diff options
Diffstat (limited to 'fs')
65 files changed, 1868 insertions, 530 deletions
@@ -724,10 +724,6 @@ config FAT_FS file system and use GNU tar's M option. GNU tar is a program available for Unix and DOS ("man tar" or "info tar"). - It is now also becoming possible to read and write compressed FAT - file systems; read <file:Documentation/filesystems/fat_cvf.txt> for - details. - The FAT support will enlarge your kernel by about 37 KB. If unsure, say Y. diff --git a/fs/affs/file.c b/fs/affs/file.c index 4aa8079..c879690 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -628,11 +628,7 @@ static int affs_prepare_write_ofs(struct file *file, struct page *page, unsigned return err; } if (to < PAGE_CACHE_SIZE) { - char *kaddr = kmap_atomic(page, KM_USER0); - - memset(kaddr + to, 0, PAGE_CACHE_SIZE - to); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, to, PAGE_CACHE_SIZE - to, KM_USER0); if (size > offset + to) { if (size < offset + PAGE_CACHE_SIZE) tmp = size & ~PAGE_CACHE_MASK; diff --git a/fs/afs/Makefile b/fs/afs/Makefile index cf83e5d..73ce561 100644 --- a/fs/afs/Makefile +++ b/fs/afs/Makefile @@ -22,6 +22,7 @@ kafs-objs := \ vlclient.o \ vlocation.o \ vnode.o \ - volume.o + volume.o \ + write.o obj-$(CONFIG_AFS_FS) := kafs.o diff --git a/fs/afs/afs_fs.h b/fs/afs/afs_fs.h index 89e0d16..2198006 100644 --- a/fs/afs/afs_fs.h +++ b/fs/afs/afs_fs.h @@ -18,6 +18,8 @@ enum AFS_FS_Operations { FSFETCHDATA = 130, /* AFS Fetch file data */ FSFETCHSTATUS = 132, /* AFS Fetch file status */ + FSSTOREDATA = 133, /* AFS Store file data */ + FSSTORESTATUS = 135, /* AFS Store file status */ FSREMOVEFILE = 136, /* AFS Remove a file */ FSCREATEFILE = 137, /* AFS Create a file */ FSRENAME = 138, /* AFS Rename or move a file or directory */ diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 9bdbf36..f64e40f 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -44,7 +44,7 @@ void afs_init_callback_state(struct afs_server *server) while (!RB_EMPTY_ROOT(&server->cb_promises)) { vnode = rb_entry(server->cb_promises.rb_node, struct afs_vnode, cb_promise); - _debug("UNPROMISE { vid=%x vn=%u uq=%u}", + _debug("UNPROMISE { vid=%x:%u uq=%u}", vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); rb_erase(&vnode->cb_promise, &server->cb_promises); vnode->cb_promised = false; @@ -84,11 +84,8 @@ void afs_broken_callback_work(struct work_struct *work) /* if the vnode's data version number changed then its contents * are different */ - if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) { - _debug("zap data {%x:%u}", - vnode->fid.vid, vnode->fid.vnode); - invalidate_remote_inode(&vnode->vfs_inode); - } + if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + afs_zap_data(vnode); } out: diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 0c1e902..2fb3127 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -55,7 +55,8 @@ const struct inode_operations afs_dir_inode_operations = { .rmdir = afs_rmdir, .rename = afs_rename, .permission = afs_permission, - .getattr = afs_inode_getattr, + .getattr = afs_getattr, + .setattr = afs_setattr, }; static struct dentry_operations afs_fs_dentry_operations = { @@ -491,7 +492,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, vnode = AFS_FS_I(dir); - _enter("{%x:%d},%p{%s},", + _enter("{%x:%u},%p{%s},", vnode->fid.vid, vnode->fid.vnode, dentry, dentry->d_name.name); ASSERTCMP(dentry->d_inode, ==, NULL); @@ -731,7 +732,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%o", + _enter("{%x:%u},{%s},%o", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); ret = -ENAMETOOLONG; @@ -796,7 +797,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s}", + _enter("{%x:%u},{%s}", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); ret = -ENAMETOOLONG; @@ -842,7 +843,7 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s}", + _enter("{%x:%u},{%s}", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); ret = -ENAMETOOLONG; @@ -916,7 +917,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, int mode, dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%o,", + _enter("{%x:%u},{%s},%o,", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); ret = -ENAMETOOLONG; @@ -983,7 +984,7 @@ static int afs_link(struct dentry *from, struct inode *dir, vnode = AFS_FS_I(from->d_inode); dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%x:%d},{%s}", + _enter("{%x:%u},{%x:%u},{%s}", vnode->fid.vid, vnode->fid.vnode, dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); @@ -1032,7 +1033,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%s", + _enter("{%x:%u},{%s},%s", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, content); @@ -1104,7 +1105,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, orig_dvnode = AFS_FS_I(old_dir); new_dvnode = AFS_FS_I(new_dir); - _enter("{%x:%d},{%x:%d},{%x:%d},{%s}", + _enter("{%x:%u},{%x:%u},{%x:%u},{%s}", orig_dvnode->fid.vid, orig_dvnode->fid.vnode, vnode->fid.vid, vnode->fid.vnode, new_dvnode->fid.vid, new_dvnode->fid.vnode, diff --git a/fs/afs/file.c b/fs/afs/file.c index ae25649..3e25795 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -15,32 +15,43 @@ #include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> +#include <linux/writeback.h> #include "internal.h" -static int afs_file_readpage(struct file *file, struct page *page); -static void afs_file_invalidatepage(struct page *page, unsigned long offset); -static int afs_file_releasepage(struct page *page, gfp_t gfp_flags); +static int afs_readpage(struct file *file, struct page *page); +static void afs_invalidatepage(struct page *page, unsigned long offset); +static int afs_releasepage(struct page *page, gfp_t gfp_flags); +static int afs_launder_page(struct page *page); const struct file_operations afs_file_operations = { .open = afs_open, .release = afs_release, .llseek = generic_file_llseek, .read = do_sync_read, + .write = do_sync_write, .aio_read = generic_file_aio_read, + .aio_write = afs_file_write, .mmap = generic_file_readonly_mmap, .sendfile = generic_file_sendfile, + .fsync = afs_fsync, }; const struct inode_operations afs_file_inode_operations = { - .getattr = afs_inode_getattr, + .getattr = afs_getattr, + .setattr = afs_setattr, .permission = afs_permission, }; const struct address_space_operations afs_fs_aops = { - .readpage = afs_file_readpage, - .set_page_dirty = __set_page_dirty_nobuffers, - .releasepage = afs_file_releasepage, - .invalidatepage = afs_file_invalidatepage, + .readpage = afs_readpage, + .set_page_dirty = afs_set_page_dirty, + .launder_page = afs_launder_page, + .releasepage = afs_releasepage, + .invalidatepage = afs_invalidatepage, + .prepare_write = afs_prepare_write, + .commit_write = afs_commit_write, + .writepage = afs_writepage, + .writepages = afs_writepages, }; /* @@ -52,7 +63,7 @@ int afs_open(struct inode *inode, struct file *file) struct key *key; int ret; - _enter("{%x:%x},", vnode->fid.vid, vnode->fid.vnode); + _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) { @@ -78,7 +89,7 @@ int afs_release(struct inode *inode, struct file *file) { struct afs_vnode *vnode = AFS_FS_I(inode); - _enter("{%x:%x},", vnode->fid.vid, vnode->fid.vnode); + _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); key_put(file->private_data); _leave(" = 0"); @@ -89,10 +100,10 @@ int afs_release(struct inode *inode, struct file *file) * deal with notification that a page was read from the cache */ #ifdef AFS_CACHING_SUPPORT -static void afs_file_readpage_read_complete(void *cookie_data, - struct page *page, - void *data, - int error) +static void afs_readpage_read_complete(void *cookie_data, + struct page *page, + void *data, + int error) { _enter("%p,%p,%p,%d", cookie_data, page, data, error); @@ -109,10 +120,10 @@ static void afs_file_readpage_read_complete(void *cookie_data, * deal with notification that a page was written to the cache */ #ifdef AFS_CACHING_SUPPORT -static void afs_file_readpage_write_complete(void *cookie_data, - struct page *page, - void *data, - int error) +static void afs_readpage_write_complete(void *cookie_data, + struct page *page, + void *data, + int error) { _enter("%p,%p,%p,%d", cookie_data, page, data, error); @@ -121,9 +132,9 @@ static void afs_file_readpage_write_complete(void *cookie_data, #endif /* - * AFS read page from file (or symlink) + * AFS read page from file, directory or symlink */ -static int afs_file_readpage(struct file *file, struct page *page) +static int afs_readpage(struct file *file, struct page *page) { struct afs_vnode *vnode; struct inode *inode; @@ -219,39 +230,17 @@ error: } /* - * get a page cookie for the specified page - */ -#ifdef AFS_CACHING_SUPPORT -int afs_cache_get_page_cookie(struct page *page, - struct cachefs_page **_page_cookie) -{ - int ret; - - _enter(""); - ret = cachefs_page_get_private(page,_page_cookie, GFP_NOIO); - - _leave(" = %d", ret); - return ret; -} -#endif - -/* * invalidate part or all of a page */ -static void afs_file_invalidatepage(struct page *page, unsigned long offset) +static void afs_invalidatepage(struct page *page, unsigned long offset) { int ret = 1; - _enter("{%lu},%lu", page->index, offset); + kenter("{%lu},%lu", page->index, offset); BUG_ON(!PageLocked(page)); if (PagePrivate(page)) { -#ifdef AFS_CACHING_SUPPORT - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - cachefs_uncache_page(vnode->cache,page); -#endif - /* We release buffers only if the entire page is being * invalidated. * The get_block cached value has been unconditionally @@ -272,25 +261,33 @@ static void afs_file_invalidatepage(struct page *page, unsigned long offset) } /* + * write back a dirty page + */ +static int afs_launder_page(struct page *page) +{ + _enter("{%lu}", page->index); + + return 0; +} + +/* * release a page and cleanup its private data */ -static int afs_file_releasepage(struct page *page, gfp_t gfp_flags) +static int afs_releasepage(struct page *page, gfp_t gfp_flags) { - struct cachefs_page *pageio; + struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); + struct afs_writeback *wb; - _enter("{%lu},%x", page->index, gfp_flags); + _enter("{{%x:%u}[%lu],%lx},%x", + vnode->fid.vid, vnode->fid.vnode, page->index, page->flags, + gfp_flags); if (PagePrivate(page)) { -#ifdef AFS_CACHING_SUPPORT - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - cachefs_uncache_page(vnode->cache, page); -#endif - - pageio = (struct cachefs_page *) page_private(page); + wb = (struct afs_writeback *) page_private(page); + ASSERT(wb != NULL); set_page_private(page, 0); ClearPagePrivate(page); - - kfree(pageio); + afs_put_writeback(wb); } _leave(" = 0"); diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index e54e6c2..025b190 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -33,8 +33,10 @@ static void xdr_decode_AFSFid(const __be32 **_bp, struct afs_fid *fid) */ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, struct afs_file_status *status, - struct afs_vnode *vnode) + struct afs_vnode *vnode, + afs_dataversion_t *store_version) { + afs_dataversion_t expected_version; const __be32 *bp = *_bp; umode_t mode; u64 data_version, size; @@ -101,7 +103,11 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, vnode->vfs_inode.i_atime = vnode->vfs_inode.i_ctime; } - if (status->data_version != data_version) { + expected_version = status->data_version; + if (store_version) + expected_version = *store_version; + + if (expected_version != data_version) { status->data_version = data_version; if (vnode && !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { _debug("vnode modified %llx on {%x:%u}", @@ -110,6 +116,8 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, set_bit(AFS_VNODE_MODIFIED, &vnode->flags); set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); } + } else if (store_version) { + status->data_version = data_version; } } @@ -156,6 +164,44 @@ static void xdr_decode_AFSVolSync(const __be32 **_bp, } /* + * encode the requested attributes into an AFSStoreStatus block + */ +static void xdr_encode_AFS_StoreStatus(__be32 **_bp, struct iattr *attr) +{ + __be32 *bp = *_bp; + u32 mask = 0, mtime = 0, owner = 0, group = 0, mode = 0; + + mask = 0; + if (attr->ia_valid & ATTR_MTIME) { + mask |= AFS_SET_MTIME; + mtime = attr->ia_mtime.tv_sec; + } + + if (attr->ia_valid & ATTR_UID) { + mask |= AFS_SET_OWNER; + owner = attr->ia_uid; + } + + if (attr->ia_valid & ATTR_GID) { + mask |= AFS_SET_GROUP; + group = attr->ia_gid; + } + + if (attr->ia_valid & ATTR_MODE) { + mask |= AFS_SET_MODE; + mode = attr->ia_mode & S_IALLUGO; + } + + *bp++ = htonl(mask); + *bp++ = htonl(mtime); + *bp++ = htonl(owner); + *bp++ = htonl(group); + *bp++ = htonl(mode); + *bp++ = 0; /* segment size */ + *_bp = bp; +} + +/* * deliver reply data to an FS.FetchStatus */ static int afs_deliver_fs_fetch_status(struct afs_call *call, @@ -175,7 +221,7 @@ static int afs_deliver_fs_fetch_status(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack(&bp, vnode); if (call->reply2) xdr_decode_AFSVolSync(&bp, call->reply2); @@ -206,7 +252,7 @@ int afs_fs_fetch_file_status(struct afs_server *server, struct afs_call *call; __be32 *bp; - _enter(",%x,{%x:%d},,", + _enter(",%x,{%x:%u},,", key_serial(key), vnode->fid.vid, vnode->fid.vnode); call = afs_alloc_flat_call(&afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); @@ -265,25 +311,20 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, call->offset = 0; call->unmarshall++; - if (call->count < PAGE_SIZE) { - page = call->reply3; - buffer = kmap_atomic(page, KM_USER0); - memset(buffer + PAGE_SIZE - call->count, 0, - call->count); - kunmap_atomic(buffer, KM_USER0); - } - /* extract the returned data */ case 2: _debug("extract data"); - page = call->reply3; - buffer = kmap_atomic(page, KM_USER0); - ret = afs_extract_data(call, skb, last, buffer, call->count); - kunmap_atomic(buffer, KM_USER0); - switch (ret) { - case 0: break; - case -EAGAIN: return 0; - default: return ret; + if (call->count > 0) { + page = call->reply3; + buffer = kmap_atomic(page, KM_USER0); + ret = afs_extract_data(call, skb, last, buffer, + call->count); + kunmap_atomic(buffer, KM_USER0); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } } call->offset = 0; @@ -300,7 +341,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, } bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack(&bp, vnode); if (call->reply2) xdr_decode_AFSVolSync(&bp, call->reply2); @@ -318,6 +359,14 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, if (!last) return 0; + if (call->count < PAGE_SIZE) { + _debug("clear"); + page = call->reply3; + buffer = kmap_atomic(page, KM_USER0); + memset(buffer + call->count, 0, PAGE_SIZE - call->count); + kunmap_atomic(buffer, KM_USER0); + } + _leave(" = 0 [done]"); return 0; } @@ -476,8 +525,8 @@ static int afs_deliver_fs_create_vnode(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply2); - xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL, NULL); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack_raw(&bp, call->reply4); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ @@ -574,7 +623,7 @@ static int afs_deliver_fs_remove(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -657,8 +706,8 @@ static int afs_deliver_fs_link(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); - xdr_decode_AFSFetchStatus(&bp, &dvnode->status, dvnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + xdr_decode_AFSFetchStatus(&bp, &dvnode->status, dvnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -746,8 +795,8 @@ static int afs_deliver_fs_symlink(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply2); - xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL, NULL); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -852,9 +901,10 @@ static int afs_deliver_fs_rename(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &orig_dvnode->status, orig_dvnode); + xdr_decode_AFSFetchStatus(&bp, &orig_dvnode->status, orig_dvnode, NULL); if (new_dvnode != orig_dvnode) - xdr_decode_AFSFetchStatus(&bp, &new_dvnode->status, new_dvnode); + xdr_decode_AFSFetchStatus(&bp, &new_dvnode->status, new_dvnode, + NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -936,3 +986,262 @@ int afs_fs_rename(struct afs_server *server, return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); } + +/* + * deliver reply data to an FS.StoreData + */ +static int afs_deliver_fs_store_data(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + struct afs_vnode *vnode = call->reply; + const __be32 *bp; + + _enter(",,%u", last); + + afs_transfer_reply(call, skb); + if (!last) { + _leave(" = 0 [more]"); + return 0; + } + + if (call->reply_size != call->reply_max) { + _leave(" = -EBADMSG [%u != %u]", + call->reply_size, call->reply_max); + return -EBADMSG; + } + + /* unmarshall the reply once we've received all of it */ + bp = call->buffer; + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, + &call->store_version); + /* xdr_decode_AFSVolSync(&bp, call->replyX); */ + + afs_pages_written_back(vnode, call); + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.StoreData operation type + */ +static const struct afs_call_type afs_RXFSStoreData = { + .name = "FS.StoreData", + .deliver = afs_deliver_fs_store_data, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * store a set of pages + */ +int afs_fs_store_data(struct afs_server *server, struct afs_writeback *wb, + pgoff_t first, pgoff_t last, + unsigned offset, unsigned to, + const struct afs_wait_mode *wait_mode) +{ + struct afs_vnode *vnode = wb->vnode; + struct afs_call *call; + loff_t size, pos, i_size; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(wb->key), vnode->fid.vid, vnode->fid.vnode); + + size = to - offset; + if (first != last) + size += (loff_t)(last - first) << PAGE_SHIFT; + pos = (loff_t)first << PAGE_SHIFT; + pos += offset; + + i_size = i_size_read(&vnode->vfs_inode); + if (pos + size > i_size) + i_size = size + pos; + + _debug("size %llx, at %llx, i_size %llx", + (unsigned long long) size, (unsigned long long) pos, + (unsigned long long) i_size); + + BUG_ON(i_size > 0xffffffff); // TODO: use 64-bit store + + call = afs_alloc_flat_call(&afs_RXFSStoreData, + (4 + 6 + 3) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->wb = wb; + call->key = wb->key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->mapping = vnode->vfs_inode.i_mapping; + call->first = first; + call->last = last; + call->first_offset = offset; + call->last_to = to; + call->send_pages = true; + call->store_version = vnode->status.data_version + 1; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + *bp++ = 0; /* mask */ + *bp++ = 0; /* mtime */ + *bp++ = 0; /* owner */ + *bp++ = 0; /* group */ + *bp++ = 0; /* unix mode */ + *bp++ = 0; /* segment size */ + + *bp++ = htonl(pos); + *bp++ = htonl(size); + *bp++ = htonl(i_size); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * deliver reply data to an FS.StoreStatus + */ +static int afs_deliver_fs_store_status(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + afs_dataversion_t *store_version; + struct afs_vnode *vnode = call->reply; + const __be32 *bp; + + _enter(",,%u", last); + + afs_transfer_reply(call, skb); + if (!last) { + _leave(" = 0 [more]"); + return 0; + } + + if (call->reply_size != call->reply_max) { + _leave(" = -EBADMSG [%u != %u]", + call->reply_size, call->reply_max); + return -EBADMSG; + } + + /* unmarshall the reply once we've received all of it */ + store_version = NULL; + if (call->operation_ID == FSSTOREDATA) + store_version = &call->store_version; + + bp = call->buffer; + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, store_version); + /* xdr_decode_AFSVolSync(&bp, call->replyX); */ + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.StoreStatus operation type + */ +static const struct afs_call_type afs_RXFSStoreStatus = { + .name = "FS.StoreStatus", + .deliver = afs_deliver_fs_store_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +static const struct afs_call_type afs_RXFSStoreData_as_Status = { + .name = "FS.StoreData", + .deliver = afs_deliver_fs_store_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * set the attributes on a file, using FS.StoreData rather than FS.StoreStatus + * so as to alter the file size also + */ +static int afs_fs_setattr_size(struct afs_server *server, struct key *key, + struct afs_vnode *vnode, struct iattr *attr, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(key), vnode->fid.vid, vnode->fid.vnode); + + ASSERT(attr->ia_valid & ATTR_SIZE); + ASSERTCMP(attr->ia_size, <=, 0xffffffff); // TODO: use 64-bit store + + call = afs_alloc_flat_call(&afs_RXFSStoreData_as_Status, + (4 + 6 + 3) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->store_version = vnode->status.data_version + 1; + call->operation_ID = FSSTOREDATA; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + xdr_encode_AFS_StoreStatus(&bp, attr); + + *bp++ = 0; /* position of start of write */ + *bp++ = 0; /* size of write */ + *bp++ = htonl(attr->ia_size); /* new file length */ + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * set the attributes on a file, using FS.StoreData if there's a change in file + * size, and FS.StoreStatus otherwise + */ +int afs_fs_setattr(struct afs_server *server, struct key *key, + struct afs_vnode *vnode, struct iattr *attr, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + if (attr->ia_valid & ATTR_SIZE) + return afs_fs_setattr_size(server, key, vnode, attr, + wait_mode); + + _enter(",%x,{%x:%u},,", + key_serial(key), vnode->fid.vid, vnode->fid.vnode); + + call = afs_alloc_flat_call(&afs_RXFSStoreStatus, + (4 + 6) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->operation_ID = FSSTORESTATUS; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTORESTATUS); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + xdr_encode_AFS_StoreStatus(&bp, attr); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} diff --git a/fs/afs/inode.c b/fs/afs/inode.c index c184a4e..515a5d1 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -125,7 +125,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, struct inode *inode; int ret; - _enter(",{%u,%u,%u},,", fid->vid, fid->vnode, fid->unique); + _enter(",{%x:%u.%u},,", fid->vid, fid->vnode, fid->unique); as = sb->s_fs_info; data.volume = as->volume; @@ -204,6 +204,19 @@ bad_inode: } /* + * mark the data attached to an inode as obsolete due to a write on the server + * - might also want to ditch all the outstanding writes and dirty pages + */ +void afs_zap_data(struct afs_vnode *vnode) +{ + _enter("zap data {%x:%u}", vnode->fid.vid, vnode->fid.vnode); + + /* nuke all the non-dirty pages that aren't locked, mapped or being + * written back */ + invalidate_remote_inode(&vnode->vfs_inode); +} + +/* * validate a vnode/inode * - there are several things we need to check * - parent dir data changes (rm, rmdir, rename, mkdir, create, link, @@ -258,10 +271,8 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) /* if the vnode's data version number changed then its contents are * different */ - if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) { - _debug("zap data {%x:%d}", vnode->fid.vid, vnode->fid.vnode); - invalidate_remote_inode(&vnode->vfs_inode); - } + if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + afs_zap_data(vnode); clear_bit(AFS_VNODE_MODIFIED, &vnode->flags); mutex_unlock(&vnode->validate_lock); @@ -278,7 +289,7 @@ error_unlock: /* * read the attributes of an inode */ -int afs_inode_getattr(struct vfsmount *mnt, struct dentry *dentry, +int afs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode; @@ -301,7 +312,7 @@ void afs_clear_inode(struct inode *inode) vnode = AFS_FS_I(inode); - _enter("{%x:%d.%d} v=%u x=%u t=%u }", + _enter("{%x:%u.%d} v=%u x=%u t=%u }", vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, @@ -323,6 +334,7 @@ void afs_clear_inode(struct inode *inode) vnode->server = NULL; } + ASSERT(list_empty(&vnode->writebacks)); ASSERT(!vnode->cb_promised); #ifdef AFS_CACHING_SUPPORT @@ -339,3 +351,47 @@ void afs_clear_inode(struct inode *inode) _leave(""); } + +/* + * set the attributes of an inode + */ +int afs_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + struct key *key; + int ret; + + _enter("{%x:%u},{n=%s},%x", + vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name, + attr->ia_valid); + + if (!(attr->ia_valid & (ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID | + ATTR_MTIME))) { + _leave(" = 0 [unsupported]"); + return 0; + } + + /* flush any dirty data outstanding on a regular file */ + if (S_ISREG(vnode->vfs_inode.i_mode)) { + filemap_write_and_wait(vnode->vfs_inode.i_mapping); + afs_writeback_all(vnode); + } + + if (attr->ia_valid & ATTR_FILE) { + key = attr->ia_file->private_data; + } else { + key = afs_request_key(vnode->volume->cell); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + goto error; + } + } + + ret = afs_vnode_setattr(vnode, key, attr); + if (!(attr->ia_valid & ATTR_FILE)) + key_put(key); + +error: + _leave(" = %d", ret); + return ret; +} diff --git a/fs/afs/internal.h b/fs/afs/internal.h index d90c158..a30d4fa 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -21,6 +21,7 @@ #define AFS_CELL_MAX_ADDRS 15 +struct pagevec; struct afs_call; typedef enum { @@ -75,12 +76,15 @@ struct afs_call { struct key *key; /* security for this call */ struct afs_server *server; /* server affected by incoming CM call */ void *request; /* request data (first part) */ - void *request2; /* request data (second part) */ + struct address_space *mapping; /* page set */ + struct afs_writeback *wb; /* writeback being performed */ void *buffer; /* reply receive buffer */ void *reply; /* reply buffer (first part) */ void *reply2; /* reply buffer (second part) */ void *reply3; /* reply buffer (third part) */ void *reply4; /* reply buffer (fourth part) */ + pgoff_t first; /* first page in mapping to deal with */ + pgoff_t last; /* last page in mapping to deal with */ enum { /* call state */ AFS_CALL_REQUESTING, /* request is being sent for outgoing call */ AFS_CALL_AWAIT_REPLY, /* awaiting reply to outgoing call */ @@ -97,14 +101,18 @@ struct afs_call { unsigned request_size; /* size of request data */ unsigned reply_max; /* maximum size of reply */ unsigned reply_size; /* current size of reply */ + unsigned first_offset; /* offset into mapping[first] */ + unsigned last_to; /* amount of mapping[last] */ unsigned short offset; /* offset into received data store */ unsigned char unmarshall; /* unmarshalling phase */ bool incoming; /* T if incoming call */ + bool send_pages; /* T if data from mapping should be sent */ u16 service_id; /* RxRPC service ID to call */ __be16 port; /* target UDP port */ __be32 operation_ID; /* operation ID for an incoming call */ u32 count; /* count for use in unmarshalling */ __be32 tmp; /* place to extract temporary data */ + afs_dataversion_t store_version; /* updated version expected from store */ }; struct afs_call_type { @@ -124,6 +132,32 @@ struct afs_call_type { }; /* + * record of an outstanding writeback on a vnode + */ +struct afs_writeback { + struct list_head link; /* link in vnode->writebacks */ + struct work_struct writer; /* work item to perform the writeback */ + struct afs_vnode *vnode; /* vnode to which this write applies */ + struct key *key; /* owner of this write */ + wait_queue_head_t waitq; /* completion and ready wait queue */ + pgoff_t first; /* first page in batch */ + pgoff_t point; /* last page in current store op */ + pgoff_t last; /* last page in batch (inclusive) */ + unsigned offset_first; /* offset into first page of start of write */ + unsigned to_last; /* offset into last page of end of write */ + int num_conflicts; /* count of conflicting writes in list */ + int usage; + bool conflicts; /* T if has dependent conflicts */ + enum { + AFS_WBACK_SYNCING, /* synchronisation being performed */ + AFS_WBACK_PENDING, /* write pending */ + AFS_WBACK_CONFLICTING, /* conflicting writes posted */ + AFS_WBACK_WRITING, /* writing back */ + AFS_WBACK_COMPLETE /* the writeback record has been unlinked */ + } state __attribute__((packed)); +}; + +/* * AFS superblock private data * - there's one superblock per volume */ @@ -305,6 +339,7 @@ struct afs_vnode { wait_queue_head_t update_waitq; /* status fetch waitqueue */ int