From aa38a711a893accf5b5192f3d705a120deaa81e0 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Fri, 18 Nov 2011 17:43:00 +0800 Subject: Btrfs: fix deadlock on metadata reservation when evicting a inode When I ran the xfstests, I found the test tasks was blocked on meta-data reservation. By debugging, I found the reason of this bug: start transaction | v reserve meta-data space | v flush delay allocation -> iput inode -> evict inode ^ | | v wait for delay allocation flush <- reserve meta-data space And besides that, the flush on evicting inode will block the thread, which is reclaiming the memory, and make oom happen easily. Fix this bug by skipping the flush step when evicting inode. Signed-off-by: Miao Xie --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8ad26b135a1..c5ccec23984 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3490,7 +3490,7 @@ void btrfs_evict_inode(struct inode *inode) * doing the truncate. */ while (1) { - ret = btrfs_block_rsv_refill(root, rsv, min_size); + ret = btrfs_block_rsv_refill_noflush(root, rsv, min_size); /* * Try and steal from the global reserve since we will -- cgit v1.2.3-18-g5258 From 42b2aa86c6670347a2a07e6d7af0e0ecc8fdbff9 Mon Sep 17 00:00:00 2001 From: "Justin P. Mattock" Date: Mon, 28 Nov 2011 20:31:00 -0800 Subject: treewide: Fix typos in various parts of the kernel, and fix some comments. The below patch fixes some typos in various parts of the kernel, as well as fixes some comments. Please let me know if I missed anything, and I will try to get it changed and resent. Signed-off-by: Justin P. Mattock Acked-by: Randy Dunlap Signed-off-by: Jiri Kosina --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 116ab67a06d..c3308c38ae7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1943,7 +1943,7 @@ enum btrfs_orphan_cleanup_state { }; /* - * This is called in transaction commmit time. If there are no orphan + * This is called in transaction commit time. If there are no orphan * files in the subvolume, it removes orphan item and frees block_rsv * structure. */ -- cgit v1.2.3-18-g5258 From f4a2f4c548296168832ad4ab7e7f7b0cd0bf1214 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Wed, 14 Dec 2011 20:12:01 -0500 Subject: Btrfs: fix wrong i_size when truncating a file to a larger size Btrfsck report error 100 after the 83th case of xfstests was run, it means the i_size of the file is wrong. The reason of this bug is that: Btrfs increased i_size of the file at the beginning, but it failed to expand the file, and failed to update the i_size to the old size because there is no enough space in the file system, so we found a wrong i_size. This patch fixes this bug by updating the i_size just when we pass the file expanding and get enough space to update i-node. Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c5ccec23984..4bbceb928af 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3370,6 +3370,8 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) static int btrfs_setsize(struct inode *inode, loff_t newsize) { + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_trans_handle *trans; loff_t oldsize = i_size_read(inode); int ret; @@ -3377,16 +3379,20 @@ static int btrfs_setsize(struct inode *inode, loff_t newsize) return 0; if (newsize > oldsize) { - i_size_write(inode, newsize); - btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); truncate_pagecache(inode, oldsize, newsize); ret = btrfs_cont_expand(inode, oldsize, newsize); - if (ret) { - btrfs_setsize(inode, oldsize); + if (ret) return ret; - } - mark_inode_dirty(inode); + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) + return PTR_ERR(trans); + + i_size_write(inode, newsize); + btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); + ret = btrfs_update_inode(trans, root, inode); + + btrfs_end_transaction_throttle(trans, root); } else { /* -- cgit v1.2.3-18-g5258 From 3642320e07444cc46327b24977d752f99706dac2 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Wed, 14 Dec 2011 20:12:02 -0500 Subject: Btrfs: fix wrong disk space information of the files Btrfsck report errors after the 83th case of xfstests was run, The error number is 400, it means the used disk space of the file is wrong. The reason of this bug is that: The file truncation may fail when the space of the file system is not enough, and leave some file extents, whose offset are beyond the end of the files. When we want to expand those files, we will drop those file extents, and put in dummy file extents, and then we should update the i-node. But btrfs forgets to do it. This patch adds the forgotten i-node update. Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4bbceb928af..f1c4bceed07 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3327,7 +3327,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) u64 hint_byte = 0; hole_size = last_byte - cur_offset; - trans = btrfs_start_transaction(root, 2); + trans = btrfs_start_transaction(root, 3); if (IS_ERR(trans)) { err = PTR_ERR(trans); break; @@ -3337,6 +3337,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) cur_offset + hole_size, &hint_byte, 1); if (err) { + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); break; } @@ -3346,6 +3347,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) 0, hole_size, 0, hole_size, 0, 0, 0); if (err) { + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); break; } @@ -3353,6 +3355,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) btrfs_drop_extent_cache(inode, hole_start, last_byte - 1, 