diff options
Diffstat (limited to 'fs')
284 files changed, 5233 insertions, 4800 deletions
diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 0ee594569dcc..85b67ffa2a43 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -286,11 +286,9 @@ static struct p9_fid *v9fs_fid_clone_with_uid(struct dentry *dentry, uid_t uid) struct p9_fid *v9fs_writeback_fid(struct dentry *dentry) { - int err, flags; + int err; struct p9_fid *fid; - struct v9fs_session_info *v9ses; - v9ses = v9fs_dentry2v9ses(dentry); fid = v9fs_fid_clone_with_uid(dentry, 0); if (IS_ERR(fid)) goto error_out; @@ -299,17 +297,8 @@ struct p9_fid *v9fs_writeback_fid(struct dentry *dentry) * dirty pages. We always request for the open fid in read-write * mode so that a partial page write which result in page * read can work. - * - * we don't have a tsyncfs operation for older version - * of protocol. So make sure the write back fid is - * opened in O_SYNC mode. */ - if (!v9fs_proto_dotl(v9ses)) - flags = O_RDWR | O_SYNC; - else - flags = O_RDWR; - - err = p9_client_open(fid, flags); + err = p9_client_open(fid, O_RDWR); if (err < 0) { p9_client_clunk(fid); fid = ERR_PTR(err); diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index 9665c2b840e6..e5ebedfc5ed8 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -116,7 +116,6 @@ struct v9fs_session_info { struct list_head slist; /* list of sessions registered with v9fs */ struct backing_dev_info bdi; struct rw_semaphore rename_sem; - struct p9_fid *root_fid; /* Used for file system sync */ }; /* cache_validity flags */ diff --git a/fs/9p/vfs_dentry.c b/fs/9p/vfs_dentry.c index b6a3b9f7fe4d..e022890c6f40 100644 --- a/fs/9p/vfs_dentry.c +++ b/fs/9p/vfs_dentry.c @@ -126,7 +126,9 @@ static int v9fs_lookup_revalidate(struct dentry *dentry, struct nameidata *nd) retval = v9fs_refresh_inode_dotl(fid, inode); else retval = v9fs_refresh_inode(fid, inode); - if (retval <= 0) + if (retval == -ENOENT) + return 0; + if (retval < 0) return retval; } out_valid: diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index ffbb113d5f33..82a7c38ddad0 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -811,7 +811,7 @@ v9fs_vfs_follow_link_dotl(struct dentry *dentry, struct nameidata *nd) fid = v9fs_fid_lookup(dentry); if (IS_ERR(fid)) { __putname(link); - link = ERR_PTR(PTR_ERR(fid)); + link = ERR_CAST(fid); goto ndset; } retval = p9_client_readlink(fid, &target); diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index f3eed3383e4f..feef6cdc1fd2 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -154,6 +154,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, retval = PTR_ERR(inode); goto release_sb; } + root = d_alloc_root(inode); if (!root) { iput(inode); @@ -185,21 +186,10 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, p9stat_free(st); kfree(st); } - v9fs_fid_add(root, fid); retval = v9fs_get_acl(inode, fid); if (retval) goto release_sb; - /* - * Add the root fid to session info. This is used - * for file system sync. We want a cloned fid here - * so that we can do a sync_filesystem after a - * shrink_dcache_for_umount - */ - v9ses->root_fid = v9fs_fid_clone(root); - if (IS_ERR(v9ses->root_fid)) { - retval = PTR_ERR(v9ses->root_fid); - goto release_sb; - } + v9fs_fid_add(root, fid); P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n"); return dget(sb->s_root); @@ -210,11 +200,15 @@ close_session: v9fs_session_close(v9ses); kfree(v9ses); return ERR_PTR(retval); + release_sb: /* - * we will do the session_close and root dentry - * release in the below call. + * we will do the session_close and root dentry release + * in the below call. But we need to clunk fid, because we haven't + * attached the fid to dentry so it won't get clunked + * automatically. */ + p9_client_clunk(fid); deactivate_locked_super(sb); return ERR_PTR(retval); } @@ -232,7 +226,7 @@ static void v9fs_kill_super(struct super_block *s) P9_DPRINTK(P9_DEBUG_VFS, " %p\n", s); kill_anon_super(s); - p9_client_clunk(v9ses->root_fid); + v9fs_session_cancel(v9ses); v9fs_session_close(v9ses); kfree(v9ses); @@ -285,14 +279,6 @@ done: return res; } -static int v9fs_sync_fs(struct super_block *sb, int wait) -{ - struct v9fs_session_info *v9ses = sb->s_fs_info; - - P9_DPRINTK(P9_DEBUG_VFS, "v9fs_sync_fs: super_block %p\n", sb); - return p9_client_sync_fs(v9ses->root_fid); -} - static int v9fs_drop_inode(struct inode *inode) { struct v9fs_session_info *v9ses; @@ -307,6 +293,51 @@ static int v9fs_drop_inode(struct inode *inode) return 1; } +static int v9fs_write_inode(struct inode *inode, + struct writeback_control *wbc) +{ + int ret; + struct p9_wstat wstat; + struct v9fs_inode *v9inode; + /* + * send an fsync request to server irrespective of + * wbc->sync_mode. + */ + P9_DPRINTK(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); + v9inode = V9FS_I(inode); + if (!v9inode->writeback_fid) + return 0; + v9fs_blank_wstat(&wstat); + + ret = p9_client_wstat(v9inode->writeback_fid, &wstat); + if (ret < 0) { + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + return ret; + } + return 0; +} + +static int v9fs_write_inode_dotl(struct inode *inode, + struct writeback_control *wbc) +{ + int ret; + struct v9fs_inode *v9inode; + /* + * send an fsync request to server irrespective of + * wbc->sync_mode. + */ + P9_DPRINTK(P9_DEBUG_VFS, "%s: inode %p\n", __func__, inode); + v9inode = V9FS_I(inode); + if (!v9inode->writeback_fid) + return 0; + ret = p9_client_fsync(v9inode->writeback_fid, 0); + if (ret < 0) { + __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + return ret; + } + return 0; +} + static const struct super_operations v9fs_super_ops = { .alloc_inode = v9fs_alloc_inode, .destroy_inode = v9fs_destroy_inode, @@ -314,17 +345,18 @@ static const struct super_operations v9fs_super_ops = { .evict_inode = v9fs_evict_inode, .show_options = generic_show_options, .umount_begin = v9fs_umount_begin, + .write_inode = v9fs_write_inode, }; static const struct super_operations v9fs_super_ops_dotl = { .alloc_inode = v9fs_alloc_inode, .destroy_inode = v9fs_destroy_inode, - .sync_fs = v9fs_sync_fs, .statfs = v9fs_statfs, .drop_inode = v9fs_drop_inode, .evict_inode = v9fs_evict_inode, .show_options = generic_show_options, .umount_begin = v9fs_umount_begin, + .write_inode = v9fs_write_inode_dotl, }; struct file_system_type v9fs_fs_type = { diff --git a/fs/adfs/map.c b/fs/adfs/map.c index d1a5932bb0f1..6935f05202ac 100644 --- a/fs/adfs/map.c +++ b/fs/adfs/map.c @@ -51,7 +51,7 @@ static DEFINE_RWLOCK(adfs_map_lock); /* * This is fun. We need to load up to 19 bits from the map at an - * arbitary bit alignment. (We're limited to 19 bits by F+ version 2). + * arbitrary bit alignment. (We're limited to 19 bits by F+ version 2). */ #define GET_FRAG_ID(_map,_start,_idmask) \ ({ \ diff --git a/fs/afs/cache.c b/fs/afs/cache.c index 0fb315dd4d2a..577763c3d88b 100644 --- a/fs/afs/cache.c +++ b/fs/afs/cache.c @@ -98,7 +98,7 @@ static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data, } /* - * provide new auxilliary cache data + * provide new auxiliary cache data */ static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data, void *buffer, uint16_t bufmax) @@ -117,7 +117,7 @@ static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data, } /* - * check that the auxilliary data indicates that the entry is still valid + * check that the auxiliary data indicates that the entry is still valid */ static enum fscache_checkaux afs_cell_cache_check_aux(void *cookie_netfs_data, const void *buffer, @@ -150,7 +150,7 @@ static uint16_t afs_vlocation_cache_get_key(const void *cookie_netfs_data, } /* - * provide new auxilliary cache data + * provide new auxiliary cache data */ static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data, void *buffer, uint16_t bufmax) @@ -172,7 +172,7 @@ static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data, } /* - * check that the auxilliary data indicates that the entry is still valid + * check that the auxiliary data indicates that the entry is still valid */ static enum fscache_checkaux afs_vlocation_cache_check_aux(void *cookie_netfs_data, @@ -283,7 +283,7 @@ static void afs_vnode_cache_get_attr(const void *cookie_netfs_data, } /* - * provide new auxilliary cache data + * provide new auxiliary cache data */ static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, void *buffer, uint16_t bufmax) @@ -309,7 +309,7 @@ static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, } /* - * check that the auxilliary data indicates that the entry is still valid + * check that the auxiliary data indicates that the entry is still valid */ static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, const void *buffer, diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 0d5eeadf6121..3c090b7555ea 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -293,7 +293,7 @@ struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz, if (!cell) { /* this should not happen unless user tries to mount * when root cell is not set. Return an impossibly - * bizzare errno to alert the user. Things like + * bizarre errno to alert the user. Things like * ENOENT might be "more appropriate" but they happen * for other reasons. */ diff --git a/fs/attr.c b/fs/attr.c index 1007ed616314..91dbe2a107f2 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -128,7 +128,7 @@ EXPORT_SYMBOL(inode_newsize_ok); * setattr_copy must be called with i_mutex held. * * setattr_copy updates the inode's metadata with that specified - * in attr. Noticably missing is inode size update, which is more complex + * in attr. Noticeably missing is inode size update, which is more complex * as it requires pagecache updates. * * The inode is not marked as dirty after this operation. The rationale is diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 96804a17bbd0..f55ae23b137e 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -612,7 +612,7 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry) * set the DMANAGED_AUTOMOUNT and DMANAGED_TRANSIT flags on the leaves * of the directory tree. There is no need to clear the automount flag * following a mount or restore it after an expire because these mounts - * are always covered. However, it is neccessary to ensure that these + * are always covered. However, it is necessary to ensure that these * flags are clear on non-empty directories to avoid unnecessary calls * during path walks. */ diff --git a/fs/befs/ChangeLog b/fs/befs/ChangeLog index ce8c787916be..75a461cfaca6 100644 --- a/fs/befs/ChangeLog +++ b/fs/befs/ChangeLog @@ -24,7 +24,7 @@ Version 0.9 (2002-03-14) Version 0.64 (2002-02-07) ========== -* Did the string comparision really right this time (btree.c) [WD] +* Did the string comparison really right this time (btree.c) [WD] * Fixed up some places where I assumed that a long int could hold a pointer value. (btree.c) [WD] @@ -114,7 +114,7 @@ Version 0.6 (2001-12-15) More flexible. Will soon be controllable at mount time (see TODO). [WD] -* Rewrote datastream positon lookups. +* Rewrote datastream position lookups. (datastream.c) [WD] * Moved the TODO list to its own file. @@ -150,7 +150,7 @@ Version 0.50 (2001-11-13) * Anton also told me that the blocksize is not allowed to be larger than the page size in linux, which is 4k i386. Oops. Added a test for (blocksize > PAGE_SIZE), and refuse to mount in that case. What this - practicaly means is that 8k blocksize volumes won't work without a major + practically means is that 8k blocksize volumes won't work without a major restructuring of the driver (or an alpha or other 64bit hardware). [WD] * Cleaned up the befs_count_blocks() function. Much smarter now. @@ -183,7 +183,7 @@ Version 0.45 (2001-10-29) structures into the generic pointer fields of the public structures with kmalloc(). put_super and put_inode free them. This allows us not to have to touch the definitions of the public structures in - include/linux/fs.h. Also, befs_inode_info is huge (becuase of the + include/linux/fs.h. Also, befs_inode_info is huge (because of the symlink string). (super.c, inode.c, befs_fs.h) [WD] * Fixed a thinko that was corrupting file reads after the first block_run @@ -404,7 +404,7 @@ Version 0.4 (2001-10-28) * Fixed compile errors on 2.4.1 kernel (WD) Resolve rejected patches - Accomodate changed NLS interface (util.h) + Accommodate changed NLS interface (util.h) Needed to include <linux/slab.h> in most files Makefile changes fs/Config.in changes diff --git a/fs/befs/befs_fs_types.h b/fs/befs/befs_fs_types.h index 7893eaa1e58c..eb557d9dc8be 100644 --- a/fs/befs/befs_fs_types.h +++ b/fs/befs/befs_fs_types.h @@ -234,7 +234,7 @@ typedef struct { } PACKED befs_btree_super; /* - * Header stucture of each btree node + * Header structure of each btree node */ typedef struct { fs64 left; diff --git a/fs/befs/btree.c b/fs/befs/btree.c index 4202db7496cb..a66c9b1136e0 100644 --- a/fs/befs/btree.c +++ b/fs/befs/btree.c @@ -5,7 +5,7 @@ * * Licensed under the GNU GPL. See the file COPYING for details. * - * 2002-02-05: Sergey S. Kostyliov added binary search withing + * 2002-02-05: Sergey S. Kostyliov added binary search within * btree nodes. * * Many thanks to: diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 06457ed8f3e7..54b8c28bebc8 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -734,7 +734,7 @@ parse_options(char *options, befs_mount_options * opts) /* This function has the responsibiltiy of getting the * filesystem ready for unmounting. - * Basicly, we free everything that we allocated in + * Basically, we free everything that we allocated in * befs_read_inode */ static void diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index f34078d702d3..303983fabfd6 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -941,9 +941,13 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) current->mm->start_stack = bprm->p; #ifdef arch_randomize_brk - if ((current->flags & PF_RANDOMIZE) && (randomize_va_space > 1)) + if ((current->flags & PF_RANDOMIZE) && (randomize_va_space > 1)) { current->mm->brk = current->mm->start_brk = arch_randomize_brk(current->mm); +#ifdef CONFIG_COMPAT_BRK + current->brk_randomized = 1; +#endif + } #endif if (current->personality & MMAP_PAGE_ZERO) { diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 811384bec8de..397d3057d336 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -717,7 +717,7 @@ static int load_flat_file(struct linux_binprm * bprm, * help simplify all this mumbo jumbo * * We've got two different sections of relocation entries. - * The first is the GOT which resides at the begining of the data segment + * The first is the GOT which resides at the beginning of the data segment * and is terminated with a -1. This one can be relocated in place. * The second is the extra relocation entries tacked after the image's * data segment. These require a little more processing as the entry is @@ -1436,7 +1436,7 @@ EXPORT_SYMBOL(bio_flush_dcache_pages); * preferred way to end I/O on a bio, it takes care of clearing * BIO_UPTODATE on error. @error is 0 on success, and and one of the * established -Exxxx (-EIO, for instance) error values in case - * something went wrong. Noone should call bi_end_io() directly on a + * something went wrong. No one should call bi_end_io() directly on a * bio unless they own it and thus know that it has an end_io * function. **/ diff --git a/fs/block_dev.c b/fs/block_dev.c index c1511c674f53..257b00e98428 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -653,7 +653,7 @@ void bd_forget(struct inode *inode) * @whole: whole block device containing @bdev, may equal @bdev * @holder: holder trying to claim @bdev * - * Test whther @bdev can be claimed by @holder. + * Test whether @bdev can be claimed by @holder. * * CONTEXT: * spin_lock(&bdev_lock). @@ -1102,6 +1102,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) if (!bdev->bd_part) goto out_clear; + ret = 0; if (disk->fops->open) { ret = disk->fops->open(bdev, mode); if (ret == -ERESTARTSYS) { @@ -1118,9 +1119,18 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) put_disk(disk); goto restart; } - if (ret) - goto out_clear; } + /* + * If the device is invalidated, rescan partition + * if open succeeded or failed with -ENOMEDIUM. + * The latter is necessary to prevent ghost + * partitions on a removed medium. + */ + if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) + rescan_partitions(disk, bdev); + if (ret) + goto out_clear; + if (!bdev->bd_openers) { bd_set_size(bdev,(loff_t)get_capacity(disk)<<9); bdi = blk_get_backing_dev_info(bdev); @@ -1128,8 +1138,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) bdi = &default_backing_dev_info; bdev_inode_switch_bdi(bdev->bd_inode, bdi); } - if (bdev->bd_invalidated) - rescan_partitions(disk, bdev); } else { struct block_device *whole; whole = bdget_disk(disk, 0); @@ -1153,13 +1161,14 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) } } else { if (bdev->bd_contains == bdev) { - if (bdev->bd_disk->fops->open) { + ret = 0; + if (bdev->bd_disk->fops->open) ret = bdev->bd_disk->fops->open(bdev, mode); - if (ret) - goto out_unlock_bdev; - } - if (bdev->bd_invalidated) + /* the same as first opener case, read comment there */ + if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM)) rescan_partitions(bdev->bd_disk, bdev); + if (ret) + goto out_unlock_bdev; } /* only one opener holds refs to the module and disk */ module_put(disk->fops->owner); diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index de34bfad9ec3..44ea5b92e1ba 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -178,16 +178,18 @@ static int btrfs_xattr_acl_set(struct dentry *dentry, const char *name, if (value) { acl = posix_acl_from_xattr(value, size); - if (acl == NULL) { - value = NULL; - size = 0; - } else if (IS_ERR(acl)) { + if (IS_ERR(acl)) return PTR_ERR(acl); + + if (acl) { + ret = posix_acl_valid(acl); + if (ret) + goto out; } } ret = btrfs_set_acl(NULL, dentry->d_inode, acl, type); - +out: posix_acl_release(acl); return ret; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 3458b5725540..8f4b81de3ae2 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -718,7 +718,7 @@ struct btrfs_space_info { u64 total_bytes; /* total bytes in the space, this doesn't take mirrors into account */ u64 bytes_used; /* total bytes used, - this does't take mirrors into account */ + this doesn't take mirrors into account */ u64 bytes_pinned; /* total bytes pinned, will be freed when the transaction finishes */ u64 bytes_reserved; /* total bytes the allocator has reserved for @@ -740,8 +740,10 @@ struct btrfs_space_info { */ unsigned long reservation_progress; - int full; /* indicates that we cannot allocate any more + int full:1; /* indicates that we cannot allocate any more chunks for this space */ + int chunk_alloc:1; /* set if we are allocating a chunk */ + int force_alloc; /* set if we need to force a chunk alloc for this space */ @@ -2576,6 +2578,11 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode, int btrfs_mark_extent_written(struct btrfs_trans_handle *trans, struct inode *inode, u64 start, u64 end); int btrfs_release_file(struct inode *inode, struct file *file); +void btrfs_drop_pages(struct page **pages, size_t num_pages); +int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode, + struct page **pages, size_t num_pages, + loff_t pos, size_t write_bytes, + struct extent_state **cached); /* tree-defrag.c */ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 8f1d44ba332f..228cf36ece83 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2824,6 +2824,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, spin_lock(&delayed_refs->lock); if (delayed_refs->num_entries == 0) { + spin_unlock(&delayed_refs->lock); printk(KERN_INFO "delayed_refs has NO entry\n"); return ret; } @@ -3057,7 +3058,7 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root) btrfs_destroy_pinned_extent(root, root->fs_info->pinned_extents); - t->use_count = 0; + atomic_set(&t->use_count, 0); list_del_init(&t->list); memset(t, 0, sizeof(*t)); kmem_cache_free(btrfs_transaction_cachep, t); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index f619c3cb13b7..9ee6bd55e16c 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -33,6 +33,25 @@ #include "locking.h" #include "free-space-cache.h" +/* control flags for do_chunk_alloc's force field + * CHUNK_ALLOC_NO_FORCE means to only allocate a chunk + * if we really need one. + * + * CHUNK_ALLOC_FORCE means it must try to allocate one + * + * CHUNK_ALLOC_LIMITED means to only try and allocate one + * if we have very few chunks already allocated. This is + * used as part of the clustering code to help make sure + * we have a good pool of storage to cluster in, without + * filling the FS with empty chunks + * + */ +enum { + CHUNK_ALLOC_NO_FORCE = 0, + CHUNK_ALLOC_FORCE = 1, + CHUNK_ALLOC_LIMITED = 2, +}; + static int update_block_group(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytenr, u64 num_bytes, int alloc); @@ -3019,7 +3038,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags, found->bytes_readonly = 0; found->bytes_may_use = 0; found->full = 0; - found->force_alloc = 0; + found->force_alloc = CHUNK_ALLOC_NO_FORCE; + found->chunk_alloc = 0; *space_info = found; list_add_rcu(&found->list, &info->space_info); atomic_set(&found->caching_threads, 0); @@ -3150,7 +3170,7 @@ again: if (!data_sinfo->full && alloc_chunk) { u64 alloc_target; - data_sinfo->force_alloc = 1; + data_sinfo->force_alloc = CHUNK_ALLOC_FORCE; spin_unlock(&data_sinfo->lock); alloc: alloc_target = btrfs_get_alloc_profile(root, 1); @@ -3160,7 +3180,8 @@ alloc: ret = do_chunk_alloc(trans, root->fs_info->extent_root, bytes + 2 * 1024 * 1024, - alloc_target, 0); + alloc_target, + CHUNK_ALLOC_NO_FORCE); btrfs_end_transaction(trans, root); if (ret < 0) { if (ret != -ENOSPC) @@ -3239,31 +3260,56 @@ static void force_metadata_allocation(struct btrfs_fs_info *info) rcu_read_lock(); list_for_each_entry_rcu(found, head, list) { if (found->flags & BTRFS_BLOCK_GROUP_METADATA) - found->force_alloc = 1; + found->force_alloc = CHUNK_ALLOC_FORCE; } rcu_read_unlock(); } static int should_alloc_chunk(struct btrfs_root *root, - struct btrfs_space_info *sinfo, u64 alloc_bytes) + struct btrfs_space_info *sinfo, u64 alloc_bytes, + int force) { u64 num_bytes = sinfo->total_bytes - sinfo->bytes_readonly; + u64 num_allocated = sinfo->bytes_used + sinfo->bytes_reserved; u64 thresh; - if (sinfo->bytes_used + sinfo->bytes_reserved + - alloc_bytes + 256 * 1024 * 1024 < num_bytes) + if (force == CHUNK_ALLOC_FORCE) + return 1; + + /* + * in limited mode, we want to have some free space up to + * about 1% of the FS size. + */ + if (force == CHUNK_ALLOC_LIMITED) { + thresh = btrfs_super_total_bytes(&root->fs_info->super_copy); + thresh = max_t(u64, 64 * 1024 * 1024, + div_factor_fine(thresh, 1)); + + if (num_bytes - num_allocated < thresh) + return 1; + } + + /* + * we have two similar checks here, one based on percentage + * and once based on a hard number of 256MB. The idea + * is that if we have a good amount of free + * room, don't allocate a chunk. A good mount is + * less than 80% utilized of the chunks we have allocated, + * or more than 256MB free + */ + if (num_allocated + alloc_bytes + 256 * 1024 * 1024 < num_bytes) return 0; - if (sinfo->bytes_used + sinfo->bytes_reserved + - alloc_bytes < div_factor(num_bytes, 8)) + if (num_allocated + alloc_bytes < div_factor(num_bytes, 8)) return 0; thresh = btrfs_super_total_bytes(&root->fs_info->super_copy); + + /* 256MB or 5% of the FS */ thresh = max_t(u64, 256 * 1024 * 1024, div_factor_fine(thresh, 5)); if (num_bytes > thresh && sinfo->bytes_used < div_factor(num_bytes, 3)) return 0; - return 1; } @@ -3273,10 +3319,9 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, { struct btrfs_space_info *space_info; struct btrfs_fs_info *fs_info = extent_root->fs_info; + int wait_for_alloc = 0; int ret = 0; - mutex_lock(&fs_info->chunk_mutex); - flags = btrfs_reduce_alloc_profile(extent_root, flags); space_info = __find_space_info(extent_root->fs_info, flags); @@ -3287,21 +3332,40 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, } BUG_ON(!space_info); +again: spin_lock(&space_info->lock); if (space_info->force_alloc) - force = 1; + force = space_info->force_alloc; if (space_info->full) { spin_unlock(&space_info->lock); - goto out; + return 0; } - if (!force && !should_alloc_chunk(extent_root, space_info, - alloc_bytes)) { + if (!should_alloc_chunk(extent_root, space_info, alloc_bytes, force)) { spin_unlock(&space_info->lock); - goto out; + return 0; + } else if (space_info->chunk_alloc) { + wait_for_alloc = 1; + } else { + space_info->chunk_alloc = 1; } + spin_unlock(&space_info->lock); + mutex_lock(&fs_info->chunk_mutex); + + /* + * The chunk_mutex is held throughout the entirety of a chunk + * allocation, so once we've acquired the chunk_mutex we know that the + * other guy is done and we need to recheck and see if we should + * allocate. + */ + if (wait_for_alloc) { + mutex_unlock(&fs_info->chunk_mutex); + wait_for_alloc = 0; + goto again; + } + /* * If we have mixed data/metadata chunks we want to make sure we keep * allocating mixed chunks instead of individual chunks. @@ -3327,9 +3391,10 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, space_info->full = 1; else ret = 1; - space_info->force_alloc = 0; + + space_info->force_alloc = CHUNK_ALLOC_NO_FORCE; + space_info->chunk_alloc = 0; spin_unlock(&space_info->lock); -out: mutex_unlock(&extent_root->fs_info->chunk_mutex); return ret; } @@ -5303,11 +5368,13 @@ loop: if (allowed_chunk_alloc) { ret = do_chunk_alloc(trans, root, num_bytes + - 2 * 1024 * 1024, data, 1); + 2 * 1024 * 1024, data, + CHUNK_ALLOC_LIMITED); allowed_chunk_alloc = 0; done_chunk_alloc = 1; - } else if (!done_chunk_alloc) { - space_info->force_alloc = 1; + } else if (!done_chunk_alloc && + space_info->force_alloc == CHUNK_ALLOC_NO_FORCE) { + space_info->force_alloc = CHUNK_ALLOC_LIMITED; } if (loop < LOOP_NO_EMPTY_SIZE) { @@ -5393,7 +5460,8 @@ again: */ if (empty_size || root->ref_cows) ret = do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes + 2 * 1024 * 1024, data, 0); + num_bytes + 2 * 1024 * 1024, data, + CHUNK_ALLOC_NO_FORCE); WARN_ON(num_bytes < root->sectorsize); ret = find_free_extent(trans, root, num_bytes, empty_size, @@ -5405,7 +5473,7 @@ again: num_bytes = num_bytes & ~(root->sectorsize - 1); num_bytes = max(num_bytes, min_alloc_size); do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes, data, 1); + num_bytes, data, CHUNK_ALLOC_FORCE); goto again; } if (ret == -ENOSPC && btrfs_test_opt(root, ENOSPC_DEBUG)) { @@ -7991,6 +8059,10 @@ static noinline int relocate_one_extent(struct btrfs_root *extent_root, u64 group_start = group->key.objectid; new_extents = kmalloc(sizeof(*new_extents), GFP_NOFS); + if (!new_extents) { + ret = -ENOMEM; + goto out; + } nr_extents = 1; ret = get_new_locations(reloc_inode, extent_key, @@ -8109,13 +8181,15 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, alloc_flags = update_block_group_flags(root, cache->flags); if (alloc_flags != cache->flags) - do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1); + do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + CHUNK_ALLOC_FORCE); ret = set_block_group_ro(cache); if (!ret) goto out; alloc_flags = get_alloc_profile(root, cache->space_info->flags); - ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1); + ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + CHUNK_ALLOC_FORCE); if (ret < 0) goto out; ret = set_block_group_ro(cache); @@ -8128,7 +8202,8 @@ int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 type) { u64 alloc_flags = get_alloc_profile(root, type); - return do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1); + return do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, + CHUNK_ALLOC_FORCE); } /* @@ -8781,23 +8856,38 @@ out: int btrfs_init_space_info(struct btrfs_fs_info *fs_info) { struct btrfs_space_info *space_info; + struct btrfs_super_block *disk_super; + u64 features; + u64 flags; + int mixed = 0; int ret; - ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM, 0, 0, - &space_info); - if (ret) - return ret; + disk_super = &fs_info->super_copy; + if (!btrfs_super_root(disk_super)) + return 1; - ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA, 0, 0, - &space_info); - if (ret) - return ret; + features = btrfs_super_incompat_flags(disk_super); + if (features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS) + mixed = 1; - ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA, 0, 0, - &space_info); + flags = BTRFS_BLOCK_GROUP_SYSTEM; + ret = update_space_info(fs_info, flags, 0, 0, &space_info); if (ret) - return ret; + goto out; + if (mixed) { + flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA; + ret = update_space_info(fs_info, flags, 0, 0, &space_info); + } else { + flags = BTRFS_BLOCK_GROUP_METADATA; + ret = update_space_info(fs_info, flags, 0, 0, &space_info); + if (ret) + goto out; + + flags = BTRFS_BLOCK_GROUP_DATA; + ret = update_space_info(fs_info, flags, 0, 0, &space_info); + } +out: return ret; } diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 20ddb28602a8..96fcfa522dab 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -10,6 +10,7 @@ #include <linux/swap.h> #include <linux/writeback.h> #include <linux/pagevec.h> +#include <linux/prefetch.h> #include "extent_io.h" #include "extent_map.h" #include "compat.h" @@ -690,6 +691,15 @@ static void cache_state(struct extent_state *state, } } +static void uncache_state(struct extent_state **cached_ptr) +{ + if (cached_ptr && (*cached_ptr)) { + struct extent_state *state = *cached_ptr; + *cached_ptr = NULL; + free_extent_state(state); + } +} + /* * set some bits on a range in the tree. This may require allocations or * sleeping, so the gfp mask is used to indicate what is allowed. @@ -940,10 +950,10 @@ static int clear_extent_new(struct extent_io_tree *tree, u64 start, u64 end, } int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end, - gfp_t mask) + struct extent_state **cached_state, gfp_t mask) { - return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, NULL, - NULL, mask); + return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, + NULL, cached_state, mask); } static int clear_extent_uptodate(struct extent_io_tree *tree, u64 start, @@ -1012,8 +1022,7 @@ int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end, mask); } -int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end, - gfp_t mask) +int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask) { return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL, mask); @@ -1735,6 +1744,9 @@ static void end_bio_extent_readpage(struct bio *bio, int err) do { struct page *page = bvec->bv_page; + struct extent_state *cached = NULL; + struct extent_state *state; + tree = &BTRFS_I(page->mapping->host)->io_tree; start = ((u64)page->index << PAGE_CACHE_SHIFT) + @@ -1749,9 +1761,20 @@ static void end_bio_extent_readpage(struct bio *bio, int err) if (++bvec <= bvec_end) prefetchw(&bvec->bv_page->flags); + spin_lock(&tree->lock); + state = find_first_extent_bit_state(tree, start, EXTENT_LOCKED); + if (state && state->start == start) { + /* + * take a reference on the state, unlock will drop + * the ref + */ + cache_state(state, &cached); + } + spin_unlock(&tree->lock); + if (uptodate && tree->ops && tree->ops->readpage_end_io_hook) { ret = tree->ops->readpage_end_io_hook(page, start, end, - NULL); + state); if (ret) uptodate = 0; } @@ -1764,15 +1787,16 @@ static void end_bio_extent_readpage(struct bio *bio, int err) test_bit(BIO_UPTODATE, &bio->bi_flags); if (err) uptodate = 0; + uncache_state(&cached); continue; } } if (uptodate) { - set_extent_uptodate(tree, start, end, + set_extent_uptodate(tree, start, end, &cached, GFP_ATOMIC); } - unlock_extent(tree, start, end, GFP_ATOMIC); + unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC); if (whole_page) { if (uptodate) { @@ -1811,6 +1835,7 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err) do { struct page *page = bvec->bv_page; + struct extent_state *cached = NULL; tree = &BTRFS_I(page->mapping->host)->io_tree; start = ((u64)page->index << PAGE_CACHE_SHIFT) + @@ -1821,13 +1846,14 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err) prefetchw(&bvec->bv_page->flags); if (uptodate) { - set_extent_uptodate(tree, start, end, GFP_ATOMIC); + set_extent_uptodate(tree, start, end, &cached, + GFP_ATOMIC); } else { ClearPageUptodate(page); SetPageError(page); } - unlock_extent(tree, start, end, GFP_ATOMIC); + unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC); } while (bvec >= bio->bi_io_vec); @@ -2016,14 +2042,17 @@ static int __extent_read_full_page(struct extent_io_tree *tree, while (cur <= end) { if (cur >= last_byte) { char *userpage; + struct extent_state *cached = NULL; + iosize = PAGE_CACHE_SIZE - page_offset; userpage = kmap_atomic(page, KM_USER0); memset(userpage + page_offset, 0, iosize); flush_dcache_page(page); kunmap_atomic(userpage, KM_USER0); set_extent_uptodate(tree, cur, cur + iosize - 1, - GFP_NOFS); - unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS); + &cached, GFP_NOFS); + unlock_extent_cached(tree, cur, cur + iosize - 1, + &cached, GFP_NOFS); break; } em = get_extent(inode, page, page_offset, cur, @@ -2063,14 +2092,17 @@ static int __extent_read_full_page(struct extent_io_tree *tree, /* we've found a hole, just zero and go on */ if (block_start == EXTENT_MAP_HOLE) { char *userpage; + struct extent_state *cached = NULL; + userpage = kmap_atomic(page, KM_USER0); memset(userpage + page_offset, 0, iosize); flush_dcache_page(page); kunmap_atomic(userpage, KM_USER0); set_extent_uptodate(tree, cur, cur + iosize - 1, - GFP_NOFS); - unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS); + &cached, GFP_NOFS); + unlock_extent_cached(tree, cur, cur + iosize - 1, + &cached, GFP_NOFS); cur = cur + iosize; page_offset += iosize; continue; @@ -2650,7 +2682,7 @@ int extent_readpages(struct extent_io_tree *tree, prefetchw(&page->flags); list_del(&page->lru); if (!add_to_page_cache_lru(page, mapping, - page->index, GFP_KERNEL)) { + page->index, GFP_NOFS)) { __extent_read_full_page(tree, page, get_extent, &bio, 0, &bio_flags); } @@ -2789,9 +2821,12 @@ int extent_prepare_write(struct extent_io_tree *tree, iocount++; block_start = block_start + iosize; } else { - set_extent_uptodate(tree, block_start, cur_end, + struct extent_state *cached = NULL; + + set_extent_uptodate(tree, block_start, cur_end, &cached, GFP_NOFS); - unlock_extent(tree, block_start, cur_end, GFP_NOFS); + unlock_extent_cached(tree, block_start, cur_end, + &cached, GFP_NOFS); block_start = cur_end + 1; } page_offset = block_start & (PAGE_CACHE_SIZE - 1); @@ -3457,7 +3492,7 @@ int set_extent_buffer_uptodate(struct extent_io_tree *tree, num_pages = num_extent_pages(eb->start, eb->len); set_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, - GFP_NOFS); + NULL, GFP_NOFS); for (i = 0; i < num_pages; i++) { page = extent_buffer_page(eb, i); if ((i == 0 && (eb->start & (PAGE_CACHE_SIZE - 1))) || @@ -3885,6 +3920,12 @@ static void move_pages(struct page *dst_page, struct page *src_page, kunmap_atomic(dst_kaddr, KM_USER0); } +static inline bool areas_overlap(unsigned long src, unsigned long dst, unsigned long len) +{ + unsigned long distance = (src > dst) ? src - dst : dst - src; + return distance < len; +} + static void copy_pages(struct page *dst_page, struct page *src_page, unsigned long dst_off, unsigned long src_off, unsigned long len) @@ -3892,10 +3933,12 @@ static void copy_pages(struct page *dst_page, struct page *src_page, char *dst_kaddr = kmap_atomic(dst_page, KM_USER0); char *src_kaddr; - if (dst_page != src_page) + if (dst_page != src_page) { src_kaddr = kmap_atomic(src_page, KM_USER1); - else + } else { src_kaddr = dst_kaddr; + BUG_ON(areas_overlap(src_off, dst_off, len)); + } memcpy(dst_kaddr + dst_off, src_kaddr + src_off, len); kunmap_atomic(dst_kaddr, KM_USER0); @@ -3970,7 +4013,7 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset, "len %lu len %lu\n", dst_offset, len, dst->len); BUG_ON(1); } - if (dst_offset < src_offset) { + if (!areas_overlap(src_offset, dst_offset, len)) { memcpy_extent_buffer(dst, dst_offset, src_offset, len); return; } diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index f62c5442835d..af2d7179c372 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -208,7 +208,7 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits, int exclusive_bits, u64 *failed_start, struct extent_state **cached_state, gfp_t mask); int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end, - gfp_t mask); + struct extent_state **cached_state, gfp_t mask); int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask); int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end, diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 2b6c12e983b3..a24a3f2fa13e 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -243,7 +243,7 @@ out: * Insert @em into @tree or perform a simple forward/backward merge with * existing mappings. The extent_map struct passed in will be inserted * into the tree directly, with an additional reference taken, or a - * reference dropped if the merge attempt was successfull. + * reference dropped if the merge attempt was successful. */ int add_extent_mapping(struct extent_map_tree *tree, struct extent_map *em) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index e621ea54a3fd..75899a01dded 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -104,7 +104,7 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages, /* * unlocks pages after btrfs_file_write is done with them */ -static noinline void btrfs_drop_pages(struct page **pages, size_t num_pages) +void btrfs_drop_pages(struct page **pages, size_t num_pages) { size_t i; for (i = 0; i < num_pages; i++) { @@ -127,16 +127,13 @@ static noinline void btrfs_drop_pages(struct page **pages, size_t num_pages) * this also makes the decision about creating an inline extent vs * doing real data extents, marking pages dirty and delalloc as required. */ -static noinline int dirty_and_release_pages(struct btrfs_root *root, - struct file *file, - struct page **pages, - size_t num_pages, - loff_t pos, - size_t write_bytes) +int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode, + struct page **pages, size_t num_pages, + loff_t pos, size_t write_bytes, + struct extent_state **cached) { int err = 0; int i; - struct inode *inode = fdentry(file)->d_inode; u64 num_bytes; u64 start_pos; u64 end_of_last_block; @@ -149,7 +146,7 @@ static noinline int dirty_and_release_pages(struct btrfs_root *root, end_of_last_block = start_pos + num_bytes - 1; err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block, - NULL); + cached); if (err) return err; @@ -992,9 +989,9 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, } if (copied > 0) { - ret = dirty_and_release_pages(root, file, pages, - dirty_pages, pos, - copied); + ret = btrfs_dirty_pages(root, inode, pages, + dirty_pages, pos, copied, + NULL); if (ret) { btrfs_delalloc_release_space(inode, dirty_pages << PAGE_CACHE_SHIFT); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index f561c953205b..63731a1fb0a1 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -508,6 +508,7 @@ int btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode; struct rb_node *node; struct list_head *pos, *n; + struct page **pages; struct page *page; struct extent_state *cached_state = NULL; struct btrfs_free_cluster *cluster = NULL; @@ -517,13 +518,13 @@ int btrfs_write_out_cache(struct btrfs_root *root, u64 start, end, len; u64 bytes = 0; u32 *crc, *checksums; - pgoff_t index = 0, last_index = 0; unsigned long first_page_offset; - int num_checksums; + int index = 0, num_pages = 0; int entries = 0; int bitmaps = 0; int ret = 0; bool next_page = false; + bool out_of_space = false; root = root->fs_info->tree_root; @@ -551,24 +552,31 @@ int btrfs_write_out_cache(struct btrfs_root *root, return 0; } - last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT; + num_pages = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> + PAGE_CACHE_SHIFT; filemap_write_and_wait(inode->i_mapping); btrfs_wait_ordered_range(inode, inode->i_size & ~(root->sectorsize - 1), (u64)-1); /* We need a checksum per page. */ - num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE; - crc = checksums = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS); + crc = checksums = kzalloc(sizeof(u32) * num_pages, GFP_NOFS); if (!crc) { iput(inode); return 0; } + pages = kzalloc(sizeof(struct page *) * num_pages, GFP_NOFS); + if (!pages) { + kfree(crc); + iput(inode); + return 0; + } + /* Since the first page has all of our checksums and our generation we * need to calculate the offset into the page that we can start writing * our entries. */ - first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64); + first_page_offset = (sizeof(u32) * num_pages) + sizeof(u64); /* Get the cluster for this block_group if it exists */ if (!list_empty(&block_group->cluster_list)) @@ -590,20 +598,18 @@ int btrfs_write_out_cache(struct btrfs_root *root, * after find_get_page at this point. Just putting this here so people * know and don't freak out. */ - while (index <= last_index) { + while (index < num_pages) { page = grab_cache_page(inode->i_mapping, index); if (!page) { - pgoff_t i = 0; + int i; - while (i < index) { - page = find_get_page(inode->i_mapping, i); - unlock_page(page); - page_cache_release(page); - page_cache_release(page); - i++; + for (i = 0; i < num_pages; i++) { + unlock_page(pages[i]); + page_cache_release(pages[i]); } goto out_free; } + pages[index] = page; index++; } @@ -631,7 +637,12 @@ int btrfs_write_out_cache(struct btrfs_root *root, offset = start_offset; } - page = find_get_page(inode->i_mapping, index); + if (index >= num_pages) { + out_of_space = true; + break; + } + + page = pages[index]; addr = kmap(page); entry = addr + start_offset; @@ -708,23 +719,6 @@ int btrfs_write_out_cache(struct btrfs_root *root, bytes += PAGE_CACHE_SIZE; - ClearPageChecked(page); - set_page_extent_mapped(page); - SetPageUptodate(page); - set_page_dirty(page); - - /* - * We need to release our reference we got for grab_cache_page, - * except for the first page which will hold our checksums, we - * do that below. - */ - if (index != 0) { - unlock_page(page); - page_cache_release(page); - } - - page_cache_release(page); - index++; } while (node || next_page); @@ -734,7 +728,11 @@ int btrfs_write_out_cache(struct btrfs_root *root, struct btrfs_free_space *entry = list_entry(pos, struct btrfs_free_space, list); - page = find_get_page(inode->i_mapping, index); + if (index >= num_pages) { + out_of_space = true; + break; + } + page = pages[index]; addr = kmap(page); memcpy(addr, entry->bitmap, PAGE_CACHE_SIZE); @@ -745,64 +743,58 @@ int btrfs_write_out_cache(struct btrfs_root *root, crc++; bytes += PAGE_CACHE_SIZE; - ClearPageChecked(page); - set_page_extent_mapped(page); - SetPageUptodate(page); - set_page_dirty(page); - unlock_page(page); - page_cache_release(page); - page_cache_release(page); list_del_init(&entry->list); index++; } + if (out_of_space) { + btrfs_drop_pages(pages, num_pages); + unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0, + i_size_read(inode) - 1, &cached_state, + GFP_NOFS); + ret = 0; + goto out_free; + } + /* Zero out the rest of the pages just to make sure */ - while (index <= last_index) { + while (index < num_pages) { void *addr; - page = find_get_page(inode->i_mapping, index); - + page = pages[index]; addr = kmap(page); memset(addr, 0, PAGE_CACHE_SIZE); kunmap(page); - ClearPageChecked(page); - set_page_extent_mapped(page); - SetPageUptodate(page); - set_page_dirty(page); - unlock_page(page); - page_cache_release(page); - page_cache_release(page); bytes += PAGE_CACHE_SIZE; index++; } - btrfs_set_extent_delalloc(inode, 0, bytes - 1, &cached_state); - /* Write the checksums and trans id to the first page */ { void *addr; u64 *gen; - page = find_get_page(inode->i_mapping, 0); + page = pages[0]; addr = kmap(page); - memcpy(addr, checksums, sizeof(u32) * num_checksums); - gen = addr + (sizeof(u32) * num_checksums); + memcpy(addr, checksums, sizeof(u32) * num_pages); + gen = addr + (sizeof(u32) * num_pages); *gen = trans->transid; kunmap(page); - ClearPageChecked(page); - set_page_extent_mapped(page); - SetPageUptodate(page); - set_page_dirty(page); - unlock_page(page); - page_cache_release(page); - page_cache_release(page); } - BTRFS_I(inode)->generation = trans->transid; + ret = btrfs_dirty_pages(root, inode, pages, num_pages, 0, + bytes, &cached_state); + btrfs_drop_pages(pages, num_pages); unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, &cached_state, GFP_NOFS); + if (ret) { + ret = 0; + goto out_free; + } + + BTRFS_I(inode)->generation = trans->transid; + filemap_write_and_wait(inode->i_mapping); key.objectid = BTRFS_FREE_SPACE_OBJECTID; @@ -853,6 +845,7 @@ out_free: BTRFS_I(inode)->generation = 0; } kfree(checksums); + kfree(pages); btrfs_update_inode(trans, root, inode); iput(inode); return ret; @@ -1775,10 +1768,13 @@ void btrfs_remove_free_space_cache(struct btrfs_block_group_cache *block_group) while ((node = rb_last(&block_group->free_space_offset)) != NULL) { info = rb_entry(node, struct btrfs_free_space, offset_index); - unlink_free_space(block_group, info); - if (info->bitmap) - kfree(info->bitmap); - kmem_cache_free(btrfs_free_space_cachep, info); + if (!info->bitmap) { + unlink_free_space(block_group, info); + kmem_cache_free(btrfs_free_space_cachep, info); + } else { + free_bitmap(block_group, info); + } + if (need_resched()) { spin_unlock(&block_group->tree_lock); cond_resched(); @@ -2308,7 +2304,7 @@ int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group, start = entry->offset; bytes = min(entry->bytes, end - start); unlink_free_space(block_group, entry); - kfree(entry); + kmem_cache_free(btrfs_free_space_cachep, entry); } spin_unlock(&block_group->tree_lock); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 65413394daef..7cd8ab0ef04d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -954,6 +954,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page, 1, 0, NULL, GFP_NOFS); while (start < end) { async_cow = kmalloc(sizeof(*async_cow), GFP_NOFS); + BUG_ON(!async_cow); async_cow->inode = inode; async_cow->root = root; async_cow->locked_page = locked_page; @@ -1770,9 +1771,12 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) add_pending_csums(trans, inode, ordered_extent->file_offset, &ordered_extent->list); - btrfs_ordered_update_i_size(inode, 0, ordered_extent); - ret = btrfs_update_inode(trans, root, inode); - BUG_ON(ret); + ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent); + if (!ret) { + ret = btrfs_update_inode(trans, root, inode); + BUG_ON(ret); + } + ret = 0; out: if (nolock) { if (trans) @@ -2321,7 +2325,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) /* * if ret == 0 means we found what we were searching for, which - * is weird, but possible, so only screw with path if we didnt + * is weird, but possible, so only screw with path if we didn't * find the key and see if we have stuff that matches */ if (ret > 0) { @@ -2590,6 +2594,13 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, struct btrfs_inode_item *item, struct inode *inode) { + if (!leaf->map_token) + map_private_extent_buffer(leaf, (unsigned long)item, + sizeof(struct btrfs_inode_item), + &leaf->map_token, &leaf->kaddr, + &leaf->map_start, &leaf->map_len, + KM_USER1); + btrfs_set_inode_uid(leaf, item, inode->i_uid); btrfs_set_inode_gid(leaf, item, inode->i_gid); btrfs_set_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size); @@ -2618,6 +2629,11 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, btrfs_set_inode_rdev(leaf, item, inode->i_rdev); btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags); btrfs_set_inode_block_group(leaf, item, BTRFS_I(inode)->block_group); + + if (leaf->map_token) { + unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); + leaf->map_token = NULL; + } } /* @@ -4207,10 +4223,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, struct btrfs_key found_key; struct btrfs_path *path; int ret; - u32 nritems; struct extent_buffer *leaf; int slot; - int advance; unsigned char d_type; int over = 0; u32 di_cur; @@ -4253,27 +4267,19 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) goto err; - advance = 0; while (1) { leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); slot = path->slots[0]; - if (advance || slot >= nritems) { - if (slot >= nritems - 1) { - ret = btrfs_next_leaf(root, path); - if (ret) - break; - leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); - slot = path->slots[0]; - } else { - slot++; - path->slots[0]++; - } + if (slot >= btrfs_header_nritems(leaf)) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) + goto err; + else if (ret > 0) + break; + continue; } - advance = 1; item = btrfs_item_nr(leaf, slot); btrfs_item_key_to_cpu(leaf, &found_key, slot); @@ -4282,7 +4288,7 @@ static int btrfs_real_readdir(struct file *filp, void *dirent, if (btrfs_key_type(&found_key) != key_type) break; if (found_key.offset < filp->f_pos) - continue; + goto next; filp->f_pos = found_key.offset; @@ -4335,6 +4341,8 @@ skip: di_cur += di_len; di = (struct btrfs_dir_item *)((char *)di + di_len); } +next: + path->slots[0]++; } /* Reached end of directory/root. Bump pos past the last item. */ @@ -4527,14 +4535,17 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, BUG_ON(!path); inode = new_inode(root->fs_info->sb); - if (!inode) + if (!inode) { + btrfs_free_path(path); return ERR_PTR(-ENOMEM); + } if (dir) { trace_btrfs_inode_request(dir); ret = btrfs_set_inode_index(dir, index); if (ret) { + btrfs_free_path(path); iput(inode); return ERR_PTR(ret); } @@ -4721,9 +4732,10 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, mode, &index); - err = PTR_ERR(inode); - if (IS_ERR(inode)) + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto out_unlock; + } err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); if (err) { @@ -4782,9 +4794,10 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, mode, &index); - err = PTR_ERR(inode); - if (IS_ERR(inode)) + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto out_unlock; + } err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); if (err) { @@ -4834,9 +4847,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, if (inode->i_nlink == ~0U) return -EMLINK; - btrfs_inc_nlink(inode); - inode->i_ctime = CURRENT_TIME; - err = btrfs_set_inode_index(dir, &index); if (err) goto fail; @@ -4852,6 +4862,9 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, goto fail; } + btrfs_inc_nlink(inode); + inode->i_ctime = CURRENT_TIME; + btrfs_set_trans_block_group(trans, dir); ihold(inode); @@ -4989,6 +5002,8 @@ static noinline int uncompress_inline(struct btrfs_path *path, inline_size = btrfs_file_extent_inline_item_len(leaf, btrfs_item_nr(leaf, path->slots[0])); tmp = kmalloc(inline_size, GFP_NOFS); + if (!tmp) + return -ENOMEM; ptr = btrfs_file_extent_inline_start(item); read_extent_buffer(leaf, tmp, ptr, inline_size); @@ -5221,7 +5236,7 @@ again: btrfs_mark_buffer_dirty(leaf); } set_extent_uptodate(io_tree, em->start, - extent_map_end(em) - 1, GFP_NOFS); + extent_map_end(em) - 1, NULL, GFP_NOFS); goto insert; } else { printk(KERN_ERR "btrfs unknown found_type %d\n", found_type); @@ -5428,17 +5443,30 @@ out: } static struct extent_map *btrfs_new_extent_direct(struct inode *inode, + struct extent_map *em, u64 start, u64 len) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; - struct extent_map *em; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; struct btrfs_key ins; u64 alloc_hint; int ret; + bool insert = false; - btrfs_drop_extent_cache(inode, start, start + len - 1, 0); + /* + * Ok if the extent map we looked up is a hole and is for the exact + * range we want, there is no reason to allocate a new one, however if + * it is not right then we need to free this one and drop the cache for + * our range. + */ + if (em->block_start != EXTENT_MAP_HOLE || em->start != start || + em->len != len) { + free_extent_map(em); + em = NULL; + insert = true; + btrfs_drop_extent_cache(inode, start, start + len - 1, 0); + } trans = btrfs_join_transaction(root, 0); if (IS_ERR(trans)) @@ -5454,10 +5482,12 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, goto out; } - em = alloc_extent_map(GFP_NOFS); if (!em) { - em = ERR_PTR(-ENOMEM); - goto out; + em = alloc_extent_map(GFP_NOFS); + if (!em) { + em = ERR_PTR(-ENOMEM); + goto out; + } } em->start = start; @@ -5467,9 +5497,15 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, em->block_start = ins.objectid; em->block_len = ins.offset; em->bdev = root->fs_info->fs_devices->latest_bdev; + + /* + * We need to do this because if we're using the original em we searched + * for, we could have EXTENT_FLAG_VACANCY set, and we don't want that. + */ + em->flags = 0; set_bit(EXTENT_FLAG_PINNED, &em->flags); - while (1) { + while (insert) { write_lock(&em_tree->lock); ret = add_extent_mapping(em_tree, em); write_unlock(&em_tree->lock); @@ -5687,8 +5723,7 @@ must_cow: * it above */ len = bh_result->b_size; - free_extent_map(em); - em = btrfs_new_extent_direct(inode, start, len); + em = btrfs_new_extent_direct(inode, em, start, len); if (IS_ERR(em)) return PTR_ERR(em); len = min(len, em->len - (start - em->start)); @@ -5851,8 +5886,10 @@ again: } add_pending_csums(trans, inode, ordered->file_offset, &ordered->list); - btrfs_ordered_update_i_size(inode, 0, ordered); - btrfs_update_inode(trans, root, inode); + ret = btrfs_ordered_update_i_size(inode, 0, ordered); + if (!ret) + btrfs_update_inode(trans, root, inode); + ret = 0; out_unlock: unlock_extent_cached(&BTRFS_I(inode)->io_tree, ordered->file_offset, ordered->file_offset + ordered->len - 1, @@ -5938,7 +5975,7 @@ static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev, static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, int rw, u64 file_offset, int skip_sum, - u32 *csums) + u32 *csums, int async_submit) { int write = rw & REQ_WRITE; struct btrfs_root *root = BTRFS_I(inode)->root; @@ -5949,13 +5986,24 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, if (ret) goto err; - if (write && !skip_sum) { + if (skip_sum) + goto map; + + if (write && async_submit) { ret = btrfs_wq_submit_bio(root->fs_info, inode, rw, bio, 0, 0, file_offset, __btrfs_submit_bio_start_direct_io, __btrfs_submit_bio_done); goto err; + } else if (write) { + /* + * If we aren't doing async submit, calculate the csum of the + * bio now. + */ + ret = btrfs_csum_one_bio(root, inode, bio, file_offset, 1); + if (ret) + goto err; } else if (!skip_sum) { ret = btrfs_lookup_bio_sums_dio(root, inode, bio, file_offset, csums); @@ -5963,7 +6011,8 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, goto err; } - ret = btrfs_map_bio(root, rw, bio, 0, 1); +map: + ret = btrfs_map_bio(root, rw, bio, 0, async_submit); err: bio_put(bio); return ret; @@ -5985,23 +6034,30 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, int nr_pages = 0; u32 *csums = dip->csums; int ret = 0; + int async_submit = 0; int write = rw & REQ_WRITE; - bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS); - if (!bio) - return -ENOMEM; - bio->bi_private = dip; - bio->bi_end_io = btrfs_end_dio_bio; - atomic_inc(&dip->pending_bios); - map_length = orig_bio->bi_size; ret = btrfs_map_block(map_tree, READ, start_sector << 9, &map_length, NULL, 0); if (ret) { - bio_put(bio); + bio_put(orig_bio); return -EIO; } + if (map_length >= orig_bio->bi_size) { + bio = orig_bio; + goto submit; + } + + async_submit = 1; + bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS); + if (!bio) + return -ENOMEM; + bio->bi_private = dip; + bio->bi_end_io = btrfs_end_dio_bio; + atomic_inc(&dip->pending_bios); + while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) { if (unlikely(map_length < submit_len + bvec->bv_len || bio_add_page(bio, bvec->bv_page, bvec->bv_len, @@ -6015,7 +6071,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, atomic_inc(&dip->pending_bios); ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum, - csums); + csums, async_submit); if (ret) { bio_put(bio); atomic_dec(&dip->pending_bios); @@ -6052,8 +6108,9 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, } } +submit: ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum, - csums); + csums, async_submit); if (!ret) return 0; @@ -6148,6 +6205,7 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io unsigned long nr_segs) { int seg; + int i; size_t size; unsigned long addr; unsigned blocksize_mask = root->sectorsize - 1; @@ -6162,8 +6220,22 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io addr = (unsigned long)iov[seg].iov_base; size = iov[seg].iov_len; end += size; - if ((addr & blocksize_mask) || (size & blocksize_mask)) + if ((addr & blocksize_mask) || (size & blocksize_mask)) goto out; + + /* If this is a write we don't need to check anymore */ + if (rw & WRITE) + continue; + + /* + * Check to make sure we don't have duplicate iov_base's in this + * iovec, if so return EINVAL, otherwise we'll get csum errors + * when reading back. + */ + for (i = seg + 1; i < nr_segs; i++) { + if (iov[seg].iov_base == iov[i].iov_base) + goto out; + } } retval = 0; out: @@ -7206,9 +7278,10 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO, &index); - err = PTR_ERR(inode); - if (IS_ERR(inode)) + if (IS_ERR(inode)) { + err = PTR_ERR(inode); goto out_unlock; + } err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name); if (err) { diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index cfc264fefdb0..2616f7ed4799 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -81,6 +81,13 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags) iflags |= FS_NOATIME_FL; if (flags & BTRFS_INODE_DIRSYNC) iflags |= FS_DIRSYNC_FL; + if (flags & BTRFS_INODE_NODATACOW) + iflags |= FS_NOCOW_FL; + + if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS)) + iflags |= FS_COMPR_FL; + else if (flags & BTRFS_INODE_NOCOMPRESS) + iflags |= FS_NOCOMP_FL; return iflags; } @@ -144,16 +151,13 @@ static int check_flags(unsigned int flags) if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ FS_NOATIME_FL | FS_NODUMP_FL | \ FS_SYNC_FL | FS_DIRSYNC_FL | \ - FS_NOCOMP_FL | FS_COMPR_FL | \ - FS_NOCOW_FL | FS_COW_FL)) + FS_NOCOMP_FL | FS_COMPR_FL | + FS_NOCOW_FL)) return -EOPNOTSUPP; if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL)) return -EINVAL; - if ((flags & FS_NOCOW_FL) && (flags & FS_COW_FL)) - return -EINVAL; - return 0; } @@ -218,6 +222,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) ip->flags |= BTRFS_INODE_DIRSYNC; else ip->flags &= ~BTRFS_INODE_DIRSYNC; + if (flags & FS_NOCOW_FL) + ip->flags |= BTRFS_INODE_NODATACOW; + else + ip->flags &= ~BTRFS_INODE_NODATACOW; /* * The COMPRESS flag can only be changed by users, while the NOCOMPRESS @@ -230,11 +238,9 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) } else if (flags & FS_COMPR_FL) { ip->flags |= BTRFS_INODE_COMPRESS; ip->flags &= ~BTRFS_INODE_NOCOMPRESS; + } else { + ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS); } - if (flags & FS_NOCOW_FL) - ip->flags |= BTRFS_INODE_NODATACOW; - else if (flags & FS_COW_FL) - ip->flags &= ~BTRFS_INODE_NODATACOW; trans = btrfs_join_transaction(root, 1); BUG_ON(IS_ERR(trans)); @@ -2287,7 +2293,7 @@ long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) struct btrfs_ioctl_space_info space; struct btrfs_ioctl_space_info *dest; struct btrfs_ioctl_space_info *dest_orig; - struct btrfs_ioctl_space_info *user_dest; + struct btrfs_ioctl_space_info __user *user_dest; struct btrfs_space_info *info; u64 types[] = {BTRFS_BLOCK_GROUP_DATA, BTRFS_BLOCK_GROUP_SYSTEM, diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 58250e09eb05..199a80134312 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2346,7 +2346,7 @@ struct btrfs_root *select_one_root(struct btrfs_trans_handle *trans, root = next->root; BUG_ON(!root); - /* no other choice for non-refernce counted tree */ + /* no other choice for non-references counted tree */ if (!root->ref_cows) return root; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 58e7de9cc90c..0ac712efcdf2 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -159,7 +159,7 @@ enum { Opt_compress_type, Opt_compress_force, Opt_compress_force_type, Opt_notreelog, Opt_ratio, Opt_flushoncommit, Opt_discard, Opt_space_cache, Opt_clear_cache, Opt_user_subvol_rm_allowed, - Opt_enospc_debug, Opt_err, + Opt_enospc_debug, Opt_subvolrootid, Opt_err, }; static match_table_t tokens = { @@ -189,6 +189,7 @@ static match_table_t tokens = { {Opt_clear_cache, "clear_cache"}, {Opt_user_subvol_rm_allowed, "user_subvol_rm_allowed"}, {Opt_enospc_debug, "enospc_debug"}, + {Opt_subvolrootid, "subvolrootid=%d"}, {Opt_err, NULL}, }; @@ -232,6 +233,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) break; case Opt_subvol: case Opt_subvolid: + case Opt_subvolrootid: case Opt_device: /* * These are parsed by btrfs_parse_early_options @@ -388,7 +390,7 @@ out: */ static int btrfs_parse_early_options(const char *options, fmode_t flags, void *holder, char **subvol_name, u64 *subvol_objectid, - struct btrfs_fs_devices **fs_devices) + u64 *subvol_rootid, struct btrfs_fs_devices **fs_devices) { substring_t args[MAX_OPT_ARGS]; char *opts, *orig, *p; @@ -429,6 +431,18 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags, *subvol_objectid = intarg; } break; + case Opt_subvolrootid: + intarg = 0; + error = match_int(&args[0], &intarg); + if (!error) { + /* we want the original fs_tree */ + if (!intarg) + *subvol_rootid = + BTRFS_FS_TREE_OBJECTID; + else + *subvol_rootid = intarg; + } + break; case Opt_device: error = btrfs_scan_one_device(match_strdup(&args[0]), flags, holder, fs_devices); @@ -736,6 +750,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, fmode_t mode = FMODE_READ; char *subvol_name = NULL; u64 subvol_objectid = 0; + u64 subvol_rootid = 0; int error = 0; if (!(flags & MS_RDONLY)) @@ -743,7 +758,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, error = btrfs_parse_early_options(data, mode, fs_type, &subvol_name, &subvol_objectid, - &fs_devices); + &subvol_rootid, &fs_devices); if (error) return ERR_PTR(error); @@ -807,15 +822,17 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, s->s_flags |= MS_ACTIVE; } - root = get_default_root(s, subvol_objectid); - if (IS_ERR(root)) { - error = PTR_ERR(root); - deactivate_locked_super(s); - goto error_free_subvol_name; - } /* if they gave us a subvolume name bind mount into that */ if (strcmp(subvol_name, ".")) { struct dentry *new_root; + + root = get_default_root(s, subvol_rootid); + if (IS_ERR(root)) { + error = PTR_ERR(root); + deactivate_locked_super(s); + goto error_free_subvol_name; + } + mutex_lock(&root->d_inode->i_mutex); new_root = lookup_one_len(subvol_name, root, strlen(subvol_name)); @@ -836,6 +853,13 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, } dput(root); root = new_root; + } else { + root = get_default_root(s, subvol_objectid); + if (IS_ERR(root)) { + error = PTR_ERR(root); + deactivate_locked_super(s); + goto error_free_subvol_name; + } } kfree(subvol_name); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 5b158da7e0bb..c571734d5e5a 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -32,10 +32,8 @@ static noinline void put_transaction(struct btrfs_transaction *transaction) { - WARN_ON(transaction->use_count == 0); - transaction->use_count--; - if (transaction->use_count == 0) { - list_del_init(&transaction->list); + WARN_ON(atomic_read(&transaction->use_count) == 0); + if (atomic_dec_and_test(&transaction->use_count)) { memset(transaction, 0, sizeof(*transaction)); kmem_cache_free(btrfs_transaction_cachep, transaction); } @@ -60,14 +58,14 @@ static noinline int join_transaction(struct btrfs_root *root) if (!cur_trans) return -ENOMEM; root->fs_info->generation++; - cur_trans->num_writers = 1; + atomic_set(&cur_trans->num_writers, 1); cur_trans->num_joined = 0; cur_trans->transid = root->fs_info->generation; init_waitqueue_head(&cur_trans->writer_wait); init_waitqueue_head(&cur_trans->commit_wait); cur_trans->in_commit = 0; cur_trans->blocked = 0; - cur_trans->use_count = 1; + atomic_set(&cur_trans->use_count, 1); cur_trans->commit_done = 0; cur_trans->start_time = get_seconds(); @@ -88,7 +86,7 @@ static noinline int join_transaction(struct btrfs_root *root) root->fs_info->running_transaction = cur_trans; spin_unlock(&root->fs_info->new_trans_lock); } else { - cur_trans->num_writers++; + atomic_inc(&cur_trans->num_writers); cur_trans->num_joined++; } @@ -145,7 +143,7 @@ static void wait_current_trans(struct btrfs_root *root) cur_trans = root->fs_info->running_transaction; if (cur_trans && cur_trans->blocked) { DEFINE_WAIT(wait); - cur_trans->use_count++; + atomic_inc(&cur_trans->use_count); while (1) { prepare_to_wait(&root->fs_info->transaction_wait, &wait, TASK_UNINTERRUPTIBLE); @@ -181,6 +179,7 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, { struct btrfs_trans_handle *h; struct btrfs_transaction *cur_trans; + int retries = 0; int ret; if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) @@ -204,7 +203,7 @@ again: } cur_trans = root->fs_info->running_transaction; - cur_trans->use_count++; + atomic_inc(&cur_trans->use_count); if (type != TRANS_JOIN_NOLOCK) mutex_unlock(&root->fs_info->trans_mutex); @@ -224,10 +223,18 @@ again: if (num_items > 0) { ret = btrfs_trans_reserve_metadata(h, root, num_items); - if (ret == -EAGAIN) { + if (ret == -EAGAIN && !retries) { + retries++; btrfs_commit_transaction(h, root); goto again; + } else if (ret == -EAGAIN) { + /* + * We have already retried and got EAGAIN, so really we + * don't have space, so set ret to -ENOSPC. + */ + ret = -ENOSPC; } + if (ret < 0) { btrfs_end_transaction(h, root); return ERR_PTR(ret); @@ -327,7 +334,7 @@ int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid) goto out_unlock; /* nothing committing|committed */ } - cur_trans->use_count++; + atomic_inc(&cur_trans->use_count); mutex_unlock(&root->fs_info->trans_mutex); wait_for_commit(root, cur_trans); @@ -457,18 +464,14 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, wake_up_process(info->transaction_kthread); } - if (lock) - mutex_lock(&info->trans_mutex); WARN_ON(cur_trans != info->running_transaction); - WARN_ON(cur_trans->num_writers < 1); - cur_trans->num_writers--; + WARN_ON(atomic_read(&cur_trans->num_writers) < 1); + atomic_dec(&cur_trans->num_writers); smp_mb(); if (waitqueue_active(&cur_trans->writer_wait)) wake_up(&cur_trans->writer_wait); put_transaction(cur_trans); - if (lock) - mutex_unlock(&info->trans_mutex); if (current->journal_info == trans) current->journal_info = NULL; @@ -1178,7 +1181,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans, /* take transaction reference */ mutex_lock(&root->fs_info->trans_mutex); cur_trans = trans->transaction; - cur_trans->use_count++; + atomic_inc(&cur_trans->use_count); mutex_unlock(&root->fs_info->trans_mutex); btrfs_end_transaction(trans, root); @@ -1237,7 +1240,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, mutex_lock(&root->fs_info->trans_mutex); if (cur_trans->in_commit) { - cur_trans->use_count++; + atomic_inc(&cur_trans->use_count); mutex_unlock(&root->fs_info->trans_mutex); btrfs_end_transaction(trans, root); @@ -1259,7 +1262,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, prev_trans = list_entry(cur_trans->list.prev, struct btrfs_transaction, list); if (!prev_trans->commit_done) { - prev_trans->use_count++; + atomic_inc(&prev_trans->use_count); mutex_unlock(&root->fs_info->trans_mutex); wait_for_commit(root, prev_trans); @@ -1300,14 +1303,14 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, TASK_UNINTERRUPTIBLE); smp_mb(); - if (cur_trans->num_writers > 1) + if (atomic_read(&cur_trans->num_writers) > 1) schedule_timeout(MAX_SCHEDULE_TIMEOUT); else if (should_grow) schedule_timeout(1); mutex_lock(&root->fs_info->trans_mutex); finish_wait(&cur_trans->writer_wait, &wait); - } while (cur_trans->num_writers > 1 || + } while (atomic_read(&cur_trans->num_writers) > 1 || (should_grow && cur_trans->num_joined != joined)); ret = create_pending_snapshots(trans, root->fs_info); @@ -1394,6 +1397,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, wake_up(&cur_trans->commit_wait); + list_del_init(&cur_trans->list); put_transaction(cur_trans); put_transaction(cur_trans); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 229a594cacd5..e441acc6c584 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -27,11 +27,11 @@ struct btrfs_transaction { * total writers in this transaction, it must be zero before the * transaction can end */ - unsigned long num_writers; + atomic_t num_writers; unsigned long num_joined; int in_commit; - int use_count; + atomic_t use_count; int commit_done; int blocked; struct list_head list; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index c50271ad3157..f997ec0c1ba4 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -2209,8 +2209,10 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, log = root->log_root; path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; + if (!path) { + err = -ENOMEM; + goto out_unlock; + } di = btrfs_lookup_dir_item(trans, log, path, dir->i_ino, name, name_len, -1); @@ -2271,6 +2273,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, } fail: btrfs_free_path(path); +out_unlock: mutex_unlock(&BTRFS_I(dir)->log_mutex); if (ret == -ENOSPC) { root->fs_info->last_trans_log_full_commit = trans->transid; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 309a57b9fc85..c7367ae5a3e6 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -155,6 +155,15 @@ static noinline int run_scheduled_bios(struct btrfs_device *device) unsigned long limit; unsigned long last_waited = 0; int force_reg = 0; + struct blk_plug plug; + + /* + * this function runs all the bios we've collected for + * a particular device. We don't want to wander off to + * another device without first sending all of these down. + * So, setup a plug here and finish it off before we return + */ + blk_start_plug(&plug); bdi = blk_get_backing_dev_info(device->bdev); fs_info = device->dev_root->fs_info; @@ -294,6 +303,7 @@ loop_lock: spin_unlock(&device->io_lock); done: + blk_finish_plug(&plug); return 0; } diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index a5303b871b13..cfd660550ded 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -180,11 +180,10 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_dir_item *di; - int ret = 0, slot, advance; + int ret = 0, slot; size_t total_size = 0, size_left = size; unsigned long name_ptr; size_t name_len; - u32 nritems; /* * ok we want all objects associated with this id. @@ -204,34 +203,24 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) goto err; - advance = 0; + while (1) { leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); slot = path->slots[0]; /* this is where we start walking through the path */ - if (advance || slot >= nritems) { + if (slot >= btrfs_header_nritems(leaf)) { /* * if we've reached the last slot in this leaf we need * to go to the next leaf and reset everything */ - if (slot >= nritems-1) { - ret = btrfs_next_leaf(root, path); - if (ret) - break; - leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); - slot = path->slots[0]; - } else { - /* - * just walking through the slots on this leaf - */ - slot++; - path->slots[0]++; - } + ret = btrfs_next_leaf(root, path); + if (ret < 0) + goto err; + else if (ret > 0) + break; + continue; } - advance = 1; btrfs_item_key_to_cpu(leaf, &found_key, slot); @@ -250,7 +239,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) /* we are just looking for how big our buffer needs to be */ if (!size) - continue; + goto next; if (!buffer || (name_len + 1) > size_left) { ret = -ERANGE; @@ -263,6 +252,8 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) size_left -= name_len + 1; buffer += name_len + 1; +next: + path->slots[0]++; } ret = total_size; diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index 37fe101a4e0d..1064805e653b 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -197,7 +197,7 @@ struct fscache_object *cachefiles_grab_object(struct fscache_object *_object) } /* - * update the auxilliary data for an object object on disk + * update the auxiliary data for an object object on disk */ static void cachefiles_update_object(struct fscache_object *_object) { diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 37368ba2e67c..38b8ab554924 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -24,7 +24,7 @@ * context needs to be associated with the osd write during writeback. * * Similarly, struct ceph_inode_info maintains a set of counters to - * count dirty pages on the inode. In the absense of snapshots, + * count dirty pages on the inode. In the absence of snapshots, * i_wrbuffer_ref == i_wrbuffer_ref_head == the dirty page count. * * When a snapshot is taken (that is, when the client receives @@ -775,6 +775,13 @@ get_more_pages: ci->i_truncate_seq, ci->i_truncate_size, &inode->i_mtime, true, 1, 0); + + if (!req) { + rc = -ENOMEM; + unlock_page(page); + break; + } + max_pages = req->r_num_pages; alloc_page_vec(fsc, req); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 6b61ded701e1..2a5404c1c42f 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -765,7 +765,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) if (touch) { struct rb_node *q; - /* touch this + preceeding caps */ + /* touch this + preceding caps */ __touch_cap(cap); for (q = rb_first(&ci->i_caps); q != p; q = rb_next(q)) { @@ -819,7 +819,7 @@ int __ceph_caps_used(struct ceph_inode_info *ci) used |= CEPH_CAP_FILE_CACHE; if (ci->i_wr_ref) used |= CEPH_CAP_FILE_WR; - if (ci->i_wrbuffer_ref) + if (ci->i_wb_ref || ci->i_wrbuffer_ref) used |= CEPH_CAP_FILE_BUFFER; return used; } @@ -1331,10 +1331,11 @@ static void ceph_flush_snaps(struct ceph_inode_info *ci) } /* - * Mark caps dirty. If inode is newly dirty, add to the global dirty - * list. + * Mark caps dirty. If inode is newly dirty, return the dirty flags. + * Caller is then responsible for calling __mark_inode_dirty with the + * returned flags value. */ -void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) +int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) { struct ceph_mds_client *mdsc = ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc; @@ -1357,7 +1358,7 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) list_add(&ci->i_dirty_item, &mdsc->cap_dirty); spin_unlock(&mdsc->cap_dirty_lock); if (ci->i_flushing_caps == 0) { - igrab(inode); + ihold(inode); dirty |= I_DIRTY_SYNC; } } @@ -1365,9 +1366,8 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) if (((was | ci->i_flushing_caps) & CEPH_CAP_FILE_BUFFER) && (mask & CEPH_CAP_FILE_BUFFER)) dirty |= I_DIRTY_DATASYNC; - if (dirty) - __mark_inode_dirty(inode, dirty); __cap_delay_requeue(mdsc, ci); + return dirty; } /* @@ -1990,11 +1990,11 @@ static void __take_cap_refs(struct ceph_inode_info *ci, int got) if (got & CEPH_CAP_FILE_WR) ci->i_wr_ref++; if (got & CEPH_CAP_FILE_BUFFER) { - if (ci->i_wrbuffer_ref == 0) - igrab(&ci->vfs_inode); - ci->i_wrbuffer_ref++; - dout("__take_cap_refs %p wrbuffer %d -> %d (?)\n", - &ci->vfs_inode, ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref); + if (ci->i_wb_ref == 0) + ihold(&ci->vfs_inode); + ci->i_wb_ref++; + dout("__take_cap_refs %p wb %d -> %d (?)\n", + &ci->vfs_inode, ci->i_wb_ref-1, ci->i_wb_ref); } } @@ -2169,12 +2169,12 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) if (--ci->i_rdcache_ref == 0) last++; if (had & CEPH_CAP_FILE_BUFFER) { - if (--ci->i_wrbuffer_ref == 0) { + if (--ci->i_wb_ref == 0) { last++; put++; } - dout("put_cap_refs %p wrbuffer %d -> %d (?)\n", - inode, ci->i_wrbuffer_ref+1, ci->i_wrbuffer_ref); + dout("put_cap_refs %p wb %d -> %d (?)\n", + inode, ci->i_wb_ref+1, ci->i_wb_ref); } if (had & CEPH_CAP_FILE_WR) if (--ci->i_wr_ref == 0) { diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 159b512d5a27..203252d88d9f 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -734,9 +734,12 @@ retry_snap: } } if (ret >= 0) { + int dirty; spin_lock(&inode->i_lock); - __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); spin_unlock(&inode->i_lock); + if (dirty) + __mark_inode_dirty(inode, dirty); } out: diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index b54c97da1c43..70b6a4839c38 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -355,6 +355,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb) ci->i_rd_ref = 0; ci->i_rdcache_ref = 0; ci->i_wr_ref = 0; + ci->i_wb_ref = 0; ci->i_wrbuffer_ref = 0; ci->i_wrbuffer_ref_head = 0; ci->i_shared_gen = 0; @@ -1567,6 +1568,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) int release = 0, dirtied = 0; int mask = 0; int err = 0; + int inode_dirty_flags = 0; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; @@ -1725,13 +1727,16 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) dout("setattr %p ATTR_FILE ... hrm!