBUG -#define REMOVE_DEBUG_TRACE(x) {remove_which_error_return = (x);} -int remove_which_error_return = 0; -#else /* ! DEBUG */ -#define REMOVE_DEBUG_TRACE(x) -#endif /* ! DEBUG */ +/* + * xfs_lock_two_inodes() can only be used to lock one type of lock + * at a time - the iolock or the ilock, but not both at once. If + * we lock both at once, lockdep will report false positives saying + * we have violated locking orders. + */ +void +xfs_lock_two_inodes( + xfs_inode_t *ip0, + xfs_inode_t *ip1, + uint lock_mode) +{ + xfs_inode_t *temp; + int attempts = 0; + xfs_log_item_t *lp; + + if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL)) + ASSERT((lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) == 0); + ASSERT(ip0->i_ino != ip1->i_ino); + + if (ip0->i_ino > ip1->i_ino) { + temp = ip0; + ip0 = ip1; + ip1 = temp; + } + + again: + xfs_ilock(ip0, xfs_lock_inumorder(lock_mode, 0)); + + /* + * If the first lock we have locked is in the AIL, we must TRY to get + * the second lock. If we can't get it, we must release the first one + * and try again. + */ + lp = (xfs_log_item_t *)ip0->i_itemp; + if (lp && (lp->li_flags & XFS_LI_IN_AIL)) { + if (!xfs_ilock_nowait(ip1, xfs_lock_inumorder(lock_mode, 1))) { + xfs_iunlock(ip0, lock_mode); + if ((++attempts % 5) == 0) + delay(1); /* Don't just spin the CPU */ + goto again; + } + } else { + xfs_ilock(ip1, xfs_lock_inumorder(lock_mode, 1)); + } +} int xfs_remove( @@ -2113,6 +1893,7 @@ xfs_remove( { xfs_mount_t *mp = dp->i_mount; xfs_trans_t *tp = NULL; + int is_dir = S_ISDIR(ip->i_d.di_mode); int error = 0; xfs_bmap_free_t free_list; xfs_fsblock_t first_block; @@ -2120,8 +1901,10 @@ xfs_remove( int committed; int link_zero; uint resblks; + uint log_count; xfs_itrace_entry(dp); + xfs_itrace_entry(ip); if (XFS_FORCED_SHUTDOWN(mp)) return XFS_ERROR(EIO); @@ -2134,19 +1917,23 @@ xfs_remove( return error; } - xfs_itrace_entry(ip); - xfs_itrace_ref(ip); - error = XFS_QM_DQATTACH(mp, dp, 0); - if (!error) - error = XFS_QM_DQATTACH(mp, ip, 0); - if (error) { - REMOVE_DEBUG_TRACE(__LINE__); + if (error) goto std_return; - } - tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE); + error = XFS_QM_DQATTACH(mp, ip, 0); + if (error) + goto std_return; + + if (is_dir) { + tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR); + log_count = XFS_DEFAULT_LOG_COUNT; + } else { + tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE); + log_count = XFS_REMOVE_LOG_COUNT; + } cancel_flags = XFS_TRANS_RELEASE_LOG_RES; + /* * We try to get the real space reservation first, * allowing for directory btree deletion(s) implying @@ -2158,25 +1945,19 @@ xfs_remove( */ resblks = XFS_REMOVE_SPACE_RES(mp); error = xfs_trans_reserve(tp, resblks, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_REMOVE_LOG_COUNT); + XFS_TRANS_PERM_LOG_RES, log_count); if (error == ENOSPC) { resblks = 0; error = xfs_trans_reserve(tp, 0, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_REMOVE_LOG_COUNT); + XFS_TRANS_PERM_LOG_RES, log_count); } if (error) { ASSERT(error != ENOSPC); - REMOVE_DEBUG_TRACE(__LINE__); - xfs_trans_cancel(tp, 0); - return error; + cancel_flags = 0; + goto out_trans_cancel; } - error = xfs_lock_dir_and_entry(dp, ip); - if (error) { - REMOVE_DEBUG_TRACE(__LINE__); - xfs_trans_cancel(tp, cancel_flags); - goto std_return; - } + xfs_lock_two_inodes(dp, ip, XFS_ILOCK_EXCL); /* * At this point, we've gotten both the directory and the entry @@ -2189,46 +1970,83 @@ xfs_remove( xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); /* - * Entry must exist since we did a lookup in xfs_lock_dir_and_entry. + * If we're removing a directory perform some additional validation. */ + if (is_dir) { + ASSERT(ip->i_d.di_nlink >= 2); + if (ip->i_d.di_nlink != 2) { + error = XFS_ERROR(ENOTEMPTY); + goto out_trans_cancel; + } + if (!xfs_dir_isempty(ip)) { + error = XFS_ERROR(ENOTEMPTY); + goto out_trans_cancel; + } + } + XFS_BMAP_INIT(&free_list, &first_block); error = xfs_dir_removename(tp, dp, name, ip->i_ino, &first_block, &free_list, resblks); if (error) { ASSERT(error != ENOENT); - REMOVE_DEBUG_TRACE(__LINE__); - goto error1; + goto out_bmap_cancel; } xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + /* + * Bump the in memory generation count on the parent + * directory so that other can know that it has changed. + */ dp->i_gen++; xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); - error = xfs_droplink(tp, ip); - if (error) { - REMOVE_DEBUG_TRACE(__LINE__); - goto error1; + if (is_dir) { + /* + * Drop the link from ip's "..". + */ + error = xfs_droplink(tp, dp); + if (error) + goto out_bmap_cancel; + + /* + * Drop the link from dp to ip. + */ + error = xfs_droplink(tp, ip); + if (error) + goto out_bmap_cancel; + } else { + /* + * When removing a non-directory we need to log the parent + * inode here for the i_gen update. For a directory this is + * done implicitly by the xfs_droplink call for the ".." entry. + */ + xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE); } - /* Determine if this is the last link while + /* + * Drop the "." link from ip to self. + */ + error = xfs_droplink(tp, ip); + if (error) + goto out_bmap_cancel; + + /* + * Determine if this is the last link while * we are in the transaction. */ - link_zero = (ip)->i_d.di_nlink==0; + link_zero = (ip->i_d.di_nlink == 0); /* * If this is a synchronous mount, make sure that the * remove transaction goes to disk before returning to * the user. */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) xfs_trans_set_sync(tp); - } error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { - REMOVE_DEBUG_TRACE(__LINE__); - goto error_rele; - } + if (error) + goto out_bmap_cancel; error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); if (error) @@ -2240,38 +2058,26 @@ xfs_remove( * will get killed on last close in xfs_close() so we don't * have to worry about that. */ - if (link_zero && xfs_inode_is_filestream(ip)) + if (!is_dir && link_zero && xfs_inode_is_filestream(ip)) xfs_filestream_deassociate(ip); xfs_itrace_exit(ip); + xfs_itrace_exit(dp); -/* Fall through to std_return with error = 0 */ std_return: if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) { - (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, - dp, DM_RIGHT_NULL, - NULL, DM_RIGHT_NULL, - name->name, NULL, ip->i_d.di_mode, error, 0); + XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, dp, DM_RIGHT_NULL, + NULL, DM_RIGHT_NULL, name->name, NULL, + ip->i_d.di_mode, error, 0); } - return error; - error1: - xfs_bmap_cancel(&free_list); - cancel_flags |= XFS_TRANS_ABORT; - xfs_trans_cancel(tp, cancel_flags); - goto std_return; + return error; - error_rele: - /* - * In this case make sure to not release the inode until after - * the current transaction is aborted. Releasing it beforehand - * can cause us to go to xfs_inactive and start a recursive - * transaction which can easily deadlock with the current one. - */ + out_bmap_cancel: xfs_bmap_cancel(&free_list); cancel_flags |= XFS_TRANS_ABORT; + out_trans_cancel: xfs_trans_cancel(tp, cancel_flags); - goto std_return; } @@ -2283,7 +2089,6 @@ xfs_link( { xfs_mount_t *mp = tdp->i_mount; xfs_trans_t *tp; - xfs_inode_t *ips[2]; int error; xfs_bmap_free_t free_list; xfs_fsblock_t first_block; @@ -2331,15 +2136,7 @@ xfs_link( goto error_return; } - if (sip->i_ino < tdp->i_ino) { - ips[0] = sip; - ips[1] = tdp; - } else { - ips[0] = tdp; - ips[1] = sip; - } - - xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL); + xfs_lock_two_inodes(sip, tdp, XFS_ILOCK_EXCL); /* * Increment vnode ref counts since xfs_trans_commit & @@ -2480,7 +2277,7 @@ xfs_mkdir( * Make sure that we have allocated dquot(s) on disk. */ error = XFS_QM_DQVOPALLOC(mp, dp, - current_fsuid(credp), current_fsgid(credp), prid, + current_fsuid(), current_fsgid(), prid, XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp); if (error) goto std_return; @@ -2638,186 +2435,6 @@ std_return: } int -xfs_rmdir( - xfs_inode_t *dp, - struct xfs_name *name, - xfs_inode_t *cdp) -{ - xfs_mount_t *mp = dp->i_mount; - xfs_trans_t *tp; - int error; - xfs_bmap_free_t free_list; - xfs_fsblock_t first_block; - int cancel_flags; - int committed; - int last_cdp_link; - uint resblks; - - xfs_itrace_entry(dp); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - if (DM_EVENT_ENABLED(dp, DM_EVENT_REMOVE)) { - error = XFS_SEND_NAMESP(mp, DM_EVENT_REMOVE, - dp, DM_RIGHT_NULL, - NULL, DM_RIGHT_NULL, name->name, - NULL, cdp->i_d.di_mode, 0, 0); - if (error) - return XFS_ERROR(error); - } - - /* - * Get the dquots for the inodes. - */ - error = XFS_QM_DQATTACH(mp, dp, 0); - if (!error) - error = XFS_QM_DQATTACH(mp, cdp, 0); - if (error) { - REMOVE_DEBUG_TRACE(__LINE__); - goto std_return; - } - - tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR); - cancel_flags = XFS_TRANS_RELEASE_LOG_RES; - /* - * We try to get the real space reservation first, - * allowing for directory btree deletion(s) implying - * possible bmap insert(s). If we can't get the space - * reservation then we use 0 instead, and avoid the bmap - * btree insert(s) in the directory code by, if the bmap - * insert tries to happen, instead trimming the LAST - * block from the directory. - */ - resblks = XFS_REMOVE_SPACE_RES(mp); - error = xfs_trans_reserve(tp, resblks, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_DEFAULT_LOG_COUNT); - if (error == ENOSPC) { - resblks = 0; - error = xfs_trans_reserve(tp, 0, XFS_REMOVE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, XFS_DEFAULT_LOG_COUNT); - } - if (error) { - ASSERT(error != ENOSPC); - cancel_flags = 0; - goto error_return; - } - XFS_BMAP_INIT(&free_list, &first_block); - - /* - * Now lock the child directory inode and the parent directory - * inode in the proper order. This will take care of validating - * that the directory entry for the child directory inode has - * not changed while we were obtaining a log reservation. - */ - error = xfs_lock_dir_and_entry(dp, cdp); - if (error) { - xfs_trans_cancel(tp, cancel_flags); - goto std_return; - } - - IHOLD(dp); - xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL); - - IHOLD(cdp); - xfs_trans_ijoin(tp, cdp, XFS_ILOCK_EXCL); - - ASSERT(cdp->i_d.di_nlink >= 2); - if (cdp->i_d.di_nlink != 2) { - error = XFS_ERROR(ENOTEMPTY); - goto error_return; - } - if (!xfs_dir_isempty(cdp)) { - error = XFS_ERROR(ENOTEMPTY); - goto error_return; - } - - error = xfs_dir_removename(tp, dp, name, cdp->i_ino, - &first_block, &free_list, resblks); - if (error) - goto error1; - - xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - - /* - * Bump the in memory generation count on the parent - * directory so that other can know that it has changed. - */ - dp->i_gen++; - - /* - * Drop the link from cdp's "..". - */ - error = xfs_droplink(tp, dp); - if (error) { - goto error1; - } - - /* - * Drop the link from dp to cdp. - */ - error = xfs_droplink(tp, cdp); - if (error) { - goto error1; - } - - /* - * Drop the "." link from cdp to self. - */ - error = xfs_droplink(tp, cdp); - if (error) { - goto error1; - } - - /* Determine these before committing transaction */ - last_cdp_link = (cdp)->i_d.di_nlink==0; - - /* - * If this is a synchronous mount, make sure that the - * rmdir transaction goes to disk before returning to - * the user. - */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { - xfs_trans_set_sync(tp); - } - - error = xfs_bmap_finish (&tp, &free_list, &committed); - if (error) { - xfs_bmap_cancel(&free_list); - xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT)); - goto std_return; - } - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - if (error) { - goto std_return; - } - - - /* Fall through to std_return with error = 0 or the errno - * from xfs_trans_commit. */ - std_return: - if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) { - (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, - dp, DM_RIGHT_NULL, - NULL, DM_RIGHT_NULL, - name->name, NULL, cdp->i_d.di_mode, - error, 0); - } - return error; - - error1: - xfs_bmap_cancel(&free_list); - cancel_flags |= XFS_TRANS_ABORT; - /* FALLTHROUGH */ - - error_return: - xfs_trans_cancel(tp, cancel_flags); - goto std_return; -} - -int xfs_symlink( xfs_inode_t *dp, struct xfs_name *link_name, @@ -2886,7 +2503,7 @@ xfs_symlink( * Make sure that we have allocated dquot(s) on disk. */ error = XFS_QM_DQVOPALLOC(mp, dp, - current_fsuid(credp), current_fsgid(credp), prid, + current_fsuid(), current_fsgid(), prid, XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp); if (error) goto std_return; @@ -3181,14 +2798,13 @@ int xfs_reclaim( xfs_inode_t *ip) { - bhv_vnode_t *vp = XFS_ITOV(ip); xfs_itrace_entry(ip); - ASSERT(!VN_MAPPED(vp)); + ASSERT(!VN_MAPPED(VFS_I(ip))); /* bad inode, get out here ASAP */ - if (VN_BAD(vp)) { + if (VN_BAD(VFS_I(ip))) { xfs_ireclaim(ip); return 0; } @@ -3225,7 +2841,7 @@ xfs_reclaim( XFS_MOUNT_ILOCK(mp); spin_lock(&ip->i_flags_lock); __xfs_iflags_set(ip, XFS_IRECLAIMABLE); - vn_to_inode(vp)->i_private = NULL; + VFS_I(ip)->i_private = NULL; ip->i_vnode = NULL; spin_unlock(&ip->i_flags_lock); list_add_tail(&ip->i_reclaim, &mp->m_del_inodes); @@ -3241,8 +2857,7 @@ xfs_finish_reclaim( int sync_mode) { xfs_perag_t *pag = xfs_get_perag(ip->i_mount, ip->i_ino); - bhv_vnode_t *vp = XFS_ITOV_NULL(ip); - int error; + struct inode *vp = VFS_I(ip); if (vp && VN_BAD(vp)) goto reclaim; @@ -3285,29 +2900,16 @@ xfs_finish_reclaim( xfs_iflock(ip); } - if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { - if (ip->i_update_core || - ((ip->i_itemp != NULL) && - (ip->i_itemp->ili_format.ilf_fields != 0))) { - error = xfs_iflush(ip, sync_mode); - /* - * If we hit an error, typically because of filesystem - * shutdown, we don't need to let vn_reclaim to know - * because we're gonna reclaim the inode anyway. - */ - if (error) { - xfs_iunlock(ip, XFS_ILOCK_EXCL); - goto reclaim; - } - xfs_iflock(ip); /* synchronize with xfs_iflush_done */ - } - - ASSERT(ip->i_update_core == 0); - ASSERT(ip->i_itemp == NULL || - ip->i_itemp->ili_format.ilf_fields == 0); + /* + * In the case of a forced shutdown we rely on xfs_iflush() to + * wait for the inode to be unpinned before returning an error. + */ + if (xfs_iflush(ip, sync_mode) == 0) { + /* synchronize with xfs_iflush_done */ + xfs_iflock(ip); + xfs_ifunlock(ip); } - xfs_ifunlock(ip); xfs_iunlock(ip, XFS_ILOCK_EXCL); reclaim: @@ -3418,7 +3020,7 @@ xfs_alloc_file_space( /* Generate a DMAPI event if needed. */ if (alloc_type != 0 && offset < ip->i_size && - (attr_flags&ATTR_DMI) == 0 && + (attr_flags & XFS_ATTR_DMI) == 0 && DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) { xfs_off_t end_dmi_offset; @@ -3532,7 +3134,7 @@ retry: allocatesize_fsb -= allocated_fsb; } dmapi_enospc_check: - if (error == ENOSPC && (attr_flags & ATTR_DMI) == 0 && + if (error == ENOSPC && (attr_flags & XFS_ATTR_DMI) == 0 && DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE)) { error = XFS_SEND_NAMESP(mp, DM_EVENT_NOSPACE, ip, DM_RIGHT_NULL, @@ -3558,6 +3160,13 @@ error1: /* Just cancel transaction */ /* * Zero file bytes between startoff and endoff inclusive. * The iolock is held exclusive and no blocks are buffered. + * + * This function is used by xfs_free_file_space() to zero + * partial blocks when the range to free is not block aligned. + * When unreserving space with boundaries that are not block + * aligned we round up the start and round down the end + * boundaries and then use this function to zero the parts of + * the blocks that got dropped during the rounding. */ STATIC int xfs_zero_remaining_bytes( @@ -3574,6 +3183,17 @@ xfs_zero_remaining_bytes( int nimap; int error = 0; + /* + * Avoid doing I/O beyond eof - it's not necessary + * since nothing