update_cnt; /* number of outstanding ops that will update the * status */ + spinlock_t writeback_lock; /* lock for writebacks */ spinlock_t lock; /* waitqueue/flags lock */ unsigned long flags; #define AFS_VNODE_CB_BROKEN 0 /* set if vnode's callback was broken */ @@ -316,6 +351,8 @@ struct afs_vnode { long acl_order; /* ACL check count (callback break count) */ + struct list_head writebacks; /* alterations in pagecache that need writing */ + /* outstanding callback notification on this file */ struct rb_node server_rb; /* link in server->fs_vnodes */ struct rb_node cb_promise; /* link in server->cb_promises */ @@ -433,10 +470,6 @@ extern const struct file_operations afs_file_operations; extern int afs_open(struct inode *, struct file *); extern int afs_release(struct inode *, struct file *); -#ifdef AFS_CACHING_SUPPORT -extern int afs_cache_get_page_cookie(struct page *, struct cachefs_page **); -#endif - /* * fsclient.c */ @@ -467,6 +500,12 @@ extern int afs_fs_rename(struct afs_server *, struct key *, struct afs_vnode *, const char *, struct afs_vnode *, const char *, const struct afs_wait_mode *); +extern int afs_fs_store_data(struct afs_server *, struct afs_writeback *, + pgoff_t, pgoff_t, unsigned, unsigned, + const struct afs_wait_mode *); +extern int afs_fs_setattr(struct afs_server *, struct key *, + struct afs_vnode *, struct iattr *, + const struct afs_wait_mode *); /* * inode.c @@ -474,10 +513,10 @@ extern int afs_fs_rename(struct afs_server *, struct key *, extern struct inode *afs_iget(struct super_block *, struct key *, struct afs_fid *, struct afs_file_status *, struct afs_callback *); +extern void afs_zap_data(struct afs_vnode *); extern int afs_validate(struct afs_vnode *, struct key *); -extern int afs_inode_getattr(struct vfsmount *, struct dentry *, - struct kstat *); -extern void afs_zap_permits(struct rcu_head *); +extern int afs_getattr(struct vfsmount *, struct dentry *, struct kstat *); +extern int afs_setattr(struct dentry *, struct iattr *); extern void afs_clear_inode(struct inode *); /* @@ -533,6 +572,7 @@ extern int afs_extract_data(struct afs_call *, struct sk_buff *, bool, void *, */ 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 *); @@ -629,6 +669,9 @@ extern int afs_vnode_symlink(struct afs_vnode *, struct key *, const char *, struct afs_file_status *, struct afs_server **); extern int afs_vnode_rename(struct afs_vnode *, struct afs_vnode *, struct key *, const char *, const char *); +extern int afs_vnode_store_data(struct afs_writeback *, pgoff_t, pgoff_t, + unsigned, unsigned); +extern int afs_vnode_setattr(struct afs_vnode *, struct key *, struct iattr *); /* * volume.c @@ -645,6 +688,23 @@ extern struct afs_server *afs_volume_pick_fileserver(struct afs_vnode *); extern int afs_volume_release_fileserver(struct afs_vnode *, struct afs_server *, int); +/* + * write.c + */ +extern int afs_set_page_dirty(struct page *); +extern void afs_put_writeback(struct afs_writeback *); +extern int afs_prepare_write(struct file *, struct page *, unsigned, unsigned); +extern int afs_commit_write(struct file *, struct page *, unsigned, unsigned); +extern int afs_writepage(struct page *, struct writeback_control *); +extern int afs_writepages(struct address_space *, struct writeback_control *); +extern int afs_write_inode(struct inode *, int); +extern void afs_pages_written_back(struct afs_vnode *, struct afs_call *); +extern ssize_t afs_file_write(struct kiocb *, const struct iovec *, + unsigned long, loff_t); +extern int afs_writeback_all(struct afs_vnode *); +extern int afs_fsync(struct file *, struct dentry *, int); + + /*****************************************************************************/ /* * debug tracing @@ -726,6 +786,21 @@ do { \ } \ } while(0) +#define ASSERTRANGE(L, OP1, N, OP2, H) \ +do { \ + if (unlikely(!((L) OP1 (N)) || !((N) OP2 (H)))) { \ + printk(KERN_ERR "\n"); \ + printk(KERN_ERR "AFS: Assertion failed\n"); \ + printk(KERN_ERR "%lu "#OP1" %lu "#OP2" %lu is false\n", \ + (unsigned long)(L), (unsigned long)(N), \ + (unsigned long)(H)); \ + printk(KERN_ERR "0x%lx "#OP1" 0x%lx "#OP2" 0x%lx is false\n", \ + (unsigned long)(L), (unsigned long)(N), \ + (unsigned long)(H)); \ + BUG(); \ + } \ +} while(0) + #define ASSERTIF(C, X) \ do { \ if (unlikely((C) && !(X))) { \ @@ -758,6 +833,10 @@ do { \ do { \ } while(0) +#define ASSERTRANGE(L, OP1, N, OP2, H) \ +do { \ +} while(0) + #define ASSERTIF(C, X) \ do { \ } while(0) diff --git a/fs/afs/main.c b/fs/afs/main.c index 80ec6fd..f1f71ff 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -149,6 +149,7 @@ error_cache: afs_vlocation_purge(); afs_cell_purge(); afs_proc_cleanup(); + rcu_barrier(); printk(KERN_ERR "kAFS: failed to register: %d\n", ret); return ret; } @@ -176,6 +177,7 @@ static void __exit afs_exit(void) cachefs_unregister_netfs(&afs_cache_netfs); #endif afs_proc_cleanup(); + rcu_barrier(); } module_exit(afs_exit); diff --git a/fs/afs/misc.c b/fs/afs/misc.c index cdb9792..d1a889c 100644 --- a/fs/afs/misc.c +++ b/fs/afs/misc.c @@ -22,6 +22,7 @@ int afs_abort_to_error(u32 abort_code) { switch (abort_code) { case 13: return -EACCES; + case 27: return -EFBIG; case 30: return -EROFS; case VSALVAGE: return -EIO; case VNOVNODE: return -ENOENT; diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 034fcfd..a3684dc 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -36,7 +36,7 @@ const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, .follow_link = afs_mntpt_follow_link, .readlink = page_readlink, - .getattr = afs_inode_getattr, + .getattr = afs_getattr, }; static LIST_HEAD(afs_vfsmounts); @@ -58,7 +58,8 @@ int afs_mntpt_check_symlink(struct afs_vnode *vnode, struct key *key) char *buf; int ret; - _enter("{%u,%u}", vnode->fid.vnode, vnode->fid.unique); + _enter("{%x:%u,%u}", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); /* read the contents of the symlink into the pagecache */ page = read_mapping_page(AFS_VNODE_TO_I(vnode)->i_mapping, 0, &file); diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 222c1a3..04189c4 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -237,6 +237,70 @@ void afs_flat_call_destructor(struct afs_call *call) } /* + * attach the data from a bunch of pages on an inode to a call + */ +int afs_send_pages(struct afs_call *call, struct msghdr *msg, struct kvec *iov) +{ + struct page *pages[8]; + unsigned count, n, loop, offset, to; + pgoff_t first = call->first, last = call->last; + int ret; + + _enter(""); + + offset = call->first_offset; + call->first_offset = 0; + + do { + _debug("attach %lx-%lx", first, last); + + count = last - first + 1; + if (count > ARRAY_SIZE(pages)) + count = ARRAY_SIZE(pages); + n = find_get_pages_contig(call->mapping, first, count, pages); + ASSERTCMP(n, ==, count); + + loop = 0; + do { + msg->msg_flags = 0; + to = PAGE_SIZE; + if (first + loop >= last) + to = call->last_to; + else + msg->msg_flags = MSG_MORE; + iov->iov_base = kmap(pages[loop]) + offset; + iov->iov_len = to - offset; + offset = 0; + + _debug("- range %u-%u%s", + offset, to, msg->msg_flags ? " [more]" : ""); + msg->msg_iov = (struct iovec *) iov; + msg->msg_iovlen = 1; + + /* have to change the state *before* sending the last + * packet as RxRPC might give us the reply before it + * returns from sending the request */ + if (first + loop >= last) + call->state = AFS_CALL_AWAIT_REPLY; + ret = rxrpc_kernel_send_data(call->rxcall, msg, + to - offset); + kunmap(pages[loop]); + if (ret < 0) + break; + } while (++loop < count); + first += count; + + for (loop = 0; loop < count; loop++) + put_page(pages[loop]); + if (ret < 0) + break; + } while (first < last); + + _leave(" = %d", ret); + return ret; +} + +/* * initiate a call */ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, @@ -253,8 +317,9 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, ASSERT(call->type != NULL); ASSERT(call->type->name != NULL); - _debug("MAKE %p{%s} [%d]", - call, call->type->name, atomic_read(&afs_outstanding_calls)); + _debug("____MAKE %p{%s,%x} [%d]____", + call, call->type->name, key_serial(call->key), + atomic_read(&afs_outstanding_calls)); call->wait_mode = wait_mode; INIT_WORK(&call->async_work, afs_process_async_call); @@ -289,16 +354,23 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, msg.msg_iovlen = 1; msg.msg_control = NULL; msg.msg_controllen = 0; - msg.msg_flags = 0; + msg.msg_flags = (call->send_pages ? MSG_MORE : 0); /* have to change the state *before* sending the last packet as RxRPC * might give us the reply before it returns from sending the * request */ - call->state = AFS_CALL_AWAIT_REPLY; + if (!call->send_pages) + call->state = AFS_CALL_AWAIT_REPLY; ret = rxrpc_kernel_send_data(rxcall, &msg, call->request_size); if (ret < 0) goto error_do_abort; + if (call->send_pages) { + ret = afs_send_pages(call, &msg, iov); + if (ret < 0) + goto error_do_abort; + } + /* at this point, an async call may no longer exist as it may have * already completed */ return wait_mode->wait(call); diff --git a/fs/afs/security.c b/fs/afs/security.c index f9f424d..e0ea88b 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -109,7 +109,7 @@ void afs_clear_permits(struct afs_vnode *vnode) { struct afs_permits *permits; - _enter("{%x}", vnode->fid.vnode); + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); mutex_lock(&vnode->permits_lock); permits = vnode->permits; @@ -132,7 +132,8 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, long acl_order) struct afs_vnode *auth_vnode; int count, loop; - _enter("{%x},%x,%lx", vnode->fid.vnode, key_serial(key), acl_order); + _enter("{%x:%u},%x,%lx", + vnode->fid.vid, vnode->fid.vnode, key_serial(key), acl_order); auth_vnode = afs_get_auth_inode(vnode, key); if (IS_ERR(auth_vnode)) { @@ -220,7 +221,8 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, bool valid; int loop, ret; - _enter(""); + _enter("{%x:%u},%x", + vnode->fid.vid, vnode->fid.vnode, key_serial(key)); auth_vnode = afs_get_auth_inode(vnode, key); if (IS_ERR(auth_vnode)) { @@ -268,9 +270,9 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, _leave(" = %d", ret); return ret; } + *_access = vnode->status.caller_access; } - *_access = vnode->status.caller_access; iput(&auth_vnode->vfs_inode); _leave(" = 0 [access %x]", *_access); return 0; @@ -288,7 +290,7 @@ int afs_permission(struct inode *inode, int mask, struct nameidata *nd) struct key *key; int ret; - _enter("{{%x:%x},%lx},%x,", + _enter("{{%x:%u},%lx},%x,", vnode->fid.vid, vnode->fid.vnode, vnode->flags, mask); key = afs_request_key(vnode->volume->cell); diff --git a/fs/afs/server.c b/fs/afs/server.c index 96bb23b..231ae41 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -252,6 +252,9 @@ static void afs_destroy_server(struct afs_server *server) { _enter("%p", server); + ASSERTIF(server->cb_break_head != server->cb_break_tail, + delayed_work_pending(&server->cb_break_work)); + ASSERTCMP(server->fs_vnodes.rb_node, ==, NULL); ASSERTCMP(server->cb_promises.rb_node, ==, NULL); ASSERTCMP(server->cb_break_head, ==, server->cb_break_tail); diff --git a/fs/afs/super.c b/fs/afs/super.c index 7030d76..d24be33 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -50,6 +50,7 @@ static const struct super_operations afs_super_ops = { .statfs = simple_statfs, .alloc_inode = afs_alloc_inode, .drop_inode = generic_delete_inode, + .write_inode = afs_write_inode, .destroy_inode = afs_destroy_inode, .clear_inode = afs_clear_inode, .umount_begin = afs_umount_begin, @@ -66,7 +67,7 @@ enum { afs_opt_vol, }; -static const match_table_t afs_options_list = { +static match_table_t afs_options_list = { { afs_opt_cell, "cell=%s" }, { afs_opt_rwpath, "rwpath" }, { afs_opt_vol, "vol=%s" }, @@ -459,7 +460,9 @@ static void afs_i_init_once(void *_vnode, struct kmem_cache *cachep, init_waitqueue_head(&vnode->update_waitq); mutex_init(&vnode->permits_lock); mutex_init(&vnode->validate_lock); + spin_lock_init(&vnode->writeback_lock); spin_lock_init(&vnode->lock); + INIT_LIST_HEAD(&vnode->writebacks); INIT_WORK(&vnode->cb_broken_work, afs_broken_callback_work); } } diff --git a/fs/afs/vnode.c b/fs/afs/vnode.c index a1904ab..ec81466 100644 --- a/fs/afs/vnode.c +++ b/fs/afs/vnode.c @@ -261,7 +261,7 @@ int afs_vnode_fetch_status(struct afs_vnode *vnode, DECLARE_WAITQUEUE(myself, current); - _enter("%s,{%u,%u,%u}", + _enter("%s,{%x:%u.%u}", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); @@ -389,7 +389,7 @@ int afs_vnode_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,,,", + _enter("%s{%x:%u.%u},%x,,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -446,7 +446,7 @@ int afs_vnode_create(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s,,", + _enter("%s{%x:%u.%u},%x,%s,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -502,7 +502,7 @@ int afs_vnode_remove(struct afs_vnode *vnode, struct key *key, const char *name, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s", + _enter("%s{%x:%u.%u},%x,%s", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -557,7 +557,7 @@ extern int afs_vnode_link(struct afs_vnode *dvnode, struct afs_vnode *vnode, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%s{%u,%u,%u},%x,%s", + _enter("%s{%x:%u.%u},%s{%x:%u.%u},%x,%s", dvnode->volume->vlocation->vldb.name, dvnode->fid.vid, dvnode->fid.vnode, @@ -628,7 +628,7 @@ int afs_vnode_symlink(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s,%s,,,", + _enter("%s{%x:%u.%u},%x,%s,%s,,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -687,7 +687,7 @@ int afs_vnode_rename(struct afs_vnode *orig_dvnode, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%s{%u,%u,%u},%x,%s,%s", + _enter("%s{%x:%u.%u},%s{%u,%u,%u},%x,%s,%s", orig_dvnode->volume->vlocation->vldb.name, orig_dvnode->fid.vid, orig_dvnode->fid.vnode, @@ -753,3 +753,110 @@ no_server: _leave(" = %ld [cnt %d]", PTR_ERR(server), orig_dvnode->update_cnt); return PTR_ERR(server); } + +/* + * write to a file + */ +int afs_vnode_store_data(struct afs_writeback *wb, pgoff_t first, pgoff_t last, + unsigned offset, unsigned to) +{ + struct afs_server *server; + struct afs_vnode *vnode = wb->vnode; + int ret; + + _enter("%s{%x:%u.