\n", inode); if (dirtied) { - __ceph_mark_dirty_caps(ci, dirtied); + inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied); inode->i_ctime = CURRENT_TIME; } release &= issued; spin_unlock(&inode->i_lock); + if (inode_dirty_flags) + __mark_inode_dirty(inode, inode_dirty_flags); + if (mask) { req->r_inode = igrab(inode); req->r_inode_drop = release; diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index f60b07b0feb0..d0fae4ce9ba5 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -3304,8 +3304,8 @@ static void con_put(struct ceph_connection *con) { struct ceph_mds_session *s = con->private; + dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref) - 1); ceph_put_mds_session(s); - dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref)); } /* diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index 0aee66b92af3..24067d68a554 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -206,7 +206,7 @@ void ceph_put_snap_realm(struct ceph_mds_client *mdsc, up_write(&mdsc->snap_rwsem); } else { spin_lock(&mdsc->snap_empty_lock); - list_add(&mdsc->snap_empty, &realm->empty_item); + list_add(&realm->empty_item, &mdsc->snap_empty); spin_unlock(&mdsc->snap_empty_lock); } } @@ -342,7 +342,7 @@ static int build_snap_context(struct ceph_snap_realm *realm) num = 0; snapc->seq = realm->seq; if (parent) { - /* include any of parent's snaps occuring _after_ my + /* include any of parent's snaps occurring _after_ my parent became my parent */ for (i = 0; i < parent->cached_context->num_snaps; i++) if (parent->cached_context->snaps[i] >= diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 619fe719968f..f5cabefa98dc 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -293,7 +293,7 @@ struct ceph_inode_info { /* held references to caps */ int i_pin_ref; - int i_rd_ref, i_rdcache_ref, i_wr_ref; + int i_rd_ref, i_rdcache_ref, i_wr_ref, i_wb_ref; int i_wrbuffer_ref, i_wrbuffer_ref_head; u32 i_shared_gen; /* increment each time we get FILE_SHARED */ u32 i_rdcache_gen; /* incremented each time we get FILE_CACHE. */ @@ -506,7 +506,7 @@ static inline int __ceph_caps_dirty(struct ceph_inode_info *ci) { return ci->i_dirty_caps | ci->i_flushing_caps; } -extern void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask); +extern int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask); extern int ceph_caps_revoking(struct ceph_inode_info *ci, int mask); extern int __ceph_caps_used(struct ceph_inode_info *ci); diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 8c9eba6ef9df..f2b628696180 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -703,6 +703,7 @@ int ceph_setxattr(struct dentry *dentry, const char *name, struct ceph_inode_xattr *xattr = NULL; int issued; int required_blob_size; + int dirty; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; @@ -763,11 +764,12 @@ retry: dout("setxattr %p issued %s\n", inode, ceph_cap_string(issued)); err = __set_xattr(ci, newname, name_len, newval, val_len, 1, 1, 1, &xattr); - __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; spin_unlock(&inode->i_lock); - + if (dirty) + __mark_inode_dirty(inode, dirty); return err; do_sync: @@ -810,6 +812,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name) struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode); int issued; int err; + int dirty; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; @@ -833,12 +836,13 @@ int ceph_removexattr(struct dentry *dentry, const char *name) goto do_sync; err = __remove_xattr_by_name(ceph_inode(inode), name); - __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); + dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; spin_unlock(&inode->i_lock); - + if (dirty) + __mark_inode_dirty(inode, dirty); return err; do_sync: spin_unlock(&inode->i_lock); diff --git a/fs/cifs/AUTHORS b/fs/cifs/AUTHORS index 7f7fa3c302af..ea940b1db77b 100644 --- a/fs/cifs/AUTHORS +++ b/fs/cifs/AUTHORS @@ -35,7 +35,7 @@ Adrian Bunk (kcalloc cleanups) Miklos Szeredi Kazeon team for various fixes especially for 2.4 version. Asser Ferno (Change Notify support) -Shaggy (Dave Kleikamp) for inumerable small fs suggestions and some good cleanup +Shaggy (Dave Kleikamp) for innumerable small fs suggestions and some good cleanup Gunter Kukkukk (testing and suggestions for support of old servers) Igor Mammedov (DFS support) Jeff Layton (many, many fixes, as well as great work on the cifs Kerberos code) diff --git a/fs/cifs/README b/fs/cifs/README index fe1683590828..74ab165fc646 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -685,22 +685,6 @@ LinuxExtensionsEnabled If set to one then the client will attempt to support and want to map the uid and gid fields to values supplied at mount (rather than the actual values, then set this to zero. (default 1) -Experimental When set to 1 used to enable certain experimental - features (currently enables multipage writes - when signing is enabled, the multipage write - performance enhancement was disabled when - signing turned on in case buffer was modified - just before it was sent, also this flag will - be used to use the new experimental directory change - notification code). When set to 2 enables - an additional experimental feature, "raw ntlmssp" - session establishment support (which allows - specifying "sec=ntlmssp" on mount). The Linux cifs - module will use ntlmv2 authentication encapsulated - in "raw ntlmssp" (not using SPNEGO) when - "sec=ntlmssp" is specified on mount. - This support also requires building cifs with - the CONFIG_CIFS_EXPERIMENTAL configuration flag. These experimental features and tracing can be enabled by changing flags in /proc/fs/cifs (after the cifs module has been installed or built into the diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c index e654dfd092c3..53d57a3fe427 100644 --- a/fs/cifs/cache.c +++ b/fs/cifs/cache.c @@ -50,7 +50,7 @@ void cifs_fscache_unregister(void) */ struct cifs_server_key { uint16_t family; /* address family */ - uint16_t port; /* IP port */ + __be16 port; /* IP port */ union { struct in_addr ipv4_addr; struct in6_addr ipv6_addr; diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 65829d32128c..30d01bc90855 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -423,7 +423,6 @@ static const struct file_operations cifs_lookup_cache_proc_fops; static const struct file_operations traceSMB_proc_fops; static const struct file_operations cifs_multiuser_mount_proc_fops; static const struct file_operations cifs_security_flags_proc_fops; -static const struct file_operations cifs_experimental_proc_fops; static const struct file_operations cifs_linux_ext_proc_fops; void @@ -441,8 +440,6 @@ cifs_proc_init(void) proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops); proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops); proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops); - proc_create("Experimental", 0, proc_fs_cifs, - &cifs_experimental_proc_fops); proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, &cifs_linux_ext_proc_fops); proc_create("MultiuserMount", 0, proc_fs_cifs, @@ -469,7 +466,6 @@ cifs_proc_clean(void) remove_proc_entry("OplockEnabled", proc_fs_cifs); remove_proc_entry("SecurityFlags", proc_fs_cifs); remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs); - remove_proc_entry("Experimental", proc_fs_cifs); remove_proc_entry("LookupCacheEnabled", proc_fs_cifs); remove_proc_entry("fs/cifs", NULL); } @@ -550,45 +546,6 @@ static const struct file_operations cifs_oplock_proc_fops = { .write = cifs_oplock_proc_write, }; -static int cifs_experimental_proc_show(struct seq_file *m, void *v) -{ - seq_printf(m, "%d\n", experimEnabled); - return 0; -} - -static int cifs_experimental_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, cifs_experimental_proc_show, NULL); -} - -static ssize_t cifs_experimental_proc_write(struct file *file, - const char __user *buffer, size_t count, loff_t *ppos) -{ - char c; - int rc; - - rc = get_user(c, buffer); - if (rc) - return rc; - if (c == '0' || c == 'n' || c == 'N') - experimEnabled = 0; - else if (c == '1' || c == 'y' || c == 'Y') - experimEnabled = 1; - else if (c == '2') - experimEnabled = 2; - - return count; -} - -static const struct file_operations cifs_experimental_proc_fops = { - .owner = THIS_MODULE, - .open = cifs_experimental_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, - .write = cifs_experimental_proc_write, -}; - static int cifs_linux_ext_proc_show(struct seq_file *m, void *v) { seq_printf(m, "%d\n", linuxExtEnabled); diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 0a265ad9e426..2b68ac57d97d 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -53,7 +53,7 @@ void cifs_dfs_release_automount_timer(void) * * Extracts sharename form full UNC. * i.e. strips from UNC trailing path that is not part of share - * name and fixup missing '\' in the begining of DFS node refferal + * name and fixup missing '\' in the beginning of DFS node refferal * if necessary. * Returns pointer to share name on success or ERR_PTR on error. * Caller is responsible for freeing returned string. diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 4dfba8283165..33d221394aca 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -113,7 +113,7 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) MAX_MECH_STR_LEN + UID_KEY_LEN + (sizeof(uid_t) * 2) + CREDUID_KEY_LEN + (sizeof(uid_t) * 2) + - USER_KEY_LEN + strlen(sesInfo->userName) + + USER_KEY_LEN + strlen(sesInfo->user_name) + PID_KEY_LEN + (sizeof(pid_t) * 2) + 1; spnego_key = ERR_PTR(-ENOMEM); @@ -153,7 +153,7 @@ cifs_get_spnego_key(struct cifsSesInfo *sesInfo) sprintf(dp, ";creduid=0x%x", sesInfo->cred_uid); dp = description + strlen(description); - sprintf(dp, ";user=%s", sesInfo->userName); + sprintf(dp, ";user=%s", sesInfo->user_name); dp = description + strlen(description); sprintf(dp, ";pid=0x%x", current->pid); diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index fc0fd4fde306..1b2e180b018d 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -90,7 +90,7 @@ cifs_mapchar(char *target, const __u16 src_char, const struct nls_table *cp, case UNI_COLON: *target = ':'; break; - case UNI_ASTERIK: + case UNI_ASTERISK: *target = '*'; break; case UNI_QUESTION: @@ -264,40 +264,41 @@ cifs_strndup_from_ucs(const char *src, const int maxlen, const bool is_unicode, * names are little endian 16 bit Unicode on the wire */ int -cifsConvertToUCS(__le16 *target, const char *source, int maxlen, +cifsConvertToUCS(__le16 *target, const char *source, int srclen, const struct nls_table *cp, int mapChars) { int i, j, charlen; - int len_remaining = maxlen; char src_char; - __u16 temp; + __le16 dst_char; + wchar_t tmp; if (!mapChars) return cifs_strtoUCS(target, source, PATH_MAX, cp); - for (i = 0, j = 0; i < maxlen; j++) { + for (i = 0, j = 0; i < srclen; j++) { src_char = source[i]; + charlen = 1; switch (src_char) { case 0: - put_unaligned_le16(0, &target[j]); + put_unaligned(0, &target[j]); goto ctoUCS_out; case ':': - temp = UNI_COLON; + dst_char = cpu_to_le16(UNI_COLON); break; case '*': - temp = UNI_ASTERIK; + dst_char = cpu_to_le16(UNI_ASTERISK); break; case '?': - temp = UNI_QUESTION; + dst_char = cpu_to_le16(UNI_QUESTION); break; case '<': - temp = UNI_LESSTHAN; + dst_char = cpu_to_le16(UNI_LESSTHAN); break; case '>': - temp = UNI_GRTRTHAN; + dst_char = cpu_to_le16(UNI_GRTRTHAN); break; case '|': - temp = UNI_PIPE; + dst_char = cpu_to_le16(UNI_PIPE); break; /* * FIXME: We can not handle remapping backslash (UNI_SLASH) @@ -305,28 +306,24 @@ cifsConvertToUCS(__le16 *target, const char *source, int maxlen, * as they use backslash as separator. */ default: - charlen = cp->char2uni(source+i, len_remaining, - &temp); + charlen = cp->char2uni(source + i, srclen - i, &tmp); + dst_char = cpu_to_le16(tmp); + /* * if no match, use question mark, which at least in * some cases serves as wild card */ if (charlen < 1) { - temp = 0x003f; + dst_char = cpu_to_le16(0x003f); charlen = 1; } - len_remaining -= charlen; - /* - * character may take more than one byte in the source - * string, but will take exactly two bytes in the - * target string - */ - i += charlen; - continue; } - put_unaligned_le16(temp, &target[j]); - i++; /* move to next char in source string */ - len_remaining--; + /* + * character may take more than one byte in the source string, + * but will take exactly two bytes in the target string + */ + i += charlen; + put_unaligned(dst_char, &target[j]); } ctoUCS_out: diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h index 7fe6b52df507..644dd882a560 100644 --- a/fs/cifs/cifs_unicode.h +++ b/fs/cifs/cifs_unicode.h @@ -44,7 +44,7 @@ * reserved symbols (along with \ and /), otherwise illegal to store * in filenames in NTFS */ -#define UNI_ASTERIK (__u16) ('*' + 0xF000) +#define UNI_ASTERISK (__u16) ('*' + 0xF000) #define UNI_QUESTION (__u16) ('?' + 0xF000) #define UNI_COLON (__u16) (':' + 0xF000) #define UNI_GRTRTHAN (__u16) ('>' + 0xF000) diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index a51585f9852b..d1a016be73ba 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -30,12 +30,13 @@ #include <linux/ctype.h> #include <linux/random.h> -/* Calculate and return the CIFS signature based on the mac key and SMB PDU */ -/* the 16 byte signature must be allocated by the caller */ -/* Note we only use the 1st eight bytes */ -/* Note that the smb header signature field on input contains the - sequence number before this function is called */ - +/* + * Calculate and return the CIFS signature based on the mac key and SMB PDU. + * The 16 byte signature must be allocated by the caller. Note we only use the + * 1st eight bytes and that the smb header signature field on input contains + * the sequence number before this function is called. Also, this function + * should be called with the server->srv_mutex held. + */ static int cifs_calculate_signature(const struct smb_hdr *cifs_pdu, struct TCP_Server_Info *server, char *signature) { @@ -209,8 +210,10 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu, cpu_to_le32(expected_sequence_number); cifs_pdu->Signature.Sequence.Reserved = 0; + mutex_lock(&server->srv_mutex); rc = cifs_calculate_signature(cifs_pdu, server, what_we_think_sig_should_be); + mutex_unlock(&server->srv_mutex); if (rc) return rc; @@ -469,15 +472,15 @@ static int calc_ntlmv2_hash(struct cifsSesInfo *ses, char *ntlmv2_hash, return rc; } - /* convert ses->userName to unicode and uppercase */ - len = strlen(ses->userName); + /* convert ses->user_name to unicode and uppercase */ + len = strlen(ses->user_name); user = kmalloc(2 + (len * 2), GFP_KERNEL); if (user == NULL) { cERROR(1, "calc_ntlmv2_hash: user mem alloc failure\n"); rc = -ENOMEM; goto calc_exit_2; } - len = cifs_strtoUCS((__le16 *)user, ses->userName, len, nls_cp); + len = cifs_strtoUCS((__le16 *)user, ses->user_name, len, nls_cp); UniStrupr(user); crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index f2970136d17d..5c412b33cd7c 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -53,7 +53,6 @@ int cifsFYI = 0; int cifsERROR = 1; int traceSMB = 0; unsigned int oplockEnabled = 1; -unsigned int experimEnabled = 0; unsigned int linuxExtEnabled = 1; unsigned int lookupCacheEnabled = 1; unsigned int multiuser_mount = 0; @@ -127,6 +126,7 @@ cifs_read_super(struct super_block *sb, void *data, kfree(cifs_sb); return rc; } + cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages; #ifdef CONFIG_CIFS_DFS_UPCALL /* copy mount params to sb for use in submounts */ @@ -409,8 +409,8 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) seq_printf(s, ",multiuser"); - else if (tcon->ses->userName) - seq_printf(s, ",username=%s", tcon->ses->userName); + else if (tcon->ses->user_name) + seq_printf(s, ",username=%s", tcon->ses->user_name); if (tcon->ses->domainName) seq_printf(s, ",domain=%s", tcon->ses->domainName); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 17afb0fbcaed..a5d1106fcbde 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -37,10 +37,9 @@ #define MAX_TREE_SIZE (2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1) #define MAX_SERVER_SIZE 15 -#define MAX_SHARE_SIZE 64 /* used to be 20, this should still be enough */ -#define MAX_USERNAME_SIZE 32 /* 32 is to allow for 15 char names + null - termination then *2 for unicode versions */ -#define MAX_PASSWORD_SIZE 512 /* max for windows seems to be 256 wide chars */ +#define MAX_SHARE_SIZE 80 +#define MAX_USERNAME_SIZE 256 /* reasonable maximum for current servers */ +#define MAX_PASSWORD_SIZE 512 /* max for windows seems to be 256 wide chars */ #define CIFS_MIN_RCV_POOL 4 @@ -92,7 +91,8 @@ enum statusEnum { CifsNew = 0, CifsGood, CifsExiting, - CifsNeedReconnect + CifsNeedReconnect, + CifsNeedNegotiate }; enum securityEnum { @@ -274,7 +274,7 @@ struct cifsSesInfo { int capabilities; char serverName[SERVER_NAME_LEN_WITH_NULL * 2]; /* BB make bigger for TCP names - will ipv6 and sctp addresses fit? */ - char userName[MAX_USERNAME_SIZE + 1]; + char *user_name; char *domainName; char *password; struct session_key auth_key; @@ -817,7 +817,6 @@ GLOBAL_EXTERN unsigned int multiuser_mount; /* if enabled allows new sessions have the uid/password or Kerberos credential or equivalent for current user */ GLOBAL_EXTERN unsigned int oplockEnabled; -GLOBAL_EXTERN unsigned int experimEnabled; GLOBAL_EXTERN unsigned int lookupCacheEnabled; GLOBAL_EXTERN unsigned int global_secflags; /* if on, session setup sent with more secure ntlmssp2 challenge/resp */ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 904aa47e3515..df959bae6728 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -142,9 +142,9 @@ cifs_reconnect_tcon(struct cifsTconInfo *tcon, int smb_command) */ while (server->tcpStatus == CifsNeedReconnect) { wait_event_interruptible_timeout(server->response_q, - (server->tcpStatus == CifsGood), 10 * HZ); + (server->tcpStatus != CifsNeedReconnect), 10 * HZ); - /* is TCP session is reestablished now ?*/ + /* are we still trying to reconnect? */ if (server->tcpStatus != CifsNeedReconnect) break; @@ -729,7 +729,7 @@ CIFSSMBEcho(struct TCP_Server_Info *server) return rc; /* set up echo request */ - smb->hdr.Tid = cpu_to_le16(0xffff); + smb->hdr.Tid = 0xffff; smb->hdr.WordCount = 1; put_unaligned_le16(1, &smb->EchoCount); put_bcc_le(1, &smb->hdr); @@ -1884,10 +1884,10 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon, __constant_cpu_to_le16(CIFS_WRLCK)) pLockData->fl_type = F_WRLCK; - pLockData->fl_start = parm_data->start; - pLockData->fl_end = parm_data->start + - parm_data->length - 1; - pLockData->fl_pid = parm_data->pid; + pLockData->fl_start = le64_to_cpu(parm_data->start); + pLockData->fl_end = pLockData->fl_start + + le64_to_cpu(parm_data->length) - 1; + pLockData->fl_pid = le32_to_cpu(parm_data->pid); } } @@ -5247,7 +5247,7 @@ cifs_fill_unix_set_info(FILE_UNIX_BASIC_INFO *data_offset, * Samba server ignores set of file size to zero due to bugs in some * older clients, but we should be precise - we use SetFileSize to * set file size and do not want to truncate file size to zero - * accidently as happened on one Samba server beta by putting + * accidentally as happened on one Samba server beta by putting * zero instead of -1 here */ data_offset->EndOfFile = cpu_to_le64(NO_CHANGE_64); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 8d6c17ab593d..277262a8e82f 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -199,8 +199,7 @@ cifs_reconnect(struct TCP_Server_Info *server) } spin_unlock(&GlobalMid_Lock); - while ((server->tcpStatus != CifsExiting) && - (server->tcpStatus != CifsGood)) { + while (server->tcpStatus == CifsNeedReconnect) { try_to_freeze(); /* we should try only the port we connected to before */ @@ -212,7 +211,7 @@ cifs_reconnect(struct TCP_Server_Info *server) atomic_inc(&tcpSesReconnectCount); spin_lock(&GlobalMid_Lock); if (server->tcpStatus != CifsExiting) - server->tcpStatus = CifsGood; + server->tcpStatus = CifsNeedNegotiate; spin_unlock(&GlobalMid_Lock); } } @@ -248,24 +247,24 @@ static int check2ndT2(struct smb_hdr *pSMB, unsigned int maxBufSize) total_data_size = get_unaligned_le16(&pSMBt->t2_rsp.TotalDataCount); data_in_this_rsp = get_unaligned_le16(&pSMBt->t2_rsp.DataCount); - remaining = total_data_size - data_in_this_rsp; - - if (remaining == 0) + if (total_data_size == data_in_this_rsp) return 0; - else if (remaining < 0) { + else if (total_data_size < data_in_this_rsp) { cFYI(1, "total data %d smaller than data in frame %d", total_data_size, data_in_this_rsp); return -EINVAL; - } else { - cFYI(1, "missing %d bytes from transact2, check next response", - remaining); - if (total_data_size > maxBufSize) { - cERROR(1, "TotalDataSize %d is over maximum buffer %d", - total_data_size, maxBufSize); - return -EINVAL; - } - return remaining; } + + remaining = total_data_size - data_in_this_rsp; + + cFYI(1, "missing %d bytes from transact2, check next response", + remaining); + if (total_data_size > maxBufSize) { + cERROR(1, "TotalDataSize %d is over maximum buffer %d", + total_data_size, maxBufSize); + return -EINVAL; + } + return remaining; } static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) @@ -275,7 +274,8 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) char *data_area_of_target; char *data_area_of_buf2; int remaining; - __u16 byte_count, total_data_size, total_in_buf, total_in_buf2; + unsigned int byte_count, total_in_buf; + __u16 total_data_size, total_in_buf2; total_data_size = get_unaligned_le16(&pSMBt->t2_rsp.TotalDataCount); @@ -288,7 +288,7 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) remaining = total_data_size - total_in_buf; if (remaining < 0) - return -EINVAL; + return -EPROTO; if (remaining == 0) /* nothing to do, ignore */ return 0; @@ -309,20 +309,29 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB) data_area_of_target += total_in_buf; /* copy second buffer into end of first buffer */ - memcpy(data_area_of_target, data_area_of_buf2, total_in_buf2); total_in_buf += total_in_buf2; + /* is the result too big for the field? */ + if (total_in_buf > USHRT_MAX) + return -EPROTO; put_unaligned_le16(total_in_buf, &pSMBt->t2_rsp.DataCount); + + /* fix up the BCC */ byte_count = get_bcc_le(pTargetSMB); byte_count += total_in_buf2; + /* is the result too big for the field? */ + if (byte_count > USHRT_MAX) + return -EPROTO; put_bcc_le(byte_count, pTargetSMB); byte_count = pTargetSMB->smb_buf_length; byte_count += total_in_buf2; - - /* BB also add check that we are not beyond maximum buffer size */ - + /* don't allow buffer to overflow */ + if (byte_count > CIFSMaxBufSize) + return -ENOBUFS; pTargetSMB->smb_buf_length = byte_count; + memcpy(data_area_of_target, data_area_of_buf2, total_in_buf2); + if (remaining == total_in_buf2) { cFYI(1, "found the last secondary response"); return 0; /* we are done */ @@ -421,7 +430,7 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) pdu_length = 4; /* enough to get RFC1001 header */ incomplete_rcv: - if (echo_retries > 0 && + if (echo_retries > 0 && server->tcpStatus == CifsGood && time_after(jiffies, server->lstrp + (echo_retries * SMB_ECHO_INTERVAL))) { cERROR(1, "Server %s has not responded in %d seconds. " @@ -608,59 +617,63 @@ incomplete_rcv: list_for_each_safe(tmp, tmp2, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - if ((mid_entry->mid == smb_buffer->Mid) && - (mid_entry->midState == MID_REQUEST_SUBMITTED) && - (mid_entry->command == smb_buffer->Command)) { - if (length == 0 && - check2ndT2(smb_buffer, server->maxBuf) > 0) { - /* We have a multipart transact2 resp */ - isMultiRsp = true; - if (mid_entry->resp_buf) { - /* merge response - fix up 1st*/ - if (coalesce_t2(smb_buffer, - mid_entry->resp_buf)) { - mid_entry->multiRsp = - true; - break; - } else { - /* all parts received */ - mid_entry->multiEnd = - true; - goto multi_t2_fnd; - } + if (mid_entry->mid != smb_buffer->Mid || + mid_entry->midState != MID_REQUEST_SUBMITTED || + mid_entry->command != smb_buffer->Command) { + mid_entry = NULL; + continue; + } + + if (length == 0 && + check2ndT2(smb_buffer, server->maxBuf) > 0) { + /* We have a multipart transact2 resp */ + isMultiRsp = true; + if (mid_entry->resp_buf) { + /* merge response - fix up 1st*/ + length = coalesce_t2(smb_buffer, + mid_entry->resp_buf); + if (length > 0) { + length = 0; + mid_entry->multiRsp = true; + break; } else { - if (!isLargeBuf) { - cERROR(1, "1st trans2 resp needs bigbuf"); - /* BB maybe we can fix this up, switch - to already allocated large buffer? */ - } else { - /* Have first buffer */ - mid_entry->resp_buf = - smb_buffer; - mid_entry->largeBuf = - true; - bigbuf = NULL; - } + /* all parts received or + * packet is malformed + */ + mid_entry->multiEnd = true; + goto multi_t2_fnd; + } + } else { + if (!isLargeBuf) { + /* + * FIXME: switch to already + * allocated largebuf? + */ + cERROR(1, "1st trans2 resp " + "needs bigbuf"); + } else { + /* Have first buffer */ + mid_entry->resp_buf = + smb_buffer; + mid_entry->largeBuf = true; + bigbuf = NULL; } - break; } - mid_entry->resp_buf = smb_buffer; - mid_entry->largeBuf = isLargeBuf; + break; + } + mid_entry->resp_buf = smb_buffer; + mid_entry->largeBuf = isLargeBuf; multi_t2_fnd: - if (length == 0) - mid_entry->midState = - MID_RESPONSE_RECEIVED; - else - mid_entry->midState = - MID_RESPONSE_MALFORMED; + if (length == 0) + mid_entry->midState = MID_RESPONSE_RECEIVED; + else + mid_entry->midState = MID_RESPONSE_MALFORMED; #ifdef CONFIG_CIFS_STATS2 - mid_entry->when_received = jiffies; + mid_entry->when_received = jiffies; #endif - list_del_init(&mid_entry->qhead); - mid_entry->callback(mid_entry); - break; - } - mid_entry = NULL; + list_del_init(&mid_entry->qhead); + mid_entry->callback(mid_entry); + break; } spin_unlock(&GlobalMid_Lock); @@ -808,8 +821,7 @@ static int cifs_parse_mount_options(char *options, const char *devname, struct smb_vol *vol) { - char *value; - char *data; + char *value, *data, *end; unsigned int temp_len, i, j; char separator[2]; short int override_uid = -1; @@ -852,6 +864,7 @@ cifs_parse_mount_options(char *options, const char *devname, if (!options) return 1; + end = options + strlen(options); if (strncmp(options, "sep=", 4) == 0) { if (options[4] != 0) { separator[0] = options[4]; @@ -881,7 +894,8 @@ cifs_parse_mount_options(char *options, const char *devname, /* null user, ie anonymous, authentication */ vol->nullauth = 1; } - if (strnlen(value, 200) < 200) { + if (strnlen(value, MAX_USERNAME_SIZE) < + MAX_USERNAME_SIZE) { vol->username = value; } else { printk(KERN_WARNING "CIFS: username too long\n"); @@ -916,6 +930,7 @@ cifs_parse_mount_options(char *options, const char *devname, the only illegal character in a password is null */ if ((value[temp_len] == 0) && + (value + temp_len < end) && (value[temp_len+1] == separator[0])) { /* reinsert comma */ value[temp_len] = separator[0]; @@ -1472,7 +1487,7 @@ srcip_matches(struct sockaddr *srcaddr, struct sockaddr *rhs) static bool match_port(struct TCP_Server_Info *server, struct sockaddr *addr) { - unsigned short int port, *sport; + __be16 port, *sport; switch (addr->sa_family) { case AF_INET: @@ -1572,7 +1587,7 @@ match_security(struct TCP_Server_Info *server, struct smb_vol *vol) return false; } - /* now check if signing mode is acceptible */ + /* now check if signing mode is acceptable */ if ((secFlags & CIFSSEC_MAY_SIGN) == 0 && (server->secMode & SECMODE_SIGN_REQUIRED)) return false; @@ -1765,6 +1780,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info) module_put(THIS_MODULE); goto out_err_crypto_release; } + tcp_ses->tcpStatus = CifsNeedNegotiate; /* thread spawned, put it on the list */ spin_lock(&cifs_tcp_ses_lock); @@ -1808,7 +1824,9 @@ cifs_find_smb_ses(struct TCP_Server_Info *server, struct smb_vol *vol) break; default: /* anything else takes username/password */ - if (strncmp(ses->userName, vol->username, + if (ses->user_name == NULL) + continue; + if (strncmp(ses->user_name, vol->username, MAX_USERNAME_SIZE)) continue; if (strlen(vol->username) != 0 && @@ -1851,6 +1869,8 @@ cifs_put_smb_ses(struct cifsSesInfo *ses) cifs_put_tcp_session(server); } +static bool warned_on_ntlm; /* globals init to false automatically */ + static struct cifsSesInfo * cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) { @@ -1906,9 +1926,11 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) else sprintf(ses->serverName, "%pI4", &addr->sin_addr); - if (volume_info->username) - strncpy(ses->userName, volume_info->username, - MAX_USERNAME_SIZE); + if (volume_info->username) { + ses->user_name = kstrdup(volume_info->username, GFP_KERNEL); + if (!ses->user_name) + goto get_ses_fail; + } /* volume_info->password freed at unmount */ if (volume_info->password) { @@ -1923,6 +1945,15 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) } ses->cred_uid = volume_info->cred_uid; ses->linux_uid = volume_info->linux_uid; + + /* ntlmv2 is much stronger than ntlm security, and has been broadly + supported for many years, time to update default security mechanism */ + if ((volume_info->secFlg == 0) && warned_on_ntlm == false) { + warned_on_ntlm = true; + cERROR(1, "default security mechanism requested. The default " + "security mechanism will be upgraded from ntlm to " + "ntlmv2 in kernel release 2.6.41"); + } ses->overrideSecFlg = volume_info->secFlg; mutex_lock(&ses->session_mutex); @@ -2276,7 +2307,7 @@ static int generic_ip_connect(struct TCP_Server_Info *server) { int rc = 0; - unsigned short int sport; + __be16 sport; int slen, sfamily; struct socket *socket = server->ssocket; struct sockaddr *saddr; @@ -2361,7 +2392,7 @@ generic_ip_connect(struct TCP_Server_Info *server) static int ip_connect(struct TCP_Server_Info *server) { - unsigned short int *sport; + __be16 *sport; struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)&server->dstaddr; struct sockaddr_in *addr = (struct sockaddr_in *)&server->dstaddr; @@ -2642,6 +2673,11 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon, 0 /* not legacy */, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + + if (rc == -EOPNOTSUPP || rc == -EINVAL) + rc = SMBQueryInformation(xid, tcon, full_path, pfile_info, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); kfree(pfile_info); return rc; } @@ -2826,7 +2862,7 @@ try_mount_again: remote_path_check: /* check if a whole path (including prepath) is not remote */ - if (!rc && cifs_sb->prepathlen && tcon) { + if (!rc && tcon) { /* build_path_to_root works only when we have a valid tcon */ full_path = cifs_build_path_to_root(cifs_sb, tcon); if (full_path == NULL) { @@ -2933,7 +2969,7 @@ mount_fail_check: if (mount_data != mount_data_global) kfree(mount_data); /* If find_unc succeeded then rc == 0 so we can not end */ - /* up accidently freeing someone elses tcon struct */ + /* up accidentally freeing someone elses tcon struct */ if (tcon) cifs_put_tcon(tcon); else if (pSesInfo) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index dd5f22918c33..9ea65cf36714 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -189,7 +189,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, inode->i_sb, mode, oflags, &oplock, &fileHandle, xid); /* EIO could indicate that (posix open) operation is not supported, despite what server claimed in capability - negotation. EREMOTE indicates DFS junction, which is not + negotiation. EREMOTE indicates DFS junction, which is not handled in posix open */ if (rc == 0) { diff --git a/fs/cifs/file.c b/fs/cifs/file.c index c27d236738fc..faf59529e847 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -575,8 +575,10 @@ reopen_error_exit: int cifs_close(struct inode *inode, struct file *file) { - cifsFileInfo_put(file->private_data); - file->private_data = NULL; + if (file->private_data != NULL) { + cifsFileInfo_put(file->private_data); + file->private_data = NULL; + } /* return code from the ->release op is always ignored */ return 0; @@ -970,6 +972,9 @@ static ssize_t cifs_write(struct cifsFileInfo *open_file, total_written += bytes_written) { rc = -EAGAIN; while (rc == -EAGAIN) { + struct kvec iov[2]; + unsigned int len; + if (open_file->invalidHandle) { /* we could deadlock if we called filemap_fdatawait from here so tell @@ -979,31 +984,14 @@ static ssize_t cifs_write(struct cifsFileInfo *open_file, if (rc != 0) break; } - if (experimEnabled || (pTcon->ses->server && - ((pTcon->ses->server->secMode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) - == 0))) { - struct kvec iov[2]; - unsigned int len; - - len = min((size_t)cifs_sb->wsize, - write_size - total_written); - /* iov[0] is reserved for smb header */ - iov[1].iov_base = (char *)write_data + - total_written; - iov[1].iov_len = len; - rc = CIFSSMBWrite2(xid, pTcon, - open_file->netfid, len, - *poffset, &bytes_written, - iov, 1, 0); - } else - rc = CIFSSMBWrite(xid, pTcon, - open_file->netfid, - min_t(const int, cifs_sb->wsize, - write_size - total_written), - *poffset, &bytes_written, - write_data + total_written, - NULL, 0); + + len = min((size_t)cifs_sb->wsize, + write_size - total_written); + /* iov[0] is reserved for smb header */ + iov[1].iov_base = (char *)write_data + total_written; + iov[1].iov_len = len; + rc = CIFSSMBWrite2(xid, pTcon, open_file->netfid, len, + *poffset, &bytes_written, iov, 1, 0); } if (rc || (bytes_written == 0)) { if (total_written) @@ -1240,12 +1228,6 @@ static int cifs_writepages(struct address_space *mapping, } tcon = tlink_tcon(open_file->tlink); - if (!experimEnabled && tcon->ses->server->secMode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { - cifsFileInfo_put(open_file); - kfree(iov); - return generic_writepages(mapping, wbc); - } cifsFileInfo_put(open_file); xid = GetXid(); @@ -1980,6 +1962,24 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, return total_read; } +/* + * If the page is mmap'ed into a process' page tables, then we need to make + * sure that it doesn't change while being written back. + */ +static int +cifs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct page *page = vmf->page; + + lock_page(page); + return VM_FAULT_LOCKED; +} + +static struct vm_operations_struct cifs_file_vm_ops = { + .fault = filemap_fault, + .page_mkwrite = cifs_page_mkwrite, +}; + int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma) { int rc, xid; @@ -1991,6 +1991,8 @@ int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma) cifs_invalidate_mapping(inode); rc = generic_file_mmap(file, vma); + if (rc == 0) + vma->vm_ops = &cifs_file_vm_ops; FreeXid(xid); return rc; } @@ -2007,6 +2009,8 @@ int cifs_file_mmap(struct file *file, struct vm_area_struct *vma) return rc; } rc = generic_file_mmap(file, vma); + if (rc == 0) + vma->vm_ops = &cifs_file_vm_ops; FreeXid(xid); return rc; } diff --git a/fs/cifs/link.c b/fs/cifs/link.c index e8804d373404..ce417a9764a3 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -239,7 +239,7 @@ CIFSQueryMFSymLink(const int xid, struct cifsTconInfo *tcon, if (rc != 0) return rc; - if (file_info.EndOfFile != CIFS_MF_SYMLINK_FILE_SIZE) { + if (file_info.EndOfFile != cpu_to_le64(CIFS_MF_SYMLINK_FILE_SIZE)) { CIFSSMBClose(xid, tcon, netfid); /* it's not a symlink */ return -EINVAL; @@ -316,7 +316,7 @@ CIFSCheckMFSymlink(struct cifs_fattr *fattr, if (rc != 0) goto out; - if (file_info.EndOfFile != CIFS_MF_SYMLINK_FILE_SIZE) { + if (file_info.EndOfFile != cpu_to_le64(CIFS_MF_SYMLINK_FILE_SIZE)) { CIFSSMBClose(xid, pTcon, netfid); /* it's not a symlink */ goto out; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 2a930a752a78..0c684ae4c071 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -100,6 +100,7 @@ sesInfoFree(struct cifsSesInfo *buf_to_free) memset(buf_to_free->password, 0, strlen(buf_to_free->password)); kfree(buf_to_free->password); } + kfree(buf_to_free->user_name); kfree(buf_to_free->domainName); kfree(buf_to_free); } @@ -520,7 +521,7 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) (struct smb_com_transaction_change_notify_rsp *)buf; struct file_notify_information *pnotify; __u32 data_offset = 0; - if (pSMBr->ByteCount > sizeof(struct file_notify_information)) { + if (get_bcc_le(buf) > sizeof(struct file_notify_information)) { data_offset = le32_to_cpu(pSMBr->DataOffset); pnotify = (struct file_notify_information *) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 16765703131b..645114ad0a10 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -219,12 +219,12 @@ static void unicode_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, bcc_ptr++; } */ /* copy user */ - if (ses->userName == NULL) { + if (ses->user_name == NULL) { /* null user mount */ *bcc_ptr = 0; *(bcc_ptr+1) = 0; } else { - bytes_ret = cifs_strtoUCS((__le16 *) bcc_ptr, ses->userName, + bytes_ret = cifs_strtoUCS((__le16 *) bcc_ptr, ses->user_name, MAX_USERNAME_SIZE, nls_cp); } bcc_ptr += 2 * bytes_ret; @@ -244,12 +244,11 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, /* copy user */ /* BB what about null user mounts - check that we do this BB */ /* copy user */ - if (ses->userName == NULL) { - /* BB what about null user mounts - check that we do this BB */ - } else { - strncpy(bcc_ptr, ses->userName, MAX_USERNAME_SIZE); - } - bcc_ptr += strnlen(ses->userName, MAX_USERNAME_SIZE); + if (ses->user_name != NULL) + strncpy(bcc_ptr, ses->user_name, MAX_USERNAME_SIZE); + /* else null user mount */ + + bcc_ptr += strnlen(ses->user_name, MAX_USERNAME_SIZE); *bcc_ptr = 0; bcc_ptr++; /* account for null termination */ @@ -277,7 +276,7 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses, } static void -decode_unicode_ssetup(char **pbcc_area, __u16 bleft, struct cifsSesInfo *ses, +decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifsSesInfo *ses, const struct nls_table *nls_cp) { int len; @@ -285,19 +284,6 @@ decode_unicode_ssetup(char **pbcc_area, __u16 bleft, struct cifsSesInfo *ses, cFYI(1, "bleft %d", bleft); - /* - * Windows servers do not always double null terminate their final - * Unicode string. Check to see if there are an uneven number of bytes - * left. If so, then add an extra NULL pad byte to the end of the - * response. - * - * See section 2.7.2 in "Implementing CIFS" for details - */ - if (bleft % 2) { - data[bleft] = 0; - ++bleft; - } - kfree(ses->serverOS); ses->serverOS = cifs_strndup_from_ucs(data, bleft, true, nls_cp); cFYI(1, "serverOS=%s", ses->serverOS); @@ -405,8 +391,8 @@ static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, /* BB spec says that if AvId field of MsvAvTimestamp is populated then we must set the MIC field of the AUTHENTICATE_MESSAGE */ ses->ntlmssp->server_flags = le32_to_cpu(pblob->NegotiateFlags); - tioffset = cpu_to_le16(pblob->TargetInfoArray.BufferOffset); - tilen = cpu_to_le16(pblob->TargetInfoArray.Length); + tioffset = le32_to_cpu(pblob->TargetInfoArray.BufferOffset); + tilen = le16_to_cpu(pblob->TargetInfoArray.Length); if (tilen) { ses->auth_key.response = kmalloc(tilen, GFP_KERNEL); if (!ses->auth_key.response) { @@ -523,14 +509,14 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, tmp += len; } - if (ses->userName == NULL) { + if (ses->user_name == NULL) { sec_blob->UserName.BufferOffset = cpu_to_le32(tmp - pbuffer); sec_blob->UserName.Length = 0; sec_blob->UserName.MaximumLength = 0; tmp += 2; } else { int len; - len = cifs_strtoUCS((__le16 *)tmp, ses->userName, + len = cifs_strtoUCS((__le16 *)tmp, ses->user_name, MAX_USERNAME_SIZE, nls_cp); len *= 2; /* unicode is 2 bytes each */ sec_blob->UserName.BufferOffset = cpu_to_le32(tmp - pbuffer); @@ -930,7 +916,9 @@ ssetup_ntlmssp_authenticate: } /* BB check if Unicode and decode strings */ - if (smb_buf->Flags2 & SMBFLG2_UNICODE) { + if (bytes_remaining == 0) { + /* no string area to decode, do nothing */ + } else if (smb_buf->Flags2 & SMBFLG2_UNICODE) { /* unicode string area must be word-aligned */ if (((unsigned long) bcc_ptr - (unsigned long) smb_buf) % 2) { ++bcc_ptr; diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 90ff3cb10de3..9a37a9b6de3a 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -53,11 +53,14 @@ DEFINE_SPINLOCK(configfs_dirent_lock); static void configfs_d_iput(struct dentry * dentry, struct inode * inode) { - struct configfs_dirent * sd = dentry->d_fsdata; + struct configfs_dirent *sd = dentry->d_fsdata; if (sd) { BUG_ON(sd->s_dentry != dentry); + /* Coordinate with configfs_readdir */ + spin_lock(&configfs_dirent_lock); sd->s_dentry = NULL; + spin_unlock(&configfs_dirent_lock); configfs_put(sd); } iput(inode); @@ -689,7 +692,8 @@ static int create_default_group(struct config_group *parent_group, sd = child->d_fsdata; sd->s_type |= CONFIGFS_USET_DEFAULT; } else { - d_delete(child); + BUG_ON(child->d_inode); + d_drop(child); dput(child); } } @@ -990,7 +994,7 @@ static int configfs_dump(struct configfs_dirent *sd, int level) * This describes these functions and their helpers. * * Allow another kernel system to depend on a config_item. If this - * happens, the item cannot go away until the dependant can live without + * happens, the item cannot go away until the dependent can live without * it. The idea is to give client modules as simple an interface as * possible. When a system asks them to depend on an item, they just * call configfs_depend_item(). If the item is live and the client @@ -1545,7 +1549,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir struct configfs_dirent * parent_sd = dentry->d_fsdata; struct configfs_dirent *cursor = filp->private_data; struct list_head *p, *q = &cursor->s_sibling; - ino_t ino; + ino_t ino = 0; int i = filp->f_pos; switch (i) { @@ -1573,6 +1577,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir struct configfs_dirent *next; const char * name; int len; + struct inode *inode = NULL; next = list_entry(p, struct configfs_dirent, s_sibling); @@ -1581,9 +1586,28 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir name = configfs_get_name(next); len = strlen(name); - if (next->s_dentry) - ino = next->s_dentry->d_inode->i_ino; - else + + /* + * We'll have a dentry and an inode for + * PINNED items and for open attribute + * files. We lock here to prevent a race + * with configfs_d_iput() clearing + * s_dentry before calling iput(). + * + * Why do we go to the trouble? If + * someone has an attribute file open, + * the inode number should match until + * they close it. Beyond that, we don't + * care. + */ + spin_lock(&configfs_dirent_lock); + dentry = next->s_dentry; + if (dentry) + inode = dentry->d_inode; + if (inode) + ino = inode->i_ino; + spin_unlock(&configfs_dirent_lock); + if (!inode) ino = iunique(configfs_sb, 2); if (filldir(dirent, name, len, filp->f_pos, ino, @@ -1683,7 +1707,8 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) err = configfs_attach_group(sd->s_element, &group->cg_item, dentry); if (err) { - d_delete(dentry); + BUG_ON(dentry->d_inode); + d_drop(dentry); dput(dentry); } else { spin_lock(&configfs_dirent_lock); diff --git a/fs/dcache.c b/fs/dcache.c index ad25c4cec7d5..18b2a1f10ed8 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -35,6 +35,7 @@ #include <linux/hardirq.h> #include <linux/bit_spinlock.h> #include <linux/rculist_bl.h> +#include <linux/prefetch.h> #include "internal.h" /* @@ -99,12 +100,9 @@ static struct kmem_cache *dentry_cache __read_mostly; static unsigned int d_hash_mask __read_mostly; static unsigned int d_hash_shift __read_mostly; -struct dcache_hash_bucket { - struct hlist_bl_head head; -}; -static struct dcache_hash_bucket *dentry_hashtable __read_mostly; +static struct hlist_bl_head *dentry_hashtable __read_mostly; -static inline struct dcache_hash_bucket *d_hash(struct dentry *parent, +static inline struct hlist_bl_head *d_hash(struct dentry *parent, unsigned long hash) { hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES; @@ -112,16 +110,6 @@ static inline struct dcache_hash_bucket *d_hash(struct dentry *parent, return dentry_hashtable + (hash & D_HASHMASK); } -static inline void spin_lock_bucket(struct dcache_hash_bucket *b) -{ - bit_spin_lock(0, (unsigned long *)&b->head.first); -} - -static inline void spin_unlock_bucket(struct dcache_hash_bucket *b) -{ - __bit_spin_unlock(0, (unsigned long *)&b->head.first); -} - /* Statistics gathering. */ struct dentry_stat_t dentry_stat = { .age_limit = 45, @@ -167,8 +155,8 @@ static void d_free(struct dentry *dentry) if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); - /* if dentry was never inserted into hash, immediate free is OK */ - if (hlist_bl_unhashed(&dentry->d_hash)) + /* if dentry was never visible to RCU, immediate free is OK */ + if (!(dentry->d_flags & DCACHE_RCUACCESS)) __d_free(&dentry->d_u.d_rcu); else call_rcu(&dentry->d_u.d_rcu, __d_free); @@ -330,28 +318,19 @@ static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent) */ void __d_drop(struct dentry *dentry) { - if (!(dentry->d_flags & DCACHE_UNHASHED)) { - if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) { - bit_spin_lock(0, - (unsigned long *)&dentry->d_sb->s_anon.first); - dentry->d_flags |= DCACHE_UNHASHED; - hlist_bl_del_init(&dentry->d_hash); - __bit_spin_unlock(0, - (unsigned long *)&dentry->d_sb->s_anon.first); - } else { - struct dcache_hash_bucket *b; + if (!d_unhashed(dentry)) { + struct hlist_bl_head *b; + if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) + b = &dentry->d_sb->s_anon; + else b = d_hash(dentry->d_parent, dentry->d_name.hash); - spin_lock_bucket(b); - /* - * We may not actually need to put DCACHE_UNHASHED - * manipulations under the hash lock, but follow - * the principle of least surprise. - */ - dentry->d_flags |= DCACHE_UNHASHED; - hlist_bl_del_rcu(&dentry->d_hash); - spin_unlock_bucket(b); - dentry_rcuwalk_barrier(dentry); - } + + hlist_bl_lock(b); + __hlist_bl_del(&dentry->d_hash); + dentry->d_hash.pprev = NULL; + hlist_bl_unlock(b); + + dentry_rcuwalk_barrier(dentry); } } EXPORT_SYMBOL(__d_drop); @@ -1304,7 +1283,7 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) dname[name->len] = 0; dentry->d_count = 1; - dentry->d_flags = DCACHE_UNHASHED; + dentry->d_flags = 0; spin_lock_init(&dentry->d_lock); seqcount_init(&dentry->d_seq); dentry->d_inode = NULL; @@ -1606,10 +1585,9 @@ struct dentry *d_obtain_alias(struct inode *inode) tmp->d_inode = inode; tmp->d_flags |= DCACHE_DISCONNECTED; list_add(&tmp->d_alias, &inode->i_dentry); - bit_spin_lock(0, (unsigned long *)&tmp->d_sb->s_anon.first); - tmp->d_flags &= ~DCACHE_UNHASHED; + hlist_bl_lock(&tmp->d_sb->s_anon); hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon); - __bit_spin_unlock(0, (unsigned long *)&tmp->d_sb->s_anon.first); + hlist_bl_unlock(&tmp->d_sb->s_anon); spin_unlock(&tmp->d_lock); spin_unlock(&inode->i_lock); security_d_instantiate(tmp, inode); @@ -1789,7 +1767,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, unsigned int len = name->len; unsigned int hash = name->hash; const unsigned char *str = name->name; - struct dcache_hash_bucket *b = d_hash(parent, hash); + struct hlist_bl_head *b = d_hash(parent, hash); struct hlist_bl_node *node; struct dentry *dentry; @@ -1813,7 +1791,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, * * See Documentation/filesystems/path-lookup.txt for more details. */ - hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) { + hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { struct inode *i; const char *tname; int tlen; @@ -1908,7 +1886,7 @@ struct dentry *__d_lookup(struct dentry *parent, struct qstr *name) unsigned int len = name->len; unsigned int hash = name->hash; const unsigned char *str = name->name; - struct dcache_hash_bucket *b = d_hash(parent, hash); + struct hlist_bl_head *b = d_hash(parent, hash); struct hlist_bl_node *node; struct dentry *found = NULL; struct dentry *dentry; @@ -1935,7 +1913,7 @@ struct dentry *__d_lookup(struct dentry *parent, struct qstr *name) */ rcu_read_lock(); - hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) { + hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) { const char *tname; int tlen; @@ -2086,13 +2064,13 @@ again: } EXPORT_SYMBOL(d_delete); -static void __d_rehash(struct dentry * entry, struct dcache_hash_bucket *b) +static void __d_rehash(struct dentry * entry, struct hlist_bl_head *b) { BUG_ON(!d_unhashed(entry)); - spin_lock_bucket(b); - entry->d_flags &= ~DCACHE_UNHASHED; - hlist_bl_add_head_rcu(&entry->d_hash, &b->head); - spin_unlock_bucket(b); + hlist_bl_lock(b); + entry->d_flags |= DCACHE_RCUACCESS; + hlist_bl_add_head_rcu(&entry->d_hash, b); + hlist_bl_unlock(b); } static void _d_rehash(struct dentry * entry) @@ -2131,7 +2109,7 @@ EXPORT_SYMBOL(d_rehash); */ void dentry_update_name_case(struct dentry *dentry, struct qstr *name) { - BUG_ON(!mutex_is_locked(&dentry->d_inode->i_mutex)); + BUG_ON(!mutex_is_locked(&dentry->d_parent->d_inode->i_mutex)); BUG_ON(dentry->d_name.len != name->len); /* d_lookup gives this */ spin_lock(&dentry->d_lock); @@ -3025,7 +3003,7 @@ static void __init dcache_init_early(void) dentry_hashtable = alloc_large_system_hash("Dentry cache", - sizeof(struct dcache_hash_bucket), + sizeof(struct hlist_bl_head), dhash_entries, 13, HASH_EARLY, @@ -3034,7 +3012,7 @@ static void __init dcache_init_early(void) 0); for (loop = 0; loop < (1 << d_hash_shift); loop++) - INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head); + INIT_HLIST_BL_HEAD(dentry_hashtable + loop); } static void __init dcache_init(void) @@ -3057,7 +3035,7 @@ static void __init dcache_init(void) dentry_hashtable = alloc_large_system_hash("Dentry cache", - sizeof(struct dcache_hash_bucket), + sizeof(struct hlist_bl_head), dhash_entries, 13, 0, @@ -3066,7 +3044,7 @@ static void __init dcache_init(void) 0); for (loop = 0; loop < (1 << d_hash_shift); loop++) - INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head); + INIT_HLIST_BL_HEAD(dentry_hashtable + loop); } /* SLAB cache for __getname() consumers */ diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 89d394d8fe24..90f76575c056 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -428,26 +428,17 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { char buf[32]; - int buf_size; + size_t buf_size; + bool bv; u32 *val = file->private_data; buf_size = min(count, (sizeof(buf)-1)); if (copy_from_user(buf, user_buf, buf_size)) return -EFAULT; - switch (buf[0]) { - case 'y': - case 'Y': - case '1': - *val = 1; - break; - case 'n': - case 'N': - case '0': - *val = 0; - break; - } - + if (strtobool(buf, &bv) == 0) + *val = bv; + return count; } diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 04b8c449303f..56d6bfcc1e48 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -519,7 +519,7 @@ static void toss_rsb(struct kref *kref) } } -/* When all references to the rsb are gone it's transfered to +/* When all references to the rsb are gone it's transferred to the tossed list for later disposal. */ static void put_rsb(struct dlm_rsb *r) diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index bffa1e73b9a9..5e2c71f05e46 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -810,7 +810,7 @@ static int tcp_accept_from_sock(struct connection *con) /* * Add it to the active queue in case we got data - * beween processing the accept adding the socket + * between processing the accept adding the socket * to the read_sockets list */ if (!test_and_set_bit(CF_READ_PENDING, &addcon->flags)) diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index eda43f362616..14638235f7b2 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -304,7 +304,7 @@ static void set_master_lkbs(struct dlm_rsb *r) } /* - * Propogate the new master nodeid to locks + * Propagate the new master nodeid to locks * The NEW_MASTER flag tells dlm_recover_locks() which rsb's to consider. * The NEW_MASTER2 flag tells recover_lvb() and set_locks_purged() which * rsb's to consider. diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index d2a70a4561f9..b8d5c8091024 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1452,6 +1452,25 @@ static void set_default_header_data(struct ecryptfs_crypt_stat *crypt_stat) crypt_stat->metadata_size = ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE; } +void ecryptfs_i_size_init(const char *page_virt, struct inode *inode) +{ + struct ecryptfs_mount_crypt_stat *mount_crypt_stat; + struct ecryptfs_crypt_stat *crypt_stat; + u64 file_size; + + crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat; + mount_crypt_stat = + &ecryptfs_superblock_to_private(inode->i_sb)->mount_crypt_stat; + if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) { + file_size = i_size_read(ecryptfs_inode_to_lower(inode)); + if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) + file_size += crypt_stat->metadata_size; + } else + file_size = get_unaligned_be64(page_virt); + i_size_write(inode, (loff_t)file_size); + crypt_stat->flags |= ECRYPTFS_I_SIZE_INITIALIZED; +} + /** * ecryptfs_read_headers_virt * @page_virt: The virtual address into which to read the headers @@ -1482,6 +1501,8 @@ static int ecryptfs_read_headers_virt(char *page_virt, rc = -EINVAL; goto out; } + if (!(crypt_stat->flags & ECRYPTFS_I_SIZE_INITIALIZED)) + ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode); offset += MAGIC_ECRYPTFS_MARKER_SIZE_BYTES; rc = ecryptfs_process_flags(crypt_stat, (page_virt + offset), &bytes_read); diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index bd3cafd0949d..e70282775e2c 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -269,6 +269,7 @@ struct ecryptfs_crypt_stat { #define ECRYPTFS_ENCFN_USE_MOUNT_FNEK 0x00000800 #define ECRYPTFS_ENCFN_USE_FEK 0x00001000 #define ECRYPTFS_UNLINK_SIGS 0x00002000 +#define ECRYPTFS_I_SIZE_INITIALIZED 0x00004000 u32 flags; unsigned int file_version; size_t iv_bytes; @@ -295,6 +296,8 @@ struct ecryptfs_crypt_stat { struct ecryptfs_inode_info { struct inode vfs_inode; struct inode *wii_inode; + struct mutex lower_file_mutex; + atomic_t lower_file_count; struct file *lower_file; struct ecryptfs_crypt_stat crypt_stat; }; @@ -626,6 +629,7 @@ struct ecryptfs_open_req { int ecryptfs_interpose(struct dentry *hidden_dentry, struct dentry *this_dentry, struct super_block *sb, u32 flags); +void ecryptfs_i_size_init(const char *page_virt, struct inode *inode); int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, struct dentry *lower_dentry, struct inode *ecryptfs_dir_inode); @@ -757,7 +761,8 @@ int ecryptfs_privileged_open(struct file **lower_file, struct dentry *lower_dentry, struct vfsmount *lower_mnt, const struct cred *cred); -int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry); +int ecryptfs_get_lower_file(struct dentry *ecryptfs_dentry); +void ecryptfs_put_lower_file(struct inode *inode); int ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, size_t *packet_size, diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index cedc913d11ba..566e5472f78c 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -191,10 +191,10 @@ static int ecryptfs_open(struct inode *inode, struct file *file) | ECRYPTFS_ENCRYPTED); } mutex_unlock(&crypt_stat->cs_mutex); - rc = ecryptfs_init_persistent_file(ecryptfs_dentry); + rc = ecryptfs_get_lower_file(ecryptfs_dentry); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " - "the persistent file for the dentry with name " + "the lower file for the dentry with name " "[%s]; rc = [%d]\n", __func__, ecryptfs_dentry->d_name.name, rc); goto out_free; @@ -202,9 +202,9 @@ static int ecryptfs_open(struct inode *inode, struct file *file) if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_ACCMODE) == O_RDONLY && (file->f_flags & O_ACCMODE) != O_RDONLY) { rc = -EPERM; - printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs " + printk(KERN_WARNING "%s: Lower file is RO; eCryptfs " "file must hence be opened RO\n", __func__); - goto out_free; + goto out_put; } ecryptfs_set_file_lower( file, ecryptfs_inode_to_private(inode)->lower_file); @@ -232,10 +232,11 @@ static int ecryptfs_open(struct inode *inode, struct file *file) "Plaintext passthrough mode is not " "enabled; returning -EIO\n"); mutex_unlock(&crypt_stat->cs_mutex); - goto out_free; + goto out_put; } rc = 0; - crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); + crypt_stat->flags &= ~(ECRYPTFS_I_SIZE_INITIALIZED + | ECRYPTFS_ENCRYPTED); mutex_unlock(&crypt_stat->cs_mutex); goto out; } @@ -245,6 +246,8 @@ static int ecryptfs_open(struct inode *inode, struct file *file) "[0x%.16lx] size: [0x%.16llx]\n", inode, inode->i_ino, (unsigned long long)i_size_read(inode)); goto out; +out_put: + ecryptfs_put_lower_file(inode); out_free: kmem_cache_free(ecryptfs_file_info_cache, ecryptfs_file_to_private(file)); @@ -254,17 +257,13 @@ out: static int ecryptfs_flush(struct file *file, fl_owner_t td) { - int rc = 0; - struct file *lower_file = NULL; - - lower_file = ecryptfs_file_to_lower(file); - if (lower_file->f_op && lower_file->f_op->flush) - rc = lower_file->f_op->flush(lower_file, td); - return rc; + return file->f_mode & FMODE_WRITE + ? filemap_write_and_wait(file->f_mapping) : 0; } static int ecryptfs_release(struct inode *inode, struct file *file) { + ecryptfs_put_lower_file(inode); kmem_cache_free(ecryptfs_file_info_cache, ecryptfs_file_to_private(file)); return 0; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index f99051b7adab..4d4cc6a90cd5 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -168,19 +168,18 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) "context; rc = [%d]\n", rc); goto out; } - rc = ecryptfs_init_persistent_file(ecryptfs_dentry); + rc = ecryptfs_get_lower_file(ecryptfs_dentry); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " - "the persistent file for the dentry with name " + "the lower file for the dentry with name " "[%s]; rc = [%d]\n", __func__, ecryptfs_dentry->d_name.name, rc); goto out; } rc = ecryptfs_write_metadata(ecryptfs_dentry); - if (rc) { + if (rc) printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc); - goto out; - } + ecryptfs_put_lower_file(ecryptfs_dentry->d_inode); out: return rc; } @@ -226,11 +225,9 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, struct dentry *lower_dir_dentry; struct vfsmount *lower_mnt; struct inode *lower_inode; - struct ecryptfs_mount_crypt_stat *mount_crypt_stat; struct ecryptfs_crypt_stat *crypt_stat; char *page_virt = NULL; - u64 file_size; - int rc = 0; + int put_lower = 0, rc = 0; lower_dir_dentry = lower_dentry->d_parent; lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt( @@ -277,14 +274,15 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, rc = -ENOMEM; goto out; } - rc = ecryptfs_init_persistent_file(ecryptfs_dentry); + rc = ecryptfs_get_lower_file(ecryptfs_dentry); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " - "the persistent file for the dentry with name " + "the lower file for the dentry with name " "[%s]; rc = [%d]\n", __func__, ecryptfs_dentry->d_name.name, rc); goto out_free_kmem; } + put_lower = 1; crypt_stat = &ecryptfs_inode_to_private( ecryptfs_dentry->d_inode)->crypt_stat; /* TODO: lock for crypt_stat comparison */ @@ -302,18 +300,7 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, } crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR; } - mount_crypt_stat = &ecryptfs_superblock_to_private( - ecryptfs_dentry->d_sb)->mount_crypt_stat; - if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) { - if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) - file_size = (crypt_stat->metadata_size - + i_size_read(lower_dentry->d_inode)); - else - file_size = i_size_read(lower_dentry->d_inode); - } else { - file_size = get_unaligned_be64(page_virt); - } - i_size_write(ecryptfs_dentry->d_inode, (loff_t)file_size); + ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode); out_free_kmem: kmem_cache_free(ecryptfs_header_cache_2, page_virt); goto out; @@ -322,6 +309,8 @@ out_put: mntput(lower_mnt); d_drop(ecryptfs_dentry); out: + if (put_lower) + ecryptfs_put_lower_file(ecryptfs_dentry->d_inode); return rc; } @@ -538,8 +527,6 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) dget(lower_dentry); rc = vfs_rmdir(lower_dir_dentry->d_inode, lower_dentry); dput(lower_dentry); - if (!rc) - d_delete(lower_dentry); fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); dir->i_nlink = lower_dir_dentry->d_inode->i_nlink; unlock_dir(lower_dir_dentry); @@ -610,8 +597,8 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry, fsstack_copy_attr_all(old_dir, lower_old_dir_dentry->d_inode); out_lock: unlock_rename(lower_old_dir_dentry, lower_new_dir_dentry); - dput(lower_new_dentry->d_parent); - dput(lower_old_dentry->d_parent); + dput(lower_new_dir_dentry); + dput(lower_old_dir_dentry); dput(lower_new_dentry); dput(lower_old_dentry); return rc; @@ -759,8 +746,11 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia, if (unlikely((ia->ia_size == i_size))) { lower_ia->ia_valid &= ~ATTR_SIZE; - goto out; + return 0; } + rc = ecryptfs_get_lower_file(dentry); + if (rc) + return rc; crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; /* Switch on growing or shrinking file */ if (ia->ia_size > i_size) { @@ -838,6 +828,7 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia, lower_ia->ia_valid &= ~ATTR_SIZE; } out: + ecryptfs_put_lower_file(inode); return rc; } @@ -913,7 +904,13 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) mount_crypt_stat = &ecryptfs_superblock_to_private( dentry->d_sb)->mount_crypt_stat; + rc = ecryptfs_get_lower_file(dentry); + if (rc) { + mutex_unlock(&crypt_stat->cs_mutex); + goto out; + } rc = ecryptfs_read_metadata(dentry); + ecryptfs_put_lower_file(inode); if (rc) { if (!(mount_crypt_stat->flags & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) { @@ -927,10 +924,17 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) goto out; } rc = 0; - crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); + crypt_stat->flags &= ~(ECRYPTFS_I_SIZE_INITIALIZED + | ECRYPTFS_ENCRYPTED); } } mutex_unlock(&crypt_stat->cs_mutex); + if (S_ISREG(inode->i_mode)) { + rc = filemap_write_and_wait(inode->i_mapping); + if (rc) + goto out; + fsstack_copy_attr_all(inode, lower_inode); + } memcpy(&lower_ia, ia, sizeof(lower_ia)); if (ia->ia_valid & ATTR_FILE) lower_ia.ia_file = ecryptfs_file_to_lower(ia->ia_file); diff --git a/fs/ecryptfs/kthread.c b/fs/ecryptfs/kthread.c index 0851ab6980f5..69f994a7d524 100644 --- a/fs/ecryptfs/kthread.c +++ b/fs/ecryptfs/kthread.c @@ -44,7 +44,7 @@ static struct task_struct *ecryptfs_kthread; * @ignored: ignored * * The eCryptfs kernel thread that has the responsibility of getting - * the lower persistent file with RW permissions. + * the lower file with RW permissions. * * Returns zero on success; non-zero otherwise */ @@ -141,8 +141,8 @@ int ecryptfs_privileged_open(struct file **lower_file, int rc = 0; /* Corresponding dput() and mntput() are done when the - * persistent file is fput() when the eCryptfs inode is - * destroyed. */ + * lower file is fput() when all eCryptfs files for the inode are + * released. */ dget(lower_dentry); mntget(lower_mnt); flags |= IS_RDONLY(lower_dentry->d_inode) ? O_RDONLY : O_RDWR; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index c27c0ecf90bc..89b93389af8e 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -96,7 +96,7 @@ void __ecryptfs_printk(const char *fmt, ...) } /** - * ecryptfs_init_persistent_file + * ecryptfs_init_lower_file * @ecryptfs_dentry: Fully initialized eCryptfs dentry object, with * the lower dentry and the lower mount set * @@ -104,42 +104,70 @@ void __ecryptfs_printk(const char *fmt, ...) * inode. All I/O operations to the lower inode occur through that * file. When the first eCryptfs dentry that interposes with the first * lower dentry for that inode is created, this function creates the - * persistent file struct and associates it with the eCryptfs - * inode. When the eCryptfs inode is destroyed, the file is closed. + * lower file struct and associates it with the eCryptfs + * inode. When all eCryptfs files associated with the inode are released, the + * file is closed. * - * The persistent file will be opened with read/write permissions, if + * The lower file will be opened with read/write permissions, if * possible. Otherwise, it is opened read-only. * - * This function does nothing if a lower persistent file is already + * This function does nothing if a lower file is already * associated with the eCryptfs inode. * * Returns zero on success; non-zero otherwise */ -int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) +static int ecryptfs_init_lower_file(struct dentry *dentry, + struct file **lower_file) { const struct cred *cred = current_cred(); - struct ecryptfs_inode_info *inode_info = - ecryptfs_inode_to_private(ecryptfs_dentry->d_inode); - int rc = 0; + struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); + struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); + int rc; - if (!inode_info->lower_file) { - struct dentry *lower_dentry; - struct vfsmount *lower_mnt = - ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); + rc = ecryptfs_privileged_open(lower_file, lower_dentry, lower_mnt, + cred); + if (rc) { + printk(KERN_ERR "Error opening lower file " + "for lower_dentry [0x%p] and lower_mnt [0x%p]; " + "rc = [%d]\n", lower_dentry, lower_mnt, rc); + (*lower_file) = NULL; + } + return rc; +} - lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry); - rc = ecryptfs_privileged_open(&inode_info->lower_file, - lower_dentry, lower_mnt, cred); - if (rc) { - printk(KERN_ERR "Error opening lower persistent file " - "for lower_dentry [0x%p] and lower_mnt [0x%p]; " - "rc = [%d]\n", lower_dentry, lower_mnt, rc); - inode_info->lower_file = NULL; - } +int ecryptfs_get_lower_file(struct dentry *dentry) +{ + struct ecryptfs_inode_info *inode_info = + ecryptfs_inode_to_private(dentry->d_inode); + int count, rc = 0; + + mutex_lock(&inode_info->lower_file_mutex); + count = atomic_inc_return(&inode_info->lower_file_count); + if (WARN_ON_ONCE(count < 1)) + rc = -EINVAL; + else if (count == 1) { + rc = ecryptfs_init_lower_file(dentry, + &inode_info->lower_file); + if (rc) + atomic_set(&inode_info->lower_file_count, 0); } + mutex_unlock(&inode_info->lower_file_mutex); return rc; } +void ecryptfs_put_lower_file(struct inode *inode) +{ + struct ecryptfs_inode_info *inode_info; + + inode_info = ecryptfs_inode_to_private(inode); + if (atomic_dec_and_mutex_lock(&inode_info->lower_file_count, + &inode_info->lower_file_mutex)) { + fput(inode_info->lower_file); + inode_info->lower_file = NULL; + mutex_unlock(&inode_info->lower_file_mutex); + } +} + static struct inode *ecryptfs_get_inode(struct inode *lower_inode, struct super_block *sb) { @@ -276,7 +304,7 @@ static void ecryptfs_init_mount_crypt_stat( /** * ecryptfs_parse_options * @sb: The ecryptfs super block - * @options: The options pased to the kernel + * @options: The options passed to the kernel * * Parse mount options: * debug=N - ecryptfs_verbosity level for debug output @@ -840,7 +868,7 @@ static int __init ecryptfs_init(void) } rc = ecryptfs_init_messaging(); if (rc) { - printk(KERN_ERR "Failure occured while attempting to " + printk(KERN_ERR "Failure occurred while attempting to " "initialize the communications channel to " "ecryptfsd\n"); goto out_destroy_kthread; diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index bacc882e1ae4..245b517bf1b6 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -55,6 +55,8 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb) if (unlikely(!inode_info)) goto out; ecryptfs_init_crypt_stat(&inode_info->crypt_stat); + mutex_init(&inode_info->lower_file_mutex); + atomic_set(&inode_info->lower_file_count, 0); inode_info->lower_file = NULL; inode = &inode_info->vfs_inode; out: @@ -77,8 +79,7 @@ static void ecryptfs_i_callback(struct rcu_head *head) * * This is used during the final destruction of the inode. All * allocation of memory related to the inode, including allocated - * memory in the crypt_stat struct, will be released here. This - * function also fput()'s the persistent file for the lower inode. + * memory in the crypt_stat struct, will be released here. * There should be no chance that this deallocation will be missed. */ static void ecryptfs_destroy_inode(struct inode *inode) @@ -86,16 +87,7 @@ static void ecryptfs_destroy_inode(struct inode *inode) struct ecryptfs_inode_info *inode_info; inode_info = ecryptfs_inode_to_private(inode); - if (inode_info->lower_file) { - struct dentry *lower_dentry = - inode_info->lower_file->f_dentry; - - BUG_ON(!lower_dentry); - if (lower_dentry->d_inode) { - fput(inode_info->lower_file); - inode_info->lower_file = NULL; - } - } + BUG_ON(inode_info->lower_file); ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat); call_rcu(&inode->i_rcu, ecryptfs_i_callback); } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index ed38801b57a7..f9cfd168fbe2 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -181,7 +181,7 @@ struct eventpoll { /* * This is a single linked list that chains all the "struct epitem" that - * happened while transfering ready events to userspace w/out + * happened while transferring ready events to userspace w/out * holding ->lock. */ struct epitem *ovflist; @@ -606,7 +606,7 @@ static void ep_free(struct eventpoll *ep) * We do not need to hold "ep->mtx" here because the epoll file * is on the way to be removed and no one has references to it * anymore. The only hit might come from eventpoll_release_file() but - * holding "epmutex" is sufficent here. + * holding "epmutex" is sufficient here. */ mutex_lock(&epmutex); @@ -720,7 +720,7 @@ void eventpoll_release_file(struct file *file) /* * We don't want to get "file->f_lock" because it is not * necessary. It is not necessary because we're in the "struct file" - * cleanup path, and this means that noone is using this file anymore. + * cleanup path, and this means that no one is using this file anymore. * So, for example, epoll_ctl() cannot hit here since if we reach this * point, the file counter already went to zero and fget() would fail. * The only hit might come from ep_free() but by holding the mutex @@ -1112,7 +1112,7 @@ static int ep_send_events_proc(struct eventpoll *ep, struct list_head *head, * Trigger mode, we need to insert back inside * the ready list, so that the next call to * epoll_wait() will check again the events - * availability. At this point, noone can insert + * availability. At this point, no one can insert * into ep->rdllist besides us. The epoll_ctl() * callers are locked out by * ep_scan_ready_list() holding "mtx" and the diff --git a/fs/exofs/common.h b/fs/exofs/common.h index 5e74ad3d4009..3bbd46956d77 100644 --- a/fs/exofs/common.h +++ b/fs/exofs/common.h @@ -115,7 +115,7 @@ struct exofs_sb_stats { * Describes the raid used in the FS. It is part of the device table. * This here is taken from the pNFS-objects definition. In exofs we * use one raid policy through-out the filesystem. (NOTE: the funny - * alignment at begining. We take care of it at exofs_device_table. + * alignment at beginning. We take care of it at exofs_device_table. */ struct exofs_dt_data_map { __le32 cb_num_comps; @@ -136,7 +136,7 @@ struct exofs_dt_device_info { u8 systemid[OSD_SYSTEMID_LEN]; __le64 long_name_offset; /* If !0 then offset-in-file */ __le32 osdname_len; /* */ - u8 osdname[44]; /* Embbeded, Ususally an asci uuid */ + u8 osdname[44]; /* Embbeded, Usually an asci uuid */ } __packed; /* diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index 0d06f4e75699..8f44cef1b3ef 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -850,7 +850,7 @@ static int find_next_reservable_window( rsv_window_remove(sb, my_rsv); /* - * Let's book the whole avaliable window for now. We will check the + * Let's book the whole available window for now. We will check the * disk bitmap later and then, if there are free blocks then we adjust * the window size if it's larger than requested. * Otherwise, we will remove this node from the tree next time @@ -1357,9 +1357,9 @@ retry_alloc: goto allocated; } /* - * We may end up a bogus ealier ENOSPC error due to + * We may end up a bogus earlier ENOSPC error due to * filesystem is "full" of reservations, but - * there maybe indeed free blocks avaliable on disk + * there maybe indeed free blocks available on disk * In this case, we just forget about the reservations * just do block allocation as without reservations. */ diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index c47f706878b5..788e09a07f7e 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -305,7 +305,7 @@ static ext2_fsblk_t ext2_find_near(struct inode *inode, Indirect *ind) return ind->bh->b_blocknr; /* - * It is going to be refered from inode itself? OK, just put it into + * It is going to be referred from inode itself? OK, just put it into * the same cylinder group then. */ bg_start = ext2_group_first_block_no(inode->i_sb, ei->i_block_group); @@ -913,7 +913,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * * When we do truncate() we may have to clean the ends of several indirect * blocks but leave the blocks themselves alive. Block is partially - * truncated if some data below the new i_size is refered from it (and + * truncated if some data below the new i_size is referred from it (and * it is on the path to the first completely truncated data block, indeed). * We have to free the top of that path along with everything to the right * of the path. Since no allocation past the truncation point is possible @@ -990,7 +990,7 @@ no_top: * @p: array of block numbers * @q: points immediately past the end of array * - * We are freeing all blocks refered from that array (numbers are + * We are freeing all blocks referred from that array (numbers are * stored as little-endian 32-bit) and updating @inode->i_blocks * appropriately. */ @@ -1030,7 +1030,7 @@ static inline void ext2_free_data(struct inode *inode, __le32 *p, __le32 *q) * @q: pointer immediately past the end of array * @depth: depth of the branches to free * - * We are freeing all blocks refered from these branches (numbers are + * We are freeing all blocks referred from these branches (numbers are * stored as little-endian 32-bit) and updating @inode->i_blocks * appropriately. */ diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 7731695e65d9..0a78dae7e2cb 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1382,7 +1382,7 @@ static struct dentry *ext2_mount(struct file_system_type *fs_type, /* Read data from quotafile - avoid pagecache and such because we cannot afford * acquiring the locks... As quota files are never truncated and quota code - * itself serializes the operations (and noone else should touch the files) + * itself serializes the operations (and no one else should touch the files) * we don't have to be afraid of races */ static ssize_t ext2_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index c2e4dce984d2..529970617a21 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -35,7 +35,7 @@ * +------------------+ * * The block header is followed by multiple entry descriptors. These entry - * descriptors are variable in size, and alligned to EXT2_XATTR_PAD + * descriptors are variable in size, and aligned to EXT2_XATTR_PAD * byte boundaries. The entry descriptors are sorted by attribute name, * so that two extended attribute blocks can be compared efficiently. * diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index 153242187fce..fe52297e31ad 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -590,7 +590,7 @@ do_more: BUFFER_TRACE(debug_bh, "Deleted!"); if (!bh2jh(bitmap_bh)->b_committed_data) BUFFER_TRACE(debug_bh, - "No commited data in bitmap"); + "No committed data in bitmap"); BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap"); __brelse(debug_bh); } @@ -1063,7 +1063,7 @@ static int find_next_reservable_window( rsv_window_remove(sb, my_rsv); /* - * Let's book the whole avaliable window for now. We will check the + * Let's book the whole available window for now. We will check the * disk bitmap later and then, if there are free blocks then we adjust * the window size if it's larger than requested. * Otherwise, we will remove this node from the tree next time @@ -1456,7 +1456,7 @@ static int ext3_has_free_blocks(struct ext3_sb_info *sbi) * * ext3_should_retry_alloc() is called when ENOSPC is returned, and if * it is profitable to retry the operation, this function will wait - * for the current or commiting transaction to complete, and then + * for the current or committing transaction to complete, and then * return TRUE. * * if the total number of retries exceed three times, return FALSE. @@ -1632,9 +1632,9 @@ retry_alloc: goto allocated; } /* - * We may end up a bogus ealier ENOSPC error due to + * We may end up a bogus earlier ENOSPC error due to * filesystem is "full" of reservations, but - * there maybe indeed free blocks avaliable on disk + * there maybe indeed free blocks available on disk * In this case, we just forget about the reservations * just do block allocation as without reservations. */ diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index fe2541d250e4..68b2e43d7c35 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -2055,7 +2055,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * * When we do truncate() we may have to clean the ends of several * indirect blocks but leave the blocks themselves alive. Block is - * partially truncated if some data below the new i_size is refered + * partially truncated if some data below the new i_size is referred * from it (and it is on the path to the first completely truncated * data block, indeed). We have to free the top of that path along * with everything to the right of the path. Since no allocation @@ -2184,7 +2184,7 @@ static void ext3_clear_blocks(handle_t *handle, struct inode *inode, * @first: array of block numbers * @last: points immediately past the end of array * - * We are freeing all blocks refered from that array (numbers are stored as + * We are freeing all blocks referred from that array (numbers are stored as * little-endian 32-bit) and updating @inode->i_blocks appropriately. * * We accumulate contiguous runs of blocks to free. Conveniently, if these @@ -2272,7 +2272,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode, * @last: pointer immediately past the end of array * @depth: depth of the branches to free * - * We are freeing all blocks refered from these branches (numbers are + * We are freeing all blocks referred from these branches (numbers are * stored as little-endian 32-bit) and updating @inode->i_blocks * appropriately. */ @@ -3291,7 +3291,7 @@ static int ext3_writepage_trans_blocks(struct inode *inode) if (ext3_should_journal_data(inode)) ret = 3 * (bpp + indirects) + 2; else - ret = 2 * (bpp + indirects) + 2; + ret = 2 * (bpp + indirects) + indirects + 2; #ifdef CONFIG_QUOTA /* We know that structure was already allocated during dquot_initialize so diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 108b142e11ed..7916e4ce166a 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -1009,7 +1009,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, if (test_opt(sb, DEBUG)) printk(KERN_DEBUG "EXT3-fs: extending last group from "E3FSBLK - " upto "E3FSBLK" blocks\n", + " up to "E3FSBLK" blocks\n", o_blocks_count, n_blocks_count); if (n_blocks_count == 0 || n_blocks_count == o_blocks_count) diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 071689f86e18..3c6a9e0eadc1 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -2925,7 +2925,7 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id, /* Read data from quotafile - avoid pagecache and such because we cannot afford * acquiring the locks... As quota files are never truncated and quota code - * itself serializes the operations (and noone else should touch the files) + * itself serializes the operations (and no one else should touch the files) * we don't have to be afraid of races */ static ssize_t ext3_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 97b970e7dd13..1c67139ad4b4 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -547,7 +547,7 @@ int ext4_claim_free_blocks(struct ext4_sb_info *sbi, * * ext4_should_retry_alloc() is called when ENOSPC is returned, and if * it is profitable to retry the operation, this function will wait - * for the current or commiting transaction to complete, and then + * for the current or committing transaction to complete, and then * return TRUE. * * if the total number of retries exceed three times, return FALSE. diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index e25e99bf7ee1..d0f53538a57f 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -86,8 +86,8 @@ #ifdef CONFIG_QUOTA /* Amount of blocks needed for quota update - we know that the structure was - * allocated so we need to update only inode+data */ -#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) + * allocated so we need to update only data block */ +#define EXT4_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 1 : 0) /* Amount of blocks needed for quota insert/delete - we do some block writes * but inode, sb and group updates are done only once */ #define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index dd2cb5076ff9..4890d6f3ad15 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1729,7 +1729,7 @@ repeat: BUG_ON(npath->p_depth != path->p_depth); eh = npath[depth].p_hdr; if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) { - ext_debug("next leaf isnt full(%d)\n", + ext_debug("next leaf isn't full(%d)\n", le16_to_cpu(eh->eh_entries)); path = npath; goto repeat; @@ -2533,7 +2533,7 @@ static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex) /* * This function is called by ext4_ext_map_blocks() if someone tries to write * to an uninitialized extent. It may result in splitting the uninitialized - * extent into multiple extents (upto three - one initialized and two + * extent into multiple extents (up to three - one initialized and two * uninitialized). * There are three possibilities: * a> There is no split required: Entire extent should be initialized @@ -3174,7 +3174,7 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, path, flags); /* * Flag the inode(non aio case) or end_io struct (aio case) - * that this IO needs to convertion to written when IO is + * that this IO needs to conversion to written when IO is * completed */ if (io && !