can read beyond eof. The space will + * be zeroed when the file is extended anyway. + */ + if (startoff >= ip->i_size) + return 0; + + if (endoff > ip->i_size) + endoff = ip->i_size; + bp = xfs_buf_get_noaddr(mp->m_sb.sb_blocksize, XFS_IS_REALTIME_INODE(ip) ? mp->m_rtdev_targp : mp->m_ddev_targp); @@ -3643,7 +3263,6 @@ xfs_free_file_space( xfs_off_t len, int attr_flags) { - bhv_vnode_t *vp; int committed; int done; xfs_off_t end_dmi_offset; @@ -3663,7 +3282,6 @@ xfs_free_file_space( xfs_trans_t *tp; int need_iolock = 1; - vp = XFS_ITOV(ip); mp = ip->i_mount; xfs_itrace_entry(ip); @@ -3679,7 +3297,7 @@ xfs_free_file_space( end_dmi_offset = offset + len; endoffset_fsb = XFS_B_TO_FSBT(mp, end_dmi_offset); - if (offset < ip->i_size && (attr_flags & ATTR_DMI) == 0 && + if (offset < ip->i_size && (attr_flags & XFS_ATTR_DMI) == 0 && DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) { if (end_dmi_offset > ip->i_size) end_dmi_offset = ip->i_size; @@ -3690,7 +3308,7 @@ xfs_free_file_space( return error; } - if (attr_flags & ATTR_NOLOCK) + if (attr_flags & XFS_ATTR_NOLOCK) need_iolock = 0; if (need_iolock) { xfs_ilock(ip, XFS_IOLOCK_EXCL); @@ -3700,7 +3318,7 @@ xfs_free_file_space( rounding = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE); ioffset = offset & ~(rounding - 1); - if (VN_CACHED(vp) != 0) { + if (VN_CACHED(VFS_I(ip)) != 0) { xfs_inval_cached_trace(ip, ioffset, -1, ioffset, -1); error = xfs_flushinval_pages(ip, ioffset, -1, FI_REMAPF_LOCKED); if (error) @@ -3867,7 +3485,7 @@ xfs_change_file_space( xfs_off_t startoffset; xfs_off_t llen; xfs_trans_t *tp; - bhv_vattr_t va; + struct iattr iattr; xfs_itrace_entry(ip); @@ -3941,10 +3559,10 @@ xfs_change_file_space( break; } - va.va_mask = XFS_AT_SIZE; - va.va_size = startoffset; + iattr.ia_valid = ATTR_SIZE; + iattr.ia_size = startoffset; - error = xfs_setattr(ip, &va, attr_flags, credp); + error = xfs_setattr(ip, &iattr, attr_flags, credp); if (error) return error; @@ -3974,7 +3592,7 @@ xfs_change_file_space( xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); xfs_trans_ihold(tp, ip); - if ((attr_flags & ATTR_DMI) == 0) { + if ((attr_flags & XFS_ATTR_DMI) == 0) { ip->i_d.di_mode &= ~S_ISUID; /* diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h index 57335ba..e932a96 100644 --- a/fs/xfs/xfs_vnodeops.h +++ b/fs/xfs/xfs_vnodeops.h @@ -2,9 +2,9 @@ #define _XFS_VNODEOPS_H 1 struct attrlist_cursor_kern; -struct bhv_vattr; struct cred; struct file; +struct iattr; struct inode; struct iovec; struct kiocb; @@ -15,14 +15,18 @@ struct xfs_iomap; int xfs_open(struct xfs_inode *ip); -int xfs_setattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags, +int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags, struct cred *credp); +#define XFS_ATTR_DMI 0x01 /* invocation from a DMI function */ +#define XFS_ATTR_NONBLOCK 0x02 /* return EAGAIN if operation would block */ +#define XFS_ATTR_NOLOCK 0x04 /* Don't grab any conflicting locks */ + int xfs_readlink(struct xfs_inode *ip, char *link); int xfs_fsync(struct xfs_inode *ip); int xfs_release(struct xfs_inode *ip); int xfs_inactive(struct xfs_inode *ip); int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, - struct xfs_inode **ipp); + struct xfs_inode **ipp, struct xfs_name *ci_name); int xfs_create(struct xfs_inode *dp, struct xfs_name *name, mode_t mode, xfs_dev_t rdev, struct xfs_inode **ipp, struct cred *credp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, @@ -31,8 +35,6 @@ int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, struct xfs_name *target_name); int xfs_mkdir(struct xfs_inode *dp, struct xfs_name *dir_name, mode_t mode, struct xfs_inode **ipp, struct cred *credp); -int xfs_rmdir(struct xfs_inode *dp, struct xfs_name *name, - struct xfs_inode *cdp); int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize, xfs_off_t *offset, filldir_t filldir); int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, |