%u},%x,%lx,%lx,%x,%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(wb->key), + first, last, offset, to); + + /* this op will fetch the status */ + spin_lock(&vnode->lock); + vnode->update_cnt++; + spin_unlock(&vnode->lock); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_store_data(server, wb, first, last, offset, to, + &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) { + afs_vnode_finalise_status_update(vnode, server); + afs_put_server(server); + } else { + afs_vnode_status_update_failed(vnode, ret); + } + + _leave(" = %d", ret); + return ret; + +no_server: + spin_lock(&vnode->lock); + vnode->update_cnt--; + ASSERTCMP(vnode->update_cnt, >=, 0); + spin_unlock(&vnode->lock); + return PTR_ERR(server); +} + +/* + * set the attributes on a file + */ +int afs_vnode_setattr(struct afs_vnode *vnode, struct key *key, + struct iattr *attr) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + /* this op will fetch the status */ + spin_lock(&vnode->lock); + vnode->update_cnt++; + spin_unlock(&vnode->lock); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_setattr(server, key, vnode, attr, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) { + afs_vnode_finalise_status_update(vnode, server); + afs_put_server(server); + } else { + afs_vnode_status_update_failed(vnode, ret); + } + + _leave(" = %d", ret); + return ret; + +no_server: + spin_lock(&vnode->lock); + vnode->update_cnt--; + ASSERTCMP(vnode->update_cnt, >=, 0); + spin_unlock(&vnode->lock); + return PTR_ERR(server); +} diff --git a/fs/afs/write.c b/fs/afs/write.c new file mode 100644 index 0000000..83ff292 --- /dev/null +++ b/fs/afs/write.c @@ -0,0 +1,835 @@ +/* handling of writes to regular files and writing back to the server + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/slab.h> +#include <linux/fs.h> +#include <linux/pagemap.h> +#include <linux/writeback.h> +#include <linux/pagevec.h> +#include "internal.h" + +static int afs_write_back_from_locked_page(struct afs_writeback *wb, + struct page *page); + +/* + * mark a page as having been made dirty and thus needing writeback + */ +int afs_set_page_dirty(struct page *page) +{ + _enter(""); + return __set_page_dirty_nobuffers(page); +} + +/* + * unlink a writeback record because its usage has reached zero + * - must be called with the wb->vnode->writeback_lock held + */ +static void afs_unlink_writeback(struct afs_writeback *wb) +{ + struct afs_writeback *front; + struct afs_vnode *vnode = wb->vnode; + + list_del_init(&wb->link); + if (!list_empty(&vnode->writebacks)) { + /* if an fsync rises to the front of the queue then wake it + * up */ + front = list_entry(vnode->writebacks.next, + struct afs_writeback, link); + if (front->state == AFS_WBACK_SYNCING) { + _debug("wake up sync"); + front->state = AFS_WBACK_COMPLETE; + wake_up(&front->waitq); + } + } +} + +/* + * free a writeback record + */ +static void afs_free_writeback(struct afs_writeback *wb) +{ + _enter(""); + key_put(wb->key); + kfree(wb); +} + +/* + * dispose of a reference to a writeback record + */ +void afs_put_writeback(struct afs_writeback *wb) +{ + struct afs_vnode *vnode = wb->vnode; + + _enter("{%d}", wb->usage); + + spin_lock(&vnode->writeback_lock); + if (--wb->usage == 0) + afs_unlink_writeback(wb); + else + wb = NULL; + spin_unlock(&vnode->writeback_lock); + if (wb) + afs_free_writeback(wb); +} + +/* + * partly or wholly fill a page that's under preparation for writing + */ +static int afs_fill_page(struct afs_vnode *vnode, struct key *key, + unsigned start, unsigned len, struct page *page) +{ + int ret; + + _enter(",,%u,%u", start, len); + + ASSERTCMP(start + len, <=, PAGE_SIZE); + + ret = afs_vnode_fetch_data(vnode, key, start, len, page); + if (ret < 0) { + if (ret == -ENOENT) { + _debug("got NOENT from server" + " - marking file deleted and stale"); + set_bit(AFS_VNODE_DELETED, &vnode->flags); + ret = -ESTALE; + } + } + + _leave(" = %d", ret); + return ret; +} + +/* + * prepare a page for being written to + */ +static int afs_prepare_page(struct afs_vnode *vnode, struct page *page, + struct key *key, unsigned offset, unsigned to) +{ + unsigned eof, tail, start, stop, len; + loff_t i_size, pos; + void *p; + int ret; + + _enter(""); + + if (offset == 0 && to == PAGE_SIZE) + return 0; + + p = kmap(page); + + i_size = i_size_read(&vnode->vfs_inode); + pos = (loff_t) page->index << PAGE_SHIFT; + if (pos >= i_size) { + /* partial write, page beyond EOF */ + _debug("beyond"); + if (offset > 0) + memset(p, 0, offset); + if (to < PAGE_SIZE) + memset(p + to, 0, PAGE_SIZE - to); + kunmap(page); + return 0; + } + + if (i_size - pos >= PAGE_SIZE) { + /* partial write, page entirely before EOF */ + _debug("before"); + tail = eof = PAGE_SIZE; + } else { + /* partial write, page overlaps EOF */ + eof = i_size - pos; + _debug("overlap %u", eof); + tail = max(eof, to); + if (tail < PAGE_SIZE) + memset(p + tail, 0, PAGE_SIZE - tail); + if (offset > eof) + memset(p + eof, 0, PAGE_SIZE - eof); + } + + kunmap(p); + + ret = 0; + if (offset > 0 || eof > to) { + /* need to fill one or two bits that aren't going to be written + * (cover both fillers in one read if there are two) */ + start = (offset > 0) ? 0 : to; + stop = (eof > to) ? eof : offset; + len = stop - start; + _debug("wr=%u-%u av=0-%u rd=%u@%u", + offset, to, eof, start, len); + ret = afs_fill_page(vnode, key, start, len, page); + } + + _leave(" = %d", ret); + return ret; +} + +/* + * prepare to perform part of a write to a page + * - the caller holds the page locked, preventing it from being written out or + * modified by anyone else + */ +int afs_prepare_write(struct file *file, struct page *page, + unsigned offset, unsigned to) +{ + struct afs_writeback *candidate, *wb; + struct afs_vnode *vnode = AFS_FS_I(file->f_dentry->d_inode); + struct key *key = file->private_data; + pgoff_t index; + int ret; + + _enter("{%x:%u},{%lx},%u,%u", + vnode->fid.vid, vnode->fid.vnode, page->index, offset, to); + + candidate = kzalloc(sizeof(*candidate), GFP_KERNEL); + if (!candidate) + return -ENOMEM; + candidate->vnode = vnode; + candidate->first = candidate->last = page->index; + candidate->offset_first = offset; + candidate->to_last = to; + candidate->usage = 1; + candidate->state = AFS_WBACK_PENDING; + init_waitqueue_head(&candidate->waitq); + + if (!PageUptodate(page)) { + _debug("not up to date"); + ret = afs_prepare_page(vnode, page, key, offset, to); + if (ret < 0) { + kfree(candidate); + _leave(" = %d [prep]", ret); + return ret; + } + SetPageUptodate(page); + } + +try_again: + index = page->index; + spin_lock(&vnode->writeback_lock); + + /* see if this page is already pending a writeback under a suitable key + * - if so we can just join onto that one */ + wb = (struct afs_writeback *) page_private(page); + if (wb) { + if (wb->key == key && wb->state == AFS_WBACK_PENDING) + goto subsume_in_current_wb; + goto flush_conflicting_wb; + } + + if (index > 0) { + /* see if we can find an already pending writeback that we can + * append this page to */ + list_for_each_entry(wb, &vnode->writebacks, link) { + if (wb->last == index - 1 && wb->key == key && + wb->state == AFS_WBACK_PENDING) + goto append_to_previous_wb; + } + } + + list_add_tail(&candidate->link, &vnode->writebacks); + candidate->key = key_get(key); + spin_unlock(&vnode->writeback_lock); + SetPagePrivate(page); + set_page_private(page, (unsigned long) candidate); + _leave(" = 0 [new]"); + return 0; + +subsume_in_current_wb: + _debug("subsume"); + ASSERTRANGE(wb->first, <=, index, <=, wb->last); + if (index == wb->first && offset < wb->offset_first) + wb->offset_first = offset; + if (index == wb->last && to > wb->to_last) + wb->to_last = to; + spin_unlock(&vnode->writeback_lock); + kfree(candidate); + _leave(" = 0 [sub]"); + return 0; + +append_to_previous_wb: + _debug("append into %lx-%lx", wb->first, wb->last); + wb->usage++; + wb->last++; + wb->to_last = to; + spin_unlock(&vnode->writeback_lock); + SetPagePrivate(page); + set_page_private(page, (unsigned long) wb); + kfree(candidate); + _leave(" = 0 [app]"); + return 0; + + /* the page is currently bound to another context, so if it's dirty we + * need to flush it before we can use the new context */ +flush_conflicting_wb: + _debug("flush conflict"); + if (wb->state == AFS_WBACK_PENDING) + wb->state = AFS_WBACK_CONFLICTING; + spin_unlock(&vnode->writeback_lock); + if (PageDirty(page)) { + ret = afs_write_back_from_locked_page(wb, page); + if (ret < 0) { + afs_put_writeback(candidate); + _leave(" = %d", ret); + return ret; + } + } + + /* the page holds a ref on the writeback record */ + afs_put_writeback(wb); + set_page_private(page, 0); + ClearPagePrivate(page); + goto try_again; +} + +/* + * finalise part of a write to a page + */ +int afs_commit_write(struct file *file, struct page *page, + unsigned offset, unsigned to) +{ + struct afs_vnode *vnode = AFS_FS_I(file->f_dentry->d_inode); + loff_t i_size, maybe_i_size; + + _enter("{%x:%u},{%lx},%u,%u", + vnode->fid.vid, vnode->fid.vnode, page->index, offset, to); + + maybe_i_size = (loff_t) page->index << PAGE_SHIFT; + maybe_i_size += to; + + i_size = i_size_read(&vnode->vfs_inode); + if (maybe_i_size > i_size) { + spin_lock(&vnode->writeback_lock); + i_size = i_size_read(&vnode->vfs_inode); + if (maybe_i_size > i_size) + i_size_write(&vnode->vfs_inode, maybe_i_size); + spin_unlock(&vnode->writeback_lock); + } + + set_page_dirty(page); + + if (PageDirty(page)) + _debug("dirtied"); + + return 0; +} + +/* + * kill all the pages in the given range + */ +static void afs_kill_pages(struct afs_vnode *vnode, bool error, + pgoff_t first, pgoff_t last) +{ + struct pagevec pv; + unsigned count, loop; + + _enter("{%x:%u},%lx-%lx", + vnode->fid.vid, vnode->fid.vnode, first, last); + + pagevec_init(&pv, 0); + + do { + _debug("kill %lx-%lx", first, last); + + count = last - first + 1; + if (count > PAGEVEC_SIZE) + count = PAGEVEC_SIZE; + pv.nr = find_get_pages_contig(vnode->vfs_inode.i_mapping, + first, count, pv.pages); + ASSERTCMP(pv.nr, ==, count); + + for (loop = 0; loop < count; loop++) { + ClearPageUptodate(pv.pages[loop]); + if (error) + SetPageError(pv.pages[loop]); + end_page_writeback(pv.pages[loop]); + } + + __pagevec_release(&pv); + } while (first < last); + + _leave(""); +} + +/* + * synchronously write back the locked page and any subsequent non-locked dirty + * pages also covered by the same writeback record + */ +static int afs_write_back_from_locked_page(struct afs_writeback *wb, + struct page *primary_page) +{ + struct page *pages[8], *page; + unsigned long count; + unsigned n, offset, to; + pgoff_t start, first, last; + int loop, ret; + + _enter(",%lx", primary_page->index); + + count = 1; + if (!clear_page_dirty_for_io(primary_page)) + BUG(); + if (test_set_page_writeback(primary_page)) + BUG(); + + /* find all consecutive lockable dirty pages, stopping when we find a + * page that is not immediately lockable, is not dirty or is missing, + * or we reach the end of the range */ + start = primary_page->index; + if (start >= wb->last) + goto no_more; + start++; + do { + _debug("more %lx [%lx]", start, count); + n = wb->last - start + 1; + if (n > ARRAY_SIZE(pages)) + n = ARRAY_SIZE(pages); + n = find_get_pages_contig(wb->vnode->vfs_inode.i_mapping, + start, n, pages); + _debug("fgpc %u", n); + if (n == 0) + goto no_more; + if (pages[0]->index != start) { + for (n--; n >= 0; n--) + put_page(pages[n]); + goto no_more; + } + + for (loop = 0; loop < n; loop++) { + page = pages[loop]; + if (page->index > wb->last) + break; + if (TestSetPageLocked(page)) + break; + if (!PageDirty(page) || + page_private(page) != (unsigned long) wb) { + unlock_page(page); + break; + } + if (!clear_page_dirty_for_io(page)) + BUG(); + if (test_set_page_writeback(page)) + BUG(); + unlock_page(page); + put_page(page); + } + count += loop; + if (loop < n) { + for (; loop < n; loop++) + put_page(pages[loop]); + goto no_more; + } + + start += loop; + } while (start <= wb->last && count < 65536); + +no_more: + /* we now have a contiguous set of dirty pages, each with writeback set + * and the dirty mark cleared; the first page is locked and must remain + * so, all the rest are unlocked */ + first = primary_page->index; + last = first + count - 1; + + offset = (first == wb->first) ? wb->offset_first : 0; + to = (last == wb->last) ? wb->to_last : PAGE_SIZE; + + _debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to); + + ret = afs_vnode_store_data(wb, first, last, offset, to); + if (ret < 0) { + switch (ret) { + case -EDQUOT: + case -ENOSPC: + set_bit(AS_ENOSPC, + &wb->vnode->vfs_inode.i_mapping->flags); + break; + case -EROFS: + case -EIO: + case -EREMOTEIO: + case -EFBIG: + case -ENOENT: + case -ENOMEDIUM: + case -ENXIO: + afs_kill_pages(wb->vnode, true, first, last); + set_bit(AS_EIO, &wb->vnode->vfs_inode.i_mapping->flags); + break; + case -EACCES: + case -EPERM: + case -ENOKEY: + case -EKEYEXPIRED: + case -EKEYREJECTED: + case -EKEYREVOKED: + afs_kill_pages(wb->vnode, false, first, last); + break; + default: + break; + } + } else { + ret = count; + } + + _leave(" = %d", ret); + return ret; +} + +/* + * write a page back to the server + * - the caller locked the page for us + */ +int afs_writepage(struct page *page, struct writeback_control *wbc) +{ + struct backing_dev_info *bdi = page->mapping->backing_dev_info; + struct afs_writeback *wb; + int ret; + + _enter("{%lx},", page->index); + + if (wbc->sync_mode != WB_SYNC_NONE) + wait_on_page_writeback(page); + + if (PageWriteback(page) || !PageDirty(page)) { + unlock_page(page); + return 0; + } + + wb = (struct afs_writeback *) page_private(page); + ASSERT(wb != NULL); + + ret = afs_write_back_from_locked_page(wb, page); + unlock_page(page); + if (ret < 0) { + _leave(" = %d", ret); + return 0; + } + + wbc->nr_to_write -= ret; + if (wbc->nonblocking && bdi_write_congested(bdi)) + wbc->encountered_congestion = 1; + + _leave(" = 0"); + return 0; +} + +/* + * write a region of pages back to the server + */ +int afs_writepages_region(struct address_space *mapping, + struct writeback_control *wbc, + pgoff_t index, pgoff_t end, pgoff_t *_next) +{ + struct backing_dev_info *bdi = mapping->backing_dev_info; + struct afs_writeback *wb; + struct page *page; + int ret, n; + + _enter(",,%lx,%lx,", index, end); + + do { + n = find_get_pages_tag(mapping, &index, PAGECACHE_TAG_DIRTY, + 1, &page); + if (!n) + break; + + _debug("wback %lx", page->index); + + if (page->index > end) { + *_next = index; + page_cache_release(page); + _leave(" = 0 [%lx]", *_next); + return 0; + } + + /* at this point we hold neither mapping->tree_lock nor lock on + * the page itself: the page may be truncated or invalidated + * (changing page->mapping to NULL), or even swizzled back from + * swapper_space to tmpfs file mapping + */ + lock_page(page); + + if (page->mapping != mapping) { + unlock_page(page); + page_cache_release(page); + continue; + } + + if (wbc->sync_mode != WB_SYNC_NONE) + wait_on_page_writeback(page); + + if (PageWriteback(page) || !PageDirty(page)) { + unlock_page(page); + continue; + } + + wb = (struct afs_writeback *) page_private(page); + ASSERT(wb != NULL); + + spin_lock(&wb->vnode->writeback_lock); + wb->state = AFS_WBACK_WRITING; + spin_unlock(&wb->vnode->writeback_lock); + + ret = afs_write_back_from_locked_page(wb, page); + unlock_page(page); + page_cache_release(page); + if (ret < 0) { + _leave(" = %d", ret); + return ret; + } + + wbc->nr_to_write -= ret; + + if (wbc->nonblocking && bdi_write_congested(bdi)) { + wbc->encountered_congestion = 1; + break; + } + + cond_resched(); + } while (index < end && wbc->nr_to_write > 0); + + *_next = index; + _leave(" = 0 [%lx]", *_next); + return 0; +} + +/* + * write some of the pending data back to the server + */ +int afs_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct backing_dev_info *bdi = mapping->backing_dev_info; + pgoff_t start, end, next; + int ret; + + _enter(""); + + if (wbc->nonblocking && bdi_write_congested(bdi)) { + wbc->encountered_congestion = 1; + _leave(" = 0 [congest]"); + return 0; + } + + if (wbc->range_cyclic) { + start = mapping->writeback_index; + end = -1; + ret = afs_writepages_region(mapping, wbc, start, end, &next); + if (start > 0 && wbc->nr_to_write > 0 && ret == 0 && + !(wbc->nonblocking && wbc->encountered_congestion)) + ret = afs_writepages_region(mapping, wbc, 0, start, + &next); + mapping->writeback_index = next; + } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { + end = (pgoff_t)(LLONG_MAX >> PAGE_CACHE_SHIFT); + ret = afs_writepages_region(mapping, wbc, 0, end, &next); + if (wbc->nr_to_write > 0) + mapping->writeback_index = next; + } else { + start = wbc->range_start >> PAGE_CACHE_SHIFT; + end = wbc->range_end >> PAGE_CACHE_SHIFT; + ret = afs_writepages_region(mapping, wbc, start, end, &next); + } + + _leave(" = %d", ret); + return ret; +} + +/* + * write an inode back + */ +int afs_write_inode(struct inode *inode, int sync) +{ + struct afs_vnode *vnode = AFS_FS_I(inode); + int ret; + + _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); + + ret = 0; + if (sync) { + ret = filemap_fdatawait(inode->i_mapping); + if (ret < 0) + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + } + + _leave(" = %d", ret); + return ret; +} + +/* + * completion of write to server + */ +void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call) +{ + struct afs_writeback *wb = call->wb; + struct pagevec pv; + unsigned count, loop; + pgoff_t first = call->first, last = call->last; + bool free_wb; + + _enter("{%x:%u},{%lx-%lx}", + vnode->fid.vid, vnode->fid.vnode, first, last); + + ASSERT(wb != NULL); + + pagevec_init(&pv, 0); + + do { + _debug("attach %lx-%lx", first, last); + + count = last - first + 1; + if (count > PAGEVEC_SIZE) + count = PAGEVEC_SIZE; + pv.nr = find_get_pages_contig(call->mapping, first, count, + pv.pages); + ASSERTCMP(pv.nr, ==, count); + + spin_lock(&vnode->writeback_lock); + for (loop = 0; loop < count; loop++) { + struct page *page = pv.pages[loop]; + end_page_writeback(page); + if (page_private(page) == (unsigned long) wb) { + set_page_private(page, 0); + ClearPagePrivate(page); + wb->usage--; + } + } + free_wb = false; + if (wb->usage == 0) { + afs_unlink_writeback(wb); + free_wb = true; + } + spin_unlock(&vnode->writeback_lock); + first += count; + if (free_wb) { + afs_free_writeback(wb); + wb = NULL; + } + + __pagevec_release(&pv); + } while (first < last); + + _leave(""); +} + +/* + * write to an AFS file + */ +ssize_t afs_file_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct dentry *dentry = iocb->ki_filp->f_path.dentry; + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + ssize_t result; + size_t count = iov_length(iov, nr_segs); + int ret; + + _enter("{%x.%u},{%zu},%lu,", + vnode->fid.vid, vnode->fid.vnode, count, nr_segs); + + if (IS_SWAPFILE(&vnode->vfs_inode)) { + printk(KERN_INFO + "AFS: Attempt to write to active swap file!\n"); + return -EBUSY; + } + + if (!count) + return 0; + + result = generic_file_aio_write(iocb, iov, nr_segs, pos); + if (IS_ERR_VALUE(result)) { + _leave(" = %zd", result); + return result; + } + + /* return error values for O_SYNC and IS_SYNC() */ + if (IS_SYNC(&vnode->vfs_inode) || iocb->ki_filp->f_flags & O_SYNC) { + ret = afs_fsync(iocb->ki_filp, dentry, 1); + if (ret < 0) + result = ret; + } + + _leave(" = %zd", result); + return result; +} + +/* + * flush the vnode to the fileserver + */ +int afs_writeback_all(struct afs_vnode *vnode) +{ + struct address_space *mapping = vnode->vfs_inode.i_mapping; + struct writeback_control wbc = { + .bdi = mapping->backing_dev_info, + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .for_writepages = 1, + .range_cyclic = 1, + }; + int ret; + + _enter(""); + + ret = mapping->a_ops->writepages(mapping, &wbc); + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + + _leave(" = %d", ret); + return ret; +} + +/* + * flush any dirty pages for this process, and check for write errors. + * - the return status from this call provides a reliable indication of + * whether any write errors occurred for this process. + */ +int afs_fsync(struct file *file, struct dentry *dentry, int datasync) +{ + struct afs_writeback *wb, *xwb; + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + int ret; + + _enter("{%x:%u},{n=%s},%d", + vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name, + datasync); + + /* use a writeback record as a marker in the queue - when this reaches + * the front of the queue, all the outstanding writes are either + * completed or rejected */ + wb = kzalloc(sizeof(*wb), GFP_KERNEL); + if (!wb) + return -ENOMEM; + wb->vnode = vnode; + wb->first = 0; + wb->last = -1; + wb->offset_first = 0; + wb->to_last = PAGE_SIZE; + wb->usage = 1; + wb->state = AFS_WBACK_SYNCING; + init_waitqueue_head(&wb->waitq); + + spin_lock(&vnode->writeback_lock); + list_for_each_entry(xwb, &vnode->writebacks, link) { + if (xwb->state == AFS_WBACK_PENDING) + xwb->state = AFS_WBACK_CONFLICTING; + } + list_add_tail(&wb->link, &vnode->writebacks); + spin_unlock(&vnode->writeback_lock); + + /* push all the outstanding writebacks to the server */ + ret = afs_writeback_all(vnode); + if (ret < 0) { + afs_put_writeback(wb); + _leave(" = %d [wb]", ret); + return ret; + } + + /* wait for the preceding writes to actually complete */ + ret = wait_event_interruptible(wb->waitq, + wb->state == AFS_WBACK_COMPLETE || + vnode->writebacks.next == &wb->link); + afs_put_writeback(wb); + _leave(" = %d", ret); + return ret; +} @@ -346,10 +346,9 @@ void fastcall exit_aio(struct mm_struct *mm) wait_for_all_aios(ctx); /* - * this is an overkill, but ensures we don't leave - * the ctx on the aio_wq + * Ensure we don't leave the ctx on the aio_wq */ - flush_workqueue(aio_wq); + cancel_work_sync(&ctx->wq.work); if (1 != atomic_read(&ctx->users)) printk(KERN_DEBUG @@ -372,7 +371,7 @@ void fastcall __put_ioctx(struct kioctx *ctx) BUG_ON(ctx->reqs_active); cancel_delayed_work(&ctx->wq); - flush_workqueue(aio_wq); + cancel_work_sync(&ctx->wq.work); aio_free_ring(ctx); mmdrop(ctx->mm); ctx->mm = NULL; diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 18657f0..72d0b41 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -675,19 +675,8 @@ static ssize_t bm_status_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) { char *s = enabled ? "enabled" : "disabled"; - int len = strlen(s); - loff_t pos = *ppos; - if (pos < 0) - return -EINVAL; - if (pos >= len) - return 0; - if (len < pos + nbytes) - nbytes = len - pos; - if (copy_to_user(buf, s + pos, nbytes)) - return -EFAULT; - *ppos = pos + nbytes; - return nbytes; + return simple_read_from_buffer(buf, nbytes, ppos, s, strlen(s)); } static ssize_t bm_status_write(struct file * file, const char __user * buffer, diff --git a/fs/buffer.c b/fs/buffer.c index eb820b8..aecd057 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1846,13 +1846,8 @@ static int __block_prepare_write(struct inode *inode, struct page *page, if (block_start >= to) break; if (buffer_new(bh)) { - void *kaddr; - clear_buffer_new(bh); - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr+block_start, 0, bh->b_size); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, block_start, bh->b_size, KM_USER0); set_buffer_uptodate(bh); mark_buffer_dirty(bh); } @@ -1940,10 +1935,8 @@ int block_read_full_page(struct page *page, get_block_t *get_block) SetPageError(page); } if (!buffer_mapped(bh)) { - void *kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + i * blocksize, 0, blocksize); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, i * blocksize, blocksize, + KM_USER0); if (!err) set_buffer_uptodate(bh); continue; @@ -2086,7 +2079,6 @@ int cont_prepare_write(struct page *page, unsigned offset, long status; unsigned zerofrom; unsigned blocksize = 1 << inode->i_blkbits; - void *kaddr; while(page->index > (pgpos = *bytes>>PAGE_CACHE_SHIFT)) { status = -ENOMEM; @@ -2108,10 +2100,8 @@ int cont_prepare_write(struct page *page, unsigned offset, PAGE_CACHE_SIZE, get_block); if (status) goto out_unmap; - kaddr = kmap_atomic(new_page, KM_USER0); - memset(kaddr+zerofrom, 0, PAGE_CACHE_SIZE-zerofrom); - flush_dcache_page(new_page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, zerofrom, PAGE_CACHE_SIZE - zerofrom, + KM_USER0); generic_commit_write(NULL, new_page, zerofrom, PAGE_CACHE_SIZE); unlock_page(new_page); page_cache_release(new_page); @@ -2138,10 +2128,7 @@ int cont_prepare_write(struct page *page, unsigned offset, if (status) goto out1; if (zerofrom < offset) { - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr+zerofrom, 0, offset-zerofrom); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, zerofrom, offset - zerofrom, KM_USER0); __block_commit_write(inode, page, zerofrom, offset); } return 0; @@ -2340,10 +2327,7 @@ failed: * Error recovery is pretty slack. Clear the page and mark it dirty * so we'll later zero out any blocks which _were_ allocated. */ - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr, 0, PAGE_CACHE_SIZE); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, 0, PAGE_CACHE_SIZE, KM_USER0); SetPageUptodate(page); set_page_dirty(page); return ret; @@ -2382,7 +2366,6 @@ int nobh_writepage(struct page *page, get_block_t *get_block, loff_t i_size = i_size_read(inode); const pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; unsigned offset; - void *kaddr; int ret; /* Is the page fully inside i_size? */ @@ -2413,10 +2396,7 @@ int nobh_writepage(struct page *page, get_block_t *get_block, * the page size, the remaining memory is zeroed when mapped, and * writes to that region are not written out to the file." */ - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, KM_USER0); out: ret = mpage_writepage(page, get_block, wbc); if (ret == -EAGAIN) @@ -2437,7 +2417,6 @@ int nobh_truncate_page(struct address_space *mapping, loff_t from) unsigned to; struct page *page; const struct address_space_operations *a_ops = mapping->a_ops; - char *kaddr; int ret = 0; if ((offset & (blocksize - 1)) == 0) @@ -2451,10 +2430,8 @@ int nobh_truncate_page(struct address_space *mapping, loff_t from) to = (offset + blocksize) & ~(blocksize - 1); ret = a_ops->prepare_write(NULL, page, offset, to); if (ret == 0) { - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, + KM_USER0); /* * It would be more correct to call aops->commit_write() * here, but this is more efficient. @@ -2480,7 +2457,6 @@ int block_truncate_page(struct address_space *mapping, struct inode *inode = mapping->host; struct page *page; struct buffer_head *bh; - void *kaddr; int err; blocksize = 1 << inode->i_blkbits; @@ -2534,11 +2510,7 @@ int block_truncate_page(struct address_space *mapping, goto unlock; } - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, length); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); - + zero_user_page(page, offset, length, KM_USER0); mark_buffer_dirty(bh); err = 0; @@ -2559,7 +2531,6 @@ int block_write_full_page(struct page *page, get_block_t *get_block, loff_t i_size = i_size_read(inode); const pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; unsigned offset; - void *kaddr; /* Is the page fully inside i_size? */ if (page->index < end_index) @@ -2585,10 +2556,7 @@ int block_write_full_page(struct page *page, get_block_t *get_block, * the page size, the remaining memory is zeroed when mapped, and * writes to that region are not written out to the file." */ - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, KM_USER0); return __block_write_full_page(inode, page, get_block, wbc); } @@ -2978,7 +2946,7 @@ static void buffer_exit_cpu(int cpu) static int buffer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { - if (action == CPU_DEAD) + if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) buffer_exit_cpu((unsigned long)hcpu); return NOTIFY_OK; } diff --git a/fs/configfs/file.c b/fs/configfs/file.c index d98be5e..3527c7c 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -77,36 +77,6 @@ static int fill_read_buffer(struct dentry * dentry, struct configfs_buffer * buf return ret; } - -/** - * flush_read_buffer - push buffer to userspace. - * @buffer: data buffer for file. - * @userbuf: user-passed buffer. - * @count: number of bytes requested. - * @ppos: file