(io->flag & EXT4_IO_END_UNWRITTEN)) { @@ -3460,10 +3460,10 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, ext4_ext_mark_uninitialized(&newex); /* * io_end structure was created for every IO write to an - * uninitialized extent. To avoid unecessary conversion, + * uninitialized extent. To avoid unnecessary conversion, * here we flag the IO that really needs the conversion. * For non asycn direct IO case, flag the inode state - * that we need to perform convertion when IO is done. + * that we need to perform conversion when IO is done. */ if ((flags & EXT4_GET_BLOCKS_PRE_IO)) { if (io && !(io->flag & EXT4_IO_END_UNWRITTEN)) { diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 7f74019d6d77..e9473cbe80df 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -101,7 +101,7 @@ extern int ext4_flush_completed_IO(struct inode *inode) * to the work-to-be schedule is freed. * * Thus we need to keep the io structure still valid here after - * convertion finished. The io structure has a flag to + * conversion finished. The io structure has a flag to * avoid double converting from both fsync and background work * queue work. */ @@ -125,9 +125,11 @@ extern int ext4_flush_completed_IO(struct inode *inode) * the parent directory's parent as well, and so on recursively, if * they are also freshly created. */ -static void ext4_sync_parent(struct inode *inode) +static int ext4_sync_parent(struct inode *inode) { + struct writeback_control wbc; struct dentry *dentry = NULL; + int ret = 0; while (inode && ext4_test_inode_state(inode, EXT4_STATE_NEWENTRY)) { ext4_clear_inode_state(inode, EXT4_STATE_NEWENTRY); @@ -136,8 +138,17 @@ static void ext4_sync_parent(struct inode *inode) if (!dentry || !dentry->d_parent || !dentry->d_parent->d_inode) break; inode = dentry->d_parent->d_inode; - sync_mapping_buffers(inode->i_mapping); + ret = sync_mapping_buffers(inode->i_mapping); + if (ret) + break; + memset(&wbc, 0, sizeof(wbc)); + wbc.sync_mode = WB_SYNC_ALL; + wbc.nr_to_write = 0; /* only write out the inode */ + ret = sync_inode(inode, &wbc); + if (ret) + break; } + return ret; } /* @@ -176,7 +187,7 @@ int ext4_sync_file(struct file *file, int datasync) if (!journal) { ret = generic_file_fsync(file, datasync); if (!ret && !list_empty(&inode->i_dentry)) - ext4_sync_parent(inode); + ret = ext4_sync_parent(inode); goto out; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 1a86282b9024..f2fa5e8a582c 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2502,6 +2502,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, * for partial write. */ set_buffer_new(bh); + set_buffer_mapped(bh); } return 0; } @@ -2588,7 +2589,7 @@ static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate); * because we should have holes filled from ext4_page_mkwrite(). We even don't * need to file the inode to the transaction's list in ordered mode because if * we are writing back data added by write(), the inode is already there and if - * we are writing back data modified via mmap(), noone guarantees in which + * we are writing back data modified via mmap(), no one guarantees in which * transaction the data will hit the disk. In case we are journaling data, we * cannot start transaction directly because transaction start ranks above page * lock so we have to do some magic. @@ -2690,7 +2691,7 @@ static int ext4_writepage(struct page *page, /* * This is called via ext4_da_writepages() to - * calulate the total number of credits to reserve to fit + * calculate the total number of credits to reserve to fit * a single extent allocation into a single transaction, * ext4_da_writpeages() will loop calling this before * the block allocation. @@ -3304,7 +3305,7 @@ int ext4_alloc_da_blocks(struct inode *inode) * the pages by calling redirty_page_for_writepage() but that * would be ugly in the extreme. So instead we would need to * replicate parts of the code in the above functions, - * simplifying them becuase we wouldn't actually intend to + * simplifying them because we wouldn't actually intend to * write out the pages, but rather only collect contiguous * logical block extents, call the multi-block allocator, and * then update the buffer heads with the block allocations. @@ -3694,7 +3695,7 @@ retry: * * The unwrritten extents will be converted to written when DIO is completed. * For async direct IO, since the IO may still pending when return, we - * set up an end_io call back function, which will do the convertion + * set up an end_io call back function, which will do the conversion * when async direct IO completed. * * If the O_DIRECT write will extend the file then add this inode to the @@ -3717,7 +3718,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, * We could direct write to holes and fallocate. * * Allocated blocks to fill the hole are marked as uninitialized - * to prevent paralel buffered read to expose the stale data + * to prevent parallel buffered read to expose the stale data * before DIO complete the data IO. * * As to previously fallocated extents, ext4 get_block @@ -3778,7 +3779,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, int err; /* * for non AIO case, since the IO is already - * completed, we could do the convertion right here + * completed, we could do the conversion right here */ err = ext4_convert_unwritten_extents(inode, offset, ret); @@ -4025,7 +4026,7 @@ static inline int all_zeroes(__le32 *p, __le32 *q) * * When we do truncate() we may have to clean the ends of several * indirect blocks but leave the blocks themselves alive. Block is - * partially truncated if some data below the new i_size is refered + * partially truncated if some data below the new i_size is referred * from it (and it is on the path to the first completely truncated * data block, indeed). We have to free the top of that path along * with everything to the right of the path. Since no allocation @@ -4169,7 +4170,7 @@ out_err: * @first: array of block numbers * @last: points immediately past the end of array * - * We are freeing all blocks refered from that array (numbers are stored as + * We are freeing all blocks referred from that array (numbers are stored as * little-endian 32-bit) and updating @inode->i_blocks appropriately. * * We accumulate contiguous runs of blocks to free. Conveniently, if these @@ -4261,7 +4262,7 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, * @last: pointer immediately past the end of array * @depth: depth of the branches to free * - * We are freeing all blocks refered from these branches (numbers are + * We are freeing all blocks referred from these branches (numbers are * stored as little-endian 32-bit) and updating @inode->i_blocks * appropriately. */ @@ -4429,8 +4430,8 @@ void ext4_truncate(struct inode *inode) Indirect chain[4]; Indirect *partial; __le32 nr = 0; - int n; - ext4_lblk_t last_block; + int n = 0; + ext4_lblk_t last_block, max_block; unsigned blocksize = inode->i_sb->s_blocksize; trace_ext4_truncate_enter(inode); @@ -4455,14 +4456,18 @@ void ext4_truncate(struct inode *inode) last_block = (inode->i_size + blocksize-1) >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); + max_block = (EXT4_SB(inode->i_sb)->s_bitmap_maxbytes + blocksize-1) + >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); if (inode->i_size & (blocksize - 1)) if (ext4_block_truncate_page(handle, mapping, inode->i_size)) goto out_stop; - n = ext4_block_to_path(inode, last_block, offsets, NULL); - if (n == 0) - goto out_stop; /* error */ + if (last_block != max_block) { + n = ext4_block_to_path(inode, last_block, offsets, NULL); + if (n == 0) + goto out_stop; /* error */ + } /* * OK. This truncate is going to happen. We add the inode to the @@ -4493,7 +4498,13 @@ void ext4_truncate(struct inode *inode) */ ei->i_disksize = inode->i_size; - if (n == 1) { /* direct blocks */ + if (last_block == max_block) { + /* + * It is unnecessary to free any data blocks if last_block is + * equal to the indirect block limit. + */ + goto out_unlock; + } else if (n == 1) { /* direct blocks */ ext4_free_data(handle, inode, NULL, i_data+offsets[0], i_data + EXT4_NDIR_BLOCKS); goto do_indirects; @@ -4553,6 +4564,7 @@ do_indirects: ; } +out_unlock: up_write(&ei->i_data_sem); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); @@ -5398,13 +5410,12 @@ static int ext4_indirect_trans_blocks(struct inode *inode, int nrblocks, /* if nrblocks are contiguous */ if (chunk) { /* - * With N contiguous data blocks, it need at most - * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) indirect blocks - * 2 dindirect blocks - * 1 tindirect block + * With N contiguous data blocks, we need at most + * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) + 1 indirect blocks, + * 2 dindirect blocks, and 1 tindirect block */ - indirects = nrblocks / EXT4_ADDR_PER_BLOCK(inode->i_sb); - return indirects + 3; + return DIV_ROUND_UP(nrblocks, + EXT4_ADDR_PER_BLOCK(inode->i_sb)) + 4; } /* * if nrblocks are not contiguous, worse case, each block touch @@ -5478,7 +5489,7 @@ static int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) } /* - * Calulate the total number of credits to reserve to fit + * Calculate the total number of credits to reserve to fit * the modification of a single pages into a single transaction, * which may include multiple chunks of block allocations. * diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index a5837a837a8b..d8a16eecf1d5 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -92,7 +92,7 @@ * between CPUs. It is possible to get scheduled at this point. * * The locality group prealloc space is used looking at whether we have - * enough free space (pa_free) withing the prealloc space. + * enough free space (pa_free) within the prealloc space. * * If we can't allocate blocks via inode prealloc or/and locality group * prealloc then we look at the buddy cache. The buddy cache is represented diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index d1bafa57f483..92816b4e0f16 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -517,7 +517,7 @@ int ext4_ext_migrate(struct inode *inode) * start with one credit accounted for * superblock modification. * - * For the tmp_inode we already have commited the + * For the tmp_inode we already have committed the * trascation that created the inode. Later as and * when we add extents we extent the journal */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 22546ad7f0ae..8553dfb310af 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -242,27 +242,44 @@ static void ext4_put_nojournal(handle_t *handle) * journal_end calls result in the superblock being marked dirty, so * that sync() will call the filesystem's write_super callback if * appropriate. + * + * To avoid j_barrier hold in userspace when a user calls freeze(), + * ext4 prevents a new handle from being started by s_frozen, which + * is in an upper layer. */ handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks) { journal_t *journal; + handle_t *handle; if (sb->s_flags & MS_RDONLY) return ERR_PTR(-EROFS); - vfs_check_frozen(sb, SB_FREEZE_TRANS); - /* Special case here: if the journal has aborted behind our - * backs (eg. EIO in the commit thread), then we still need to - * take the FS itself readonly cleanly. */ journal = EXT4_SB(sb)->s_journal; - if (journal) { - if (is_journal_aborted(journal)) { - ext4_abort(sb, "Detected aborted journal"); - return ERR_PTR(-EROFS); - } - return jbd2_journal_start(journal, nblocks); + handle = ext4_journal_current_handle(); + + /* + * If a handle has been started, it should be allowed to + * finish, otherwise deadlock could happen between freeze + * and others(e.g. truncate) due to the restart of the + * journal handle if the filesystem is forzen and active + * handles are not stopped. + */ + if (!handle) + vfs_check_frozen(sb, SB_FREEZE_TRANS); + + if (!journal) + return ext4_get_nojournal(); + /* + * Special case here: if the journal has aborted behind our + * backs (eg. EIO in the commit thread), then we still need to + * take the FS itself readonly cleanly. + */ + if (is_journal_aborted(journal)) { + ext4_abort(sb, "Detected aborted journal"); + return ERR_PTR(-EROFS); } - return ext4_get_nojournal(); + return jbd2_journal_start(journal, nblocks); } /* @@ -617,7 +634,7 @@ __acquires(bitlock) * filesystem will have already been marked read/only and the * journal has been aborted. We return 1 as a hint to callers * who might what to use the return value from - * ext4_grp_locked_error() to distinguish beween the + * ext4_grp_locked_error() to distinguish between the * ERRORS_CONT and ERRORS_RO case, and perhaps return more * aggressively from the ext4 function in question, with a * more appropriate error code. @@ -2975,6 +2992,12 @@ static int ext4_register_li_request(struct super_block *sb, mutex_unlock(&ext4_li_info->li_list_mtx); sbi->s_li_request = elr; + /* + * set elr to NULL here since it has been inserted to + * the request_list and the removal and free of it is + * handled by ext4_clear_request_list from now on. + */ + elr = NULL; if (!(ext4_li_info->li_state & EXT4_LAZYINIT_RUNNING)) { ret = ext4_run_lazyinit_thread(); @@ -3385,6 +3408,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) get_random_bytes(&sbi->s_next_generation, sizeof(u32)); spin_lock_init(&sbi->s_next_gen_lock); + init_timer(&sbi->s_err_report); + sbi->s_err_report.function = print_daily_error_info; + sbi->s_err_report.data = (unsigned long) sb; + err = percpu_counter_init(&sbi->s_freeblocks_counter, ext4_count_free_blocks(sb)); if (!err) { @@ -3646,9 +3673,6 @@ no_journal: "Opts: %s%s%s", descr, sbi->s_es->s_mount_opts, *sbi->s_es->s_mount_opts ? "; " : "", orig_data); - init_timer(&sbi->s_err_report); - sbi->s_err_report.function = print_daily_error_info; - sbi->s_err_report.data = (unsigned long) sb; if (es->s_error_count) mod_timer(&sbi->s_err_report, jiffies + 300*HZ); /* 5 minutes */ @@ -3672,6 +3696,7 @@ failed_mount_wq: sbi->s_journal = NULL; } failed_mount3: + del_timer(&sbi->s_err_report); if (sbi->s_flex_groups) { if (is_vmalloc_addr(sbi->s_flex_groups)) vfree(sbi->s_flex_groups); @@ -4138,6 +4163,11 @@ static int ext4_sync_fs(struct super_block *sb, int wait) /* * LVM calls this function before a (read-only) snapshot is created. This * gives us a chance to flush the journal completely and mark the fs clean. + * + * Note that only this function cannot bring a filesystem to be in a clean + * state independently, because ext4 prevents a new handle from being started + * by @sb->s_frozen, which stays in an upper layer. It thus needs help from + * the upper layer. */ static int ext4_freeze(struct super_block *sb) { @@ -4614,17 +4644,30 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, static int ext4_quota_off(struct super_block *sb, int type) { + struct inode *inode = sb_dqopt(sb)->files[type]; + handle_t *handle; + /* Force all delayed allocation blocks to be allocated. * Caller already holds s_umount sem */ if (test_opt(sb, DELALLOC)) sync_filesystem(sb); + /* Update modification times of quota files when userspace can + * start looking at them */ + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) + goto out; + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + ext4_mark_inode_dirty(handle, inode); + ext4_journal_stop(handle); + +out: return dquot_quota_off(sb, type); } /* Read data from quotafile - avoid pagecache and such because we cannot afford * acquiring the locks... As quota files are never truncated and quota code - * itself serializes the operations (and noone else should touch the files) + * itself serializes the operations (and no one else should touch the files) * we don't have to be afraid of races */ static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) @@ -4714,9 +4757,8 @@ out: if (inode->i_size < off + len) { i_size_write(inode, off + len); EXT4_I(inode)->i_disksize = inode->i_size; + ext4_mark_inode_dirty(handle, inode); } - inode->i_mtime = inode->i_ctime = CURRENT_TIME; - ext4_mark_inode_dirty(handle, inode); mutex_unlock(&inode->i_mutex); return len; } diff --git a/fs/fhandle.c b/fs/fhandle.c index bf93ad2bee07..6b088641f5bf 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -7,6 +7,7 @@ #include <linux/exportfs.h> #include <linux/fs_struct.h> #include <linux/fsnotify.h> +#include <linux/personality.h> #include <asm/uaccess.h> #include "internal.h" diff --git a/fs/file.c b/fs/file.c index 0be344755c02..4c6992d8f3ba 100644 --- a/fs/file.c +++ b/fs/file.c @@ -9,6 +9,7 @@ #include <linux/module.h> #include <linux/fs.h> #include <linux/mm.h> +#include <linux/mmzone.h> #include <linux/time.h> #include <linux/sched.h> #include <linux/slab.h> @@ -39,14 +40,17 @@ int sysctl_nr_open_max = 1024 * 1024; /* raised later */ */ static DEFINE_PER_CPU(struct fdtable_defer, fdtable_defer_list); -static inline void *alloc_fdmem(unsigned int size) +static void *alloc_fdmem(unsigned int size) { - void *data; - - data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN); - if (data != NULL) - return data; - + /* + * Very large allocations can stress page reclaim, so fall back to + * vmalloc() if the allocation size will be considered "large" by the VM. + */ + if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) { + void *data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN); + if (data != NULL) + return data; + } return vmalloc(size); } diff --git a/fs/filesystems.c b/fs/filesystems.c index 751d6b255a12..0845f84f2a5f 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -110,14 +110,13 @@ int unregister_filesystem(struct file_system_type * fs) *tmp = fs->next; fs->next = NULL; write_unlock(&file_systems_lock); + synchronize_rcu(); return 0; } tmp = &(*tmp)->next; } write_unlock(&file_systems_lock); - synchronize_rcu(); - return -EINVAL; } diff --git a/fs/freevxfs/vxfs_fshead.c b/fs/freevxfs/vxfs_fshead.c index 78948b4b1894..c9a6a94e58e9 100644 --- a/fs/freevxfs/vxfs_fshead.c +++ b/fs/freevxfs/vxfs_fshead.c @@ -164,7 +164,7 @@ vxfs_read_fshead(struct super_block *sbp) goto out_free_pfp; } if (!VXFS_ISILT(VXFS_INO(infp->vsi_stilist))) { - printk(KERN_ERR "vxfs: structual list inode is of wrong type (%x)\n", + printk(KERN_ERR "vxfs: structural list inode is of wrong type (%x)\n", VXFS_INO(infp->vsi_stilist)->vii_mode & VXFS_TYPE_MASK); goto out_iput_stilist; } diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index 6c5131d592f0..3360f1e678ad 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -162,7 +162,7 @@ vxfs_find_entry(struct inode *ip, struct dentry *dp, struct page **ppp) /** * vxfs_inode_by_name - find inode number for dentry * @dip: directory to search in - * @dp: dentry we seach for + * @dp: dentry we search for * * Description: * vxfs_inode_by_name finds out the inode number of diff --git a/fs/freevxfs/vxfs_olt.h b/fs/freevxfs/vxfs_olt.h index d8324296486f..b7b3af502615 100644 --- a/fs/freevxfs/vxfs_olt.h +++ b/fs/freevxfs/vxfs_olt.h @@ -60,7 +60,7 @@ enum { * * The Object Location Table header is placed at the beginning of each * OLT extent. It is used to fing certain filesystem-wide metadata, e.g. - * the inital inode list, the fileset header or the device configuration. + * the initial inode list, the fileset header or the device configuration. */ struct vxfs_olt { u_int32_t olt_magic; /* magic number */ diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index b5ed541fb137..34591ee804b5 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -144,7 +144,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, * * Description: * This does WB_SYNC_NONE opportunistic writeback. The IO is only - * started when this function returns, we make no guarentees on + * started when this function returns, we make no guarantees on * completion. Caller need not hold sb s_umount semaphore. * */ diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index c6ba49bd95b3..b32eb29a4e6f 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -174,7 +174,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) if (!inode) return 0; - if (nd->flags & LOOKUP_RCU) + if (nd && (nd->flags & LOOKUP_RCU)) return -ECHILD; fc = get_fuse_conn(inode); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 6ea00734984e..82a66466a24c 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -523,7 +523,7 @@ static int fuse_readpage(struct file *file, struct page *page) goto out; /* - * Page writeback can extend beyond the liftime of the + * Page writeback can extend beyond the lifetime of the * page-cache page, so make sure we read a properly synced * page. */ diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile index f3d23ef4e876..86128202384f 100644 --- a/fs/gfs2/Makefile +++ b/fs/gfs2/Makefile @@ -1,9 +1,9 @@ ccflags-y := -I$(src) obj-$(CONFIG_GFS2_FS) += gfs2.o gfs2-y := acl.o bmap.o dir.o xattr.o glock.o \ - glops.o inode.o log.o lops.o main.o meta_io.o \ + glops.o log.o lops.o main.o meta_io.o \ aops.o dentry.o export.o file.o \ - ops_fstype.o ops_inode.o quota.o \ + ops_fstype.o inode.o quota.o \ recovery.o rgrp.o super.o sys.o trans.o util.o gfs2-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index c71995b111bf..802ac5eeba28 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -884,8 +884,8 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping, } brelse(dibh); - gfs2_trans_end(sdp); failed: + gfs2_trans_end(sdp); if (al) { gfs2_inplace_release(ip); gfs2_quota_unlock(ip); @@ -1076,8 +1076,8 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) bd = bh->b_private; if (bd && bd->bd_ail) goto cannot_release; - gfs2_assert_warn(sdp, !buffer_pinned(bh)); - gfs2_assert_warn(sdp, !buffer_dirty(bh)); + if (buffer_pinned(bh) || buffer_dirty(bh)) + goto not_possible; bh = bh->b_this_page; } while(bh != head); gfs2_log_unlock(sdp); @@ -1107,6 +1107,10 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) } while (bh != head); return try_to_free_buffers(page); + +not_possible: /* Should never happen */ + WARN_ON(buffer_dirty(bh)); + WARN_ON(buffer_pinned(bh)); cannot_release: gfs2_log_unlock(sdp); return 0; diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index ef3dc4b9fae2..74add2ddcc3f 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1136,7 +1136,7 @@ void gfs2_trim_blocks(struct inode *inode) * earlier versions of GFS2 have a bug in the stuffed file reading * code which will result in a buffer overrun if the size is larger * than the max stuffed file size. In order to prevent this from - * occuring, such files are unstuffed, but in other cases we can + * occurring, such files are unstuffed, but in other cases we can * just update the inode size directly. * * Returns: 0 on success, or -ve on error diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 5c356d09c321..091ee4779538 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -82,12 +82,9 @@ struct qstr gfs2_qdot __read_mostly; struct qstr gfs2_qdotdot __read_mostly; -typedef int (*leaf_call_t) (struct gfs2_inode *dip, u32 index, u32 len, - u64 leaf_no, void *data); typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent, const struct qstr *name, void *opaque); - int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block, struct buffer_head **bhp) { @@ -1506,7 +1503,7 @@ struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name) inode = gfs2_inode_lookup(dir->i_sb, be16_to_cpu(dent->de_type), be64_to_cpu(dent->de_inum.no_addr), - be64_to_cpu(dent->de_inum.no_formal_ino)); + be64_to_cpu(dent->de_inum.no_formal_ino), 0); brelse(bh); return inode; } @@ -1600,7 +1597,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) */ int gfs2_dir_add(struct inode *inode, const struct qstr *name, - const struct gfs2_inode *nip, unsigned type) + const struct gfs2_inode *nip) { struct gfs2_inode *ip = GFS2_I(inode); struct buffer_head *bh; @@ -1616,7 +1613,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, return PTR_ERR(dent); dent = gfs2_init_dirent(inode, dent, name, bh); gfs2_inum_out(nip, dent); - dent->de_type = cpu_to_be16(type); + dent->de_type = cpu_to_be16(IF2DT(nip->i_inode.i_mode)); if (ip->i_diskflags & GFS2_DIF_EXHASH) { leaf = (struct gfs2_leaf *)bh->b_data; be16_add_cpu(&leaf->lf_entries, 1); @@ -1628,6 +1625,8 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_entries++; ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME; + if (S_ISDIR(nip->i_inode.i_mode)) + inc_nlink(&ip->i_inode); gfs2_dinode_out(ip, bh->b_data); brelse(bh); error = 0; @@ -1672,8 +1671,9 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, * Returns: 0 on success, error code on failure */ -int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) +int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry) { + const struct qstr *name = &dentry->d_name; struct gfs2_dirent *dent, *prev = NULL; struct buffer_head *bh; int error; @@ -1714,6 +1714,8 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) gfs2_trans_add_bh(dip->i_gl, bh, 1); dip->i_entries--; dip->i_inode.i_mtime = dip->i_inode.i_ctime = CURRENT_TIME; + if (S_ISDIR(dentry->d_inode->i_mode)) + drop_nlink(&dip->i_inode); gfs2_dinode_out(dip, bh->b_data); brelse(bh); mark_inode_dirty(&dip->i_inode); @@ -1768,94 +1770,20 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, } /** - * foreach_leaf - call a function for each leaf in a directory - * @dip: the directory - * @lc: the function to call for each each - * @data: private data to pass to it - * - * Returns: errno - */ - -static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data) -{ - struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct buffer_head *bh; - struct gfs2_leaf *leaf; - u32 hsize, len; - u32 ht_offset, lp_offset, ht_offset_cur = -1; - u32 index = 0; - __be64 *lp; - u64 leaf_no; - int error = 0; - - hsize = 1 << dip->i_depth; - if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) { - gfs2_consist_inode(dip); - return -EIO; - } - - lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS); - if (!lp) - return -ENOMEM; - - while (index < hsize) { - lp_offset = index & (sdp->sd_hash_ptrs - 1); - ht_offset = index - lp_offset; - - if (ht_offset_cur != ht_offset) { - error = gfs2_dir_read_data(dip, (char *)lp, - ht_offset * sizeof(__be64), - sdp->sd_hash_bsize, 1); - if (error != sdp->sd_hash_bsize) { - if (error >= 0) - error = -EIO; - goto out; - } - ht_offset_cur = ht_offset; - } - - leaf_no = be64_to_cpu(lp[lp_offset]); - if (leaf_no) { - error = get_leaf(dip, leaf_no, &bh); - if (error) - goto out; - leaf = (struct gfs2_leaf *)bh->b_data; - len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth)); - brelse(bh); - - error = lc(dip, index, len, leaf_no, data); - if (error) - goto out; - - index = (index & ~(len - 1)) + len; - } else - index++; - } - - if (index != hsize) { - gfs2_consist_inode(dip); - error = -EIO; - } - -out: - kfree(lp); - - return error; -} - -/** * leaf_dealloc - Deallocate a directory leaf * @dip: the directory * @index: the hash table offset in the directory * @len: the number of pointers to this leaf * @leaf_no: the leaf number - * @data: not used + * @leaf_bh: buffer_head for the starting leaf + * last_dealloc: 1 if this is the final dealloc for the leaf, else 0 * * Returns: errno */ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, - u64 leaf_no, void *data) + u64 leaf_no, struct buffer_head *leaf_bh, + int last_dealloc) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_leaf *tmp_leaf; @@ -1887,14 +1815,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_qs; /* Count the number of leaves */ + bh = leaf_bh; for (blk = leaf_no; blk; blk = nblk) { - error = get_leaf(dip, blk, &bh); - if (error) - goto out_rlist; + if (blk != leaf_no) { + error = get_leaf(dip, blk, &bh); + if (error) + goto out_rlist; + } tmp_leaf = (struct gfs2_leaf *)bh->b_data; nblk = be64_to_cpu(tmp_leaf->lf_next); - brelse(bh); + if (blk != leaf_no) + brelse(bh); gfs2_rlist_add(sdp, &rlist, blk); l_blocks++; @@ -1918,13 +1850,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, if (error) goto out_rg_gunlock; + bh = leaf_bh; + for (blk = leaf_no; blk; blk = nblk) { - error = get_leaf(dip, blk, &bh); - if (error) - goto out_end_trans; + if (blk != leaf_no) { + error = get_leaf(dip, blk, &bh); + if (error) + goto out_end_trans; + } tmp_leaf = (struct gfs2_leaf *)bh->b_data; nblk = be64_to_cpu(tmp_leaf->lf_next); - brelse(bh); + if (blk != leaf_no) + brelse(bh); gfs2_free_meta(dip, blk, 1); gfs2_add_inode_blocks(&dip->i_inode, -1); @@ -1942,6 +1879,10 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_end_trans; gfs2_trans_add_bh(dip->i_gl, dibh, 1); + /* On the last dealloc, make this a regular file in case we crash. + (We don't want to free these blocks a second time.) */ + if (last_dealloc) + dip->i_inode.i_mode = S_IFREG; gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); @@ -1975,29 +1916,67 @@ int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct buffer_head *bh; - int error; + struct gfs2_leaf *leaf; + u32 hsize, len; + u32 ht_offset, lp_offset, ht_offset_cur = -1; + u32 index = 0, next_index; + __be64 *lp; + u64 leaf_no; + int error = 0, last; - /* Dealloc on-disk leaves to FREEMETA state */ - error = foreach_leaf(dip, leaf_dealloc, NULL); - if (error) - return error; + hsize = 1 << dip->i_depth; + if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) { + gfs2_consist_inode(dip); + return -EIO; + } - /* Make this a regular file in case we crash. - (We don't want to free these blocks a second time.) */ + lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS); + if (!lp) + return -ENOMEM; - error = gfs2_trans_begin(sdp, RES_DINODE, 0); - if (error) - return error; + while (index < hsize) { + lp_offset = index & (sdp->sd_hash_ptrs - 1); + ht_offset = index - lp_offset; - error = gfs2_meta_inode_buffer(dip, &bh); - if (!error) { - gfs2_trans_add_bh(dip->i_gl, bh, 1); - ((struct gfs2_dinode *)bh->b_data)->di_mode = - cpu_to_be32(S_IFREG); - brelse(bh); + if (ht_offset_cur != ht_offset) { + error = gfs2_dir_read_data(dip, (char *)lp, + ht_offset * sizeof(__be64), + sdp->sd_hash_bsize, 1); + if (error != sdp->sd_hash_bsize) { + if (error >= 0) + error = -EIO; + goto out; + } + ht_offset_cur = ht_offset; + } + + leaf_no = be64_to_cpu(lp[lp_offset]); + if (leaf_no) { + error = get_leaf(dip, leaf_no, &bh); + if (error) + goto out; + leaf = (struct gfs2_leaf *)bh->b_data; + len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth)); + + next_index = (index & ~(len - 1)) + len; + last = ((next_index >= hsize) ? 1 : 0); + error = leaf_dealloc(dip, index, len, leaf_no, bh, + last); + brelse(bh); + if (error) + goto out; + index = next_index; + } else + index++; } - gfs2_trans_end(sdp); + if (index != hsize) { + gfs2_consist_inode(dip); + error = -EIO; + } + +out: + kfree(lp); return error; } diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index a98f644bd3df..e686af11becd 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -22,8 +22,8 @@ extern struct inode *gfs2_dir_search(struct inode *dir, extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename, const struct gfs2_inode *ip); extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename, - const struct gfs2_inode *ip, unsigned int type); -extern int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename); + const struct gfs2_inode *ip); +extern int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry); extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque, filldir_t filldir); extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index b5a5e60df0d5..fe9945f2ff72 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -139,7 +139,7 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, struct gfs2_sbd *sdp = sb->s_fs_info; struct inode *inode; - inode = gfs2_ilookup(sb, inum->no_addr); + inode = gfs2_ilookup(sb, inum->no_addr, 0); if (inode) { if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) { iput(inode); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index b2682e073eee..a9f5cbe45cd9 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -545,18 +545,10 @@ static int gfs2_close(struct inode *inode, struct file *file) /** * gfs2_fsync - sync the dirty data for a file (across the cluster) * @file: the file that points to the dentry (we ignore this) - * @dentry: the dentry that points to the inode to sync + * @datasync: set if we can ignore timestamp changes * - * The VFS will flush "normal" data for us. We only need to worry - * about metadata here. For journaled data, we just do a log flush - * as we can't avoid it. Otherwise we can just bale out if datasync - * is set. For stuffed inodes we must flush the log in order to - * ensure that all data is on disk. - * - * The call to write_inode_now() is there to write back metadata and - * the inode itself. It does also try and write the data, but thats - * (hopefully) a no-op due to the VFS having already called filemap_fdatawrite() - * for us. + * The VFS will flush data for us. We only need to worry + * about metadata here. * * Returns: errno */ @@ -565,22 +557,20 @@ static int gfs2_fsync(struct file *file, int datasync) { struct inode *inode = file->f_mapping->host; int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC); - int ret = 0; - - if (gfs2_is_jdata(GFS2_I(inode))) { - gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); - return 0; - } + struct gfs2_inode *ip = GFS2_I(inode); + int ret; - if (sync_state != 0) { - if (!datasync) - ret = write_inode_now(inode, 0); + if (datasync) + sync_state &= ~I_DIRTY_SYNC; - if (gfs2_is_stuffed(GFS2_I(inode))) - gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); + if (sync_state) { + ret = sync_inode_metadata(inode, 1); + if (ret) + return ret; + gfs2_ail_flush(ip->i_gl); } - return ret; + return 0; } /** @@ -617,18 +607,51 @@ static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov, return generic_file_aio_write(iocb, iov, nr_segs, pos); } -static void empty_write_end(struct page *page, unsigned from, - unsigned to) +static int empty_write_end(struct page *page, unsigned from, + unsigned to, int mode) { - struct gfs2_inode *ip = GFS2_I(page->mapping->host); + struct inode *inode = page->mapping->host; + struct gfs2_inode *ip = GFS2_I(inode); + struct buffer_head *bh; + unsigned offset, blksize = 1 << inode->i_blkbits; + pgoff_t end_index = i_size_read(inode) >> PAGE_CACHE_SHIFT; zero_user(page, from, to-from); mark_page_accessed(page); - if (!gfs2_is_writeback(ip)) - gfs2_page_add_databufs(ip, page, from, to); + if (page->index < end_index || !(mode & FALLOC_FL_KEEP_SIZE)) { + if (!gfs2_is_writeback(ip)) + gfs2_page_add_databufs(ip, page, from, to); - block_commit_write(page, from, to); + block_commit_write(page, from, to); + return 0; + } + + offset = 0; + bh = page_buffers(page); + while (offset < to) { + if (offset >= from) { + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + clear_buffer_new(bh); + write_dirty_buffer(bh, WRITE); + } + offset += blksize; + bh = bh->b_this_page; + } + + offset = 0; + bh = page_buffers(page); + while (offset < to) { + if (offset >= from) { + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) + return -EIO; + } + offset += blksize; + bh = bh->b_this_page; + } + return 0; } static int needs_empty_write(sector_t block, struct inode *inode) @@ -643,7 +666,8 @@ static int needs_empty_write(sector_t block, struct inode *inode) return !buffer_mapped(&bh_map); } -static int write_empty_blocks(struct page *page, unsigned from, unsigned to) +static int write_empty_blocks(struct page *page, unsigned from, unsigned to, + int mode) { struct inode *inode = page->mapping->host; unsigned start, end, next, blksize; @@ -668,7 +692,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to) gfs2_block_map); if (unlikely(ret)) return ret; - empty_write_end(page, start, end); + ret = empty_write_end(page, start, end, mode); + if (unlikely(ret)) + return ret; end = 0; } start = next; @@ -682,7 +708,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to) ret = __block_write_begin(page, start, end - start, gfs2_block_map); if (unlikely(ret)) return ret; - empty_write_end(page, start, end); + ret = empty_write_end(page, start, end, mode); + if (unlikely(ret)) + return ret; } return 0; @@ -731,7 +759,7 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len, if (curr == end) to = end_offset; - error = write_empty_blocks(page, from, to); + error = write_empty_blocks(page, from, to, mode); if (!error && offset + to > inode->i_size && !(mode & FALLOC_FL_KEEP_SIZE)) { i_size_write(inode, offset + to); @@ -788,6 +816,7 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t bytes, max_bytes; struct gfs2_alloc *al; int error; + loff_t bsize_mask = ~((loff_t)sdp->sd_sb.sb_bsize - 1); loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift; next = (next + 1) << sdp->sd_sb.sb_bsize_shift; @@ -795,13 +824,15 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, if (mode & ~FALLOC_FL_KEEP_SIZE) return -EOPNOTSUPP; - offset = (offset >> sdp->sd_sb.sb_bsize_shift) << - sdp->sd_sb.sb_bsize_shift; + offset &= bsize_mask; len = next - offset; bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2; if (!bytes) bytes = UINT_MAX; + bytes &= bsize_mask; + if (bytes == 0) + bytes = sdp->sd_sb.sb_bsize; gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh); error = gfs2_glock_nq(&ip->i_gh); @@ -832,6 +863,9 @@ retry: if (error) { if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) { bytes >>= 1; + bytes &= bsize_mask; + if (bytes == 0) + bytes = sdp->sd_sb.sb_bsize; goto retry; } goto out_qunlock; diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index e2431313491f..a2a6abbccc07 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -93,14 +93,12 @@ static unsigned int gl_hash(const struct gfs2_sbd *sdp, static inline void spin_lock_bucket(unsigned int hash) { - struct hlist_bl_head *bl = &gl_hash_table[hash]; - bit_spin_lock(0, (unsigned long *)bl); + hlist_bl_lock(&gl_hash_table[hash]); } static inline void spin_unlock_bucket(unsigned int hash) { - struct hlist_bl_head *bl = &gl_hash_table[hash]; - __bit_spin_unlock(0, (unsigned long *)bl); + hlist_bl_unlock(&gl_hash_table[hash]); } static void gfs2_glock_dealloc(struct rcu_head *rcu) @@ -145,14 +143,9 @@ static int demote_ok(const struct gfs2_glock *gl) { const struct gfs2_glock_operations *glops = gl->gl_ops; - /* assert_spin_locked(&gl->gl_spin); */ - if (gl->gl_state == LM_ST_UNLOCKED) return 0; - if (test_bit(GLF_LFLUSH, &gl->gl_flags)) - return 0; - if ((gl->gl_name.ln_type != LM_TYPE_INODE) && - !list_empty(&gl->gl_holders)) + if (!list_empty(&gl->gl_holders)) return 0; if (glops->go_demote_ok) return glops->go_demote_ok(gl); @@ -160,6 +153,31 @@ static int demote_ok(const struct gfs2_glock *gl) } +void gfs2_glock_add_to_lru(struct gfs2_glock *gl) +{ + spin_lock(&lru_lock); + + if (!list_empty(&gl->gl_lru)) + list_del_init(&gl->gl_lru); + else + atomic_inc(&lru_count); + + list_add_tail(&gl->gl_lru, &lru_list); + set_bit(GLF_LRU, &gl->gl_flags); + spin_unlock(&lru_lock); +} + +static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl) +{ + spin_lock(&lru_lock); + if (!list_empty(&gl->gl_lru)) { + list_del_init(&gl->gl_lru); + atomic_dec(&lru_count); + clear_bit(GLF_LRU, &gl->gl_flags); + } + spin_unlock(&lru_lock); +} + /** * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list * @gl: the glock @@ -170,24 +188,8 @@ static int demote_ok(const struct gfs2_glock *gl) static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl) { - if (demote_ok(gl)) { - spin_lock(&lru_lock); - - if (!list_empty(&gl->gl_lru)) - list_del_init(&gl->gl_lru); - else - atomic_inc(&lru_count); - - list_add_tail(&gl->gl_lru, &lru_list); - spin_unlock(&lru_lock); - } -} - -void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl) -{ - spin_lock(&gl->gl_spin); - __gfs2_glock_schedule_for_reclaim(gl); - spin_unlock(&gl->gl_spin); + if (demote_ok(gl)) + gfs2_glock_add_to_lru(gl); } /** @@ -219,12 +221,7 @@ void gfs2_glock_put(struct gfs2_glock *gl) spin_lock_bucket(gl->gl_hash); hlist_bl_del_rcu(&gl->gl_list); spin_unlock_bucket(gl->gl_hash); - spin_lock(&lru_lock); - if (!list_empty(&gl->gl_lru)) { - list_del_init(&gl->gl_lru); - atomic_dec(&lru_count); - } - spin_unlock(&lru_lock); + gfs2_glock_remove_from_lru(gl); GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders)); GLOCK_BUG_ON(gl, mapping && mapping->nrpages); trace_gfs2_glock_put(gl); @@ -544,11 +541,6 @@ __acquires(&gl->gl_spin) clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); gfs2_glock_hold(gl); - if (target != LM_ST_UNLOCKED && (gl->gl_state == LM_ST_SHARED || - gl->gl_state == LM_ST_DEFERRED) && - !(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) - lck_flags |= LM_FLAG_TRY_1CB; - if (sdp->sd_lockstruct.ls_ops->lm_lock) { /* lock_dlm */ ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); @@ -650,7 +642,7 @@ static void delete_work_func(struct work_struct *work) /* Note: Unsafe to dereference ip as we don't hold right refs/locks */ if (ip) - inode = gfs2_ilookup(sdp->sd_vfs, no_addr); + inode = gfs2_ilookup(sdp->sd_vfs, no_addr, 1); else inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED); if (inode && !IS_ERR(inode)) { @@ -1027,6 +1019,9 @@ int gfs2_glock_nq(struct gfs2_holder *gh) if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) return -EIO; + if (test_bit(GLF_LRU, &gl->gl_flags)) + gfs2_glock_remove_from_lru(gl); + spin_lock(&gl->gl_spin); add_to_queue(gh); if ((LM_FLAG_NOEXP & gh->gh_flags) && @@ -1084,7 +1079,8 @@ void gfs2_glock_dq(struct gfs2_holder *gh) !test_bit(GLF_DEMOTE, &gl->gl_flags)) fast_path = 1; } - __gfs2_glock_schedule_for_reclaim(gl); + if (!test_bit(GLF_LFLUSH, &gl->gl_flags)) + __gfs2_glock_schedule_for_reclaim(gl); trace_gfs2_glock_queue(gh, 0); spin_unlock(&gl->gl_spin); if (likely(fast_path)) @@ -1123,7 +1119,7 @@ void gfs2_glock_dq_uninit(struct gfs2_holder *gh) * @number: the lock number * @glops: the glock operations for the type of glock * @state: the state to acquire the glock in - * @flags: modifier flags for the aquisition + * @flags: modifier flags for the acquisition * @gh: the struct gfs2_holder * * Returns: errno @@ -1367,6 +1363,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m while(nr && !list_empty(&lru_list)) { gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru); list_del_init(&gl->gl_lru); + clear_bit(GLF_LRU, &gl->gl_flags); atomic_dec(&lru_count); /* Test for being demotable */ @@ -1389,6 +1386,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m } nr_skipped++; list_add(&gl->gl_lru, &skipped); + set_bit(GLF_LRU, &gl->gl_flags); } list_splice(&skipped, &lru_list); atomic_add(nr_skipped, &lru_count); @@ -1461,12 +1459,7 @@ static void thaw_glock(struct gfs2_glock *gl) static void clear_glock(struct gfs2_glock *gl) { - spin_lock(&lru_lock); - if (!list_empty(&gl->gl_lru)) { - list_del_init(&gl->gl_lru); - atomic_dec(&lru_count); - } - spin_unlock(&lru_lock); + gfs2_glock_remove_from_lru(gl); spin_lock(&gl->gl_spin); if (gl->gl_state != LM_ST_UNLOCKED) @@ -1601,9 +1594,11 @@ static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh) return 0; } -static const char *gflags2str(char *buf, const unsigned long *gflags) +static const char *gflags2str(char *buf, const struct gfs2_glock *gl) { + const unsigned long *gflags = &gl->gl_flags; char *p = buf; + if (test_bit(GLF_LOCK, gflags)) *p++ = 'l'; if (test_bit(GLF_DEMOTE, gflags)) @@ -1626,6 +1621,10 @@ static const char *gflags2str(char *buf, const unsigned long *gflags) *p++ = 'F'; if (test_bit(GLF_QUEUED, gflags)) *p++ = 'q'; + if (test_bit(GLF_LRU, gflags)) + *p++ = 'L'; + if (gl->gl_object) + *p++ = 'o'; *p = 0; return buf; } @@ -1660,14 +1659,15 @@ static int __dump_glock(struct seq_file *seq, const struct gfs2_glock *gl) dtime *= 1000000/HZ; /* demote time in uSec */ if (!test_bit(GLF_DEMOTE, &gl->gl_flags)) dtime = 0; - gfs2_print_dbg(seq, "G: s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d r:%d\n", + gfs2_print_dbg(seq, "G: s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d v:%d r:%d\n", state2str(gl->gl_state), gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number, - gflags2str(gflags_buf, &gl->gl_flags), + gflags2str(gflags_buf, gl), state2str(gl->gl_target), state2str(gl->gl_demote_state), dtime, atomic_read(&gl->gl_ail_count), + atomic_read(&gl->gl_revokes), atomic_read(&gl->gl_ref)); list_for_each_entry(gh, &gl->gl_holders, gh_list) { diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index aea160690e94..6b2f757b9281 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -225,11 +225,10 @@ static inline int gfs2_glock_nq_init(struct gfs2_glock *gl, extern void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state); extern void gfs2_glock_complete(struct gfs2_glock *gl, int ret); -extern void gfs2_reclaim_glock(struct gfs2_sbd *sdp); extern void gfs2_gl_hash_clear(struct gfs2_sbd *sdp); extern void gfs2_glock_finish_truncate(struct gfs2_inode *ip); extern void gfs2_glock_thaw(struct gfs2_sbd *sdp); -extern void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl); +extern void gfs2_glock_add_to_lru(struct gfs2_glock *gl); extern void gfs2_glock_free(struct gfs2_glock *gl); extern int __init gfs2_glock_init(void); diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 3754e3cbf02b..8ef70f464731 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -28,33 +28,18 @@ #include "trans.h" /** - * ail_empty_gl - remove all buffers for a given lock from the AIL + * __gfs2_ail_flush - remove all buffers for a given lock from the AIL * @gl: the glock * * None of the buffers should be dirty, locked, or pinned. */ -static void gfs2_ail_empty_gl(struct gfs2_glock *gl) +static void __gfs2_ail_flush(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_sbd; struct list_head *head = &gl->gl_ail_list; struct gfs2_bufdata *bd; struct buffer_head *bh; - struct gfs2_trans tr; - - memset(&tr, 0, sizeof(tr)); - tr.tr_revokes = atomic_read(&gl->gl_ail_count); - - if (!tr.tr_revokes) - return; - - /* A shortened, inline version of gfs2_trans_begin() */ - tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64)); - tr.tr_ip = (unsigned long)__builtin_return_address(0); - INIT_LIST_HEAD(&tr.tr_list_buf); - gfs2_log_reserve(sdp, tr.tr_reserved); - BUG_ON(current->journal_info); - current->journal_info = &tr; spin_lock(&sdp->sd_ail_lock); while (!list_empty(head)) { @@ -76,11 +61,51 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl) } gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count)); spin_unlock(&sdp->sd_ail_lock); +} + + +static void gfs2_ail_empty_gl(struct gfs2_glock *gl) +{ + struct gfs2_sbd *sdp = gl->gl_sbd; + struct gfs2_trans tr; + + memset(&tr, 0, sizeof(tr)); + tr.tr_revokes = atomic_read(&gl->gl_ail_count); + + if (!tr.tr_revokes) + return; + + /* A shortened, inline version of gfs2_trans_begin() */ + tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64)); + tr.tr_ip = (unsigned long)__builtin_return_address(0); + INIT_LIST_HEAD(&tr.tr_list_buf); + gfs2_log_reserve(sdp, tr.tr_reserved); + BUG_ON(current->journal_info); + current->journal_info = &tr; + + __gfs2_ail_flush(gl); gfs2_trans_end(sdp); gfs2_log_flush(sdp, NULL); } +void gfs2_ail_flush(struct gfs2_glock *gl) +{ + struct gfs2_sbd *sdp = gl->gl_sbd; + unsigned int revokes = atomic_read(&gl->gl_ail_count); + int ret; + + if (!revokes) + return; + + ret = gfs2_trans_begin(sdp, 0, revokes); + if (ret) + return; + __gfs2_ail_flush(gl); + gfs2_trans_end(sdp); + gfs2_log_flush(sdp, NULL); +} + /** * rgrp_go_sync - sync out the metadata for this glock * @gl: the glock @@ -227,6 +252,119 @@ static int inode_go_demote_ok(const struct gfs2_glock *gl) } /** + * gfs2_set_nlink - Set the inode's link count based on on-disk info + * @inode: The inode in question + * @nlink: The link count + * + * If the link count has hit zero, it must never be raised, whatever the + * on-disk inode might say. When new struct inodes are created the link + * count is set to 1, so that we can safely use this test even when reading + * in on disk information for the first time. + */ + +static void gfs2_set_nlink(struct inode *inode, u32 nlink) +{ + /* + * We will need to review setting the nlink count here in the + * light of the forthcoming ro bind mount work. This is a reminder + * to do that. + */ + if ((inode->i_nlink != nlink) && (inode->i_nlink != 0)) { + if (nlink == 0) + clear_nlink(inode); + else + inode->i_nlink = nlink; + } +} + +static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) +{ + const struct gfs2_dinode *str = buf; + struct timespec atime; + u16 height, depth; + + if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) + goto corrupt; + ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino); + ip->i_inode.i_mode = be32_to_cpu(str->di_mode); + ip->i_inode.i_rdev = 0; + switch (ip->i_inode.i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), + be32_to_cpu(str->di_minor)); + break; + }; + + ip->i_inode.i_uid = be32_to_cpu(str->di_uid); + ip->i_inode.i_gid = be32_to_cpu(str->di_gid); + gfs2_set_nlink(&ip->i_inode, be32_to_cpu(str->di_nlink)); + i_size_write(&ip->i_inode, be64_to_cpu(str->di_size)); + gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); + atime.tv_sec = be64_to_cpu(str->di_atime); + atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); + if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0) + ip->i_inode.i_atime = atime; + ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); + ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec); + ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); + ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec); + + ip->i_goal = be64_to_cpu(str->di_goal_meta); + ip->i_generation = be64_to_cpu(str->di_generation); + + ip->i_diskflags = be32_to_cpu(str->di_flags); + gfs2_set_inode_flags(&ip->i_inode); + height = be16_to_cpu(str->di_height); + if (unlikely(height > GFS2_MAX_META_HEIGHT)) + goto corrupt; + ip->i_height = (u8)height; + + depth = be16_to_cpu(str->di_depth); + if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) + goto corrupt; + ip->i_depth = (u8)depth; + ip->i_entries = be32_to_cpu(str->di_entries); + + ip->i_eattr = be64_to_cpu(str->di_eattr); + if (S_ISREG(ip->i_inode.i_mode)) + gfs2_set_aops(&ip->i_inode); + + return 0; +corrupt: + gfs2_consist_inode(ip); + return -EIO; +} + +/** + * gfs2_inode_refresh - Refresh the incore copy of the dinode + * @ip: The GFS2 inode + * + * Returns: errno + */ + +int gfs2_inode_refresh(struct gfs2_inode *ip) +{ + struct buffer_head *dibh; + int error; + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + return error; + + if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) { + brelse(dibh); + return -EIO; + } + + error = gfs2_dinode_in(ip, dibh->b_data); + brelse(dibh); + clear_bit(GIF_INVALID, &ip->i_flags); + + return error; +} + +/** * inode_go_lock - operation done after an inode lock is locked by a process * @gl: the glock * @flags: @@ -385,6 +523,10 @@ static int trans_go_demote_ok(const struct gfs2_glock *gl) static void iopen_go_callback(struct gfs2_glock *gl) { struct gfs2_inode *ip = (struct gfs2_inode *)gl->gl_object; + struct gfs2_sbd *sdp = gl->gl_sbd; + + if (sdp->sd_vfs->s_flags & MS_RDONLY) + return; if (gl->gl_demote_state == LM_ST_UNLOCKED && gl->gl_state == LM_ST_SHARED && ip) { diff --git a/fs/gfs2/glops.h b/fs/gfs2/glops.h index b3aa2e3210fd..6fce409b5a50 100644 --- a/fs/gfs2/glops.h +++ b/fs/gfs2/glops.h @@ -23,4 +23,6 @@ extern const struct gfs2_glock_operations gfs2_quota_glops; extern const struct gfs2_glock_operations gfs2_journal_glops; extern const struct gfs2_glock_operations *gfs2_glops_list[]; +extern void gfs2_ail_flush(struct gfs2_glock *gl); + #endif /* __GLOPS_DOT_H__ */ diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 870a89d6d4dc..0a064e91ac70 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -20,7 +20,6 @@ #define DIO_WAIT 0x00000010 #define DIO_METADATA 0x00000020 -#define DIO_ALL 0x00000100 struct gfs2_log_operations; struct gfs2_log_element; @@ -200,6 +199,8 @@ enum { GLF_INITIAL = 10, GLF_FROZEN = 11, GLF_QUEUED = 12, + GLF_LRU = 13, + GLF_OBJECT = 14, /* Used only for tracing */ }; struct gfs2_glock { @@ -234,6 +235,7 @@ struct gfs2_glock { struct list_head gl_ail_list; atomic_t gl_ail_count; + atomic_t gl_revokes; struct delayed_work gl_work; struct work_struct gl_delete; struct rcu_head gl_rcu; @@ -374,8 +376,6 @@ struct gfs2_ail { unsigned int ai_first; struct list_head ai_ail1_list; struct list_head ai_ail2_list; - - u64 ai_sync_gen; }; struct gfs2_journal_extent { @@ -488,7 +488,6 @@ struct gfs2_sb_host { char sb_lockproto[GFS2_LOCKNAME_LEN]; char sb_locktable[GFS2_LOCKNAME_LEN]; - u8 sb_uuid[16]; }; /* @@ -654,7 +653,6 @@ struct gfs2_sbd { spinlock_t sd_ail_lock; struct list_head sd_ail1_list; struct list_head sd_ail2_list; - u64 sd_ail_sync_gen; /* Replay stuff */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 97d54a28776a..03e0c529063e 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -1,23 +1,25 @@ /* * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2011 Red Hat, Inc. All rights reserved. * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions * of the GNU General Public License version 2. */ -#include <linux/sched.h> #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/completion.h> #include <linux/buffer_head.h> +#include <linux/namei.h> +#include <linux/mm.h> +#include <linux/xattr.h> #include <linux/posix_acl.h> -#include <linux/sort.h> #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> +#include <linux/fiemap.h> #include <linux/security.h> -#include <linux/time.h> +#include <asm/uaccess.h> #include "gfs2.h" #include "incore.h" @@ -26,51 +28,70 @@ #include "dir.h" #include "xattr.h" #include "glock.h" -#include "glops.h" #include "inode.h" -#include "log.h" #include "meta_io.h" #include "quota.h" #include "rgrp.h" #include "trans.h" #include "util.h" +#include "super.h" +#include "glops.h" -struct gfs2_inum_range_host { - u64 ir_start; - u64 ir_length; +struct gfs2_skip_data { + u64 no_addr; + int skipped; + int non_block; }; static int iget_test(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - u64 *no_addr = opaque; + struct gfs2_skip_data *data = opaque; - if (ip->i_no_addr == *no_addr) + if (ip->i_no_addr == data->no_addr) { + if (data->non_block && + inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) { + data->skipped = 1; + return 0; + } return 1; - + } return 0; } static int iget_set(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - u64 *no_addr = opaque; + struct gfs2_skip_data *data = opaque; - inode->i_ino = (unsigned long)*no_addr; - ip->i_no_addr = *no_addr; + if (data->skipped) + return -ENOENT; + inode->i_ino = (unsigned long)(data->no_addr); + ip->i_no_addr = data->no_addr; return 0; } -struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr) +struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int non_block) { unsigned long hash = (unsigned long)no_addr; - return ilookup5(sb, hash, iget_test, &no_addr); + struct gfs2_skip_data data; + + data.no_addr = no_addr; + data.skipped = 0; + data.non_block = non_block; + return ilookup5(sb, hash, iget_test, &data); } -static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr) +static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr, + int non_block) { + struct gfs2_skip_data data; unsigned long hash = (unsigned long)no_addr; - return iget5_locked(sb, hash, iget_test, iget_set, &no_addr); + + data.no_addr = no_addr; + data.skipped = 0; + data.non_block = non_block; + return iget5_locked(sb, hash, iget_test, iget_set, &data); } /** @@ -111,19 +132,20 @@ static void gfs2_set_iop(struct inode *inode) * @sb: The super block * @no_addr: The inode number * @type: The type of the inode + * non_block: Can we block on inodes that are being freed? * * Returns: A VFS inode, or an error */ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, - u64 no_addr, u64 no_formal_ino) + u64 no_addr, u64 no_formal_ino, int non_block) { struct inode *inode; struct gfs2_inode *ip; struct gfs2_glock *io_gl = NULL; int error; - inode = gfs2_iget(sb, no_addr); + inode = gfs2_iget(sb, no_addr, non_block); ip = GFS2_I(inode); if (!inode) @@ -185,11 +207,12 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, { struct super_block *sb = sdp->sd_vfs; struct gfs2_holder i_gh; - struct inode *inode; + struct inode *inode = NULL; int error; + /* Must not read in block until block type is verified */ error = gfs2_glock_nq_num(sdp, no_addr, &gfs2_inode_glops, - LM_ST_SHARED, LM_FLAG_ANY, &i_gh); + LM_ST_EXCLUSIVE, GL_SKIP, &i_gh); if (error) return ERR_PTR(error); @@ -197,7 +220,7 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, if (error) goto fail; - inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0); + inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0, 1); if (IS_ERR(inode)) goto fail; @@ -222,203 +245,6 @@ fail_iput: goto fail; } -static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) -{ - const struct gfs2_dinode *str = buf; - struct timespec atime; - u16 height, depth; - - if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr))) - goto corrupt; - ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino); - ip->i_inode.i_mode = be32_to_cpu(str->di_mode); - ip->i_inode.i_rdev = 0; - switch (ip->i_inode.i_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), - be32_to_cpu(str->di_minor)); - break; - }; - - ip->i_inode.i_uid = be32_to_cpu(str->di_uid); - ip->i_inode.i_gid = be32_to_cpu(str->di_gid); - /* - * We will need to review setting the nlink count here in the - * light of the forthcoming ro bind mount work. This is a reminder - * to do that. - */ - ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink); - i_size_write(&ip->i_inode, be64_to_cpu(str->di_size)); - gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks)); - atime.tv_sec = be64_to_cpu(str->di_atime); - atime.tv_nsec = be32_to_cpu(str->di_atime_nsec); - if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0) - ip->i_inode.i_atime = atime; - ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); - ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec); - ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); - ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec); - - ip->i_goal = be64_to_cpu(str->di_goal_meta); - ip->i_generation = be64_to_cpu(str->di_generation); - - ip->i_diskflags = be32_to_cpu(str->di_flags); - gfs2_set_inode_flags(&ip->i_inode); - height = be16_to_cpu(str->di_height); - if (unlikely(height > GFS2_MAX_META_HEIGHT)) - goto corrupt; - ip->i_height = (u8)height; - - depth = be16_to_cpu(str->di_depth); - if (unlikely(depth > GFS2_DIR_MAX_DEPTH)) - goto corrupt; - ip->i_depth = (u8)depth; - ip->i_entries = be32_to_cpu(str->di_entries); - - ip->i_eattr = be64_to_cpu(str->di_eattr); - if (S_ISREG(ip->i_inode.i_mode)) - gfs2_set_aops(&ip->i_inode); - - return 0; -corrupt: - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; -} - -/** - * gfs2_inode_refresh - Refresh the incore copy of the dinode - * @ip: The GFS2 inode - * - * Returns: errno - */ - -int gfs2_inode_refresh(struct gfs2_inode *ip) -{ - struct buffer_head *dibh; - int error; - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - return error; - - if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) { - brelse(dibh); - return -EIO; - } - - error = gfs2_dinode_in(ip, dibh->b_data); - brelse(dibh); - clear_bit(GIF_INVALID, &ip->i_flags); - - return error; -} - -int gfs2_dinode_dealloc(struct gfs2_inode *ip) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct gfs2_alloc *al; - struct gfs2_rgrpd *rgd; - int error; - - if (gfs2_get_inode_blocks(&ip->i_inode) != 1) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - al = gfs2_alloc_get(ip); - if (!al) - return -ENOMEM; - - error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); - if (error) - goto out; - - error = gfs2_rindex_hold(sdp, &al->al_ri_gh); - if (error) - goto out_qs; - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - if (!rgd) { - gfs2_consist_inode(ip); - error = -EIO; - goto out_rindex_relse; - } - - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, - &al->al_rgd_gh); - if (error) - goto out_rindex_relse; - - error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA, 1); - if (error) - goto out_rg_gunlock; - - set_bit(GLF_DIRTY, &ip->i_gl->gl_flags); - set_bit(GLF_LFLUSH, &ip->i_gl->gl_flags); - - gfs2_free_di(rgd, ip); - - gfs2_trans_end(sdp); - -out_rg_gunlock: - gfs2_glock_dq_uninit(&al->al_rgd_gh); -out_rindex_relse: - gfs2_glock_dq_uninit(&al->al_ri_gh); -out_qs: - gfs2_quota_unhold(ip); -out: - gfs2_alloc_put(ip); - return error; -} - -/** - * gfs2_change_nlink - Change nlink count on inode - * @ip: The GFS2 inode - * @diff: The change in the nlink count required - * - * Returns: errno - */ -int gfs2_change_nlink(struct gfs2_inode *ip, int diff) -{ - struct buffer_head *dibh; - u32 nlink; - int error; - - BUG_ON(diff != 1 && diff != -1); - nlink = ip->i_inode.i_nlink + diff; - - /* If we are reducing the nlink count, but the new value ends up being - bigger than the old one, we must have underflowed. */ - if (diff < 0 && nlink > ip->i_inode.i_nlink) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - return error; - - if (diff > 0) - inc_nlink(&ip->i_inode); - else - drop_nlink(&ip->i_inode); - - ip->i_inode.i_ctime = CURRENT_TIME; - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - mark_inode_dirty(&ip->i_inode); - - if (ip->i_inode.i_nlink == 0) - gfs2_unlink_di(&ip->i_inode); /* mark inode unlinked */ - - return error; -} struct inode *gfs2_lookup_simple(struct inode *dip, const char *name) { @@ -517,7 +343,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, /* Don't create entries in an unlinked directory */ if (!dip->i_inode.i_nlink) - return -EPERM; + return -ENOENT; error = gfs2_dir_check(&dip->i_inode, name, NULL); switch (error) { @@ -587,21 +413,44 @@ out: return error; } +static void gfs2_init_dir(struct buffer_head *dibh, + const struct gfs2_inode *parent) +{ + struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; + struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1); + + gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent); + dent->de_inum = di->di_num; /* already GFS2 endian */ + dent->de_type = cpu_to_be16(DT_DIR); + + dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1)); + gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent); + gfs2_inum_out(parent, dent); + dent->de_type = cpu_to_be16(DT_DIR); + +} + /** * init_dinode - Fill in a new dinode structure - * @dip: the directory this inode is being created in + * @dip: The directory this inode is being created in * @gl: The glock covering the new inode - * @inum: the inode number - * @mode: the file permissions - * @uid: - * @gid: + * @inum: The inode number + * @mode: The file permissions + * @uid: The uid of the new inode + * @gid: The gid of the new inode + * @generation: The generation number of the new inode + * @dev: The device number (if a device node) + * @symname: The symlink destination (if a symlink) + * @size: The inode size (ignored for directories) + * @bhp: The buffer head (returned to caller) * */ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, const struct gfs2_inum_host *inum, unsigned int mode, unsigned int uid, unsigned int gid, - const u64 *generation, dev_t dev, struct buffer_head **bhp) + const u64 *generation, dev_t dev, const char *symname, + unsigned size, struct buffer_head **bhp) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_dinode *di; @@ -620,7 +469,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_uid = cpu_to_be32(uid); di->di_gid = cpu_to_be32(gid); di->di_nlink = 0; - di->di_size = 0; + di->di_size = cpu_to_be64(size); di->di_blocks = cpu_to_be64(1); di->di_atime = di->di_mtime = di->di_ctime = cpu_to_be64(tv.tv_sec); di->di_major = cpu_to_be32(MAJOR(dev)); @@ -628,16 +477,6 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_goal_meta = di->di_goal_data = cpu_to_be64(inum->no_addr); di->di_generation = cpu_to_be64(*generation); di->di_flags = 0; - - if (S_ISREG(mode)) { - if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) || - gfs2_tune_get(sdp, gt_new_files_jdata)) - di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); - } else if (S_ISDIR(mode)) { - di->di_flags |= cpu_to_be32(dip->i_diskflags & - GFS2_DIF_INHERIT_JDATA); - } - di->__pad1 = 0; di->di_payload_format = cpu_to_be32(S_ISDIR(mode) ? GFS2_FORMAT_DE : 0); di->di_height = 0; @@ -651,7 +490,26 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_mtime_nsec = cpu_to_be32(tv.tv_nsec); di->di_ctime_nsec = cpu_to_be32(tv.tv_nsec); memset(&di->di_reserved, 0, sizeof(di->di_reserved)); - + + switch(mode & S_IFMT) { + case S_IFREG: + if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) || + gfs2_tune_get(sdp, gt_new_files_jdata)) + di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); + break; + case S_IFDIR: + di->di_flags |= cpu_to_be32(dip->i_diskflags & + GFS2_DIF_INHERIT_JDATA); + di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA); + di->di_size = cpu_to_be64(sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)); + di->di_entries = cpu_to_be32(2); + gfs2_init_dir(dibh, dip); + break; + case S_IFLNK: + memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, size); + break; + } + set_buffer_uptodate(dibh); *bhp = dibh; @@ -659,7 +517,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, unsigned int mode, const struct gfs2_inum_host *inum, - const u64 *generation, dev_t dev, struct buffer_head **bhp) + const u64 *generation, dev_t dev, const char *symname, + unsigned int size, struct buffer_head **bhp) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); unsigned int uid, gid; @@ -681,7 +540,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, if (error) goto out_quota; - init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, bhp); + init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, symname, size, bhp); gfs2_quota_change(dip, +1, uid, gid); gfs2_trans_end(sdp); @@ -735,14 +594,16 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, goto fail_quota_locks; } - error = gfs2_dir_add(&dip->i_inode, name, ip, IF2DT(ip->i_inode.i_mode)); + error = gfs2_dir_add(&dip->i_inode, name, ip); if (error) goto fail_end_trans; error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto fail_end_trans; - ip->i_inode.i_nlink = 1; + inc_nlink(&ip->i_inode); + if (S_ISDIR(ip->i_inode.i_mode)) + inc_nlink(&ip->i_inode); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -789,27 +650,25 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip, } /** - * gfs2_createi - Create a new inode - * @ghs: An array of two holders - * @name: The name of the new file - * @mode: the permissions on the new inode + * gfs2_create_inode - Create a new inode + * @dir: The parent directory + * @dentry: The new dentry + * @mode: The permissions on the new inode + * @dev: For device nodes, this is the device number + * @symname: For symlinks, this is the link destination + * @size: The initial size of the inode (ignored for directories) * - * @ghs[0] is an initialized holder for the directory - * @ghs[1] is the holder for the inode lock - * - * If the return value is not NULL, the glocks on both the directory and the new - * file are held. A transaction has been started and an inplace reservation - * is held, as well. - * - * Returns: An inode + * Returns: 0 on success, or error code */ -struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode, dev_t dev) +static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, + unsigned int mode, dev_t dev, const char *symname, + unsigned int size) { + const struct qstr *name = &dentry->d_name; + struct gfs2_holder ghs[2]; struct inode *inode = NULL; - struct gfs2_inode *dip = ghs->gh_gl->gl_object; - struct inode *dir = &dip->i_inode; + struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_inum_host inum = { .no_addr = 0, .no_formal_ino = 0 }; int error; @@ -817,10 +676,9 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, struct buffer_head *bh = NULL; if (!name->len || name->len > GFS2_FNAMESIZE) - return ERR_PTR(-ENAMETOOLONG); + return -ENAMETOOLONG; - gfs2_holder_reinit(LM_ST_EXCLUSIVE, 0, ghs); - error = gfs2_glock_nq(ghs); + error = gfs2_glock_nq_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); if (error) goto fail; @@ -838,12 +696,12 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (error) goto fail_gunlock; - error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, &bh); + error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, symname, size, &bh); if (error) goto fail_gunlock2; inode = gfs2_inode_lookup(dir->i_sb, IF2DT(mode), inum.no_addr, - inum.no_formal_ino); + inum.no_formal_ino, 0); if (IS_ERR(inode)) goto fail_gunlock2; @@ -865,18 +723,852 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (bh) brelse(bh); - return inode; + + gfs2_trans_end(sdp); + if (dip->i_alloc->al_rgd) + gfs2_inplace_release(dip); + gfs2_quota_unlock(dip); + gfs2_alloc_put(dip); + gfs2_glock_dq_uninit_m(2, ghs); + mark_inode_dirty(inode); + d_instantiate(dentry, inode); + return 0; fail_gunlock2: gfs2_glock_dq_uninit(ghs + 1); if (inode && !IS_ERR(inode)) iput(inode); fail_gunlock: - gfs2_glock_dq(ghs); + gfs2_glock_dq_uninit(ghs); fail: if (bh) brelse(bh); - return ERR_PTR(error); + return error; +} + +/** + * gfs2_create - Create a file + * @dir: The directory in which to create the file + * @dentry: The dentry of the new file + * @mode: The mode of the new file + * + * Returns: errno + */ + +static int gfs2_create(struct inode *dir, struct dentry *dentry, + int mode, struct nameidata *nd) +{ + struct inode *inode; + int ret; + + for (;;) { + ret = gfs2_create_inode(dir, dentry, S_IFREG | mode, 0, NULL, 0); + if (ret != -EEXIST || (nd && (nd->flags & LOOKUP_EXCL))) + return ret; + + inode = gfs2_lookupi(dir, &dentry->d_name, 0); + if (inode) { + if (!IS_ERR(inode)) + break; + return PTR_ERR(inode); + } + } + + d_instantiate(dentry, inode); + return 0; +} + +/** + * gfs2_lookup - Look up a filename in a directory and return its inode + * @dir: The directory inode + * @dentry: The dentry of the new inode + * @nd: passed from Linux VFS, ignored by us + * + * Called by the VFS layer. Lock dir and call gfs2_lookupi() + * + * Returns: errno + */ + +static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, + struct nameidata *nd) +{ + struct inode *inode = NULL; + + inode = gfs2_lookupi(dir, &dentry->d_name, 0); + if (inode && IS_ERR(inode)) + return ERR_CAST(inode); + + if (inode) { + struct gfs2_glock *gl = GFS2_I(inode)->i_gl; + struct gfs2_holder gh; + int error; + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) { + iput(inode); + return ERR_PTR(error); + } + gfs2_glock_dq_uninit(&gh); + return d_splice_alias(inode, dentry); + } + d_add(dentry, inode); + + return NULL; +} + +/** + * gfs2_link - Link to a file + * @old_dentry: The inode to link + * @dir: Add link to this directory + * @dentry: The name of the link + * + * Link the inode in "old_dentry" into the directory "dir" with the + * name in "dentry". + * + * Returns: errno + */ + +static int gfs2_link(struct dentry *old_dentry, struct inode *dir, + struct dentry *dentry) +{ + struct gfs2_inode *dip = GFS2_I(dir); + struct gfs2_sbd *sdp = GFS2_SB(dir); + struct inode *inode = old_dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder ghs[2]; + struct buffer_head *dibh; + int alloc_required; + int error; + + if (S_ISDIR(inode->i_mode)) + return -EPERM; + + gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); + + error = gfs2_glock_nq(ghs); /* parent */ + if (error) + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; + + error = -ENOENT; + if (inode->i_nlink == 0) + goto out_gunlock; + + error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0); + if (error) + goto out_gunlock; + + error = gfs2_dir_check(dir, &dentry->d_name, NULL); + switch (error) { + case -ENOENT: + break; + case 0: + error = -EEXIST; + default: + goto out_gunlock; + } + + error = -EINVAL; + if (!dip->i_inode.i_nlink) + goto out_gunlock; + error = -EFBIG; + if (dip->i_entries == (u32)-1) + goto out_gunlock; + error = -EPERM; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + goto out_gunlock; + error = -EINVAL; + if (!ip->i_inode.i_nlink) + goto out_gunlock; + error = -EMLINK; + if (ip->i_inode.i_nlink == (u32)-1) + goto out_gunlock; + + alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); + if (error < 0) + goto out_gunlock; + error = 0; + + if (alloc_required) { + struct gfs2_alloc *al = gfs2_alloc_get(dip); + if (!al) { + error = -ENOMEM; + goto out_gunlock; + } + + error = gfs2_quota_lock_check(dip); + if (error) + goto out_alloc; + + al->al_requested = sdp->sd_max_dirres; + + error = gfs2_inplace_reserve(dip); + if (error) + goto out_gunlock_q; + + error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + + gfs2_rg_blocks(al) + + 2 * RES_DINODE + RES_STATFS + + RES_QUOTA, 0); + if (error) + goto out_ipres; + } else { + error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0); + if (error) + goto out_ipres; + } + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + goto out_end_trans; + + error = gfs2_dir_add(dir, &dentry->d_name, ip); + if (error) + goto out_brelse; + + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + inc_nlink(&ip->i_inode); + ip->i_inode.i_ctime = CURRENT_TIME; + gfs2_dinode_out(ip, dibh->b_data); + mark_inode_dirty(&ip->i_inode); + +out_brelse: + brelse(dibh); +out_end_trans: + gfs2_trans_end(sdp); +out_ipres: + if (alloc_required) + gfs2_inplace_release(dip); +out_gunlock_q: + if (alloc_required) + gfs2_quota_unlock(dip); +out_alloc: + if (alloc_required) + gfs2_alloc_put(dip); +out_gunlock: + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_glock_dq(ghs); +out_parent: + gfs2_holder_uninit(ghs); + gfs2_holder_uninit(ghs + 1); + if (!error) { + ihold(inode); + d_instantiate(dentry, inode); + mark_inode_dirty(inode); + } + return error; +} + +/* + * gfs2_unlink_ok - check to see that a inode is still in a directory + * @dip: the directory + * @name: the name of the file + * @ip: the inode + * + * Assumes that the lock on (at least) @dip is held. + * + * Returns: 0 if the parent/child relationship is correct, errno if it isn't + */ + +static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, + const struct gfs2_inode *ip) +{ + int error; + + if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) + return -EPERM; + + if ((dip->i_inode.i_mode & S_ISVTX) && + dip->i_inode.i_uid != current_fsuid() && + ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER)) + return -EPERM; + + if (IS_APPEND(&dip->i_inode)) + return -EPERM; + + error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); + if (error) + return error; + + error = gfs2_dir_check(&dip->i_inode, name, ip); + if (error) + return error; + + return 0; +} + +/** + * gfs2_unlink_inode - Removes an inode from its parent dir and unlinks it + * @dip: The parent directory + * @name: The name of the entry in the parent directory + * @bh: The inode buffer for the inode to be removed + * @inode: The inode to be removed + * + * Called with all the locks and in a transaction. This will only be + * called for a directory after it has been checked to ensure it is empty. + * + * Returns: 0 on success, or an error + */ + +static int gfs2_unlink_inode(struct gfs2_inode *dip, + const struct dentry *dentry, + struct buffer_head *bh) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + int error; + + error = gfs2_dir_del(dip, dentry); + if (error) + return error; + + ip->i_entries = 0; + inode->i_ctime = CURRENT_TIME; + if (S_ISDIR(inode->i_mode)) + clear_nlink(inode); + else + drop_nlink(inode); + gfs2_trans_add_bh(ip->i_gl, bh, 1); + gfs2_dinode_out(ip, bh->b_data); + mark_inode_dirty(inode); + if (inode->i_nlink == 0) + gfs2_unlink_di(inode); + return 0; +} + + +/** + * gfs2_unlink - Unlink an inode (this does rmdir as well) + * @dir: The inode of the directory containing the inode to unlink + * @dentry: The file itself + * + * This routine