From c3473e830074ef04f974f2829690942dd8580619 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Tue, 19 Jun 2012 10:59:00 -0400 Subject: Btrfs: fix dio write vs buffered read race Miao pointed out there's a problem with mixing dio writes and buffered reads. If the read happens between us invalidating the page range and actually locking the extent we can bring in pages into page cache. Then once the write finishes if somebody tries to read again it will just find uptodate pages and we'll read stale data. So we need to lock the extent and check for uptodate bits in the range. If there are uptodate bits we need to unlock and invalidate again. This will keep this race from happening since we will hold the extent locked until we create the ordered extent, and then teh read side always waits for ordered extents. There was also a race in how we updated i_size, previously we were relying on the generic DIO stuff to adjust the i_size after the DIO had completed, but this happens outside of the extent lock which means reads could come in and not see the updated i_size. So instead move this work into where we create the extents, and then this way the update ordered i_size stuff works properly in the endio handlers. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 50 insertions(+), 5 deletions(-) (limited to 'fs/btrfs/inode.c') diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 4a4f2d59a64b..6971fb5fc859 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5904,8 +5904,17 @@ map: bh_result->b_size = len; bh_result->b_bdev = em->bdev; set_buffer_mapped(bh_result); - if (create && !test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) - set_buffer_new(bh_result); + if (create) { + if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) + set_buffer_new(bh_result); + + /* + * Need to update the i_size under the extent lock so buffered + * readers will get the updated i_size when we unlock. + */ + if (start + len > i_size_read(inode)) + i_size_write(inode, start + len); + } free_extent_map(em); @@ -6388,12 +6397,48 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, */ ordered = btrfs_lookup_ordered_range(inode, lockstart, lockend - lockstart + 1); - if (!ordered) + + /* + * We need to make sure there are no buffered pages in this + * range either, we could have raced between the invalidate in + * generic_file_direct_write and locking the extent. The + * invalidate needs to happen so that reads after a write do not + * get stale data. + */ + if (!ordered && (!writing || + !test_range_bit(&BTRFS_I(inode)->io_tree, + lockstart, lockend, EXTENT_UPTODATE, 0, + cached_state))) break; + unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, lockend, &cached_state, GFP_NOFS); - btrfs_start_ordered_extent(inode, ordered, 1); - btrfs_put_ordered_extent(ordered); + + if (ordered) { + btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); + } else { + /* Screw you mmap */ + ret = filemap_write_and_wait_range(file->f_mapping, + lockstart, + lockend); + if (ret) + goto out; + + /* + * If we found a page that couldn't be invalidated just + * fall back to buffered. + */ + ret = invalidate_inode_pages2_range(file->f_mapping, + lockstart >> PAGE_CACHE_SHIFT, + lockend >> PAGE_CACHE_SHIFT); + if (ret) { + if (ret == -EBUSY) + ret = 0; + goto out; + } + } + cond_resched(); } -- cgit v1.2.3 From 6bf02314d9a5c29f6ec30285b9ad5361c2d4c85a Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 25 Jun 2012 21:59:09 -0600 Subject: Btrfs: fix wrong check during log recovery When we're evicting an inode during log recovery, we need to ensure that the inode is not in orphan state any more, which means inode's run_time flags has _no_ BTRFS_INODE_HAS_ORPHAN_ITEM. Thus, the BUG_ON was triggered because of a wrong check for the flags. Reviewed-by: David Sterba Signed-off-by: Liu Bo 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 6971fb5fc859..9f07bd121f66 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3754,7 +3754,7 @@ void btrfs_evict_inode(struct inode *inode) btrfs_wait_ordered_range(inode, 0, (u64)-1); if (root->fs_info->log_root_recovering) { - BUG_ON(!test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, + BUG_ON(test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM, &BTRFS_I(inode)->runtime_flags)); goto no_delete; } -- cgit v1.2.3 From b3d9b7a3c752dc4b6976a4ff7b8298887a5b734d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 9 Jun 2012 13:51:19 -0400 Subject: vfs: switch i_dentry/d_alias to hlist 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 a7d1921ac76b..a101572f1cea 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6987,7 +6987,7 @@ void btrfs_destroy_inode(struct inode *inode) struct btrfs_ordered_extent *ordered; struct btrfs_root *root = BTRFS_I(inode)->root; - WARN_ON(!list_empty(&inode->i_dentry)); + WARN_ON(!hlist_empty(&inode->i_dentry)); WARN_ON(inode->i_data.nrpages); WARN_ON(BTRFS_I(inode)->outstanding_extents); WARN_ON(BTRFS_I(inode)->reserved_extents); -- cgit v1.2.3 From 00cd8dd3bf95f2cc8435b4cac01d9995635c6d0b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 10 Jun 2012 17:13:09 -0400 Subject: stop passing nameidata to ->lookup() Just the flags; only NFS cares even about that, but there are legitimate uses for such argument. And getting rid of that completely would require splitting ->lookup() into a couple of methods (at least), so let's leave that alone for now... 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 a101572f1cea..e5f1f81b2d65 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4247,7 +4247,7 @@ static void btrfs_dentry_release(struct dentry *dentry) } static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct dentry *ret; -- cgit v1.2.3 From ebfc3b49a7ac25920cb5be5445f602e51d2ea559 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 10 Jun 2012 18:05:36 -0400 Subject: don't pass nameidata to ->create() boolean "does it have to be exclusive?" flag is passed instead; Local filesystem should just ignore it - the object is guaranteed not to be there yet. 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 e5f1f81b2d65..fb8d671d00e6 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4893,7 +4893,7 @@ out_unlock: } static int btrfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, struct nameidata *nd) + umode_t mode, bool excl) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(dir)->root; -- cgit v1.2.3