position. - * - * Copy the buffer we filled in fill_read_buffer() to userspace. - * This is done at the reader's leisure, copying and advancing - * the amount they specify each time. - * This may be called continuously until the buffer is empty. - */ -static int flush_read_buffer(struct configfs_buffer * buffer, char __user * buf, - size_t count, loff_t * ppos) -{ - int error; - - if (*ppos > buffer->count) - return 0; - - if (count > (buffer->count - *ppos)) - count = buffer->count - *ppos; - - error = copy_to_user(buf,buffer->page + *ppos,count); - if (!error) - *ppos += count; - return error ? -EFAULT : count; -} - /** * configfs_read_file - read an attribute. * @file: file pointer. @@ -139,7 +109,8 @@ configfs_read_file(struct file *file, char __user *buf, size_t count, loff_t *pp } pr_debug("%s: count = %zd, ppos = %lld, buf = %s\n", __FUNCTION__, count, *ppos, buffer->page); - retval = flush_read_buffer(buffer,buf,count,ppos); + retval = simple_read_from_buffer(buf, count, ppos, buffer->page, + buffer->count); out: up(&buffer->sem); return retval; diff --git a/fs/direct-io.c b/fs/direct-io.c index d9d0833..8593f3d 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -439,7 +439,7 @@ static int dio_bio_complete(struct dio *dio, struct bio *bio) * Wait on and process all in-flight BIOs. This must only be called once * all bios have been issued so that the refcount can only decrease. * This just waits for all bios to make it through dio_bio_complete. IO - * errors are propogated through dio->io_error and should be propogated via + * errors are propagated through dio->io_error and should be propagated via * dio_complete(). */ static void dio_await_completion(struct dio *dio) @@ -867,7 +867,6 @@ static int do_direct_IO(struct dio *dio) do_holes: /* Handle holes */ if (!buffer_mapped(map_bh)) { - char *kaddr; loff_t i_size_aligned; /* AKPM: eargh, -ENOTBLK is a hack */ @@ -888,11 +887,8 @@ do_holes: page_cache_release(page); goto out; } - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + (block_in_page << blkbits), - 0, 1 << blkbits); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, block_in_page << blkbits, + 1 << blkbits, KM_USER0); dio->block_in_file++; block_in_page++; goto next_block; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index e1bb031..a6cb617 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1767,7 +1767,6 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, struct inode *inode = mapping->host; struct buffer_head *bh; int err = 0; - void *kaddr; blocksize = inode->i_sb->s_blocksize; length = blocksize - (offset & (blocksize - 1)); @@ -1779,10 +1778,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, */ if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode) && PageUptodate(page)) { - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, length); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, length, KM_USER0); set_page_dirty(page); goto unlock; } @@ -1835,11 +1831,7 @@ static int ext3_block_truncate_page(handle_t *handle, struct page *page, goto unlock; } - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, length); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); - + zero_user_page(page, offset, length, KM_USER0); BUFFER_TRACE(bh, "zeroed end of block"); err = 0; diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c index 0208cc7..47552d4 100644 --- a/fs/jbd/checkpoint.c +++ b/fs/jbd/checkpoint.c @@ -1,5 +1,5 @@ /* - * linux/fs/checkpoint.c + * linux/fs/jbd/checkpoint.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 * diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c index 11563fe..2a5f4b8 100644 --- a/fs/jbd/recovery.c +++ b/fs/jbd/recovery.c @@ -1,5 +1,5 @@ /* - * linux/fs/recovery.c + * linux/fs/jbd/recovery.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 * diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index a68cbb6..824e3b7 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -1,5 +1,5 @@ /* - * linux/fs/revoke.c + * linux/fs/jbd/revoke.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 2000 * diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index f9822fc..772b653 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -1,5 +1,5 @@ /* - * linux/fs/transaction.c + * linux/fs/jbd/transaction.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1998 * diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 68039fa..3fccde7 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -1,5 +1,5 @@ /* - * linux/fs/checkpoint.c + * linux/fs/jbd2/checkpoint.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 * diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 9f10aca..395c92a 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -1,5 +1,5 @@ /* - * linux/fs/recovery.c + * linux/fs/jbd2/recovery.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 * diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 1e864dc..9246e76 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -1,5 +1,5 @@ /* - * linux/fs/revoke.c + * linux/fs/jbd2/revoke.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 2000 * diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index e347d8c0..7946ff4 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1,5 +1,5 @@ /* - * linux/fs/transaction.c + * linux/fs/jbd2/transaction.c * * Written by Stephen C. Tweedie <sct@redhat.com>, 1998 * diff --git a/fs/jffs2/histo_mips.h b/fs/jffs2/histo_mips.h deleted file mode 100644 index fa3dac1..0000000 --- a/fs/jffs2/histo_mips.h +++ /dev/null @@ -1,2 +0,0 @@ -#define BIT_DIVIDER_MIPS 1043 -static int bits_mips[8] = { 277,249,290,267,229,341,212,241}; /* mips32 */ diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 6aff389..4884d5e 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -219,9 +219,9 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info *tn) { uint32_t fn_end = tn->fn->ofs + tn->fn->size; - struct jffs2_tmp_dnode_info *insert_point = NULL, *this; + struct jffs2_tmp_dnode_info *this; - dbg_readinode("insert fragment %#04x-%#04x, ver %u\n", tn->fn->ofs, fn_end, tn->version); + dbg_readinode("insert fragment %#04x-%#04x, ver %u at %08x\n", tn->fn->ofs, fn_end, tn->version, ref_offset(tn->fn->raw)); /* If a node has zero dsize, we only have to keep if it if it might be the node with highest version -- i.e. the one which will end up as f->metadata. @@ -240,23 +240,16 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, /* Find the earliest node which _may_ be relevant to this one */ this = jffs2_lookup_tn(&rii->tn_root, tn->fn->ofs); - if (!this) { - /* First addition to empty tree. $DEITY how I love the easy cases */ - rb_link_node(&tn->rb, NULL, &rii->tn_root.rb_node); - rb_insert_color(&tn->rb, &rii->tn_root); - dbg_readinode("keep new frag\n"); - return 0; - } - - /* If we add a new node it'll be somewhere under here. */ - insert_point = this; - - /* If the node is coincident with another at a lower address, - back up until the other node is found. It may be relevant */ - while (tn->overlapped) - tn = tn_prev(tn); + if (this) { + /* If the node is coincident with another at a lower address, + back up until the other node is found. It may be relevant */ + while (this->overlapped) + this = tn_prev(this); - dbg_readinode("'this' found %#04x-%#04x (%s)\n", this->fn->ofs, this->fn->ofs + this->fn->size, this->fn ? "data" : "hole"); + /* First node should never be marked overlapped */ + BUG_ON(!this); + dbg_readinode("'this' found %#04x-%#04x (%s)\n", this->fn->ofs, this->fn->ofs + this->fn->size, this->fn ? "data" : "hole"); + } while (this) { if (this->fn->ofs > fn_end) @@ -274,11 +267,10 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, return 0; } else { /* Who cares if the new one is good; keep it for now anyway. */ + dbg_readinode("Like new node. Throw away old\n"); rb_replace_node(&this->rb, &tn->rb, &rii->tn_root); - /* Same overlapping from in front and behind */ - tn->overlapped = this->overlapped; jffs2_kill_tn(c, this); - dbg_readinode("Like new node. Throw away old\n"); + /* Same overlapping from in front and behind */ return 0; } } @@ -291,13 +283,8 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, jffs2_kill_tn(c, tn); return 0; } - /* ... and is good. Kill 'this'... */ - rb_replace_node(&this->rb, &tn->rb, &rii->tn_root); - tn->overlapped = this->overlapped; - jffs2_kill_tn(c, this); - /* ... and any subsequent nodes which are also overlapped */ - this = tn_next(tn); - while (this && this->fn->ofs + this->fn->size < fn_end) { + /* ... and is good. Kill 'this' and any subsequent nodes which are also overlapped */ + while (this && this->fn->ofs + this->fn->size <= fn_end) { struct jffs2_tmp_dnode_info *next = tn_next(this); if (this->version < tn->version) { tn_erase(this, &rii->tn_root); @@ -308,8 +295,8 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, } this = next; } - dbg_readinode("Done inserting new\n"); - return 0; + dbg_readinode("Done killing overlapped nodes\n"); + continue; } if (this->version > tn->version && this->fn->ofs <= tn->fn->ofs && @@ -321,29 +308,21 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, return 0; } /* ... but 'this' was bad. Replace it... */ - rb_replace_node(&this->rb, &tn->rb, &rii->tn_root); dbg_readinode("Bad CRC on old overlapping node. Kill it\n"); + tn_erase(this, &rii->tn_root); jffs2_kill_tn(c, this); - return 0; + break; } - /* We want to be inserted under the last node which is - either at a lower offset _or_ has a smaller range */ - if (this->fn->ofs < tn->fn->ofs || - (this->fn->ofs == tn->fn->ofs && - this->fn->size <= tn->fn->size)) - insert_point = this; this = tn_next(this); } - dbg_readinode("insert_point %p, ver %d, 0x%x-0x%x, ov %d\n", - insert_point, insert_point->version, insert_point->fn->ofs, - insert_point->fn->ofs+insert_point->fn->size, - insert_point->overlapped); + /* We neither completely obsoleted nor were completely - obsoleted by an earlier node. Insert under insert_point */ + obsoleted by an earlier node. Insert into the tree */ { - struct rb_node *parent = &insert_point->rb; - struct rb_node **link = &parent; + struct rb_node *parent; + struct rb_node **link = &rii->tn_root.rb_node; + struct jffs2_tmp_dnode_info *insert_point = NULL; while (*link) { parent = *link; @@ -359,6 +338,7 @@ static int jffs2_add_tn_to_tree(struct jffs2_sb_info *c, rb_link_node(&tn->rb, &insert_point->rb, link); rb_insert_color(&tn->rb, &rii->tn_root); } + /* If there's anything behind that overlaps us, note it */ this = tn_prev(tn); if (this) { @@ -457,7 +437,7 @@ static int jffs2_build_inode_fragtree(struct jffs2_sb_info *c, this = tn_last(&rii->tn_root); while (this) { dbg_readinode("tn %p ver %d range 0x%x-0x%x ov %d\n", this, this->version, this->fn->ofs, - this->fn->ofs+this->fn->size, this->overlapped); + this->fn->ofs+this->fn->size, this->overlapped); this = tn_prev(this); } #endif @@ -483,7 +463,7 @@ static int jffs2_build_inode_fragtree(struct jffs2_sb_info *c, vers_next = tn_prev(this); eat_last(&ver_root, &this->rb); if (check_tn_node(c, this)) { - dbg_readinode("node ver %x, 0x%x-0x%x failed CRC\n", + dbg_readinode("node ver %d, 0x%x-0x%x failed CRC\n", this->version, this->fn->ofs, this->fn->ofs+this->fn->size); jffs2_kill_tn(c, this); @@ -496,7 +476,7 @@ static int jffs2_build_inode_fragtree(struct jffs2_sb_info *c, high_ver = this->version; rii->latest_ref = this->fn->raw; } - dbg_readinode("Add %p (v %x, 0x%x-0x%x, ov %d) to fragtree\n", + dbg_readinode("Add %p (v %d, 0x%x-0x%x, ov %d) to fragtree\n", this, this->version, this->fn->ofs, this->fn->ofs+this->fn->size, this->overlapped); @@ -850,7 +830,7 @@ static inline int read_dnode(struct jffs2_sb_info *c, struct jffs2_raw_node_ref return ret; } #ifdef JFFS2_DBG_READINODE_MESSAGES - dbg_readinode("After adding ver %d:\n", tn->version); + dbg_readinode("After adding ver %d:\n", je32_to_cpu(rd->version)); tn = tn_first(&rii->tn_root); while (tn) { dbg_readinode("%p: v %d r 0x%x-0x%x ov %d\n", diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index c556e85..91d1d0f 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -637,7 +637,10 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) memset(c->wbuf,0xff,c->wbuf_pagesize); /* adjust write buffer offset, else we get a non contiguous write bug */ - c->wbuf_ofs += c->wbuf_pagesize; + if (SECTOR_ADDR(c->wbuf_ofs) == SECTOR_ADDR(c->wbuf_ofs+c->wbuf_pagesize)) + c->wbuf_ofs += c->wbuf_pagesize; + else + c->wbuf_ofs = 0xffffffff; c->wbuf_len = 0; return 0; } diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index 82b0544..f3b1ebb 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -1507,7 +1507,7 @@ dbAllocAG(struct bmap * bmp, int agno, s64 nblocks, int l2nb, s64 * results) if (l2nb < budmin) { /* search the lower level dmap control pages to get - * the starting block number of the the dmap that + * the starting block number of the dmap that * contains or starts off the free space. */ if ((rc = diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index c465607..c653022 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -386,7 +386,7 @@ int diRead(struct inode *ip) return -EIO; } - /* locate the the disk inode requested */ + /* locate the disk inode requested */ dp = (struct dinode *) mp->data; dp += rel_inode; @@ -1407,7 +1407,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip) inum = pip->i_ino + 1; ino = inum & (INOSPERIAG - 1); - /* back off the the hint if it is outside of the iag */ + /* back off the hint if it is outside of the iag */ if (ino == 0) inum = pip->i_ino; diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index 6a3f00d..44a2f33 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -1960,7 +1960,7 @@ static void lbmfree(struct lbuf * bp) /* * NAME: lbmRedrive * - * FUNCTION: add a log buffer to the the log redrive list + * FUNCTION: add a log buffer to the log redrive list * * PARAMETER: * bp - log buffer @@ -159,7 +159,10 @@ int dcache_readdir(struct file * filp, void * dirent, filldir_t filldir) continue; spin_unlock(&dcache_lock); - if (filldir(dirent, next->d_name.name, next->d_name.len, filp->f_pos, next->d_inode->i_ino, dt_type(next->d_inode)) < 