uses the type of the inode as a flag to figure out + * whether this is an unlink or an rmdir. + * + * Returns: errno + */ + +static int gfs2_unlink(struct inode *dir, struct dentry *dentry) +{ + struct gfs2_inode *dip = GFS2_I(dir); + struct gfs2_sbd *sdp = GFS2_SB(dir); + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct buffer_head *bh; + struct gfs2_holder ghs[3]; + struct gfs2_rgrpd *rgd; + struct gfs2_holder ri_gh; + int error; + + error = gfs2_rindex_hold(sdp, &ri_gh); + if (error) + return error; + + gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); + + rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); + gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); + + + error = gfs2_glock_nq(ghs); /* parent */ + if (error) + goto out_parent; + + error = gfs2_glock_nq(ghs + 1); /* child */ + if (error) + goto out_child; + + error = -ENOENT; + if (inode->i_nlink == 0) + goto out_rgrp; + + if (S_ISDIR(inode->i_mode)) { + error = -ENOTEMPTY; + if (ip->i_entries > 2 || inode->i_nlink > 2) + goto out_rgrp; + } + + error = gfs2_glock_nq(ghs + 2); /* rgrp */ + if (error) + goto out_rgrp; + + error = gfs2_unlink_ok(dip, &dentry->d_name, ip); + if (error) + goto out_gunlock; + + error = gfs2_trans_begin(sdp, 2*RES_DINODE + 3*RES_LEAF + RES_RG_BIT, 0); + if (error) + goto out_gunlock; + + error = gfs2_meta_inode_buffer(ip, &bh); + if (error) + goto out_end_trans; + + error = gfs2_unlink_inode(dip, dentry, bh); + brelse(bh); + +out_end_trans: + gfs2_trans_end(sdp); +out_gunlock: + gfs2_glock_dq(ghs + 2); +out_rgrp: + gfs2_holder_uninit(ghs + 2); + gfs2_glock_dq(ghs + 1); +out_child: + gfs2_holder_uninit(ghs + 1); + gfs2_glock_dq(ghs); +out_parent: + gfs2_holder_uninit(ghs); + gfs2_glock_dq_uninit(&ri_gh); + return error; +} + +/** + * gfs2_symlink - Create a symlink + * @dir: The directory to create the symlink in + * @dentry: The dentry to put the symlink in + * @symname: The thing which the link points to + * + * Returns: errno + */ + +static int gfs2_symlink(struct inode *dir, struct dentry *dentry, + const char *symname) +{ + struct gfs2_sbd *sdp = GFS2_SB(dir); + unsigned int size; + + size = strlen(symname); + if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1) + return -ENAMETOOLONG; + + return gfs2_create_inode(dir, dentry, S_IFLNK | S_IRWXUGO, 0, symname, size); +} + +/** + * gfs2_mkdir - Make a directory + * @dir: The parent directory of the new one + * @dentry: The dentry of the new directory + * @mode: The mode of the new directory + * + * Returns: errno + */ + +static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) +{ + return gfs2_create_inode(dir, dentry, S_IFDIR | mode, 0, NULL, 0); +} + +/** + * gfs2_mknod - Make a special file + * @dir: The directory in which the special file will reside + * @dentry: The dentry of the special file + * @mode: The mode of the special file + * @dev: The device specification of the special file + * + */ + +static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, + dev_t dev) +{ + return gfs2_create_inode(dir, dentry, mode, dev, NULL, 0); +} + +/* + * gfs2_ok_to_move - check if it's ok to move a directory to another directory + * @this: move this + * @to: to here + * + * Follow @to back to the root and make sure we don't encounter @this + * Assumes we already hold the rename lock. + * + * Returns: errno + */ + +static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) +{ + struct inode *dir = &to->i_inode; + struct super_block *sb = dir->i_sb; + struct inode *tmp; + int error = 0; + + igrab(dir); + + for (;;) { + if (dir == &this->i_inode) { + error = -EINVAL; + break; + } + if (dir == sb->s_root->d_inode) { + error = 0; + break; + } + + tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1); + if (IS_ERR(tmp)) { + error = PTR_ERR(tmp); + break; + } + + iput(dir); + dir = tmp; + } + + iput(dir); + + return error; +} + +/** + * gfs2_rename - Rename a file + * @odir: Parent directory of old file name + * @odentry: The old dentry of the file + * @ndir: Parent directory of new file name + * @ndentry: The new dentry of the file + * + * Returns: errno + */ + +static int gfs2_rename(struct inode *odir, struct dentry *odentry, + struct inode *ndir, struct dentry *ndentry) +{ + struct gfs2_inode *odip = GFS2_I(odir); + struct gfs2_inode *ndip = GFS2_I(ndir); + struct gfs2_inode *ip = GFS2_I(odentry->d_inode); + struct gfs2_inode *nip = NULL; + struct gfs2_sbd *sdp = GFS2_SB(odir); + struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh; + struct gfs2_rgrpd *nrgd; + unsigned int num_gh; + int dir_rename = 0; + int alloc_required = 0; + unsigned int x; + int error; + + if (ndentry->d_inode) { + nip = GFS2_I(ndentry->d_inode); + if (ip == nip) + return 0; + } + + error = gfs2_rindex_hold(sdp, &ri_gh); + if (error) + return error; + + if (odip != ndip) { + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, + 0, &r_gh); + if (error) + goto out; + + if (S_ISDIR(ip->i_inode.i_mode)) { + dir_rename = 1; + /* don't move a dirctory into it's subdir */ + error = gfs2_ok_to_move(ip, ndip); + if (error) + goto out_gunlock_r; + } + } + + num_gh = 1; + gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); + if (odip != ndip) { + gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + } + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + + if (nip) { + gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); + num_gh++; + /* grab the resource lock for unlink flag twiddling + * this is the case of the target file already existing + * so we unlink before doing the rename + */ + nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr); + if (nrgd) + gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); + } + + for (x = 0; x < num_gh; x++) { + error = gfs2_glock_nq(ghs + x); + if (error) + goto out_gunlock; + } + + error = -ENOENT; + if (ip->i_inode.i_nlink == 0) + goto out_gunlock; + + /* Check out the old directory */ + + error = gfs2_unlink_ok(odip, &odentry->d_name, ip); + if (error) + goto out_gunlock; + + /* Check out the new directory */ + + if (nip) { + error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip); + if (error) + goto out_gunlock; + + if (nip->i_inode.i_nlink == 0) { + error = -EAGAIN; + goto out_gunlock; + } + + if (S_ISDIR(nip->i_inode.i_mode)) { + if (nip->i_entries < 2) { + gfs2_consist_inode(nip); + error = -EIO; + goto out_gunlock; + } + if (nip->i_entries > 2) { + error = -ENOTEMPTY; + goto out_gunlock; + } + } + } else { + error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0); + if (error) + goto out_gunlock; + + error = gfs2_dir_check(ndir, &ndentry->d_name, NULL); + switch (error) { + case -ENOENT: + error = 0; + break; + case 0: + error = -EEXIST; + default: + goto out_gunlock; + }; + + if (odip != ndip) { + if (!ndip->i_inode.i_nlink) { + error = -ENOENT; + goto out_gunlock; + } + if (ndip->i_entries == (u32)-1) { + error = -EFBIG; + goto out_gunlock; + } + if (S_ISDIR(ip->i_inode.i_mode) && + ndip->i_inode.i_nlink == (u32)-1) { + error = -EMLINK; + goto out_gunlock; + } + } + } + + /* Check out the dir to be renamed */ + + if (dir_rename) { + error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0); + if (error) + goto out_gunlock; + } + + if (nip == NULL) + alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name); + error = alloc_required; + if (error < 0) + goto out_gunlock; + error = 0; + + if (alloc_required) { + struct gfs2_alloc *al = gfs2_alloc_get(ndip); + if (!al) { + error = -ENOMEM; + goto out_gunlock; + } + + error = gfs2_quota_lock_check(ndip); + if (error) + goto out_alloc; + + al->al_requested = sdp->sd_max_dirres; + + error = gfs2_inplace_reserve_ri(ndip); + if (error) + goto out_gunlock_q; + + error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + + gfs2_rg_blocks(al) + + 4 * RES_DINODE + 4 * RES_LEAF + + RES_STATFS + RES_QUOTA + 4, 0); + if (error) + goto out_ipreserv; + } else { + error = gfs2_trans_begin(sdp, 4 * RES_DINODE + + 5 * RES_LEAF + 4, 0); + if (error) + goto out_gunlock; + } + + /* Remove the target file, if it exists */ + + if (nip) { + struct buffer_head *bh; + error = gfs2_meta_inode_buffer(nip, &bh); + if (error) + goto out_end_trans; + error = gfs2_unlink_inode(ndip, ndentry, bh); + brelse(bh); + } + + if (dir_rename) { + error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR); + if (error) + goto out_end_trans; + } else { + struct buffer_head *dibh; + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) + goto out_end_trans; + ip->i_inode.i_ctime = CURRENT_TIME; + gfs2_trans_add_bh(ip->i_gl, dibh, 1); + gfs2_dinode_out(ip, dibh->b_data); + brelse(dibh); + } + + error = gfs2_dir_del(odip, odentry); + if (error) + goto out_end_trans; + + error = gfs2_dir_add(ndir, &ndentry->d_name, ip); + if (error) + goto out_end_trans; + +out_end_trans: + gfs2_trans_end(sdp); +out_ipreserv: + if (alloc_required) + gfs2_inplace_release(ndip); +out_gunlock_q: + if (alloc_required) + gfs2_quota_unlock(ndip); +out_alloc: + if (alloc_required) + gfs2_alloc_put(ndip); +out_gunlock: + while (x--) { + gfs2_glock_dq(ghs + x); + gfs2_holder_uninit(ghs + x); + } +out_gunlock_r: + if (r_gh.gh_gl) + gfs2_glock_dq_uninit(&r_gh); +out: + gfs2_glock_dq_uninit(&ri_gh); + return error; +} + +/** + * gfs2_follow_link - Follow a symbolic link + * @dentry: The dentry of the link + * @nd: Data that we pass to vfs_follow_link() + * + * This can handle symlinks of any size. + * + * Returns: 0 on success or error code + */ + +static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + struct gfs2_inode *ip = GFS2_I(dentry->d_inode); + struct gfs2_holder i_gh; + struct buffer_head *dibh; + unsigned int size; + char *buf; + int error; + + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); + error = gfs2_glock_nq(&i_gh); + if (error) { + gfs2_holder_uninit(&i_gh); + nd_set_link(nd, ERR_PTR(error)); + return NULL; + } + + size = (unsigned int)i_size_read(&ip->i_inode); + if (size == 0) { + gfs2_consist_inode(ip); + buf = ERR_PTR(-EIO); + goto out; + } + + error = gfs2_meta_inode_buffer(ip, &dibh); + if (error) { + buf = ERR_PTR(error); + goto out; + } + + buf = kzalloc(size + 1, GFP_NOFS); + if (!buf) + buf = ERR_PTR(-ENOMEM); + else + memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), size); + brelse(dibh); +out: + gfs2_glock_dq_uninit(&i_gh); + nd_set_link(nd, buf); + return NULL; +} + +static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p) +{ + char *s = nd_get_link(nd); + if (!IS_ERR(s)) + kfree(s); +} + +/** + * gfs2_permission - + * @inode: The inode + * @mask: The mask to be tested + * @flags: Indicates whether this is an RCU path walk or not + * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. + * + * Returns: errno + */ + +int gfs2_permission(struct inode *inode, int mask, unsigned int flags) +{ + struct gfs2_inode *ip; + struct gfs2_holder i_gh; + int error; + int unlock = 0; + + + ip = GFS2_I(inode); + if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); + if (error) + return error; + unlock = 1; + } + + if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) + error = -EACCES; + else + error = generic_permission(inode, mask, flags, gfs2_check_acl); + if (unlock) + gfs2_glock_dq_uninit(&i_gh); + + return error; } static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) @@ -902,8 +1594,6 @@ static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) * @ip: * @attr: * - * Called with a reference on the vnode. - * * Returns: errno */ @@ -923,60 +1613,280 @@ int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) return error; } -void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) -{ - struct gfs2_dinode *str = buf; - - str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); - str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); - str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); - str->di_num.no_addr = cpu_to_be64(ip->i_no_addr); - str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino); - str->di_mode = cpu_to_be32(ip->i_inode.i_mode); - str->di_uid = cpu_to_be32(ip->i_inode.i_uid); - str->di_gid = cpu_to_be32(ip->i_inode.i_gid); - str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); - str->di_size = cpu_to_be64(i_size_read(&ip->i_inode)); - str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode)); - str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); - str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); - str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); - - str->di_goal_meta = cpu_to_be64(ip->i_goal); - str->di_goal_data = cpu_to_be64(ip->i_goal); - str->di_generation = cpu_to_be64(ip->i_generation); - - str->di_flags = cpu_to_be32(ip->i_diskflags); - str->di_height = cpu_to_be16(ip->i_height); - str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && - !(ip->i_diskflags & GFS2_DIF_EXHASH) ? - GFS2_FORMAT_DE : 0); - str->di_depth = cpu_to_be16(ip->i_depth); - str->di_entries = cpu_to_be32(ip->i_entries); - - str->di_eattr = cpu_to_be64(ip->i_eattr); - str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); - str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec); - str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec); -} - -void gfs2_dinode_print(const struct gfs2_inode *ip) -{ - printk(KERN_INFO " no_formal_ino = %llu\n", - (unsigned long long)ip->i_no_formal_ino); - printk(KERN_INFO " no_addr = %llu\n", - (unsigned long long)ip->i_no_addr); - printk(KERN_INFO " i_size = %llu\n", - (unsigned long long)i_size_read(&ip->i_inode)); - printk(KERN_INFO " blocks = %llu\n", - (unsigned long long)gfs2_get_inode_blocks(&ip->i_inode)); - printk(KERN_INFO " i_goal = %llu\n", - (unsigned long long)ip->i_goal); - printk(KERN_INFO " i_diskflags = 0x%.8X\n", ip->i_diskflags); - printk(KERN_INFO " i_height = %u\n", ip->i_height); - printk(KERN_INFO " i_depth = %u\n", ip->i_depth); - printk(KERN_INFO " i_entries = %u\n", ip->i_entries); - printk(KERN_INFO " i_eattr = %llu\n", - (unsigned long long)ip->i_eattr); +static int setattr_chown(struct inode *inode, struct iattr *attr) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); + u32 ouid, ogid, nuid, ngid; + int error; + + ouid = inode->i_uid; + ogid = inode->i_gid; + nuid = attr->ia_uid; + ngid = attr->ia_gid; + + if (!(attr->ia_valid & ATTR_UID) || ouid == nuid) + ouid = nuid = NO_QUOTA_CHANGE; + if (!(attr->ia_valid & ATTR_GID) || ogid == ngid) + ogid = ngid = NO_QUOTA_CHANGE; + + if (!gfs2_alloc_get(ip)) + return -ENOMEM; + + error = gfs2_quota_lock(ip, nuid, ngid); + if (error) + goto out_alloc; + + if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { + error = gfs2_quota_check(ip, nuid, ngid); + if (error) + goto out_gunlock_q; + } + + error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0); + if (error) + goto out_gunlock_q; + + error = gfs2_setattr_simple(ip, attr); + if (error) + goto out_end_trans; + + if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { + u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); + gfs2_quota_change(ip, -blocks, ouid, ogid); + gfs2_quota_change(ip, blocks, nuid, ngid); + } + +out_end_trans: + gfs2_trans_end(sdp); +out_gunlock_q: + gfs2_quota_unlock(ip); +out_alloc: + gfs2_alloc_put(ip); + return error; } +/** + * gfs2_setattr - Change attributes on an inode + * @dentry: The dentry which is changing + * @attr: The structure describing the change + * + * The VFS layer wants to change one or more of an inodes attributes. Write + * that change out to disk. + * + * Returns: errno + */ + +static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder i_gh; + int error; + + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); + if (error) + return error; + + error = -EPERM; + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + goto out; + + error = inode_change_ok(inode, attr); + if (error) + goto out; + + if (attr->ia_valid & ATTR_SIZE) + error = gfs2_setattr_size(inode, attr->ia_size); + else if (attr->ia_valid & (ATTR_UID | ATTR_GID)) + error = setattr_chown(inode, attr); + else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode)) + error = gfs2_acl_chmod(ip, attr); + else + error = gfs2_setattr_simple(ip, attr); + +out: + gfs2_glock_dq_uninit(&i_gh); + if (!error) + mark_inode_dirty(inode); + return error; +} + +/** + * gfs2_getattr - Read out an inode's attributes + * @mnt: The vfsmount the inode is being accessed from + * @dentry: The dentry to stat + * @stat: The inode's stats + * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. Note that its the NFS + * readdirplus operation which causes this to be called (from filldir) + * with the glock already held. + * + * Returns: errno + */ + +static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int error; + int unlock = 0; + + if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) + return error; + unlock = 1; + } + + generic_fillattr(inode, stat); + if (unlock) + gfs2_glock_dq_uninit(&gh); + + return 0; +} + +static int gfs2_setxattr(struct dentry *dentry, const char *name, + const void *data, size_t size, int flags) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_setxattr(dentry, name, data, size, flags); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name, + void *data, size_t size) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_getxattr(dentry, name, data, size); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static int gfs2_removexattr(struct dentry *dentry, const char *name) +{ + struct inode *inode = dentry->d_inode; + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq(&gh); + if (ret == 0) { + ret = generic_removexattr(dentry, name); + gfs2_glock_dq(&gh); + } + gfs2_holder_uninit(&gh); + return ret; +} + +static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, + u64 start, u64 len) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder gh; + int ret; + + ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC); + if (ret) + return ret; + + mutex_lock(&inode->i_mutex); + + ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh); + if (ret) + goto out; + + if (gfs2_is_stuffed(ip)) { + u64 phys = ip->i_no_addr << inode->i_blkbits; + u64 size = i_size_read(inode); + u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED| + FIEMAP_EXTENT_DATA_INLINE; + phys += sizeof(struct gfs2_dinode); + phys += start; + if (start + len > size) + len = size - start; + if (start < size) + ret = fiemap_fill_next_extent(fieinfo, start, phys, + len, flags); + if (ret == 1) + ret = 0; + } else { + ret = __generic_block_fiemap(inode, fieinfo, start, len, + gfs2_block_map); + } + + gfs2_glock_dq_uninit(&gh); +out: + mutex_unlock(&inode->i_mutex); + return ret; +} + +const struct inode_operations gfs2_file_iops = { + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + +const struct inode_operations gfs2_dir_iops = { + .create = gfs2_create, + .lookup = gfs2_lookup, + .link = gfs2_link, + .unlink = gfs2_unlink, + .symlink = gfs2_symlink, + .mkdir = gfs2_mkdir, + .rmdir = gfs2_unlink, + .mknod = gfs2_mknod, + .rename = gfs2_rename, + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + +const struct inode_operations gfs2_symlink_iops = { + .readlink = generic_readlink, + .follow_link = gfs2_follow_link, + .put_link = gfs2_put_link, + .permission = gfs2_permission, + .setattr = gfs2_setattr, + .getattr = gfs2_getattr, + .setxattr = gfs2_setxattr, + .getxattr = gfs2_getxattr, + .listxattr = gfs2_listxattr, + .removexattr = gfs2_removexattr, + .fiemap = gfs2_fiemap, +}; + diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 3e00a66e7cbd..31606076f701 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -97,26 +97,21 @@ err: } extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, - u64 no_addr, u64 no_formal_ino); + u64 no_addr, u64 no_formal_ino, + int non_block); extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, u64 *no_formal_ino, unsigned int blktype); -extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr); +extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int nonblock); extern int gfs2_inode_refresh(struct gfs2_inode *ip); -extern int gfs2_dinode_dealloc(struct gfs2_inode *inode); -extern int gfs2_change_nlink(struct gfs2_inode *ip, int diff); extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, int is_root); -extern struct inode *gfs2_createi(struct gfs2_holder *ghs, - const struct qstr *name, - unsigned int mode, dev_t dev); extern int gfs2_permission(struct inode *inode, int mask, unsigned int flags); extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); -extern void gfs2_dinode_print(const struct gfs2_inode *ip); extern const struct inode_operations gfs2_file_iops; extern const struct inode_operations gfs2_dir_iops; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 5b102c1887fd..cec26c00b50d 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -18,6 +18,7 @@ #include <linux/kthread.h> #include <linux/freezer.h> #include <linux/bio.h> +#include <linux/writeback.h> #include "gfs2.h" #include "incore.h" @@ -83,55 +84,97 @@ void gfs2_remove_from_ail(struct gfs2_bufdata *bd) /** * gfs2_ail1_start_one - Start I/O on a part of the AIL * @sdp: the filesystem - * @tr: the part of the AIL + * @wbc: The writeback control structure + * @ai: The ail structure * */ -static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) +static int gfs2_ail1_start_one(struct gfs2_sbd *sdp, + struct writeback_control *wbc, + struct gfs2_ail *ai) __releases(&sdp->sd_ail_lock) __acquires(&sdp->sd_ail_lock) { + struct gfs2_glock *gl = NULL; + struct address_space *mapping; struct gfs2_bufdata *bd, *s; struct buffer_head *bh; - int retry; - do { - retry = 0; + list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, bd_ail_st_list) { + bh = bd->bd_bh; - list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, - bd_ail_st_list) { - bh = bd->bd_bh; + gfs2_assert(sdp, bd->bd_ail == ai); - gfs2_assert(sdp, bd->bd_ail == ai); + if (!buffer_busy(bh)) { + if (!buffer_uptodate(bh)) + gfs2_io_error_bh(sdp, bh); + list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); + continue; + } - if (!buffer_busy(bh)) { - if (!buffer_uptodate(bh)) - gfs2_io_error_bh(sdp, bh); - list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); - continue; - } + if (!buffer_dirty(bh)) + continue; + if (gl == bd->bd_gl) + continue; + gl = bd->bd_gl; + list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list); + mapping = bh->b_page->mapping; + if (!mapping) + continue; + spin_unlock(&sdp->sd_ail_lock); + generic_writepages(mapping, wbc); + spin_lock(&sdp->sd_ail_lock); + if (wbc->nr_to_write <= 0) + break; + return 1; + } - if (!buffer_dirty(bh)) - continue; + return 0; +} - list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list); - get_bh(bh); - spin_unlock(&sdp->sd_ail_lock); - lock_buffer(bh); - if (test_clear_buffer_dirty(bh)) { - bh->b_end_io = end_buffer_write_sync; - submit_bh(WRITE_SYNC, bh); - } else { - unlock_buffer(bh); - brelse(bh); - } - spin_lock(&sdp->sd_ail_lock); - - retry = 1; +/** + * gfs2_ail1_flush - start writeback of some ail1 entries + * @sdp: The super block + * @wbc: The writeback control structure + * + * Writes back some ail1 entries, according to the limits in the + * writeback control structure + */ + +void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc) +{ + struct list_head *head = &sdp->sd_ail1_list; + struct gfs2_ail *ai; + + trace_gfs2_ail_flush(sdp, wbc, 1); + spin_lock(&sdp->sd_ail_lock); +restart: + list_for_each_entry_reverse(ai, head, ai_list) { + if (wbc->nr_to_write <= 0) break; - } - } while (retry); + if (gfs2_ail1_start_one(sdp, wbc, ai)) + goto restart; + } + spin_unlock(&sdp->sd_ail_lock); + trace_gfs2_ail_flush(sdp, wbc, 0); +} + +/** + * gfs2_ail1_start - start writeback of all ail1 entries + * @sdp: The superblock + */ + +static void gfs2_ail1_start(struct gfs2_sbd *sdp) +{ + struct writeback_control wbc = { + .sync_mode = WB_SYNC_NONE, + .nr_to_write = LONG_MAX, + .range_start = 0, + .range_end = LLONG_MAX, + }; + + return gfs2_ail1_flush(sdp, &wbc); } /** @@ -141,7 +184,7 @@ __acquires(&sdp->sd_ail_lock) * */ -static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int flags) +static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai) { struct gfs2_bufdata *bd, *s; struct buffer_head *bh; @@ -149,71 +192,37 @@ static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int fl list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, bd_ail_st_list) { bh = bd->bd_bh; - gfs2_assert(sdp, bd->bd_ail == ai); - - if (buffer_busy(bh)) { - if (flags & DIO_ALL) - continue; - else - break; - } - + if (buffer_busy(bh)) + continue; if (!buffer_uptodate(bh)) gfs2_io_error_bh(sdp, bh); - list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list); } - return list_empty(&ai->ai_ail1_list); } -static void gfs2_ail1_start(struct gfs2_sbd *sdp) -{ - struct list_head *head; - u64 sync_gen; - struct gfs2_ail *ai; - int done = 0; - - spin_lock(&sdp->sd_ail_lock); - head = &sdp->sd_ail1_list; - if (list_empty(head)) { - spin_unlock(&sdp->sd_ail_lock); - return; - } - sync_gen = sdp->sd_ail_sync_gen++; - - while(!done) { - done = 1; - list_for_each_entry_reverse(ai, head, ai_list) { - if (ai->ai_sync_gen >= sync_gen) - continue; - ai->ai_sync_gen = sync_gen; - gfs2_ail1_start_one(sdp, ai); /* This may drop ail lock */ - done = 0; - break; - } - } - - spin_unlock(&sdp->sd_ail_lock); -} +/** + * gfs2_ail1_empty - Try to empty the ail1 lists + * @sdp: The superblock + * + * Tries to empty the ail1 lists, starting with the oldest first + */ -static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags) +static int gfs2_ail1_empty(struct gfs2_sbd *sdp) { struct gfs2_ail *ai, *s; int ret; spin_lock(&sdp->sd_ail_lock); - list_for_each_entry_safe_reverse(ai, s, &sdp->sd_ail1_list, ai_list) { - if (gfs2_ail1_empty_one(sdp, ai, flags)) + gfs2_ail1_empty_one(sdp, ai); + if (list_empty(&ai->ai_ail1_list)) list_move(&ai->ai_list, &sdp->sd_ail2_list); - else if (!(flags & DIO_ALL)) + else break; } - ret = list_empty(&sdp->sd_ail1_list); - spin_unlock(&sdp->sd_ail_lock); return ret; @@ -574,7 +583,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull) set_buffer_uptodate(bh); clear_buffer_dirty(bh); - gfs2_ail1_empty(sdp, 0); + gfs2_ail1_empty(sdp); tail = current_tail(sdp); lh = (struct gfs2_log_header *)bh->b_data; @@ -869,7 +878,7 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp) gfs2_log_flush(sdp, NULL); for (;;) { gfs2_ail1_start(sdp); - if (gfs2_ail1_empty(sdp, DIO_ALL)) + if (gfs2_ail1_empty(sdp)) break; msleep(10); } @@ -905,17 +914,15 @@ int gfs2_logd(void *data) preflush = atomic_read(&sdp->sd_log_pinned); if (gfs2_jrnl_flush_reqd(sdp) || t == 0) { - gfs2_ail1_empty(sdp, DIO_ALL); + gfs2_ail1_empty(sdp); gfs2_log_flush(sdp, NULL); - gfs2_ail1_empty(sdp, DIO_ALL); } if (gfs2_ail_flush_reqd(sdp)) { gfs2_ail1_start(sdp); io_schedule(); - gfs2_ail1_empty(sdp, 0); + gfs2_ail1_empty(sdp); gfs2_log_flush(sdp, NULL); - gfs2_ail1_empty(sdp, DIO_ALL); } wake_up(&sdp->sd_log_waitq); diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 0d007f920234..ab0621698b73 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -12,6 +12,7 @@ #include <linux/list.h> #include <linux/spinlock.h> +#include <linux/writeback.h> #include "incore.h" /** @@ -59,6 +60,7 @@ extern struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp, extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl); extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans); extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd); +extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc); extern void gfs2_log_shutdown(struct gfs2_sbd *sdp); extern void gfs2_meta_syncfs(struct gfs2_sbd *sdp); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 51d27f00ebb4..05bbb124699f 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -40,7 +40,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh) { struct gfs2_bufdata *bd; - gfs2_assert_withdraw(sdp, test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)); + BUG_ON(!current->journal_info); clear_buffer_dirty(bh); if (test_set_buffer_pinned(bh)) @@ -65,6 +65,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh) * @sdp: the filesystem the buffer belongs to * @bh: The buffer to unpin * @ai: + * @flags: The inode dirty flags * */ @@ -73,10 +74,8 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, { struct gfs2_bufdata *bd = bh->b_private; - gfs2_assert_withdraw(sdp, buffer_uptodate(bh)); - - if (!buffer_pinned(bh)) - gfs2_assert_withdraw(sdp, 0); + BUG_ON(!buffer_uptodate(bh)); + BUG_ON(!buffer_pinned(bh)); lock_buffer(bh); mark_buffer_dirty(bh); @@ -95,8 +94,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list); spin_unlock(&sdp->sd_ail_lock); - if (test_and_clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags)) - gfs2_glock_schedule_for_reclaim(bd->bd_gl); + clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags); trace_gfs2_pin(bd, 0); unlock_buffer(bh); atomic_dec(&sdp->sd_log_pinned); @@ -322,12 +320,16 @@ static void buf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass) static void revoke_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le) { + struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le); + struct gfs2_glock *gl = bd->bd_gl; struct gfs2_trans *tr; tr = current->journal_info; tr->tr_touched = 1; tr->tr_num_revoke++; sdp->sd_log_num_revoke++; + atomic_inc(&gl->gl_revokes); + set_bit(GLF_LFLUSH, &gl->gl_flags); list_add(&le->le_list, &sdp->sd_log_le_revoke); } @@ -350,9 +352,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) ld->ld_data1 = cpu_to_be32(sdp->sd_log_num_revoke); offset = sizeof(struct gfs2_log_descriptor); - while (!list_empty(head)) { - bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list); - list_del_init(&bd->bd_le.le_list); + list_for_each_entry(bd, head, bd_le.le_list) { sdp->sd_log_num_revoke--; if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) { @@ -367,8 +367,6 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) } *(__be64 *)(bh->b_data + offset) = cpu_to_be64(bd->bd_blkno); - kmem_cache_free(gfs2_bufdata_cachep, bd); - offset += sizeof(u64); } gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); @@ -376,6 +374,22 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) submit_bh(WRITE_SYNC, bh); } +static void revoke_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) +{ + struct list_head *head = &sdp->sd_log_le_revoke; + struct gfs2_bufdata *bd; + struct gfs2_glock *gl; + + while (!list_empty(head)) { + bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list); + list_del_init(&bd->bd_le.le_list); + gl = bd->bd_gl; + atomic_dec(&gl->gl_revokes); + clear_bit(GLF_LFLUSH, &gl->gl_flags); + kmem_cache_free(gfs2_bufdata_cachep, bd); + } +} + static void revoke_lo_before_scan(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head, int pass) { @@ -749,6 +763,7 @@ const struct gfs2_log_operations gfs2_buf_lops = { const struct gfs2_log_operations gfs2_revoke_lops = { .lo_add = revoke_lo_add, .lo_before_commit = revoke_lo_before_commit, + .lo_after_commit = revoke_lo_after_commit, .lo_before_scan = revoke_lo_before_scan, .lo_scan_elements = revoke_lo_scan_elements, .lo_after_scan = revoke_lo_after_scan, diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 888a5f5a1a58..cfa327d33194 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -53,6 +53,7 @@ static void gfs2_init_glock_once(void *foo) INIT_LIST_HEAD(&gl->gl_lru); INIT_LIST_HEAD(&gl->gl_ail_list); atomic_set(&gl->gl_ail_count, 0); + atomic_set(&gl->gl_revokes, 0); } static void gfs2_init_gl_aspace_once(void *foo) diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 675349b5a133..747238cd9f96 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -31,6 +31,7 @@ #include "rgrp.h" #include "trans.h" #include "util.h" +#include "trace_gfs2.h" static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc) { @@ -310,6 +311,7 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, int struct gfs2_bufdata *bd = bh->b_private; if (test_clear_buffer_pinned(bh)) { + trace_gfs2_pin(bd, 0); atomic_dec(&sdp->sd_log_pinned); list_del_init(&bd->bd_le.le_list); if (meta) { diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h index 6a1d9ba16411..22c526593131 100644 --- a/fs/gfs2/meta_io.h +++ b/fs/gfs2/meta_io.h @@ -77,8 +77,6 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen); #define buffer_busy(bh) \ ((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock) | (1ul << BH_Pinned))) -#define buffer_in_io(bh) \ -((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock))) #endif /* __DIO_DOT_H__ */ diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 42ef24355afb..8ac9ae189b53 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -126,8 +126,10 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) * changed. */ -static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) +static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent) { + struct gfs2_sb_host *sb = &sdp->sd_sb; + if (sb->sb_magic != GFS2_MAGIC || sb->sb_type != GFS2_METATYPE_SB) { if (!silent) @@ -157,8 +159,10 @@ static void end_bio_io_page(struct bio *bio, int error) unlock_page(page); } -static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) +static void gfs2_sb_in(struct gfs2_sbd *sdp, const void *buf) { + struct gfs2_sb_host *sb = &sdp->sd_sb; + struct super_block *s = sdp->sd_vfs; const struct gfs2_sb *str = buf; sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic); @@ -175,7 +179,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN); memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); - memcpy(sb->sb_uuid, str->sb_uuid, 16); + memcpy(s->s_uuid, str->sb_uuid, 16); } /** @@ -197,7 +201,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) * Returns: 0 on success or error */ -static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) +static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent) { struct super_block *sb = sdp->sd_vfs; struct gfs2_sb *p; @@ -227,10 +231,10 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector) return -EIO; } p = kmap(page); - gfs2_sb_in(&sdp->sd_sb, p); + gfs2_sb_in(sdp, p); kunmap(page); __free_page(page); - return 0; + return gfs2_check_sb(sdp, silent); } /** @@ -247,17 +251,13 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent) unsigned int x; int error; - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent); if (error) { if (!silent) fs_err(sdp, "can't read superblock\n"); return error; } - error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); - if (error) - return error; - sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT; sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift; @@ -340,14 +340,10 @@ static int init_names(struct gfs2_sbd *sdp, int silent) /* Try to autodetect */ if (!proto[0] || !table[0]) { - error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift); + error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent); if (error) return error; - error = gfs2_check_sb(sdp, &sdp->sd_sb, silent); - if (error) - goto out; - if (!proto[0]) proto = sdp->sd_sb.sb_lockproto; if (!table[0]) @@ -364,7 +360,6 @@ static int init_names(struct gfs2_sbd *sdp, int silent) while ((table = strchr(table, '/'))) *table = '_'; -out: return error; } @@ -430,7 +425,7 @@ static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr, struct dentry *dentry; struct inode *inode; - inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0); + inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0); if (IS_ERR(inode)) { fs_err(sdp, "can't read in %s inode: %ld\n", name, PTR_ERR(inode)); return PTR_ERR(inode); @@ -1119,8 +1114,7 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent if (sdp->sd_args.ar_statfs_quantum) { sdp->sd_tune.gt_statfs_slow = 0; sdp->sd_tune.gt_statfs_quantum = sdp->sd_args.ar_statfs_quantum; - } - else { + } else { sdp->sd_tune.gt_statfs_slow = 1; sdp->sd_tune.gt_statfs_quantum = 30; } diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c deleted file mode 100644 index 09e436a50723..000000000000 --- a/fs/gfs2/ops_inode.c +++ /dev/null @@ -1,1344 +0,0 @@ -/* - * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. - * - * This copyrighted material is made available to anyone wishing to use, - * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License version 2. - */ - -#include <linux/slab.h> -#include <linux/spinlock.h> -#include <linux/completion.h> -#include <linux/buffer_head.h> -#include <linux/namei.h> -#include <linux/mm.h> -#include <linux/xattr.h> -#include <linux/posix_acl.h> -#include <linux/gfs2_ondisk.h> -#include <linux/crc32.h> -#include <linux/fiemap.h> -#include <asm/uaccess.h> - -#include "gfs2.h" -#include "incore.h" -#include "acl.h" -#include "bmap.h" -#include "dir.h" -#include "xattr.h" -#include "glock.h" -#include "inode.h" -#include "meta_io.h" -#include "quota.h" -#include "rgrp.h" -#include "trans.h" -#include "util.h" -#include "super.h" - -/** - * gfs2_create - Create a file - * @dir: The directory in which to create the file - * @dentry: The dentry of the new file - * @mode: The mode of the new file - * - * Returns: errno - */ - -static int gfs2_create(struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - for (;;) { - inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0); - if (!IS_ERR(inode)) { - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - gfs2_glock_dq_uninit_m(2, ghs); - mark_inode_dirty(inode); - break; - } else if (PTR_ERR(inode) != -EEXIST || - (nd && nd->flags & LOOKUP_EXCL)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - inode = gfs2_lookupi(dir, &dentry->d_name, 0); - if (inode) { - if (!IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - break; - } else { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - } - } - - d_instantiate(dentry, inode); - - return 0; -} - -/** - * gfs2_lookup - Look up a filename in a directory and return its inode - * @dir: The directory inode - * @dentry: The dentry of the new inode - * @nd: passed from Linux VFS, ignored by us - * - * Called by the VFS layer. Lock dir and call gfs2_lookupi() - * - * Returns: errno - */ - -static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) -{ - struct inode *inode = NULL; - - inode = gfs2_lookupi(dir, &dentry->d_name, 0); - if (inode && IS_ERR(inode)) - return ERR_CAST(inode); - - if (inode) { - struct gfs2_glock *gl = GFS2_I(inode)->i_gl; - struct gfs2_holder gh; - int error; - error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (error) { - iput(inode); - return ERR_PTR(error); - } - gfs2_glock_dq_uninit(&gh); - return d_splice_alias(inode, dentry); - } - d_add(dentry, inode); - - return NULL; -} - -/** - * gfs2_link - Link to a file - * @old_dentry: The inode to link - * @dir: Add link to this directory - * @dentry: The name of the link - * - * Link the inode in "old_dentry" into the directory "dir" with the - * name in "dentry". - * - * Returns: errno - */ - -static int gfs2_link(struct dentry *old_dentry, struct inode *dir, - struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct inode *inode = old_dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder ghs[2]; - int alloc_required; - int error; - - if (S_ISDIR(inode->i_mode)) - return -EPERM; - - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_check(dir, &dentry->d_name, NULL); - switch (error) { - case -ENOENT: - break; - case 0: - error = -EEXIST; - default: - goto out_gunlock; - } - - error = -EINVAL; - if (!dip->i_inode.i_nlink) - goto out_gunlock; - error = -EFBIG; - if (dip->i_entries == (u32)-1) - goto out_gunlock; - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out_gunlock; - error = -EINVAL; - if (!ip->i_inode.i_nlink) - goto out_gunlock; - error = -EMLINK; - if (ip->i_inode.i_nlink == (u32)-1) - goto out_gunlock; - - alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); - if (error < 0) - goto out_gunlock; - error = 0; - - if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(dip); - if (!al) { - error = -ENOMEM; - goto out_gunlock; - } - - error = gfs2_quota_lock_check(dip); - if (error) - goto out_alloc; - - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve(dip); - if (error) - goto out_gunlock_q; - - error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + - gfs2_rg_blocks(al) + - 2 * RES_DINODE + RES_STATFS + - RES_QUOTA, 0); - if (error) - goto out_ipres; - } else { - error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0); - if (error) - goto out_ipres; - } - - error = gfs2_dir_add(dir, &dentry->d_name, ip, IF2DT(inode->i_mode)); - if (error) - goto out_end_trans; - - error = gfs2_change_nlink(ip, +1); - -out_end_trans: - gfs2_trans_end(sdp); -out_ipres: - if (alloc_required) - gfs2_inplace_release(dip); -out_gunlock_q: - if (alloc_required) - gfs2_quota_unlock(dip); -out_alloc: - if (alloc_required) - gfs2_alloc_put(dip); -out_gunlock: - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_holder_uninit(ghs + 1); - if (!error) { - ihold(inode); - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - } - return error; -} - -/* - * gfs2_unlink_ok - check to see that a inode is still in a directory - * @dip: the directory - * @name: the name of the file - * @ip: the inode - * - * Assumes that the lock on (at least) @dip is held. - * - * Returns: 0 if the parent/child relationship is correct, errno if it isn't - */ - -static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, - const struct gfs2_inode *ip) -{ - int error; - - if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) - return -EPERM; - - if ((dip->i_inode.i_mode & S_ISVTX) && - dip->i_inode.i_uid != current_fsuid() && - ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER)) - return -EPERM; - - if (IS_APPEND(&dip->i_inode)) - return -EPERM; - - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); - if (error) - return error; - - error = gfs2_dir_check(&dip->i_inode, name, ip); - if (error) - return error; - - return 0; -} - -/** - * gfs2_unlink - Unlink a file - * @dir: The inode of the directory containing the file to unlink - * @dentry: The file itself - * - * Unlink a file. Call gfs2_unlinki() - * - * Returns: errno - */ - -static int gfs2_unlink(struct inode *dir, struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder ghs[3]; - struct gfs2_rgrpd *rgd; - struct gfs2_holder ri_gh; - int error; - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_glock_nq(ghs + 2); /* rgrp */ - if (error) - goto out_rgrp; - - error = gfs2_unlink_ok(dip, &dentry->d_name, ip); - if (error) - goto out_gunlock; - - error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_del(dip, &dentry->d_name); - if (error) - goto out_end_trans; - - error = gfs2_change_nlink(ip, -1); - -out_end_trans: - gfs2_trans_end(sdp); -out_gunlock: - gfs2_glock_dq(ghs + 2); -out_rgrp: - gfs2_holder_uninit(ghs + 2); - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_holder_uninit(ghs + 1); - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_symlink - Create a symlink - * @dir: The directory to create the symlink in - * @dentry: The dentry to put the symlink in - * @symname: The thing which the link points to - * - * Returns: errno - */ - -static int gfs2_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) -{ - struct gfs2_inode *dip = GFS2_I(dir), *ip; - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - struct buffer_head *dibh; - int size; - int error; - - /* Must be stuffed with a null terminator for gfs2_follow_link() */ - size = strlen(symname); - if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1) - return -ENAMETOOLONG; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO, 0); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - ip = ghs[1].gh_gl->gl_object; - - i_size_write(inode, size); - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(ip, dibh->b_data); - memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, - size); - brelse(dibh); - } - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/** - * gfs2_mkdir - Make a directory - * @dir: The parent directory of the new one - * @dentry: The dentry of the new directory - * @mode: The mode of the new directory - * - * Returns: errno - */ - -static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) -{ - struct gfs2_inode *dip = GFS2_I(dir), *ip; - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - struct buffer_head *dibh; - int error; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode, 0); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - ip = ghs[1].gh_gl->gl_object; - - ip->i_inode.i_nlink = 2; - i_size_write(inode, sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)); - ip->i_diskflags |= GFS2_DIF_JDATA; - ip->i_entries = 2; - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data; - struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent); - dent->de_inum = di->di_num; /* already GFS2 endian */ - dent->de_type = cpu_to_be16(DT_DIR); - di->di_entries = cpu_to_be32(1); - - dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1)); - gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent); - - gfs2_inum_out(dip, dent); - dent->de_type = cpu_to_be16(DT_DIR); - - gfs2_dinode_out(ip, di); - - brelse(dibh); - } - - error = gfs2_change_nlink(dip, +1); - gfs2_assert_withdraw(sdp, !error); /* dip already pinned */ - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/** - * gfs2_rmdiri - Remove a directory - * @dip: The parent directory of the directory to be removed - * @name: The name of the directory to be removed - * @ip: The GFS2 inode of the directory to be removed - * - * Assumes Glocks on dip and ip are held - * - * Returns: errno - */ - -static int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, - struct gfs2_inode *ip) -{ - int error; - - if (ip->i_entries != 2) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - return -EIO; - } - - error = gfs2_dir_del(dip, name); - if (error) - return error; - - error = gfs2_change_nlink(dip, -1); - if (error) - return error; - - error = gfs2_dir_del(ip, &gfs2_qdot); - if (error) - return error; - - error = gfs2_dir_del(ip, &gfs2_qdotdot); - if (error) - return error; - - /* It looks odd, but it really should be done twice */ - error = gfs2_change_nlink(ip, -1); - if (error) - return error; - - error = gfs2_change_nlink(ip, -1); - if (error) - return error; - - return error; -} - -/** - * gfs2_rmdir - Remove a directory - * @dir: The parent directory of the directory to be removed - * @dentry: The dentry of the directory to remove - * - * Remove a directory. Call gfs2_rmdiri() - * - * Returns: errno - */ - -static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder ghs[3]; - struct gfs2_rgrpd *rgd; - struct gfs2_holder ri_gh; - int error; - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1); - - rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); - gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); - - error = gfs2_glock_nq(ghs); /* parent */ - if (error) - goto out_parent; - - error = gfs2_glock_nq(ghs + 1); /* child */ - if (error) - goto out_child; - - error = gfs2_glock_nq(ghs + 2); /* rgrp */ - if (error) - goto out_rgrp; - - error = gfs2_unlink_ok(dip, &dentry->d_name, ip); - if (error) - goto out_gunlock; - - if (ip->i_entries < 2) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(ip); - error = -EIO; - goto out_gunlock; - } - if (ip->i_entries > 2) { - error = -ENOTEMPTY; - goto out_gunlock; - } - - error = gfs2_trans_begin(sdp, 2 * RES_DINODE + 3 * RES_LEAF + RES_RG_BIT, 0); - if (error) - goto out_gunlock; - - error = gfs2_rmdiri(dip, &dentry->d_name, ip); - - gfs2_trans_end(sdp); - -out_gunlock: - gfs2_glock_dq(ghs + 2); -out_rgrp: - gfs2_holder_uninit(ghs + 2); - gfs2_glock_dq(ghs + 1); -out_child: - gfs2_holder_uninit(ghs + 1); - gfs2_glock_dq(ghs); -out_parent: - gfs2_holder_uninit(ghs); - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_mknod - Make a special file - * @dir: The directory in which the special file will reside - * @dentry: The dentry of the special file - * @mode: The mode of the special file - * @rdev: The device specification of the special file - * - */ - -static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, - dev_t dev) -{ - struct gfs2_inode *dip = GFS2_I(dir); - struct gfs2_sbd *sdp = GFS2_SB(dir); - struct gfs2_holder ghs[2]; - struct inode *inode; - - gfs2_holder_init(dip->i_gl, 0, 0, ghs); - - inode = gfs2_createi(ghs, &dentry->d_name, mode, dev); - if (IS_ERR(inode)) { - gfs2_holder_uninit(ghs); - return PTR_ERR(inode); - } - - gfs2_trans_end(sdp); - if (dip->i_alloc->al_rgd) - gfs2_inplace_release(dip); - gfs2_quota_unlock(dip); - gfs2_alloc_put(dip); - - gfs2_glock_dq_uninit_m(2, ghs); - - d_instantiate(dentry, inode); - mark_inode_dirty(inode); - - return 0; -} - -/* - * gfs2_ok_to_move - check if it's ok to move a directory to another directory - * @this: move this - * @to: to here - * - * Follow @to back to the root and make sure we don't encounter @this - * Assumes we already hold the rename lock. - * - * Returns: errno - */ - -static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to) -{ - struct inode *dir = &to->i_inode; - struct super_block *sb = dir->i_sb; - struct inode *tmp; - int error = 0; - - igrab(dir); - - for (;;) { - if (dir == &this->i_inode) { - error = -EINVAL; - break; - } - if (dir == sb->s_root->d_inode) { - error = 0; - break; - } - - tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1); - if (IS_ERR(tmp)) { - error = PTR_ERR(tmp); - break; - } - - iput(dir); - dir = tmp; - } - - iput(dir); - - return error; -} - -/** - * gfs2_rename - Rename a file - * @odir: Parent directory of old file name - * @odentry: The old dentry of the file - * @ndir: Parent directory of new file name - * @ndentry: The new dentry of the file - * - * Returns: errno - */ - -static int gfs2_rename(struct inode *odir, struct dentry *odentry, - struct inode *ndir, struct dentry *ndentry) -{ - struct gfs2_inode *odip = GFS2_I(odir); - struct gfs2_inode *ndip = GFS2_I(ndir); - struct gfs2_inode *ip = GFS2_I(odentry->d_inode); - struct gfs2_inode *nip = NULL; - struct gfs2_sbd *sdp = GFS2_SB(odir); - struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh; - struct gfs2_rgrpd *nrgd; - unsigned int num_gh; - int dir_rename = 0; - int alloc_required = 0; - unsigned int x; - int error; - - if (ndentry->d_inode) { - nip = GFS2_I(ndentry->d_inode); - if (ip == nip) - return 0; - } - - error = gfs2_rindex_hold(sdp, &ri_gh); - if (error) - return error; - - if (odip != ndip) { - error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, - 0, &r_gh); - if (error) - goto out; - - if (S_ISDIR(ip->i_inode.i_mode)) { - dir_rename = 1; - /* don't move a dirctory into it's subdir */ - error = gfs2_ok_to_move(ip, ndip); - if (error) - goto out_gunlock_r; - } - } - - num_gh = 1; - gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); - if (odip != ndip) { - gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - } - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - - if (nip) { - gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh); - num_gh++; - /* grab the resource lock for unlink flag twiddling - * this is the case of the target file already existing - * so we unlink before doing the rename - */ - nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr); - if (nrgd) - gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++); - } - - for (x = 0; x < num_gh; x++) { - error = gfs2_glock_nq(ghs + x); - if (error) - goto out_gunlock; - } - - /* Check out the old directory */ - - error = gfs2_unlink_ok(odip, &odentry->d_name, ip); - if (error) - goto out_gunlock; - - /* Check out the new directory */ - - if (nip) { - error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip); - if (error) - goto out_gunlock; - - if (S_ISDIR(nip->i_inode.i_mode)) { - if (nip->i_entries < 2) { - if (gfs2_consist_inode(nip)) - gfs2_dinode_print(nip); - error = -EIO; - goto out_gunlock; - } - if (nip->i_entries > 2) { - error = -ENOTEMPTY; - goto out_gunlock; - } - } - } else { - error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0); - if (error) - goto out_gunlock; - - error = gfs2_dir_check(ndir, &ndentry->d_name, NULL); - switch (error) { - case -ENOENT: - error = 0; - break; - case 0: - error = -EEXIST; - default: - goto out_gunlock; - }; - - if (odip != ndip) { - if (!ndip->i_inode.i_nlink) { - error = -EINVAL; - goto out_gunlock; - } - if (ndip->i_entries == (u32)-1) { - error = -EFBIG; - goto out_gunlock; - } - if (S_ISDIR(ip->i_inode.i_mode) && - ndip->i_inode.i_nlink == (u32)-1) { - error = -EMLINK; - goto out_gunlock; - } - } - } - - /* Check out the dir to be renamed */ - - if (dir_rename) { - error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0); - if (error) - goto out_gunlock; - } - - if (nip == NULL) - alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name); - error = alloc_required; - if (error < 0) - goto out_gunlock; - error = 0; - - if (alloc_required) { - struct gfs2_alloc *al = gfs2_alloc_get(ndip); - if (!al) { - error = -ENOMEM; - goto out_gunlock; - } - - error = gfs2_quota_lock_check(ndip); - if (error) - goto out_alloc; - - al->al_requested = sdp->sd_max_dirres; - - error = gfs2_inplace_reserve_ri(ndip); - if (error) - goto out_gunlock_q; - - error = gfs2_trans_begin(sdp, sdp->sd_max_dirres + - gfs2_rg_blocks(al) + - 4 * RES_DINODE + 4 * RES_LEAF + - RES_STATFS + RES_QUOTA + 4, 0); - if (error) - goto out_ipreserv; - } else { - error = gfs2_trans_begin(sdp, 4 * RES_DINODE + - 5 * RES_LEAF + 4, 0); - if (error) - goto out_gunlock; - } - - /* Remove the target file, if it exists */ - - if (nip) { - if (S_ISDIR(nip->i_inode.i_mode)) - error = gfs2_rmdiri(ndip, &ndentry->d_name, nip); - else { - error = gfs2_dir_del(ndip, &ndentry->d_name); - if (error) - goto out_end_trans; - error = gfs2_change_nlink(nip, -1); - } - if (error) - goto out_end_trans; - } - - if (dir_rename) { - error = gfs2_change_nlink(ndip, +1); - if (error) - goto out_end_trans; - error = gfs2_change_nlink(odip, -1); - if (error) - goto out_end_trans; - - error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR); - if (error) - goto out_end_trans; - } else { - struct buffer_head *dibh; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto out_end_trans; - ip->i_inode.i_ctime = CURRENT_TIME; - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - } - - error = gfs2_dir_del(odip, &odentry->d_name); - if (error) - goto out_end_trans; - - error = gfs2_dir_add(ndir, &ndentry->d_name, ip, IF2DT(ip->i_inode.i_mode)); - if (error) - goto out_end_trans; - -out_end_trans: - gfs2_trans_end(sdp); -out_ipreserv: - if (alloc_required) - gfs2_inplace_release(ndip); -out_gunlock_q: - if (alloc_required) - gfs2_quota_unlock(ndip); -out_alloc: - if (alloc_required) - gfs2_alloc_put(ndip); -out_gunlock: - while (x--) { - gfs2_glock_dq(ghs + x); - gfs2_holder_uninit(ghs + x); - } -out_gunlock_r: - if (r_gh.gh_gl) - gfs2_glock_dq_uninit(&r_gh); -out: - gfs2_glock_dq_uninit(&ri_gh); - return error; -} - -/** - * gfs2_follow_link - Follow a symbolic link - * @dentry: The dentry of the link - * @nd: Data that we pass to vfs_follow_link() - * - * This can handle symlinks of any size. - * - * Returns: 0 on success or error code - */ - -static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd) -{ - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); - struct gfs2_holder i_gh; - struct buffer_head *dibh; - unsigned int x, size; - char *buf; - int error; - - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh); - error = gfs2_glock_nq(&i_gh); - if (error) { - gfs2_holder_uninit(&i_gh); - nd_set_link(nd, ERR_PTR(error)); - return NULL; - } - - size = (unsigned int)i_size_read(&ip->i_inode); - if (size == 0) { - gfs2_consist_inode(ip); - buf = ERR_PTR(-EIO); - goto out; - } - - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) { - buf = ERR_PTR(error); - goto out; - } - - x = size + 1; - buf = kmalloc(x, GFP_NOFS); - if (!buf) - buf = ERR_PTR(-ENOMEM); - else - memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), x); - brelse(dibh); -out: - gfs2_glock_dq_uninit(&i_gh); - nd_set_link(nd, buf); - return NULL; -} - -static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p) -{ - char *s = nd_get_link(nd); - if (!IS_ERR(s)) - kfree(s); -} - -/** - * gfs2_permission - - * @inode: The inode - * @mask: The mask to be tested - * @flags: Indicates whether this is an RCU path walk or not - * - * This may be called from the VFS directly, or from within GFS2 with the - * inode locked, so we look to see if the glock is already locked and only - * lock the glock if its not already been done. - * - * Returns: errno - */ - -int gfs2_permission(struct inode *inode, int mask, unsigned int flags) -{ - struct gfs2_inode *ip; - struct gfs2_holder i_gh; - int error; - int unlock = 0; - - - ip = GFS2_I(inode); - if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { - if (flags & IPERM_FLAG_RCU) - return -ECHILD; - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); - if (error) - return error; - unlock = 1; - } - - if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) - error = -EACCES; - else - error = generic_permission(inode, mask, flags, gfs2_check_acl); - if (unlock) - gfs2_glock_dq_uninit(&i_gh); - - return error; -} - -static int setattr_chown(struct inode *inode, struct iattr *attr) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); - u32 ouid, ogid, nuid, ngid; - int error; - - ouid = inode->i_uid; - ogid = inode->i_gid; - nuid = attr->ia_uid; - ngid = attr->ia_gid; - - if (!(attr->ia_valid & ATTR_UID) || ouid == nuid) - ouid = nuid = NO_QUOTA_CHANGE; - if (!(attr->ia_valid & ATTR_GID) || ogid == ngid) - ogid = ngid = NO_QUOTA_CHANGE; - - if (!gfs2_alloc_get(ip)) - return -ENOMEM; - - error = gfs2_quota_lock(ip, nuid, ngid); - if (error) - goto out_alloc; - - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { - error = gfs2_quota_check(ip, nuid, ngid); - if (error) - goto out_gunlock_q; - } - - error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0); - if (error) - goto out_gunlock_q; - - error = gfs2_setattr_simple(ip, attr); - if (error) - goto out_end_trans; - - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { - u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); - gfs2_quota_change(ip, -blocks, ouid, ogid); - gfs2_quota_change(ip, blocks, nuid, ngid); - } - -out_end_trans: - gfs2_trans_end(sdp); -out_gunlock_q: - gfs2_quota_unlock(ip); -out_alloc: - gfs2_alloc_put(ip); - return error; -} - -/** - * gfs2_setattr - Change attributes on an inode - * @dentry: The dentry which is changing - * @attr: The structure describing the change - * - * The VFS layer wants to change one or more of an inodes attributes. Write - * that change out to disk. - * - * Returns: errno - */ - -static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder i_gh; - int error; - - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); - if (error) - return error; - - error = -EPERM; - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out; - - error = inode_change_ok(inode, attr); - if (error) - goto out; - - if (attr->ia_valid & ATTR_SIZE) - error = gfs2_setattr_size(inode, attr->ia_size); - else if (attr->ia_valid & (ATTR_UID | ATTR_GID)) - error = setattr_chown(inode, attr); - else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode)) - error = gfs2_acl_chmod(ip, attr); - else - error = gfs2_setattr_simple(ip, attr); - -out: - gfs2_glock_dq_uninit(&i_gh); - if (!error) - mark_inode_dirty(inode); - return error; -} - -/** - * gfs2_getattr - Read out an inode's attributes - * @mnt: The vfsmount the inode is being accessed from - * @dentry: The dentry to stat - * @stat: The inode's stats - * - * This may be called from the VFS directly, or from within GFS2 with the - * inode locked, so we look to see if the glock is already locked and only - * lock the glock if its not already been done. Note that its the NFS - * readdirplus operation which causes this to be called (from filldir) - * with the glock already held. - * - * Returns: errno - */ - -static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, - struct kstat *stat) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int error; - int unlock = 0; - - if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (error) - return error; - unlock = 1; - } - - generic_fillattr(inode, stat); - if (unlock) - gfs2_glock_dq_uninit(&gh); - - return 0; -} - -static int gfs2_setxattr(struct dentry *dentry, const char *name, - const void *data, size_t size, int flags) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_setxattr(dentry, name, data, size, flags); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name, - void *data, size_t size) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_getxattr(dentry, name, data, size); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static int gfs2_removexattr(struct dentry *dentry, const char *name) -{ - struct inode *inode = dentry->d_inode; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); - ret = gfs2_glock_nq(&gh); - if (ret == 0) { - ret = generic_removexattr(dentry, name); - gfs2_glock_dq(&gh); - } - gfs2_holder_uninit(&gh); - return ret; -} - -static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, - u64 start, u64 len) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder gh; - int ret; - - ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC); - if (ret) - return ret; - - mutex_lock(&inode->i_mutex); - - ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh); - if (ret) - goto out; - - if (gfs2_is_stuffed(ip)) { - u64 phys = ip->i_no_addr << inode->i_blkbits; - u64 size = i_size_read(inode); - u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED| - FIEMAP_EXTENT_DATA_INLINE; - phys += sizeof(struct gfs2_dinode); - phys += start; - if (start + len > size) - len = size - start; - if (start < size) - ret = fiemap_fill_next_extent(fieinfo, start, phys, - len, flags); - if (ret == 1) - ret = 0; - } else { - ret = __generic_block_fiemap(inode, fieinfo, start, len, - gfs2_block_map); - } - - gfs2_glock_dq_uninit(&gh); -out: - mutex_unlock(&inode->i_mutex); - return ret; -} - -const struct inode_operations gfs2_file_iops = { - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - -const struct inode_operations gfs2_dir_iops = { - .create = gfs2_create, - .lookup = gfs2_lookup, - .link = gfs2_link, - .unlink = gfs2_unlink, - .symlink = gfs2_symlink, - .mkdir = gfs2_mkdir, - .rmdir = gfs2_rmdir, - .mknod = gfs2_mknod, - .rename = gfs2_rename, - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - -const struct inode_operations gfs2_symlink_iops = { - .readlink = generic_readlink, - .follow_link = gfs2_follow_link, - .put_link = gfs2_put_link, - .permission = gfs2_permission, - .setattr = gfs2_setattr, - .getattr = gfs2_getattr, - .setxattr = gfs2_setxattr, - .getxattr = gfs2_getxattr, - .listxattr = gfs2_listxattr, - .removexattr = gfs2_removexattr, - .fiemap = gfs2_fiemap, -}; - diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index cf930cd9664a..7273ad3c85ba 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -78,10 +78,11 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal, static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, unsigned char *buf2, unsigned int offset, - unsigned int buflen, u32 block, + struct gfs2_bitmap *bi, u32 block, unsigned char new_state) { unsigned char *byte1, *byte2, *end, cur_state; + unsigned int buflen = bi->bi_len; const unsigned int bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE; byte1 = buf1 + offset + (block / GFS2_NBBY); @@ -92,6 +93,16 @@ static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1, cur_state = (*byte1 >> bit) & GFS2_BIT_MASK; if (unlikely(!valid_change[new_state * 4 + cur_state])) { + printk(KERN_WARNING "GFS2: buf_blk = 0x%llx old_state=%d, " + "new_state=%d\n", + (unsigned long long)block, cur_state, new_state); + printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%lx\n", + (unsigned long long)rgd->rd_addr, + (unsigned long)bi->bi_start); + printk(KERN_WARNING "GFS2: bi_offset=0x%lx bi_len=0x%lx\n", + (unsigned long)bi->bi_offset, + (unsigned long)bi->bi_len); + dump_stack(); gfs2_consist_rgrpd(rgd); return; } @@ -381,6 +392,7 @@ static void clear_rgrpdi(struct gfs2_sbd *sdp) if (gl) { gl->gl_object = NULL; + gfs2_glock_add_to_lru(gl); gfs2_glock_put(gl); } @@ -945,7 +957,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip /* rgblk_search can return a block < goal, so we need to keep it marching forward. */ no_addr = block + rgd->rd_data0; - goal++; + goal = max(block + 1, goal + 1); if (*last_unlinked != NO_BLOCK && no_addr <= *last_unlinked) continue; if (no_addr == skip) @@ -971,7 +983,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip found++; /* Limit reclaim to sensible number of tasks */ - if (found > 2*NR_CPUS) + if (found > NR_CPUS) return; } @@ -1365,7 +1377,7 @@ skip: gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi->bi_len, blk, new_state); + bi, blk, new_state); goal = blk; while (*n < elen) { goal++; @@ -1375,7 +1387,7 @@ skip: GFS2_BLKST_FREE) break; gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset, - bi->bi_len, goal, new_state); + bi, goal, new_state); (*n)++; } out: @@ -1432,7 +1444,7 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart, } gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1); gfs2_setbit(rgd, bi->bi_bh->b_data, NULL, bi->bi_offset, - bi->bi_len, buf_blk, new_state); + bi, buf_blk, new_state); } return rgd; diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index ec73ed70bae1..ed540e7018be 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -23,6 +23,7 @@ #include <linux/time.h> #include <linux/wait.h> #include <linux/writeback.h> +#include <linux/backing-dev.h> #include "gfs2.h" #include "incore.h" @@ -657,7 +658,7 @@ out: * @sdp: the file system * * This function flushes data and meta data for all machines by - * aquiring the transaction log exclusively. All journals are + * acquiring the transaction log exclusively. All journals are * ensured to be in a clean state as well. * * Returns: errno @@ -700,11 +701,47 @@ void gfs2_unfreeze_fs(struct gfs2_sbd *sdp) mutex_unlock(&sdp->sd_freeze_lock); } +void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) +{ + struct gfs2_dinode *str = buf; + + str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); + str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); + str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); + str->di_num.no_addr = cpu_to_be64(ip->i_no_addr); + str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino); + str->di_mode = cpu_to_be32(ip->i_inode.i_mode); + str->di_uid = cpu_to_be32(ip->i_inode.i_uid); + str->di_gid = cpu_to_be32(ip->i_inode.i_gid); + str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); + str->di_size = cpu_to_be64(i_size_read(&ip->i_inode)); + str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode)); + str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); + str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); + str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); + + str->di_goal_meta = cpu_to_be64(ip->i_goal); + str->di_goal_data = cpu_to_be64(ip->i_goal); + str->di_generation = cpu_to_be64(ip->i_generation); + + str->di_flags = cpu_to_be32(ip->i_diskflags); + str->di_height = cpu_to_be16(ip->i_height); + str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && + !(ip->i_diskflags & GFS2_DIF_EXHASH) ? + GFS2_FORMAT_DE : 0); + str->di_depth = cpu_to_be16(ip->i_depth); + str->di_entries = cpu_to_be32(ip->i_entries); + + str->di_eattr = cpu_to_be64(ip->i_eattr); + str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec); + str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec); + str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec); +} /** * gfs2_write_inode - Make sure the inode is stable on the disk * @inode: The inode - * @sync: synchronous write flag + * @wbc: The writeback control structure * * Returns: errno */ @@ -713,15 +750,17 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); + struct address_space *metamapping = gfs2_glock2aspace(ip->i_gl); + struct backing_dev_info *bdi = metamapping->backing_dev_info; struct gfs2_holder gh; struct buffer_head *bh; struct timespec atime; struct gfs2_dinode *di; - int ret = 0; + int ret = -EAGAIN; - /* Check this is a "normal" inode, etc */ + /* Skip timestamp update, if this is from a memalloc */ if (current->flags & PF_MEMALLOC) - return 0; + goto do_flush; ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); if (ret) goto do_flush; @@ -745,6 +784,13 @@ do_unlock: do_flush: if (wbc->sync_mode == WB_SYNC_ALL) gfs2_log_flush(GFS2_SB(inode), ip->i_gl); + filemap_fdatawrite(metamapping); + if (bdi->dirty_exceeded) + gfs2_ail1_flush(sdp, wbc); + if (!ret && (wbc->sync_mode == WB_SYNC_ALL)) + ret = filemap_fdatawait(metamapping); + if (ret) + mark_inode_dirty_sync(inode); return ret; } @@ -874,8 +920,9 @@ restart: static int gfs2_sync_fs(struct super_block *sb, int wait) { - if (wait && sb->s_fs_info) - gfs2_log_flush(sb->s_fs_info, NULL); + struct gfs2_sbd *sdp = sb->s_fs_info; + if (wait && sdp) + gfs2_log_flush(sdp, NULL); return 0; } @@ -1308,6 +1355,78 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) return 0; } +static void gfs2_final_release_pages(struct gfs2_inode *ip) +{ + struct inode *inode = &ip->i_inode; + struct gfs2_glock *gl = ip->i_gl; + + truncate_inode_pages(gfs2_glock2aspace(ip->i_gl), 0); + truncate_inode_pages(&inode->i_data, 0); + + if (atomic_read(&gl->gl_revokes) == 0) { + clear_bit(GLF_LFLUSH, &gl->gl_flags); + clear_bit(GLF_DIRTY, &gl->gl_flags); + } +} + +static int gfs2_dinode_dealloc(struct gfs2_inode *ip) +{ + struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + struct gfs2_alloc *al; + struct gfs2_rgrpd *rgd; + int error; + + if (gfs2_get_inode_blocks(&ip->i_inode) != 1) { + gfs2_consist_inode(ip); + return -EIO; + } + + al = gfs2_alloc_get(ip); + if (!al) + return -ENOMEM; + + error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE); + if (error) + goto out; + + error = gfs2_rindex_hold(sdp, &al->al_ri_gh); + if (error) + goto out_qs; + + rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr); + if (!rgd) { + gfs2_consist_inode(ip); + error = -EIO; + goto out_rindex_relse; + } + + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, + &al->al_rgd_gh); + if (error) + goto out_rindex_relse; + + error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA, + sdp->sd_jdesc->jd_blocks); + if (error) + goto out_rg_gunlock; + + gfs2_free_di(rgd, ip); + + gfs2_final_release_pages(ip); + + gfs2_trans_end(sdp); + +out_rg_gunlock: + gfs2_glock_dq_uninit(&al->al_rgd_gh); +out_rindex_relse: + gfs2_glock_dq_uninit(&al->al_ri_gh); +out_qs: + gfs2_quota_unhold(ip); +out: + gfs2_alloc_put(ip); + return error; +} + /* * We have to (at the moment) hold the inodes main lock to cover * the gap between unlocking the shared lock on the iopen lock and @@ -1318,15 +1437,17 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt) static void gfs2_evict_inode(struct inode *inode) { - struct gfs2_sbd *sdp = inode->i_sb->s_fs_info; + struct super_block *sb = inode->i_sb; + struct gfs2_sbd *sdp = sb->s_fs_info; struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_holder gh; int error; - if (inode->i_nlink) + if (inode->i_nlink || (sb->s_flags & MS_RDONLY)) goto out; - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + /* Must not read inode block until block type has been verified */ + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &gh); if (unlikely(error)) { gfs2_glock_dq_uninit(&ip->i_iopen_gh); goto out; @@ -1336,6 +1457,12 @@ static void gfs2_evict_inode(struct inode *inode) if (error) goto out_truncate; + if (test_bit(GIF_INVALID, &ip->i_flags)) { + error = gfs2_inode_refresh(ip); + if (error) + goto out_truncate; + } + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; gfs2_glock_dq_wait(&ip->i_iopen_gh); gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &ip->i_iopen_gh); @@ -1363,15 +1490,13 @@ static void gfs2_evict_inode(struct inode *inode) } error = gfs2_dinode_dealloc(ip); - if (error) - goto out_unlock; + goto out_unlock; out_truncate: error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks); if (error) goto out_unlock; - /* Needs to be done before glock release & also in a transaction */ - truncate_inode_pages(&inode->i_data, 0); + gfs2_final_release_pages(ip); gfs2_trans_end(sdp); out_unlock: @@ -1386,6 +1511,7 @@ out: end_writeback(inode); ip->i_gl->gl_object = NULL; + gfs2_glock_add_to_lru(ip->i_gl); gfs2_glock_put(ip->i_gl); ip->i_gl = NULL; if (ip->i_iopen_gh.gh_gl) { diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 748ccb557c18..e20eab37bc80 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -81,7 +81,8 @@ static int gfs2_uuid_valid(const u8 *uuid) static ssize_t uuid_show(struct gfs2_sbd *sdp, char *buf) { - const u8 *uuid = sdp->sd_sb.sb_uuid; + struct super_block *s = sdp->sd_vfs; + const u8 *uuid = s->s_uuid; buf[0] = '\0'; if (!gfs2_uuid_valid(uuid)) return 0; @@ -616,7 +617,8 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj, struct kobj_uevent_env *env) { struct gfs2_sbd *sdp = container_of(kobj, struct gfs2_sbd, sd_kobj); - const u8 *uuid = sdp->sd_sb.sb_uuid; + struct super_block *s = sdp->sd_vfs; + const u8 *uuid = s->s_uuid; add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name); add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name); diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h index cedb0bb96d96..5d07609ec57d 100644 --- a/fs/gfs2/trace_gfs2.h +++ b/fs/gfs2/trace_gfs2.h @@ -10,6 +10,7 @@ #include <linux/buffer_head.h> #include <linux/dlmconstants.h> #include <linux/gfs2_ondisk.h> +#include <linux/writeback.h> #include "incore.h" #include "glock.h" @@ -40,7 +41,9 @@ {(1UL << GLF_REPLY_PENDING), "r" }, \ {(1UL << GLF_INITIAL), "I" }, \ {(1UL << GLF_FROZEN), "F" }, \ - {(1UL << GLF_QUEUED), "q" }) + {(1UL << GLF_QUEUED), "q" }, \ + {(1UL << GLF_LRU), "L" }, \ + {(1UL << GLF_OBJECT), "o" }) #ifndef NUMPTY #define NUMPTY @@ -94,7 +97,7 @@ TRACE_EVENT(gfs2_glock_state_change, __entry->new_state = glock_trace_state(new_state); __entry->tgt_state = glock_trace_state(gl->gl_target); __entry->dmt_state = glock_trace_state(gl->gl_demote_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld state %s to %s tgt:%s dmt:%s flags:%s", @@ -127,7 +130,7 @@ TRACE_EVENT(gfs2_glock_put, __entry->gltype = gl->gl_name.ln_type; __entry->glnum = gl->gl_name.ln_number; __entry->cur_state = glock_trace_state(gl->gl_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld state %s => %s flags:%s", @@ -161,7 +164,7 @@ TRACE_EVENT(gfs2_demote_rq, __entry->glnum = gl->gl_name.ln_number; __entry->cur_state = glock_trace_state(gl->gl_state); __entry->dmt_state = glock_trace_state(gl->gl_demote_state); - __entry->flags = gl->gl_flags; + __entry->flags = gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0); ), TP_printk("%u,%u glock %d:%lld demote %s to %s flags:%s", @@ -318,6 +321,33 @@ TRACE_EVENT(gfs2_log_blocks, MINOR(__entry->dev), __entry->blocks) ); +/* Writing back the AIL */ +TRACE_EVENT(gfs2_ail_flush, + + TP_PROTO(const struct gfs2_sbd *sdp, const struct writeback_control *wbc, int start), + + TP_ARGS(sdp, wbc, start), + + TP_STRUCT__entry( + __field( dev_t, dev ) + __field( int, start ) + __field( int, sync_mode ) + __field( long, nr_to_write ) + ), + + TP_fast_assign( + __entry->dev = sdp->sd_vfs->s_dev; + __entry->start = start; + __entry->sync_mode = wbc->sync_mode; + __entry->nr_to_write = wbc->nr_to_write; + ), + + TP_printk("%u,%u ail flush %s %s %ld", MAJOR(__entry->dev), + MINOR(__entry->dev), __entry->start ? "start" : "end", + __entry->sync_mode == WB_SYNC_ALL ? "all" : "none", + __entry->nr_to_write) +); + /* Section 3 - bmap * * Objectives: diff --git a/fs/hpfs/Kconfig b/fs/hpfs/Kconfig index 0c39dc3ef7d7..56bd15c5bf6c 100644 --- a/fs/hpfs/Kconfig +++ b/fs/hpfs/Kconfig @@ -1,7 +1,6 @@ config HPFS_FS tristate "OS/2 HPFS file system support" depends on BLOCK - depends on BROKEN || !PREEMPT help OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS is the file system used for organizing files on OS/2 hard disk diff --git a/fs/hpfs/alloc.c b/fs/hpfs/alloc.c index 5503e2c28910..7a5eb2c718c8 100644 --- a/fs/hpfs/alloc.c +++ b/fs/hpfs/alloc.c @@ -8,8 +8,6 @@ #include "hpfs_fn.h" -static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec); - /* * Check if a sector is allocated in bitmap * This is really slow. Turned on only if chk==2 @@ -18,9 +16,9 @@ static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec); static int chk_if_allocated(struct super_block *s, secno sec, char *msg) { struct quad_buffer_head qbh; - unsigned *bmp; + u32 *bmp; if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "chk"))) goto fail; - if ((bmp[(sec & 0x3fff) >> 5] >> (sec & 0x1f)) & 1) { + if ((cpu_to_le32(bmp[(sec & 0x3fff) >> 5]) >> (sec & 0x1f)) & 1) { hpfs_error(s, "sector '%s' - %08x not allocated in bitmap", msg, sec); goto fail1; } @@ -28,7 +26,7 @@ static int chk_if_allocated(struct super_block *s, secno sec, char *msg) if (sec >= hpfs_sb(s)->sb_dirband_start && sec < hpfs_sb(s)->sb_dirband_start + hpfs_sb(s)->sb_dirband_size) { unsigned ssec = (sec - hpfs_sb(s)->sb_dirband_start) / 4; if (!(bmp = hpfs_map_dnode_bitmap(s, &qbh))) goto fail; - if ((bmp[ssec >> 5] >> (ssec & 0x1f)) & 1) { + if ((le32_to_cpu(bmp[ssec >> 5]) >> (ssec & 0x1f)) & 1) { hpfs_error(s, "sector '%s' - %08x not allocated in directory bitmap", msg, sec); goto fail1; } @@ -75,7 +73,6 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne hpfs_error(s, "Bad allocation size: %d", n); return 0; } - lock_super(s); if (bs != ~0x3fff) { if (!