0); + btrfs_update_inode(trans, root, inode); btrfs_end_transaction(trans, root); } free_extent_map(em); -- cgit v1.2.3-18-g5258 From f8e9e0b07be0464e12366631da3da73a1a62449c Mon Sep 17 00:00:00 2001 From: Arne Jansen Date: Wed, 14 Dec 2011 20:12:02 -0500 Subject: btrfs: keep orphans for subvolume deletion Since we have the free space caches, btrfs_orphan_cleanup also runs for the tree_root. Unfortunately this also cleans up the orphans used to mark subvol deletions in progress. Currently if a subvol deletion gets interrupted twice by umount/mount, the deletion will not be continued and the space permanently lost, though it would be possible to write a tool to recover those lost subvol deletions. This patch checks if the orphan belongs to a subvol (dead root) and skips the deletion. Signed-off-by: Arne Jansen Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index f1c4bceed07..4a31493d97a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2158,6 +2158,38 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) if (ret && ret != -ESTALE) goto out; + if (ret == -ESTALE && root == root->fs_info->tree_root) { + struct btrfs_root *dead_root; + struct btrfs_fs_info *fs_info = root->fs_info; + int is_dead_root = 0; + + /* + * this is an orphan in the tree root. Currently these + * could come from 2 sources: + * a) a snapshot deletion in progress + * b) a free space cache inode + * We need to distinguish those two, as the snapshot + * orphan must not get deleted. + * find_dead_roots already ran before us, so if this + * is a snapshot deletion, we should find the root + * in the dead_roots list + */ + spin_lock(&fs_info->trans_lock); + list_for_each_entry(dead_root, &fs_info->dead_roots, + root_list) { + if (dead_root->root_key.objectid == + found_key.objectid) { + is_dead_root = 1; + break; + } + } + spin_unlock(&fs_info->trans_lock); + if (is_dead_root) { + /* prevent this orphan from being found again */ + key.offset = found_key.objectid - 1; + continue; + } + } /* * Inode is already gone but the orphan item is still there, * kill the orphan item. -- cgit v1.2.3-18-g5258 From ad19db71f498fd858dd84ce603efcf97e321f184 Mon Sep 17 00:00:00 2001 From: Casey Schaufler Date: Thu, 15 Dec 2011 10:09:07 -0500 Subject: BTRFS: Establish i_ops before calling d_instantiate The Smack LSM hook for security_d_instantiate checks the inode's i_op->getxattr value to determine if the containing filesystem supports extended attributes. The BTRFS filesystem sets the inode's i_op value only after it has instantiated the inode. This results in Smack incorrectly giving new BTRFS inodes attributes from the filesystem defaults on the assumption that values can't be stored on the filesystem. This patch moves the assignment of inode operation vectors ahead of the calls to d_instantiate, letting Smack know that the filesystem supports extended attributes. There should be no impact on the performance or behavior of BTRFS. Signed-off-by: Casey Schaufler Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 31 ++++++++++++++++++++++++++----- 1 file changed, 26 insertions(+), 5 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4a31493d97a..d4a9195c7f0 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4596,11 +4596,18 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + + inode->i_op = &btrfs_special_inode_operations; err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { - inode->i_op = &btrfs_special_inode_operations; init_special_inode(inode, inode->i_mode, rdev); btrfs_update_inode(trans, root, inode); } @@ -4654,14 +4661,21 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + inode->i_fop = &btrfs_file_operations; + inode->i_op = &btrfs_file_inode_operations; + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { inode->i_mapping->a_ops = &btrfs_aops; inode->i_mapping->backing_dev_info = &root->fs_info->bdi; - inode->i_fop = &btrfs_file_operations; - inode->i_op = &btrfs_file_inode_operations; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; } out_unlock: @@ -7117,14 +7131,21 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock; } + /* + * If the active LSM wants to access the inode during + * d_instantiate it needs these. Smack checks to see + * if the filesystem supports xattrs by looking at the + * ops vector. + */ + inode->i_fop = &btrfs_file_operations; + inode->i_op = &btrfs_file_inode_operations; + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { inode->i_mapping->a_ops = &btrfs_aops; inode->i_mapping->backing_dev_info = &root->fs_info->bdi; - inode->i_fop = &btrfs_file_operations; - inode->i_op = &btrfs_file_inode_operations; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; } if (drop_inode) -- cgit v1.2.3-18-g5258 From 22c44fe65adacd20a174f3f54686509ee94ef7be Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Wed, 30 Nov 2011 10:45:38 -0500 Subject: Btrfs: deal with enospc from dirtying inodes properly Now that we're properly keeping track of delayed inode space we've been getting a lot of warnings out of btrfs_dirty_inode() when running xfstest 83. This is because a bunch of people call mark_inode_dirty, which is void so we can't return ENOSPC. This needs to be fixed in a few areas 1) file_update_time - this updates the mtime and such when writing to a file, which will call mark_inode_dirty. So copy file_update_time into btrfs so we can call btrfs_dirty_inode directly and return an error if we get one appropriately. 2) fix symlinks to use btrfs_setattr for ->setattr. For some reason we weren't setting ->setattr for symlinks, even though we should have been. This catches one of the cases where we were getting errors in mark_inode_dirty. 