0) + if (filldir(dirent, next->d_name.name, + next->d_name.len, filp->f_pos, + next->d_inode->i_ino, + dt_type(next->d_inode)) < 0) return 0; spin_lock(&dcache_lock); /* next is still alive */ @@ -284,11 +284,9 @@ do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages, } if (first_hole != blocks_per_page) { - char *kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + (first_hole << blkbits), 0, - PAGE_CACHE_SIZE - (first_hole << blkbits)); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, first_hole << blkbits, + PAGE_CACHE_SIZE - (first_hole << blkbits), + KM_USER0); if (first_hole == 0) { SetPageUptodate(page); unlock_page(page); @@ -576,14 +574,11 @@ page_is_mapped: * written out to the file." */ unsigned offset = i_size & (PAGE_CACHE_SIZE - 1); - char *kaddr; if (page->index > end_index || !offset) goto confused; - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, PAGE_CACHE_SIZE - offset, + KM_USER0); } /* @@ -1152,14 +1152,12 @@ static int fastcall do_path_lookup(int dfd, const char *name, fput_light(file, fput_needed); } - current->total_link_count = 0; - retval = link_path_walk(name, nd); + + retval = path_walk(name, nd); out: - if (likely(retval == 0)) { - if (unlikely(!audit_dummy_context() && nd && nd->dentry && + if (unlikely(!retval && !audit_dummy_context() && nd->dentry && nd->dentry->d_inode)) audit_inode(name, nd->dentry->d_inode); - } out_fail: return retval; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 625d8e5..3df4288 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -38,7 +38,6 @@ #include "delegation.h" #include "iostat.h" -#define NFS_PARANOIA 1 /* #define NFS_DEBUG_VERBOSE 1 */ static int nfs_opendir(struct inode *, struct file *); @@ -650,12 +649,15 @@ int nfs_fsync_dir(struct file *filp, struct dentry *dentry, int datasync) */ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) { + unsigned long verf; + if (IS_ROOT(dentry)) return 1; - if ((NFS_I(dir)->cache_validity & NFS_INO_INVALID_ATTR) != 0 - || nfs_attribute_timeout(dir)) + verf = (unsigned long)dentry->d_fsdata; + if (nfs_caches_unstable(dir) + || verf != NFS_I(dir)->cache_change_attribute) return 0; - return nfs_verify_change_attribute(dir, (unsigned long)dentry->d_fsdata); + return 1; } static inline void nfs_set_verifier(struct dentry * dentry, unsigned long verf) @@ -665,8 +667,7 @@ static inline void nfs_set_verifier(struct dentry * dentry, unsigned long verf) static void nfs_refresh_verifier(struct dentry * dentry, unsigned long verf) { - if (time_after(verf, (unsigned long)dentry->d_fsdata)) - nfs_set_verifier(dentry, verf); + nfs_set_verifier(dentry, verf); } /* @@ -765,6 +766,10 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd) nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE); inode = dentry->d_inode; + /* Revalidate parent directory attribute cache */ + if (nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0) + goto out_zap_parent; + if (!inode) { if (nfs_neg_need_reval(dir, dentry, nd)) goto out_bad; @@ -778,10 +783,6 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd) goto out_bad; } - /* Revalidate parent directory attribute cache */ - if (nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0) - goto out_zap_parent; - /* Force a full look up iff the parent directory has changed */ if (nfs_check_verifier(dir, dentry)) { if (nfs_lookup_verify_inode(inode, nd)) @@ -1360,11 +1361,6 @@ static int nfs_sillyrename(struct inode *dir, struct dentry *dentry) atomic_read(&dentry->d_count)); nfs_inc_stats(dir, NFSIOS_SILLYRENAME); -#ifdef NFS_PARANOIA -if (!dentry->d_inode) -printk("NFS: silly-renaming %s/%s, negative dentry??\n", -dentry->d_parent->d_name.name, dentry->d_name.name); -#endif /* * We don't allow a dentry to be silly-renamed twice. */ @@ -1681,16 +1677,9 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, new_inode = NULL; /* instantiate the replacement target */ d_instantiate(new_dentry, NULL); - } else if (atomic_read(&new_dentry->d_count) > 1) { - /* dentry still busy? */ -#ifdef NFS_PARANOIA - printk("nfs_rename: target %s/%s busy, d_count=%d\n", - new_dentry->d_parent->d_name.name, - new_dentry->d_name.name, - atomic_read(&new_dentry->d_count)); -#endif + } else if (atomic_read(&new_dentry->d_count) > 1) + /* dentry still busy? */ goto out; - } } else drop_nlink(new_inode); diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 2347785..d1cbf0a 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -41,7 +41,6 @@ #include "internal.h" #define NFSDBG_FACILITY NFSDBG_CLIENT -#define NFS_PARANOIA 1 /* * get an NFS2/NFS3 root dentry from the root filehandle diff --git a/fs/nfs/idmap.c b/fs/nfs/idmap.c index 9d4a6b2..d11eb055 100644 --- a/fs/nfs/idmap.c +++ b/fs/nfs/idmap.c @@ -272,7 +272,7 @@ nfs_idmap_id(struct idmap *idmap, struct idmap_hashtable *h, set_current_state(TASK_UNINTERRUPTIBLE); mutex_unlock(&idmap->idmap_im_lock); schedule(); - current->state = TASK_RUNNING; + __set_current_state(TASK_RUNNING); remove_wait_queue(&idmap->idmap_wq, &wq); mutex_lock(&idmap->idmap_im_lock); @@ -333,7 +333,7 @@ nfs_idmap_name(struct idmap *idmap, struct idmap_hashtable *h, set_current_state(TASK_UNINTERRUPTIBLE); mutex_unlock(&idmap->idmap_im_lock); schedule(); - current->state = TASK_RUNNING; + __set_current_state(TASK_RUNNING); remove_wait_queue(&idmap->idmap_wq, &wq); mutex_lock(&idmap->idmap_im_lock); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 1e9a915..2a3fd95 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -48,7 +48,6 @@ #include "internal.h" #define NFSDBG_FACILITY NFSDBG_VFS -#define NFS_PARANOIA 1 static void nfs_invalidate_inode(struct inode *); static int nfs_update_inode(struct inode *, struct nfs_fattr *); @@ -1075,10 +1074,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* * Big trouble! The inode has become a different object. */ -#ifdef NFS_PARANOIA printk(KERN_DEBUG "%s: inode %ld mode changed, %07o to %07o\n", __FUNCTION__, inode->i_ino, inode->i_mode, fattr->mode); -#endif out_err: /* * No need to worry about unhashing the dentry, as the diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index abd9f8b..cd3ca7b 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -26,7 +26,6 @@ #include "internal.h" #define NFSDBG_FACILITY NFSDBG_XDR -/* #define NFS_PARANOIA 1 */ /* Mapping from NFS error code to "errno" error code. */ #define errno_NFSERR_IO EIO diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index b8c28f2..938f371 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -224,7 +224,8 @@ static int nfs4_stat_to_errno(int); encode_getattr_maxsz) #define NFS4_dec_setattr_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 3) + op_decode_hdr_maxsz + 3 + \ + nfs4_fattr_maxsz) #define NFS4_enc_fsinfo_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_fsinfo_maxsz) @@ -2079,9 +2080,11 @@ out: #define READ_BUF(nbytes) do { \ p = xdr_inline_decode(xdr, nbytes); \ - if (!p) { \ - printk(KERN_WARNING "%s: reply buffer overflowed in line %d.", \ - __FUNCTION__, __LINE__); \ + if (unlikely(!p)) { \ + printk(KERN_INFO "%s: prematurely hit end of receive" \ + " buffer\n", __FUNCTION__); \ + printk(KERN_INFO "%s: xdr->p=%p, bytes=%u, xdr->end=%p\n", \ + __FUNCTION__, xdr->p, nbytes, xdr->end); \ return -EIO; \ } \ } while (0) diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 3889501..e12054c 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -20,8 +20,6 @@ #include "internal.h" -#define NFS_PARANOIA 1 - static struct kmem_cache *nfs_page_cachep; static inline struct nfs_page * @@ -167,11 +165,6 @@ nfs_release_request(struct nfs_page *req) if (!atomic_dec_and_test(&req->wb_count)) return; -#ifdef NFS_PARANOIA - BUG_ON (!list_empty(&req->wb_list)); - BUG_ON (NFS_WBACK_BUSY(req)); -#endif - /* Release struct file or cached credential */ nfs_clear_request(req); put_nfs_open_context(req->wb_context); diff --git a/fs/nfsd/Makefile b/fs/nfsd/Makefile index ce341dc..9b118ee 100644 --- a/fs/nfsd/Makefile +++ b/fs/nfsd/Makefile @@ -11,4 +11,3 @@ nfsd-$(CONFIG_NFSD_V3) += nfs3proc.o nfs3xdr.o nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o nfsd-$(CONFIG_NFSD_V4) += nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \ nfs4acl.o nfs4callback.o nfs4recover.o -nfsd-objs := $(nfsd-y) diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 6f24768..79bd03b 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -469,6 +469,13 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) nd.dentry = NULL; exp.ex_path = NULL; + /* fs locations */ + exp.ex_fslocs.locations = NULL; + exp.ex_fslocs.locations_count = 0; + exp.ex_fslocs.migrated = 0; + + exp.ex_uuid = NULL; + if (mesg[mlen-1] != '\n') return -EINVAL; mesg[mlen-1] = 0; @@ -509,13 +516,6 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) if (exp.h.expiry_time == 0) goto out; - /* fs locations */ - exp.ex_fslocs.locations = NULL; - exp.ex_fslocs.locations_count = 0; - exp.ex_fslocs.migrated = 0; - - exp.ex_uuid = NULL; - /* flags */ err = get_int(&mesg, &an_int); if (err == -ENOENT) diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 7f5bad0..eac8283 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -177,7 +177,7 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, if (max_blocksize < resp->count) resp->count = max_blocksize; - svc_reserve(rqstp, ((1 + NFS3_POST_OP_ATTR_WORDS + 3)<<2) + resp->count +4); + svc_reserve_auth(rqstp, ((1 + NFS3_POST_OP_ATTR_WORDS + 3)<<2) + resp->count +4); fh_copy(&resp->fh, &argp->fh); nfserr = nfsd_read(rqstp, &resp->fh, NULL, diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 7e4bb0a..10f6e7d 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -239,7 +239,7 @@ static __be32 * encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; - if (dentry && dentry->d_inode != NULL) { + if (dentry && dentry->d_inode) { int err; struct kstat stat; @@ -300,9 +300,9 @@ int nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_sattrargs *args) { - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_sattr3(p, &args->attrs))) + if (!(p = decode_fh(p, &args->fh))) return 0; + p = decode_sattr3(p, &args->attrs); if ((args->check_guard = ntohl(*p++)) != 0) { struct timespec time; @@ -343,9 +343,9 @@ nfs3svc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, int v,pn; u32 max_blocksize = svc_max_payload(rqstp); - if (!(p = decode_fh(p, &args->fh)) - || !(p = xdr_decode_hyper(p, &args->offset))) + if (!(p = decode_fh(p, &args->fh))) return 0; + p = xdr_decode_hyper(p, &args->offset); len = args->count = ntohl(*p++); @@ -369,28 +369,44 @@ int nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeargs *args) { - unsigned int len, v, hdr; + unsigned int len, v, hdr, dlen; u32 max_blocksize = svc_max_payload(rqstp); - if (!(p = decode_fh(p, &args->fh)) - || !(p = xdr_decode_hyper(p, &args->offset))) + if (!(p = decode_fh(p, &args->fh))) return 0; + p = xdr_decode_hyper(p, &args->offset); args->count = ntohl(*p++); args->stable = ntohl(*p++); len = args->len = ntohl(*p++); + /* + * The count must equal the amount of data passed. + */ + if (args->count != args->len) + return 0; + /* + * Check to make sure that we got the right number of + * bytes. + */ hdr = (void*)p - rqstp->rq_arg.head[0].iov_base; - if (rqstp->rq_arg.len < hdr || - rqstp->rq_arg.len - hdr < len) + dlen = rqstp->rq_arg.head[0].iov_len + rqstp->rq_arg.page_len + - hdr; + /* + * Round the length of the data which was specified up to + * the next multiple of XDR units and then compare that + * against the length which was actually received. + */ + if (dlen != XDR_QUADLEN(len)*4) return 0; + if (args->count > max_blocksize) { + args->count = max_blocksize; + len = args->len = max_blocksize; + } rqstp->rq_vec[0].iov_base = (void*)p; rqstp->rq_vec[0].iov_len = rqstp->rq_arg.head[0].iov_len - hdr; - - if (len > max_blocksize) - len = max_blocksize; - v= 0; + v = 0; while (len > rqstp->rq_vec[v].iov_len) { len -= rqstp->rq_vec[v].iov_len; v++; @@ -398,9 +414,8 @@ nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, rqstp->rq_vec[v].iov_len = PAGE_SIZE; } rqstp->rq_vec[v].iov_len = len; - args->vlen = v+1; - - return args->count == args->len && rqstp->rq_vec[0].iov_len > 0; + args->vlen = v + 1; + return 1; } int @@ -414,8 +429,7 @@ nfs3svc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, switch (args->createmode = ntohl(*p++)) { case NFS3_CREATE_UNCHECKED: case NFS3_CREATE_GUARDED: - if (!(p = decode_sattr3(p, &args->attrs))) - return 0; + p = decode_sattr3(p, &args->attrs); break; case NFS3_CREATE_EXCLUSIVE: args->verf = p; @@ -431,10 +445,10 @@ int nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len)) - || !(p = decode_sattr3(p, &args->attrs))) + if (!(p = decode_fh(p, &args->fh)) || + !(p = decode_filename(p, &args->name, &args->len))) return 0; + p = decode_sattr3(p, &args->attrs); return xdr_argsize_check(rqstp, p); } @@ -448,11 +462,12 @@ nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, char *old, *new; struct kvec *vec; - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_filename(p, &args->fname, &args->flen)) - || !(p = decode_sattr3(p, &args->attrs)) + if (!(p = decode_fh(p, &args->ffh)) || + !(p = decode_filename(p, &args->fname, &args->flen)) ) return 0; + p = decode_sattr3(p, &args->attrs); + /* now decode the pathname, which might be larger than the first page. * As we have to check for nul's anyway, we copy it into a new page * This page appears in the rq_res.pages list, but as pages_len is always @@ -502,10 +517,8 @@ nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, __be32 *p, args->ftype = ntohl(*p++); if (args->ftype == NF3BLK || args->ftype == NF3CHR - || args->ftype == NF3SOCK || args->ftype == NF3FIFO) { - if (!