(bmp = hpfs_map_bitmap(s, near >> 14, &qbh, "aib"))) goto uls; } else { @@ -85,10 +82,6 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne ret = bs + nr; goto rt; } - /*if (!tstbits(bmp, nr + n, n + forward)) { - ret = bs + nr + n; - goto rt; - }*/ q = nr + n; b = 0; while ((a = tstbits(bmp, q, n + forward)) != 0) { q += a; @@ -105,14 +98,14 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne goto rt; } nr >>= 5; - /*for (i = nr + 1; i != nr; i++, i &= 0x1ff) {*/ + /*for (i = nr + 1; i != nr; i++, i &= 0x1ff) */ i = nr; do { - if (!bmp[i]) goto cont; - if (n + forward >= 0x3f && bmp[i] != -1) goto cont; + if (!le32_to_cpu(bmp[i])) goto cont; + if (n + forward >= 0x3f && le32_to_cpu(bmp[i]) != 0xffffffff) goto cont; q = i<<5; if (i > 0) { - unsigned k = bmp[i-1]; + unsigned k = le32_to_cpu(bmp[i-1]); while (k & 0x80000000) { q--; k <<= 1; } @@ -132,18 +125,17 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne } while (i != nr); rt: if (ret) { - if (hpfs_sb(s)->sb_chk && ((ret >> 14) != (bs >> 14) || (bmp[(ret & 0x3fff) >> 5] | ~(((1 << n) - 1) << (ret & 0x1f))) != 0xffffffff)) { + if (hpfs_sb(s)->sb_chk && ((ret >> 14) != (bs >> 14) || (le32_to_cpu(bmp[(ret & 0x3fff) >> 5]) | ~(((1 << n) - 1) << (ret & 0x1f))) != 0xffffffff)) { hpfs_error(s, "Allocation doesn't work! Wanted %d, allocated at %08x", n, ret); ret = 0; goto b; } - bmp[(ret & 0x3fff) >> 5] &= ~(((1 << n) - 1) << (ret & 0x1f)); + bmp[(ret & 0x3fff) >> 5] &= cpu_to_le32(~(((1 << n) - 1) << (ret & 0x1f))); hpfs_mark_4buffers_dirty(&qbh); } b: hpfs_brelse4(&qbh); uls: - unlock_super(s); return ret; } @@ -155,7 +147,7 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne * sectors */ -secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forward, int lock) +secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forward) { secno sec; int i; @@ -167,7 +159,6 @@ secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forwa forward = -forward; f_p = 1; } - if (lock) hpfs_lock_creation(s); n_bmps = (sbi->sb_fs_size + 0x4000 - 1) >> 14; if (near && near < sbi->sb_fs_size) { if ((sec = alloc_in_bmp(s, near, n, f_p ? forward : forward/4))) goto ret; @@ -214,18 +205,17 @@ secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forwa ret: if (sec && f_p) { for (i = 0; i < forward; i++) { - if (!hpfs_alloc_if_possible_nolock(s, sec + i + 1)) { + if (!hpfs_alloc_if_possible(s, sec + i + 1)) { hpfs_error(s, "Prealloc doesn't work! Wanted %d, allocated at %08x, can't allocate %d", forward, sec, i); sec = 0; break; } } } - if (lock) hpfs_unlock_creation(s); return sec; } -static secno alloc_in_dirband(struct super_block *s, secno near, int lock) +static secno alloc_in_dirband(struct super_block *s, secno near) { unsigned nr = near; secno sec; @@ -236,49 +226,35 @@ static secno alloc_in_dirband(struct super_block *s, secno near, int lock) nr = sbi->sb_dirband_start + sbi->sb_dirband_size - 4; nr -= sbi->sb_dirband_start; nr >>= 2; - if (lock) hpfs_lock_creation(s); sec = alloc_in_bmp(s, (~0x3fff) | nr, 1, 0); - if (lock) hpfs_unlock_creation(s); if (!sec) return 0; return ((sec & 0x3fff) << 2) + sbi->sb_dirband_start; } /* Alloc sector if it's free */ -static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec) +int hpfs_alloc_if_possible(struct super_block *s, secno sec) { struct quad_buffer_head qbh; - unsigned *bmp; - lock_super(s); + u32 *bmp; if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "aip"))) goto end; - if (bmp[(sec & 0x3fff) >> 5] & (1 << (sec & 0x1f))) { - bmp[(sec & 0x3fff) >> 5] &= ~(1 << (sec & 0x1f)); + if (le32_to_cpu(bmp[(sec & 0x3fff) >> 5]) & (1 << (sec & 0x1f))) { + bmp[(sec & 0x3fff) >> 5] &= cpu_to_le32(~(1 << (sec & 0x1f))); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); - unlock_super(s); return 1; } hpfs_brelse4(&qbh); end: - unlock_super(s); return 0; } -int hpfs_alloc_if_possible(struct super_block *s, secno sec) -{ - int r; - hpfs_lock_creation(s); - r = hpfs_alloc_if_possible_nolock(s, sec); - hpfs_unlock_creation(s); - return r; -} - /* Free sectors in bitmaps */ void hpfs_free_sectors(struct super_block *s, secno sec, unsigned n) { struct quad_buffer_head qbh; - unsigned *bmp; + u32 *bmp; struct hpfs_sb_info *sbi = hpfs_sb(s); /*printk("2 - ");*/ if (!n) return; @@ -286,26 +262,22 @@ void hpfs_free_sectors(struct super_block *s, secno sec, unsigned n) hpfs_error(s, "Trying to free reserved sector %08x", sec); return; } - lock_super(s); sbi->sb_max_fwd_alloc += n > 0xffff ? 0xffff : n; if (sbi->sb_max_fwd_alloc > 0xffffff) sbi->sb_max_fwd_alloc = 0xffffff; new_map: if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "free"))) { - unlock_super(s); return; } new_tst: - if ((bmp[(sec & 0x3fff) >> 5] >> (sec & 0x1f) & 1)) { + if ((le32_to_cpu(bmp[(sec & 0x3fff) >> 5]) >> (sec & 0x1f) & 1)) { hpfs_error(s, "sector %08x not allocated", sec); hpfs_brelse4(&qbh); - unlock_super(s); return; } - bmp[(sec & 0x3fff) >> 5] |= 1 << (sec & 0x1f); + bmp[(sec & 0x3fff) >> 5] |= cpu_to_le32(1 << (sec & 0x1f)); if (!--n) { hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); - unlock_super(s); return; } if (!(++sec & 0x3fff)) { @@ -327,13 +299,13 @@ int hpfs_check_free_dnodes(struct super_block *s, int n) int n_bmps = (hpfs_sb(s)->sb_fs_size + 0x4000 - 1) >> 14; int b = hpfs_sb(s)->sb_c_bitmap & 0x0fffffff; int i, j; - unsigned *bmp; + u32 *bmp; struct quad_buffer_head qbh; if ((bmp = hpfs_map_dnode_bitmap(s, &qbh))) { for (j = 0; j < 512; j++) { unsigned k; - if (!bmp[j]) continue; - for (k = bmp[j]; k; k >>= 1) if (k & 1) if (!--n) { + if (!le32_to_cpu(bmp[j])) continue; + for (k = le32_to_cpu(bmp[j]); k; k >>= 1) if (k & 1) if (!--n) { hpfs_brelse4(&qbh); return 0; } @@ -352,10 +324,10 @@ int hpfs_check_free_dnodes(struct super_block *s, int n) chk_bmp: if (bmp) { for (j = 0; j < 512; j++) { - unsigned k; - if (!bmp[j]) continue; + u32 k; + if (!le32_to_cpu(bmp[j])) continue; for (k = 0xf; k; k <<= 4) - if ((bmp[j] & k) == k) { + if ((le32_to_cpu(bmp[j]) & k) == k) { if (!--n) { hpfs_brelse4(&qbh); return 0; @@ -379,44 +351,40 @@ void hpfs_free_dnode(struct super_block *s, dnode_secno dno) hpfs_free_sectors(s, dno, 4); } else { struct quad_buffer_head qbh; - unsigned *bmp; + u32 *bmp; unsigned ssec = (dno - hpfs_sb(s)->sb_dirband_start) / 4; - lock_super(s); if (!(bmp = hpfs_map_dnode_bitmap(s, &qbh))) { - unlock_super(s); return; } - bmp[ssec >> 5] |= 1 << (ssec & 0x1f); + bmp[ssec >> 5] |= cpu_to_le32(1 << (ssec & 0x1f)); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); - unlock_super(s); } } struct dnode *hpfs_alloc_dnode(struct super_block *s, secno near, - dnode_secno *dno, struct quad_buffer_head *qbh, - int lock) + dnode_secno *dno, struct quad_buffer_head *qbh) { struct dnode *d; if (hpfs_count_one_bitmap(s, hpfs_sb(s)->sb_dmap) > FREE_DNODES_ADD) { - if (!(*dno = alloc_in_dirband(s, near, lock))) - if (!(*dno = hpfs_alloc_sector(s, near, 4, 0, lock))) return NULL; + if (!(*dno = alloc_in_dirband(s, near))) + if (!(*dno = hpfs_alloc_sector(s, near, 4, 0))) return NULL; } else { - if (!(*dno = hpfs_alloc_sector(s, near, 4, 0, lock))) - if (!(*dno = alloc_in_dirband(s, near, lock))) return NULL; + if (!(*dno = hpfs_alloc_sector(s, near, 4, 0))) + if (!(*dno = alloc_in_dirband(s, near))) return NULL; } if (!(d = hpfs_get_4sectors(s, *dno, qbh))) { hpfs_free_dnode(s, *dno); return NULL; } memset(d, 0, 2048); - d->magic = DNODE_MAGIC; - d->first_free = 52; + d->magic = cpu_to_le32(DNODE_MAGIC); + d->first_free = cpu_to_le32(52); d->dirent[0] = 32; d->dirent[2] = 8; d->dirent[30] = 1; d->dirent[31] = 255; - d->self = *dno; + d->self = cpu_to_le32(*dno); return d; } @@ -424,16 +392,16 @@ struct fnode *hpfs_alloc_fnode(struct super_block *s, secno near, fnode_secno *f struct buffer_head **bh) { struct fnode *f; - if (!(*fno = hpfs_alloc_sector(s, near, 1, FNODE_ALLOC_FWD, 1))) return NULL; + if (!(*fno = hpfs_alloc_sector(s, near, 1, FNODE_ALLOC_FWD))) return NULL; if (!(f = hpfs_get_sector(s, *fno, bh))) { hpfs_free_sectors(s, *fno, 1); return NULL; } memset(f, 0, 512); - f->magic = FNODE_MAGIC; - f->ea_offs = 0xc4; + f->magic = cpu_to_le32(FNODE_MAGIC); + f->ea_offs = cpu_to_le16(0xc4); f->btree.n_free_nodes = 8; - f->btree.first_free = 8; + f->btree.first_free = cpu_to_le16(8); return f; } @@ -441,16 +409,16 @@ struct anode *hpfs_alloc_anode(struct super_block *s, secno near, anode_secno *a struct buffer_head **bh) { struct anode *a; - if (!(*ano = hpfs_alloc_sector(s, near, 1, ANODE_ALLOC_FWD, 1))) return NULL; + if (!(*ano = hpfs_alloc_sector(s, near, 1, ANODE_ALLOC_FWD))) return NULL; if (!(a = hpfs_get_sector(s, *ano, bh))) { hpfs_free_sectors(s, *ano, 1); return NULL; } memset(a, 0, 512); - a->magic = ANODE_MAGIC; - a->self = *ano; + a->magic = cpu_to_le32(ANODE_MAGIC); + a->self = cpu_to_le32(*ano); a->btree.n_free_nodes = 40; a->btree.n_used_nodes = 0; - a->btree.first_free = 8; + a->btree.first_free = cpu_to_le16(8); return a; } diff --git a/fs/hpfs/anode.c b/fs/hpfs/anode.c index 6a2f04bf3df0..08b503e8ed29 100644 --- a/fs/hpfs/anode.c +++ b/fs/hpfs/anode.c @@ -22,8 +22,8 @@ secno hpfs_bplus_lookup(struct super_block *s, struct inode *inode, if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, a, &c1, &c2, "hpfs_bplus_lookup")) return -1; if (btree->internal) { for (i = 0; i < btree->n_used_nodes; i++) - if (btree->u.internal[i].file_secno > sec) { - a = btree->u.internal[i].down; + if (le32_to_cpu(btree->u.internal[i].file_secno) > sec) { + a = le32_to_cpu(btree->u.internal[i].down); brelse(bh); if (!(anode = hpfs_map_anode(s, a, &bh))) return -1; btree = &anode->btree; @@ -34,18 +34,18 @@ secno hpfs_bplus_lookup(struct super_block *s, struct inode *inode, return -1; } for (i = 0; i < btree->n_used_nodes; i++) - if (btree->u.external[i].file_secno <= sec && - btree->u.external[i].file_secno + btree->u.external[i].length > sec) { - a = btree->u.external[i].disk_secno + sec - btree->u.external[i].file_secno; + if (le32_to_cpu(btree->u.external[i].file_secno) <= sec && + le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) > sec) { + a = le32_to_cpu(btree->u.external[i].disk_secno) + sec - le32_to_cpu(btree->u.external[i].file_secno); if (hpfs_sb(s)->sb_chk) if (hpfs_chk_sectors(s, a, 1, "data")) { brelse(bh); return -1; } if (inode) { struct hpfs_inode_info *hpfs_inode = hpfs_i(inode); - hpfs_inode->i_file_sec = btree->u.external[i].file_secno; - hpfs_inode->i_disk_sec = btree->u.external[i].disk_secno; - hpfs_inode->i_n_secs = btree->u.external[i].length; + hpfs_inode->i_file_sec = le32_to_cpu(btree->u.external[i].file_secno); + hpfs_inode->i_disk_sec = le32_to_cpu(btree->u.external[i].disk_secno); + hpfs_inode->i_n_secs = le32_to_cpu(btree->u.external[i].length); } brelse(bh); return a; @@ -83,8 +83,8 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi return -1; } if (btree->internal) { - a = btree->u.internal[n].down; - btree->u.internal[n].file_secno = -1; + a = le32_to_cpu(btree->u.internal[n].down); + btree->u.internal[n].file_secno = cpu_to_le32(-1); mark_buffer_dirty(bh); brelse(bh); if (hpfs_sb(s)->sb_chk) @@ -94,15 +94,15 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi goto go_down; } if (n >= 0) { - if (btree->u.external[n].file_secno + btree->u.external[n].length != fsecno) { + if (le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length) != fsecno) { hpfs_error(s, "allocated size %08x, trying to add sector %08x, %cnode %08x", - btree->u.external[n].file_secno + btree->u.external[n].length, fsecno, + le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length), fsecno, fnod?'f':'a', node); brelse(bh); return -1; } - if (hpfs_alloc_if_possible(s, se = btree->u.external[n].disk_secno + btree->u.external[n].length)) { - btree->u.external[n].length++; + if (hpfs_alloc_if_possible(s, se = le32_to_cpu(btree->u.external[n].disk_secno) + le32_to_cpu(btree->u.external[n].length))) { + btree->u.external[n].length = cpu_to_le32(le32_to_cpu(btree->u.external[n].length) + 1); mark_buffer_dirty(bh); brelse(bh); return se; @@ -115,20 +115,20 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi } se = !fnod ? node : (node + 16384) & ~16383; } - if (!(se = hpfs_alloc_sector(s, se, 1, fsecno*ALLOC_M>ALLOC_FWD_MAX ? ALLOC_FWD_MAX : fsecno*ALLOC_M<ALLOC_FWD_MIN ? ALLOC_FWD_MIN : fsecno*ALLOC_M, 1))) { + if (!(se = hpfs_alloc_sector(s, se, 1, fsecno*ALLOC_M>ALLOC_FWD_MAX ? ALLOC_FWD_MAX : fsecno*ALLOC_M<ALLOC_FWD_MIN ? ALLOC_FWD_MIN : fsecno*ALLOC_M))) { brelse(bh); return -1; } - fs = n < 0 ? 0 : btree->u.external[n].file_secno + btree->u.external[n].length; + fs = n < 0 ? 0 : le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length); if (!btree->n_free_nodes) { - up = a != node ? anode->up : -1; + up = a != node ? le32_to_cpu(anode->up) : -1; if (!(anode = hpfs_alloc_anode(s, a, &na, &bh1))) { brelse(bh); hpfs_free_sectors(s, se, 1); return -1; } if (a == node && fnod) { - anode->up = node; + anode->up = cpu_to_le32(node); anode->btree.fnode_parent = 1; anode->btree.n_used_nodes = btree->n_used_nodes; anode->btree.first_free = btree->first_free; @@ -137,9 +137,9 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi btree->internal = 1; btree->n_free_nodes = 11; btree->n_used_nodes = 1; - btree->first_free = (char *)&(btree->u.internal[1]) - (char *)btree; - btree->u.internal[0].file_secno = -1; - btree->u.internal[0].down = na; + btree->first_free = cpu_to_le16((char *)&(btree->u.internal[1]) - (char *)btree); + btree->u.internal[0].file_secno = cpu_to_le32(-1); + btree->u.internal[0].down = cpu_to_le32(na); mark_buffer_dirty(bh); } else if (!(ranode = hpfs_alloc_anode(s, /*a*/0, &ra, &bh2))) { brelse(bh); @@ -153,15 +153,15 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi btree = &anode->btree; } btree->n_free_nodes--; n = btree->n_used_nodes++; - btree->first_free += 12; - btree->u.external[n].disk_secno = se; - btree->u.external[n].file_secno = fs; - btree->u.external[n].length = 1; + btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 12); + btree->u.external[n].disk_secno = cpu_to_le32(se); + btree->u.external[n].file_secno = cpu_to_le32(fs); + btree->u.external[n].length = cpu_to_le32(1); mark_buffer_dirty(bh); brelse(bh); if ((a == node && fnod) || na == -1) return se; c2 = 0; - while (up != -1) { + while (up != (anode_secno)-1) { struct anode *new_anode; if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, up, &c1, &c2, "hpfs_add_sector_to_btree #2")) return -1; @@ -174,47 +174,47 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi } if (btree->n_free_nodes) { btree->n_free_nodes--; n = btree->n_used_nodes++; - btree->first_free += 8; - btree->u.internal[n].file_secno = -1; - btree->u.internal[n].down = na; - btree->u.internal[n-1].file_secno = fs; + btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 8); + btree->u.internal[n].file_secno = cpu_to_le32(-1); + btree->u.internal[n].down = cpu_to_le32(na); + btree->u.internal[n-1].file_secno = cpu_to_le32(fs); mark_buffer_dirty(bh); brelse(bh); brelse(bh2); hpfs_free_sectors(s, ra, 1); if ((anode = hpfs_map_anode(s, na, &bh))) { - anode->up = up; + anode->up = cpu_to_le32(up); anode->btree.fnode_parent = up == node && fnod; mark_buffer_dirty(bh); brelse(bh); } return se; } - up = up != node ? anode->up : -1; - btree->u.internal[btree->n_used_nodes - 1].file_secno = /*fs*/-1; + up = up != node ? le32_to_cpu(anode->up) : -1; + btree->u.internal[btree->n_used_nodes - 1].file_secno = cpu_to_le32(/*fs*/-1); mark_buffer_dirty(bh); brelse(bh); a = na; if ((new_anode = hpfs_alloc_anode(s, a, &na, &bh))) { anode = new_anode; - /*anode->up = up != -1 ? up : ra;*/ + /*anode->up = cpu_to_le32(up != -1 ? up : ra);*/ anode->btree.internal = 1; anode->btree.n_used_nodes = 1; anode->btree.n_free_nodes = 59; - anode->btree.first_free = 16; - anode->btree.u.internal[0].down = a; - anode->btree.u.internal[0].file_secno = -1; + anode->btree.first_free = cpu_to_le16(16); + anode->btree.u.internal[0].down = cpu_to_le32(a); + anode->btree.u.internal[0].file_secno = cpu_to_le32(-1); mark_buffer_dirty(bh); brelse(bh); if ((anode = hpfs_map_anode(s, a, &bh))) { - anode->up = na; + anode->up = cpu_to_le32(na); mark_buffer_dirty(bh); brelse(bh); } } else na = a; } if ((anode = hpfs_map_anode(s, na, &bh))) { - anode->up = node; + anode->up = cpu_to_le32(node); if (fnod) anode->btree.fnode_parent = 1; mark_buffer_dirty(bh); brelse(bh); @@ -232,14 +232,14 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi } btree = &fnode->btree; } - ranode->up = node; - memcpy(&ranode->btree, btree, btree->first_free); + ranode->up = cpu_to_le32(node); + memcpy(&ranode->btree, btree, le16_to_cpu(btree->first_free)); if (fnod) ranode->btree.fnode_parent = 1; ranode->btree.n_free_nodes = (ranode->btree.internal ? 60 : 40) - ranode->btree.n_used_nodes; if (ranode->btree.internal) for (n = 0; n < ranode->btree.n_used_nodes; n++) { struct anode *unode; - if ((unode = hpfs_map_anode(s, ranode->u.internal[n].down, &bh1))) { - unode->up = ra; + if ((unode = hpfs_map_anode(s, le32_to_cpu(ranode->u.internal[n].down), &bh1))) { + unode->up = cpu_to_le32(ra); unode->btree.fnode_parent = 0; mark_buffer_dirty(bh1); brelse(bh1); @@ -248,11 +248,11 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi btree->internal = 1; btree->n_free_nodes = fnod ? 10 : 58; btree->n_used_nodes = 2; - btree->first_free = (char *)&btree->u.internal[2] - (char *)btree; - btree->u.internal[0].file_secno = fs; - btree->u.internal[0].down = ra; - btree->u.internal[1].file_secno = -1; - btree->u.internal[1].down = na; + btree->first_free = cpu_to_le16((char *)&btree->u.internal[2] - (char *)btree); + btree->u.internal[0].file_secno = cpu_to_le32(fs); + btree->u.internal[0].down = cpu_to_le32(ra); + btree->u.internal[1].file_secno = cpu_to_le32(-1); + btree->u.internal[1].down = cpu_to_le32(na); mark_buffer_dirty(bh); brelse(bh); mark_buffer_dirty(bh2); @@ -279,7 +279,7 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree) go_down: d2 = 0; while (btree1->internal) { - ano = btree1->u.internal[pos].down; + ano = le32_to_cpu(btree1->u.internal[pos].down); if (level) brelse(bh); if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, ano, &d1, &d2, "hpfs_remove_btree #1")) @@ -290,7 +290,7 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree) pos = 0; } for (i = 0; i < btree1->n_used_nodes; i++) - hpfs_free_sectors(s, btree1->u.external[i].disk_secno, btree1->u.external[i].length); + hpfs_free_sectors(s, le32_to_cpu(btree1->u.external[i].disk_secno), le32_to_cpu(btree1->u.external[i].length)); go_up: if (!level) return; brelse(bh); @@ -298,13 +298,13 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree) if (hpfs_stop_cycles(s, ano, &c1, &c2, "hpfs_remove_btree #2")) return; hpfs_free_sectors(s, ano, 1); oano = ano; - ano = anode->up; + ano = le32_to_cpu(anode->up); if (--level) { if (!(anode = hpfs_map_anode(s, ano, &bh))) return; btree1 = &anode->btree; } else btree1 = btree; for (i = 0; i < btree1->n_used_nodes; i++) { - if (btree1->u.internal[i].down == oano) { + if (le32_to_cpu(btree1->u.internal[i].down) == oano) { if ((pos = i + 1) < btree1->n_used_nodes) goto go_down; else @@ -411,7 +411,7 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs) if (fno) { btree->n_free_nodes = 8; btree->n_used_nodes = 0; - btree->first_free = 8; + btree->first_free = cpu_to_le16(8); btree->internal = 0; mark_buffer_dirty(bh); } else hpfs_free_sectors(s, f, 1); @@ -421,22 +421,22 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs) while (btree->internal) { nodes = btree->n_used_nodes + btree->n_free_nodes; for (i = 0; i < btree->n_used_nodes; i++) - if (btree->u.internal[i].file_secno >= secs) goto f; + if (le32_to_cpu(btree->u.internal[i].file_secno) >= secs) goto f; brelse(bh); hpfs_error(s, "internal btree %08x doesn't end with -1", node); return; f: for (j = i + 1; j < btree->n_used_nodes; j++) - hpfs_ea_remove(s, btree->u.internal[j].down, 1, 0); + hpfs_ea_remove(s, le32_to_cpu(btree->u.internal[j].down), 1, 0); btree->n_used_nodes = i + 1; btree->n_free_nodes = nodes - btree->n_used_nodes; - btree->first_free = 8 + 8 * btree->n_used_nodes; + btree->first_free = cpu_to_le16(8 + 8 * btree->n_used_nodes); mark_buffer_dirty(bh); - if (btree->u.internal[i].file_secno == secs) { + if (btree->u.internal[i].file_secno == cpu_to_le32(secs)) { brelse(bh); return; } - node = btree->u.internal[i].down; + node = le32_to_cpu(btree->u.internal[i].down); brelse(bh); if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, node, &c1, &c2, "hpfs_truncate_btree")) @@ -446,25 +446,25 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs) } nodes = btree->n_used_nodes + btree->n_free_nodes; for (i = 0; i < btree->n_used_nodes; i++) - if (btree->u.external[i].file_secno + btree->u.external[i].length >= secs) goto ff; + if (le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) >= secs) goto ff; brelse(bh); return; ff: - if (secs <= btree->u.external[i].file_secno) { + if (secs <= le32_to_cpu(btree->u.external[i].file_secno)) { hpfs_error(s, "there is an allocation error in file %08x, sector %08x", f, secs); if (i) i--; } - else if (btree->u.external[i].file_secno + btree->u.external[i].length > secs) { - hpfs_free_sectors(s, btree->u.external[i].disk_secno + secs - - btree->u.external[i].file_secno, btree->u.external[i].length - - secs + btree->u.external[i].file_secno); /* I hope gcc optimizes this :-) */ - btree->u.external[i].length = secs - btree->u.external[i].file_secno; + else if (le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) > secs) { + hpfs_free_sectors(s, le32_to_cpu(btree->u.external[i].disk_secno) + secs - + le32_to_cpu(btree->u.external[i].file_secno), le32_to_cpu(btree->u.external[i].length) + - secs + le32_to_cpu(btree->u.external[i].file_secno)); /* I hope gcc optimizes this :-) */ + btree->u.external[i].length = cpu_to_le32(secs - le32_to_cpu(btree->u.external[i].file_secno)); } for (j = i + 1; j < btree->n_used_nodes; j++) - hpfs_free_sectors(s, btree->u.external[j].disk_secno, btree->u.external[j].length); + hpfs_free_sectors(s, le32_to_cpu(btree->u.external[j].disk_secno), le32_to_cpu(btree->u.external[j].length)); btree->n_used_nodes = i + 1; btree->n_free_nodes = nodes - btree->n_used_nodes; - btree->first_free = 8 + 12 * btree->n_used_nodes; + btree->first_free = cpu_to_le16(8 + 12 * btree->n_used_nodes); mark_buffer_dirty(bh); brelse(bh); } @@ -480,12 +480,12 @@ void hpfs_remove_fnode(struct super_block *s, fnode_secno fno) struct extended_attribute *ea_end; if (!(fnode = hpfs_map_fnode(s, fno, &bh))) return; if (!fnode->dirflag) hpfs_remove_btree(s, &fnode->btree); - else hpfs_remove_dtree(s, fnode->u.external[0].disk_secno); + else hpfs_remove_dtree(s, le32_to_cpu(fnode->u.external[0].disk_secno)); ea_end = fnode_end_ea(fnode); for (ea = fnode_ea(fnode); ea < ea_end; ea = next_ea(ea)) if (ea->indirect) hpfs_ea_remove(s, ea_sec(ea), ea->anode, ea_len(ea)); - hpfs_ea_ext_remove(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l); + hpfs_ea_ext_remove(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l)); brelse(bh); hpfs_free_sectors(s, fno, 1); } diff --git a/fs/hpfs/buffer.c b/fs/hpfs/buffer.c index 793cb9d943d2..9ecde27d1e29 100644 --- a/fs/hpfs/buffer.c +++ b/fs/hpfs/buffer.c @@ -9,22 +9,6 @@ #include <linux/slab.h> #include "hpfs_fn.h" -void hpfs_lock_creation(struct super_block *s) -{ -#ifdef DEBUG_LOCKS - printk("lock creation\n"); -#endif - mutex_lock(&hpfs_sb(s)->hpfs_creation_de); -} - -void hpfs_unlock_creation(struct super_block *s) -{ -#ifdef DEBUG_LOCKS - printk("unlock creation\n"); -#endif - mutex_unlock(&hpfs_sb(s)->hpfs_creation_de); -} - /* Map a sector into a buffer and return pointers to it and to the buffer. */ void *hpfs_map_sector(struct super_block *s, unsigned secno, struct buffer_head **bhp, @@ -32,6 +16,8 @@ void *hpfs_map_sector(struct super_block *s, unsigned secno, struct buffer_head { struct buffer_head *bh; + hpfs_lock_assert(s); + cond_resched(); *bhp = bh = sb_bread(s, secno); @@ -50,6 +36,8 @@ void *hpfs_get_sector(struct super_block *s, unsigned secno, struct buffer_head struct buffer_head *bh; /*return hpfs_map_sector(s, secno, bhp, 0);*/ + hpfs_lock_assert(s); + cond_resched(); if ((*bhp = bh = sb_getblk(s, secno)) != NULL) { @@ -70,6 +58,8 @@ void *hpfs_map_4sectors(struct super_block *s, unsigned secno, struct quad_buffe struct buffer_head *bh; char *data; + hpfs_lock_assert(s); + cond_resched(); if (secno & 3) { @@ -125,6 +115,8 @@ void *hpfs_get_4sectors(struct super_block *s, unsigned secno, { cond_resched(); + hpfs_lock_assert(s); + if (secno & 3) { printk("HPFS: hpfs_get_4sectors: unaligned read\n"); return NULL; diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c index b3d7c0ddb609..f46ae025bfb5 100644 --- a/fs/hpfs/dir.c +++ b/fs/hpfs/dir.c @@ -88,9 +88,9 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) hpfs_error(inode->i_sb, "not a directory, fnode %08lx", (unsigned long)inode->i_ino); } - if (hpfs_inode->i_dno != fno->u.external[0].disk_secno) { + if (hpfs_inode->i_dno != le32_to_cpu(fno->u.external[0].disk_secno)) { e = 1; - hpfs_error(inode->i_sb, "corrupted inode: i_dno == %08x, fnode -> dnode == %08x", hpfs_inode->i_dno, fno->u.external[0].disk_secno); + hpfs_error(inode->i_sb, "corrupted inode: i_dno == %08x, fnode -> dnode == %08x", hpfs_inode->i_dno, le32_to_cpu(fno->u.external[0].disk_secno)); } brelse(bh); if (e) { @@ -156,7 +156,7 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) goto again; } tempname = hpfs_translate_name(inode->i_sb, de->name, de->namelen, lc, de->not_8x3); - if (filldir(dirent, tempname, de->namelen, old_pos, de->fnode, DT_UNKNOWN) < 0) { + if (filldir(dirent, tempname, de->namelen, old_pos, le32_to_cpu(de->fnode), DT_UNKNOWN) < 0) { filp->f_pos = old_pos; if (tempname != de->name) kfree(tempname); hpfs_brelse4(&qbh); @@ -221,7 +221,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name * Get inode number, what we're after. */ - ino = de->fnode; + ino = le32_to_cpu(de->fnode); /* * Go find or make an inode. @@ -236,7 +236,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name hpfs_init_inode(result); if (de->directory) hpfs_read_inode(result); - else if (de->ea_size && hpfs_sb(dir->i_sb)->sb_eas) + else if (le32_to_cpu(de->ea_size) && hpfs_sb(dir->i_sb)->sb_eas) hpfs_read_inode(result); else { result->i_mode |= S_IFREG; @@ -250,8 +250,6 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name hpfs_result = hpfs_i(result); if (!de->directory) hpfs_result->i_parent_dir = dir->i_ino; - hpfs_decide_conv(result, name, len); - if (de->has_acl || de->has_xtd_perm) if (!(dir->i_sb->s_flags & MS_RDONLY)) { hpfs_error(result->i_sb, "ACLs or XPERM found. This is probably HPFS386. This driver doesn't support it now. Send me some info on these structures"); goto bail1; @@ -263,19 +261,19 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name */ if (!result->i_ctime.tv_sec) { - if (!(result->i_ctime.tv_sec = local_to_gmt(dir->i_sb, de->creation_date))) + if (!(result->i_ctime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->creation_date)))) result->i_ctime.tv_sec = 1; result->i_ctime.tv_nsec = 0; - result->i_mtime.tv_sec = local_to_gmt(dir->i_sb, de->write_date); + result->i_mtime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->write_date)); result->i_mtime.tv_nsec = 0; - result->i_atime.tv_sec = local_to_gmt(dir->i_sb, de->read_date); + result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->read_date)); result->i_atime.tv_nsec = 0; - hpfs_result->i_ea_size = de->ea_size; + hpfs_result->i_ea_size = le32_to_cpu(de->ea_size); if (!hpfs_result->i_ea_mode && de->read_only) result->i_mode &= ~0222; if (!de->directory) { if (result->i_size == -1) { - result->i_size = de->file_size; + result->i_size = le32_to_cpu(de->file_size); result->i_data.a_ops = &hpfs_aops; hpfs_i(result)->mmu_private = result->i_size; /* diff --git a/fs/hpfs/dnode.c b/fs/hpfs/dnode.c index 9b2ffadfc8c4..1e0e2ac30fd3 100644 --- a/fs/hpfs/dnode.c +++ b/fs/hpfs/dnode.c @@ -14,11 +14,11 @@ static loff_t get_pos(struct dnode *d, struct hpfs_dirent *fde) struct hpfs_dirent *de_end = dnode_end_de(d); int i = 1; for (de = dnode_first_de(d); de < de_end; de = de_next_de(de)) { - if (de == fde) return ((loff_t) d->self << 4) | (loff_t)i; + if (de == fde) return ((loff_t) le32_to_cpu(d->self) << 4) | (loff_t)i; i++; } printk("HPFS: get_pos: not_found\n"); - return ((loff_t)d->self << 4) | (loff_t)1; + return ((loff_t)le32_to_cpu(d->self) << 4) | (loff_t)1; } void hpfs_add_pos(struct inode *inode, loff_t *pos) @@ -130,29 +130,30 @@ static void set_last_pointer(struct super_block *s, struct dnode *d, dnode_secno { struct hpfs_dirent *de; if (!(de = dnode_last_de(d))) { - hpfs_error(s, "set_last_pointer: empty dnode %08x", d->self); + hpfs_error(s, "set_last_pointer: empty dnode %08x", le32_to_cpu(d->self)); return; } if (hpfs_sb(s)->sb_chk) { if (de->down) { hpfs_error(s, "set_last_pointer: dnode %08x has already last pointer %08x", - d->self, de_down_pointer(de)); + le32_to_cpu(d->self), de_down_pointer(de)); return; } - if (de->length != 32) { - hpfs_error(s, "set_last_pointer: bad last dirent in dnode %08x", d->self); + if (le16_to_cpu(de->length) != 32) { + hpfs_error(s, "set_last_pointer: bad last dirent in dnode %08x", le32_to_cpu(d->self)); return; } } if (ptr) { - if ((d->first_free += 4) > 2048) { - hpfs_error(s,"set_last_pointer: too long dnode %08x", d->self); - d->first_free -= 4; + d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + 4); + if (le32_to_cpu(d->first_free) > 2048) { + hpfs_error(s, "set_last_pointer: too long dnode %08x", le32_to_cpu(d->self)); + d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) - 4); return; } - de->length = 36; + de->length = cpu_to_le16(36); de->down = 1; - *(dnode_secno *)((char *)de + 32) = ptr; + *(dnode_secno *)((char *)de + 32) = cpu_to_le32(ptr); } } @@ -168,7 +169,7 @@ struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d, for (de = dnode_first_de(d); de < de_end; de = de_next_de(de)) { int c = hpfs_compare_names(s, name, namelen, de->name, de->namelen, de->last); if (!c) { - hpfs_error(s, "name (%c,%d) already exists in dnode %08x", *name, namelen, d->self); + hpfs_error(s, "name (%c,%d) already exists in dnode %08x", *name, namelen, le32_to_cpu(d->self)); return NULL; } if (c < 0) break; @@ -176,15 +177,14 @@ struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d, memmove((char *)de + d_size, de, (char *)de_end - (char *)de); memset(de, 0, d_size); if (down_ptr) { - *(int *)((char *)de + d_size - 4) = down_ptr; + *(dnode_secno *)((char *)de + d_size - 4) = cpu_to_le32(down_ptr); de->down = 1; } - de->length = d_size; - if (down_ptr) de->down = 1; + de->length = cpu_to_le16(d_size); de->not_8x3 = hpfs_is_name_long(name, namelen); de->namelen = namelen; memcpy(de->name, name, namelen); - d->first_free += d_size; + d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + d_size); return de; } @@ -194,25 +194,25 @@ static void hpfs_delete_de(struct super_block *s, struct dnode *d, struct hpfs_dirent *de) { if (de->last) { - hpfs_error(s, "attempt to delete last dirent in dnode %08x", d->self); + hpfs_error(s, "attempt to delete last dirent in dnode %08x", le32_to_cpu(d->self)); return; } - d->first_free -= de->length; - memmove(de, de_next_de(de), d->first_free + (char *)d - (char *)de); + d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) - le16_to_cpu(de->length)); + memmove(de, de_next_de(de), le32_to_cpu(d->first_free) + (char *)d - (char *)de); } static void fix_up_ptrs(struct super_block *s, struct dnode *d) { struct hpfs_dirent *de; struct hpfs_dirent *de_end = dnode_end_de(d); - dnode_secno dno = d->self; + dnode_secno dno = le32_to_cpu(d->self); for (de = dnode_first_de(d); de < de_end; de = de_next_de(de)) if (de->down) { struct quad_buffer_head qbh; struct dnode *dd; if ((dd = hpfs_map_dnode(s, de_down_pointer(de), &qbh))) { - if (dd->up != dno || dd->root_dnode) { - dd->up = dno; + if (le32_to_cpu(dd->up) != dno || dd->root_dnode) { + dd->up = cpu_to_le32(dno); dd->root_dnode = 0; hpfs_mark_4buffers_dirty(&qbh); } @@ -262,7 +262,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, kfree(nname); return 1; } - if (d->first_free + de_size(namelen, down_ptr) <= 2048) { + if (le32_to_cpu(d->first_free) + de_size(namelen, down_ptr) <= 2048) { loff_t t; copy_de(de=hpfs_add_de(i->i_sb, d, name, namelen, down_ptr), new_de); t = get_pos(d, de); @@ -286,11 +286,11 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, kfree(nname); return 1; } - memcpy(nd, d, d->first_free); + memcpy(nd, d, le32_to_cpu(d->first_free)); copy_de(de = hpfs_add_de(i->i_sb, nd, name, namelen, down_ptr), new_de); for_all_poss(i, hpfs_pos_ins, get_pos(nd, de), 1); h = ((char *)dnode_last_de(nd) - (char *)nd) / 2 + 10; - if (!(ad = hpfs_alloc_dnode(i->i_sb, d->up, &adno, &qbh1, 0))) { + if (!(ad = hpfs_alloc_dnode(i->i_sb, le32_to_cpu(d->up), &adno, &qbh1))) { hpfs_error(i->i_sb, "unable to alloc dnode - dnode tree will be corrupted"); hpfs_brelse4(&qbh); kfree(nd); @@ -313,20 +313,21 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, down_ptr = adno; set_last_pointer(i->i_sb, ad, de->down ? de_down_pointer(de) : 0); de = de_next_de(de); - memmove((char *)nd + 20, de, nd->first_free + (char *)nd - (char *)de); - nd->first_free -= (char *)de - (char *)nd - 20; - memcpy(d, nd, nd->first_free); + memmove((char *)nd + 20, de, le32_to_cpu(nd->first_free) + (char *)nd - (char *)de); + nd->first_free = cpu_to_le32(le32_to_cpu(nd->first_free) - ((char *)de - (char *)nd - 20)); + memcpy(d, nd, le32_to_cpu(nd->first_free)); for_all_poss(i, hpfs_pos_del, (loff_t)dno << 4, pos); fix_up_ptrs(i->i_sb, ad); if (!d->root_dnode) { - dno = ad->up = d->up; + ad->up = d->up; + dno = le32_to_cpu(ad->up); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); goto go_up; } - if (!(rd = hpfs_alloc_dnode(i->i_sb, d->up, &rdno, &qbh2, 0))) { + if (!(rd = hpfs_alloc_dnode(i->i_sb, le32_to_cpu(d->up), &rdno, &qbh2))) { hpfs_error(i->i_sb, "unable to alloc dnode - dnode tree will be corrupted"); hpfs_brelse4(&qbh); hpfs_brelse4(&qbh1); @@ -338,7 +339,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, i->i_blocks += 4; rd->root_dnode = 1; rd->up = d->up; - if (!(fnode = hpfs_map_fnode(i->i_sb, d->up, &bh))) { + if (!(fnode = hpfs_map_fnode(i->i_sb, le32_to_cpu(d->up), &bh))) { hpfs_free_dnode(i->i_sb, rdno); hpfs_brelse4(&qbh); hpfs_brelse4(&qbh1); @@ -347,10 +348,11 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, kfree(nname); return 1; } - fnode->u.external[0].disk_secno = rdno; + fnode->u.external[0].disk_secno = cpu_to_le32(rdno); mark_buffer_dirty(bh); brelse(bh); - d->up = ad->up = hpfs_i(i)->i_dno = rdno; + hpfs_i(i)->i_dno = rdno; + d->up = ad->up = cpu_to_le32(rdno); d->root_dnode = ad->root_dnode = 0; hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); @@ -373,7 +375,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, int hpfs_add_dirent(struct inode *i, const unsigned char *name, unsigned namelen, - struct hpfs_dirent *new_de, int cdepth) + struct hpfs_dirent *new_de) { struct hpfs_inode_info *hpfs_inode = hpfs_i(i); struct dnode *d; @@ -403,7 +405,6 @@ int hpfs_add_dirent(struct inode *i, } } hpfs_brelse4(&qbh); - if (!cdepth) hpfs_lock_creation(i->i_sb); if (hpfs_check_free_dnodes(i->i_sb, FREE_DNODES_ADD)) { c = 1; goto ret; @@ -411,7 +412,6 @@ int hpfs_add_dirent(struct inode *i, i->i_version++; c = hpfs_add_to_dnode(i, dno, name, namelen, new_de, 0); ret: - if (!cdepth) hpfs_unlock_creation(i->i_sb); return c; } @@ -437,9 +437,9 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to) return 0; if (!(dnode = hpfs_map_dnode(i->i_sb, dno, &qbh))) return 0; if (hpfs_sb(i->i_sb)->sb_chk) { - if (dnode->up != chk_up) { + if (le32_to_cpu(dnode->up) != chk_up) { hpfs_error(i->i_sb, "move_to_top: up pointer from %08x should be %08x, is %08x", - dno, chk_up, dnode->up); + dno, chk_up, le32_to_cpu(dnode->up)); hpfs_brelse4(&qbh); return 0; } @@ -455,7 +455,7 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to) hpfs_brelse4(&qbh); } while (!(de = dnode_pre_last_de(dnode))) { - dnode_secno up = dnode->up; + dnode_secno up = le32_to_cpu(dnode->up); hpfs_brelse4(&qbh); hpfs_free_dnode(i->i_sb, dno); i->i_size -= 2048; @@ -474,8 +474,8 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to) hpfs_brelse4(&qbh); return 0; } - dnode->first_free -= 4; - de->length -= 4; + dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4); + de->length = cpu_to_le16(le16_to_cpu(de->length) - 4); de->down = 0; hpfs_mark_4buffers_dirty(&qbh); dno = up; @@ -483,12 +483,12 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to) t = get_pos(dnode, de); for_all_poss(i, hpfs_pos_subst, t, 4); for_all_poss(i, hpfs_pos_subst, t + 1, 5); - if (!(nde = kmalloc(de->length, GFP_NOFS))) { + if (!(nde = kmalloc(le16_to_cpu(de->length), GFP_NOFS))) { hpfs_error(i->i_sb, "out of memory for dirent - directory will be corrupted"); hpfs_brelse4(&qbh); return 0; } - memcpy(nde, de, de->length); + memcpy(nde, de, le16_to_cpu(de->length)); ddno = de->down ? de_down_pointer(de) : 0; hpfs_delete_de(i->i_sb, dnode, de); set_last_pointer(i->i_sb, dnode, ddno); @@ -517,11 +517,11 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) try_it_again: if (hpfs_stop_cycles(i->i_sb, dno, &c1, &c2, "delete_empty_dnode")) return; if (!(dnode = hpfs_map_dnode(i->i_sb, dno, &qbh))) return; - if (dnode->first_free > 56) goto end; - if (dnode->first_free == 52 || dnode->first_free == 56) { + if (le32_to_cpu(dnode->first_free) > 56) goto end; + if (le32_to_cpu(dnode->first_free) == 52 || le32_to_cpu(dnode->first_free) == 56) { struct hpfs_dirent *de_end; int root = dnode->root_dnode; - up = dnode->up; + up = le32_to_cpu(dnode->up); de = dnode_first_de(dnode); down = de->down ? de_down_pointer(de) : 0; if (hpfs_sb(i->i_sb)->sb_chk) if (root && !down) { @@ -545,13 +545,13 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) return; } if ((d1 = hpfs_map_dnode(i->i_sb, down, &qbh1))) { - d1->up = up; + d1->up = cpu_to_le32(up); d1->root_dnode = 1; hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); } if ((fnode = hpfs_map_fnode(i->i_sb, up, &bh))) { - fnode->u.external[0].disk_secno = down; + fnode->u.external[0].disk_secno = cpu_to_le32(down); mark_buffer_dirty(bh); brelse(bh); } @@ -570,22 +570,22 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) for_all_poss(i, hpfs_pos_subst, ((loff_t)dno << 4) | 1, ((loff_t)up << 4) | p); if (!down) { de->down = 0; - de->length -= 4; - dnode->first_free -= 4; + de->length = cpu_to_le16(le16_to_cpu(de->length) - 4); + dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4); memmove(de_next_de(de), (char *)de_next_de(de) + 4, - (char *)dnode + dnode->first_free - (char *)de_next_de(de)); + (char *)dnode + le32_to_cpu(dnode->first_free) - (char *)de_next_de(de)); } else { struct dnode *d1; struct quad_buffer_head qbh1; - *(dnode_secno *) ((void *) de + de->length - 4) = down; + *(dnode_secno *) ((void *) de + le16_to_cpu(de->length) - 4) = down; if ((d1 = hpfs_map_dnode(i->i_sb, down, &qbh1))) { - d1->up = up; + d1->up = cpu_to_le32(up); hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); } } } else { - hpfs_error(i->i_sb, "delete_empty_dnode: dnode %08x, first_free == %03x", dno, dnode->first_free); + hpfs_error(i->i_sb, "delete_empty_dnode: dnode %08x, first_free == %03x", dno, le32_to_cpu(dnode->first_free)); goto end; } @@ -596,18 +596,18 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) struct quad_buffer_head qbh1; if (!de_next->down) goto endm; ndown = de_down_pointer(de_next); - if (!(de_cp = kmalloc(de->length, GFP_NOFS))) { + if (!(de_cp = kmalloc(le16_to_cpu(de->length), GFP_NOFS))) { printk("HPFS: out of memory for dtree balancing\n"); goto endm; } - memcpy(de_cp, de, de->length); + memcpy(de_cp, de, le16_to_cpu(de->length)); hpfs_delete_de(i->i_sb, dnode, de); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | p, 4); for_all_poss(i, hpfs_pos_del, ((loff_t)up << 4) | p, 1); if (de_cp->down) if ((d1 = hpfs_map_dnode(i->i_sb, de_down_pointer(de_cp), &qbh1))) { - d1->up = ndown; + d1->up = cpu_to_le32(ndown); hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); } @@ -635,7 +635,7 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) struct hpfs_dirent *del = dnode_last_de(d1); dlp = del->down ? de_down_pointer(del) : 0; if (!dlp && down) { - if (d1->first_free > 2044) { + if (le32_to_cpu(d1->first_free) > 2044) { if (hpfs_sb(i->i_sb)->sb_chk >= 2) { printk("HPFS: warning: unbalanced dnode tree, see hpfs.txt 4 more info\n"); printk("HPFS: warning: terminating balancing operation\n"); @@ -647,38 +647,38 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) printk("HPFS: warning: unbalanced dnode tree, see hpfs.txt 4 more info\n"); printk("HPFS: warning: goin'on\n"); } - del->length += 4; + del->length = cpu_to_le16(le16_to_cpu(del->length) + 4); del->down = 1; - d1->first_free += 4; + d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) + 4); } if (dlp && !down) { - del->length -= 4; + del->length = cpu_to_le16(le16_to_cpu(del->length) - 4); del->down = 0; - d1->first_free -= 4; + d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) - 4); } else if (down) - *(dnode_secno *) ((void *) del + del->length - 4) = down; + *(dnode_secno *) ((void *) del + le16_to_cpu(del->length) - 4) = cpu_to_le32(down); } else goto endm; - if (!(de_cp = kmalloc(de_prev->length, GFP_NOFS))) { + if (!(de_cp = kmalloc(le16_to_cpu(de_prev->length), GFP_NOFS))) { printk("HPFS: out of memory for dtree balancing\n"); hpfs_brelse4(&qbh1); goto endm; } hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); - memcpy(de_cp, de_prev, de_prev->length); + memcpy(de_cp, de_prev, le16_to_cpu(de_prev->length)); hpfs_delete_de(i->i_sb, dnode, de_prev); if (!de_prev->down) { - de_prev->length += 4; + de_prev->length = cpu_to_le16(le16_to_cpu(de_prev->length) + 4); de_prev->down = 1; - dnode->first_free += 4; + dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) + 4); } - *(dnode_secno *) ((void *) de_prev + de_prev->length - 4) = ndown; + *(dnode_secno *) ((void *) de_prev + le16_to_cpu(de_prev->length) - 4) = cpu_to_le32(ndown); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | (p - 1), 4); for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | p, ((loff_t)up << 4) | (p - 1)); if (down) if ((d1 = hpfs_map_dnode(i->i_sb, de_down_pointer(de), &qbh1))) { - d1->up = ndown; + d1->up = cpu_to_le32(ndown); hpfs_mark_4buffers_dirty(&qbh1); hpfs_brelse4(&qbh1); } @@ -701,7 +701,6 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de, { struct dnode *dnode = qbh->data; dnode_secno down = 0; - int lock = 0; loff_t t; if (de->first || de->last) { hpfs_error(i->i_sb, "hpfs_remove_dirent: attempt to delete first or last dirent in dnode %08x", dno); @@ -710,11 +709,8 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de, } if (de->down) down = de_down_pointer(de); if (depth && (de->down || (de == dnode_first_de(dnode) && de_next_de(de)->last))) { - lock = 1; - hpfs_lock_creation(i->i_sb); if (hpfs_check_free_dnodes(i->i_sb, FREE_DNODES_DEL)) { hpfs_brelse4(qbh); - hpfs_unlock_creation(i->i_sb); return 2; } } @@ -727,11 +723,9 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de, dnode_secno a = move_to_top(i, down, dno); for_all_poss(i, hpfs_pos_subst, 5, t); if (a) delete_empty_dnode(i, a); - if (lock) hpfs_unlock_creation(i->i_sb); return !a; } delete_empty_dnode(i, dno); - if (lock) hpfs_unlock_creation(i->i_sb); return 0; } @@ -751,8 +745,8 @@ void hpfs_count_dnodes(struct super_block *s, dnode_secno dno, int *n_dnodes, ptr = 0; go_up: if (!(dnode = hpfs_map_dnode(s, dno, &qbh))) return; - if (hpfs_sb(s)->sb_chk) if (odno && odno != -1 && dnode->up != odno) - hpfs_error(s, "hpfs_count_dnodes: bad up pointer; dnode %08x, down %08x points to %08x", odno, dno, dnode->up); + if (hpfs_sb(s)->sb_chk) if (odno && odno != -1 && le32_to_cpu(dnode->up) != odno) + hpfs_error(s, "hpfs_count_dnodes: bad up pointer; dnode %08x, down %08x points to %08x", odno, dno, le32_to_cpu(dnode->up)); de = dnode_first_de(dnode); if (ptr) while(1) { if (de->down) if (de_down_pointer(de) == ptr) goto process_de; @@ -776,7 +770,7 @@ void hpfs_count_dnodes(struct super_block *s, dnode_secno dno, int *n_dnodes, if (!de->first && !de->last && n_items) (*n_items)++; if ((de = de_next_de(de)) < dnode_end_de(dnode)) goto next_de; ptr = dno; - dno = dnode->up; + dno = le32_to_cpu(dnode->up); if (dnode->root_dnode) { hpfs_brelse4(&qbh); return; @@ -824,8 +818,8 @@ dnode_secno hpfs_de_as_down_as_possible(struct super_block *s, dnode_secno dno) return d; if (!(de = map_nth_dirent(s, d, 1, &qbh, NULL))) return dno; if (hpfs_sb(s)->sb_chk) - if (up && ((struct dnode *)qbh.data)->up != up) - hpfs_error(s, "hpfs_de_as_down_as_possible: bad up pointer; dnode %08x, down %08x points to %08x", up, d, ((struct dnode *)qbh.data)->up); + if (up && le32_to_cpu(((struct dnode *)qbh.data)->up) != up) + hpfs_error(s, "hpfs_de_as_down_as_possible: bad up pointer; dnode %08x, down %08x points to %08x", up, d, le32_to_cpu(((struct dnode *)qbh.data)->up)); if (!de->down) { hpfs_brelse4(&qbh); return d; @@ -874,7 +868,7 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp, /* Going up */ if (dnode->root_dnode) goto bail; - if (!(up_dnode = hpfs_map_dnode(inode->i_sb, dnode->up, &qbh0))) + if (!(up_dnode = hpfs_map_dnode(inode->i_sb, le32_to_cpu(dnode->up), &qbh0))) goto bail; end_up_de = dnode_end_de(up_dnode); @@ -882,16 +876,16 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp, for (up_de = dnode_first_de(up_dnode); up_de < end_up_de; up_de = de_next_de(up_de)) { if (!(++c & 077)) hpfs_error(inode->i_sb, - "map_pos_dirent: pos crossed dnode boundary; dnode = %08x", dnode->up); + "map_pos_dirent: pos crossed dnode boundary; dnode = %08x", le32_to_cpu(dnode->up)); if (up_de->down && de_down_pointer(up_de) == dno) { - *posp = ((loff_t) dnode->up << 4) + c; + *posp = ((loff_t) le32_to_cpu(dnode->up) << 4) + c; hpfs_brelse4(&qbh0); return de; } } hpfs_error(inode->i_sb, "map_pos_dirent: pointer to dnode %08x not found in parent dnode %08x", - dno, dnode->up); + dno, le32_to_cpu(dnode->up)); hpfs_brelse4(&qbh0); bail: @@ -1017,17 +1011,17 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno, /*name2[15] = 0xff;*/ name1len = 15; name2len = 256; } - if (!(upf = hpfs_map_fnode(s, f->up, &bh))) { + if (!(upf = hpfs_map_fnode(s, le32_to_cpu(f->up), &bh))) { kfree(name2); return NULL; } if (!upf->dirflag) { brelse(bh); - hpfs_error(s, "fnode %08x has non-directory parent %08x", fno, f->up); + hpfs_error(s, "fnode %08x has non-directory parent %08x", fno, le32_to_cpu(f->up)); kfree(name2); return NULL; } - dno = upf->u.external[0].disk_secno; + dno = le32_to_cpu(upf->u.external[0].disk_secno); brelse(bh); go_down: downd = 0; @@ -1049,7 +1043,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno, return NULL; } next_de: - if (de->fnode == fno) { + if (le32_to_cpu(de->fnode) == fno) { kfree(name2); return de; } @@ -1065,7 +1059,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno, goto go_down; } f: - if (de->fnode == fno) { + if (le32_to_cpu(de->fnode) == fno) { kfree(name2); return de; } @@ -1074,7 +1068,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno, if ((de = de_next_de(de)) < de_end) goto next_de; if (d->root_dnode) goto not_found; downd = dno; - dno = d->up; + dno = le32_to_cpu(d->up); hpfs_brelse4(qbh); if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, downd, &d1, &d2, "map_fnode_dirent #2")) { diff --git a/fs/hpfs/ea.c b/fs/hpfs/ea.c index 45e53d972b42..d8b84d113c89 100644 --- a/fs/hpfs/ea.c +++ b/fs/hpfs/ea.c @@ -24,7 +24,7 @@ void hpfs_ea_ext_remove(struct super_block *s, secno a, int ano, unsigned len) } if (hpfs_ea_read(s, a, ano, pos, 4, ex)) return; if (ea->indirect) { - if (ea->valuelen != 8) { + if (ea_valuelen(ea) != 8) { hpfs_error(s, "ea->indirect set while ea->valuelen!=8, %s %08x, pos %08x", ano ? "anode" : "sectors", a, pos); return; @@ -33,7 +33,7 @@ void hpfs_ea_ext_remove(struct super_block *s, secno a, int ano, unsigned len) return; hpfs_ea_remove(s, ea_sec(ea), ea->anode, ea_len(ea)); } - pos += ea->namelen + ea->valuelen + 5; + pos += ea->namelen + ea_valuelen(ea) + 5; } if (!ano) hpfs_free_sectors(s, a, (len+511) >> 9); else { @@ -76,24 +76,24 @@ int hpfs_read_ea(struct super_block *s, struct fnode *fnode, char *key, unsigned pos; int ano, len; secno a; + char ex[4 + 255 + 1 + 8]; struct extended_attribute *ea; struct extended_attribute *ea_end = fnode_end_ea(fnode); for (ea = fnode_ea(fnode); ea < ea_end; ea = next_ea(ea)) if (!strcmp(ea->name, key)) { if (ea->indirect) goto indirect; - if (ea->valuelen >= size) + if (ea_valuelen(ea) >= size) return -EINVAL; - memcpy(buf, ea_data(ea), ea->valuelen); - buf[ea->valuelen] = 0; + memcpy(buf, ea_data(ea), ea_valuelen(ea)); + buf[ea_valuelen(ea)] = 0; return 0; } - a = fnode->ea_secno; - len = fnode->ea_size_l; + a = le32_to_cpu(fnode->ea_secno); + len = le32_to_cpu(fnode->ea_size_l); ano = fnode->ea_anode; pos = 0; while (pos < len) { - char ex[4 + 255 + 1 + 8]; ea = (struct extended_attribute *)ex; if (pos + 4 > len) { hpfs_error(s, "EAs don't end correctly, %s %08x, len %08x", @@ -106,14 +106,14 @@ int hpfs_read_ea(struct super_block *s, struct fnode *fnode, char *key, if (!strcmp(ea->name, key)) { if (ea->indirect) goto indirect; - if (ea->valuelen >= size) + if (ea_valuelen(ea) >= size) return -EINVAL; - if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea->valuelen, buf)) + if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea_valuelen(ea), buf)) return -EIO; - buf[ea->valuelen] = 0; + buf[ea_valuelen(ea)] = 0; return 0; } - pos += ea->namelen + ea->valuelen + 5; + pos += ea->namelen + ea_valuelen(ea) + 5; } return -ENOENT; indirect: @@ -138,16 +138,16 @@ char *hpfs_get_ea(struct super_block *s, struct fnode *fnode, char *key, int *si if (!strcmp(ea->name, key)) { if (ea->indirect) return get_indirect_ea(s, ea->anode, ea_sec(ea), *size = ea_len(ea)); - if (!(ret = kmalloc((*size = ea->valuelen) + 1, GFP_NOFS))) { + if (!(ret = kmalloc((*size = ea_valuelen(ea)) + 1, GFP_NOFS))) { printk("HPFS: out of memory for EA\n"); return NULL; } - memcpy(ret, ea_data(ea), ea->valuelen); - ret[ea->valuelen] = 0; + memcpy(ret, ea_data(ea), ea_valuelen(ea)); + ret[ea_valuelen(ea)] = 0; return ret; } - a = fnode->ea_secno; - len = fnode->ea_size_l; + a = le32_to_cpu(fnode->ea_secno); + len = le32_to_cpu(fnode->ea_size_l); ano = fnode->ea_anode; pos = 0; while (pos < len) { @@ -164,18 +164,18 @@ char *hpfs_get_ea(struct super_block *s, struct fnode *fnode, char *key, int *si if (!strcmp(ea->name, key)) { if (ea->indirect) return get_indirect_ea(s, ea->anode, ea_sec(ea), *size = ea_len(ea)); - if (!(ret = kmalloc((*size = ea->valuelen) + 1, GFP_NOFS))) { + if (!(ret = kmalloc((*size = ea_valuelen(ea)) + 1, GFP_NOFS))) { printk("HPFS: out of memory for EA\n"); return NULL; } - if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea->valuelen, ret)) { + if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea_valuelen(ea), ret)) { kfree(ret); return NULL; } - ret[ea->valuelen] = 0; + ret[ea_valuelen(ea)] = 0; return ret; } - pos += ea->namelen + ea->valuelen + 5; + pos += ea->namelen + ea_valuelen(ea) + 5; } return NULL; } @@ -202,13 +202,13 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key, if (ea->indirect) { if (ea_len(ea) == size) set_indirect_ea(s, ea->anode, ea_sec(ea), data, size); - } else if (ea->valuelen == size) { + } else if (ea_valuelen(ea) == size) { memcpy(ea_data(ea), data, size); } return; } - a = fnode->ea_secno; - len = fnode->ea_size_l; + a = le32_to_cpu(fnode->ea_secno); + len = le32_to_cpu(fnode->ea_size_l); ano = fnode->ea_anode; pos = 0; while (pos < len) { @@ -228,68 +228,70 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key, set_indirect_ea(s, ea->anode, ea_sec(ea), data, size); } else { - if (ea->valuelen == size) + if (ea_valuelen(ea) == size) hpfs_ea_write(s, a, ano, pos + 4 + ea->namelen + 1, size, data); } return; } - pos += ea->namelen + ea->valuelen + 5; + pos += ea->namelen + ea_valuelen(ea) + 5; } - if (!fnode->ea_offs) { - /*if (fnode->ea_size_s) { + if (!le16_to_cpu(fnode->ea_offs)) { + /*if (le16_to_cpu(fnode->ea_size_s)) { hpfs_error(s, "fnode %08x: ea_size_s == %03x, ea_offs == 0", - inode->i_ino, fnode->ea_size_s); + inode->i_ino, le16_to_cpu(fnode->ea_size_s)); return; }*/ - fnode->ea_offs = 0xc4; + fnode->ea_offs = cpu_to_le16(0xc4); } - if (fnode->ea_offs < 0xc4 || fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200) { + if (le16_to_cpu(fnode->ea_offs) < 0xc4 || le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) > 0x200) { hpfs_error(s, "fnode %08lx: ea_offs == %03x, ea_size_s == %03x", (unsigned long)inode->i_ino, - fnode->ea_offs, fnode->ea_size_s); + le32_to_cpu(fnode->ea_offs), le16_to_cpu(fnode->ea_size_s)); return; } - if ((fnode->ea_size_s || !fnode->ea_size_l) && - fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s + strlen(key) + size + 5 <= 0x200) { + if ((le16_to_cpu(fnode->ea_size_s) || !le32_to_cpu(fnode->ea_size_l)) && + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) + strlen(key) + size + 5 <= 0x200) { ea = fnode_end_ea(fnode); *(char *)ea = 0; ea->namelen = strlen(key); - ea->valuelen = size; + ea->valuelen_lo = size; + ea->valuelen_hi = size >> 8; strcpy(ea->name, key); memcpy(ea_data(ea), data, size); - fnode->ea_size_s += strlen(key) + size + 5; + fnode->ea_size_s = cpu_to_le16(le16_to_cpu(fnode->ea_size_s) + strlen(key) + size + 5); goto ret; } /* Most the code here is 99.9993422% unused. I hope there are no bugs. But what .. HPFS.IFS has also bugs in ea management. */ - if (fnode->ea_size_s && !fnode->ea_size_l) { + if (le16_to_cpu(fnode->ea_size_s) && !le32_to_cpu(fnode->ea_size_l)) { secno n; struct buffer_head *bh; char *data; - if (!(n = hpfs_alloc_sector(s, fno, 1, 0, 1))) return; + if (!(n = hpfs_alloc_sector(s, fno, 1, 0))) return; if (!(data = hpfs_get_sector(s, n, &bh))) { hpfs_free_sectors(s, n, 1); return; } - memcpy(data, fnode_ea(fnode), fnode->ea_size_s); - fnode->ea_size_l = fnode->ea_size_s; - fnode->ea_size_s = 0; - fnode->ea_secno = n; - fnode->ea_anode = 0; + memcpy(data, fnode_ea(fnode), le16_to_cpu(fnode->ea_size_s)); + fnode->ea_size_l = cpu_to_le32(le16_to_cpu(fnode->ea_size_s)); + fnode->ea_size_s = cpu_to_le16(0); + fnode->ea_secno = cpu_to_le32(n); + fnode->ea_anode = cpu_to_le32(0); mark_buffer_dirty(bh); brelse(bh); } - pos = fnode->ea_size_l + 5 + strlen(key) + size; - len = (fnode->ea_size_l + 511) >> 9; + pos = le32_to_cpu(fnode->ea_size_l) + 5 + strlen(key) + size; + len = (le32_to_cpu(fnode->ea_size_l) + 511) >> 9; if (pos >= 30000) goto bail; while (((pos + 511) >> 9) > len) { if (!len) { - if (!(fnode->ea_secno = hpfs_alloc_sector(s, fno, 1, 0, 1))) - goto bail; + secno q = hpfs_alloc_sector(s, fno, 1, 0); + if (!q) goto bail; + fnode->ea_secno = cpu_to_le32(q); fnode->ea_anode = 0; len++; } else if (!fnode->ea_anode) { - if (hpfs_alloc_if_possible(s, fnode->ea_secno + len)) { + if (hpfs_alloc_if_possible(s, le32_to_cpu(fnode->ea_secno) + len)) { len++; } else { /* Aargh... don't know how to create ea anodes :-( */ @@ -298,26 +300,26 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key, anode_secno a_s; if (!(anode = hpfs_alloc_anode(s, fno, &a_s, &bh))) goto bail; - anode->up = fno; + anode->up = cpu_to_le32(fno); anode->btree.fnode_parent = 1; anode->btree.n_free_nodes--; anode->btree.n_used_nodes++; - anode->btree.first_free += 12; - anode->u.external[0].disk_secno = fnode->ea_secno; - anode->u.external[0].file_secno = 0; - anode->u.external[0].length = len; + anode->btree.first_free = cpu_to_le16(le16_to_cpu(anode->btree.first_free) + 12); + anode->u.external[0].disk_secno = cpu_to_le32(le32_to_cpu(fnode->ea_secno)); + anode->u.external[0].file_secno = cpu_to_le32(0); + anode->u.external[0].length = cpu_to_le32(len); mark_buffer_dirty(bh); brelse(bh); fnode->ea_anode = 1; - fnode->ea_secno = a_s;*/ + fnode->ea_secno = cpu_to_le32(a_s);*/ secno new_sec; int i; - if (!(new_sec = hpfs_alloc_sector(s, fno, 1, 1 - ((pos + 511) >> 9), 1))) + if (!(new_sec = hpfs_alloc_sector(s, fno, 1, 1 - ((pos + 511) >> 9)))) goto bail; for (i = 0; i < len; i++) { struct buffer_head *bh1, *bh2; void *b1, *b2; - if (!(b1 = hpfs_map_sector(s, fnode->ea_secno + i, &bh1, len - i - 1))) { + if (!(b1 = hpfs_map_sector(s, le32_to_cpu(fnode->ea_secno) + i, &bh1, len - i - 1))) { hpfs_free_sectors(s, new_sec, (pos + 511) >> 9); goto bail; } @@ -331,13 +333,13 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key, mark_buffer_dirty(bh2); brelse(bh2); } - hpfs_free_sectors(s, fnode->ea_secno, len); - fnode->ea_secno = new_sec; + hpfs_free_sectors(s, le32_to_cpu(fnode->ea_secno), len); + fnode->ea_secno = cpu_to_le32(new_sec); len = (pos + 511) >> 9; } } if (fnode->ea_anode) { - if (hpfs_add_sector_to_btree(s, fnode->ea_secno, + if (hpfs_add_sector_to_btree(s, le32_to_cpu(fnode->ea_secno), 0, len) != -1) { len++; } else { @@ -349,17 +351,17 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key, h[1] = strlen(key); h[2] = size & 0xff; h[3] = size >> 8; - if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l, 4, h)) goto bail; - if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l + 4, h[1] + 1, key)) goto bail; - if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l + 5 + h[1], size, data)) goto bail; - fnode->ea_size_l = pos; + if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l), 4, h)) goto bail; + if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l) + 4, h[1] + 1, key)) goto bail; + if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l) + 5 + h[1], size, data)) goto bail; + fnode->ea_size_l = cpu_to_le32(pos); ret: hpfs_i(inode)->i_ea_size += 5 + strlen(key) + size; return; bail: - if (fnode->ea_secno) - if (fnode->ea_anode) hpfs_truncate_btree(s, fnode->ea_secno, 1, (fnode->ea_size_l + 511) >> 9); - else hpfs_free_sectors(s, fnode->ea_secno + ((fnode->ea_size_l + 511) >> 9), len - ((fnode->ea_size_l + 511) >> 9)); - else fnode->ea_secno = fnode->ea_size_l = 0; + if (le32_to_cpu(fnode->ea_secno)) + if (fnode->ea_anode) hpfs_truncate_btree(s, le32_to_cpu(fnode->ea_secno), 1, (le32_to_cpu(fnode->ea_size_l) + 511) >> 9); + else hpfs_free_sectors(s, le32_to_cpu(fnode->ea_secno) + ((le32_to_cpu(fnode->ea_size_l) + 511) >> 9), len - ((le32_to_cpu(fnode->ea_size_l) + 511) >> 9)); + else fnode->ea_secno = fnode->ea_size_l = cpu_to_le32(0); } diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c index 9b9eb6933e43..89c500ee5213 100644 --- a/fs/hpfs/file.c +++ b/fs/hpfs/file.c @@ -20,8 +20,8 @@ static int hpfs_file_release(struct inode *inode, struct file *file) int hpfs_file_fsync(struct file *file, int datasync) { - /*return file_fsync(file, datasync);*/ - return 0; /* Don't fsync :-) */ + struct inode *inode = file->f_mapping->host; + return sync_blockdev(inode->i_sb->s_bdev); } /* @@ -48,38 +48,46 @@ static secno hpfs_bmap(struct inode *inode, unsigned file_secno) static void hpfs_truncate(struct inode *i) { if (IS_IMMUTABLE(i)) return /*-EPERM*/; - hpfs_lock(i->i_sb); + hpfs_lock_assert(i->i_sb); + hpfs_i(i)->i_n_secs = 0; i->i_blocks = 1 + ((i->i_size + 511) >> 9); hpfs_i(i)->mmu_private = i->i_size; hpfs_truncate_btree(i->i_sb, i->i_ino, 1, ((i->i_size + 511) >> 9)); hpfs_write_inode(i); hpfs_i(i)->i_n_secs = 0; - hpfs_unlock(i->i_sb); } static int hpfs_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create) { + int r; secno s; + hpfs_lock(inode->i_sb); s = hpfs_bmap(inode, iblock); if (s) { map_bh(bh_result, inode->i_sb, s); - return 0; + goto ret_0; } - if (!create) return 0; + if (!create) goto ret_0; if (iblock<<9 != hpfs_i(inode)->mmu_private) { BUG(); - return -EIO; + r = -EIO; + goto ret_r; } if ((s = hpfs_add_sector_to_btree(inode->i_sb, inode->i_ino, 1, inode->i_blocks - 1)) == -1) { hpfs_truncate_btree(inode->i_sb, inode->i_ino, 1, inode->i_blocks - 1); - return -ENOSPC; + r = -ENOSPC; + goto ret_r; } inode->i_blocks++; hpfs_i(inode)->mmu_private += 512; set_buffer_new(bh_result); map_bh(bh_result, inode->i_sb, s); - return 0; + ret_0: + r = 0; + ret_r: + hpfs_unlock(inode->i_sb); + return r; } static int hpfs_writepage(struct page *page, struct writeback_control *wbc) @@ -130,8 +138,11 @@ static ssize_t hpfs_file_write(struct file *file, const char __user *buf, ssize_t retval; retval = do_sync_write(file, buf, count, ppos); - if (retval > 0) + if (retval > 0) { + hpfs_lock(file->f_path.dentry->d_sb); hpfs_i(file->f_path.dentry->d_inode)->i_dirty = 1; + hpfs_unlock(file->f_path.dentry->d_sb); + } return retval; } diff --git a/fs/hpfs/hpfs.h b/fs/hpfs/hpfs.h index 0e84c73cd9c4..8b0650aae328 100644 --- a/fs/hpfs/hpfs.h +++ b/fs/hpfs/hpfs.h @@ -19,9 +19,13 @@ For definitive information on HPFS, ask somebody else -- this is guesswork. There are certain to be many mistakes. */ +#if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN) +#error unknown endian +#endif + /* Notation */ -typedef unsigned secno; /* sector number, partition relative */ +typedef u32 secno; /* sector number, partition relative */ typedef secno dnode_secno; /* sector number of a dnode */ typedef secno fnode_secno; /* sector number of an fnode */ @@ -38,28 +42,28 @@ typedef u32 time32_t; /* 32-bit time_t type */ struct hpfs_boot_block { - unsigned char jmp[3]; - unsigned char oem_id[8]; - unsigned char bytes_per_sector[2]; /* 512 */ - unsigned char sectors_per_cluster; - unsigned char n_reserved_sectors[2]; - unsigned char n_fats; - unsigned char n_rootdir_entries[2]; - unsigned char n_sectors_s[2]; - unsigned char media_byte; - unsigned short sectors_per_fat; - unsigned short sectors_per_track; - unsigned short heads_per_cyl; - unsigned int n_hidden_sectors; - unsigned int n_sectors_l; /* size of partition */ - unsigned char drive_number; - unsigned char mbz; - unsigned char sig_28h; /* 28h */ - unsigned char vol_serno[4]; - unsigned char vol_label[11]; - unsigned char sig_hpfs[8]; /* "HPFS " */ - unsigned char pad[448]; - unsigned short magic; /* aa55 */ + u8 jmp[3]; + u8 oem_id[8]; + u8 bytes_per_sector[2]; /* 512 */ + u8 sectors_per_cluster; + u8 n_reserved_sectors[2]; + u8 n_fats; + u8 n_rootdir_entries[2]; + u8 n_sectors_s[2]; + u8 media_byte; + u16 sectors_per_fat; + u16 sectors_per_track; + u16 heads_per_cyl; + u32 n_hidden_sectors; + u32 n_sectors_l; /* size of partition */ + u8 drive_number; + u8 mbz; + u8 sig_28h; /* 28h */ + u8 vol_serno[4]; + u8 vol_label[11]; + u8 sig_hpfs[8]; /* "HPFS " */ + u8 pad[448]; + u16 magic; /* aa55 */ }; @@ -71,31 +75,29 @@ struct hpfs_boot_block struct hpfs_super_block { - unsigned magic; /* f995 e849 */ - unsigned magic1; /* fa53 e9c5, more magic? */ - /*unsigned huh202;*/ /* ?? 202 = N. of B. in 1.00390625 S.*/ - char version; /* version of a filesystem usually 2 */ - char funcversion; /* functional version - oldest version + u32 magic; /* f995 e849 */ + u32 magic1; /* fa53 e9c5, more magic? */ + u8 version; /* version of a filesystem usually 2 */ + u8 funcversion; /* functional version - oldest version of filesystem that can understand this disk */ - unsigned short int zero; /* 0 */ + u16 zero; /* 0 */ fnode_secno root; /* fnode of root directory */ secno n_sectors; /* size of filesystem */ - unsigned n_badblocks; /* number of bad blocks */ + u32 n_badblocks; /* number of bad blocks */ secno bitmaps; /* pointers to free space bit maps */ - unsigned zero1; /* 0 */ + u32 zero1; /* 0 */ secno badblocks; /* bad block list */ - unsigned zero3; /* 0 */ + u32 zero3; /* 0 */ time32_t last_chkdsk; /* date last checked, 0 if never */ - /*unsigned zero4;*/ /* 0 */ - time32_t last_optimize; /* date last optimized, 0 if never */ + time32_t last_optimize; /* date last optimized, 0 if never */ secno n_dir_band; /* number of sectors in dir band */ secno dir_band_start; /* first sector in dir band */ secno dir_band_end; /* last sector in dir band */ secno dir_band_bitmap; /* free space map, 1 dnode per bit */ - char volume_name[32]; /* not used */ + u8 volume_name[32]; /* not used */ secno user_id_table; /* 8 preallocated sectors - user id */ - unsigned zero6[103]; /* 0 */ + u32 zero6[103]; /* 0 */ }; @@ -107,44 +109,65 @@ struct hpfs_super_block struct hpfs_spare_block { - unsigned magic; /* f991 1849 */ - unsigned magic1; /* fa52 29c5, more magic? */ - - unsigned dirty: 1; /* 0 clean, 1 "improperly stopped" */ - /*unsigned flag1234: 4;*/ /* unknown flags */ - unsigned sparedir_used: 1; /* spare dirblks used */ - unsigned hotfixes_used: 1; /* hotfixes used */ - unsigned bad_sector: 1; /* bad sector, corrupted disk (???) */ - unsigned bad_bitmap: 1; /* bad bitmap */ - unsigned fast: 1; /* partition was fast formatted */ - unsigned old_wrote: 1; /* old version wrote to partion */ - unsigned old_wrote_1: 1; /* old version wrote to partion (?) */ - unsigned install_dasd_limits: 1; /* HPFS386 flags */ - unsigned resynch_dasd_limits: 1; - unsigned dasd_limits_operational: 1; - unsigned multimedia_active: 1; - unsigned dce_acls_active: 1; - unsigned dasd_limits_dirty: 1; - unsigned flag67: 2; - unsigned char mm_contlgulty; - unsigned char unused; + u32 magic; /* f991 1849 */ + u32 magic1; /* fa52 29c5, more magic? */ + +#ifdef __LITTLE_ENDIAN + u8 dirty: 1; /* 0 clean, 1 "improperly stopped" */ + u8 sparedir_used: 1; /* spare dirblks used */ + u8 hotfixes_used: 1; /* hotfixes used */ + u8 bad_sector: 1; /* bad sector, corrupted disk (???) */ + u8 bad_bitmap: 1; /* bad bitmap */ + u8 fast: 1; /* partition was fast formatted */ + u8 old_wrote: 1; /* old version wrote to partion */ + u8 old_wrote_1: 1; /* old version wrote to partion (?) */ +#else + u8 old_wrote_1: 1; /* old version wrote to partion (?) */ + u8 old_wrote: 1; /* old version wrote to partion */ + u8 fast: 1; /* partition was fast formatted */ + u8 bad_bitmap: 1; /* bad bitmap */ + u8 bad_sector: 1; /* bad sector, corrupted disk (???) */ + u8 hotfixes_used: 1; /* hotfixes used */ + u8 sparedir_used: 1; /* spare dirblks used */ + u8 dirty: 1; /* 0 clean, 1 "improperly stopped" */ +#endif + +#ifdef __LITTLE_ENDIAN + u8 install_dasd_limits: 1; /* HPFS386 flags */ + u8 resynch_dasd_limits: 1; + u8 dasd_limits_operational: 1; + u8 multimedia_active: 1; + u8 dce_acls_active: 1; + u8 dasd_limits_dirty: 1; + u8 flag67: 2; +#else + u8 flag67: 2; + u8 dasd_limits_dirty: 1; + u8 dce_acls_active: 1; + u8 multimedia_active: 1; + u8 dasd_limits_operational: 1; + u8 resynch_dasd_limits: 1; + u8 install_dasd_limits: 1; /* HPFS386 flags */ +#endif + + u8 mm_contlgulty; + u8 unused; secno hotfix_map; /* info about remapped bad sectors */ - unsigned n_spares_used; /* number of hotfixes */ - unsigned n_spares; /* number of spares in hotfix map */ - unsigned n_dnode_spares_free; /* spare dnodes unused */ - unsigned n_dnode_spares; /* length of spare_dnodes[] list, + u32 n_spares_used; /* number of hotfixes */ + u32 n_spares; /* number of spares in hotfix map */ + u32 n_dnode_spares_free; /* spare dnodes unused */ + u32 n_dnode_spares; /* length of spare_dnodes[] list, follows in this block*/ secno code_page_dir; /* code page directory block */ - unsigned n_code_pages; /* number of code pages */ - /*unsigned large_numbers[2];*/ /* ?? */ - unsigned super_crc; /* on HPFS386 and LAN Server this is + u32 n_code_pages; /* number of code pages */ + u32 super_crc; /* on HPFS386 and LAN Server this is checksum of superblock, on normal OS/2 unused */ - unsigned spare_crc; /* on HPFS386 checksum of spareblock */ - unsigned zero1[15]; /* unused */ + u32 spare_crc; /* on HPFS386 checksum of spareblock */ + u32 zero1[15]; /* unused */ dnode_secno spare_dnodes[100]; /* emergency free dnode list */ - unsigned zero2[1]; /* room for more? */ + u32 zero2[1]; /* room for more? */ }; /* The bad block list is 4 sectors long. The first word must be zero, @@ -179,18 +202,18 @@ struct hpfs_spare_block struct code_page_directory { - unsigned magic; /* 4945 21f7 */ - unsigned n_code_pages; /* number of pointers following */ - unsigned zero1[2]; + u32 magic; /* 4945 21f7 */ + u32 n_code_pages; /* number of pointers following */ + u32 zero1[2]; struct { - unsigned short ix; /* index */ - unsigned short code_page_number; /* code page number */ - unsigned bounds; /* matches corresponding word + u16 ix; /* index */ + u16 code_page_number; /* code page number */ + u32 bounds; /* matches corresponding word in data block */ secno code_page_data; /* sector number of a code_page_data containing c.p. array */ - unsigned short index; /* index in c.p. array in that sector*/ - unsigned short unknown; /* some unknown value; usually 0; + u16 index; /* index in c.p. array in that sector*/ + u16 unknown; /* some unknown value; usually 0; 2 in Japanese version */ } array[31]; /* unknown length */ }; @@ -201,21 +224,21 @@ struct code_page_directory struct code_page_data { - unsigned magic; /* 8945 21f7 */ - unsigned n_used; /* # elements used in c_p_data[] */ - unsigned bounds[3]; /* looks a bit like + u32 magic; /* 8945 21f7 */ + u32 n_used; /* # elements used in c_p_data[] */ + u32 bounds[3]; /* looks a bit like (beg1,end1), (beg2,end2) one byte each */ - unsigned short offs[3]; /* offsets from start of sector + u16 offs[3]; /* offsets from start of sector to start of c_p_data[ix] */ struct { - unsigned short ix; /* index */ - unsigned short code_page_number; /* code page number */ - unsigned short unknown; /* the same as in cp directory */ - unsigned char map[128]; /* upcase table for chars 80..ff */ - unsigned short zero2; + u16 ix; /* index */ + u16 code_page_number; /* code page number */ + u16 unknown; /* the same as in cp directory */ + u8 map[128]; /* upcase table for chars 80..ff */ + u16 zero2; } code_page[3]; - unsigned char incognita[78]; + u8 incognita[78]; }; @@ -255,50 +278,84 @@ struct code_page_data #define DNODE_MAGIC 0x77e40aae struct dnode { - unsigned magic; /* 77e4 0aae */ - unsigned first_free; /* offset from start of dnode to + u32 magic; /* 77e4 0aae */ + u32 first_free; /* offset from start of dnode to first free dir entry */ - unsigned root_dnode:1; /* Is it root dnode? */ - unsigned increment_me:31; /* some kind of activity counter? - Neither HPFS.IFS nor CHKDSK cares +#ifdef __LITTLE_ENDIAN + u8 root_dnode: 1; /* Is it root dnode? */ + u8 increment_me: 7; /* some kind of activity counter? */ + /* Neither HPFS.IFS nor CHKDSK cares + if you change this word */ +#else + u8 increment_me: 7; /* some kind of activity counter? */ + /* Neither HPFS.IFS nor CHKDSK cares if you change this word */ + u8 root_dnode: 1; /* Is it root dnode? */ +#endif + u8 increment_me2[3]; secno up; /* (root dnode) directory's fnode (nonroot) parent dnode */ dnode_secno self; /* pointer to this dnode */ - unsigned char dirent[2028]; /* one or more dirents */ + u8 dirent[2028]; /* one or more dirents */ }; struct hpfs_dirent { - unsigned short length; /* offset to next dirent */ - unsigned first: 1; /* set on phony ^A^A (".") entry */ - unsigned has_acl: 1; - unsigned down: 1; /* down pointer present (after name) */ - unsigned last: 1; /* set on phony \377 entry */ - unsigned has_ea: 1; /* entry has EA */ - unsigned has_xtd_perm: 1; /* has extended perm list (???) */ - unsigned has_explicit_acl: 1; - unsigned has_needea: 1; /* ?? some EA has NEEDEA set + u16 length; /* offset to next dirent */ + +#ifdef __LITTLE_ENDIAN + u8 first: 1; /* set on phony ^A^A (".") entry */ + u8 has_acl: 1; + u8 down: 1; /* down pointer present (after name) */ + u8 last: 1; /* set on phony \377 entry */ + u8 has_ea: 1; /* entry has EA */ + u8 has_xtd_perm: 1; /* has extended perm list (???) */ + u8 has_explicit_acl: 1; + u8 has_needea: 1; /* ?? some EA has NEEDEA set + I have no idea why this is + interesting in a dir entry */ +#else + u8 has_needea: 1; /* ?? some EA has NEEDEA set I have no idea why this is interesting in a dir entry */ - unsigned read_only: 1; /* dos attrib */ - unsigned hidden: 1; /* dos attrib */ - unsigned system: 1; /* dos attrib */ - unsigned flag11: 1; /* would be volume label dos attrib */ - unsigned directory: 1; /* dos attrib */ - unsigned archive: 1; /* dos attrib */ - unsigned not_8x3: 1; /* name is not 8.3 */ - unsigned flag15: 1; + u8 has_explicit_acl: 1; + u8 has_xtd_perm: 1; /* has extended perm list (???) */ + u8 has_ea: 1; /* entry has EA */ + u8 last: 1; /* set on phony \377 entry */ + u8 down: 1; /* down pointer present (after name) */ + u8 has_acl: 1; + u8 first: 1; /* set on phony ^A^A (".") entry */ +#endif + +#ifdef __LITTLE_ENDIAN + u8 read_only: 1; /* dos attrib */ + u8 hidden: 1; /* dos attrib */ + u8 system: 1; /* dos attrib */ + u8 flag11: 1; /* would be volume label dos attrib */ + u8 directory: 1; /* dos attrib */ + u8 archive: 1; /* dos attrib */ + u8 not_8x3: 1; /* name is not 8.3 */ + u8 flag15: 1; +#else + u8 flag15: 1; + u8 not_8x3: 1; /* name is not 8.3 */ + u8 archive: 1; /* dos attrib */ + u8 directory: 1; /* dos attrib */ + u8 flag11: 1; /* would be volume label dos attrib */ + u8 system: 1; /* dos attrib */ + u8 hidden: 1; /* dos attrib */ + u8 read_only: 1; /* dos attrib */ +#endif + fnode_secno fnode; /* fnode giving allocation info */ time32_t write_date; /* mtime */ - unsigned file_size; /* file length, bytes */ + u32 file_size; /* file length, bytes */ time32_t read_date; /* atime */ time32_t creation_date; /* ctime */ - unsigned ea_size; /* total EA length, bytes */ - unsigned char no_of_acls : 3; /* number of ACL's */ - unsigned char reserver : 5; - unsigned char ix; /* code page index (of filename), see + u32 ea_size; /* total EA length, bytes */ + u8 no_of_acls; /* number of ACL's (low 3 bits) */ + u8 ix; /* code page index (of filename), see struct code_page_data */ - unsigned char namelen, name[1]; /* file name */ + u8 namelen, name[1]; /* file name */ /* dnode_secno down; btree down pointer, if present, follows name on next word boundary, or maybe it precedes next dirent, which is on a word boundary. */ @@ -318,38 +375,50 @@ struct hpfs_dirent { struct bplus_leaf_node { - unsigned file_secno; /* first file sector in extent */ - unsigned length; /* length, sectors */ + u32 file_secno; /* first file sector in extent */ + u32 length; /* length, sectors */ secno disk_secno; /* first corresponding disk sector */ }; struct bplus_internal_node { - unsigned file_secno; /* subtree maps sectors < this */ + u32 file_secno; /* subtree maps sectors < this */ anode_secno down; /* pointer to subtree */ }; struct bplus_header { - unsigned hbff: 1; /* high bit of first free entry offset */ - unsigned flag1: 1; - unsigned flag2: 1; - unsigned flag3: 1; - unsigned flag4: 1; - unsigned fnode_parent: 1; /* ? we're pointed to by an fnode, +#ifdef __LITTLE_ENDIAN + u8 hbff: 1; /* high bit of first free entry offset */ + u8 flag1234: 4; + u8 fnode_parent: 1; /* ? we're pointed to by an fnode, the data btree or some ea or the main ea bootage pointer ea_secno */ /* also can get set in fnodes, which may be a chkdsk glitch or may mean this bit is irrelevant in fnodes, or this interpretation is all wet */ - unsigned binary_search: 1; /* suggest binary search (unused) */ - unsigned internal: 1; /* 1 -> (internal) tree of anodes + u8 binary_search: 1; /* suggest binary search (unused) */ + u8 internal: 1; /* 1 -> (internal) tree of anodes + 0 -> (leaf) list of extents */ +#else + u8 internal: 1; /* 1 -> (internal) tree of anodes 0 -> (leaf) list of extents */ - unsigned char fill[3]; - unsigned char n_free_nodes; /* free nodes in following array */ - unsigned char n_used_nodes; /* used nodes in following array */ - unsigned short first_free; /* offset from start of header to + u8 binary_search: 1; /* suggest binary search (unused) */ + u8 fnode_parent: 1; /* ? we're pointed to by an fnode, + the data btree or some ea or the + main ea bootage pointer ea_secno */ + /* also can get set in fnodes, which + may be a chkdsk glitch or may mean + this bit is irrelevant in fnodes, + or this interpretation is all wet */ + u8 flag1234: 4; + u8 hbff: 1; /* high bit of first free entry offset */ +#endif + u8 fill[3]; + u8 n_free_nodes; /* free nodes in following array */ + u8 n_used_nodes; /* used nodes in following array */ + u16 first_free; /* offset from start of header to first free node in array */ union { struct bplus_internal_node internal[0]; /* (internal) 2-word entries giving @@ -369,37 +438,38 @@ struct bplus_header struct fnode { - unsigned magic; /* f7e4 0aae */ - unsigned zero1[2]; /* read history */ - unsigned char len, name[15]; /* true length, truncated name */ + u32 magic; /* f7e4 0aae */ + u32 zero1[2]; /* read history */ + u8 len, name[15]; /* true length, truncated name */ fnode_secno up; /* pointer to file's directory fnode */ - /*unsigned zero2[3];*/ secno acl_size_l; secno acl_secno; - unsigned short acl_size_s; - char acl_anode; - char zero2; /* history bit count */ - unsigned ea_size_l; /* length of disk-resident ea's */ + u16 acl_size_s; + u8 acl_anode; + u8 zero2; /* history bit count */ + u32 ea_size_l; /* length of disk-resident ea's */ secno ea_secno; /* first sector of disk-resident ea's*/ - unsigned short ea_size_s; /* length of fnode-resident ea's */ - - unsigned flag0: 1; - unsigned ea_anode: 1; /* 1 -> ea_secno is an anode */ - unsigned flag2: 1; - unsigned flag3: 1; - unsigned flag4: 1; - unsigned flag5: 1; - unsigned flag6: 1; - unsigned flag7: 1; - unsigned dirflag: 1; /* 1 -> directory. first & only extent + u16 ea_size_s; /* length of fnode-resident ea's */ + +#ifdef __LITTLE_ENDIAN + u8 flag0: 1; + u8 ea_anode: 1; /* 1 -> ea_secno is an anode */ + u8 flag234567: 6; +#else + u8 flag234567: 6; + u8 ea_anode: 1; /* 1 -> ea_secno is an anode */ + u8 flag0: 1; +#endif + +#ifdef __LITTLE_ENDIAN + u8 dirflag: 1; /* 1 -> directory. first & only extent points to dnode. */ - unsigned flag9: 1; - unsigned flag10: 1; - unsigned flag11: 1; - unsigned flag12: 1; - unsigned flag13: 1; - unsigned flag14: 1; - unsigned flag15: 1; + u8 flag9012345: 7; +#else + u8 flag9012345: 7; + u8 dirflag: 1; /* 1 -> directory. first & only extent + points to dnode. */ +#endif struct bplus_header btree; /* b+ tree, 8 extents or 12 subtrees */ union { @@ -407,17 +477,16 @@ struct fnode struct bplus_internal_node internal[12]; } u; - unsigned file_size; /* file length, bytes */ - unsigned n_needea; /* number of EA's with NEEDEA set */ - char user_id[16]; /* unused */ - unsigned short ea_offs; /* offset from start of fnode + u32 file_size; /* file length, bytes */ + u32 n_needea; /* number of EA's with NEEDEA set */ + u8 user_id[16]; /* unused */ + u16 ea_offs; /* offset from start of fnode to first fnode-resident ea */ - char dasd_limit_treshhold; - char dasd_limit_delta; - unsigned dasd_limit; - unsigned dasd_usage; - /*unsigned zero5[2];*/ - unsigned char ea[316]; /* zero or more EA's, packed together + u8 dasd_limit_treshhold; + u8 dasd_limit_delta; + u32 dasd_limit; + u32 dasd_usage; + u8 ea[316]; /* zero or more EA's, packed together with no alignment padding. (Do not use this name, get here via fnode + ea_offs. I think.) */ @@ -430,7 +499,7 @@ struct fnode struct anode { - unsigned magic; /* 37e4 0aae */ + u32 magic; /* 37e4 0aae */ anode_secno self; /* pointer to this anode */ secno up; /* parent anode or fnode */ @@ -440,7 +509,7 @@ struct anode struct bplus_internal_node internal[60]; } u; - unsigned fill[3]; /* unused */ + u32 fill[3]; /* unused */ }; @@ -461,25 +530,31 @@ struct anode struct extended_attribute { - unsigned indirect: 1; /* 1 -> value gives sector number +#ifdef __LITTLE_ENDIAN + u8 indirect: 1; /* 1 -> value gives sector number where real value starts */ - unsigned anode: 1; /* 1 -> sector is an anode + u8 anode: 1; /* 1 -> sector is an anode + that points to fragmented value */ + u8 flag23456: 5; + u8 needea: 1; /* required ea */ +#else + u8 needea: 1; /* required ea */ + u8 flag23456: 5; + u8 anode: 1; /* 1 -> sector is an anode that points to fragmented value */ - unsigned flag2: 1; - unsigned flag3: 1; - unsigned flag4: 1; - unsigned flag5: 1; - unsigned flag6: 1; - unsigned needea: 1; /* required ea */ - unsigned char namelen; /* length of name, bytes */ - unsigned short valuelen; /* length of value, bytes */ - unsigned char name[0]; + u8 indirect: 1; /* 1 -> value gives sector number + where real value starts */ +#endif + u8 namelen; /* length of name, bytes */ + u8 valuelen_lo; /* length of value, bytes */ + u8 valuelen_hi; /* length of value, bytes */ + u8 name[0]; /* - unsigned char name[namelen]; ascii attrib name - unsigned char nul; terminating '\0', not counted - unsigned char value[valuelen]; value, arbitrary + u8 name[namelen]; ascii attrib name + u8 nul; terminating '\0', not counted + u8 value[valuelen]; value, arbitrary if this.indirect, valuelen is 8 and the value is - unsigned length; real length of value, bytes + u32 length; real length of value, bytes secno secno; sector address where it starts if this.anode, the above sector number is the root of an anode tree which points to the value. diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h index c15adbca07ff..dd552f862c8f 100644 --- a/fs/hpfs/hpfs_fn.h +++ b/fs/hpfs/hpfs_fn.h @@ -13,6 +13,7 @@ #include <linux/pagemap.h> #include <linux/buffer_head.h> #include <linux/slab.h> +#include <asm/unaligned.h> #include "hpfs.h" @@ -51,18 +52,16 @@ struct hpfs_inode_info { unsigned i_disk_sec; /* (files) minimalist cache of alloc info */ unsigned i_n_secs; /* (files) minimalist cache of alloc info */ unsigned i_ea_size; /* size of extended attributes */ - unsigned i_conv : 2; /* (files) crlf->newline hackery */ unsigned i_ea_mode : 1; /* file's permission is stored in ea */ unsigned i_ea_uid : 1; /* file's uid is stored in ea */ unsigned i_ea_gid : 1; /* file's gid is stored in ea */ unsigned i_dirty : 1; - struct mutex i_mutex; - struct mutex i_parent_mutex; loff_t **i_rddir_off; struct inode vfs_inode; }; struct hpfs_sb_info { + struct mutex hpfs_mutex; /* global hpfs lock */ ino_t sb_root; /* inode number of root dir */ unsigned sb_fs_size; /* file system size, sectors */ unsigned sb_bitmaps; /* sector number of bitmap list */ @@ -74,7 +73,6 @@ struct hpfs_sb_info { uid_t sb_uid; /* uid from mount options */ gid_t sb_gid; /* gid from mount options */ umode_t sb_mode; /* mode from mount options */ - unsigned sb_conv : 2; /* crlf->newline hackery */ unsigned sb_eas : 2; /* eas: 0-ignore, 1-ro, 2-rw */ unsigned sb_err : 2; /* on errs: 0-cont, 1-ro, 2-panic */ unsigned sb_chk : 2; /* checks: 0-no, 1-normal, 2-strict */ @@ -87,20 +85,9 @@ struct hpfs_sb_info { unsigned *sb_bmp_dir; /* main bitmap directory */ unsigned sb_c_bitmap; /* current bitmap */ unsigned sb_max_fwd_alloc; /* max forwad allocation */ - struct mutex hpfs_creation_de; /* when creating dirents, nobody else - can alloc blocks */ - /*unsigned sb_mounting : 1;*/ int sb_timeshift; }; -/* - * conv= options - */ - -#define CONV_BINARY 0 /* no conversion */ -#define CONV_TEXT 1 /* crlf->newline */ -#define CONV_AUTO 2 /* decide based on file contents */ - /* Four 512-byte buffers and the 2k block obtained by concatenating them */ struct quad_buffer_head { @@ -113,7 +100,7 @@ struct quad_buffer_head { static inline dnode_secno de_down_pointer (struct hpfs_dirent *de) { CHKCOND(de->down,("HPFS: de_down_pointer: !de->down\n")); - return *(dnode_secno *) ((void *) de + de->length - 4); + return le32_to_cpu(*(dnode_secno *) ((void *) de + le16_to_cpu(de->length) - 4)); } /* The first dir entry in a dnode */ @@ -127,41 +114,46 @@ static inline struct hpfs_dirent *dnode_first_de (struct dnode *dnode) static inline struct hpfs_dirent *dnode_end_de (struct dnode *dnode) { - CHKCOND(dnode->first_free>=0x14 && dnode->first_free<=0xa00,("HPFS: dnode_end_de: dnode->first_free = %d\n",(int)dnode->first_free)); - return (void *) dnode + dnode->first_free; + CHKCOND(le32_to_cpu(dnode->first_free)>=0x14 && le32_to_cpu(dnode->first_free)<=0xa00,("HPFS: dnode_end_de: dnode->first_free = %x\n",(unsigned)le32_to_cpu(dnode->first_free))); + return (void *) dnode + le32_to_cpu(dnode->first_free); } /* The dir entry after dir entry de */ static inline struct hpfs_dirent *de_next_de (struct hpfs_dirent *de) { - CHKCOND(de->length>=0x20 && de->length<0x800,("HPFS: de_next_de: de->length = %d\n",(int)de->length)); - return (void *) de + de->length; + CHKCOND(le16_to_cpu(de->length)>=0x20 && le16_to_cpu(de->length)<0x800,("HPFS: de_next_de: de->length = %x\n",(unsigned)le16_to_cpu(de->length))); + return (void *) de + le16_to_cpu(de->length); } static inline struct extended_attribute *fnode_ea(struct fnode *fnode) { - return (struct extended_attribute *)((char *)fnode + fnode->ea_offs + fnode->acl_size_s); + return (struct extended_attribute *)((char *)fnode + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s)); } static inline struct extended_attribute *fnode_end_ea(struct fnode *fnode) { - return (struct extended_attribute *)((char *)fnode + fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s); + return (struct extended_attribute *)((char *)fnode + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s)); +} + +static unsigned ea_valuelen(struct extended_attribute *ea) +{ + return ea->valuelen_lo + 256 * ea->valuelen_hi; } static inline struct extended_attribute *next_ea(struct extended_attribute *ea) { - return (struct extended_attribute *)((char *)ea + 5 + ea->namelen + ea->valuelen); + return (struct extended_attribute *)((char *)ea + 5 + ea->namelen + ea_valuelen(ea)); } static inline secno ea_sec(struct extended_attribute *ea) { - return *(secno *)((char *)ea + 9 + ea->namelen); + return le32_to_cpu(get_unaligned((secno *)((char *)ea + 9 + ea->namelen))); } static inline secno ea_len(struct extended_attribute *ea) { - return *(secno *)((char *)ea + 5 + ea->namelen); + return le32_to_cpu(get_unaligned((secno *)((char *)ea + 5 + ea->namelen))); } static inline char *ea_data(struct extended_attribute *ea) @@ -186,13 +178,13 @@ static inline void copy_de(struct hpfs_dirent *dst, struct hpfs_dirent *src) dst->not_8x3 = n; } -static inline unsigned tstbits(unsigned *bmp, unsigned b, unsigned n) +static inline unsigned tstbits(u32 *bmp, unsigned b, unsigned n) { int i; if ((b >= 0x4000) || (b + n - 1 >= 0x4000)) return n; - if (!