3) Fix btrfs_setattr and btrfs_setsize to call btrfs_dirty_inode directly instead of mark_inode_dirty. This lets us return errors properly for truncate and chown/anything related to setattr. 4) Add a new btrfs_fs_dirty_inode which will just call btrfs_dirty_inode and print an error if we have one. The only remaining user we can't control for this is touch_atime(), but we don't really want to keep people from walking down the tree if we don't have space to save the atime update, so just complain but don't worry about it. With this patch xfstests 83 complains a handful of times instead of hundreds of times. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 80 ++++++++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 61 insertions(+), 19 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c5ccec23984..eec33b9c953 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -38,6 +38,7 @@ #include #include #include +#include #include "compat.h" #include "ctree.h" #include "disk-io.h" @@ -3386,7 +3387,7 @@ static int btrfs_setsize(struct inode *inode, loff_t newsize) return ret; } - mark_inode_dirty(inode); + ret = btrfs_dirty_inode(inode); } else { /* @@ -3426,9 +3427,9 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid) { setattr_copy(inode, attr); - mark_inode_dirty(inode); + err = btrfs_dirty_inode(inode); - if (attr->ia_valid & ATTR_MODE) + if (!err && attr->ia_valid & ATTR_MODE) err = btrfs_acl_chmod(inode); } @@ -4204,42 +4205,80 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) * FIXME, needs more benchmarking...there are no reasons other than performance * to keep or drop this code. */ -void btrfs_dirty_inode(struct inode *inode, int flags) +int btrfs_dirty_inode(struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; int ret; if (BTRFS_I(inode)->dummy_inode) - return; + return 0; trans = btrfs_join_transaction(root); - BUG_ON(IS_ERR(trans)); + if (IS_ERR(trans)) + return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, inode); if (ret && ret == -ENOSPC) { /* whoops, lets try again with the full transaction */ btrfs_end_transaction(trans, root); trans = btrfs_start_transaction(root, 1); - if (IS_ERR(trans)) { - printk_ratelimited(KERN_ERR "btrfs: fail to " - "dirty inode %llu error %ld\n", - (unsigned long long)btrfs_ino(inode), - PTR_ERR(trans)); - return; - } + if (IS_ERR(trans)) + return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, inode); - if (ret) { - printk_ratelimited(KERN_ERR "btrfs: fail to " - "dirty inode %llu error %d\n", - (unsigned long long)btrfs_ino(inode), - ret); - } } btrfs_end_transaction(trans, root); if (BTRFS_I(inode)->delayed_node) btrfs_balance_delayed_items(root); + + return ret; +} + +/* + * This is a copy of file_update_time. We need this so we can return error on + * ENOSPC for updating the inode in the case of file write and mmap writes. + */ +int btrfs_update_time(struct file *file) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct timespec now; + int ret; + enum { S_MTIME = 1, S_CTIME = 2, S_VERSION = 4 } sync_it = 0; + + /* First try to exhaust all avenues to not sync */ + if (IS_NOCMTIME(inode)) + return 0; + + now = current_fs_time(inode->i_sb); + if (!timespec_equal(&inode->i_mtime, &now)) + sync_it = S_MTIME; + + if (!timespec_equal(&inode->i_ctime, &now)) + sync_it |= S_CTIME; + + if (IS_I_VERSION(inode)) + sync_it |= S_VERSION; + + if (!sync_it) + return 0; + + /* Finally allowed to write? Takes lock. */ + if (mnt_want_write_file(file)) + return 0; + + /* Only change inode inside the lock region */ + if (sync_it & S_VERSION) + inode_inc_iversion(inode); + if (sync_it & S_CTIME) + inode->i_ctime = now; + if (sync_it & S_MTIME) + inode->i_mtime = now; + ret = btrfs_dirty_inode(inode); + if (!ret) + mark_inode_dirty_sync(inode); + mnt_drop_write(file->f_path.mnt); + return ret; } /* @@ -6304,6 +6343,8 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) u64 page_end; ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); + if (!ret) + ret = btrfs_update_time(vma->vm_file); if (ret) { if (ret == -ENOMEM) ret = VM_FAULT_OOM; @@ -7353,6 +7394,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = { .follow_link = page_follow_link_light, .put_link = page_put_link, .getattr = btrfs_getattr, + .setattr = btrfs_setattr, .permission = btrfs_permission, .setxattr = btrfs_setxattr, .getxattr = btrfs_getxattr, -- cgit v1.2.3-18-g5258 From 660d3f6cde552323578b85fc5a09a6742f1fe804 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 9 Dec 2011 11:18:51 -0500 Subject: Btrfs: fix how we do delalloc reservations and how we free reservations on error Running xfstests 269 with some tracing my scripts kept spitting out errors about releasing bytes that we didn't actually have reserved. This took me down a huge rabbit hole and it turns out the way we deal with reserved_extents is wrong, we need to only be setting it if the reservation succeeds, otherwise the free() method will come in and unreserve space that isn't actually reserved yet, which can lead to other warnings and such. The math was all working out right in the end, but it caused all sorts of other issues in addition to making my scripts yell and scream and generally make it impossible for me to track down the original issue I was looking for. The other problem is with our error handling in the reservation code. There are two cases that we need to deal with 1) We raced with free. In this case free won't free anything because csum_bytes