(p = decode_sattr3(p, &args->attrs))) - return 0; - } + || args->ftype == NF3SOCK || args->ftype == NF3FIFO) + p = decode_sattr3(p, &args->attrs); if (args->ftype == NF3BLK || args->ftype == NF3CHR) { args->major = ntohl(*p++); diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index 673a53c..cc3b7ba 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -137,7 +137,6 @@ struct ace_container { static short ace2type(struct nfs4_ace *); static void _posix_to_nfsv4_one(struct posix_acl *, struct nfs4_acl *, unsigned int); -void nfs4_acl_add_ace(struct nfs4_acl *, u32, u32, u32, int, uid_t); struct nfs4_acl * nfs4_acl_posix_to_nfsv4(struct posix_acl *pacl, struct posix_acl *dpacl, @@ -785,21 +784,6 @@ nfs4_acl_new(int n) return acl; } -void -nfs4_acl_add_ace(struct nfs4_acl *acl, u32 type, u32 flag, u32 access_mask, - int whotype, uid_t who) -{ - struct nfs4_ace *ace = acl->aces + acl->naces; - - ace->type = type; - ace->flag = flag; - ace->access_mask = access_mask; - ace->whotype = whotype; - ace->who = who; - - acl->naces++; -} - static struct { char *string; int stringlen; @@ -851,6 +835,5 @@ nfs4_acl_write_who(int who, char *p) } EXPORT_SYMBOL(nfs4_acl_new); -EXPORT_SYMBOL(nfs4_acl_add_ace); EXPORT_SYMBOL(nfs4_acl_get_whotype); EXPORT_SYMBOL(nfs4_acl_write_who); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 678f3be..3cc8ce4 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -1326,8 +1326,6 @@ do_recall(void *__dp) { struct nfs4_delegation *dp = __dp; - daemonize("nfsv4-recall"); - nfsd4_cb_recall(dp); return 0; } diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 739dd3c..6ca2d24 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -323,7 +323,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, * */ - u8 version = 1; + u8 version; u8 fsid_type = 0; struct inode * inode = dentry->d_inode; struct dentry *parent = dentry->d_parent; @@ -341,15 +341,59 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, * the reference filehandle (if it is in the same export) * or the export options. */ + retry: + version = 1; if (ref_fh && ref_fh->fh_export == exp) { version = ref_fh->fh_handle.fh_version; - if (version == 0xca) + fsid_type = ref_fh->fh_handle.fh_fsid_type; + + if (ref_fh == fhp) + fh_put(ref_fh); + ref_fh = NULL; + + switch (version) { + case 0xca: fsid_type = FSID_DEV; - else - fsid_type = ref_fh->fh_handle.fh_fsid_type; - /* We know this version/type works for this export - * so there is no need for further checks. + break; + case 1: + break; + default: + goto retry; + } + + /* Need to check that this type works for this + * export point. As the fsid -> filesystem mapping + * was guided by user-space, there is no guarantee + * that the filesystem actually supports that fsid + * type. If it doesn't we loop around again without + * ref_fh set. */ + switch(fsid_type) { + case FSID_DEV: + if (!old_valid_dev(ex_dev)) + goto retry; + /* FALL THROUGH */ + case FSID_MAJOR_MINOR: + case FSID_ENCODE_DEV: + if (!(exp->ex_dentry->d_inode->i_sb->s_type->fs_flags + & FS_REQUIRES_DEV)) + goto retry; + break; + case FSID_NUM: + if (! (exp->ex_flags & NFSEXP_FSID)) + goto retry; + break; + case FSID_UUID8: + case FSID_UUID16: + if (!root_export) + goto retry; + /* fall through */ + case FSID_UUID4_INUM: + case FSID_UUID16_INUM: + if (exp->ex_uuid == NULL) + goto retry; + break; + } } else if (exp->ex_uuid) { if (fhp->fh_maxsize >= 64) { if (root_export) diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 5cc2eec..b2c7147 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -155,7 +155,7 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, argp->count); argp->count = NFSSVC_MAXBLKSIZE_V2; } - svc_reserve(rqstp, (19<<2) + argp->count + 4); + svc_reserve_auth(rqstp, (19<<2) + argp->count + 4); resp->count = argp->count; nfserr = nfsd_read(rqstp, fh_copy(&resp->fh, &argp->fh), NULL, diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 0c24b9e..cb3e7fa 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -231,9 +231,10 @@ int nfssvc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_sattrargs *args) { - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_sattr(p, &args->attrs))) + p = decode_fh(p, &args->fh); + if (!p) return 0; + p = decode_sattr(p, &args->attrs); return xdr_argsize_check(rqstp, p); } @@ -284,8 +285,9 @@ int nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_writeargs *args) { - unsigned int len; + unsigned int len, hdr, dlen; int v; + if (!(p = decode_fh(p, &args->fh))) return 0; @@ -293,11 +295,30 @@ nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, args->offset = ntohl(*p++); /* offset */ p++; /* totalcount */ len = args->len = ntohl(*p++); - rqstp->rq_vec[0].iov_base = (void*)p; - rqstp->rq_vec[0].iov_len = rqstp->rq_arg.head[0].iov_len - - (((void*)p) - rqstp->rq_arg.head[0].iov_base); + /* + * The protocol specifies a maximum of 8192 bytes. + */ if (len > NFSSVC_MAXBLKSIZE_V2) - len = NFSSVC_MAXBLKSIZE_V2; + return 0; + + /* + * Check to make sure that we got the right number of + * bytes. + */ + hdr = (void*)p - rqstp->rq_arg.head[0].iov_base; + dlen = rqstp->rq_arg.head[0].iov_len + rqstp->rq_arg.page_len + - hdr; + + /* + * Round the length of the data which was specified up to + * the next multiple of XDR units and then compare that + * against the length which was actually received. + */ + if (dlen != XDR_QUADLEN(len)*4) + return 0; + + rqstp->rq_vec[0].iov_base = (void*)p; + rqstp->rq_vec[0].iov_len = rqstp->rq_arg.head[0].iov_len - hdr; v = 0; while (len > rqstp->rq_vec[v].iov_len) { len -= rqstp->rq_vec[v].iov_len; @@ -306,18 +327,18 @@ nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, rqstp->rq_vec[v].iov_len = PAGE_SIZE; } rqstp->rq_vec[v].iov_len = len; - args->vlen = v+1; - return rqstp->rq_vec[0].iov_len > 0; + args->vlen = v + 1; + return 1; } int nfssvc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_createargs *args) { - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len)) - || !(p = decode_sattr(p, &args->attrs))) + if ( !(p = decode_fh(p, &args->fh)) + || !(p = decode_filename(p, &args->name, &args->len))) return 0; + p = decode_sattr(p, &args->attrs); return xdr_argsize_check(rqstp, p); } @@ -361,11 +382,11 @@ int nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_symlinkargs *args) { - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_filename(p, &args->fname, &args->flen)) - || !(p = decode_pathname(p, &args->tname, &args->tlen)) - || !(p = decode_sattr(p, &args->attrs))) + if ( !(p = decode_fh(p, &args->ffh)) + || !(p = decode_filename(p, &args->fname, &args->flen)) + || !(p = decode_pathname(p, &args->tname, &args->tlen))) return 0; + p = decode_sattr(p, &args->attrs); return xdr_argsize_check(rqstp, p); } diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index ab45db5..9e451a6 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -1059,20 +1059,12 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode maping blocks, since there is none, so we just zero out remaining parts of first and last pages in write area (if needed) */ if ((pos & ~((loff_t) PAGE_CACHE_SIZE - 1)) > inode->i_size) { - if (from != 0) { /* First page needs to be partially zeroed */ - char *kaddr = kmap_atomic(prepared_pages[0], KM_USER0); - memset(kaddr, 0, from); - kunmap_atomic(kaddr, KM_USER0); - flush_dcache_page(prepared_pages[0]); - } - if (to != PAGE_CACHE_SIZE) { /* Last page needs to be partially zeroed */ - char *kaddr = - kmap_atomic(prepared_pages[num_pages - 1], - KM_USER0); - memset(kaddr + to, 0, PAGE_CACHE_SIZE - to); - kunmap_atomic(kaddr, KM_USER0); - flush_dcache_page(prepared_pages[num_pages - 1]); - } + if (from != 0) /* First page needs to be partially zeroed */ + zero_user_page(prepared_pages[0], 0, from, KM_USER0); + + if (to != PAGE_CACHE_SIZE) /* Last page needs to be partially zeroed */ + zero_user_page(prepared_pages[num_pages-1], to, + PAGE_CACHE_SIZE - to, KM_USER0); /* Since all blocks are new - use already calculated value */ return blocks; @@ -1199,13 +1191,9 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode ll_rw_block(READ, 1, &bh); *wait_bh++ = bh; } else { /* Not mapped, zero it */ - char *kaddr = - kmap_atomic(prepared_pages[0], - KM_USER0); - memset(kaddr + block_start, 0, - from - block_start); - kunmap_atomic(kaddr, KM_USER0); - flush_dcache_page(prepared_pages[0]); + zero_user_page(prepared_pages[0], + block_start, + from - block_start, KM_USER0); set_buffer_uptodate(bh); } } @@ -1237,13 +1225,8 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode ll_rw_block(READ, 1, &bh); *wait_bh++ = bh; } else { /* Not mapped, zero it */ - char *kaddr = - kmap_atomic(prepared_pages - [num_pages - 1], - KM_USER0); - memset(kaddr + to, 0, block_end - to); - kunmap_atomic(kaddr, KM_USER0); - flush_dcache_page(prepared_pages[num_pages - 1]); + zero_user_page(prepared_pages[num_pages-1], + to, block_end - to, KM_USER0); set_buffer_uptodate(bh); } } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 9fcbfe3..1272d11 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2148,13 +2148,8 @@ int reiserfs_truncate_file(struct inode *p_s_inode, int update_timestamps) length = offset & (blocksize - 1); /* if we are not on a block boundary */ if (length) { - char *kaddr; - length = blocksize - length; - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + offset, 0, length); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, offset, length, KM_USER0); if (buffer_mapped(bh) && bh->b_blocknr != 0) { mark_buffer_dirty(bh); } @@ -2370,7 +2365,6 @@ static int reiserfs_write_full_page(struct page *page, ** last byte in the file */ if (page->index >= end_index) { - char *kaddr; unsigned last_offset; last_offset = inode->i_size & (PAGE_CACHE_SIZE - 1); @@ -2379,10 +2373,7 @@ static int reiserfs_write_full_page(struct page *page, unlock_page(page); return 0; } - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + last_offset, 0, PAGE_CACHE_SIZE - last_offset); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); + zero_user_page(page, last_offset, PAGE_CACHE_SIZE - last_offset, KM_USER0); } bh = head; block = page->index << (PAGE_CACHE_SHIFT - s->s_blocksize_bits); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index e073fd8..f25086a 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1110,7 +1110,7 @@ static int flush_commit_list(struct super_block *s, if (!barrier) { /* If there was a write error in the journal - we can't commit * this transaction - it will be invalid and, if successful, - * will just end up propogating the write error out to + * will just end up propagating the write error out to * the file system. */ if (likely(!retval && !reiserfs_is_journal_aborted (journal))) { if (buffer_dirty(jl->j_commit_bh)) @@ -1125,7 +1125,7 @@ static int flush_commit_list(struct super_block *s, /* If there was a write error in the journal - we can't commit this * transaction - it will be invalid and, if successful, will just end - * up propogating the write error out to the filesystem. */ + * up propagating the write error out to the filesystem. */ if (unlikely(!buffer_uptodate(jl->j_commit_bh))) { #ifdef CONFIG_REISERFS_CHECK reiserfs_warning(s, "journal-615: buffer write failed"); diff --git a/fs/select.c b/fs/select.c index d862241..a974082 100644 --- a/fs/select.c +++ b/fs/select.c @@ -64,7 +64,7 @@ EXPORT_SYMBOL(poll_initwait); static void free_poll_entry(struct poll_table_entry *entry) { - remove_wait_queue(entry->wait_address,&entry->wait); + remove_wait_queue(entry->wait_address, &entry->wait); fput(entry->filp); } @@ -128,7 +128,7 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, entry->filp = filp; entry->wait_address = wait_address; init_waitqueue_entry(&entry->wait, current); - add_wait_queue(wait_address,&entry->wait); + add_wait_queue(wait_address, &entry->wait); } #define FDS_IN(fds, n) (fds->in + n) diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 0e637ad..b502c71 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -111,36 +111,6 @@ static int fill_read_buffer(struct dentry * dentry, struct sysfs_buffer * buffer return ret; } - -/** - * flush_read_buffer - push buffer to userspace. - * @buffer: data buffer for file. - * @buf: user-passed buffer. - * @count: number of bytes requested. - * @ppos: file position. - * - * Copy the buffer we filled in fill_read_buffer() to userspace. - * This is done at the reader's leisure, copying and advancing - * the amount they specify each time. - * This may be called continuously until the buffer is empty. - */ -static int flush_read_buffer(struct sysfs_buffer * buffer, char __user * buf, - size_t count, loff_t * ppos) -{ - int error; - - if (*ppos > buffer->count) - return 0; - - if (count > (buffer->count - *ppos)) - count = buffer->count - *ppos; - - error = copy_to_user(buf,buffer->page + *ppos,count); - if (!error) - *ppos += count; - return error ? -EFAULT : count; -} - /** * sysfs_read_file - read an attribute. * @file: file pointer. @@ -177,7 +147,8 @@ sysfs_read_file(struct file *file, char __user *buf, size_t count, loff_t *ppos) } pr_debug("%s: count = %zd, ppos = %lld, buf = %s\n", __FUNCTION__, count, *ppos, buffer->page); - retval = flush_read_buffer(buffer,buf,count,ppos); + retval = simple_read_from_buffer(buf, count, ppos, buffer->page, + buffer->count); out: up(&buffer->sem); return retval; diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 7775ddc..e725ddd 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -809,7 +809,7 @@ xfs_inumbers( xfs_buf_relse(agbp); agbp = NULL; /* - * Move up the the last inode in the current + * Move up the last inode in the current * chunk. The lookup_ge will always get * us the first inode in the next chunk. */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index f5aa3ef..a96bde6 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1734,11 +1734,13 @@ xfs_icsb_cpu_notify( per_cpu_ptr(mp->m_sb_cnts, (unsigned long)hcpu); switch (action) { case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: /* Easy Case - initialize the area and locks, and * then rebalance when online does everything else for us. */ memset(cntp, 0, sizeof(xfs_icsb_cnts_t)); break; case CPU_ONLINE: + case CPU_ONLINE_FROZEN: xfs_icsb_lock(mp); xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0, 0); xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0, 0); @@ -1746,6 +1748,7 @@ xfs_icsb_cpu_notify( xfs_icsb_unlock(mp); break; case CPU_DEAD: + case CPU_DEAD_FROZEN: /* Disable all the counters, then fold the dead cpu's * count into the total on the global superblock and * re-enable the counters. */ |