((bmp[(b & 0x3fff) >> 5] >> (b & 0x1f)) & 1)) return 1; + if (!((le32_to_cpu(bmp[(b & 0x3fff) >> 5]) >> (b & 0x1f)) & 1)) return 1; for (i = 1; i < n; i++) - if (/*b+i < 0x4000 &&*/ !((bmp[((b+i) & 0x3fff) >> 5] >> ((b+i) & 0x1f)) & 1)) + if (!((le32_to_cpu(bmp[((b+i) & 0x3fff) >> 5]) >> ((b+i) & 0x1f)) & 1)) return i + 1; return 0; } @@ -200,12 +192,12 @@ static inline unsigned tstbits(unsigned *bmp, unsigned b, unsigned n) /* alloc.c */ int hpfs_chk_sectors(struct super_block *, secno, int, char *); -secno hpfs_alloc_sector(struct super_block *, secno, unsigned, int, int); +secno hpfs_alloc_sector(struct super_block *, secno, unsigned, int); int hpfs_alloc_if_possible(struct super_block *, secno); void hpfs_free_sectors(struct super_block *, secno, unsigned); int hpfs_check_free_dnodes(struct super_block *, int); void hpfs_free_dnode(struct super_block *, secno); -struct dnode *hpfs_alloc_dnode(struct super_block *, secno, dnode_secno *, struct quad_buffer_head *, int); +struct dnode *hpfs_alloc_dnode(struct super_block *, secno, dnode_secno *, struct quad_buffer_head *); struct fnode *hpfs_alloc_fnode(struct super_block *, secno, fnode_secno *, struct buffer_head **); struct anode *hpfs_alloc_anode(struct super_block *, secno, anode_secno *, struct buffer_head **); @@ -222,8 +214,6 @@ void hpfs_remove_fnode(struct super_block *, fnode_secno fno); /* buffer.c */ -void hpfs_lock_creation(struct super_block *); -void hpfs_unlock_creation(struct super_block *); void *hpfs_map_sector(struct super_block *, unsigned, struct buffer_head **, int); void *hpfs_get_sector(struct super_block *, unsigned, struct buffer_head **); void *hpfs_map_4sectors(struct super_block *, unsigned, struct quad_buffer_head *, int); @@ -247,7 +237,7 @@ void hpfs_del_pos(struct inode *, loff_t *); struct hpfs_dirent *hpfs_add_de(struct super_block *, struct dnode *, const unsigned char *, unsigned, secno); int hpfs_add_dirent(struct inode *, const unsigned char *, unsigned, - struct hpfs_dirent *, int); + struct hpfs_dirent *); int hpfs_remove_dirent(struct inode *, dnode_secno, struct hpfs_dirent *, struct quad_buffer_head *, int); void hpfs_count_dnodes(struct super_block *, dnode_secno, int *, int *, int *); dnode_secno hpfs_de_as_down_as_possible(struct super_block *, dnode_secno dno); @@ -303,7 +293,6 @@ int hpfs_compare_names(struct super_block *, const unsigned char *, unsigned, const unsigned char *, unsigned, int); int hpfs_is_name_long(const unsigned char *, unsigned); void hpfs_adjust_length(const unsigned char *, unsigned *); -void hpfs_decide_conv(struct inode *, const unsigned char *, unsigned); /* namei.c */ @@ -346,21 +335,26 @@ static inline time32_t gmt_to_local(struct super_block *s, time_t t) /* * Locking: * - * hpfs_lock() is a leftover from the big kernel lock. - * Right now, these functions are empty and only left - * for documentation purposes. The file system no longer - * works on SMP systems, so the lock is not needed - * any more. + * hpfs_lock() locks the whole filesystem. It must be taken + * on any method called by the VFS. * - * If someone is interested in making it work again, this - * would be the place to start by adding a per-superblock - * mutex and fixing all the bugs and performance issues - * caused by that. + * We don't do any per-file locking anymore, it is hard to + * review and HPFS is not performance-sensitive anyway. */ static inline void hpfs_lock(struct super_block *s) { + struct hpfs_sb_info *sbi = hpfs_sb(s); + mutex_lock(&sbi->hpfs_mutex); } static inline void hpfs_unlock(struct super_block *s) { + struct hpfs_sb_info *sbi = hpfs_sb(s); + mutex_unlock(&sbi->hpfs_mutex); +} + +static inline void hpfs_lock_assert(struct super_block *s) +{ + struct hpfs_sb_info *sbi = hpfs_sb(s); + WARN_ON(!mutex_is_locked(&sbi->hpfs_mutex)); } diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c index 87f1f787e767..338cd8368451 100644 --- a/fs/hpfs/inode.c +++ b/fs/hpfs/inode.c @@ -17,7 +17,6 @@ void hpfs_init_inode(struct inode *i) i->i_uid = hpfs_sb(sb)->sb_uid; i->i_gid = hpfs_sb(sb)->sb_gid; i->i_mode = hpfs_sb(sb)->sb_mode; - hpfs_inode->i_conv = hpfs_sb(sb)->sb_conv; i->i_size = -1; i->i_blocks = -1; @@ -116,8 +115,8 @@ void hpfs_read_inode(struct inode *i) i->i_mode |= S_IFDIR; i->i_op = &hpfs_dir_iops; i->i_fop = &hpfs_dir_ops; - hpfs_inode->i_parent_dir = fnode->up; - hpfs_inode->i_dno = fnode->u.external[0].disk_secno; + hpfs_inode->i_parent_dir = le32_to_cpu(fnode->up); + hpfs_inode->i_dno = le32_to_cpu(fnode->u.external[0].disk_secno); if (hpfs_sb(sb)->sb_chk >= 2) { struct buffer_head *bh0; if (hpfs_map_fnode(sb, hpfs_inode->i_parent_dir, &bh0)) brelse(bh0); @@ -133,7 +132,7 @@ void hpfs_read_inode(struct inode *i) i->i_op = &hpfs_file_iops; i->i_fop = &hpfs_file_ops; i->i_nlink = 1; - i->i_size = fnode->file_size; + i->i_size = le32_to_cpu(fnode->file_size); i->i_blocks = ((i->i_size + 511) >> 9) + 1; i->i_data.a_ops = &hpfs_aops; hpfs_i(i)->mmu_private = i->i_size; @@ -144,7 +143,7 @@ void hpfs_read_inode(struct inode *i) static void hpfs_write_inode_ea(struct inode *i, struct fnode *fnode) { struct hpfs_inode_info *hpfs_inode = hpfs_i(i); - /*if (fnode->acl_size_l || fnode->acl_size_s) { + /*if (le32_to_cpu(fnode->acl_size_l) || le16_to_cpu(fnode->acl_size_s)) { Some unknown structures like ACL may be in fnode, we'd better not overwrite them hpfs_error(i->i_sb, "fnode %08x has some unknown HPFS386 stuctures", i->i_ino); @@ -187,9 +186,7 @@ void hpfs_write_inode(struct inode *i) kfree(hpfs_inode->i_rddir_off); hpfs_inode->i_rddir_off = NULL; } - mutex_lock(&hpfs_inode->i_parent_mutex); if (!i->i_nlink) { - mutex_unlock(&hpfs_inode->i_parent_mutex); return; } parent = iget_locked(i->i_sb, hpfs_inode->i_parent_dir); @@ -200,14 +197,9 @@ void hpfs_write_inode(struct inode *i) hpfs_read_inode(parent); unlock_new_inode(parent); } - mutex_lock(&hpfs_inode->i_mutex); hpfs_write_inode_nolock(i); - mutex_unlock(&hpfs_inode->i_mutex); iput(parent); - } else { - mark_inode_dirty(i); } - mutex_unlock(&hpfs_inode->i_parent_mutex); } void hpfs_write_inode_nolock(struct inode *i) @@ -226,30 +218,30 @@ void hpfs_write_inode_nolock(struct inode *i) } } else de = NULL; if (S_ISREG(i->i_mode)) { - fnode->file_size = i->i_size; - if (de) de->file_size = i->i_size; + fnode->file_size = cpu_to_le32(i->i_size); + if (de) de->file_size = cpu_to_le32(i->i_size); } else if (S_ISDIR(i->i_mode)) { - fnode->file_size = 0; - if (de) de->file_size = 0; + fnode->file_size = cpu_to_le32(0); + if (de) de->file_size = cpu_to_le32(0); } hpfs_write_inode_ea(i, fnode); if (de) { - de->write_date = gmt_to_local(i->i_sb, i->i_mtime.tv_sec); - de->read_date = gmt_to_local(i->i_sb, i->i_atime.tv_sec); - de->creation_date = gmt_to_local(i->i_sb, i->i_ctime.tv_sec); + de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec)); + de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec)); + de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec)); de->read_only = !(i->i_mode & 0222); - de->ea_size = hpfs_inode->i_ea_size; + de->ea_size = cpu_to_le32(hpfs_inode->i_ea_size); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); } if (S_ISDIR(i->i_mode)) { if ((de = map_dirent(i, hpfs_inode->i_dno, "\001\001", 2, NULL, &qbh))) { - de->write_date = gmt_to_local(i->i_sb, i->i_mtime.tv_sec); - de->read_date = gmt_to_local(i->i_sb, i->i_atime.tv_sec); - de->creation_date = gmt_to_local(i->i_sb, i->i_ctime.tv_sec); + de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec)); + de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec)); + de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec)); de->read_only = !(i->i_mode & 0222); - de->ea_size = /*hpfs_inode->i_ea_size*/0; - de->file_size = 0; + de->ea_size = cpu_to_le32(/*hpfs_inode->i_ea_size*/0); + de->file_size = cpu_to_le32(0); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); } else @@ -269,6 +261,10 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr) hpfs_lock(inode->i_sb); if (inode->i_ino == hpfs_sb(inode->i_sb)->sb_root) goto out_unlock; + if ((attr->ia_valid & ATTR_UID) && attr->ia_uid >= 0x10000) + goto out_unlock; + if ((attr->ia_valid & ATTR_GID) && attr->ia_gid >= 0x10000) + goto out_unlock; if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size > inode->i_size) goto out_unlock; @@ -284,7 +280,6 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr) } setattr_copy(inode, attr); - mark_inode_dirty(inode); hpfs_write_inode(inode); diff --git a/fs/hpfs/map.c b/fs/hpfs/map.c index 840d033ecee8..a790821366a7 100644 --- a/fs/hpfs/map.c +++ b/fs/hpfs/map.c @@ -21,7 +21,7 @@ unsigned int *hpfs_map_bitmap(struct super_block *s, unsigned bmp_block, hpfs_error(s, "hpfs_map_bitmap called with bad parameter: %08x at %s", bmp_block, id); return NULL; } - sec = hpfs_sb(s)->sb_bmp_dir[bmp_block]; + sec = le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[bmp_block]); if (!sec || sec > hpfs_sb(s)->sb_fs_size-4) { hpfs_error(s, "invalid bitmap block pointer %08x -> %08x at %s", bmp_block, sec, id); return NULL; @@ -46,18 +46,18 @@ unsigned char *hpfs_load_code_page(struct super_block *s, secno cps) struct code_page_data *cpd; struct code_page_directory *cp = hpfs_map_sector(s, cps, &bh, 0); if (!cp) return NULL; - if (cp->magic != CP_DIR_MAGIC) { - printk("HPFS: Code page directory magic doesn't match (magic = %08x)\n", cp->magic); + if (le32_to_cpu(cp->magic) != CP_DIR_MAGIC) { + printk("HPFS: Code page directory magic doesn't match (magic = %08x)\n", le32_to_cpu(cp->magic)); brelse(bh); return NULL; } - if (!cp->n_code_pages) { + if (!le32_to_cpu(cp->n_code_pages)) { printk("HPFS: n_code_pages == 0\n"); brelse(bh); return NULL; } - cpds = cp->array[0].code_page_data; - cpi = cp->array[0].index; + cpds = le32_to_cpu(cp->array[0].code_page_data); + cpi = le16_to_cpu(cp->array[0].index); brelse(bh); if (cpi >= 3) { @@ -66,12 +66,12 @@ unsigned char *hpfs_load_code_page(struct super_block *s, secno cps) } if (!(cpd = hpfs_map_sector(s, cpds, &bh, 0))) return NULL; - if ((unsigned)cpd->offs[cpi] > 0x178) { + if (le16_to_cpu(cpd->offs[cpi]) > 0x178) { printk("HPFS: Code page index out of sector\n"); brelse(bh); return NULL; } - ptr = (unsigned char *)cpd + cpd->offs[cpi] + 6; + ptr = (unsigned char *)cpd + le16_to_cpu(cpd->offs[cpi]) + 6; if (!(cp_table = kmalloc(256, GFP_KERNEL))) { printk("HPFS: out of memory for code page table\n"); brelse(bh); @@ -125,7 +125,7 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea if (hpfs_sb(s)->sb_chk) { struct extended_attribute *ea; struct extended_attribute *ea_end; - if (fnode->magic != FNODE_MAGIC) { + if (le32_to_cpu(fnode->magic) != FNODE_MAGIC) { hpfs_error(s, "bad magic on fnode %08lx", (unsigned long)ino); goto bail; @@ -138,7 +138,7 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea (unsigned long)ino); goto bail; } - if (fnode->btree.first_free != + if (le16_to_cpu(fnode->btree.first_free) != 8 + fnode->btree.n_used_nodes * (fnode->btree.internal ? 8 : 12)) { hpfs_error(s, "bad first_free pointer in fnode %08lx", @@ -146,12 +146,12 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea goto bail; } } - if (fnode->ea_size_s && ((signed int)fnode->ea_offs < 0xc4 || - (signed int)fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200)) { + if (le16_to_cpu(fnode->ea_size_s) && (le16_to_cpu(fnode->ea_offs) < 0xc4 || + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) > 0x200)) { hpfs_error(s, "bad EA info in fnode %08lx: ea_offs == %04x ea_size_s == %04x", (unsigned long)ino, - fnode->ea_offs, fnode->ea_size_s); + le16_to_cpu(fnode->ea_offs), le16_to_cpu(fnode->ea_size_s)); goto bail; } ea = fnode_ea(fnode); @@ -178,16 +178,20 @@ struct anode *hpfs_map_anode(struct super_block *s, anode_secno ano, struct buff if (hpfs_sb(s)->sb_chk) if (hpfs_chk_sectors(s, ano, 1, "anode")) return NULL; if ((anode = hpfs_map_sector(s, ano, bhp, ANODE_RD_AHEAD))) if (hpfs_sb(s)->sb_chk) { - if (anode->magic != ANODE_MAGIC || anode->self != ano) { + if (le32_to_cpu(anode->magic) != ANODE_MAGIC) { hpfs_error(s, "bad magic on anode %08x", ano); goto bail; } + if (le32_to_cpu(anode->self) != ano) { + hpfs_error(s, "self pointer invalid on anode %08x", ano); + goto bail; + } if ((unsigned)anode->btree.n_used_nodes + (unsigned)anode->btree.n_free_nodes != (anode->btree.internal ? 60 : 40)) { hpfs_error(s, "bad number of nodes in anode %08x", ano); goto bail; } - if (anode->btree.first_free != + if (le16_to_cpu(anode->btree.first_free) != 8 + anode->btree.n_used_nodes * (anode->btree.internal ? 8 : 12)) { hpfs_error(s, "bad first_free pointer in anode %08x", ano); goto bail; @@ -219,26 +223,26 @@ struct dnode *hpfs_map_dnode(struct super_block *s, unsigned secno, unsigned p, pp = 0; unsigned char *d = (unsigned char *)dnode; int b = 0; - if (dnode->magic != DNODE_MAGIC) { + if (le32_to_cpu(dnode->magic) != DNODE_MAGIC) { hpfs_error(s, "bad magic on dnode %08x", secno); goto bail; } - if (dnode->self != secno) - hpfs_error(s, "bad self pointer on dnode %08x self = %08x", secno, dnode->self); + if (le32_to_cpu(dnode->self) != secno) + hpfs_error(s, "bad self pointer on dnode %08x self = %08x", secno, le32_to_cpu(dnode->self)); /* Check dirents - bad dirents would cause infinite loops or shooting to memory */ - if (dnode->first_free > 2048/* || dnode->first_free < 84*/) { - hpfs_error(s, "dnode %08x has first_free == %08x", secno, dnode->first_free); + if (le32_to_cpu(dnode->first_free) > 2048) { + hpfs_error(s, "dnode %08x has first_free == %08x", secno, le32_to_cpu(dnode->first_free)); goto bail; } - for (p = 20; p < dnode->first_free; p += d[p] + (d[p+1] << 8)) { + for (p = 20; p < le32_to_cpu(dnode->first_free); p += d[p] + (d[p+1] << 8)) { struct hpfs_dirent *de = (struct hpfs_dirent *)((char *)dnode + p); - if (de->length > 292 || (de->length < 32) || (de->length & 3) || p + de->length > 2048) { + if (le16_to_cpu(de->length) > 292 || (le16_to_cpu(de->length) < 32) || (le16_to_cpu(de->length) & 3) || p + le16_to_cpu(de->length) > 2048) { hpfs_error(s, "bad dirent size in dnode %08x, dirent %03x, last %03x", secno, p, pp); goto bail; } - if (((31 + de->namelen + de->down*4 + 3) & ~3) != de->length) { - if (((31 + de->namelen + de->down*4 + 3) & ~3) < de->length && s->s_flags & MS_RDONLY) goto ok; + if (((31 + de->namelen + de->down*4 + 3) & ~3) != le16_to_cpu(de->length)) { + if (((31 + de->namelen + de->down*4 + 3) & ~3) < le16_to_cpu(de->length) && s->s_flags & MS_RDONLY) goto ok; hpfs_error(s, "namelen does not match dirent size in dnode %08x, dirent %03x, last %03x", secno, p, pp); goto bail; } @@ -251,7 +255,7 @@ struct dnode *hpfs_map_dnode(struct super_block *s, unsigned secno, pp = p; } - if (p != dnode->first_free) { + if (p != le32_to_cpu(dnode->first_free)) { hpfs_error(s, "size on last dirent does not match first_free; dnode %08x", secno); goto bail; } @@ -277,7 +281,7 @@ dnode_secno hpfs_fnode_dno(struct super_block *s, ino_t ino) if (!fnode) return 0; - dno = fnode->u.external[0].disk_secno; + dno = le32_to_cpu(fnode->u.external[0].disk_secno); brelse(bh); return dno; } diff --git a/fs/hpfs/name.c b/fs/hpfs/name.c index f24736d7a439..9acdf338def0 100644 --- a/fs/hpfs/name.c +++ b/fs/hpfs/name.c @@ -8,39 +8,6 @@ #include "hpfs_fn.h" -static const char *text_postfix[]={ -".ASM", ".BAS", ".BAT", ".C", ".CC", ".CFG", ".CMD", ".CON", ".CPP", ".DEF", -".DOC", ".DPR", ".ERX", ".H", ".HPP", ".HTM", ".HTML", ".JAVA", ".LOG", ".PAS", -".RC", ".TEX", ".TXT", ".Y", ""}; - -static const char *text_prefix[]={ -"AUTOEXEC.", "CHANGES", "COPYING", "CONFIG.", "CREDITS", "FAQ", "FILE_ID.DIZ", -"MAKEFILE", "READ.ME", "README", "TERMCAP", ""}; - -void hpfs_decide_conv(struct inode *inode, const unsigned char *name, unsigned len) -{ - struct hpfs_inode_info *hpfs_inode = hpfs_i(inode); - int i; - if (hpfs_inode->i_conv != CONV_AUTO) return; - for (i = 0; *text_postfix[i]; i++) { - int l = strlen(text_postfix[i]); - if (l <= len) - if (!hpfs_compare_names(inode->i_sb, text_postfix[i], l, name + len - l, l, 0)) - goto text; - } - for (i = 0; *text_prefix[i]; i++) { - int l = strlen(text_prefix[i]); - if (l <= len) - if (!hpfs_compare_names(inode->i_sb, text_prefix[i], l, name, l, 0)) - goto text; - } - hpfs_inode->i_conv = CONV_BINARY; - return; - text: - hpfs_inode->i_conv = CONV_TEXT; - return; -} - static inline int not_allowed_char(unsigned char c) { return c<' ' || c=='"' || c=='*' || c=='/' || c==':' || c=='<' || diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index d5f8c8a19023..1f05839c27a7 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -29,7 +29,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh); if (!fnode) goto bail; - dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0, 1); + dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0); if (!dnode) goto bail1; memset(&dee, 0, sizeof dee); @@ -37,8 +37,8 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (!(mode & 0222)) dee.read_only = 1; /*dee.archive = 0;*/ dee.hidden = name[0] == '.'; - dee.fnode = fno; - dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds()); + dee.fnode = cpu_to_le32(fno); + dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds())); result = new_inode(dir->i_sb); if (!result) goto bail2; @@ -46,7 +46,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) result->i_ino = fno; hpfs_i(result)->i_parent_dir = dir->i_ino; hpfs_i(result)->i_dno = dno; - result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date); + result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)); result->i_ctime.tv_nsec = 0; result->i_mtime.tv_nsec = 0; result->i_atime.tv_nsec = 0; @@ -60,8 +60,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (dee.read_only) result->i_mode &= ~0222; - mutex_lock(&hpfs_i(dir)->i_mutex); - r = hpfs_add_dirent(dir, name, len, &dee, 0); + r = hpfs_add_dirent(dir, name, len, &dee); if (r == 1) goto bail3; if (r == -1) { @@ -70,21 +69,21 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) } fnode->len = len; memcpy(fnode->name, name, len > 15 ? 15 : len); - fnode->up = dir->i_ino; + fnode->up = cpu_to_le32(dir->i_ino); fnode->dirflag = 1; fnode->btree.n_free_nodes = 7; fnode->btree.n_used_nodes = 1; - fnode->btree.first_free = 0x14; - fnode->u.external[0].disk_secno = dno; - fnode->u.external[0].file_secno = -1; + fnode->btree.first_free = cpu_to_le16(0x14); + fnode->u.external[0].disk_secno = cpu_to_le32(dno); + fnode->u.external[0].file_secno = cpu_to_le32(-1); dnode->root_dnode = 1; - dnode->up = fno; + dnode->up = cpu_to_le32(fno); de = hpfs_add_de(dir->i_sb, dnode, "\001\001", 2, 0); - de->creation_date = de->write_date = de->read_date = gmt_to_local(dir->i_sb, get_seconds()); + de->creation_date = de->write_date = de->read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds())); if (!(mode & 0222)) de->read_only = 1; de->first = de->directory = 1; /*de->hidden = de->system = 0;*/ - de->fnode = fno; + de->fnode = cpu_to_le32(fno); mark_buffer_dirty(bh); brelse(bh); hpfs_mark_4buffers_dirty(&qbh0); @@ -101,11 +100,9 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) hpfs_write_inode_nolock(result); } d_instantiate(dentry, result); - mutex_unlock(&hpfs_i(dir)->i_mutex); hpfs_unlock(dir->i_sb); return 0; bail3: - mutex_unlock(&hpfs_i(dir)->i_mutex); iput(result); bail2: hpfs_brelse4(&qbh0); @@ -140,8 +137,8 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc if (!(mode & 0222)) dee.read_only = 1; dee.archive = 1; dee.hidden = name[0] == '.'; - dee.fnode = fno; - dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds()); + dee.fnode = cpu_to_le32(fno); + dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds())); result = new_inode(dir->i_sb); if (!result) @@ -154,9 +151,8 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc result->i_op = &hpfs_file_iops; result->i_fop = &hpfs_file_ops; result->i_nlink = 1; - hpfs_decide_conv(result, name, len); hpfs_i(result)->i_parent_dir = dir->i_ino; - result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date); + result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)); result->i_ctime.tv_nsec = 0; result->i_mtime.tv_nsec = 0; result->i_atime.tv_nsec = 0; @@ -168,8 +164,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc result->i_data.a_ops = &hpfs_aops; hpfs_i(result)->mmu_private = 0; - mutex_lock(&hpfs_i(dir)->i_mutex); - r = hpfs_add_dirent(dir, name, len, &dee, 0); + r = hpfs_add_dirent(dir, name, len, &dee); if (r == 1) goto bail2; if (r == -1) { @@ -178,7 +173,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc } fnode->len = len; memcpy(fnode->name, name, len > 15 ? 15 : len); - fnode->up = dir->i_ino; + fnode->up = cpu_to_le32(dir->i_ino); mark_buffer_dirty(bh); brelse(bh); @@ -193,12 +188,10 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc hpfs_write_inode_nolock(result); } d_instantiate(dentry, result); - mutex_unlock(&hpfs_i(dir)->i_mutex); hpfs_unlock(dir->i_sb); return 0; bail2: - mutex_unlock(&hpfs_i(dir)->i_mutex); iput(result); bail1: brelse(bh); @@ -232,8 +225,8 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t if (!(mode & 0222)) dee.read_only = 1; dee.archive = 1; dee.hidden = name[0] == '.'; - dee.fnode = fno; - dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds()); + dee.fnode = cpu_to_le32(fno); + dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds())); result = new_inode(dir->i_sb); if (!result) @@ -242,7 +235,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t hpfs_init_inode(result); result->i_ino = fno; hpfs_i(result)->i_parent_dir = dir->i_ino; - result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date); + result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)); result->i_ctime.tv_nsec = 0; result->i_mtime.tv_nsec = 0; result->i_atime.tv_nsec = 0; @@ -254,8 +247,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t result->i_blocks = 1; init_special_inode(result, mode, rdev); - mutex_lock(&hpfs_i(dir)->i_mutex); - r = hpfs_add_dirent(dir, name, len, &dee, 0); + r = hpfs_add_dirent(dir, name, len, &dee); if (r == 1) goto bail2; if (r == -1) { @@ -264,19 +256,17 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t } fnode->len = len; memcpy(fnode->name, name, len > 15 ? 15 : len); - fnode->up = dir->i_ino; + fnode->up = cpu_to_le32(dir->i_ino); mark_buffer_dirty(bh); insert_inode_hash(result); hpfs_write_inode_nolock(result); d_instantiate(dentry, result); - mutex_unlock(&hpfs_i(dir)->i_mutex); brelse(bh); hpfs_unlock(dir->i_sb); return 0; bail2: - mutex_unlock(&hpfs_i(dir)->i_mutex); iput(result); bail1: brelse(bh); @@ -310,8 +300,8 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy memset(&dee, 0, sizeof dee); dee.archive = 1; dee.hidden = name[0] == '.'; - dee.fnode = fno; - dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds()); + dee.fnode = cpu_to_le32(fno); + dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds())); result = new_inode(dir->i_sb); if (!result) @@ -319,7 +309,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy result->i_ino = fno; hpfs_init_inode(result); hpfs_i(result)->i_parent_dir = dir->i_ino; - result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date); + result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date)); result->i_ctime.tv_nsec = 0; result->i_mtime.tv_nsec = 0; result->i_atime.tv_nsec = 0; @@ -333,8 +323,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy result->i_op = &page_symlink_inode_operations; result->i_data.a_ops = &hpfs_symlink_aops; - mutex_lock(&hpfs_i(dir)->i_mutex); - r = hpfs_add_dirent(dir, name, len, &dee, 0); + r = hpfs_add_dirent(dir, name, len, &dee); if (r == 1) goto bail2; if (r == -1) { @@ -343,7 +332,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy } fnode->len = len; memcpy(fnode->name, name, len > 15 ? 15 : len); - fnode->up = dir->i_ino; + fnode->up = cpu_to_le32(dir->i_ino); hpfs_set_ea(result, fnode, "SYMLINK", symlink, strlen(symlink)); mark_buffer_dirty(bh); brelse(bh); @@ -352,11 +341,9 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy hpfs_write_inode_nolock(result); d_instantiate(dentry, result); - mutex_unlock(&hpfs_i(dir)->i_mutex); hpfs_unlock(dir->i_sb); return 0; bail2: - mutex_unlock(&hpfs_i(dir)->i_mutex); iput(result); bail1: brelse(bh); @@ -374,7 +361,6 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry) struct hpfs_dirent *de; struct inode *inode = dentry->d_inode; dnode_secno dno; - fnode_secno fno; int r; int rep = 0; int err; @@ -382,8 +368,6 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry) hpfs_lock(dir->i_sb); hpfs_adjust_length(name, &len); again: - mutex_lock(&hpfs_i(inode)->i_parent_mutex); - mutex_lock(&hpfs_i(dir)->i_mutex); err = -ENOENT; de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh); if (!de) @@ -397,7 +381,6 @@ again: if (de->directory) goto out1; - fno = de->fnode; r = hpfs_remove_dirent(dir, dno, de, &qbh, 1); switch (r) { case 1: @@ -410,8 +393,6 @@ again: if (rep++) break; - mutex_unlock(&hpfs_i(dir)->i_mutex); - mutex_unlock(&hpfs_i(inode)->i_parent_mutex); dentry_unhash(dentry); if (!d_unhashed(dentry)) { dput(dentry); @@ -445,8 +426,6 @@ again: out1: hpfs_brelse4(&qbh); out: - mutex_unlock(&hpfs_i(dir)->i_mutex); - mutex_unlock(&hpfs_i(inode)->i_parent_mutex); hpfs_unlock(dir->i_sb); return err; } @@ -459,15 +438,12 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry) struct hpfs_dirent *de; struct inode *inode = dentry->d_inode; dnode_secno dno; - fnode_secno fno; int n_items = 0; int err; int r; hpfs_adjust_length(name, &len); hpfs_lock(dir->i_sb); - mutex_lock(&hpfs_i(inode)->i_parent_mutex); - mutex_lock(&hpfs_i(dir)->i_mutex); err = -ENOENT; de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh); if (!de) @@ -486,7 +462,6 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry) if (n_items) goto out1; - fno = de->fnode; r = hpfs_remove_dirent(dir, dno, de, &qbh, 1); switch (r) { case 1: @@ -505,8 +480,6 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry) out1: hpfs_brelse4(&qbh); out: - mutex_unlock(&hpfs_i(dir)->i_mutex); - mutex_unlock(&hpfs_i(inode)->i_parent_mutex); hpfs_unlock(dir->i_sb); return err; } @@ -568,12 +541,6 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, hpfs_lock(i->i_sb); /* order doesn't matter, due to VFS exclusion */ - mutex_lock(&hpfs_i(i)->i_parent_mutex); - if (new_inode) - mutex_lock(&hpfs_i(new_inode)->i_parent_mutex); - mutex_lock(&hpfs_i(old_dir)->i_mutex); - if (new_dir != old_dir) - mutex_lock(&hpfs_i(new_dir)->i_mutex); /* Erm? Moving over the empty non-busy directory is perfectly legal */ if (new_inode && S_ISDIR(new_inode->i_mode)) { @@ -610,9 +577,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_dir == old_dir) hpfs_brelse4(&qbh); - hpfs_lock_creation(i->i_sb); - if ((r = hpfs_add_dirent(new_dir, new_name, new_len, &de, 1))) { - hpfs_unlock_creation(i->i_sb); + if ((r = hpfs_add_dirent(new_dir, new_name, new_len, &de))) { if (r == -1) hpfs_error(new_dir->i_sb, "hpfs_rename: dirent already exists!"); err = r == 1 ? -ENOSPC : -EFSERROR; if (new_dir != old_dir) hpfs_brelse4(&qbh); @@ -621,20 +586,17 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_dir == old_dir) if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, old_name, old_len, &dno, &qbh))) { - hpfs_unlock_creation(i->i_sb); hpfs_error(i->i_sb, "lookup succeeded but map dirent failed at #2"); err = -ENOENT; goto end1; } if ((r = hpfs_remove_dirent(old_dir, dno, dep, &qbh, 0))) { - hpfs_unlock_creation(i->i_sb); hpfs_error(i->i_sb, "hpfs_rename: could not remove dirent"); err = r == 2 ? -ENOSPC : -EFSERROR; goto end1; } - hpfs_unlock_creation(i->i_sb); - + end: hpfs_i(i)->i_parent_dir = new_dir->i_ino; if (S_ISDIR(i->i_mode)) { @@ -642,22 +604,14 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, drop_nlink(old_dir); } if ((fnode = hpfs_map_fnode(i->i_sb, i->i_ino, &bh))) { - fnode->up = new_dir->i_ino; + fnode->up = cpu_to_le32(new_dir->i_ino); fnode->len = new_len; memcpy(fnode->name, new_name, new_len>15?15:new_len); if (new_len < 15) memset(&fnode->name[new_len], 0, 15 - new_len); mark_buffer_dirty(bh); brelse(bh); } - hpfs_i(i)->i_conv = hpfs_sb(i->i_sb)->sb_conv; - hpfs_decide_conv(i, new_name, new_len); end1: - if (old_dir != new_dir) - mutex_unlock(&hpfs_i(new_dir)->i_mutex); - mutex_unlock(&hpfs_i(old_dir)->i_mutex); - mutex_unlock(&hpfs_i(i)->i_parent_mutex); - if (new_inode) - mutex_unlock(&hpfs_i(new_inode)->i_parent_mutex); hpfs_unlock(i->i_sb); return err; } diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index c89b40808587..98580a3b5005 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -18,15 +18,16 @@ /* Mark the filesystem dirty, so that chkdsk checks it when os/2 booted */ -static void mark_dirty(struct super_block *s) +static void mark_dirty(struct super_block *s, int remount) { - if (hpfs_sb(s)->sb_chkdsk && !(s->s_flags & MS_RDONLY)) { + if (hpfs_sb(s)->sb_chkdsk && (remount || !(s->s_flags & MS_RDONLY))) { struct buffer_head *bh; struct hpfs_spare_block *sb; if ((sb = hpfs_map_sector(s, 17, &bh, 0))) { sb->dirty = 1; sb->old_wrote = 0; mark_buffer_dirty(bh); + sync_dirty_buffer(bh); brelse(bh); } } @@ -40,10 +41,12 @@ static void unmark_dirty(struct super_block *s) struct buffer_head *bh; struct hpfs_spare_block *sb; if (s->s_flags & MS_RDONLY) return; + sync_blockdev(s->s_bdev); if ((sb = hpfs_map_sector(s, 17, &bh, 0))) { sb->dirty = hpfs_sb(s)->sb_chkdsk > 1 - hpfs_sb(s)->sb_was_error; sb->old_wrote = hpfs_sb(s)->sb_chkdsk >= 2 && !hpfs_sb(s)->sb_was_error; mark_buffer_dirty(bh); + sync_dirty_buffer(bh); brelse(bh); } } @@ -63,13 +66,13 @@ void hpfs_error(struct super_block *s, const char *fmt, ...) if (!hpfs_sb(s)->sb_was_error) { if (hpfs_sb(s)->sb_err == 2) { printk("; crashing the system because you wanted it\n"); - mark_dirty(s); + mark_dirty(s, 0); panic("HPFS panic"); } else if (hpfs_sb(s)->sb_err == 1) { if (s->s_flags & MS_RDONLY) printk("; already mounted read-only\n"); else { printk("; remounting read-only\n"); - mark_dirty(s); + mark_dirty(s, 0); s->s_flags |= MS_RDONLY; } } else if (s->s_flags & MS_RDONLY) printk("; going on - but anything won't be destroyed because it's read-only\n"); @@ -102,9 +105,12 @@ static void hpfs_put_super(struct super_block *s) { struct hpfs_sb_info *sbi = hpfs_sb(s); + hpfs_lock(s); + unmark_dirty(s); + hpfs_unlock(s); + kfree(sbi->sb_cp_table); kfree(sbi->sb_bmp_dir); - unmark_dirty(s); s->s_fs_info = NULL; kfree(sbi); } @@ -129,7 +135,7 @@ static unsigned count_bitmaps(struct super_block *s) n_bands = (hpfs_sb(s)->sb_fs_size + 0x3fff) >> 14; count = 0; for (n = 0; n < n_bands; n++) - count += hpfs_count_one_bitmap(s, hpfs_sb(s)->sb_bmp_dir[n]); + count += hpfs_count_one_bitmap(s, le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[n])); return count; } @@ -188,8 +194,6 @@ static void init_once(void *foo) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; - mutex_init(&ei->i_mutex); - mutex_init(&ei->i_parent_mutex); inode_init_once(&ei->vfs_inode); } @@ -218,7 +222,6 @@ static void destroy_inodecache(void) enum { Opt_help, Opt_uid, Opt_gid, Opt_umask, Opt_case_lower, Opt_case_asis, - Opt_conv_binary, Opt_conv_text, Opt_conv_auto, Opt_check_none, Opt_check_normal, Opt_check_strict, Opt_err_cont, Opt_err_ro, Opt_err_panic, Opt_eas_no, Opt_eas_ro, Opt_eas_rw, @@ -233,9 +236,6 @@ static const match_table_t tokens = { {Opt_umask, "umask=%o"}, {Opt_case_lower, "case=lower"}, {Opt_case_asis, "case=asis"}, - {Opt_conv_binary, "conv=binary"}, - {Opt_conv_text, "conv=text"}, - {Opt_conv_auto, "conv=auto"}, {Opt_check_none, "check=none"}, {Opt_check_normal, "check=normal"}, {Opt_check_strict, "check=strict"}, @@ -253,7 +253,7 @@ static const match_table_t tokens = { }; static int parse_opts(char *opts, uid_t *uid, gid_t *gid, umode_t *umask, - int *lowercase, int *conv, int *eas, int *chk, int *errs, + int *lowercase, int *eas, int *chk, int *errs, int *chkdsk, int *timeshift) { char *p; @@ -295,15 +295,6 @@ static int parse_opts(char *opts, uid_t *uid, gid_t *gid, umode_t *umask, case Opt_case_asis: *lowercase = 0; break; - case Opt_conv_binary: - *conv = CONV_BINARY; - break; - case Opt_conv_text: - *conv = CONV_TEXT; - break; - case Opt_conv_auto: - *conv = CONV_AUTO; - break; case Opt_check_none: *chk = 0; break; @@ -370,9 +361,6 @@ HPFS filesystem options:\n\ umask=xxx set mode of files that don't have mode specified in eas\n\ case=lower lowercase all files\n\ case=asis do not lowercase files (default)\n\ - conv=binary do not convert CR/LF -> LF (default)\n\ - conv=auto convert only files with known text extensions\n\ - conv=text convert all files\n\ check=none no fs checks - kernel may crash on corrupted filesystem\n\ check=normal do some checks - it should not crash (default)\n\ check=strict do extra time-consuming checks, used for debugging\n\ @@ -394,7 +382,7 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) uid_t uid; gid_t gid; umode_t umask; - int lowercase, conv, eas, chk, errs, chkdsk, timeshift; + int lowercase, eas, chk, errs, chkdsk, timeshift; int o; struct hpfs_sb_info *sbi = hpfs_sb(s); char *new_opts = kstrdup(data, GFP_KERNEL); @@ -405,11 +393,11 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) lock_super(s); uid = sbi->sb_uid; gid = sbi->sb_gid; umask = 0777 & ~sbi->sb_mode; - lowercase = sbi->sb_lowercase; conv = sbi->sb_conv; + lowercase = sbi->sb_lowercase; eas = sbi->sb_eas; chk = sbi->sb_chk; chkdsk = sbi->sb_chkdsk; errs = sbi->sb_err; timeshift = sbi->sb_timeshift; - if (!(o = parse_opts(data, &uid, &gid, &umask, &lowercase, &conv, + if (!(o = parse_opts(data, &uid, &gid, &umask, &lowercase, &eas, &chk, &errs, &chkdsk, ×hift))) { printk("HPFS: bad mount options.\n"); goto out_err; @@ -427,11 +415,11 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) sbi->sb_uid = uid; sbi->sb_gid = gid; sbi->sb_mode = 0777 & ~umask; - sbi->sb_lowercase = lowercase; sbi->sb_conv = conv; + sbi->sb_lowercase = lowercase; sbi->sb_eas = eas; sbi->sb_chk = chk; sbi->sb_chkdsk = chkdsk; sbi->sb_err = errs; sbi->sb_timeshift = timeshift; - if (!(*flags & MS_RDONLY)) mark_dirty(s); + if (!(*flags & MS_RDONLY)) mark_dirty(s, 1); replace_mount_options(s, new_opts); @@ -471,7 +459,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) uid_t uid; gid_t gid; umode_t umask; - int lowercase, conv, eas, chk, errs, chkdsk, timeshift; + int lowercase, eas, chk, errs, chkdsk, timeshift; dnode_secno root_dno; struct hpfs_dirent *de = NULL; @@ -479,11 +467,6 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) int o; - if (num_possible_cpus() > 1) { - printk(KERN_ERR "HPFS is not SMP safe\n"); - return -EINVAL; - } - save_mount_options(s, options); sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); @@ -495,20 +478,20 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) sbi->sb_bmp_dir = NULL; sbi->sb_cp_table = NULL; - mutex_init(&sbi->hpfs_creation_de); + mutex_init(&sbi->hpfs_mutex); + hpfs_lock(s); uid = current_uid(); gid = current_gid(); umask = current_umask(); lowercase = 0; - conv = CONV_BINARY; eas = 2; chk = 1; errs = 1; chkdsk = 1; timeshift = 0; - if (!(o = parse_opts(options, &uid, &gid, &umask, &lowercase, &conv, + if (!(o = parse_opts(options, &uid, &gid, &umask, &lowercase, &eas, &chk, &errs, &chkdsk, ×hift))) { printk("HPFS: bad mount options.\n"); goto bail0; @@ -526,9 +509,9 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) if (!(spareblock = hpfs_map_sector(s, 17, &bh2, 0))) goto bail3; /* Check magics */ - if (/*bootblock->magic != BB_MAGIC - ||*/ superblock->magic != SB_MAGIC - || spareblock->magic != SP_MAGIC) { + if (/*le16_to_cpu(bootblock->magic) != BB_MAGIC + ||*/ le32_to_cpu(superblock->magic) != SB_MAGIC + || le32_to_cpu(spareblock->magic) != SP_MAGIC) { if (!silent) printk("HPFS: Bad magic ... probably not HPFS\n"); goto bail4; } @@ -549,19 +532,18 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) s->s_op = &hpfs_sops; s->s_d_op = &hpfs_dentry_operations; - sbi->sb_root = superblock->root; - sbi->sb_fs_size = superblock->n_sectors; - sbi->sb_bitmaps = superblock->bitmaps; - sbi->sb_dirband_start = superblock->dir_band_start; - sbi->sb_dirband_size = superblock->n_dir_band; - sbi->sb_dmap = superblock->dir_band_bitmap; + sbi->sb_root = le32_to_cpu(superblock->root); + sbi->sb_fs_size = le32_to_cpu(superblock->n_sectors); + sbi->sb_bitmaps = le32_to_cpu(superblock->bitmaps); + sbi->sb_dirband_start = le32_to_cpu(superblock->dir_band_start); + sbi->sb_dirband_size = le32_to_cpu(superblock->n_dir_band); + sbi->sb_dmap = le32_to_cpu(superblock->dir_band_bitmap); sbi->sb_uid = uid; sbi->sb_gid = gid; sbi->sb_mode = 0777 & ~umask; sbi->sb_n_free = -1; sbi->sb_n_free_dnodes = -1; sbi->sb_lowercase = lowercase; - sbi->sb_conv = conv; sbi->sb_eas = eas; sbi->sb_chk = chk; sbi->sb_chkdsk = chkdsk; @@ -573,7 +555,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) sbi->sb_max_fwd_alloc = 0xffffff; /* Load bitmap directory */ - if (!(sbi->sb_bmp_dir = hpfs_load_bitmap_directory(s, superblock->bitmaps))) + if (!(sbi->sb_bmp_dir = hpfs_load_bitmap_directory(s, le32_to_cpu(superblock->bitmaps)))) goto bail4; /* Check for general fs errors*/ @@ -591,20 +573,20 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) mark_buffer_dirty(bh2); } - if (spareblock->hotfixes_used || spareblock->n_spares_used) { + if (le32_to_cpu(spareblock->hotfixes_used) || le32_to_cpu(spareblock->n_spares_used)) { if (errs >= 2) { printk("HPFS: Hotfixes not supported here, try chkdsk\n"); - mark_dirty(s); + mark_dirty(s, 0); goto bail4; } hpfs_error(s, "hotfixes not supported here, try chkdsk"); if (errs == 0) printk("HPFS: Proceeding, but your filesystem will be probably corrupted by this driver...\n"); else printk("HPFS: This driver may read bad files or crash when operating on disk with hotfixes.\n"); } - if (spareblock->n_dnode_spares != spareblock->n_dnode_spares_free) { + if (le32_to_cpu(spareblock->n_dnode_spares) != le32_to_cpu(spareblock->n_dnode_spares_free)) { if (errs >= 2) { printk("HPFS: Spare dnodes used, try chkdsk\n"); - mark_dirty(s); + mark_dirty(s, 0); goto bail4; } hpfs_error(s, "warning: spare dnodes used, try chkdsk"); @@ -612,26 +594,26 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) } if (chk) { unsigned a; - if (superblock->dir_band_end - superblock->dir_band_start + 1 != superblock->n_dir_band || - superblock->dir_band_end < superblock->dir_band_start || superblock->n_dir_band > 0x4000) { + if (le32_to_cpu(superblock->dir_band_end) - le32_to_cpu(superblock->dir_band_start) + 1 != le32_to_cpu(superblock->n_dir_band) || + le32_to_cpu(superblock->dir_band_end) < le32_to_cpu(superblock->dir_band_start) || le32_to_cpu(superblock->n_dir_band) > 0x4000) { hpfs_error(s, "dir band size mismatch: dir_band_start==%08x, dir_band_end==%08x, n_dir_band==%08x", - superblock->dir_band_start, superblock->dir_band_end, superblock->n_dir_band); + le32_to_cpu(superblock->dir_band_start), le32_to_cpu(superblock->dir_band_end), le32_to_cpu(superblock->n_dir_band)); goto bail4; } a = sbi->sb_dirband_size; sbi->sb_dirband_size = 0; - if (hpfs_chk_sectors(s, superblock->dir_band_start, superblock->n_dir_band, "dir_band") || - hpfs_chk_sectors(s, superblock->dir_band_bitmap, 4, "dir_band_bitmap") || - hpfs_chk_sectors(s, superblock->bitmaps, 4, "bitmaps")) { - mark_dirty(s); + if (hpfs_chk_sectors(s, le32_to_cpu(superblock->dir_band_start), le32_to_cpu(superblock->n_dir_band), "dir_band") || + hpfs_chk_sectors(s, le32_to_cpu(superblock->dir_band_bitmap), 4, "dir_band_bitmap") || + hpfs_chk_sectors(s, le32_to_cpu(superblock->bitmaps), 4, "bitmaps")) { + mark_dirty(s, 0); goto bail4; } sbi->sb_dirband_size = a; } else printk("HPFS: You really don't want any checks? You are crazy...\n"); /* Load code page table */ - if (spareblock->n_code_pages) - if (!(sbi->sb_cp_table = hpfs_load_code_page(s, spareblock->code_page_dir))) + if (le32_to_cpu(spareblock->n_code_pages)) + if (!(sbi->sb_cp_table = hpfs_load_code_page(s, le32_to_cpu(spareblock->code_page_dir)))) printk("HPFS: Warning: code page support is disabled\n"); brelse(bh2); @@ -660,13 +642,13 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) if (!de) hpfs_error(s, "unable to find root dir"); else { - root->i_atime.tv_sec = local_to_gmt(s, de->read_date); + root->i_atime.tv_sec = local_to_gmt(s, le32_to_cpu(de->read_date)); root->i_atime.tv_nsec = 0; - root->i_mtime.tv_sec = local_to_gmt(s, de->write_date); + root->i_mtime.tv_sec = local_to_gmt(s, le32_to_cpu(de->write_date)); root->i_mtime.tv_nsec = 0; - root->i_ctime.tv_sec = local_to_gmt(s, de->creation_date); + root->i_ctime.tv_sec = local_to_gmt(s, le32_to_cpu(de->creation_date)); root->i_ctime.tv_nsec = 0; - hpfs_i(root)->i_ea_size = de->ea_size; + hpfs_i(root)->i_ea_size = le16_to_cpu(de->ea_size); hpfs_i(root)->i_parent_dir = root->i_ino; if (root->i_size == -1) root->i_size = 2048; @@ -674,6 +656,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent) root->i_blocks = 5; hpfs_brelse4(&qbh); } + hpfs_unlock(s); return 0; bail4: brelse(bh2); @@ -681,6 +664,7 @@ bail3: brelse(bh1); bail2: brelse(bh0); bail1: bail0: + hpfs_unlock(s); kfree(sbi->sb_bmp_dir); kfree(sbi->sb_cp_table); s->s_fs_info = NULL; diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index da871ee084d3..69b180459463 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -362,7 +362,7 @@ void journal_commit_transaction(journal_t *journal) * we do not require it to remember exactly which old buffers it * has reserved. This is consistent with the existing behaviour * that multiple journal_get_write_access() calls to the same - * buffer are perfectly permissable. + * buffer are perfectly permissible. */ while (commit_transaction->t_reserved_list) { jh = commit_transaction->t_reserved_list; diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index eb11601f2e00..b3713afaaa9e 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -770,7 +770,7 @@ journal_t * journal_init_dev(struct block_device *bdev, journal->j_wbufsize = n; journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { - printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n", __func__); goto out_err; } @@ -831,7 +831,7 @@ journal_t * journal_init_inode (struct inode *inode) journal->j_wbufsize = n; journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { - printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n", __func__); goto out_err; } diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index d29018307e2e..305a90763154 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -71,7 +71,7 @@ * switching hash tables under them. For operations on the lists of entries in * the hash table j_revoke_lock is used. * - * Finally, also replay code uses the hash tables but at this moment noone else + * Finally, also replay code uses the hash tables but at this moment no one else * can touch them (filesystem isn't mounted yet) and hence no locking is * needed. */ diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 5b2e4c30a2a1..60d2319651b2 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -1392,7 +1392,7 @@ int journal_stop(handle_t *handle) * by 30x or more... * * We try and optimize the sleep time against what the underlying disk - * can do, instead of having a static sleep time. This is usefull for + * can do, instead of having a static sleep time. This is useful for * the case where our storage is so fast that it is more optimal to go * ahead and force a flush and wait for the transaction to be committed * than it is to wait for an arbitrary amount of time for new writers to diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index fa36d7662b21..6e28000a4b21 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -105,6 +105,8 @@ static int journal_submit_commit_record(journal_t *journal, int ret; struct timespec now = current_kernel_time(); + *cbh = NULL; + if (is_journal_aborted(journal)) return 0; @@ -403,7 +405,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) * we do not require it to remember exactly which old buffers it * has reserved. This is consistent with the existing behaviour * that multiple jbd2_journal_get_write_access() calls to the same - * buffer are perfectly permissable. + * buffer are perfectly permissible. */ while (commit_transaction->t_reserved_list) { jh = commit_transaction->t_reserved_list; @@ -806,7 +808,7 @@ wait_for_iobuf: if (err) __jbd2_journal_abort_hard(journal); } - if (!err && !is_journal_aborted(journal)) + if (cbh) err = journal_wait_on_commit_record(journal, cbh); if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT) && diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 90407b8fece7..e0ec3db1c395 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -917,7 +917,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_wbufsize = n; journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { - printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n", __func__); goto out_err; } @@ -983,7 +983,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_wbufsize = n; journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); if (!journal->j_wbuf) { - printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + printk(KERN_ERR "%s: Can't allocate bhs for commit thread\n", __func__); goto out_err; } @@ -2413,10 +2413,12 @@ const char *jbd2_dev_to_name(dev_t device) new_dev = kmalloc(sizeof(struct devname_cache), GFP_KERNEL); if (!new_dev) return "NODEV-ALLOCFAILURE"; /* Something non-NULL */ + bd = bdget(device); spin_lock(&devname_cache_lock); if (devcache[i]) { if (devcache[i]->device == device) { kfree(new_dev); + bdput(bd); ret = devcache[i]->devname; spin_unlock(&devname_cache_lock); return ret; @@ -2425,7 +2427,6 @@ const char *jbd2_dev_to_name(dev_t device) } devcache[i] = new_dev; devcache[i]->device = device; - bd = bdget(device); if (bd) { bdevname(bd, devcache[i]->devname); bdput(bd); diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 9ad321fd63fd..69fd93588118 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -71,7 +71,7 @@ * switching hash tables under them. For operations on the lists of entries in * the hash table j_revoke_lock is used. * - * Finally, also replay code uses the hash tables but at this moment noone else + * Finally, also replay code uses the hash tables but at this moment no one else * can touch them (filesystem isn't mounted yet) and hence no locking is * needed. */ diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 1d1191050f99..05fa77a23711 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -1403,7 +1403,7 @@ int jbd2_journal_stop(handle_t *handle) /* * Once we drop t_updates, if it goes to zero the transaction - * could start commiting on us and eventually disappear. So + * could start committing on us and eventually disappear. So * once we do this, we must not dereference transaction * pointer again. */ diff --git a/fs/jffs2/TODO b/fs/jffs2/TODO index 5d3ea4070f01..ca28964abd4b 100644 --- a/fs/jffs2/TODO +++ b/fs/jffs2/TODO @@ -11,7 +11,7 @@ - checkpointing (do we need this? scan is quite fast) - make the scan code populate real inodes so read_inode just after mount doesn't have to read the flash twice for large files. - Make this a per-inode option, changable with chattr, so you can + Make this a per-inode option, changeable with chattr, so you can decide which inodes should be in-core immediately after mount. - test, test, test diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index d32ee9412cb9..2ab1a0d91210 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -24,7 +24,7 @@ * * Returns: 0 if the data CRC is correct; * 1 - if incorrect; - * error code if an error occured. + * error code if an error occurred. */ static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info *tn) { diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c index 800171dca53b..e537fb0e0184 100644 --- a/fs/jffs2/summary.c +++ b/fs/jffs2/summary.c @@ -121,7 +121,7 @@ int jffs2_sum_add_inode_mem(struct jffs2_summary *s, struct jffs2_raw_inode *ri, temp->nodetype = ri->nodetype; temp->inode = ri->ino; temp->version = ri->version; - temp->offset = cpu_to_je32(ofs); /* relative offset from the begining of the jeb */ + temp->offset = cpu_to_je32(ofs); /* relative offset from the beginning of the jeb */ temp->totlen = ri->totlen; temp->next = NULL; @@ -139,7 +139,7 @@ int jffs2_sum_add_dirent_mem(struct jffs2_summary *s, struct jffs2_raw_dirent *r temp->nodetype = rd->nodetype; temp->totlen = rd->totlen; - temp->offset = cpu_to_je32(ofs); /* relative from the begining of the jeb */ + temp->offset = cpu_to_je32(ofs); /* relative from the beginning of the jeb */ temp->pino = rd->pino; temp->version = rd->version; temp->ino = rd->ino; diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 07ee1546b2fa..4515bea0268f 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -1116,7 +1116,7 @@ int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, /* * On NAND we try to mark this block bad. If the block was erased more - * than MAX_ERASE_FAILURES we mark it finaly bad. + * than MAX_ERASE_FAILURES we mark it finally bad. * Don't care about failures. This block remains on the erase-pending * or badblock list as long as nobody manipulates the flash with * a bootloader or something like that. diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index c92ea3b3ea5e..4496872cf4e7 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -1649,7 +1649,7 @@ static int dbFindCtl(struct bmap * bmp, int l2nb, int level, s64 * blkno) } /* search the tree within the dmap control page for - * sufficent free space. if sufficient free space is found, + * sufficient free space. if sufficient free space is found, * dbFindLeaf() returns the index of the leaf at which * free space was found. */ @@ -2744,7 +2744,7 @@ static int dbJoin(dmtree_t * tp, int leafno, int newval) /* check which (leafno or buddy) is the left buddy. * the left buddy gets to claim the blocks resulting * from the join while the right gets to claim none. - * the left buddy is also eligable to participate in + * the left buddy is also eligible to participate in * a join at the next higher level while the right * is not. * diff --git a/fs/jfs/jfs_extent.c b/fs/jfs/jfs_extent.c index 5d3bbd10f8db..e5fe8506ed16 100644 --- a/fs/jfs/jfs_extent.c +++ b/fs/jfs/jfs_extent.c @@ -126,7 +126,7 @@ extAlloc(struct inode *ip, s64 xlen, s64 pno, xad_t * xp, bool abnr) /* allocate the disk blocks for the extent. initially, extBalloc() * will try to allocate disk blocks for the requested size (xlen). - * if this fails (xlen contiguous free blocks not avaliable), it'll + * if this fails (xlen contiguous free blocks not available), it'll * try to allocate a smaller number of blocks (producing a smaller * extent), with this smaller number of blocks consisting of the * requested number of blocks rounded down to the next smaller @@ -481,7 +481,7 @@ int extFill(struct inode *ip, xad_t * xp) * * initially, we will try to allocate disk blocks for the * requested size (nblocks). if this fails (nblocks - * contiguous free blocks not avaliable), we'll try to allocate + * contiguous free blocks not available), we'll try to allocate * a smaller number of blocks (producing a smaller extent), with * this smaller number of blocks consisting of the requested * number of blocks rounded down to the next smaller power of 2 @@ -575,7 +575,7 @@ extBalloc(struct inode *ip, s64 hint, s64 * nblocks, s64 * blkno) * to a new set of blocks. If moving the extent, we initially * will try to allocate disk blocks for the requested size * (newnblks). if this fails (new contiguous free blocks not - * avaliable), we'll try to allocate a smaller number of + * available), we'll try to allocate a smaller number of * blocks (producing a smaller extent), with this smaller * number of blocks consisting of the requested number of * blocks rounded down to the next smaller power of 2 diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 3a09423b6c22..ed53a4740168 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -1069,7 +1069,7 @@ int diFree(struct inode *ip) */ if (iagp->nfreeexts == cpu_to_le32(EXTSPERIAG - 1)) { /* in preparation for removing the iag from the - * ag extent free list, read the iags preceeding + * ag extent free list, read the iags preceding * and following the iag on the ag extent free * list. */ @@ -1095,7 +1095,7 @@ int diFree(struct inode *ip) int inofreefwd = le32_to_cpu(iagp->inofreefwd); /* in preparation for removing the iag from the - * ag inode free list, read the iags preceeding + * ag inode free list, read the iags preceding * and following the iag on the ag inode free * list. before reading these iags, we must make * sure that we already don't have them in hand @@ -1681,7 +1681,7 @@ diAllocAG(struct inomap * imap, int agno, bool dir, struct inode *ip) * try to allocate a new extent of free inodes. */ if (addext) { - /* if free space is not avaliable for this new extent, try + /* if free space is not available for this new extent, try * below to allocate a free and existing (already backed) * inode from the ag. */ @@ -2036,7 +2036,7 @@ static int diAllocBit(struct inomap * imap, struct iag * iagp, int ino) /* check if this is the last free inode within the iag. * if so, it will have to be removed from the ag free - * inode list, so get the iags preceeding and following + * inode list, so get the iags preceding and following * it on the list. */ if (iagp->nfreeinos == cpu_to_le32(1)) { @@ -2208,7 +2208,7 @@ static int diNewExt(struct inomap * imap, struct iag * iagp, int extno) /* check if this is the last free extent within the * iag. if so, the iag must be removed from the ag - * free extent list, so get the iags preceeding and + * free extent list, so get the iags preceding and * following the iag on this list. */ if (iagp->nfreeexts == cpu_to_le32(1)) { @@ -2504,7 +2504,7 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp) } - /* get the next avaliable iag number */ + /* get the next available iag number */ iagno = imap->im_nextiag; /* make sure that we have not exceeded the maximum inode @@ -2615,7 +2615,7 @@ diNewIAG(struct inomap * imap, int *iagnop, int agno, struct metapage ** mpp) duplicateIXtree(sb, blkno, xlen, &xaddr); - /* update the next avaliable iag number */ + /* update the next available iag number */ imap->im_nextiag += 1; /* Add the iag to the iag free list so we don't lose the iag diff --git a/fs/jfs/jfs_logmgr.h b/fs/jfs/jfs_logmgr.h index 9236bc49ae7f..e38c21598850 100644 --- a/fs/jfs/jfs_logmgr.h +++ b/fs/jfs/jfs_logmgr.h @@ -288,7 +288,7 @@ struct lrd { /* * SYNCPT: log sync point * - * replay log upto syncpt address specified; + * replay log up to syncpt address specified; */ struct { __le32 sync; /* 4: syncpt address (0 = here) */ diff --git a/fs/jfs/jfs_metapage.h b/fs/jfs/jfs_metapage.h index d94f8d9e87d7..a78beda85f68 100644 --- a/fs/jfs/jfs_metapage.h +++ b/fs/jfs/jfs_metapage.h @@ -75,7 +75,7 @@ extern void grab_metapage(struct metapage *); extern void force_metapage(struct metapage *); /* - * hold_metapage and put_metapage are used in conjuction. The page lock + * hold_metapage and put_metapage are used in conjunction. The page lock * is not dropped between the two, so no other threads can get or release * the metapage */ diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 9466957ec841..f6cc0c09ec63 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -636,7 +636,7 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, * the inode of the page and available to all anonymous * transactions until txCommit() time at which point * they are transferred to the transaction tlock list of - * the commiting transaction of the inode) + * the committing transaction of the inode) */ if (xtid == 0) { tlck->tid = tid; diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c index 1aba0039f1c9..8ea5efb5a34e 100644 --- a/fs/jfs/resize.c +++ b/fs/jfs/resize.c @@ -57,7 +57,7 @@ * 2. compute new FSCKSize from new LVSize; * 3. set new FSSize as MIN(FSSize, LVSize-(LogSize+FSCKSize)) where * assert(new FSSize >= old FSSize), - * i.e., file system must not be shrinked; + * i.e., file system must not be shrunk; */ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) { @@ -182,7 +182,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) */ newFSSize = newLVSize - newLogSize - newFSCKSize; - /* file system cannot be shrinked */ + /* file system cannot be shrunk */ if (newFSSize < bmp->db_mapsize) { rc = -EINVAL; goto out; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index eeca48a031ab..06c8a67cbe76 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -644,7 +644,7 @@ static int jfs_show_options(struct seq_file *seq, struct vfsmount *vfs) /* Read data from quotafile - avoid pagecache and such because we cannot afford * acquiring the locks... As quota files are never truncated and quota code - * itself serializes the operations (and noone else should touch the files) + * itself serializes the operations (and no one else should touch the files) * we don't have to be afraid of races */ static ssize_t jfs_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c index 1adc8d455f0e..df0de27c2733 100644 --- a/fs/logfs/dev_bdev.c +++ b/fs/logfs/dev_bdev.c @@ -10,6 +10,7 @@ #include <linux/blkdev.h> #include <linux/buffer_head.h> #include <linux/gfp.h> +#include <linux/prefetch.h> #define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1)) diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c index 7466e9dcc8c5..339e17e9133d 100644 --- a/fs/logfs/dev_mtd.c +++ b/fs/logfs/dev_mtd.c @@ -60,7 +60,7 @@ static int mtd_write(struct super_block *sb, loff_t ofs, size_t len, void *buf) * asynchronous properties. So just to prevent the first implementor of such * a thing from breaking logfs in 2350, we do the usual pointless dance to * declare a completion variable and wait for completion before returning - * from mtd_erase(). What an excercise in futility! + * from mtd_erase(). What an exercise in futility! */ static void logfs_erase_callback(struct erase_info *ei) { diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index f9ddf0c388c8..9ed89d1663f8 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -92,7 +92,7 @@ static int beyond_eof(struct inode *inode, loff_t bix) * so short names (len <= 9) don't even occupy the complete 32bit name * space. A prime >256 ensures short names quickly spread the 32bit * name space. Add about 26 for the estimated amount of information - * of each character and pick a prime nearby, preferrably a bit-sparse + * of each character and pick a prime nearby, preferably a bit-sparse * one. */ static u32 hash_32(const char *s, int len, u32 seed) diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c index ee99a9f5dfd3..9e22085231b3 100644 --- a/fs/logfs/readwrite.c +++ b/fs/logfs/readwrite.c @@ -1616,7 +1616,7 @@ int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, err = logfs_write_buf(inode, page, flags); if (!err && shrink_level(gc_level) == 0) { /* Rewrite cannot mark the inode dirty but has to - * write it immediatly. + * write it immediately. * Q: Can't we just create an alias for the inode * instead? And if not, why not? */ diff --git a/fs/logfs/super.c b/fs/logfs/super.c index 33435e4b14d2..ce03a182c771 100644 --- a/fs/logfs/super.c +++ b/fs/logfs/super.c @@ -480,10 +480,6 @@ static int logfs_read_sb(struct super_block *sb, int read_only) !read_only) return -EIO; - mutex_init(&super->s_dirop_mutex); - mutex_init(&super->s_object_alias_mutex); - INIT_LIST_HEAD(&super->s_freeing_list); - ret = logfs_init_rw(sb); if (ret) return ret; @@ -601,6 +597,10 @@ static struct dentry *logfs_mount(struct file_system_type *type, int flags, if (!super) return ERR_PTR(-ENOMEM); + mutex_init(&super->s_dirop_mutex); + mutex_init(&super->s_object_alias_mutex); + INIT_LIST_HEAD(&super->s_freeing_list); + if (!devname) err = logfs_get_sb_bdev(super, type, devname); else if (strncmp(devname, "mtd", 3)) diff --git a/fs/mbcache.c b/fs/mbcache.c index a25444ab2baf..2f174be06555 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -542,7 +542,7 @@ __mb_cache_entry_find(struct list_head *l, struct list_head *head, * mb_cache_entry_find_first() * * Find the first cache entry on a given device with a certain key in - * an additional index. Additonal matches can be found with + * an additional index. Additional matches can be found with * mb_cache_entry_find_next(). Returns NULL if no match was found. The * returned cache entry is locked for shared access ("multiple readers"). * diff --git a/fs/namei.c b/fs/namei.c index 3cb616d38d9c..e3c4f112ebf7 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -70,7 +70,7 @@ * name indicated by the symlink. The old code always complained that the * name already exists, due to not following the symlink even if its target * is nonexistent. The new semantics affects also mknod() and link() when - * the name is a symlink pointing to a non-existant name. + * the name is a symlink pointing to a non-existent name. * * I don't know which semantics is the right one, since I have no access * to standards. But I found by trial that HP-UX 9.0 has the full "new" @@ -179,7 +179,7 @@ EXPORT_SYMBOL(putname); static int acl_permission_check(struct inode *inode, int mask, unsigned int flags, int (*check_acl)(struct inode *inode, int mask, unsigned int flags)) { - umode_t mode = inode->i_mode; + unsigned int mode = inode->i_mode; mask &= MAY_READ | MAY_WRITE | MAY_EXEC; @@ -697,6 +697,7 @@ static __always_inline void set_root_rcu(struct nameidata *nd) do { seq = read_seqcount_begin(&fs->seq); nd->root = fs->root; + nd->seq = __read_seqcount_begin(&nd->root.dentry->d_seq); } while (read_seqcount_retry(&fs->seq, seq)); } } diff --git a/fs/namespace.c b/fs/namespace.c index 7dba2ed03429..d99bcf59e4c2 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1030,18 +1030,6 @@ const struct seq_operations mounts_op = { .show = show_vfsmnt }; -static int uuid_is_nil(u8 *uuid) -{ - int i; - u8 *cp = (u8 *)uuid; - - for (i = 0; i < 16; i++) { - if (*cp++) - return 0; - } - return 1; -} - static int show_mountinfo(struct seq_file *m, void *v) { struct proc_mounts *p = m->private; @@ -1085,10 +1073,6 @@ static int show_mountinfo(struct seq_file *m, void *v) if (IS_MNT_UNBINDABLE(mnt)) seq_puts(m, " unbindable"); - if (!uuid_is_nil(mnt->mnt_sb->s_uuid)) - /* print the uuid */ - seq_printf(m, " uuid:%pU", mnt->mnt_sb->s_uuid); - /* Filesystem specific data */ seq_puts(m, " - "); show_type(m, sb); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 00a1d1c3d3a4..0250e4ce4893 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -596,7 +596,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) /* server->priv.data = NULL; */ server->m = data; - /* Althought anything producing this is buggy, it happens + /* Although anything producing this is buggy, it happens now because of PATH_MAX changes.. */ if (server->m.time_out < 1) { server->m.time_out = 10; diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 14e0f9371d14..00ecf62ce7c1 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -241,7 +241,7 @@ static __be32 decode_layoutrecall_args(struct svc_rqst *rqstp, args->cbl_layout_type = ntohl(*p++); /* Depite the spec's xdr, iomode really belongs in the FILE switch, - * as it is unuseable and ignored with the other types. + * as it is unusable and ignored with the other types. */ iomode = ntohl(*p++); args->cbl_layoutchanged = ntohl(*p++); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 3ac5bd695e5e..2f093ed16980 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -301,7 +301,7 @@ nfs_file_mmap(struct file * file, struct vm_area_struct * vma) * disk, but it retrieves and clears ctx->error after synching, despite * the two being set at the same time in nfs_context_set_write_error(). * This is because the former is used to notify the _next_ call to - * nfs_file_write() that a write error occured, and hence cause it to + * nfs_file_write() that a write error occurred, and hence cause it to * fall back to doing a synchronous write. */ static int diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 9166fcb66da2..1f063bacd285 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -119,7 +119,7 @@ Elong: } #ifdef CONFIG_NFS_V4 -static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors, struct inode *inode) +static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors) { struct gss_api_mech *mech; struct xdr_netobj oid; @@ -148,67 +148,64 @@ static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors, return pseudoflavor; } -static rpc_authflavor_t nfs_negotiate_security(const struct dentry *parent, const struct dentry *dentry) +static int nfs_negotiate_security(const struct dentry *parent, + const struct dentry *dentry, + rpc_authflavor_t *flavor) { - int status = 0; struct page *page; struct nfs4_secinfo_flavors *flavors; int (*secinfo)(struct inode *, const struct qstr *, struct nfs4_secinfo_flavors *); - rpc_authflavor_t flavor = RPC_AUTH_UNIX; + int ret = -EPERM; secinfo = NFS_PROTO(parent->d_inode)->secinfo; if (secinfo != NULL) { page = alloc_page(GFP_KERNEL); if (!page) { - status = -ENOMEM; + ret = -ENOMEM; goto out; } flavors = page_address(page); - status = secinfo(parent->d_inode, &dentry->d_name, flavors); - flavor = nfs_find_best_sec(flavors, dentry->d_inode); + ret = secinfo(parent->d_inode, &dentry->d_name, flavors); + *flavor = nfs_find_best_sec(flavors); put_page(page); } - return flavor; - out: - status = -ENOMEM; - return status; + return ret; } -static rpc_authflavor_t nfs_lookup_with_sec(struct nfs_server *server, struct dentry *parent, - struct dentry *dentry, struct path *path, - struct nfs_fh *fh, struct nfs_fattr *fattr) +static int nfs_lookup_with_sec(struct nfs_server *server, struct dentry *parent, + struct dentry *dentry, struct path *path, + struct nfs_fh *fh, struct nfs_fattr *fattr, + rpc_authflavor_t *flavor) { - rpc_authflavor_t flavor; struct rpc_clnt *clone; struct rpc_auth *auth; int err; - flavor = nfs_negotiate_security(parent, path->dentry); - if (flavor < 0) + err = nfs_negotiate_security(parent, path->dentry, flavor); + if (err < 0) goto out; clone = rpc_clone_client(server->client); - auth = rpcauth_create(flavor, clone); + auth = rpcauth_create(*flavor, clone); if (!auth) { - flavor = -EIO; + err = -EIO; goto out_shutdown; } err = server->nfs_client->rpc_ops->lookup(clone, parent->d_inode, &path->dentry->d_name, fh, fattr); - if (err < 0) - flavor = err; out_shutdown: rpc_shutdown_client(clone); out: - return flavor; + return err; } #else /* CONFIG_NFS_V4 */ -static inline rpc_authflavor_t nfs_lookup_with_sec(struct nfs_server *server, - struct dentry *parent, struct dentry *dentry, - struct path *path, struct nfs_fh *fh, - struct nfs_fattr *fattr) +static inline int nfs_lookup_with_sec(struct nfs_server *server, + struct dentry *parent, struct dentry *dentry, + struct path *path, struct nfs_fh *fh, + struct nfs_fattr *fattr, + rpc_authflavor_t *flavor) { return -EPERM; } @@ -234,7 +231,7 @@ struct vfsmount *nfs_d_automount(struct path *path) struct nfs_fh *fh = NULL; struct nfs_fattr *fattr = NULL; int err; - rpc_authflavor_t flavor = 1; + rpc_authflavor_t flavor = RPC_AUTH_UNIX; dprintk("--> nfs_d_automount()\n"); @@ -255,13 +252,8 @@ struct vfsmount *nfs_d_automount(struct path *path) err = server->nfs_client->rpc_ops->lookup(server->client, parent->d_inode, &path->dentry->d_name, fh, fattr); - if (err == -EPERM) { - flavor = nfs_lookup_with_sec(server, parent, path->dentry, path, fh, fattr); - if (flavor < 0) - err = flavor; - else - err = 0; - } + if (err == -EPERM && NFS_PROTO(parent->d_inode)->secinfo != NULL) + err = nfs_lookup_with_sec(server, parent, path->dentry, path, fh, fattr, &flavor); dput(parent); if (err != 0) { mnt = ERR_PTR(err); diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index e1c261ddd65d..c4a69833dd0d 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -47,6 +47,7 @@ enum nfs4_client_state { NFS4CLNT_LAYOUTRECALL, NFS4CLNT_SESSION_RESET, NFS4CLNT_RECALL_SLOT, + NFS4CLNT_LEASE_CONFIRM, }; enum nfs4_session_state { diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 6f8192f4cfc7..be79dc9f386d 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -117,6 +117,8 @@ static int filelayout_async_handle_error(struct rpc_task *task, case -EKEYEXPIRED: rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX); break; + case -NFS4ERR_RETRY_UNCACHED_REP: + break; default: dprintk("%s DS error. Retry through MDS %d\n", __func__, task->tk_status); @@ -416,7 +418,8 @@ static int filelayout_check_layout(struct pnfs_layout_hdr *lo, struct nfs4_filelayout_segment *fl, struct nfs4_layoutget_res *lgr, - struct nfs4_deviceid *id) + struct nfs4_deviceid *id, + gfp_t gfp_flags) { struct nfs4_file_layout_dsaddr *dsaddr; int status = -EINVAL; @@ -439,7 +442,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, /* find and reference the deviceid */ dsaddr = nfs4_fl_find_get_deviceid(id); if (dsaddr == NULL) { - dsaddr = get_device_info(lo->plh_inode, id); + dsaddr = get_device_info(lo->plh_inode, id, gfp_flags); if (dsaddr == NULL) goto out; } @@ -500,7 +503,8 @@ static int filelayout_decode_layout(struct pnfs_layout_hdr *flo, struct nfs4_filelayout_segment *fl, struct nfs4_layoutget_res *lgr, - struct nfs4_deviceid *id) + struct nfs4_deviceid *id, + gfp_t gfp_flags) { struct xdr_stream stream; struct xdr_buf buf = { @@ -516,7 +520,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, dprintk("%s: set_layout_map Begin\n", __func__); - scratch = alloc_page(GFP_KERNEL); + scratch = alloc_page(gfp_flags); if (!scratch) return -ENOMEM; @@ -554,13 +558,13 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, goto out_err; fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), - GFP_KERNEL); + gfp_flags); if (!fl->fh_array) goto out_err; for (i = 0; i < fl->num_fh; i++) { /* Do we want to use a mempool here? */ - fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), GFP_KERNEL); + fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), gfp_flags); if (!fl->fh_array[i]) goto out_err_free; @@ -605,19 +609,20 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg) static struct pnfs_layout_segment * filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, - struct nfs4_layoutget_res *lgr) + struct nfs4_layoutget_res *lgr, + gfp_t gfp_flags) { struct nfs4_filelayout_segment *fl; int rc; struct nfs4_deviceid id; dprintk("--> %s\n", __func__); - fl = kzalloc(sizeof(*fl), GFP_KERNEL); + fl = kzalloc(sizeof(*fl), gfp_flags); if (!fl) return NULL; - rc = filelayout_decode_layout(layoutid, fl, lgr, &id); - if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id)) { + rc = filelayout_decode_layout(layoutid, fl, lgr, &id, gfp_flags); + if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id, gfp_flags)) { _filelayout_free_lseg(fl); return NULL; } @@ -633,7 +638,7 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, int size = (fl->stripe_type == STRIPE_SPARSE) ? fl->dsaddr->ds_num : fl->dsaddr->stripe_count; - fl->commit_buckets = kcalloc(size, sizeof(struct list_head), GFP_KERNEL); + fl->commit_buckets = kcalloc(size, sizeof(struct list_head), gfp_flags); if (!fl->commit_buckets) { filelayout_free_lseg(&fl->generic_hdr); return NULL; diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h index 085a354e0f08..2b461d77b43a 100644 --- a/fs/nfs/nfs4filelayout.h +++ b/fs/nfs/nfs4filelayout.h @@ -33,7 +33,7 @@ #include "pnfs.h" /* - * Field testing shows we need to support upto 4096 stripe indices. + * Field testing shows we need to support up to 4096 stripe indices. * We store each index as a u8 (u32 on the wire) to keep the memory footprint * reasonable. This in turn means we support a maximum of 256 * RFC 5661 multipath_list4 structures. @@ -104,6 +104,6 @@ extern struct nfs4_file_layout_dsaddr * nfs4_fl_find_get_deviceid(struct nfs4_deviceid *dev_id); extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr); struct nfs4_file_layout_dsaddr * -get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id); +get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags); #endif /* FS_NFS_NFS4FILELAYOUT_H */ diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c index de5350f2b249..db07c7af1395 100644 --- a/fs/nfs/nfs4filelayoutdev.c +++ b/fs/nfs/nfs4filelayoutdev.c @@ -225,11 +225,11 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr) } static struct nfs4_pnfs_ds * -nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port) +nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port, gfp_t gfp_flags) { struct nfs4_pnfs_ds *tmp_ds, *ds; - ds = kzalloc(sizeof(*tmp_ds), GFP_KERNEL); + ds = kzalloc(sizeof(*tmp_ds), gfp_flags); if (!ds) goto out; @@ -261,7 +261,7 @@ out: * Currently only support ipv4, and one multi-path address. */ static struct nfs4_pnfs_ds * -decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode) +decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode, gfp_t gfp_flags) { struct nfs4_pnfs_ds *ds = NULL; char *buf; @@ -303,7 +303,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode) rlen); goto out_err; } - buf = kmalloc(rlen + 1, GFP_KERNEL); + buf = kmalloc(rlen + 1, gfp_flags); if (!buf) { dprintk("%s: Not enough memory\n", __func__); goto out_err; @@ -333,7 +333,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode) sscanf(pstr, "-%d-%d", &tmp[0], &tmp[1]); port = htons((tmp[0] << 8) | (tmp[1])); - ds = nfs4_pnfs_ds_add(inode, ip_addr, port); + ds = nfs4_pnfs_ds_add(inode, ip_addr, port, gfp_flags); dprintk("%s: Decoded address and port %s\n", __func__, buf); out_free: kfree(buf); @@ -343,7 +343,7 @@ out_err: /* Decode opaque device data and return the result */ static struct nfs4_file_layout_dsaddr* -decode_device(struct inode *ino, struct pnfs_device *pdev) +decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags) { int i; u32 cnt, num; @@ -362,7 +362,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev) struct page *scratch; /* set up xdr stream */ - scratch = alloc_page(GFP_KERNEL); + scratch = alloc_page(gfp_flags); if (!scratch) goto out_err; @@ -384,7 +384,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev) } /* read stripe indices */ - stripe_indices = kcalloc(cnt, sizeof(u8), GFP_KERNEL); + stripe_indices = kcalloc(cnt, sizeof(u8), gfp_flags); if (!stripe_indices) goto out_err_free_scratch; @@ -423,7 +423,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev) dsaddr = kzalloc(sizeof(*dsaddr) + (sizeof(struct nfs4_pnfs_ds *) * (num - 1)), - GFP_KERNEL); + gfp_flags); if (!dsaddr) goto out_err_free_stripe_indices; @@ -452,7 +452,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev) for (j = 0; j < mp_count; j++) { if (j == 0) { dsaddr->ds_list[i] = decode_and_add_ds(&stream, - ino); + ino, gfp_flags); if (dsaddr->ds_list[i] == NULL) goto out_err_free_deviceid; } else { @@ -503,12 +503,12 @@ out_err: * available devices. */ static struct nfs4_file_layout_dsaddr * -decode_and_add_device(struct inode *inode, struct pnfs_device *dev) +decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_flags) { struct nfs4_file_layout_dsaddr *d, *new; long hash; - new = decode_device(inode, dev); + new = decode_device(inode, dev, gfp_flags); if (!new) { printk(KERN_WARNING "%s: Could not decode or add device\n", __func__); @@ -537,7 +537,7 @@ decode_and_add_device(struct inode *inode, struct pnfs_device *dev) * of available devices, and return it. */ struct nfs4_file_layout_dsaddr * -get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id) +get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags) { struct pnfs_device *pdev = NULL; u32 max_resp_sz; @@ -556,17 +556,17 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id) dprintk("%s inode %p max_resp_sz %u max_pages %d\n", __func__, inode, max_resp_sz, max_pages); - pdev = kzalloc(sizeof(struct pnfs_device), GFP_KERNEL); + pdev = kzalloc(sizeof(struct pnfs_device), gfp_flags); if (pdev == NULL) return NULL; - pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL); + pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags); if (pages == NULL) { kfree(pdev); return NULL; } for (i = 0; i < max_pages; i++) { - pages[i] = alloc_page(GFP_KERNEL); + pages[i] = alloc_page(gfp_flags); if (!pages[i]) goto out_free; } @@ -587,7 +587,7 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id) * Found new device, need to decode it and then add it to the * list of known devices for this mountpoint. */ - dsaddr = decode_and_add_device(inode, pdev); + dsaddr = decode_and_add_device(inode, pdev, gfp_flags); out_free: for (i = 0; i < max_pages; i++) __free_page(pages[i]); diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index dfd1e6d7e6c3..cf1b339c3937 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -46,6 +46,7 @@ #include <linux/nfs4.h> #include <linux/nfs_fs.h> #include <linux/nfs_page.h> +#include <linux/nfs_mount.h> #include <linux/namei.h> #include <linux/mount.h> #include <linux/module.h> @@ -299,6 +300,7 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc ret = nfs4_delay(server->client, &exception->timeout); if (ret != 0) break; + case -NFS4ERR_RETRY_UNCACHED_REP: case -NFS4ERR_OLD_STATEID: exception->retry = 1; break; @@ -443,8 +445,8 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res * if (res->sr_status == 1) res->sr_status = NFS_OK; - /* -ERESTARTSYS can result in skipping nfs41_sequence_setup */ - if (!res->sr_slot) + /* don't increment the sequence number if the task wasn't sent */ + if (!RPC_WAS_SENT(task)) goto out; /* Check the SEQUENCE operation status */ @@ -2185,9 +2187,14 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs4_exception exception = { }; int err; do { - err = nfs4_handle_exception(server, - _nfs4_lookup_root(server, fhandle, info), - &exception); + err = _nfs4_lookup_root(server, fhandle, info); + switch (err) { + case 0: + case -NFS4ERR_WRONGSEC: + break; + default: + err = nfs4_handle_exception(server, err, &exception); + } } while (exception.retry); return err; } @@ -2204,31 +2211,51 @@ static int nfs4_lookup_root_sec(struct nfs_server *server, struct nfs_fh *fhandl goto out; } ret = nfs4_lookup_root(server, fhandle, info); - if (ret < 0) - ret = -EAGAIN; out: return ret; } -/* - * get the file handle for the "/" directory on the server - */ -static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, +static int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *info) { int i, len, status = 0; - rpc_authflavor_t flav_array[NFS_MAX_SECFLAVORS + 2]; + rpc_authflavor_t flav_array[NFS_MAX_SECFLAVORS]; - flav_array[0] = RPC_AUTH_UNIX; - len = gss_mech_list_pseudoflavors(&flav_array[1]); - flav_array[1+len] = RPC_AUTH_NULL; - len += 2; + len = gss_mech_list_pseudoflavors(&flav_array[0]); + flav_array[len] = RPC_AUTH_NULL; + len += 1; for (i = 0; i < len; i++) { status = nfs4_lookup_root_sec(server, fhandle, info, flav_array[i]); - if (status == 0) - break; + if (status == -NFS4ERR_WRONGSEC || status == -EACCES) + continue; + break; } + /* + * -EACCESS could mean that the user doesn't have correct permissions + * to access the mount. It could also mean that we tried to mount + * with a gss auth flavor, but rpc.gssd isn't running. Either way, + * existing mount programs don't handle -EACCES very well so it should + * be mapped to -EPERM instead. + */ + if (status == -EACCES) + status = -EPERM; + return status; +} + +/* + * get the file handle for the "/" directory on the server + */ +static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, + struct nfs_fsinfo *info) +{ + int status = nfs4_lookup_root(server, fhandle, info); + if ((status == -NFS4ERR_WRONGSEC) && !