is modified before we dro the lock in our reservation path, so free rightly doesn't release any space because the reservation code may be depending on that reservation. However if we fail, we need the reservation side to do the free at that point since that space is no longer in use. So as it stands the code was doing this fine and it worked out, except in case #2 2) We don't race with free. Nobody comes in and changes anything, and our reservation fails. In this case we didn't reserve anything anyway and we just need to clean up csum_bytes but not free anything. So we keep track of csum_bytes before we drop the lock and if it hasn't changed we know we can just decrement csum_bytes and carry on. Because of the case where we can race with free()'s since we have to drop our spin_lock to do the reservation, I'm going to serialize all reservations with the i_mutex. We already get this for free in the heavy use paths, truncate and file write all hold the i_mutex, just needed to add it to page_mkwrite and various ioctl/balance things. With this patch my space leak scripts no longer scream bloody murder. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index eec33b9c953..8938174e6bc 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2192,7 +2192,14 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) continue; } nr_truncate++; + /* + * Need to hold the imutex for reservation purposes, not + * a huge deal here but I have a WARN_ON in + * btrfs_delalloc_reserve_space to catch offenders. + */ + mutex_lock(&inode->i_mutex); ret = btrfs_truncate(inode); + mutex_unlock(&inode->i_mutex); } else { nr_unlink++; } @@ -6342,7 +6349,10 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) u64 page_start; u64 page_end; + /* Need this to keep space reservations serialized */ + mutex_lock(&inode->i_mutex); ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); + mutex_unlock(&inode->i_mutex); if (!ret) ret = btrfs_update_time(vma->vm_file); if (ret) { -- cgit v1.2.3-18-g5258 From 7041ee97281c30a78658904140c7bd9373a36142 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 9 Dec 2011 13:26:22 -0500 Subject: Btrfs: fix leaked space in truncate We were occasionaly leaking space when running xfstest 269. This is because if we failed to start the transaction in the truncate loop we'd just goto out, but we need to break so that the inode is removed from the orphan list and the space is properly freed. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8938174e6bc..6349c63a4b3 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6566,8 +6566,9 @@ static int btrfs_truncate(struct inode *inode) /* Just need the 1 for updating the inode */ trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - err = PTR_ERR(trans); - goto out; + ret = err = PTR_ERR(trans); + trans = NULL; + break; } } -- cgit v1.2.3-18-g5258 From ee4d89f0c4967c624c92516fcc37b41069bfdc23 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 13 Dec 2011 12:55:58 -0500 Subject: Btrfs: don't panic if orphan item already exists I've been hitting this BUG_ON() in btrfs_orphan_add when running xfstest 269 in a loop. This is because we will add an orphan item, do the truncate, the truncate will fail for whatever reason (*cough*ENOSPC*cough*) and then we're left with an orphan item still in the fs. Then we come back later to do another truncate and it blows up because we already have an orphan item. This is ok so just fix the BUG_ON() to only BUG() if ret is not EEXIST. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6349c63a4b3..b212f391cea 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2032,7 +2032,7 @@ int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode) /* insert an orphan item to track this unlinked/truncated file */ if (insert >= 1) { ret = btrfs_insert_orphan_item(trans, root, btrfs_ino(inode)); - BUG_ON(ret); + BUG_ON(ret && ret != -EEXIST); } /* insert an orphan item to track subvolume contains orphan files */ -- cgit v1.2.3-18-g5258 From 66d7e7f09f77456fe68683247d77721032a00ee5 Mon Sep 17 00:00:00 2001 From: Arne Jansen Date: Mon, 12 Sep 2011 15:26:38 +0200 Subject: Btrfs: mark delayed refs as for cow Add a for_cow parameter to add_delayed_*_ref and pass the appropriate value from every call site. The for_cow parameter will later on be used to determine if a ref will change anything with respect to qgroups. Delayed refs coming from relocation are always counted as for_cow, as they don't change subvol quota. Also pass in the fs_info for later use. btrfs_find_all_roots() will use this as an optimization, as changes that are for_cow will not change anything with respect to which root points to a certain leaf. Thus, we don't need to add the current sequence number to those delayed refs. Signed-off-by: Arne Jansen Signed-off-by: Jan Schmidt --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index c5ccec23984..ea819386b86 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3139,7 +3139,7 @@ delete: ret = btrfs_free_extent(trans, root, extent_start, extent_num_bytes, 0, btrfs_header_owner(leaf), - ino, extent_offset); + ino, extent_offset, 0); BUG_ON(ret); } -- cgit v1.2.3-18-g5258 From 08c422c27f855d27b0b3d9fa30ebd938d4ae6f1f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 23 Dec 2011 07:58:13 -0500 Subject: Btrfs: call d_instantiate after all ops are setup This closes races where btrfs is calling d_instantiate too soon during inode creation. All of the callers of btrfs_add_nondir are updated to instantiate after the inode is fully setup in memory. Signed-off-by: Al Viro Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 740e67bbe24..13b0542015f 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4590,10 +4590,6 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans, int err = btrfs_add_link(trans, dir, inode, dentry->d_name.name, dentry->d_name.len, backref, index); - if (!err) { - d_instantiate(dentry, inode); - return 0; - } if (err > 0) err = -EEXIST; return err; @@ -4655,6 +4651,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, else { init_special_inode(inode, inode->i_mode, rdev); btrfs_update_inode(trans, root, inode); + d_instantiate(dentry, inode); } out_unlock: nr = trans->blocks_used; @@ -4722,6 +4719,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, inode->i_mapping->a_ops = &btrfs_aops; inode->i_mapping->backing_dev_info = &root->fs_info->bdi; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; + d_instantiate(dentry, inode); } out_unlock: nr = trans->blocks_used; @@ -4779,6 +4777,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *parent = dentry->d_parent; err = btrfs_update_inode(trans, root, inode); BUG_ON(err); + d_instantiate(dentry, inode); btrfs_log_new_name(trans, inode, NULL, parent); } @@ -7245,6 +7244,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, drop_inode = 1; out_unlock: + if (!err) + d_instantiate(dentry, inode); nr = trans->blocks_used; btrfs_end_transaction_throttle(trans, root); if (drop_inode) { -- cgit v1.2.3-18-g5258 From 6b520e0565422966cdf1c3759bd73df77b0f248c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 12 Dec 2011 15:51:45 -0500 Subject: vfs: fix the stupidity with i_dentry in inode destructors Seeing that just about every destructor got that INIT_LIST_HEAD() copied into it, there is no point whatsoever keeping this INIT_LIST_HEAD in inode_init_once(); the cost of taking it into inode_init_always() will be negligible for pipes and sockets and negative for everything else. Not to mention the removal of boilerplate code from ->destroy_inode() instances... Signed-off-by: Al Viro --- fs/btrfs/inode.c | 1 - 1 file changed, 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index fd1a06df5bc..f8ff9738558 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6761,7 +6761,6 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) static void btrfs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); - INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); } -- cgit v1.2.3-18-g5258 From 18bb1db3e7607e4a997d50991a6f9fa5b0f8722c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 26 Jul 2011 01:41:39 -0400 Subject: switch vfs_mkdir() and ->mkdir() to umode_t vfs_mkdir() gets int, but immediately drops everything that might not fit into umode_t and that's the only caller of ->mkdir()... Signed-off-by: Al Viro --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index f8ff9738558..e30de56e6b6 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4792,7 +4792,7 @@ fail: return err; } -static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct inode *inode = NULL; struct btrfs_trans_handle *trans; -- cgit v1.2.3-18-g5258 From 4acdaf27ebe2034c342f3be57ef49aed1ad885ef Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 26 Jul 2011 01:42:34 -0400 Subject: switch ->create() to umode_t vfs_create() ignores everything outside of 16bit subset of its mode argument; switching it to umode_t is obviously equivalent and it's the only caller of the method Signed-off-by: Al Viro --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index e30de56e6b6..19630aacb32 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4665,7 +4665,7 @@ out_unlock: } static int btrfs_create(struct inode *dir, struct dentry *dentry, - int mode, struct nameidata *nd) + umode_t mode, struct nameidata *nd) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; -- cgit v1.2.3-18-g5258 From 1a67aafb5f72a436ca044293309fa7e6351d6a35 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 26 Jul 2011 01:52:52 -0400 Subject: switch ->mknod() to umode_t Signed-off-by: Al Viro --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 19630aacb32..0060875d6af 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4596,7 +4596,7 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans, } static int btrfs_mknod(struct inode *dir, struct dentry *dentry, - int mode, dev_t rdev) + umode_t mode, dev_t rdev) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; -- cgit v1.2.3-18-g5258 From 175a4eb7ea531cdbf6d574f5d5ba9aa0f5e8ed13 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 26 Jul 2011 03:30:54 -0400 Subject: fs: propagate umode_t, misc bits Signed-off-by: Al Viro --- fs/btrfs/inode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 0060875d6af..2f426a51e60 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4412,8 +4412,8 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *dir, const char *name, int name_len, - u64 ref_objectid, u64 objectid, int mode, - u64 *index) + u64 ref_objectid, u64 objectid, + umode_t mode, u64 *index) { struct inode *inode; struct btrfs_inode_item *inode_item; -- cgit v1.2.3-18-g5258 From 6bf7e080d5bcb0d399ee38ce3dabbfad64448192 Mon Sep 17 00:00:00 2001 From: Jan Schmidt Date: Thu, 1 Dec 2011 14:35:19 +0100 Subject: Btrfs: make sure we're not using obsolete code in btrfs_get_extent There's code in btrfs_get_extent that should never be used. This patch turns a WARN_ON(1) into a BUG(), hoping we can remove the transaction code from btrfs_get_extent soon. Signed-off-by: Jan Schmidt --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ea819386b86..603d740f0f1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5022,7 +5022,7 @@ again: } flush_dcache_page(page); } else if (create && PageUptodate(page)) { - WARN_ON(1); + BUG(); if (!trans) { kunmap(page); free_extent_map(em); -- cgit v1.2.3-18-g5258 From 7ad85bb76a61801362701b77c5cee5aa09f35369 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 12 Jan 2012 19:10:12 -0500 Subject: Btrfs: do not use btrfs_end_transaction_throttle everywhere A user reported a problem where things like open with O_CREAT would take up to 30 seconds when he had nfs activity on the same mount. This is because all of our quick metadata operations, like create, symlink etc all do btrfs_end_transaction_throttle, which if the transaction is blocked will wait for the commit to complete before it returns. This adds a ridiculous amount of latency and isn't really needed. The normal btrfs_end_transaction will mark the transaction as blocked and wake the transaction kthread up if it thinks the transaction needs to end (this being in the running out of global reserve space scenario), and this is all that is really needed since we've already done everything we're going to do, we just need to return. This should help people with the latency they were seeing when using synchronous heavy workloads. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index acc4ff39ca4..5f8ba210c0a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2845,7 +2845,7 @@ static void __unlink_end_trans(struct btrfs_trans_handle *trans, BUG_ON(!root->fs_info->enospc_unlink); root->fs_info->enospc_unlink = 0; } - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); } static int btrfs_unlink(struct inode *dir, struct dentry *dentry) @@ -3434,7 +3434,7 @@ static int btrfs_setsize(struct inode *inode, loff_t newsize) i_size_write(inode, newsize); btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); ret = btrfs_update_inode(trans, root, inode); - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); } else { /* @@ -4655,7 +4655,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, } out_unlock: nr = trans->blocks_used; - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); btrfs_btree_balance_dirty(root, nr); if (drop_inode) { inode_dec_link_count(inode); @@ -4723,7 +4723,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, } out_unlock: nr = trans->blocks_used; - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); if (drop_inode) { inode_dec_link_count(inode); iput(inode); @@ -4782,7 +4782,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, } nr = trans->blocks_used; - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); fail: if (drop_inode) { inode_dec_link_count(inode); @@ -4848,7 +4848,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) out_fail: nr = trans->blocks_used; - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); if (drop_on_err) iput(inode); btrfs_btree_balance_dirty(root, nr); @@ -6668,7 +6668,7 @@ end_trans: err = ret; nr = trans->blocks_used; - ret = btrfs_end_transaction_throttle(trans, root); + ret = btrfs_end_transaction(trans, root); btrfs_btree_balance_dirty(root, nr); } @@ -7075,7 +7075,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, btrfs_end_log_trans(root); } out_fail: - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); out_notrans: if (old_ino == BTRFS_FIRST_FREE_OBJECTID) up_read(&root->fs_info->subvol_sem); @@ -7247,7 +7247,7 @@ out_unlock: if (!err) d_instantiate(dentry, inode); nr = trans->blocks_used; - btrfs_end_transaction_throttle(trans, root); + btrfs_end_transaction(trans, root); if (drop_inode) { inode_dec_link_count(inode); iput(inode); -- cgit v1.2.3-18-g5258 From f70a9a6b94af86fca069a7552ab672c31b457786 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Thu, 12 Jan 2012 19:10:12 -0500 Subject: Btrfs: fix btrfsck error 400 when truncating a compressed Reproduce steps: # mkfs.btrfs /dev/sdb5 # mount /dev/sdb5 -o compress=lzo /mnt # dd if=/dev/zero of=/mnt/tmpfile bs=128K count=1 # sync # truncate -s 64K /mnt/tmpfile root 5 inode 257 errors 400 This is because of the wrong if condition, which is used to check if we should subtract the bytes of the dropped range from i_blocks/i_bytes of i-node or not. When we truncate a compressed extent, btrfs substracts the bytes of the whole extent, it's wrong. We should substract the real size that we truncate, no matter it is a compressed extent or not. Fix it. Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5f8ba210c0a..946a7f1b329 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3009,7 +3009,6 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, int pending_del_nr = 0; int pending_del_slot = 0; int extent_type = -1; - int encoding; int ret; int err = 0; u64 ino = btrfs_ino(inode); @@ -3059,7 +3058,6 @@ search_again: leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); found_type = btrfs_key_type(&found_key); - encoding = 0; if (found_key.objectid != ino) break; @@ -3072,10 +3070,6 @@ search_again: fi = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_file_extent_item); extent_type = btrfs_file_extent_type(leaf, fi); - encoding = btrfs_file_extent_compression(leaf, fi); - encoding |= btrfs_file_extent_encryption(leaf, fi); - encoding |= btrfs_file_extent_other_encoding(leaf, fi); - if (extent_type != BTRFS_FILE_EXTENT_INLINE) { item_end += btrfs_file_extent_num_bytes(leaf, fi); @@ -3103,7 +3097,7 @@ search_again: if (extent_type != BTRFS_FILE_EXTENT_INLINE) { u64 num_dec; extent_start = btrfs_file_extent_disk_bytenr(leaf, fi); - if (!del_item && !encoding) { + if (!del_item) { u64 orig_num_bytes = btrfs_file_extent_num_bytes(leaf, fi); extent_num_bytes = new_size - -- cgit v1.2.3-18-g5258 From ec39e180fd3188c983c94603634bfcd019f42ae7 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 12 Jan 2012 19:10:12 -0500 Subject: Btrfs: release space on error in page_mkwrite If updating the inode gave us an ENOSPC we were just returning in page_mkwrite, which is a problem since we make our reservation right before trying to update the inode, so fix the out label so that we actually free our reservation. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 946a7f1b329..85fd86ea983 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6488,8 +6488,8 @@ out_unlock: if (!ret) return VM_FAULT_LOCKED; unlock_page(page); - btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE); out: + btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE); return ret; } -- cgit v1.2.3-18-g5258 From 90290e19820e3323ce6b9c2888eeb68bf29c278b Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 2 Dec 2011 15:44:12 -0500 Subject: Btrfs: protect orphan block rsv with spin_lock We've been seeing warnings coming out of the orphan commit stuff forever from ceph. Turns out it's because we're racing with checking if the orphan block reserve is set, because we clear it outside of the spin_lock. So leave the normal fastpath checks where they are, but take the spin_lock and _recheck_ to make sure we haven't had an orphan block rsv added in the meantime. Then clear the root's orphan block rsv and release the lock. With this patch a user said the warnings went away and they usually showed up pretty soon after he started ceph. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 85fd86ea983..619742d3716 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1951,12 +1951,28 @@ enum btrfs_orphan_cleanup_state { void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, struct btrfs_root *root) { + struct btrfs_block_rsv *block_rsv; int ret; if (!list_empty(&root->orphan_list) || root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE) return; + spin_lock(&root->orphan_lock); + if (!list_empty(&root->orphan_list)) { + spin_unlock(&root->orphan_lock); + return; + } + + if (root->orphan_cleanup_state != ORPHAN_CLEANUP_DONE) { + spin_unlock(&root->orphan_lock); + return; + } + + block_rsv = root->orphan_block_rsv; + root->orphan_block_rsv = NULL; + spin_unlock(&root->orphan_lock); + if (root->orphan_item_inserted && btrfs_root_refs(&root->root_item) > 0) { ret = btrfs_del_orphan_item(trans, root->fs_info->tree_root, @@ -1965,10 +1981,9 @@ void btrfs_orphan_commit_root(struct btrfs_trans_handle *trans, root->orphan_item_inserted = 0; } - if (root->orphan_block_rsv) { - WARN_ON(root->orphan_block_rsv->size > 0); - btrfs_free_block_rsv(root, root->orphan_block_rsv); - root->orphan_block_rsv = NULL; + if (block_rsv) { + WARN_ON(block_rsv->size > 0); + btrfs_free_block_rsv(root, block_rsv); } } -- cgit v1.2.3-18-g5258 From f248679e86fead40cc78e724c7181d6bec1a2046 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 13 Jan 2012 12:09:22 -0500 Subject: Btrfs: add a delalloc mutex to inodes for delalloc reservations I was using i_mutex for this, but we're getting bogus lockdep warnings by doing that and theres no real way to get rid of those, so just stop using i_mutex to protect delalloc metadata reservations and use a delalloc mutex instead. This shouldn't be contended often at all, only if you are writing and mmap writing to the file at the same time. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 619742d3716..5977987abdb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2239,14 +2239,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) continue; } nr_truncate++; - /* - * Need to hold the imutex for reservation purposes, not - * a huge deal here but I have a WARN_ON in - * btrfs_delalloc_reserve_space to catch offenders. - */ - mutex_lock(&inode->i_mutex); ret = btrfs_truncate(inode); - mutex_unlock(&inode->i_mutex); } else { nr_unlink++; } @@ -6411,10 +6404,7 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) u64 page_start; u64 page_end; - /* Need this to keep space reservations serialized */ - mutex_lock(&inode->i_mutex); ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); - mutex_unlock(&inode->i_mutex); if (!ret) ret = btrfs_update_time(vma->vm_file); if (ret) { @@ -6758,6 +6748,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) extent_io_tree_init(&ei->io_tree, &inode->i_data); extent_io_tree_init(&ei->io_failure_tree, &inode->i_data); mutex_init(&ei->log_mutex); + mutex_init(&ei->delalloc_mutex); btrfs_ordered_inode_tree_init(&ei->ordered_tree); INIT_LIST_HEAD(&ei->i_orphan); INIT_LIST_HEAD(&ei->delalloc_inodes); -- cgit v1.2.3-18-g5258 From 9998eb703490589c3e8f1bf09b15203156776edb Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Wed, 25 Jan 2012 13:47:40 -0500 Subject: Btrfs: fix reservations in btrfs_page_mkwrite Josef fixed btrfs_page_mkwrite to properly release reserved extents if there was an error. But if we fail to get a reservation and we fail to dirty the inode (for ENOSPC reasons), we'll end up trying to release a reservation we never had. This makes sure we only release if we were able to reserve. Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5977987abdb..7405753ec5d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6401,18 +6401,23 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) unsigned long zero_start; loff_t size; int ret; + int reserved = 0; u64 page_start; u64 page_end; ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); - if (!ret) + if (!ret) { ret = btrfs_update_time(vma->vm_file); + reserved = 1; + } if (ret) { if (ret == -ENOMEM) ret = VM_FAULT_OOM; else /* -ENOSPC, -EIO, etc */ ret = VM_FAULT_SIGBUS; - goto out; + if (reserved) + goto out; + goto out_noreserve; } ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */ @@ -6495,6 +6500,7 @@ out_unlock: unlock_page(page); out: btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE); +out_noreserve: return ret; } -- cgit v1.2.3-18-g5258 From 87826df0ec36fc28884b4ddbb3f3af41c4c2008f Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Wed, 15 Feb 2012 16:23:57 +0100 Subject: btrfs: delalloc for page dirtied out-of-band in fixup worker We encountered an issue that was easily observable on s/390 systems but could really happen anywhere. The timing just seemed to hit reliably on s/390 with limited memory. The gist is that when an unexpected set_page_dirty() happened, we'd run into the BUG() in btrfs_writepage_fixup_worker since it wasn't properly set up for delalloc. This patch does the following: - Performs the missing delalloc in the fixup worker - Allow the start hook to return -EBUSY which informs __extent_writepage that it should mark the page skipped and not to redirty it. This is required since the fixup worker can fail with -ENOSPC and the page will have already been redirtied. That causes an Oops in drop_outstanding_extents later. Retrying the fixup worker could lead to an infinite loop. Deferring the page redirty also saves us some cycles since the page would be stuck in a resubmit-redirty loop until the fixup worker completes. It's not harmful, just wasteful. - If the fixup worker fails, we mark the page and mapping as errored, and end the writeback, similar to what we would do had the page actually been submitted to writeback. Signed-off-by: Jeff Mahoney --- fs/btrfs/inode.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7405753ec5d..bf392e53261 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1555,6 +1555,7 @@ static void btrfs_writepage_fixup_worker(struct btrfs_work *work) struct inode *inode; u64 page_start; u64 page_end; + int ret; fixup = container_of(work, struct btrfs_writepage_fixup, work); page = fixup->page; @@ -1582,12 +1583,21 @@ again: page_end, &cached_state, GFP_NOFS); unlock_page(page); btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); goto again; } - BUG(); + ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); + if (ret) { + mapping_set_error(page->mapping, ret); + end_extent_writepage(page, ret, page_start, page_end); + ClearPageChecked(page); + goto out; + } + btrfs_set_extent_delalloc(inode, page_start, page_end, &cached_state); ClearPageChecked(page); + set_page_dirty(page); out: unlock_extent_cached(&BTRFS_I(inode)->io_tree, page_start, page_end, &cached_state, GFP_NOFS); @@ -1630,7 +1640,7 @@ static int btrfs_writepage_start_hook(struct page *page, u64 start, u64 end) fixup->work.func = btrfs_writepage_fixup_worker; fixup->page = page; btrfs_queue_worker(&root->fs_info->fixup_workers, &fixup->work); - return -EAGAIN; + return -EBUSY; } static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, -- cgit v1.2.3-18-g5258 From 12fc9d0923ca70ae8960bccebac09d5c12f8c4d4 Mon Sep 17 00:00:00 2001 From: Florian Albrechtskirchinger Date: Fri, 10 Feb 2012 22:15:54 +0100 Subject: btrfs: honor umask when creating subvol root Set the subvol root inode permissions based on the current umask. --- fs/btrfs/inode.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index bf392e53261..6e0ee9b0d74 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6706,8 +6706,10 @@ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, int err; u64 index = 0; - inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, new_dirid, - new_dirid, S_IFDIR | 0700, &index); + inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, + new_dirid, new_dirid, + S_IFDIR | (~current_umask() & S_IRWXUGO), + &index); if (IS_ERR(inode)) return PTR_ERR(inode); inode->i_op = &btrfs_dir_inode_operations; -- cgit v1.2.3-18-g5258 From fe66a05a06795bd3b788404d69ea7709f46a1609 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Mon, 20 Feb 2012 08:40:56 -0500 Subject: Btrfs: improve error handling for btrfs_insert_dir_item callers This allows us to gracefully continue if we aren't able to insert directory items, both for normal files/dirs and snapshots. Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6e0ee9b0d74..cbeb2e36ceb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4585,7 +4585,8 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, ret = btrfs_insert_dir_item(trans, root, name, name_len, parent_inode, &key, btrfs_inode_type(inode), index); - BUG_ON(ret); + if (ret) + goto fail_dir_item; btrfs_i_size_write(parent_inode, parent_inode->i_size + name_len * 2); @@ -4593,6 +4594,23 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, ret = btrfs_update_inode(trans, root, parent_inode); } return ret; + +fail_dir_item: + if (unlikely(ino == BTRFS_FIRST_FREE_OBJECTID)) { + u64 local_index; + int err; + err = btrfs_del_root_ref(trans, root->fs_info->tree_root, + key.objectid, root->root_key.objectid, + parent_ino, &local_index, name, name_len); + + } else if (add_backref) { + u64 local_index; + int err; + + err = btrfs_del_inode_ref(trans, root, name, name_len, + ino, parent_ino, &local_index); + } + return ret; } static int btrfs_add_nondir(struct btrfs_trans_handle *trans, -- cgit v1.2.3-18-g5258