(server->flags & NFS_MOUNT_SECFLAVOUR)) + /* + * A status of -NFS4ERR_WRONGSEC will be mapped to -EPERM + * by nfs4_map_errors() as this function exits. + */ + status = nfs4_find_root_sec(server, fhandle, info); if (status == 0) status = nfs4_server_capabilities(server, fhandle); if (status == 0) @@ -3669,6 +3696,7 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, rpc_delay(task, NFS4_POLL_RETRY_MAX); task->tk_status = 0; return -EAGAIN; + case -NFS4ERR_RETRY_UNCACHED_REP: case -NFS4ERR_OLD_STATEID: task->tk_status = 0; return -EAGAIN; @@ -3725,21 +3753,20 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, sizeof(setclientid.sc_uaddr), "%s.%u.%u", clp->cl_ipaddr, port >> 8, port & 255); - status = rpc_call_sync(clp->cl_rpcclient, &msg, 0); + status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (status != -NFS4ERR_CLID_INUSE) break; - if (signalled()) + if (loop != 0) { + ++clp->cl_id_uniquifier; break; - if (loop++ & 1) - ssleep(clp->cl_lease_time / HZ + 1); - else - if (++clp->cl_id_uniquifier == 0) - break; + } + ++loop; + ssleep(clp->cl_lease_time / HZ + 1); } return status; } -static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, +int nfs4_proc_setclientid_confirm(struct nfs_client *clp, struct nfs4_setclientid_res *arg, struct rpc_cred *cred) { @@ -3754,7 +3781,7 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, int status; now = jiffies; - status = rpc_call_sync(clp->cl_rpcclient, &msg, 0); + status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (status == 0) { spin_lock(&clp->cl_lock); clp->cl_lease_time = fsinfo.lease_time * HZ; @@ -3764,26 +3791,6 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp, return status; } -int nfs4_proc_setclientid_confirm(struct nfs_client *clp, - struct nfs4_setclientid_res *arg, - struct rpc_cred *cred) -{ - long timeout = 0; - int err; - do { - err = _nfs4_proc_setclientid_confirm(clp, arg, cred); - switch (err) { - case 0: - return err; - case -NFS4ERR_RESOURCE: - /* The IBM lawyers misread another document! */ - case -NFS4ERR_DELAY: - err = nfs4_delay(clp->cl_rpcclient, &timeout); - } - } while (err == 0); - return err; -} - struct nfs4_delegreturndata { struct nfs4_delegreturnargs args; struct nfs4_delegreturnres res; @@ -4788,7 +4795,7 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred) init_utsname()->domainname, clp->cl_rpcclient->cl_auth->au_flavor); - status = rpc_call_sync(clp->cl_rpcclient, &msg, 0); + status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (!status) status = nfs4_check_cl_exchange_flags(clp->cl_exchange_flags); dprintk("<-- %s status= %d\n", __func__, status); @@ -4839,6 +4846,8 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status); rpc_delay(task, NFS4_POLL_RETRY_MIN); task->tk_status = 0; + /* fall through */ + case -NFS4ERR_RETRY_UNCACHED_REP: nfs_restart_rpc(task, data->clp); return; } @@ -4871,7 +4880,8 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) .rpc_client = clp->cl_rpcclient, .rpc_message = &msg, .callback_ops = &nfs4_get_lease_time_ops, - .callback_data = &data + .callback_data = &data, + .flags = RPC_TASK_TIMEOUT, }; int status; @@ -5173,7 +5183,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp) nfs4_init_channel_attrs(&args); args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN); - status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0); + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (!status) /* Verify the session's negotiated channel_attrs values */ @@ -5196,20 +5206,10 @@ int nfs4_proc_create_session(struct nfs_client *clp) int status; unsigned *ptr; struct nfs4_session *session = clp->cl_session; - long timeout = 0; - int err; dprintk("--> %s clp=%p session=%p\n", __func__, clp, session); - do { - status = _nfs4_proc_create_session(clp); - if (status == -NFS4ERR_DELAY) { - err = nfs4_delay(clp->cl_rpcclient, &timeout); - if (err) - status = err; - } - } while (status == -NFS4ERR_DELAY); - + status = _nfs4_proc_create_session(clp); if (status) goto out; @@ -5250,7 +5250,7 @@ int nfs4_proc_destroy_session(struct nfs4_session *session) msg.rpc_argp = session; msg.rpc_resp = NULL; msg.rpc_cred = NULL; - status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0); + status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT); if (status) printk(KERN_WARNING @@ -5483,6 +5483,8 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf break; case -NFS4ERR_DELAY: rpc_delay(task, NFS4_POLL_RETRY_MAX); + /* fall through */ + case -NFS4ERR_RETRY_UNCACHED_REP: return -EAGAIN; default: nfs4_schedule_lease_recovery(clp); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index a6804f704d9d..036f5adc9e1f 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -64,10 +64,15 @@ static LIST_HEAD(nfs4_clientid_list); int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) { - struct nfs4_setclientid_res clid; + struct nfs4_setclientid_res clid = { + .clientid = clp->cl_clientid, + .confirm = clp->cl_confirm, + }; unsigned short port; int status; + if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state)) + goto do_confirm; port = nfs_callback_tcpport; if (clp->cl_addr.ss_family == AF_INET6) port = nfs_callback_tcpport6; @@ -75,10 +80,14 @@ int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid); if (status != 0) goto out; + clp->cl_clientid = clid.clientid; + clp->cl_confirm = clid.confirm; + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); +do_confirm: status = nfs4_proc_setclientid_confirm(clp, &clid, cred); if (status != 0) goto out; - clp->cl_clientid = clid.clientid; + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); nfs4_schedule_state_renewal(clp); out: return status; @@ -230,13 +239,18 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) { int status; + if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state)) + goto do_confirm; nfs4_begin_drain_session(clp); status = nfs4_proc_exchange_id(clp, cred); if (status != 0) goto out; + set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); +do_confirm: status = nfs4_proc_create_session(clp); if (status != 0) goto out; + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); nfs41_setup_state_renewal(clp); nfs_mark_client_ready(clp, NFS_CS_READY); out: @@ -1584,20 +1598,23 @@ static int nfs4_recall_slot(struct nfs_client *clp) { return 0; } */ static void nfs4_set_lease_expired(struct nfs_client *clp, int status) { - if (nfs4_has_session(clp)) { - switch (status) { - case -NFS4ERR_DELAY: - case -NFS4ERR_CLID_INUSE: - case -EAGAIN: - break; + switch (status) { + case -NFS4ERR_CLID_INUSE: + case -NFS4ERR_STALE_CLIENTID: + clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state); + break; + case -NFS4ERR_DELAY: + case -ETIMEDOUT: + case -EAGAIN: + ssleep(1); + break; - case -EKEYEXPIRED: - nfs4_warn_keyexpired(clp->cl_hostname); - case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery - * in nfs4_exchange_id */ - default: - return; - } + case -EKEYEXPIRED: + nfs4_warn_keyexpired(clp->cl_hostname); + case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery + * in nfs4_exchange_id */ + default: + return; } set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state); } @@ -1607,7 +1624,7 @@ static void nfs4_state_manager(struct nfs_client *clp) int status = 0; /* Ensure exclusive access to NFSv4 state */ - for(;;) { + do { if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) { /* We're going to have to re-establish a clientid */ status = nfs4_reclaim_lease(clp); @@ -1691,7 +1708,7 @@ static void nfs4_state_manager(struct nfs_client *clp) break; if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0) break; - } + } while (atomic_read(&clp->cl_count) > 1); return; out_error: printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s" diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index dddfb5795d7b..c3ccd2c46834 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -1452,26 +1452,25 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args, static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr) { - uint32_t attrs[2] = {0, 0}; + uint32_t attrs[2] = { + FATTR4_WORD0_RDATTR_ERROR, + FATTR4_WORD1_MOUNTED_ON_FILEID, + }; uint32_t dircount = readdir->count >> 1; __be32 *p; if (readdir->plus) { attrs[0] |= FATTR4_WORD0_TYPE|FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE| - FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE; + FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE|FATTR4_WORD0_FILEID; attrs[1] |= FATTR4_WORD1_MODE|FATTR4_WORD1_NUMLINKS|FATTR4_WORD1_OWNER| FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV| FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS| FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY; dircount >>= 1; } - attrs[0] |= FATTR4_WORD0_RDATTR_ERROR|FATTR4_WORD0_FILEID; - attrs[1] |= FATTR4_WORD1_MOUNTED_ON_FILEID; - /* Switch to mounted_on_fileid if the server supports it */ - if (readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID) - attrs[0] &= ~FATTR4_WORD0_FILEID; - else - attrs[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; + /* Use mounted_on_fileid only if the server supports it */ + if (!(readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)) + attrs[0] |= FATTR4_WORD0_FILEID; p = reserve_space(xdr, 12+NFS4_VERIFIER_SIZE+20); *p++ = cpu_to_be32(OP_READDIR); @@ -3140,7 +3139,7 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma goto out_overflow; xdr_decode_hyper(p, fileid); bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID; - ret = NFS_ATTR_FATTR_FILEID; + ret = NFS_ATTR_FATTR_MOUNTED_ON_FILEID; } dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid); return ret; @@ -4002,7 +4001,6 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, { int status; umode_t fmode = 0; - uint64_t fileid; uint32_t type; status = decode_attr_type(xdr, bitmap, &type); @@ -4101,13 +4099,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, goto xdr_error; fattr->valid |= status; - status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid); + status = decode_attr_mounted_on_fileid(xdr, bitmap, &fattr->mounted_on_fileid); if (status < 0) goto xdr_error; - if (status != 0 && !(fattr->valid & status)) { - fattr->fileid = fileid; - fattr->valid |= status; - } + fattr->valid |= status; xdr_error: dprintk("%s: xdr returned %d\n", __func__, -status); @@ -4838,17 +4833,21 @@ static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res) struct nfs4_secinfo_flavor *sec_flavor; int status; __be32 *p; - int i; + int i, num_flavors; status = decode_op_hdr(xdr, OP_SECINFO); + if (status) + goto out; p = xdr_inline_decode(xdr, 4); if (unlikely(!p)) goto out_overflow; - res->flavors->num_flavors = be32_to_cpup(p); - for (i = 0; i < res->flavors->num_flavors; i++) { + res->flavors->num_flavors = 0; + num_flavors = be32_to_cpup(p); + + for (i = 0; i < num_flavors; i++) { sec_flavor = &res->flavors->flavors[i]; - if ((char *)&sec_flavor[1] - (char *)res > PAGE_SIZE) + if ((char *)&sec_flavor[1] - (char *)res->flavors > PAGE_SIZE) break; p = xdr_inline_decode(xdr, 4); @@ -4857,13 +4856,15 @@ static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res) sec_flavor->flavor = be32_to_cpup(p); if (sec_flavor->flavor == RPC_AUTH_GSS) { - if (decode_secinfo_gss(xdr, sec_flavor)) - break; + status = decode_secinfo_gss(xdr, sec_flavor); + if (status) + goto out; } + res->flavors->num_flavors++; } - return 0; - +out: + return status; out_overflow: print_overflow_msg(__func__, xdr); return -EIO; @@ -6408,7 +6409,9 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh, entry->server, 1) < 0) goto out_overflow; - if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID) + if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) + entry->ino = entry->fattr->mounted_on_fileid; + else if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID) entry->ino = entry->fattr->fileid; entry->d_type = DT_UNKNOWN; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index d9ab97269ce6..f57f5281a520 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -383,6 +383,7 @@ pnfs_destroy_all_layouts(struct nfs_client *clp) plh_layouts); dprintk("%s freeing layout for inode %lu\n", __func__, lo->plh_inode->i_ino); + list_del_init(&lo->plh_layouts); pnfs_destroy_layout(NFS_I(lo->plh_inode)); } } @@ -466,7 +467,8 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo, static struct pnfs_layout_segment * send_layoutget(struct pnfs_layout_hdr *lo, struct nfs_open_context *ctx, - u32 iomode) + u32 iomode, + gfp_t gfp_flags) { struct inode *ino = lo->plh_inode; struct nfs_server *server = NFS_SERVER(ino); @@ -479,7 +481,7 @@ send_layoutget(struct pnfs_layout_hdr *lo, dprintk("--> %s\n", __func__); BUG_ON(ctx == NULL); - lgp = kzalloc(sizeof(*lgp), GFP_KERNEL); + lgp = kzalloc(sizeof(*lgp), gfp_flags); if (lgp == NULL) return NULL; @@ -487,12 +489,12 @@ send_layoutget(struct pnfs_layout_hdr *lo, max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz; max_pages = max_resp_sz >> PAGE_SHIFT; - pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL); + pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags); if (!pages) goto out_err_free; for (i = 0; i < max_pages; i++) { - pages[i] = alloc_page(GFP_KERNEL); + pages[i] = alloc_page(gfp_flags); if (!pages[i]) goto out_err_free; } @@ -508,6 +510,7 @@ send_layoutget(struct pnfs_layout_hdr *lo, lgp->args.layout.pages = pages; lgp->args.layout.pglen = max_pages * PAGE_SIZE; lgp->lsegpp = &lseg; + lgp->gfp_flags = gfp_flags; /* Synchronously retrieve layout information from server and * store in lseg. @@ -665,11 +668,11 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo, } static struct pnfs_layout_hdr * -alloc_init_layout_hdr(struct inode *ino) +alloc_init_layout_hdr(struct inode *ino, gfp_t gfp_flags) { struct pnfs_layout_hdr *lo; - lo = kzalloc(sizeof(struct pnfs_layout_hdr), GFP_KERNEL); + lo = kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags); if (!lo) return NULL; atomic_set(&lo->plh_refcount, 1); @@ -681,7 +684,7 @@ alloc_init_layout_hdr(struct inode *ino) } static struct pnfs_layout_hdr * -pnfs_find_alloc_layout(struct inode *ino) +pnfs_find_alloc_layout(struct inode *ino, gfp_t gfp_flags) { struct nfs_inode *nfsi = NFS_I(ino); struct pnfs_layout_hdr *new = NULL; @@ -696,7 +699,7 @@ pnfs_find_alloc_layout(struct inode *ino) return nfsi->layout; } spin_unlock(&ino->i_lock); - new = alloc_init_layout_hdr(ino); + new = alloc_init_layout_hdr(ino, gfp_flags); spin_lock(&ino->i_lock); if (likely(nfsi->layout == NULL)) /* Won the race? */ @@ -756,7 +759,8 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode) struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, - enum pnfs_iomode iomode) + enum pnfs_iomode iomode, + gfp_t gfp_flags) { struct nfs_inode *nfsi = NFS_I(ino); struct nfs_client *clp = NFS_SERVER(ino)->nfs_client; @@ -767,7 +771,7 @@ pnfs_update_layout(struct inode *ino, if (!pnfs_enabled_sb(NFS_SERVER(ino))) return NULL; spin_lock(&ino->i_lock); - lo = pnfs_find_alloc_layout(ino); + lo = pnfs_find_alloc_layout(ino, gfp_flags); if (lo == NULL) { dprintk("%s ERROR: can't get pnfs_layout_hdr\n", __func__); goto out_unlock; @@ -807,7 +811,7 @@ pnfs_update_layout(struct inode *ino, spin_unlock(&clp->cl_lock); } - lseg = send_layoutget(lo, ctx, iomode); + lseg = send_layoutget(lo, ctx, iomode, gfp_flags); if (!lseg && first) { spin_lock(&clp->cl_lock); list_del_init(&lo->plh_layouts); @@ -846,7 +850,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) goto out; } /* Inject layout blob into I/O device driver */ - lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res); + lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res, lgp->gfp_flags); if (!lseg || IS_ERR(lseg)) { if (!lseg) status = -ENOMEM; @@ -899,7 +903,8 @@ static int pnfs_read_pg_test(struct nfs_pageio_descriptor *pgio, /* This is first coelesce call for a series of nfs_pages */ pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, prev->wb_context, - IOMODE_READ); + IOMODE_READ, + GFP_KERNEL); } return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req); } @@ -921,7 +926,8 @@ static int pnfs_write_pg_test(struct nfs_pageio_descriptor *pgio, /* This is first coelesce call for a series of nfs_pages */ pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, prev->wb_context, - IOMODE_RW); + IOMODE_RW, + GFP_NOFS); } return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req); } @@ -1004,6 +1010,7 @@ pnfs_set_layoutcommit(struct nfs_write_data *wdata) { struct nfs_inode *nfsi = NFS_I(wdata->inode); loff_t end_pos = wdata->args.offset + wdata->res.count; + bool mark_as_dirty = false; spin_lock(&nfsi->vfs_inode.i_lock); if (!test_and_set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) { @@ -1011,13 +1018,18 @@ pnfs_set_layoutcommit(struct nfs_write_data *wdata) get_lseg(wdata->lseg); wdata->lseg->pls_lc_cred = get_rpccred(wdata->args.context->state->owner->so_cred); - mark_inode_dirty_sync(wdata->inode); + mark_as_dirty = true; dprintk("%s: Set layoutcommit for inode %lu ", __func__, wdata->inode->i_ino); } if (end_pos > wdata->lseg->pls_end_pos) wdata->lseg->pls_end_pos = end_pos; spin_unlock(&nfsi->vfs_inode.i_lock); + + /* if pnfs_layoutcommit_inode() runs between inode locks, the next one + * will be a noop because NFS_INO_LAYOUTCOMMIT will not be set */ + if (mark_as_dirty) + mark_inode_dirty_sync(wdata->inode); } EXPORT_SYMBOL_GPL(pnfs_set_layoutcommit); diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index bc4827202e7a..0c015bad9e7a 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -70,7 +70,7 @@ struct pnfs_layoutdriver_type { const u32 id; const char *name; struct module *owner; - struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr); + struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags); void (*free_lseg) (struct pnfs_layout_segment *lseg); /* test for nfs page cache coalescing */ @@ -126,7 +126,7 @@ void get_layout_hdr(struct pnfs_layout_hdr *lo); void put_lseg(struct pnfs_layout_segment *lseg); struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, - enum pnfs_iomode access_type); + enum pnfs_iomode access_type, gfp_t gfp_flags); void set_pnfs_layoutdriver(struct nfs_server *, u32 id); void unset_pnfs_layoutdriver(struct nfs_server *); enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *, @@ -245,7 +245,7 @@ static inline void put_lseg(struct pnfs_layout_segment *lseg) static inline struct pnfs_layout_segment * pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, - enum pnfs_iomode access_type) + enum pnfs_iomode access_type, gfp_t gfp_flags) { return NULL; } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 7cded2b12a05..2bcf0dc306a1 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -288,7 +288,7 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc) atomic_set(&req->wb_complete, requests); BUG_ON(desc->pg_lseg != NULL); - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL); ClearPageError(page); offset = 0; nbytes = desc->pg_count; @@ -351,7 +351,7 @@ static int nfs_pagein_one(struct nfs_pageio_descriptor *desc) } req = nfs_list_entry(data->pages.next); if ((!lseg) && list_is_singular(&data->pages)) - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL); ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count, 0, lseg); diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 2b8e9a5e366a..e288f06d3fa7 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1004,6 +1004,7 @@ static int nfs_parse_security_flavors(char *value, return 0; } + mnt->flags |= NFS_MOUNT_SECFLAVOUR; mnt->auth_flavor_len = 1; return 1; } @@ -1976,6 +1977,15 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) if (error < 0) goto out; + /* + * noac is a special case. It implies -o sync, but that's not + * necessarily reflected in the mtab options. do_remount_sb + * will clear MS_SYNCHRONOUS if -o sync wasn't specified in the + * remount options, so we have to explicitly reset it. + */ + if (data->flags & NFS_MOUNT_NOAC) + *flags |= MS_SYNCHRONOUS; + /* compare new mount options with old ones */ error = nfs_compare_remount_data(nfss, data); out: @@ -2235,8 +2245,7 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type, if (!s->s_root) { /* initial superblock/root creation */ nfs_fill_super(s, data); - nfs_fscache_get_super_cookie( - s, data ? data->fscache_uniq : NULL, NULL); + nfs_fscache_get_super_cookie(s, data->fscache_uniq, NULL); } mntroot = nfs_get_root(s, mntfh, dev_name); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index af0c6279a4a7..49c715b4ac92 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -542,11 +542,15 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, pgoff_t idx_start, u if (!nfs_need_commit(nfsi)) return 0; + spin_lock(&inode->i_lock); ret = nfs_scan_list(nfsi, dst, idx_start, npages, NFS_PAGE_TAG_COMMIT); if (ret > 0) nfsi->ncommit -= ret; + spin_unlock(&inode->i_lock); + if (nfs_need_commit(NFS_I(inode))) __mark_inode_dirty(inode, I_DIRTY_DATASYNC); + return ret; } #else @@ -676,7 +680,6 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, req = nfs_setup_write_request(ctx, page, offset, count); if (IS_ERR(req)) return PTR_ERR(req); - nfs_mark_request_dirty(req); /* Update file length */ nfs_grow_file(page, offset, count); nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes); @@ -936,7 +939,7 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc) atomic_set(&req->wb_complete, requests); BUG_ON(desc->pg_lseg); - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS); ClearPageError(page); offset = 0; nbytes = desc->pg_count; @@ -1010,7 +1013,7 @@ static int nfs_flush_one(struct nfs_pageio_descriptor *desc) } req = nfs_list_entry(data->pages.next); if ((!lseg) && list_is_singular(&data->pages)) - lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW); + lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS); if ((desc->pg_ioflags & FLUSH_COND_STABLE) && (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit)) @@ -1414,8 +1417,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) task->tk_pid, task->tk_status); /* Call the NFS version-specific code */ - if (NFS_PROTO(data->inode)->commit_done(task, data) != 0) - return; + NFS_PROTO(data->inode)->commit_done(task, data); } void nfs_commit_release_pages(struct nfs_write_data *data) @@ -1483,9 +1485,7 @@ int nfs_commit_inode(struct inode *inode, int how) res = nfs_commit_set_lock(NFS_I(inode), may_wait); if (res <= 0) goto out_mark_dirty; - spin_lock(&inode->i_lock); res = nfs_scan_commit(inode, &head, 0, 0); - spin_unlock(&inode->i_lock); if (res) { int error; diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index ec0f277be7f5..6940439bd609 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -173,7 +173,7 @@ xdr_nfsace_decode(struct xdr_array2_desc *desc, void *elem) return -EINVAL; break; case ACL_MASK: - /* Solaris sometimes sets additonal bits in the mask */ + /* Solaris sometimes sets additional bits in the mask */ entry->e_perm &= S_IRWXO; break; default: diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 0c6d81670137..7c831a2731fa 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -38,7 +38,6 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) exp_readlock(); nfserr = nfsd_open(rqstp, &fh, S_IFREG, NFSD_MAY_LOCK, filp); fh_put(&fh); - rqstp->rq_client = NULL; exp_readunlock(); /* We return nlm error codes as nlm doesn't know * about nfsd, but nfsd does know about nlm.. diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 7e84a852cdae..ad48faca20fc 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -702,7 +702,7 @@ nfs3svc_encode_readres(struct svc_rqst *rqstp, __be32 *p, *p++ = htonl(resp->eof); *p++ = htonl(resp->count); /* xdr opaque count */ xdr_ressize_check(rqstp, p); - /* now update rqstp->rq_res to reflect data aswell */ + /* now update rqstp->rq_res to reflect data as well */ rqstp->rq_res.page_len = resp->count; if (resp->count & 3) { /* need to pad the tail */ diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index fbde6f79922e..4cf04e11c66c 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -258,6 +258,7 @@ static void nfs4_put_deleg_lease(struct nfs4_file *fp) if (atomic_dec_and_test(&fp->fi_delegees)) { vfs_setlease(fp->fi_deleg_file, F_UNLCK, &fp->fi_lease); fp->fi_lease = NULL; + fput(fp->fi_deleg_file); fp->fi_deleg_file = NULL; } } @@ -397,9 +398,12 @@ static void unhash_generic_stateid(struct nfs4_stateid *stp) static void free_generic_stateid(struct nfs4_stateid *stp) { - int oflag = nfs4_access_bmap_to_omode(stp); + int oflag; - nfs4_file_put_access(stp->st_file, oflag); + if (stp->st_access_bmap) { + oflag = nfs4_access_bmap_to_omode(stp); + nfs4_file_put_access(stp->st_file, oflag); + } put_nfs4_file(stp->st_file); kmem_cache_free(stateid_slab, stp); } @@ -3055,7 +3059,7 @@ check_special_stateids(svc_fh *current_fh, stateid_t *stateid, int flags) if (ONE_STATEID(stateid) && (flags & RD_STATE)) return nfs_ok; else if (locks_in_grace()) { - /* Answer in remaining cases depends on existance of + /* Answer in remaining cases depends on existence of * conflicting state; so we must wait out the grace period. */ return nfserr_grace; } else if (flags & WR_STATE) @@ -3675,7 +3679,7 @@ find_lockstateowner_str(struct inode *inode, clientid_t *clid, /* * Alloc a lock owner structure. * Called in nfsd4_lock - therefore, OPEN and OPEN_CONFIRM (if needed) has - * occured. + * occurred. * * strhashval = lock_ownerstr_hashval */ diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 4ce005dbf3e6..65ec595e2226 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -451,7 +451,7 @@ nfssvc_encode_readres(struct svc_rqst *rqstp, __be32 *p, *p++ = htonl(resp->count); xdr_ressize_check(rqstp, p); - /* now update rqstp->rq_res to reflect data aswell */ + /* now update rqstp->rq_res to reflect data as well */ rqstp->rq_res.page_len = resp->count; if (resp->count & 3) { /* need to pad the tail */ diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 2e1cebde90df..129f3c9f62d5 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1363,7 +1363,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; if (!(iap->ia_valid & ATTR_MODE)) iap->ia_mode = 0; - err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE); + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC); if (err) goto out; @@ -1385,6 +1385,13 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dchild)) goto out_nfserr; + /* If file doesn't exist, check for permissions to create one */ + if (!dchild->d_inode) { + err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE); + if (err) + goto out; + } + err = fh_compose(resfhp, fhp->fh_export, dchild, fhp); if (err) goto out; diff --git a/fs/nilfs2/alloc.c b/fs/nilfs2/alloc.c index 0a0a66d98cce..f7684483785e 100644 --- a/fs/nilfs2/alloc.c +++ b/fs/nilfs2/alloc.c @@ -646,7 +646,7 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems) unsigned long group, group_offset; int i, j, n, ret; - for (i = 0; i < nitems; i += n) { + for (i = 0; i < nitems; i = j) { group = nilfs_palloc_group(inode, entry_nrs[i], &group_offset); ret = nilfs_palloc_get_desc_block(inode, group, 0, &desc_bh); if (ret < 0) diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 6b1305dc26c0..9fde1c00a296 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -164,7 +164,7 @@ static int process_access_response(struct fsnotify_group *group, fd, response); /* * make sure the response is valid, if invalid we do nothing and either - * userspace can send a valid responce or we will clean it up after the + * userspace can send a valid response or we will clean it up after the * timeout */ switch (response) { diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c index 0348d0c8f65e..e3cbd746f64a 100644 --- a/fs/notify/inotify/inotify_fsnotify.c +++ b/fs/notify/inotify/inotify_fsnotify.c @@ -194,7 +194,7 @@ static int idr_callback(int id, void *p, void *data) static void inotify_free_group_priv(struct fsnotify_group *group) { - /* ideally the idr is empty and we won't hit the BUG in teh callback */ + /* ideally the idr is empty and we won't hit the BUG in the callback */ idr_for_each(&group->inotify_data.idr, idr_callback, group); idr_remove_all(&group->inotify_data.idr); idr_destroy(&group->inotify_data.idr); diff --git a/fs/notify/mark.c b/fs/notify/mark.c index 50c00856f730..252ab1f6452b 100644 --- a/fs/notify/mark.c +++ b/fs/notify/mark.c @@ -24,7 +24,7 @@ * referencing this object. The object typically will live inside the kernel * with a refcnt of 2, one for each list it is on (i_list, g_list). Any task * which can find this object holding the appropriete locks, can take a reference - * and the object itself is guarenteed to survive until the reference is dropped. + * and the object itself is guaranteed to survive until the reference is dropped. * * LOCKING: * There are 3 spinlocks involved with fsnotify inode marks and they MUST diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index f5094ee224c1..f14fde2b03d6 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -197,7 +197,7 @@ err_out: } else if (ctx_needs_reset) { /* * If there is no attribute list, restoring the search context - * is acomplished simply by copying the saved context back over + * is accomplished simply by copying the saved context back over * the caller supplied context. If there is an attribute list, * things are more complicated as we need to deal with mapping * of mft records and resulting potential changes in pointers. @@ -1181,7 +1181,7 @@ not_found: * for, i.e. if one wants to add the attribute to the mft record this is the * correct place to insert its attribute list entry into. * - * When -errno != -ENOENT, an error occured during the lookup. @ctx->attr is + * When -errno != -ENOENT, an error occurred during the lookup. @ctx->attr is * then undefined and in particular you should not rely on it not changing. */ int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name, diff --git a/fs/ntfs/compress.c b/fs/ntfs/compress.c index ef9ed854255c..ee4144ce5d7c 100644 --- a/fs/ntfs/compress.c +++ b/fs/ntfs/compress.c @@ -501,7 +501,7 @@ int ntfs_read_compressed_block(struct page *page) VCN start_vcn = (((s64)index << PAGE_CACHE_SHIFT) & ~cb_size_mask) >> vol->cluster_size_bits; /* - * The first vcn after the last wanted vcn (minumum alignment is again + * The first vcn after the last wanted vcn (minimum alignment is again * PAGE_CACHE_SIZE. */ VCN end_vcn = ((((s64)(index + 1UL) << PAGE_CACHE_SHIFT) + cb_size - 1) diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 0b56c6b7ec01..c05d6dcf77a4 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -622,7 +622,7 @@ static int ntfs_read_locked_inode(struct inode *vi) */ /* Everyone gets all permissions. */ vi->i_mode |= S_IRWXUGO; - /* If read-only, noone gets write permissions. */ + /* If read-only, no one gets write permissions. */ if (IS_RDONLY(vi)) vi->i_mode &= ~S_IWUGO; if (m->flags & MFT_RECORD_IS_DIRECTORY) { @@ -2529,7 +2529,7 @@ retry_truncate: * specifies that the behaviour is unspecified thus we do not * have to do anything. This means that in our implementation * in the rare case that the file is mmap()ped and a write - * occured into the mmap()ped region just beyond the file size + * occurred into the mmap()ped region just beyond the file size * and writepage has not yet been called to write out the page * (which would clear the area beyond the file size) and we now * extend the file size to incorporate this dirty region diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h index 8b2549f672bf..faece7190866 100644 --- a/fs/ntfs/layout.h +++ b/fs/ntfs/layout.h @@ -286,7 +286,7 @@ typedef le16 MFT_RECORD_FLAGS; * fragmented. Volume free space includes the empty part of the mft zone and * when the volume's free 88% are used up, the mft zone is shrunk by a factor * of 2, thus making more space available for more files/data. This process is - * repeated everytime there is no more free space except for the mft zone until + * repeated every time there is no more free space except for the mft zone until * there really is no more free space. */ @@ -1657,13 +1657,13 @@ typedef enum { * pointed to by the Owner field was provided by a defaulting mechanism * rather than explicitly provided by the original provider of the * security descriptor. This may affect the treatment of the SID with - * respect to inheritence of an owner. + * respect to inheritance of an owner. * * SE_GROUP_DEFAULTED - This boolean flag, when set, indicates that the SID in * the Group field was provided by a defaulting mechanism rather than * explicitly provided by the original provider of the security * descriptor. This may affect the treatment of the SID with respect to - * inheritence of a primary group. + * inheritance of a primary group. * * SE_DACL_PRESENT - This boolean flag, when set, indicates that the security * descriptor contains a discretionary ACL. If this flag is set and the @@ -1674,7 +1674,7 @@ typedef enum { * pointed to by the Dacl field was provided by a defaulting mechanism * rather than explicitly provided by the original provider of the * security descriptor. This may affect the treatment of the ACL with - * respect to inheritence of an ACL. This flag is ignored if the + * respect to inheritance of an ACL. This flag is ignored if the * DaclPresent flag is not set. * * SE_SACL_PRESENT - This boolean flag, when set, indicates that the security @@ -1686,7 +1686,7 @@ typedef enum { * pointed to by the Sacl field was provided by a defaulting mechanism * rather than explicitly provided by the original provider of the * security descriptor. This may affect the treatment of the ACL with - * respect to inheritence of an ACL. This flag is ignored if the + * respect to inheritance of an ACL. This flag is ignored if the * SaclPresent flag is not set. * * SE_SELF_RELATIVE - This boolean flag, when set, indicates that the security @@ -2283,7 +2283,7 @@ typedef struct { // the key_length is zero, then the vcn immediately // follows the INDEX_ENTRY_HEADER. Regardless of // key_length, the address of the 8-byte boundary - // alligned vcn of INDEX_ENTRY{_HEADER} *ie is given by + // aligned vcn of INDEX_ENTRY{_HEADER} *ie is given by // (char*)ie + le16_to_cpu(ie*)->length) - sizeof(VCN), // where sizeof(VCN) can be hardcoded as 8 if wanted. */ } __attribute__ ((__packed__)) INDEX_ENTRY; diff --git a/fs/ntfs/logfile.c b/fs/ntfs/logfile.c index 4dadcdf3d451..c71de292c5ad 100644 --- a/fs/ntfs/logfile.c +++ b/fs/ntfs/logfile.c @@ -669,7 +669,7 @@ err_out: * of cases where we think that a volume is dirty when in fact it is clean. * This should only affect volumes that have not been shutdown cleanly but did * not have any pending, non-check-pointed i/o, i.e. they were completely idle - * at least for the five seconds preceeding the unclean shutdown. + * at least for the five seconds preceding the unclean shutdown. * * This function assumes that the $LogFile journal has already been consistency * checked by a call to ntfs_check_logfile() and in particular if the $LogFile diff --git a/fs/ntfs/logfile.h b/fs/ntfs/logfile.h index b5a6f08bd35c..aa2b6ac3f0a4 100644 --- a/fs/ntfs/logfile.h +++ b/fs/ntfs/logfile.h @@ -222,7 +222,7 @@ typedef struct { /* 24*/ sle64 file_size; /* Usable byte size of the log file. If the restart_area_offset + the offset of the file_size are > 510 then corruption has - occured. This is the very first check when + occurred. This is the very first check when starting with the restart_area as if it fails it means that some of the above values will be corrupted by the multi sector diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 326e7475a22a..382857f9c7db 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -73,7 +73,7 @@ static inline MFT_RECORD *map_mft_record_page(ntfs_inode *ni) if (index > end_index || (i_size & ~PAGE_CACHE_MASK) < ofs + vol->mft_record_size) { page = ERR_PTR(-ENOENT); - ntfs_error(vol->sb, "Attemt to read mft record 0x%lx, " + ntfs_error(vol->sb, "Attempt to read mft record 0x%lx, " "which is beyond the end of the mft. " "This is probably a bug in the ntfs " "driver.", ni->mft_no); @@ -1442,7 +1442,7 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol) // Note: It will need to be a special mft record and if none of // those are available it gets rather complicated... ntfs_error(vol->sb, "Not enough space in this mft record to " - "accomodate extended mft bitmap attribute " + "accommodate extended mft bitmap attribute " "extent. Cannot handle this yet."); ret = -EOPNOTSUPP; goto undo_alloc; @@ -1879,7 +1879,7 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol) // and we would then need to update all references to this mft // record appropriately. This is rather complicated... ntfs_error(vol->sb, "Not enough space in this mft record to " - "accomodate extended mft data attribute " + "accommodate extended mft data attribute " "extent. Cannot handle this yet."); ret = -EOPNOTSUPP; goto undo_alloc; @@ -2357,7 +2357,7 @@ ntfs_inode *ntfs_mft_record_alloc(ntfs_volume *vol, const int mode, } #ifdef DEBUG read_lock_irqsave(&mftbmp_ni->size_lock, flags); - ntfs_debug("Status of mftbmp after initialized extention: " + ntfs_debug("Status of mftbmp after initialized extension: " "allocated_size 0x%llx, data_size 0x%llx, " "initialized_size 0x%llx.", (long long)mftbmp_ni->allocated_size, diff --git a/fs/ntfs/runlist.c b/fs/ntfs/runlist.c index 56a9a6d25a2a..eac7d6788a10 100644 --- a/fs/ntfs/runlist.c +++ b/fs/ntfs/runlist.c @@ -1243,7 +1243,7 @@ err_out: * write. * * This is used when building the mapping pairs array of a runlist to compress - * a given logical cluster number (lcn) or a specific run length to the minumum + * a given logical cluster number (lcn) or a specific run length to the minimum * size possible. * * Return the number of bytes written on success. On error, i.e. the diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 29099a07b9fe..b52706da4645 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -458,7 +458,7 @@ static int ntfs_remount(struct super_block *sb, int *flags, char *opt) * the volume on boot and updates them. * * When remounting read-only, mark the volume clean if no volume errors - * have occured. + * have occurred. */ if ((sb->s_flags & MS_RDONLY) && !(*flags & MS_RDONLY)) { static const char *es = ". Cannot remount read-write."; @@ -1269,7 +1269,7 @@ static int check_windows_hibernation_status(ntfs_volume *vol) "hibernated on the volume."); return 0; } - /* A real error occured. */ + /* A real error occurred. */ ntfs_error(vol->sb, "Failed to find inode number for " "hiberfil.sys."); return ret; @@ -1370,7 +1370,7 @@ static bool load_and_init_quota(ntfs_volume *vol) NVolSetQuotaOutOfDate(vol); return true; } - /* A real error occured. */ + /* A real error occurred. */ ntfs_error(vol->sb, "Failed to find inode number for $Quota."); return false; } @@ -1454,7 +1454,7 @@ not_enabled: NVolSetUsnJrnlStamped(vol); return true; } - /* A real error occured. */ + /* A real error occurred. */ ntfs_error(vol->sb, "Failed to find inode number for " "$UsnJrnl."); return false; @@ -2292,7 +2292,7 @@ static void ntfs_put_super(struct super_block *sb) ntfs_commit_inode(vol->mft_ino); /* - * If a read-write mount and no volume errors have occured, mark the + * If a read-write mount and no volume errors have occurred, mark the * volume clean. Also, re-commit all affected inodes. */ if (!(sb->s_flags & MS_RDONLY)) { @@ -2496,7 +2496,7 @@ static s64 get_nr_free_clusters(ntfs_volume *vol) if (vol->nr_clusters & 63) nr_free += 64 - (vol->nr_clusters & 63); up_read(&vol->lcnbmp_lock); - /* If errors occured we may well have gone below zero, fix this. */ + /* If errors occurred we may well have gone below zero, fix this. */ if (nr_free < 0) nr_free = 0; ntfs_debug("Exiting."); @@ -2561,7 +2561,7 @@ static unsigned long __get_nr_free_mft_records(ntfs_volume *vol, } ntfs_debug("Finished reading $MFT/$BITMAP, last index = 0x%lx.", index - 1); - /* If errors occured we may well have gone below zero, fix this. */ + /* If errors occurred we may well have gone below zero, fix this. */ if (nr_free < 0) nr_free = 0; ntfs_debug("Exiting."); diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index b27a0d86f8c5..48aa9c7401c7 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -4519,7 +4519,7 @@ set_tail_append: } /* - * Helper function called at the begining of an insert. + * Helper function called at the beginning of an insert. * * This computes a few things that are commonly used in the process of * inserting into the btree: diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h index eceb456037c1..75cf3ad987a6 100644 --- a/fs/ocfs2/aops.h +++ b/fs/ocfs2/aops.h @@ -71,7 +71,7 @@ static inline void ocfs2_iocb_set_rw_locked(struct kiocb *iocb, int level) /* * Using a named enum representing lock types in terms of #N bit stored in - * iocb->private, which is going to be used for communication bewteen + * iocb->private, which is going to be used for communication between * ocfs2_dio_end_io() and ocfs2_file_aio_write/read(). */ enum ocfs2_iocb_lock_bits { diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 2461eb3272ed..9a3e6bbff27b 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -539,25 +539,41 @@ static int o2hb_verify_crc(struct o2hb_region *reg, /* We want to make sure that nobody is heartbeating on top of us -- * this will help detect an invalid configuration. */ -static int o2hb_check_last_timestamp(struct o2hb_region *reg) +static void o2hb_check_last_timestamp(struct o2hb_region *reg) { - int node_num, ret; struct o2hb_disk_slot *slot; struct o2hb_disk_heartbeat_block *hb_block; + char *errstr; - node_num = o2nm_this_node(); - - ret = 1; - slot = ®->hr_slots[node_num]; + slot = ®->hr_slots[o2nm_this_node()]; /* Don't check on our 1st timestamp */ - if (slot->ds_last_time) { - hb_block = slot->ds_raw_block; + if (!slot->ds_last_time) + return; - if (le64_to_cpu(hb_block->hb_seq) != slot->ds_last_time) - ret = 0; - } + hb_block = slot->ds_raw_block; + if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time && + le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation && + hb_block->hb_node == slot->ds_node_num) + return; - return ret; +#define ERRSTR1 "Another node is heartbeating on device" +#define ERRSTR2 "Heartbeat generation mismatch on device" +#define ERRSTR3 "Heartbeat sequence mismatch on device" + + if (hb_block->hb_node != slot->ds_node_num) + errstr = ERRSTR1; + else if (le64_to_cpu(hb_block->hb_generation) != + slot->ds_last_generation) + errstr = ERRSTR2; + else + errstr = ERRSTR3; + + mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), " + "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name, + slot->ds_node_num, (unsigned long long)slot->ds_last_generation, + (unsigned long long)slot->ds_last_time, hb_block->hb_node, + (unsigned long long)le64_to_cpu(hb_block->hb_generation), + (unsigned long long)le64_to_cpu(hb_block->hb_seq)); } static inline void o2hb_prepare_block(struct o2hb_region *reg, @@ -983,9 +999,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) /* With an up to date view of the slots, we can check that no * other node has been improperly configured to heartbeat in * our slot. */ - if (!o2hb_check_last_timestamp(reg)) - mlog(ML_ERROR, "Device \"%s\": another node is heartbeating " - "in our slot!\n", reg->hr_dev_name); + o2hb_check_last_timestamp(reg); /* fill in the proper info for our next heartbeat */ o2hb_prepare_block(reg, reg->hr_generation); @@ -999,8 +1013,8 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) } i = -1; - while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { - + while((i = find_next_bit(configured_nodes, + O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { change |= o2hb_check_slot(reg, ®->hr_slots[i]); } @@ -1690,6 +1704,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, struct file *filp = NULL; struct inode *inode = NULL; ssize_t ret = -EINVAL; + int live_threshold; if (reg->hr_bdev) goto out; @@ -1766,8 +1781,18 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, * A node is considered live after it has beat LIVE_THRESHOLD * times. We're not steady until we've given them a chance * _after_ our first read. + * The default threshold is bare minimum so as to limit the delay + * during mounts. For global heartbeat, the threshold doubled for the + * first region. */ - atomic_set(®->hr_steady_iterations, O2HB_LIVE_THRESHOLD + 1); + live_threshold = O2HB_LIVE_THRESHOLD; + if (o2hb_global_heartbeat_active()) { + spin_lock(&o2hb_live_lock); + if (o2hb_pop_count(&o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1) + live_threshold <<= 1; + spin_unlock(&o2hb_live_lock); + } + atomic_set(®->hr_steady_iterations, live_threshold + 1); hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s", reg->hr_item.ci_name); @@ -2275,7 +2300,7 @@ void o2hb_free_hb_set(struct config_group *group) kfree(hs); } -/* hb callback registration and issueing */ +/* hb callback registration and issuing */ static struct o2hb_callback *hbcall_from_type(enum o2hb_callback_type type) { diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index a87366750f23..8f9cea1597af 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -89,7 +89,7 @@ static void o2quo_fence_self(void) }; } -/* Indicate that a timeout occured on a hearbeat region write. The +/* Indicate that a timeout occurred on a hearbeat region write. The * other nodes in the cluster may consider us dead at that time so we * want to "fence" ourselves so that we don't scribble on the disk * after they think they've recovered us. This can't solve all @@ -261,7 +261,7 @@ void o2quo_hb_still_up(u8 node) spin_unlock(&qs->qs_lock); } -/* This is analagous to hb_up. as a node's connection comes up we delay the +/* This is analogous to hb_up. as a node's connection comes up we delay the * quorum decision until we see it heartbeating. the hold will be droped in * hb_up or hb_down. it might be perpetuated by con_err until hb_down. if * it's already heartbeating we we might be dropping a hold that conn_up got. diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index ee04ff5ee603..db5ee4b4f47a 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -565,7 +565,7 @@ static void o2net_set_nn_state(struct o2net_node *nn, * the work queue actually being up. */ if (!valid && o2net_wq) { unsigned long delay; - /* delay if we're withing a RECONNECT_DELAY of the + /* delay if we're within a RECONNECT_DELAY of the * last attempt */ delay = (nn->nn_last_connect_attempt + msecs_to_jiffies(o2net_reconnect_delay())) diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 9fe5b8fd658f..8582e3f4f120 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2868,7 +2868,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, bytes = blocks_wanted << sb->s_blocksize_bits; struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); struct ocfs2_inode_info *oi = OCFS2_I(dir); - struct ocfs2_alloc_context *data_ac; + struct ocfs2_alloc_context *data_ac = NULL; struct ocfs2_alloc_context *meta_ac = NULL; struct buffer_head *dirdata_bh = NULL; struct buffer_head *dx_root_bh = NULL; diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 7540a492eaba..3b179d6cbde0 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1614,7 +1614,8 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) spin_unlock(&dlm->spinlock); /* Support for global heartbeat and node info was added in 1.1 */ - if (dlm_protocol.pv_major > 1 || dlm_protocol.pv_minor > 0) { + if (dlm->dlm_locking_proto.pv_major > 1 || + dlm->dlm_locking_proto.pv_minor > 0) { status = dlm_send_nodeinfo(dlm, ctxt->yes_resp_map); if (status) { mlog_errno(status); diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 9d67610dfc74..84d166328cf7 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -808,7 +808,7 @@ lookup: dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); mle = NULL; - /* this is lame, but we cant wait on either + /* this is lame, but we can't wait on either * the mle or lockres waitqueue here */ if (mig) msleep(100); @@ -843,7 +843,7 @@ lookup: /* finally add the lockres to its hash bucket */ __dlm_insert_lockres(dlm, res); - /* since this lockres is new it doesnt not require the spinlock */ + /* since this lockres is new it doesn't not require the spinlock */ dlm_lockres_grab_inflight_ref_new(dlm, res); /* if this node does not become the master make sure to drop @@ -2574,6 +2574,9 @@ fail: res->state &= ~DLM_LOCK_RES_MIGRATING; wake = 1; spin_unlock(&res->spinlock); + if (dlm_is_host_down(ret)) + dlm_wait_for_node_death(dlm, target, + DLM_NODE_DEATH_WAIT_MAX); goto leave; } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 41565ae52856..89659d6dc206 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1607,6 +1607,9 @@ static void ocfs2_calc_trunc_pos(struct inode *inode, range = le32_to_cpu(rec->e_cpos) + ocfs2_rec_clusters(el, rec); if (le32_to_cpu(rec->e_cpos) >= trunc_start) { + /* + * remove an entire extent record. + */ *trunc_cpos = le32_to_cpu(rec->e_cpos); /* * Skip holes if any. @@ -1617,7 +1620,16 @@ static void ocfs2_calc_trunc_pos(struct inode *inode, *blkno = le64_to_cpu(rec->e_blkno); *trunc_end = le32_to_cpu(rec->e_cpos); } else if (range > trunc_start) { + /* + * remove a partial extent record, which means we're + * removing the last extent record. + */ *trunc_cpos = trunc_start; + /* + * skip hole if any. + */ + if (range < *trunc_end) + *trunc_end = range; *trunc_len = *trunc_end - trunc_start; coff = trunc_start - le32_to_cpu(rec->e_cpos); *blkno = le64_to_cpu(rec->e_blkno) + diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 177d3a6c2a5f..b4c8bb6b8d28 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -835,7 +835,7 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode) /* If we have allowd wipe of this inode for another node, it * will be marked here so we can safely skip it. Recovery will - * cleanup any inodes we might inadvertantly skip here. */ + * cleanup any inodes we might inadvertently skip here. */ if (oi->ip_flags & OCFS2_INODE_SKIP_DELETE) goto bail_unlock; @@ -917,7 +917,7 @@ static int ocfs2_query_inode_wipe(struct inode *inode, * the inode open lock in ocfs2_read_locked_inode(). When we * get to ->delete_inode(), each node tries to convert it's * lock to an exclusive. Trylocks are serialized by the inode - * meta data lock. If the upconvert suceeds, we know the inode + * meta data lock. If the upconvert succeeds, we know the inode * is no longer live and can be deleted. * * Though we call this with the meta data lock held, the diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index dcc2d9327150..295d56454e8b 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1260,6 +1260,9 @@ void ocfs2_complete_mount_recovery(struct ocfs2_super *osb) { struct ocfs2_journal *journal = osb->journal; + if (ocfs2_is_hard_readonly(osb)) + return; + /* No need to queue up our truncate_log as regular cleanup will catch * that */ ocfs2_queue_recovery_completion(journal, osb->slot_num, @@ -1368,7 +1371,7 @@ skip_recovery: mlog_errno(status); /* Now it is right time to recover quotas... We have to do this under - * superblock lock so that noone can start using the slot (and crash) + * superblock lock so that no one can start using the slot (and crash) * before we recover it */ for (i = 0; i < rm_quota_used; i++) { qrec = ocfs2_begin_quota_recovery(osb, rm_quota[i]); diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h index 6180da1e37e6..68cf2f6d3c6a 100644 --- a/fs/ocfs2/journal.h +++ b/fs/ocfs2/journal.h @@ -215,7 +215,7 @@ static inline void ocfs2_checkpoint_inode(struct inode *inode) /* WARNING: This only kicks off a single * checkpoint. If someone races you and adds more * metadata to the journal, you won't know, and will - * wind up waiting *alot* longer than necessary. Right + * wind up waiting *a lot* longer than necessary. Right * now we only use this in clear_inode so that's * OK. */ ocfs2_start_checkpoint(osb); diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 28f2cc1080d8..e5d738cd9cc0 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -2128,7 +2128,7 @@ leave: } /** - * ocfs2_prep_new_orphaned_file() - Prepare the orphan dir to recieve a newly + * ocfs2_prep_new_orphaned_file() - Prepare the orphan dir to receive a newly * allocated file. This is different from the typical 'add to orphan dir' * operation in that the inode does not yet exist. This is a problem because * the orphan dir stringifies the inode block number to come up with it's diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index bf2e7764920e..938387a10d5d 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -441,7 +441,7 @@ static unsigned char ocfs2_type_by_mode[S_IFMT >> S_SHIFT] = { struct ocfs2_block_check { /*00*/ __le32 bc_crc32e; /* 802.3 Ethernet II CRC32 */ __le16 bc_ecc; /* Single-error-correction parity vector. - This is a simple Hamming code dependant + This is a simple Hamming code dependent on the blocksize. OCFS2's maximum blocksize, 4K, requires 16 parity bits, so we fit in __le16. */ @@ -750,7 +750,7 @@ struct ocfs2_dinode { after an unclean shutdown */ } journal1; - } id1; /* Inode type dependant 1 */ + } id1; /* Inode type dependent 1 */ /*C0*/ union { struct ocfs2_super_block i_super; struct ocfs2_local_alloc i_lab; @@ -1019,7 +1019,7 @@ struct ocfs2_xattr_entry { __le16 xe_name_offset; /* byte offset from the 1st entry in the local xattr storage(inode, xattr block or xattr bucket). */ - __u8 xe_name_len; /* xattr name len, does't include prefix. */ + __u8 xe_name_len; /* xattr name len, doesn't include prefix. */ __u8 xe_type; /* the low 7 bits indicate the name prefix * type and the highest bit indicates whether * the EA is stored in the local storage. */ diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c index 279aef68025b..92fcd575775a 100644 --- a/fs/ocfs2/quota_global.c +++ b/fs/ocfs2/quota_global.c @@ -556,7 +556,7 @@ int __ocfs2_sync_dquot(struct dquot *dquot, int freeing) spin_unlock(&dq_data_lock); err = ocfs2_qinfo_lock(info, freeing); if (err < 0) { - mlog(ML_ERROR, "Failed to lock quota info, loosing quota write" + mlog(ML_ERROR, "Failed to lock quota info, losing quota write" " (type=%d, id=%u)\n", dquot->dq_type, (unsigned)dquot->dq_id); goto out; diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h index 1e49cc29d06c..42c2b804f3fd 100644 --- a/fs/ocfs2/reservations.h +++ b/fs/ocfs2/reservations.h @@ -29,7 +29,7 @@ struct ocfs2_alloc_reservation { struct rb_node r_node; - unsigned int r_start; /* Begining of current window */ + unsigned int r_start; /* Beginning of current window */ unsigned int r_len; /* Length of the window */ unsigned int r_last_len; /* Length of most recent alloc */ diff --git a/fs/ocfs2/stackglue.h b/fs/ocfs2/stackglue.h index 8ce7398ae1d2..1ec56fdb8d0d 100644 --- a/fs/ocfs2/stackglue.h +++ b/fs/ocfs2/stackglue.h @@ -126,7 +126,7 @@ struct ocfs2_stack_operations { * * ->connect() must not return until it is guaranteed that * - * - Node down notifications for the filesystem will be recieved + * - Node down notifications for the filesystem will be received * and passed to conn->cc_recovery_handler(). * - Locking requests for the filesystem will be processed. */ diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index ab6e2061074f..ba5d97e4a73e 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1511,7 +1511,7 @@ static int ocfs2_cluster_group_search(struct inode *inode, max_bits = le16_to_cpu(gd->bg_bits); /* Tail groups in cluster bitmaps which aren't cpg - * aligned are prone to partial extention by a failed + * aligned are prone to partial extension by a failed * fs resize. If the file system resize never got to * update the dinode cluster count, then we don't want * to trust any clusters past it, regardless of what @@ -2459,7 +2459,7 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle, /* The alloc_bh comes from ocfs2_free_dinode() or * ocfs2_free_clusters(). The callers have all locked the * allocator and gotten alloc_bh from the lock call. This - * validates the dinode buffer. Any corruption that has happended + * validates the dinode buffer. Any corruption that has happened * is a code bug. */ BUG_ON(!OCFS2_IS_VALID_DINODE(fe)); BUG_ON((count + start_bit) > ocfs2_bits_per_group(cl)); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 69fa11b35aa4..5a521c748859 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -78,7 +78,7 @@ static struct kmem_cache *ocfs2_inode_cachep = NULL; struct kmem_cache *ocfs2_dquot_cachep; struct kmem_cache *ocfs2_qf_chunk_cachep; -/* OCFS2 needs to schedule several differnt types of work which +/* OCFS2 needs to schedule several different types of work which * require cluster locking, disk I/O, recovery waits, etc. Since these * types of work tend to be heavy we avoid using the kernel events * workqueue and schedule on our own. */ diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index 57a215dc2d9b..81ecf9c0bf0a 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -3554,7 +3554,7 @@ int ocfs2_xattr_set(struct inode *inode, down_write(&OCFS2_I(inode)->ip_xattr_sem); /* * Scan inode and external block to find the same name - * extended attribute and collect search infomation. + * extended attribute and collect search information. */ ret = ocfs2_xattr_ibody_find(inode, name_index, name, &xis); if (ret) @@ -3578,7 +3578,7 @@ int ocfs2_xattr_set(struct inode *inode, goto cleanup; } - /* Check whether the value is refcounted and do some prepartion. */ + /* Check whether the value is refcounted and do some preparation. */ if (OCFS2_I(inode)->ip_dyn_features & OCFS2_HAS_REFCOUNT_FL && (!xis.not_found || !xbs.not_found)) { ret = ocfs2_prepare_refcount_xattr(inode, di, &xi, diff --git a/fs/partitions/check.c b/fs/partitions/check.c index ac546975031f..d545e97d99c3 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -500,7 +500,7 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno, /* everything is up and running, commence */ rcu_assign_pointer(ptbl->part[partno], p); - /* suppress uevent if the disk supresses it */ + /* suppress uevent if the disk suppresses it */ if (!dev_get_uevent_suppress(ddev)) kobject_uevent(&pdev->kobj, KOBJ_ADD); @@ -585,7 +585,7 @@ rescan: /* * If any partition code tried to read beyond EOD, try * unlocking native capacity even if partition table is - * sucessfully read as we could be missing some partitions. + * successfully read as we could be missing some partitions. */ if (state->access_beyond_eod) { printk(KERN_WARNING diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c index ac0ccb5026a2..19d6750d1d6c 100644 --- a/fs/partitions/efi.c +++ b/fs/partitions/efi.c @@ -348,6 +348,12 @@ static int is_gpt_valid(struct parsed_partitions *state, u64 lba, goto fail; } + /* Check that sizeof_partition_entry has the correct value */ + if (le32_to_cpu((*gpt)->sizeof_partition_entry) != sizeof(gpt_entry)) { + pr_debug("GUID Partitition Entry Size check failed.\n"); + goto fail; + } + if (!(*ptes = alloc_read_gpt_entries(state, *gpt))) goto fail; diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c index b10e3540d5b7..ce4f62440425 100644 --- a/fs/partitions/ldm.c +++ b/fs/partitions/ldm.c @@ -1299,6 +1299,11 @@ static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags) BUG_ON (!data || !frags); + if (size < 2 * VBLK_SIZE_HEAD) { + ldm_error("Value of size is to small."); + return false; + } + group = get_unaligned_be32(data + 0x08); rec = get_unaligned_be16(data + 0x0C); num = get_unaligned_be16(data + 0x0E); @@ -1306,6 +1311,10 @@ static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags) ldm_error ("A VBLK claims to have %d parts.", num); return false; } + if (rec >= num) { + ldm_error("REC value (%d) exceeds NUM value (%d)", rec, num); + return false; + } list_for_each (item, frags) { f = list_entry (item, struct frag, list); @@ -1334,10 +1343,9 @@ found: f->map |= (1 << rec); - if (num > 0) { - data += VBLK_SIZE_HEAD; - size -= VBLK_SIZE_HEAD; - } + data += VBLK_SIZE_HEAD; + size -= VBLK_SIZE_HEAD; + memcpy (f->data+rec*(size-VBLK_SIZE_HEAD)+VBLK_SIZE_HEAD, data, size); return true; diff --git a/fs/proc/base.c b/fs/proc/base.c index 5a670c11aeac..dfa532730e55 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -220,7 +220,7 @@ static struct mm_struct *__check_mem_permission(struct task_struct *task) } /* - * Noone else is allowed. + * No one else is allowed. */ mmput(mm); return ERR_PTR(-EPERM); @@ -3124,11 +3124,16 @@ static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldi /* for the /proc/ directory itself, after non-process stuff has been done */ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) { - unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY; - struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode); + unsigned int nr; + struct task_struct *reaper; struct tgid_iter iter; struct pid_namespace *ns; + if (filp->f_pos >= PID_MAX_LIMIT + TGID_OFFSET) + goto out_no_task; + nr = filp->f_pos - FIRST_PROCESS_ENTRY; + + reaper = get_proc_task(filp->f_path.dentry->d_inode); if (!reaper) goto out_no_task; diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 2e7addfd9803..318d8654989b 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -214,7 +214,7 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) int flags = vma->vm_flags; unsigned long ino = 0; unsigned long long pgoff = 0; - unsigned long start; + unsigned long start, end; dev_t dev = 0; int len; @@ -227,13 +227,15 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) /* We don't show the stack guard page in /proc/maps */ start = vma->vm_start; - if (vma->vm_flags & VM_GROWSDOWN) - if (!vma_stack_continue(vma->vm_prev, vma->vm_start)) - start += PAGE_SIZE; + if (stack_guard_page_start(vma, start)) + start += PAGE_SIZE; + end = vma->vm_end; + if (stack_guard_page_end(vma, end)) + end -= PAGE_SIZE; seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", start, - vma->vm_end, + end, flags & VM_READ ? 'r' : '-', flags & VM_WRITE ? 'w' : '-', flags & VM_EXEC ? 'x' : '-', diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig index 867d0ac026ce..8007ae7c0d8c 100644 --- a/fs/pstore/Kconfig +++ b/fs/pstore/Kconfig @@ -1,5 +1,5 @@ config PSTORE - bool "Persistant store support" + bool "Persistent store support" default n help This option enables generic access to platform level diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index fcc8ae75d874..d3c032f5fa0a 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -442,7 +442,7 @@ EXPORT_SYMBOL(dquot_acquire); */ int dquot_commit(struct dquot *dquot) { - int ret = 0, ret2 = 0; + int ret = 0; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dqopt->dqio_mutex); @@ -454,15 +454,10 @@ int dquot_commit(struct dquot *dquot) spin_unlock(&dq_list_lock); /* Inactive dquot can be only if there was error during read/init * => we have better not writing it */ - if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) { + if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot); - if (info_dirty(&dqopt->info[dquot->dq_type])) { - ret2 = dqopt->ops[dquot->dq_type]->write_file_info( - dquot->dq_sb, dquot->dq_type); - } - if (ret >= 0) - ret = ret2; - } + else + ret = -EIO; out_sem: mutex_unlock(&dqopt->dqio_mutex); return ret; @@ -956,7 +951,7 @@ static inline int dqput_blocks(struct dquot *dquot) /* * Remove references to dquots from inode and add dquot to list for freeing - * if we have the last referece to dquot + * if we have the last reference to dquot * We can't race with anybody because we hold dqptr_sem for writing... */ static int remove_inode_dquot_ref(struct inode *inode, int type, diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 9eead2c796b7..fbb0b478a346 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -112,6 +112,7 @@ int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize) SetPageDirty(page); unlock_page(page); + put_page(page); } return 0; diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index c77514bd5776..c5e82ece7c6c 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -1,7 +1,7 @@ /* ** Write ahead logging implementation copyright Chris Mason 2000 ** -** The background commits make this code very interelated, and +** The background commits make this code very interrelated, and ** overly complex. I need to rethink things a bit....The major players: ** ** journal_begin -- call with the number of blocks you expect to log. @@ -2725,7 +2725,7 @@ int journal_init(struct super_block *sb, const char *j_dev_name, REISERFS_DISK_OFFSET_IN_BYTES / sb->s_blocksize + 2); - /* Sanity check to see is the standard journal fitting withing first bitmap + /* Sanity check to see is the standard journal fitting within first bitmap (actual for small blocksizes) */ if (!SB_ONDISK_JOURNAL_DEVICE(sb) && (SB_JOURNAL_1st_RESERVED_BLOCK(sb) + diff --git a/fs/reiserfs/lock.c b/fs/reiserfs/lock.c index b87aa2c1afc1..7df1ce48203a 100644 --- a/fs/reiserfs/lock.c +++ b/fs/reiserfs/lock.c @@ -15,7 +15,7 @@ * for this mutex, no need for a system wide mutex facility. * * Also this lock is often released before a call that could block because - * reiserfs performances were partialy based on the release while schedule() + * reiserfs performances were partially based on the release while schedule() * property of the Bkl. */ void reiserfs_write_lock(struct super_block *s) diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 0aab04f46827..b216ff6be1c9 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -393,7 +393,7 @@ void add_save_link(struct reiserfs_transaction_handle *th, /* body of "save" link */ link = INODE_PKEY(inode)->k_dir_id; - /* put "save" link inot tree, don't charge quota to anyone */ + /* put "save" link into tree, don't charge quota to anyone */ retval = reiserfs_insert_item(th, &path, &key, &ih, NULL, (char *)&link); if (retval) { @@ -2104,7 +2104,7 @@ out: /* Read data from quotafile - avoid pagecache and such because we cannot afford * acquiring the locks... As quota files are never truncated and quota code - * itself serializes the operations (and noone else should touch the files) + * itself serializes the operations (and no one else should touch the files) * we don't have to be afraid of races */ static ssize_t reiserfs_quota_read(struct super_block *sb, int type, char *data, size_t len, loff_t off) diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 5c11ca82b782..47d2a4498b03 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -396,7 +396,7 @@ static struct page *reiserfs_get_page(struct inode *dir, size_t n) struct address_space *mapping = dir->i_mapping; struct page *page; /* We can deadlock if we try to free dentries, - and an unlink/rmdir has just occured - GFP_NOFS avoids this */ + and an unlink/rmdir has just occurred - GFP_NOFS avoids this */ mapping_set_gfp_mask(mapping, GFP_NOFS); page = read_mapping_page(mapping, n >> PAGE_CACHE_SHIFT, NULL); if (!IS_ERR(page)) { diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index 26b15ae34d6f..c37b520132ff 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -104,7 +104,7 @@ struct squashfs_cache_entry *squashfs_cache_get(struct super_block *sb, entry = &cache->entry[i]; /* - * Initialise choosen cache entry, and fill it in from + * Initialise chosen cache entry, and fill it in from * disk. */ cache->unused--; @@ -286,7 +286,7 @@ cleanup: /* - * Copy upto length bytes from cache entry to buffer starting at offset bytes + * Copy up to length bytes from cache entry to buffer starting at offset bytes * into the cache entry. If there's not length bytes then copy the number of * bytes available. In all cases return the number of bytes copied. */ diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index da3fefe91a8f..1ad8c93c1b85 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -24,13 +24,6 @@ #include "sysfs.h" -/* used in crash dumps to help with debugging */ -static char last_sysfs_file[PATH_MAX]; -void sysfs_printk_last_file(void) -{ - printk(KERN_EMERG "last sysfs file: %s\n", last_sysfs_file); -} - /* * There's one sysfs_buffer for each open file and one * sysfs_open_dirent for each sysfs_dirent with one or more open @@ -337,11 +330,6 @@ static int sysfs_open_file(struct inode *inode, struct file *file) struct sysfs_buffer *buffer; const struct sysfs_ops *ops; int error = -EACCES; - char *p; - - p = d_path(&file->f_path, last_sysfs_file, sizeof(last_sysfs_file)); - if (!IS_ERR(p)) - memmove(last_sysfs_file, p, strlen(p) + 1); /* need attr_sd for attr and ops, its parent for kobj */ if (!sysfs_get_active(attr_sd)) diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index c8769dc222d8..194414f8298c 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -101,9 +101,9 @@ int sysfs_create_group(struct kobject *kobj, } /** - * sysfs_update_group - given a directory kobject, create an attribute group - * @kobj: The kobject to create the group on - * @grp: The attribute group to create + * sysfs_update_group - given a directory kobject, update an attribute group + * @kobj: The kobject to update the group on + * @grp: The attribute group to update * * This function updates an attribute group. Unlike * sysfs_create_group(), it will explicitly not warn or error if any diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig index d7440904be17..f8b0160da2da 100644 --- a/fs/ubifs/Kconfig +++ b/fs/ubifs/Kconfig @@ -47,7 +47,7 @@ config UBIFS_FS_DEBUG bool "Enable debugging support" depends on UBIFS_FS select DEBUG_FS - select KALLSYMS_ALL + select KALLSYMS help This option enables UBIFS debugging support. It makes sure various assertions, self-checks, debugging messages and test modes are compiled diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index c8ff0d1ae5d3..8b3a7da531eb 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -147,7 +147,7 @@ static int make_free_space(struct ubifs_info *c) if (liab2 < liab1) return -EAGAIN; - dbg_budg("new liability %lld (not shrinked)", liab2); + dbg_budg("new liability %lld (not shrunk)", liab2); /* Liability did not shrink again, try GC */ dbg_budg("Run GC"); diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c index b148fbc80f8d..1bd01ded7123 100644 --- a/fs/ubifs/commit.c +++ b/fs/ubifs/commit.c @@ -577,7 +577,7 @@ int dbg_check_old_index(struct ubifs_info *c, struct ubifs_zbranch *zroot) size_t sz; if (!(ubifs_chk_flags & UBIFS_CHK_OLD_IDX)) - goto out; + return 0; INIT_LIST_HEAD(&list); diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index f25a7339f800..004d3745dc45 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -972,11 +972,39 @@ void dbg_dump_index(struct ubifs_info *c) void dbg_save_space_info(struct ubifs_info *c) { struct ubifs_debug_info *d = c->dbg; - - ubifs_get_lp_stats(c, &d->saved_lst); + int freeable_cnt; spin_lock(&c->space_lock); + memcpy(&d->saved_lst, &c->lst, sizeof(struct ubifs_lp_stats)); + + /* + * We use a dirty hack here and zero out @c->freeable_cnt, because it + * affects the free space calculations, and UBIFS might not know about + * all freeable eraseblocks. Indeed, we know about freeable eraseblocks + * only when we read their lprops, and we do this only lazily, upon the + * need. So at any given point of time @c->freeable_cnt might be not + * exactly accurate. + * + * Just one example about the issue we hit when we did not zero + * @c->freeable_cnt. + * 1. The file-system is mounted R/O, c->freeable_cnt is %0. We save the + * amount of free space in @d->saved_free + * 2. We re-mount R/W, which makes UBIFS to read the "lsave" + * information from flash, where we cache LEBs from various + * categories ('ubifs_remount_fs()' -> 'ubifs_lpt_init()' + * -> 'lpt_init_wr()' -> 'read_lsave()' -> 'ubifs_lpt_lookup()' + * -> 'ubifs_get_pnode()' -> 'update_cats()' + * -> 'ubifs_add_to_cat()'). + * 3. Lsave contains a freeable eraseblock, and @c->freeable_cnt + * becomes %1. + * 4. We calculate the amount of free space when the re-mount is + * finished in 'dbg_check_space_info()' and it does not match + * @d->saved_free. + */ + freeable_cnt = c->freeable_cnt; + c->freeable_cnt = 0; d->saved_free = ubifs_get_free_space_nolock(c); + c->freeable_cnt = freeable_cnt; spin_unlock(&c->space_lock); } @@ -993,12 +1021,15 @@ int dbg_check_space_info(struct ubifs_info *c) { struct ubifs_debug_info *d = c->dbg; struct ubifs_lp_stats lst; - long long avail, free; + long long free; + int freeable_cnt; spin_lock(&c->space_lock); - avail = ubifs_calc_available(c, c->min_idx_lebs); + freeable_cnt = c->freeable_cnt; + c->freeable_cnt = 0; + free = ubifs_get_free_space_nolock(c); + c->freeable_cnt = freeable_cnt; spin_unlock(&c->space_lock); - free = ubifs_get_free_space(c); if (free != d->saved_free) { ubifs_err("free space changed from %lld to %lld", @@ -2806,40 +2837,38 @@ int dbg_debugfs_init_fs(struct ubifs_info *c) struct ubifs_debug_info *d = c->dbg; sprintf(d->dfs_dir_name, "ubi%d_%d", c->vi.ubi_num, c->vi.vol_id); - d->dfs_dir = debugfs_create_dir(d->dfs_dir_name, dfs_rootdir); - if (IS_ERR(d->dfs_dir)) { - err = PTR_ERR(d->dfs_dir); - ubifs_err("cannot create \"%s\" debugfs directory, error %d\n", - d->dfs_dir_name, err); + fname = d->dfs_dir_name; + dent = debugfs_create_dir(fname, dfs_rootdir); + if (IS_ERR_OR_NULL(dent)) goto out; - } + d->dfs_dir = dent; fname = "dump_lprops"; dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); - if (IS_ERR(dent)) + if (IS_ERR_OR_NULL(dent)) goto out_remove; d->dfs_dump_lprops = dent; fname = "dump_budg"; dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); - if (IS_ERR(dent)) + if (IS_ERR_OR_NULL(dent)) goto out_remove; d->dfs_dump_budg = dent; fname = "dump_tnc"; dent = debugfs_create_file(fname, S_IWUSR, d->dfs_dir, c, &dfs_fops); - if (IS_ERR(dent)) + if (IS_ERR_OR_NULL(dent)) goto out_remove; d->dfs_dump_tnc = dent; return 0; out_remove: - err = PTR_ERR(dent); - ubifs_err("cannot create \"%s\" debugfs directory, error %d\n", - fname, err); debugfs_remove_recursive(d->dfs_dir); out: + err = dent ? PTR_ERR(dent) : -ENODEV; + ubifs_err("cannot create \"%s\" debugfs directory, error %d\n", + fname, err); return err; } diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index 919f0de29d8f..e6493cac193d 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -23,6 +23,12 @@ #ifndef __UBIFS_DEBUG_H__ #define __UBIFS_DEBUG_H__ +/* Checking helper functions */ +typedef int (*dbg_leaf_callback)(struct ubifs_info *c, + struct ubifs_zbranch *zbr, void *priv); +typedef int (*dbg_znode_callback)(struct ubifs_info *c, + struct ubifs_znode *znode, void *priv); + #ifdef CONFIG_UBIFS_FS_DEBUG /** @@ -270,11 +276,6 @@ void dbg_dump_tnc(struct ubifs_info *c); void dbg_dump_index(struct ubifs_info *c); void dbg_dump_lpt_lebs(const struct ubifs_info *c); -/* Checking helper functions */ -typedef int (*dbg_leaf_callback)(struct ubifs_info *c, - struct ubifs_zbranch *zbr, void *priv); -typedef int (*dbg_znode_callback)(struct ubifs_info *c, - struct ubifs_znode *znode, void *priv); int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb, dbg_znode_callback znode_cb, void *priv); @@ -295,7 +296,6 @@ int dbg_check_idx_size(struct ubifs_info *c, long long idx_size); int dbg_check_filesystem(struct ubifs_info *c); void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat, int add_pos); -int dbg_check_lprops(struct ubifs_info *c); int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode, int row, int col); int dbg_check_inode_size(struct ubifs_info *c, const struct inode *inode, @@ -401,58 +401,94 @@ void dbg_debugfs_exit_fs(struct ubifs_info *c); #define DBGKEY(key) ((char *)(key)) #define DBGKEY1(key) ((char *)(key)) -#define ubifs_debugging_init(c) 0 -#define ubifs_debugging_exit(c) ({}) - -#define dbg_ntype(type) "" -#define dbg_cstate(cmt_state) "" -#define dbg_jhead(jhead) "" -#define dbg_get_key_dump(c, key) ({}) -#define dbg_dump_inode(c, inode) ({}) -#define dbg_dump_node(c, node) ({}) -#define dbg_dump_lpt_node(c, node, lnum, offs) ({}) -#define dbg_dump_budget_req(req) ({}) -#define dbg_dump_lstats(lst) ({}) -#define dbg_dump_budg(c) ({}) -#define dbg_dump_lprop(c, lp) ({}) -#define dbg_dump_lprops(c) ({}) -#define dbg_dump_lpt_info(c) ({}) -#define dbg_dump_leb(c, lnum) ({}) -#define dbg_dump_znode(c, znode) ({}) -#define dbg_dump_heap(c, heap, cat) ({}) -#define dbg_dump_pnode(c, pnode, parent, iip) ({}) -#define dbg_dump_tnc(c) ({}) -#define dbg_dump_index(c) ({}) -#define dbg_dump_lpt_lebs(c) ({}) - -#define dbg_walk_index(c, leaf_cb, znode_cb, priv) 0 -#define dbg_old_index_check_init(c, zroot) 0 -#define dbg_save_space_info(c) ({}) -#define dbg_check_space_info(c) 0 -#define dbg_check_old_index(c, zroot) 0 -#define dbg_check_cats(c) 0 -#define dbg_check_ltab(c) 0 -#define dbg_chk_lpt_free_spc(c) 0 -#define dbg_chk_lpt_sz(c, action, len) 0 -#define dbg_check_synced_i_size(inode) 0 -#define dbg_check_dir_size(c, dir) 0 -#define dbg_check_tnc(c, x) 0 -#define dbg_check_idx_size(c, idx_size) 0 -#define dbg_check_filesystem(c) 0 -#define dbg_check_heap(c, heap, cat, add_pos) ({}) -#define dbg_check_lprops(c) 0 -#define dbg_check_lpt_nodes(c, cnode, row, col) 0 -#define dbg_check_inode_size(c, inode, size) 0 -#define dbg_check_data_nodes_order(c, head) 0 -#define dbg_check_nondata_nodes_order(c, head) 0 -#define dbg_force_in_the_gaps_enabled 0 -#define dbg_force_in_the_gaps() 0 -#define dbg_failure_mode 0 - -#define dbg_debugfs_init() 0 -#define dbg_debugfs_exit() -#define dbg_debugfs_init_fs(c) 0 -#define dbg_debugfs_exit_fs(c) 0 +static inline int ubifs_debugging_init(struct ubifs_info *c) { return 0; } +static inline void ubifs_debugging_exit(struct ubifs_info *c) { return; } +static inline const char *dbg_ntype(int type) { return ""; } +static inline const char *dbg_cstate(int cmt_state) { return ""; } +static inline const char *dbg_jhead(int jhead) { return ""; } +static inline const char * +dbg_get_key_dump(const struct ubifs_info *c, + const union ubifs_key *key) { return ""; } +static inline void dbg_dump_inode(const struct ubifs_info *c, + const struct inode *inode) { return; } +static inline void dbg_dump_node(const struct ubifs_info *c, + const void *node) { return; } +static inline void dbg_dump_lpt_node(const struct ubifs_info *c, + void *node, int lnum, + int offs) { return; } +static inline void +dbg_dump_budget_req(const struct ubifs_budget_req *req) { return; } +static inline void +dbg_dump_lstats(const struct ubifs_lp_stats *lst) { return; } +static inline void dbg_dump_budg(struct ubifs_info *c) { return; } +static inline void dbg_dump_lprop(const struct ubifs_info *c, + const struct ubifs_lprops *lp) { return; } +static inline void dbg_dump_lprops(struct ubifs_info *c) { return; } +static inline void dbg_dump_lpt_info(struct ubifs_info *c) { return; } +static inline void dbg_dump_leb(const struct ubifs_info *c, + int lnum) { return; } +static inline void +dbg_dump_znode(const struct ubifs_info *c, + const struct ubifs_znode *znode) { return; } +static inline void dbg_dump_heap(struct ubifs_info *c, + struct ubifs_lpt_heap *heap, + int cat) { return; } +static inline void dbg_dump_pnode(struct ubifs_info *c, + struct ubifs_pnode *pnode, + struct ubifs_nnode *parent, + int iip) { return; } +static inline void dbg_dump_tnc(struct ubifs_info *c) { return; } +static inline void dbg_dump_index(struct ubifs_info *c) { return; } +static inline void dbg_dump_lpt_lebs(const struct ubifs_info *c) { return; } + +static inline int dbg_walk_index(struct ubifs_info *c, + dbg_leaf_callback leaf_cb, + dbg_znode_callback znode_cb, + void *priv) { return 0; } +static inline void dbg_save_space_info(struct ubifs_info *c) { return; } +static inline int dbg_check_space_info(struct ubifs_info *c) { return 0; } +static inline int dbg_check_lprops(struct ubifs_info *c) { return 0; } +static inline int +dbg_old_index_check_init(struct ubifs_info *c, + struct ubifs_zbranch *zroot) { return 0; } +static inline int +dbg_check_old_index(struct ubifs_info *c, + struct ubifs_zbranch *zroot) { return 0; } +static inline int dbg_check_cats(struct ubifs_info *c) { return 0; } +static inline int dbg_check_ltab(struct ubifs_info *c) { return 0; } +static inline int dbg_chk_lpt_free_spc(struct ubifs_info *c) { return 0; } +static inline int dbg_chk_lpt_sz(struct ubifs_info *c, + int action, int len) { return 0; } +static inline int dbg_check_synced_i_size(struct inode *inode) { return 0; } +static inline int dbg_check_dir_size(struct ubifs_info *c, + const struct inode *dir) { return 0; } +static inline int dbg_check_tnc(struct ubifs_info *c, int extra) { return 0; } +static inline int dbg_check_idx_size(struct ubifs_info *c, + long long idx_size) { return 0; } +static inline int dbg_check_filesystem(struct ubifs_info *c) { return 0; } +static inline void dbg_check_heap(struct ubifs_info *c, + struct ubifs_lpt_heap *heap, + int cat, int add_pos) { return; } +static inline int dbg_check_lpt_nodes(struct ubifs_info *c, + struct ubifs_cnode *cnode, int row, int col) { return 0; } +static inline int dbg_check_inode_size(struct ubifs_info *c, + const struct inode *inode, + loff_t size) { return 0; } +static inline int +dbg_check_data_nodes_order(struct ubifs_info *c, + struct list_head *head) { return 0; } +static inline int +dbg_check_nondata_nodes_order(struct ubifs_info *c, + struct list_head *head) { return 0; } + +static inline int dbg_force_in_the_gaps(void) { return 0; } +#define dbg_force_in_the_gaps_enabled 0 +#define dbg_failure_mode 0 + +static inline int dbg_debugfs_init(void) { return 0; } +static inline void dbg_debugfs_exit(void) { return; } +static inline int dbg_debugfs_init_fs(struct ubifs_info *c) { return 0; } +static inline int dbg_debugfs_exit_fs(struct ubifs_info *c) { return 0; } #endif /* !CONFIG_UBIFS_FS_DEBUG */ #endif /* !__UBIFS_DEBUG_H__ */ diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 28be1e6a65e8..b286db79c686 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1312,6 +1312,9 @@ int ubifs_fsync(struct file *file, int datasync) dbg_gen("syncing inode %lu", inode->i_ino); + if (inode->i_sb->s_flags & MS_RDONLY) + return 0; + /* * VFS has already synchronized dirty pages for this inode. Synchronize * the inode unless this is a 'datasync()' call. diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c index 4d0cb1241460..40fa780ebea7 100644 --- a/fs/ubifs/log.c +++ b/fs/ubifs/log.c @@ -175,26 +175,6 @@ void ubifs_add_bud(struct ubifs_info *c, struct ubifs_bud *bud) } /** - * ubifs_create_buds_lists - create journal head buds lists for remount rw. - * @c: UBIFS file-system description object - */ -void ubifs_create_buds_lists(struct ubifs_info *c) -{ - struct rb_node *p; - - spin_lock(&c->buds_lock); - p = rb_first(&c->buds); - while (p) { - struct ubifs_bud *bud = rb_entry(p, struct ubifs_bud, rb); - struct ubifs_jhead *jhead = &c->jheads[bud->jhead]; - - list_add_tail(&bud->list, &jhead->buds_list); - p = rb_next(p); - } - spin_unlock(&c->buds_lock); -} - -/** * ubifs_add_bud_to_log - add a new bud to the log. * @c: UBIFS file-system description object * @jhead: journal head the bud belongs to diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c index 72775d35b99e..ef5155e109a2 100644 --- a/fs/ubifs/lpt.c +++ b/fs/ubifs/lpt.c @@ -1270,10 +1270,9 @@ static int read_pnode(struct ubifs_info *c, struct ubifs_nnode *parent, int iip) lnum = branch->lnum; offs = branch->offs; pnode = kzalloc(sizeof(struct ubifs_pnode), GFP_NOFS); - if (!pnode) { - err = -ENOMEM; - goto out; - } + if (!pnode) + return -ENOMEM; + if (lnum == 0) { /* * This pnode was not written which just means that the LEB diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c index 936f2cbfe6b6..3dbad6fbd1eb 100644 --- a/fs/ubifs/recovery.c +++ b/fs/ubifs/recovery.c @@ -317,6 +317,32 @@ int ubifs_recover_master_node(struct ubifs_info *c) goto out_free; } memcpy(c->rcvrd_mst_node, c->mst_node, UBIFS_MST_NODE_SZ); + + /* + * We had to recover the master node, which means there was an + * unclean reboot. However, it is possible that the master node + * is clean at this point, i.e., %UBIFS_MST_DIRTY is not set. + * E.g., consider the following chain of events: + * + * 1. UBIFS was cleanly unmounted, so the master node is clean + * 2. UBIFS is being mounted R/W and starts changing the master + * node in the first (%UBIFS_MST_LNUM). A power cut happens, + * so this LEB ends up with some amount of garbage at the + * end. + * 3. UBIFS is being mounted R/O. We reach this place and + * recover the master node from the second LEB + * (%UBIFS_MST_LNUM + 1). But we cannot update the media + * because we are being mounted R/O. We have to defer the + * operation. + * 4. However, this master node (@c->mst_node) is marked as + * clean (since the step 1). And if we just return, the + * mount code will be confused and won't recover the master + * node when it is re-mounter R/W later. + * + * Thus, to force the recovery by marking the master node as + * dirty. + */ + c->mst_node->flags |= cpu_to_le32(UBIFS_MST_DIRTY); } else { /* Write the recovered master node */ c->max_sqnum = le64_to_cpu(mst->ch.sqnum) - 1; diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c index eed0fcff8d73..d3d6d365bfc1 100644 --- a/fs/ubifs/replay.c +++ b/fs/ubifs/replay.c @@ -59,6 +59,7 @@ enum { * @new_size: truncation new size * @free: amount of free space in a bud * @dirty: amount of dirty space in a bud from padding and deletion nodes + * @jhead: journal head number of the bud * * UBIFS journal replay must compare node sequence numbers, which means it must * build a tree of node information to insert into the TNC. @@ -80,6 +81,7 @@ struct replay_entry { struct { int free; int dirty; + int jhead; }; }; }; @@ -159,6 +161,11 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r) err = PTR_ERR(lp); goto out; } + + /* Make sure the journal head points to the latest bud */ + err = ubifs_wbuf_seek_nolock(&c->jheads[r->jhead].wbuf, r->lnum, + c->leb_size - r->free, UBI_SHORTTERM); + out: ubifs_release_lprops(c); return err; @@ -627,10 +634,6 @@ static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead, ubifs_assert(sleb->endpt - offs >= used); ubifs_assert(sleb->endpt % c->min_io_size == 0); - if (sleb->endpt + c->min_io_size <= c->leb_size && !c->ro_mount) - err = ubifs_wbuf_seek_nolock(&c->jheads[jhead].wbuf, lnum, - sleb->endpt, UBI_SHORTTERM); - *dirty = sleb->endpt - offs - used; *free = c->leb_size - sleb->endpt; @@ -653,12 +656,14 @@ out_dump: * @sqnum: sequence number * @free: amount of free space in bud * @dirty: amount of dirty space from padding and deletion nodes + * @jhead: journal head number for the bud * * This function inserts a reference node to the replay tree and returns zero * in case of success or a negative error code in case of failure. */ static int insert_ref_node(struct ubifs_info *c, int lnum, int offs, - unsigned long long sqnum, int free, int dirty) + unsigned long long sqnum, int free, int dirty, + int jhead) { struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL; struct replay_entry *r; @@ -688,6 +693,7 @@ static int insert_ref_node(struct ubifs_info *c, int lnum, int offs, r->flags = REPLAY_REF; r->free = free; r->dirty = dirty; + r->jhead = jhead; rb_link_node(&r->rb, parent, p); rb_insert_color(&r->rb, &c->replay_tree); @@ -712,7 +718,7 @@ static int replay_buds(struct ubifs_info *c) if (err) return err; err = insert_ref_node(c, b->bud->lnum, b->bud->start, b->sqnum, - free, dirty); + free, dirty, b->bud->jhead); if (err) return err; } diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 6ddd9973e681..04ad07f4fcc3 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1257,12 +1257,12 @@ static int mount_ubifs(struct ubifs_info *c) goto out_free; } + err = alloc_wbufs(c); + if (err) + goto out_cbuf; + sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id); if (!c->ro_mount) { - err = alloc_wbufs(c); - if (err) - goto out_cbuf; - /* Create background thread */ c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name); if (IS_ERR(c->bgt)) { @@ -1568,6 +1568,7 @@ static int ubifs_remount_rw(struct ubifs_info *c) mutex_lock(&c->umount_mutex); dbg_save_space_info(c); c->remounting_rw = 1; + c->ro_mount = 0; err = check_free_space(c); if (err) @@ -1630,12 +1631,6 @@ static int ubifs_remount_rw(struct ubifs_info *c) if (err) goto out; - err = alloc_wbufs(c); - if (err) - goto out; - - ubifs_create_buds_lists(c); - /* Create background thread */ c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name); if (IS_ERR(c->bgt)) { @@ -1670,19 +1665,30 @@ static int ubifs_remount_rw(struct ubifs_info *c) if (err) goto out; + dbg_gen("re-mounted read-write"); + c->remounting_rw = 0; + if (c->need_recovery) { c->need_recovery = 0; ubifs_msg("deferred recovery completed"); + } else { + /* + * Do not run the debugging space check if the were doing + * recovery, because when we saved the information we had the + * file-system in a state where the TNC and lprops has been + * modified in memory, but all the I/O operations (including a + * commit) were deferred. So the file-system was in + * "non-committed" state. Now the file-system is in committed + * state, and of course the amount of free space will change + * because, for example, the old index size was imprecise. + */ + err = dbg_check_space_info(c); } - - dbg_gen("re-mounted read-write"); - c->ro_mount = 0; - c->remounting_rw = 0; - err = dbg_check_space_info(c); mutex_unlock(&c->umount_mutex); return err; out: + c->ro_mount = 1; vfree(c->orph_buf); c->orph_buf = NULL; if (c->bgt) { @@ -1732,7 +1738,6 @@ static void ubifs_remount_ro(struct ubifs_info *c) if (err) ubifs_ro_mode(c, err); - free_wbufs(c); vfree(c->orph_buf); c->orph_buf = NULL; kfree(c->write_reserve_buf); @@ -1760,10 +1765,12 @@ static void ubifs_put_super(struct super_block *sb) * of the media. For example, there will be dirty inodes if we failed * to write them back because of I/O errors. */ - ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0); - ubifs_assert(c->budg_idx_growth == 0); - ubifs_assert(c->budg_dd_growth == 0); - ubifs_assert(c->budg_data_growth == 0); + if (!c->ro_error) { + ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0); + ubifs_assert(c->budg_idx_growth == 0); + ubifs_assert(c->budg_dd_growth == 0); + ubifs_assert(c->budg_data_growth == 0); + } /* * The 'c->umount_lock' prevents races between UBIFS memory shrinker diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 27a4babe7df0..e765743cf9f3 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -78,7 +78,7 @@ static int ufs_block_to_path(struct inode *inode, sector_t i_block, sector_t off /* * Returns the location of the fragment from - * the begining of the filesystem. + * the beginning of the filesystem. */ static u64 ufs_frag_map(struct inode *inode, sector_t frag, bool needs_lock) diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 7693d6293404..3915ade6f9a8 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -483,9 +483,9 @@ static int ufs_parse_options (char * options, unsigned * mount_options) } /* - * Diffrent types of UFS hold fs_cstotal in different - * places, and use diffrent data structure for it. - * To make things simplier we just copy fs_cstotal to ufs_sb_private_info + * Different types of UFS hold fs_cstotal in different + * places, and use different data structure for it. + * To make things simpler we just copy fs_cstotal to ufs_sb_private_info */ static void ufs_setup_cstotal(struct super_block *sb) { diff --git a/fs/xattr.c b/fs/xattr.c index a19acdb81cd1..f1ef94974dea 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -666,7 +666,7 @@ generic_setxattr(struct dentry *dentry, const char *name, const void *value, siz handler = xattr_resolve_name(dentry->d_sb->s_xattr, &name); if (!handler) return -EOPNOTSUPP; - return handler->set(dentry, name, value, size, 0, handler->flags); + return handler->set(dentry, name, value, size, flags, handler->flags); } /* diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 52dbd14260ba..79ce38be15a1 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1295,7 +1295,7 @@ xfs_get_blocks_direct( * If the private argument is non-NULL __xfs_get_blocks signals us that we * need to issue a transaction to convert the range from unwritten to written * extents. In case this is regular synchronous I/O we just call xfs_end_io - * to do this and we are done. But in case this was a successfull AIO + * to do this and we are done. But in case this was a successful AIO * request this handler is called from interrupt context, from which we * can't start transactions. In that case offload the I/O completion to * the workqueues we also use for buffered I/O completion. diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 596bb2c9de42..9ef9ed2cfe2e 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -120,7 +120,7 @@ xfs_buf_lru_add( * The unlocked check is safe here because it only occurs when there are not * b_lru_ref counts left on the inode under the pag->pag_buf_lock. it is there * to optimise the shrinker removing the buffer from the LRU and calling - * xfs_buf_free(). i.e. it removes an unneccessary round trip on the + * xfs_buf_free(). i.e. it removes an unnecessary round trip on the * bt_lru_lock. */ STATIC void @@ -293,7 +293,6 @@ xfs_buf_allocate_memory( size_t nbytes, offset; gfp_t gfp_mask = xb_to_gfp(flags); unsigned short page_count, i; - pgoff_t first; xfs_off_t end; int error; @@ -333,7 +332,6 @@ use_alloc_page: return error; offset = bp->b_offset; - first = bp->b_file_offset >> PAGE_SHIFT; bp->b_flags |= _XBF_PAGES; for (i = 0; i < bp->b_page_count; i++) { @@ -380,7 +378,7 @@ out_free_pages: } /* - * Map buffer into kernel address-space if nessecary. + * Map buffer into kernel address-space if necessary. */ STATIC int _xfs_buf_map_pages( @@ -657,8 +655,6 @@ xfs_buf_readahead( xfs_off_t ioff, size_t isize) { - struct backing_dev_info *bdi; - if (bdi_read_congested(target->bt_bdi)) return; @@ -919,8 +915,6 @@ xfs_buf_lock( if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) xfs_log_force(bp->b_target->bt_mount, 0); - if (atomic_read(&bp->b_io_remaining)) - blk_flush_plug(current); down(&bp->b_sema); XB_SET_OWNER(bp); @@ -1309,8 +1303,6 @@ xfs_buf_iowait( { trace_xfs_buf_iowait(bp, _RET_IP_); - if (atomic_read(&bp->b_io_remaining)) - blk_flush_plug(current); wait_for_completion(&bp->b_iowait); trace_xfs_buf_iowait_done(bp, _RET_IP_); @@ -1747,8 +1739,8 @@ xfsbufd( do { long age = xfs_buf_age_centisecs * msecs_to_jiffies(10); long tout = xfs_buf_timer_centisecs * msecs_to_jiffies(10); - int count = 0; struct list_head tmp; + struct blk_plug plug; if (unlikely(freezing(current))) { set_bit(XBT_FORCE_SLEEP, &target->bt_flags); @@ -1764,16 +1756,15 @@ xfsbufd( xfs_buf_delwri_split(target, &tmp, age); list_sort(NULL, &tmp, xfs_buf_cmp); + + blk_start_plug(&plug); while (!list_empty(&tmp)) { struct xfs_buf *bp; bp = list_first_entry(&tmp, struct xfs_buf, b_list); list_del_init(&bp->b_list); xfs_bdstrat_cb(bp); - count++; } - if (count) - blk_flush_plug(current); - + blk_finish_plug(&plug); } while (!kthread_should_stop()); return 0; @@ -1793,6 +1784,7 @@ xfs_flush_buftarg( int pincount = 0; LIST_HEAD(tmp_list); LIST_HEAD(wait_list); + struct blk_plug plug; xfs_buf_runall_queues(xfsconvertd_workqueue); xfs_buf_runall_queues(xfsdatad_workqueue); @@ -1807,6 +1799,8 @@ xfs_flush_buftarg( * we do that after issuing all the IO. */ list_sort(NULL, &tmp_list, xfs_buf_cmp); + + blk_start_plug(&plug); while (!list_empty(&tmp_list)) { bp = list_first_entry(&tmp_list, struct xfs_buf, b_list); ASSERT(target == bp->b_target); @@ -1817,10 +1811,10 @@ xfs_flush_buftarg( } xfs_bdstrat_cb(bp); } + blk_finish_plug(&plug); if (wait) { - /* Expedite and wait for IO to complete. */ - blk_flush_plug(current); + /* Wait for IO to complete. */ while (!list_empty(&wait_list)) { bp = list_first_entry(&wait_list, struct xfs_buf, b_list); diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 52aadfbed132..f4213ba1ff85 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -381,7 +381,7 @@ xfs_aio_write_isize_update( /* * If this was a direct or synchronous I/O that failed (such as ENOSPC) then - * part of the I/O may have been written to disk before the error occured. In + * part of the I/O may have been written to disk before the error occurred. In * this case the on-disk file size may have been adjusted beyond the in-memory * file size and now needs to be truncated back. */ diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 9ff7fc603d2f..dd21784525a8 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -70,7 +70,7 @@ xfs_synchronize_times( /* * If the linux inode is valid, mark it dirty. - * Used when commiting a dirty inode into a transaction so that + * Used when committing a dirty inode into a transaction so that * the inode will get written back by the linux code */ void diff --git a/fs/xfs/linux-2.6/xfs_message.c b/fs/xfs/linux-2.6/xfs_message.c index 508e06fd7d1e..9f76cceb678d 100644 --- a/fs/xfs/linux-2.6/xfs_message.c +++ b/fs/xfs/linux-2.6/xfs_message.c @@ -28,53 +28,49 @@ /* * XFS logging functions */ -static int +static void __xfs_printk( const char *level, const struct xfs_mount *mp, struct va_format *vaf) { - if (mp && mp->m_fsname) - return printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf); - return printk("%sXFS: %pV\n", level, vaf); + if (mp && mp->m_fsname) { + printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf); + return; + } + printk("%sXFS: %pV\n", level, vaf); } -int xfs_printk( +void xfs_printk( const char *level, const struct xfs_mount *mp, const char *fmt, ...) { struct va_format vaf; va_list args; - int r; va_start(args, fmt); vaf.fmt = fmt; vaf.va = &args; - r = __xfs_printk(level, mp, &vaf); + __xfs_printk(level, mp, &vaf); va_end(args); - - return r; } #define define_xfs_printk_level(func, kern_level) \ -int func(const struct xfs_mount *mp, const char *fmt, ...) \ +void func(const struct xfs_mount *mp, const char *fmt, ...) \ { \ struct va_format vaf; \ va_list args; \ - int r; \ \ va_start(args, fmt); \ \ vaf.fmt = fmt; \ vaf.va = &args; \ \ - r = __xfs_printk(kern_level, mp, &vaf); \ + __xfs_printk(kern_level, mp, &vaf); \ va_end(args); \ - \ - return r; \ } \ define_xfs_printk_level(xfs_emerg, KERN_EMERG); @@ -88,7 +84,7 @@ define_xfs_printk_level(xfs_info, KERN_INFO); define_xfs_printk_level(xfs_debug, KERN_DEBUG); #endif -int +void xfs_alert_tag( const struct xfs_mount *mp, int panic_tag, @@ -97,7 +93,6 @@ xfs_alert_tag( struct va_format vaf; va_list args; int do_panic = 0; - int r; if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) { xfs_printk(KERN_ALERT, mp, @@ -110,12 +105,10 @@ xfs_alert_tag( vaf.fmt = fmt; vaf.va = &args; - r = __xfs_printk(KERN_ALERT, mp, &vaf); + __xfs_printk(KERN_ALERT, mp, &vaf); va_end(args); BUG_ON(do_panic); - - return r; } void diff --git a/fs/xfs/linux-2.6/xfs_message.h b/fs/xfs/linux-2.6/xfs_message.h index e77ffa16745b..f1b3fc1b6c4e 100644 --- a/fs/xfs/linux-2.6/xfs_message.h +++ b/fs/xfs/linux-2.6/xfs_message.h @@ -3,32 +3,34 @@ struct xfs_mount; -extern int xfs_printk(const char *level, const struct xfs_mount *mp, +extern void xfs_printk(const char *level, const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 3, 4))); -extern int xfs_emerg(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_emerg(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_alert(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_alert(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_alert_tag(const struct xfs_mount *mp, int tag, +extern void xfs_alert_tag(const struct xfs_mount *mp, int tag, const char *fmt, ...) __attribute__ ((format (printf, 3, 4))); -extern int xfs_crit(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_crit(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_err(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_err(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_warn(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_warn(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_notice(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_notice(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); -extern int xfs_info(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_info(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); #ifdef DEBUG -extern int xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) +extern void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) __attribute__ ((format (printf, 2, 3))); #else -#define xfs_debug(mp, fmt, ...) (0) +static inline void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) +{ +} #endif extern void assfail(char *expr, char *f, int l); diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 1ba5c451da36..b38e58d02299 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -816,75 +816,6 @@ xfs_setup_devices( return 0; } -/* - * XFS AIL push thread support - */ -void -xfsaild_wakeup( - struct xfs_ail *ailp, - xfs_lsn_t threshold_lsn) -{ - /* only ever move the target forwards */ - if (XFS_LSN_CMP(threshold_lsn, ailp->xa_target) > 0) { - ailp->xa_target = threshold_lsn; - wake_up_process(ailp->xa_task); - } -} - -STATIC int -xfsaild( - void *data) -{ - struct xfs_ail *ailp = data; - xfs_lsn_t last_pushed_lsn = 0; - long tout = 0; /* milliseconds */ - - while (!kthread_should_stop()) { - /* - * for short sleeps indicating congestion, don't allow us to - * get woken early. Otherwise all we do is bang on the AIL lock - * without making progress. - */ - if (tout && tout <= 20) - __set_current_state(TASK_KILLABLE); - else - __set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(tout ? - msecs_to_jiffies(tout) : MAX_SCHEDULE_TIMEOUT); - - /* swsusp */ - try_to_freeze(); - - ASSERT(ailp->xa_mount->m_log); - if (XFS_FORCED_SHUTDOWN(ailp->xa_mount)) - continue; - - tout = xfsaild_push(ailp, &last_pushed_lsn); - } - - return 0; -} /* xfsaild */ - -int -xfsaild_start( - struct xfs_ail *ailp) -{ - ailp->xa_target = 0; - ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild/%s", - ailp->xa_mount->m_fsname); - if (IS_ERR(ailp->xa_task)) - return -PTR_ERR(ailp->xa_task); - return 0; -} - -void -xfsaild_stop( - struct xfs_ail *ailp) -{ - kthread_stop(ailp->xa_task); -} - - /* Catch misguided souls that try to use this interface on XFS */ STATIC struct inode * xfs_fs_alloc_inode( @@ -1191,22 +1122,12 @@ xfs_fs_sync_fs( return -error; if (laptop_mode) { - int prev_sync_seq = mp->m_sync_seq; - /* * The disk must be active because we're syncing. * We schedule xfssyncd now (now that the disk is * active) instead of later (when it might not be). */ - wake_up_process(mp->m_sync_task); - /* - * We have to wait for the sync iteration to complete. - * If we don't, the disk activity caused by the sync - * will come after the sync is completed, and that - * triggers another sync from laptop mode. - */ - wait_event(mp->m_wait_single_sync_task, - mp->m_sync_seq != prev_sync_seq); + flush_delayed_work_sync(&mp->m_sync_work); } return 0; @@ -1490,9 +1411,6 @@ xfs_fs_fill_super( spin_lock_init(&mp->m_sb_lock); mutex_init(&mp->m_growlock); atomic_set(&mp->m_active_trans, 0); - INIT_LIST_HEAD(&mp->m_sync_list); - spin_lock_init(&mp->m_sync_lock); - init_waitqueue_head(&mp->m_wait_single_sync_task); mp->m_super = sb; sb->s_fs_info = mp; @@ -1799,6 +1717,38 @@ xfs_destroy_zones(void) } STATIC int __init +xfs_init_workqueues(void) +{ + /* + * max_active is set to 8 to give enough concurency to allow + * multiple work operations on each CPU to run. This allows multiple + * filesystems to be running sync work concurrently, and scales with + * the number of CPUs in the system. + */ + xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8); + if (!xfs_syncd_wq) + goto out; + + xfs_ail_wq = alloc_workqueue("xfsail", WQ_CPU_INTENSIVE, 8); + if (!xfs_ail_wq) + goto out_destroy_syncd; + + return 0; + +out_destroy_syncd: + destroy_workqueue(xfs_syncd_wq); +out: + return -ENOMEM; +} + +STATIC void +xfs_destroy_workqueues(void) +{ + destroy_workqueue(xfs_ail_wq); + destroy_workqueue(xfs_syncd_wq); +} + +STATIC int __init init_xfs_fs(void) { int error; @@ -1813,10 +1763,14 @@ init_xfs_fs(void) if (error) goto out; - error = xfs_mru_cache_init(); + error = xfs_init_workqueues(); if (error) goto out_destroy_zones; + error = xfs_mru_cache_init(); + if (error) + goto out_destroy_wq; + error = xfs_filestream_init(); if (error) goto out_mru_cache_uninit; @@ -1833,6 +1787,10 @@ init_xfs_fs(void) if (error) goto out_cleanup_procfs; + error = xfs_init_workqueues(); + if (error) + goto out_sysctl_unregister; + vfs_initquota(); error = register_filesystem(&xfs_fs_type); @@ -1850,6 +1808,8 @@ init_xfs_fs(void) xfs_filestream_uninit(); out_mru_cache_uninit: xfs_mru_cache_uninit(); + out_destroy_wq: + xfs_destroy_workqueues(); out_destroy_zones: xfs_destroy_zones(); out: @@ -1866,6 +1826,7 @@ exit_xfs_fs(void) xfs_buf_terminate(); xfs_filestream_uninit(); xfs_mru_cache_uninit(); + xfs_destroy_workqueues(); xfs_destroy_zones(); } diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 594cd822d84d..3e898a48122d 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -22,6 +22,7 @@ #include "xfs_log.h" #include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_trans_priv.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_mount.h" @@ -39,6 +40,8 @@ #include <linux/kthread.h> #include <linux/freezer.h> +struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ + /* * The inode lookup is done in batches to keep the amount of lock traffic and * radix tree lookups to a minimum. The batch size is a trade off between @@ -401,7 +404,7 @@ xfs_quiesce_fs( /* * Second stage of a quiesce. The data is already synced, now we have to take * care of the metadata. New transactions are already blocked, so we need to - * wait for any remaining transactions to drain out before proceding. + * wait for any remaining transactions to drain out before proceeding. */ void xfs_quiesce_attr( @@ -431,62 +434,12 @@ xfs_quiesce_attr( xfs_unmountfs_writesb(mp); } -/* - * Enqueue a work item to be picked up by the vfs xfssyncd thread. - * Doing this has two advantages: - * - It saves on stack space, which is tight in certain situations - * - It can be used (with care) as a mechanism to avoid deadlocks. - * Flushing while allocating in a full filesystem requires both. - */ -STATIC void -xfs_syncd_queue_work( - struct xfs_mount *mp, - void *data, - void (*syncer)(struct xfs_mount *, void *), - struct completion *completion) +static void +xfs_syncd_queue_sync( + struct xfs_mount *mp) { - struct xfs_sync_work *work; - - work = kmem_alloc(sizeof(struct xfs_sync_work), KM_SLEEP); - INIT_LIST_HEAD(&work->w_list); - work->w_syncer = syncer; - work->w_data = data; - work->w_mount = mp; - work->w_completion = completion; - spin_lock(&mp->m_sync_lock); - list_add_tail(&work->w_list, &mp->m_sync_list); - spin_unlock(&mp->m_sync_lock); - wake_up_process(mp->m_sync_task); -} - -/* - * Flush delayed allocate data, attempting to free up reserved space - * from existing allocations. At this point a new allocation attempt - * has failed with ENOSPC and we are in the process of scratching our - * heads, looking about for more room... - */ -STATIC void -xfs_flush_inodes_work( - struct xfs_mount *mp, - void *arg) -{ - struct inode *inode = arg; - xfs_sync_data(mp, SYNC_TRYLOCK); - xfs_sync_data(mp, SYNC_TRYLOCK | SYNC_WAIT); - iput(inode); -} - -void -xfs_flush_inodes( - xfs_inode_t *ip) -{ - struct inode *inode = VFS_I(ip); - DECLARE_COMPLETION_ONSTACK(completion); - - igrab(inode); - xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inodes_work, &completion); - wait_for_completion(&completion); - xfs_log_force(ip->i_mount, XFS_LOG_SYNC); + queue_delayed_work(xfs_syncd_wq, &mp->m_sync_work, + msecs_to_jiffies(xfs_syncd_centisecs * 10)); } /* @@ -496,9 +449,10 @@ xfs_flush_inodes( */ STATIC void xfs_sync_worker( - struct xfs_mount *mp, - void *unused) + struct work_struct *work) { + struct xfs_mount *mp = container_of(to_delayed_work(work), + struct xfs_mount, m_sync_work); int error; if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { @@ -508,73 +462,106 @@ xfs_sync_worker( error = xfs_fs_log_dummy(mp); else xfs_log_force(mp, 0); - xfs_reclaim_inodes(mp, 0); error = xfs_qm_sync(mp, SYNC_TRYLOCK); + + /* start pushing all the metadata that is currently dirty */ + xfs_ail_push_all(mp->m_ail); } - mp->m_sync_seq++; - wake_up(&mp->m_wait_single_sync_task); + + /* queue us up again */ + xfs_syncd_queue_sync(mp); } -STATIC int -xfssyncd( - void *arg) +/* + * Queue a new inode reclaim pass if there are reclaimable inodes and there + * isn't a reclaim pass already in progress. By default it runs every 5s based + * on the xfs syncd work default of 30s. Perhaps this should have it's own + * tunable, but that can be done if this method proves to be ineffective or too + * aggressive. + */ +static void +xfs_syncd_queue_reclaim( + struct xfs_mount *mp) { - struct xfs_mount *mp = arg; - long timeleft; - xfs_sync_work_t *work, *n; - LIST_HEAD (tmp); - - set_freezable(); - timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10); - for (;;) { - if (list_empty(&mp->m_sync_list)) - timeleft = schedule_timeout_interruptible(timeleft); - /* swsusp */ - try_to_freeze(); - if (kthread_should_stop() && list_empty(&mp->m_sync_list)) - break; - spin_lock(&mp->m_sync_lock); - /* - * We can get woken by laptop mode, to do a sync - - * that's the (only!) case where the list would be - * empty with time remaining. - */ - if (!timeleft || list_empty(&mp->m_sync_list)) { - if (!timeleft) - timeleft = xfs_syncd_centisecs * - msecs_to_jiffies(10); - INIT_LIST_HEAD(&mp->m_sync_work.w_list); - list_add_tail(&mp->m_sync_work.w_list, - &mp->m_sync_list); - } - list_splice_init(&mp->m_sync_list, &tmp); - spin_unlock(&mp->m_sync_lock); + /* + * We can have inodes enter reclaim after we've shut down the syncd + * workqueue during unmount, so don't allow reclaim work to be queued + * during unmount. + */ + if (!(mp->m_super->s_flags & MS_ACTIVE)) + return; - list_for_each_entry_safe(work, n, &tmp, w_list) { - (*work->w_syncer)(mp, work->w_data); - list_del(&work->w_list); - if (work == &mp->m_sync_work) - continue; - if (work->w_completion) - complete(work->w_completion); - kmem_free(work); - } + rcu_read_lock(); + if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) { + queue_delayed_work(xfs_syncd_wq, &mp->m_reclaim_work, + msecs_to_jiffies(xfs_syncd_centisecs / 6 * 10)); } + rcu_read_unlock(); +} - return 0; +/* + * This is a fast pass over the inode cache to try to get reclaim moving on as + * many inodes as possible in a short period of time. It kicks itself every few + * seconds, as well as being kicked by the inode cache shrinker when memory + * goes low. It scans as quickly as possible avoiding locked inodes or those + * already being flushed, and once done schedules a future pass. + */ +STATIC void +xfs_reclaim_worker( + struct work_struct *work) +{ + struct xfs_mount *mp = container_of(to_delayed_work(work), + struct xfs_mount, m_reclaim_work); + + xfs_reclaim_inodes(mp, SYNC_TRYLOCK); + xfs_syncd_queue_reclaim(mp); +} + +/* + * Flush delayed allocate data, attempting to free up reserved space + * from existing allocations. At this point a new allocation attempt + * has failed with ENOSPC and we are in the process of scratching our + * heads, looking about for more room. + * + * Queue a new data flush if there isn't one already in progress and + * wait for completion of the flush. This means that we only ever have one + * inode flush in progress no matter how many ENOSPC events are occurring and + * so will prevent the system from bogging down due to every concurrent + * ENOSPC event scanning all the active inodes in the system for writeback. + */ +void +xfs_flush_inodes( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + + queue_work(xfs_syncd_wq, &mp->m_flush_work); + flush_work_sync(&mp->m_flush_work); +} + +STATIC void +xfs_flush_worker( + struct work_struct *work) +{ + struct xfs_mount *mp = container_of(work, + struct xfs_mount, m_flush_work); + + xfs_sync_data(mp, SYNC_TRYLOCK); + xfs_sync_data(mp, SYNC_TRYLOCK | SYNC_WAIT); } int xfs_syncd_init( struct xfs_mount *mp) { - mp->m_sync_work.w_syncer = xfs_sync_worker; - mp->m_sync_work.w_mount = mp; - mp->m_sync_work.w_completion = NULL; - mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd/%s", mp->m_fsname); - if (IS_ERR(mp->m_sync_task)) - return -PTR_ERR(mp->m_sync_task); + INIT_WORK(&mp->m_flush_work, xfs_flush_worker); + INIT_DELAYED_WORK(&mp->m_sync_work, xfs_sync_worker); + INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker); + + xfs_syncd_queue_sync(mp); + xfs_syncd_queue_reclaim(mp); + return 0; } @@ -582,7 +569,9 @@ void xfs_syncd_stop( struct xfs_mount *mp) { - kthread_stop(mp->m_sync_task); + cancel_delayed_work_sync(&mp->m_sync_work); + cancel_delayed_work_sync(&mp->m_reclaim_work); + cancel_work_sync(&mp->m_flush_work); } void @@ -601,6 +590,10 @@ __xfs_inode_set_reclaim_tag( XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), XFS_ICI_RECLAIM_TAG); spin_unlock(&ip->i_mount->m_perag_lock); + + /* schedule periodic background inode reclaim */ + xfs_syncd_queue_reclaim(ip->i_mount); + trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno, -1, _RET_IP_); } @@ -933,6 +926,7 @@ restart: XFS_LOOKUP_BATCH, XFS_ICI_RECLAIM_TAG); if (!nr_found) { + done = 1; rcu_read_unlock(); break; } @@ -1017,7 +1011,13 @@ xfs_reclaim_inodes( } /* - * Shrinker infrastructure. + * Inode cache shrinker. + * + * When called we make sure that there is a background (fast) inode reclaim in + * progress, while we will throttle the speed of reclaim via doiing synchronous + * reclaim of inodes. That means if we come across dirty inodes, we wait for + * them to be cleaned, which we hope will not be very long due to the + * background walker having already kicked the IO off on those dirty inodes. */ static int xfs_reclaim_inode_shrink( @@ -1032,10 +1032,15 @@ xfs_reclaim_inode_shrink( mp = container_of(shrink, struct xfs_mount, m_inode_shrink); if (nr_to_scan) { + /* kick background reclaimer and push the AIL */ + xfs_syncd_queue_reclaim(mp); + xfs_ail_push_all(mp->m_ail); + if (!(gfp_mask & __GFP_FS)) return -1; - xfs_reclaim_inodes_ag(mp, SYNC_TRYLOCK, &nr_to_scan); + xfs_reclaim_inodes_ag(mp, SYNC_TRYLOCK | SYNC_WAIT, + &nr_to_scan); /* terminate if we don't exhaust the scan */ if (nr_to_scan > 0) return -1; diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h index 32ba6628290c..e3a6ad27415f 100644 --- a/fs/xfs/linux-2.6/xfs_sync.h +++ b/fs/xfs/linux-2.6/xfs_sync.h @@ -32,6 +32,8 @@ typedef struct xfs_sync_work { #define SYNC_WAIT 0x0001 /* wait for i/o to complete */ #define SYNC_TRYLOCK 0x0002 /* only try to lock inodes */ +extern struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ + int xfs_syncd_init(struct xfs_mount *mp); void xfs_syncd_stop(struct xfs_mount *mp); diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index 7e2416478503..6fa214603819 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -600,7 +600,7 @@ xfs_qm_dqread( /* * Reservation counters are defined as reservation plus current usage - * to avoid having to add everytime. + * to avoid having to add every time. */ dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount); dqp->q_res_icount = be64_to_cpu(ddqp->d_icount); diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 254ee062bd7d..69228aa8605a 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -461,12 +461,10 @@ xfs_qm_dqflush_all( struct xfs_quotainfo *q = mp->m_quotainfo; int recl; struct xfs_dquot *dqp; - int niters; int error; if (!q) return 0; - niters = 0; again: mutex_lock(&q->qi_dqlist_lock); list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { @@ -1314,14 +1312,9 @@ xfs_qm_dqiter_bufs( { xfs_buf_t *bp; int error; - int notcommitted; - int incr; int type; ASSERT(blkcnt > 0); - notcommitted = 0; - incr = (blkcnt > XFS_QM_MAX_DQCLUSTER_LOGSZ) ? - XFS_QM_MAX_DQCLUSTER_LOGSZ : blkcnt; type = flags & XFS_QMOPT_UQUOTA ? XFS_DQ_USER : (flags & XFS_QMOPT_PQUOTA ? XFS_DQ_PROJ : XFS_DQ_GROUP); error = 0; diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h index c9446f1c726d..567b29b9f1b3 100644 --- a/fs/xfs/quota/xfs_qm.h +++ b/fs/xfs/quota/xfs_qm.h @@ -65,11 +65,6 @@ extern kmem_zone_t *qm_dqtrxzone; * block in the dquot/xqm code. */ #define XFS_DQUOT_CLUSTER_SIZE_FSB (xfs_filblks_t)1 -/* - * When doing a quotacheck, we log dquot clusters of this many FSBs at most - * in a single transaction. We don't want to ask for too huge a log reservation. - */ -#define XFS_QM_MAX_DQCLUSTER_LOGSZ 3 typedef xfs_dqhash_t xfs_dqlist_t; diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index 774d7ec6df8e..a0a829addca9 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -134,7 +134,7 @@ xfs_qm_newmount( */ if (quotaondisk && !XFS_QM_NEED_QUOTACHECK(mp)) { /* - * If an error occured, qm_mount_quotas code + * If an error occurred, qm_mount_quotas code * has already disabled quotas. So, just finish * mounting, and get on with the boring life * without disk quotas. diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index c82f06778a27..2dadb15d5ca9 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -172,7 +172,7 @@ xfs_qm_scall_quotaoff( /* * Next we make the changes in the quota flag in the mount struct. * This isn't protected by a particular lock directly, because we - * don't want to take a mrlock everytime we depend on quotas being on. + * don't want to take a mrlock every time we depend on quotas being on. */ mp->m_qflags &= ~(flags); @@ -313,14 +313,12 @@ xfs_qm_scall_quotaon( { int error; uint qf; - uint accflags; __int64_t sbflags; flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD); /* * Switching on quota accounting must be done at mount time. */ - accflags = flags & XFS_ALL_QUOTA_ACCT; flags &= ~(XFS_ALL_QUOTA_ACCT); sbflags = 0; @@ -354,7 +352,7 @@ xfs_qm_scall_quotaon( return XFS_ERROR(EINVAL); } /* - * If everything's upto-date incore, then don't waste time. + * If everything's up to-date incore, then don't waste time. */ if ((mp->m_qflags & flags) == flags) return XFS_ERROR(EEXIST); diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index 4bc3c649aee4..27d64d752eab 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -2395,17 +2395,33 @@ xfs_free_extent( memset(&args, 0, sizeof(xfs_alloc_arg_t)); args.tp = tp; args.mp = tp->t_mountp; + + /* + * validate that the block number is legal - the enables us to detect + * and handle a silent filesystem corruption rather than crashing. + */ args.agno = XFS_FSB_TO_AGNO(args.mp, bno); - ASSERT(args.agno < args.mp->m_sb.sb_agcount); + if (args.agno >= args.mp->m_sb.sb_agcount) + return EFSCORRUPTED; + args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno); + if (args.agbno >= args.mp->m_sb.sb_agblocks) + return EFSCORRUPTED; + args.pag = xfs_perag_get(args.mp, args.agno); - if ((error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING))) + ASSERT(args.pag); + + error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING); + if (error) goto error0; -#ifdef DEBUG - ASSERT(args.agbp != NULL); - ASSERT((args.agbno + len) <= - be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)); -#endif + + /* validate the extent size is legal now we have the agf locked */ + if (args.agbno + len > + be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)) { + error = EFSCORRUPTED; + goto error0; + } + error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0); error0: xfs_perag_put(args.pag); diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index e5413d96f1af..7b7e005e3dcc 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -992,7 +992,7 @@ xfs_buf_iodone_callbacks( lasttarg = XFS_BUF_TARGET(bp); /* - * If the write was asynchronous then noone will be looking for the + * If the write was asynchronous then no one will be looking for the * error. Clear the error state and write the buffer out again. * * During sync or umount we'll write all pending buffers again diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 742c8330994a..a37480a6e023 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -2789,7 +2789,7 @@ xfs_iflush( /* * We can't flush the inode until it is unpinned, so wait for it if we - * are allowed to block. We know noone new can pin it, because we are + * are allowed to block. We know no one new can pin it, because we are * holding the inode lock shared and you need to hold it exclusively to * pin the inode. * diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index f753200cef8d..ff4e2a30227d 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -111,7 +111,7 @@ struct xfs_imap { * Generally, we do not want to hold the i_rlock while holding the * i_ilock. Hierarchy is i_iolock followed by i_rlock. * - * xfs_iptr_t contains all the inode fields upto and including the + * xfs_iptr_t contains all the inode fields up to and including the * i_mnext and i_mprev fields, it is used as a marker in the inode * chain off the mount structure by xfs_sync calls. */ @@ -336,7 +336,7 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags) /* * Project quota id helpers (previously projid was 16bit only - * and using two 16bit values to hold new 32bit projid was choosen + * and using two 16bit values to hold new 32bit projid was chosen * to retain compatibility with "old" filesystems). */ static inline prid_t diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 46cc40131d4a..576fdfe81d60 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -198,6 +198,41 @@ xfs_inode_item_size( } /* + * xfs_inode_item_format_extents - convert in-core extents to on-disk form + * + * For either the data or attr fork in extent format, we need to endian convert + * the in-core extent as we place them into the on-disk inode. In this case, we + * need to do this conversion before we write the extents into the log. Because + * we don't have the disk inode to write into here, we allocate a buffer and + * format the extents into it via xfs_iextents_copy(). We free the buffer in + * the unlock routine after the copy for the log has been made. + * + * In the case of the data fork, the in-core and on-disk fork sizes can be + * different due to delayed allocation extents. We only log on-disk extents + * here, so always use the physical fork size to determine the size of the + * buffer we need to allocate. + */ +STATIC void +xfs_inode_item_format_extents( + struct xfs_inode *ip, + struct xfs_log_iovec *vecp, + int whichfork, + int type) +{ + xfs_bmbt_rec_t *ext_buffer; + + ext_buffer = kmem_alloc(XFS_IFORK_SIZE(ip, whichfork), KM_SLEEP); + if (whichfork == XFS_DATA_FORK) + ip->i_itemp->ili_extents_buf = ext_buffer; + else + ip->i_itemp->ili_aextents_buf = ext_buffer; + + vecp->i_addr = ext_buffer; + vecp->i_len = xfs_iextents_copy(ip, ext_buffer, whichfork); + vecp->i_type = type; +} + +/* * This is called to fill in the vector of log iovecs for the * given inode log item. It fills the first item with an inode * log format structure, the second with the on-disk inode structure, @@ -213,7 +248,6 @@ xfs_inode_item_format( struct xfs_inode *ip = iip->ili_inode; uint nvecs; size_t data_bytes; - xfs_bmbt_rec_t *ext_buffer; xfs_mount_t *mp; vecp->i_addr = &iip->ili_format; @@ -320,22 +354,8 @@ xfs_inode_item_format( } else #endif { - /* - * There are delayed allocation extents - * in the inode, or we need to convert - * the extents to on disk format. - * Use xfs_iextents_copy() - * to copy only the real extents into - * a separate buffer. We'll free the - * buffer in the unlock routine. - */ - ext_buffer = kmem_alloc(ip->i_df.if_bytes, - KM_SLEEP); - iip->ili_extents_buf = ext_buffer; - vecp->i_addr = ext_buffer; - vecp->i_len = xfs_iextents_copy(ip, ext_buffer, - XFS_DATA_FORK); - vecp->i_type = XLOG_REG_TYPE_IEXT; + xfs_inode_item_format_extents(ip, vecp, + XFS_DATA_FORK, XLOG_REG_TYPE_IEXT); } ASSERT(vecp->i_len <= ip->i_df.if_bytes); iip->ili_format.ilf_dsize = vecp->i_len; @@ -445,19 +465,12 @@ xfs_inode_item_format( */ vecp->i_addr = ip->i_afp->if_u1.if_extents; vecp->i_len = ip->i_afp->if_bytes; + vecp->i_type = XLOG_REG_TYPE_IATTR_EXT; #else ASSERT(iip->ili_aextents_buf == NULL); - /* - * Need to endian flip before logging - */ - ext_buffer = kmem_alloc(ip->i_afp->if_bytes, - KM_SLEEP); - iip->ili_aextents_buf = ext_buffer; - vecp->i_addr = ext_buffer; - vecp->i_len = xfs_iextents_copy(ip, ext_buffer, - XFS_ATTR_FORK); + xfs_inode_item_format_extents(ip, vecp, + XFS_ATTR_FORK, XLOG_REG_TYPE_IATTR_EXT); #endif - vecp->i_type = XLOG_REG_TYPE_IATTR_EXT; iip->ili_format.ilf_asize = vecp->i_len; vecp++; nvecs++; diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index dc1882adaf54..751e94fe1f77 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -204,7 +204,6 @@ xfs_bulkstat( xfs_agi_t *agi; /* agi header data */ xfs_agino_t agino; /* inode # in allocation group */ xfs_agnumber_t agno; /* allocation group number */ - xfs_daddr_t bno; /* inode cluster start daddr */ int chunkidx; /* current index into inode chunk */ int clustidx; /* current index into inode cluster */ xfs_btree_cur_t *cur; /* btree cursor for ialloc btree */ @@ -463,7 +462,6 @@ xfs_bulkstat( mp->m_sb.sb_inopblog); } ino = XFS_AGINO_TO_INO(mp, agno, agino); - bno = XFS_AGB_TO_DADDR(mp, agno, agbno); /* * Skip if this inode is free. */ diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 25efa9b8a602..b612ce4520ae 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -761,7 +761,7 @@ xfs_log_need_covered(xfs_mount_t *mp) break; case XLOG_STATE_COVER_NEED: case XLOG_STATE_COVER_NEED2: - if (!xfs_trans_ail_tail(log->l_ailp) && + if (!xfs_ail_min_lsn(log->l_ailp) && xlog_iclogs_empty(log)) { if (log->l_covered_state == XLOG_STATE_COVER_NEED) log->l_covered_state = XLOG_STATE_COVER_DONE; @@ -801,7 +801,7 @@ xlog_assign_tail_lsn( xfs_lsn_t tail_lsn; struct log *log = mp->m_log; - tail_lsn = xfs_trans_ail_tail(mp->m_ail); + tail_lsn = xfs_ail_min_lsn(mp->m_ail); if (!tail_lsn) tail_lsn = atomic64_read(&log->l_last_sync_lsn); @@ -1239,7 +1239,7 @@ xlog_grant_push_ail( * the filesystem is shutting down. */ if (!XLOG_FORCED_SHUTDOWN(log)) - xfs_trans_ail_push(log->l_ailp, threshold_lsn); + xfs_ail_push(log->l_ailp, threshold_lsn); } /* @@ -3407,6 +3407,17 @@ xlog_verify_dest_ptr( xfs_emerg(log->l_mp, "%s: invalid ptr", __func__); } +/* + * Check to make sure the grant write head didn't just over lap the tail. If + * the cycles are the same, we can't be overlapping. Otherwise, make sure that + * the cycles differ by exactly one and check the byte count. + * + * This check is run unlocked, so can give false positives. Rather than assert + * on failures, use a warn-once flag and a panic tag to allow the admin to + * determine if they want to panic the machine when such an error occurs. For + * debug kernels this will have the same effect as using an assert but, unlinke + * an assert, it can be turned off at runtime. + */ STATIC void xlog_verify_grant_tail( struct log *log) @@ -3414,17 +3425,22 @@ xlog_verify_grant_tail( int tail_cycle, tail_blocks; int cycle, space; - /* - * Check to make sure the grant write head didn't just over lap the - * tail. If the cycles are the same, we can't be overlapping. - * Otherwise, make sure that the cycles differ by exactly one and - * check the byte count. - */ xlog_crack_grant_head(&log->l_grant_write_head, &cycle, &space); xlog_crack_atomic_lsn(&log->l_tail_lsn, &tail_cycle, &tail_blocks); if (tail_cycle != cycle) { - ASSERT(cycle - 1 == tail_cycle); - ASSERT(space <= BBTOB(tail_blocks)); + if (cycle - 1 != tail_cycle && + !(log->l_flags & XLOG_TAIL_WARN)) { + xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES, + "%s: cycle - 1 != tail_cycle", __func__); + log->l_flags |= XLOG_TAIL_WARN; + } + + if (space > BBTOB(tail_blocks) && + !(log->l_flags & XLOG_TAIL_WARN)) { + xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES, + "%s: space > BBTOB(tail_blocks)", __func__); + log->l_flags |= XLOG_TAIL_WARN; + } } } diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 15dbf1f9c2be..5864850e9e34 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -144,6 +144,7 @@ static inline uint xlog_get_client_id(__be32 i) #define XLOG_RECOVERY_NEEDED 0x4 /* log was recovered */ #define XLOG_IO_ERROR 0x8 /* log hit an I/O error, and being shutdown */ +#define XLOG_TAIL_WARN 0x10 /* log tail verify warning issued */ #ifdef __KERNEL__ /* @@ -570,7 +571,7 @@ int xlog_write(struct log *log, struct xfs_log_vec *log_vector, * When we crack an atomic LSN, we sample it first so that the value will not * change while we are cracking it into the component values. This means we * will always get consistent component values to work from. This should always - * be used to smaple and crack LSNs taht are stored and updated in atomic + * be used to sample and crack LSNs that are stored and updated in atomic * variables. */ static inline void diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 0c4a5618e7af..5cc464a17c93 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -101,7 +101,7 @@ xlog_get_bp( /* * We do log I/O in units of log sectors (a power-of-2 * multiple of the basic block size), so we round up the - * requested size to acommodate the basic blocks required + * requested size to accommodate the basic blocks required * for complete log sectors. * * In addition, the buffer may be used for a non-sector- @@ -112,7 +112,7 @@ xlog_get_bp( * an issue. Nor will this be a problem if the log I/O is * done in basic blocks (sector size 1). But otherwise we * extend the buffer by one extra log sector to ensure - * there's space to accomodate this possiblility. + * there's space to accommodate this possibility. */ if (nbblks > 1 && log->l_sectBBsize > 1) nbblks += log->l_sectBBsize; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index a62e8971539d..19af0ab0d0c6 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -203,12 +203,9 @@ typedef struct xfs_mount { struct mutex m_icsb_mutex; /* balancer sync lock */ #endif struct xfs_mru_cache *m_filestream; /* per-mount filestream data */ - struct task_struct *m_sync_task; /* generalised sync thread */ - xfs_sync_work_t m_sync_work; /* work item for VFS_SYNC */ - struct list_head m_sync_list; /* sync thread work item list */ - spinlock_t m_sync_lock; /* work item list lock */ - int m_sync_seq; /* sync thread generation no. */ - wait_queue_head_t m_wait_single_sync_task; + struct delayed_work m_sync_work; /* background sync work */ + struct delayed_work m_reclaim_work; /* background inode reclaim */ + struct work_struct m_flush_work; /* background inode flush */ __int64_t m_update_flags; /* sb flags we need to update on the next remount,rw */ struct shrinker m_inode_shrink; /* inode reclaim shrinker */ diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 12aff9584e29..5fc2380092c8 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -28,74 +28,138 @@ #include "xfs_trans_priv.h" #include "xfs_error.h" -STATIC void xfs_ail_splice(struct xfs_ail *, struct list_head *, xfs_lsn_t); -STATIC void xfs_ail_delete(struct xfs_ail *, xfs_log_item_t *); -STATIC xfs_log_item_t * xfs_ail_min(struct xfs_ail *); -STATIC xfs_log_item_t * xfs_ail_next(struct xfs_ail *, xfs_log_item_t *); +struct workqueue_struct *xfs_ail_wq; /* AIL workqueue */ #ifdef DEBUG -STATIC void xfs_ail_check(struct xfs_ail *, xfs_log_item_t *); -#else +/* + * Check that the list is sorted as it should be. + */ +STATIC void +xfs_ail_check( + struct xfs_ail *ailp, + xfs_log_item_t *lip) +{ + xfs_log_item_t *prev_lip; + + if (list_empty(&ailp->xa_ail)) + return; + + /* + * Check the next and previous entries are valid. + */ + ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); + prev_lip = list_entry(lip->li_ail.prev, xfs_log_item_t, li_ail); + if (&prev_lip->li_ail != &ailp->xa_ail) + ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); + + prev_lip = list_entry(lip->li_ail.next, xfs_log_item_t, li_ail); + if (&prev_lip->li_ail != &ailp->xa_ail) + ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) >= 0); + + +#ifdef XFS_TRANS_DEBUG + /* + * Walk the list checking lsn ordering, and that every entry has the + * XFS_LI_IN_AIL flag set. This is really expensive, so only do it + * when specifically debugging the transaction subsystem. + */ + prev_lip = list_entry(&ailp->xa_ail, xfs_log_item_t, li_ail); + list_for_each_entry(lip, &ailp->xa_ail, li_ail) { + if (&prev_lip->li_ail != &ailp->xa_ail) + ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); + ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); + prev_lip = lip; + } +#endif /* XFS_TRANS_DEBUG */ +} +#else /* !DEBUG */ #define xfs_ail_check(a,l) #endif /* DEBUG */ +/* + * Return a pointer to the first item in the AIL. If the AIL is empty, then + * return NULL. + */ +static xfs_log_item_t * +xfs_ail_min( + struct xfs_ail *ailp) +{ + if (list_empty(&ailp->xa_ail)) + return NULL; + + return list_first_entry(&ailp->xa_ail, xfs_log_item_t, li_ail); +} + + /* + * Return a pointer to the last item in the AIL. If the AIL is empty, then + * return NULL. + */ +static xfs_log_item_t * +xfs_ail_max( + struct xfs_ail *ailp) +{ + if (list_empty(&ailp->xa_ail)) + return NULL; + + return list_entry(ailp->xa_ail.prev, xfs_log_item_t, li_ail); +} + +/* + * Return a pointer to the item which follows the given item in the AIL. If + * the given item is the last item in the list, then return NULL. + */ +static xfs_log_item_t * +xfs_ail_next( + struct xfs_ail *ailp, + xfs_log_item_t *lip) +{ + if (lip->li_ail.next == &ailp->xa_ail) + return NULL; + + return list_first_entry(&lip->li_ail, xfs_log_item_t, li_ail); +} /* - * This is called by the log manager code to determine the LSN - * of the tail of the log. This is exactly the LSN of the first - * item in the AIL. If the AIL is empty, then this function - * returns 0. + * This is called by the log manager code to determine the LSN of the tail of + * the log. This is exactly the LSN of the first item in the AIL. If the AIL + * is empty, then this function returns 0. * - * We need the AIL lock in order to get a coherent read of the - * lsn of the last item in the AIL. + * We need the AIL lock in order to get a coherent read of the lsn of the last + * item in the AIL. */ xfs_lsn_t -xfs_trans_ail_tail( +xfs_ail_min_lsn( struct xfs_ail *ailp) { - xfs_lsn_t lsn; + xfs_lsn_t lsn = 0; xfs_log_item_t *lip; spin_lock(&ailp->xa_lock); lip = xfs_ail_min(ailp); - if (lip == NULL) { - lsn = (xfs_lsn_t)0; - } else { + if (lip) lsn = lip->li_lsn; - } spin_unlock(&ailp->xa_lock); return lsn; } /* - * xfs_trans_push_ail - * - * This routine is called to move the tail of the AIL forward. It does this by - * trying to flush items in the AIL whose lsns are below the given - * threshold_lsn. - * - * the push is run asynchronously in a separate thread, so we return the tail - * of the log right now instead of the tail after the push. This means we will - * either continue right away, or we will sleep waiting on the async thread to - * do its work. - * - * We do this unlocked - we only need to know whether there is anything in the - * AIL at the time we are called. We don't need to access the contents of - * any of the objects, so the lock is not needed. + * Return the maximum lsn held in the AIL, or zero if the AIL is empty. */ -void -xfs_trans_ail_push( - struct xfs_ail *ailp, - xfs_lsn_t threshold_lsn) +static xfs_lsn_t +xfs_ail_max_lsn( + struct xfs_ail *ailp) { - xfs_log_item_t *lip; + xfs_lsn_t lsn = 0; + xfs_log_item_t *lip; - lip = xfs_ail_min(ailp); - if (lip && !XFS_FORCED_SHUTDOWN(ailp->xa_mount)) { - if (XFS_LSN_CMP(threshold_lsn, ailp->xa_target) > 0) - xfsaild_wakeup(ailp, threshold_lsn); - } + spin_lock(&ailp->xa_lock); + lip = xfs_ail_max(ailp); + if (lip) + lsn = lip->li_lsn; + spin_unlock(&ailp->xa_lock); + + return lsn; } /* @@ -236,35 +300,78 @@ out: } /* - * xfsaild_push does the work of pushing on the AIL. Returning a timeout of - * zero indicates that the caller should sleep until woken. + * splice the log item list into the AIL at the given LSN. */ -long -xfsaild_push( - struct xfs_ail *ailp, - xfs_lsn_t *last_lsn) +static void +xfs_ail_splice( + struct xfs_ail *ailp, + struct list_head *list, + xfs_lsn_t lsn) { - long tout = 0; - xfs_lsn_t last_pushed_lsn = *last_lsn; - xfs_lsn_t target = ailp->xa_target; - xfs_lsn_t lsn; - xfs_log_item_t *lip; - int flush_log, count, stuck; - xfs_mount_t *mp = ailp->xa_mount; + xfs_log_item_t *next_lip; + + /* If the list is empty, just insert the item. */ + if (list_empty(&ailp->xa_ail)) { + list_splice(list, &ailp->xa_ail); + return; + } + + list_for_each_entry_reverse(next_lip, &ailp->xa_ail, li_ail) { + if (XFS_LSN_CMP(next_lip->li_lsn, lsn) <= 0) + break; + } + + ASSERT(&next_lip->li_ail == &ailp->xa_ail || + XFS_LSN_CMP(next_lip->li_lsn, lsn) <= 0); + + list_splice_init(list, &next_lip->li_ail); +} + +/* + * Delete the given item from the AIL. Return a pointer to the item. + */ +static void +xfs_ail_delete( + struct xfs_ail *ailp, + xfs_log_item_t *lip) +{ + xfs_ail_check(ailp, lip); + list_del(&lip->li_ail); + xfs_trans_ail_cursor_clear(ailp, lip); +} + +/* + * xfs_ail_worker does the work of pushing on the AIL. It will requeue itself + * to run at a later time if there is more work to do to complete the push. + */ +STATIC void +xfs_ail_worker( + struct work_struct *work) +{ + struct xfs_ail *ailp = container_of(to_delayed_work(work), + struct xfs_ail, xa_work); + xfs_mount_t *mp = ailp->xa_mount; struct xfs_ail_cursor *cur = &ailp->xa_cursors; - int push_xfsbufd = 0; + xfs_log_item_t *lip; + xfs_lsn_t lsn; + xfs_lsn_t target; + long tout = 10; + int flush_log = 0; + int stuck = 0; + int count = 0; + int push_xfsbufd = 0; spin_lock(&ailp->xa_lock); + target = ailp->xa_target; xfs_trans_ail_cursor_init(ailp, cur); - lip = xfs_trans_ail_cursor_first(ailp, cur, *last_lsn); + lip = xfs_trans_ail_cursor_first(ailp, cur, ailp->xa_last_pushed_lsn); if (!lip || XFS_FORCED_SHUTDOWN(mp)) { /* * AIL is empty or our push has reached the end. */ xfs_trans_ail_cursor_done(ailp, cur); spin_unlock(&ailp->xa_lock); - *last_lsn = 0; - return tout; + goto out_done; } XFS_STATS_INC(xs_push_ail); @@ -281,8 +388,7 @@ xfsaild_push( * lots of contention on the AIL lists. */ lsn = lip->li_lsn; - flush_log = stuck = count = 0; - while ((XFS_LSN_CMP(lip->li_lsn, target) < 0)) { + while ((XFS_LSN_CMP(lip->li_lsn, target) <= 0)) { int lock_result; /* * If we can lock the item without sleeping, unlock the AIL @@ -301,13 +407,13 @@ xfsaild_push( case XFS_ITEM_SUCCESS: XFS_STATS_INC(xs_push_ail_success); IOP_PUSH(lip); - last_pushed_lsn = lsn; + ailp->xa_last_pushed_lsn = lsn; break; case XFS_ITEM_PUSHBUF: XFS_STATS_INC(xs_push_ail_pushbuf); IOP_PUSHBUF(lip); - last_pushed_lsn = lsn; + ailp->xa_last_pushed_lsn = lsn; push_xfsbufd = 1; break; @@ -319,7 +425,7 @@ xfsaild_push( case XFS_ITEM_LOCKED: XFS_STATS_INC(xs_push_ail_locked); - last_pushed_lsn = lsn; + ailp->xa_last_pushed_lsn = lsn; stuck++; break; @@ -374,9 +480,27 @@ xfsaild_push( wake_up_process(mp->m_ddev_targp->bt_task); } + /* assume we have more work to do in a short while */ +out_done: if (!count) { /* We're past our target or empty, so idle */ - last_pushed_lsn = 0; + ailp->xa_last_pushed_lsn = 0; + + /* + * We clear the XFS_AIL_PUSHING_BIT first before checking + * whether the target has changed. If the target has changed, + * this pushes the requeue race directly onto the result of the + * atomic test/set bit, so we are guaranteed that either the + * the pusher that changed the target or ourselves will requeue + * the work (but not both). + */ + clear_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags); + smp_rmb(); + if (XFS_LSN_CMP(ailp->xa_target, target) == 0 || + test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags)) + return; + + tout = 50; } else if (XFS_LSN_CMP(lsn, target) >= 0) { /* * We reached the target so wait a bit longer for I/O to @@ -384,7 +508,7 @@ xfsaild_push( * start the next scan from the start of the AIL. */ tout = 50; - last_pushed_lsn = 0; + ailp->xa_last_pushed_lsn = 0; } else if ((stuck * 100) / count > 90) { /* * Either there is a lot of contention on the AIL or we @@ -396,14 +520,61 @@ xfsaild_push( * continuing from where we were. */ tout = 20; - } else { - /* more to do, but wait a short while before continuing */ - tout = 10; } - *last_lsn = last_pushed_lsn; - return tout; + + /* There is more to do, requeue us. */ + queue_delayed_work(xfs_syncd_wq, &ailp->xa_work, + msecs_to_jiffies(tout)); +} + +/* + * This routine is called to move the tail of the AIL forward. It does this by + * trying to flush items in the AIL whose lsns are below the given + * threshold_lsn. + * + * The push is run asynchronously in a workqueue, which means the caller needs + * to handle waiting on the async flush for space to become available. + * We don't want to interrupt any push that is in progress, hence we only queue + * work if we set the pushing bit approriately. + * + * We do this unlocked - we only need to know whether there is anything in the + * AIL at the time we are called. We don't need to access the contents of + * any of the objects, so the lock is not needed. + */ +void +xfs_ail_push( + struct xfs_ail *ailp, + xfs_lsn_t threshold_lsn) +{ + xfs_log_item_t *lip; + + lip = xfs_ail_min(ailp); + if (!lip || XFS_FORCED_SHUTDOWN(ailp->xa_mount) || + XFS_LSN_CMP(threshold_lsn, ailp->xa_target) <= 0) + return; + + /* + * Ensure that the new target is noticed in push code before it clears + * the XFS_AIL_PUSHING_BIT. + */ + smp_wmb(); + xfs_trans_ail_copy_lsn(ailp, &ailp->xa_target, &threshold_lsn); + if (!test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags)) + queue_delayed_work(xfs_syncd_wq, &ailp->xa_work, 0); } +/* + * Push out all items in the AIL immediately + */ +void +xfs_ail_push_all( + struct xfs_ail *ailp) +{ + xfs_lsn_t threshold_lsn = xfs_ail_max_lsn(ailp); + + if (threshold_lsn) + xfs_ail_push(ailp, threshold_lsn); +} /* * This is to be called when an item is unlocked that may have @@ -615,7 +786,6 @@ xfs_trans_ail_init( xfs_mount_t *mp) { struct xfs_ail *ailp; - int error; ailp = kmem_zalloc(sizeof(struct xfs_ail), KM_MAYFAIL); if (!ailp) @@ -624,15 +794,9 @@ xfs_trans_ail_init( ailp->xa_mount = mp; INIT_LIST_HEAD(&ailp->xa_ail); spin_lock_init(&ailp->xa_lock); - error = xfsaild_start(ailp); - if (error) - goto out_free_ailp; + INIT_DELAYED_WORK(&ailp->xa_work, xfs_ail_worker); mp->m_ail = ailp; return 0; - -out_free_ailp: - kmem_free(ailp); - return error; } void @@ -641,124 +805,6 @@ xfs_trans_ail_destroy( { struct xfs_ail *ailp = mp->m_ail; - xfsaild_stop(ailp); + cancel_delayed_work_sync(&ailp->xa_work); kmem_free(ailp); } - -/* - * splice the log item list into the AIL at the given LSN. - */ -STATIC void -xfs_ail_splice( - struct xfs_ail *ailp, - struct list_head *list, - xfs_lsn_t lsn) -{ - xfs_log_item_t *next_lip; - - /* - * If the list is empty, just insert the item. - */ - if (list_empty(&ailp->xa_ail)) { - list_splice(list, &ailp->xa_ail); - return; - } - - list_for_each_entry_reverse(next_lip, &ailp->xa_ail, li_ail) { - if (XFS_LSN_CMP(next_lip->li_lsn, lsn) <= 0) - break; - } - - ASSERT((&next_lip->li_ail == &ailp->xa_ail) || - (XFS_LSN_CMP(next_lip->li_lsn, lsn) <= 0)); - - list_splice_init(list, &next_lip->li_ail); - return; -} - -/* - * Delete the given item from the AIL. Return a pointer to the item. - */ -STATIC void -xfs_ail_delete( - struct xfs_ail *ailp, - xfs_log_item_t *lip) -{ - xfs_ail_check(ailp, lip); - list_del(&lip->li_ail); - xfs_trans_ail_cursor_clear(ailp, lip); -} - -/* - * Return a pointer to the first item in the AIL. - * If the AIL is empty, then return NULL. - */ -STATIC xfs_log_item_t * -xfs_ail_min( - struct xfs_ail *ailp) -{ - if (list_empty(&ailp->xa_ail)) - return NULL; - - return list_first_entry(&ailp->xa_ail, xfs_log_item_t, li_ail); -} - -/* - * Return a pointer to the item which follows - * the given item in the AIL. If the given item - * is the last item in the list, then return NULL. - */ -STATIC xfs_log_item_t * -xfs_ail_next( - struct xfs_ail *ailp, - xfs_log_item_t *lip) -{ - if (lip->li_ail.next == &ailp->xa_ail) - return NULL; - - return list_first_entry(&lip->li_ail, xfs_log_item_t, li_ail); -} - -#ifdef DEBUG -/* - * Check that the list is sorted as it should be. - */ -STATIC void -xfs_ail_check( - struct xfs_ail *ailp, - xfs_log_item_t *lip) -{ - xfs_log_item_t *prev_lip; - - if (list_empty(&ailp->xa_ail)) - return; - - /* - * Check the next and previous entries are valid. - */ - ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); - prev_lip = list_entry(lip->li_ail.prev, xfs_log_item_t, li_ail); - if (&prev_lip->li_ail != &ailp->xa_ail) - ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); - - prev_lip = list_entry(lip->li_ail.next, xfs_log_item_t, li_ail); - if (&prev_lip->li_ail != &ailp->xa_ail) - ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) >= 0); - - -#ifdef XFS_TRANS_DEBUG - /* - * Walk the list checking lsn ordering, and that every entry has the - * XFS_LI_IN_AIL flag set. This is really expensive, so only do it - * when specifically debugging the transaction subsystem. - */ - prev_lip = list_entry(&ailp->xa_ail, xfs_log_item_t, li_ail); - list_for_each_entry(lip, &ailp->xa_ail, li_ail) { - if (&prev_lip->li_ail != &ailp->xa_ail) - ASSERT(XFS_LSN_CMP(prev_lip->li_lsn, lip->li_lsn) <= 0); - ASSERT((lip->li_flags & XFS_LI_IN_AIL) != 0); - prev_lip = lip; - } -#endif /* XFS_TRANS_DEBUG */ -} -#endif /* DEBUG */ diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c index 16084d8ea231..048b0c689d3e 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/xfs_trans_inode.c @@ -81,7 +81,7 @@ xfs_trans_ijoin( * * * Grabs a reference to the inode which will be dropped when the transaction - * is commited. The inode will also be unlocked at that point. The inode + * is committed. The inode will also be unlocked at that point. The inode * must be locked, and it cannot be associated with any transaction. */ void diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 35162c238fa3..6b164e9e9a1f 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -65,16 +65,22 @@ struct xfs_ail_cursor { struct xfs_ail { struct xfs_mount *xa_mount; struct list_head xa_ail; - uint xa_gen; - struct task_struct *xa_task; xfs_lsn_t xa_target; struct xfs_ail_cursor xa_cursors; spinlock_t xa_lock; + struct delayed_work xa_work; + xfs_lsn_t xa_last_pushed_lsn; + unsigned long xa_flags; }; +#define XFS_AIL_PUSHING_BIT 0 + /* * From xfs_trans_ail.c */ + +extern struct workqueue_struct *xfs_ail_wq; /* AIL workqueue */ + void xfs_trans_ail_update_bulk(struct xfs_ail *ailp, struct xfs_log_item **log_items, int nr_items, xfs_lsn_t lsn) __releases(ailp->xa_lock); @@ -98,12 +104,13 @@ xfs_trans_ail_delete( xfs_trans_ail_delete_bulk(ailp, &lip, 1); } -void xfs_trans_ail_push(struct xfs_ail *, xfs_lsn_t); +void xfs_ail_push(struct xfs_ail *, xfs_lsn_t); +void xfs_ail_push_all(struct xfs_ail *); +xfs_lsn_t xfs_ail_min_lsn(struct xfs_ail *ailp); + void xfs_trans_unlocked_item(struct xfs_ail *, xfs_log_item_t *); -xfs_lsn_t xfs_trans_ail_tail(struct xfs_ail *ailp); - struct xfs_log_item *xfs_trans_ail_cursor_first(struct xfs_ail *ailp, struct xfs_ail_cursor *cur, xfs_lsn_t lsn); @@ -112,11 +119,6 @@ struct xfs_log_item *xfs_trans_ail_cursor_next(struct xfs_ail *ailp, void xfs_trans_ail_cursor_done(struct xfs_ail *ailp, struct xfs_ail_cursor *cur); -long xfsaild_push(struct xfs_ail *, xfs_lsn_t *); -void xfsaild_wakeup(struct xfs_ail *, xfs_lsn_t); -int xfsaild_start(struct xfs_ail *); -void xfsaild_stop(struct xfs_ail *); - #if BITS_PER_LONG != 64 static inline void xfs_trans_ail_copy_lsn( diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index c48b4217ec47..b7a5fe7c52c8 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -953,7 +953,7 @@ xfs_release( * If we previously truncated this file and removed old data * in the process, we want to initiate "early" writeout on * the last close. This is an attempt to combat the notorious - * NULL files problem which is particularly noticable from a + * NULL files problem which is particularly noticeable from a * truncate down, buffered (re-)write (delalloc), followed by * a crash. What we are effectively doing here is * significantly reducing the time window where we'd otherwise @@ -982,7 +982,7 @@ xfs_release( * * Further, check if the inode is being opened, written and * closed frequently and we have delayed allocation blocks - * oustanding (e.g. streaming writes from the NFS server), + * outstanding (e.g. streaming writes from the NFS server), * truncating the blocks past EOF will cause fragmentation to * occur. * |