linux-toradex.git/fs/btrfs/extent-tree.c, branch v4.9.105

btrfs: Fix possible softlock on single core machines

2018-05-30T05:50:41+00:00

[ Upstream commit 1e1c50a929bc9e49bc3f9935b92450d9e69f8158 ]

do_chunk_alloc implements a loop checking whether there is a pending
chunk allocation and if so causes the caller do loop. Generally this
loop is executed only once, however testing with btrfs/072 on a single
core vm machines uncovered an extreme case where the system could loop
indefinitely. This is due to a missing cond_resched when loop which
doesn't give a chance to the previous chunk allocator finish its job.

The fix is to simply add the missing cond_resched.

Fixes: 6d74119f1a3e ("Btrfs: avoid taking the chunk_mutex in do_chunk_alloc")
Signed-off-by: Nikolay Borisov 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix missing error return in btrfs_drop_snapshot

2017-12-14T08:28:12+00:00

commit e19182c0fff451e3744c1107d98f072e7ca377a0 upstream.

If btrfs_del_root fails in btrfs_drop_snapshot, we'll pick up the
error but then return 0 anyway due to mixing err and ret.

Fixes: 79787eaab4612 ("btrfs: replace many BUG_ONs with proper error handling")
Signed-off-by: Jeff Mahoney 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: clear space cache inode generation always

2017-12-05T10:24:32+00:00

commit 8e138e0d92c6c9d3d481674fb14e3439b495be37 upstream.

We discovered a box that had double allocations, and suspected the space
cache may be to blame.  While auditing the write out path I noticed that
if we've already setup the space cache we will just carry on.  This
means that any error we hit after cache_save_setup before we go to
actually write the cache out we won't reset the inode generation, so
whatever was already written will be considered correct, except it'll be
stale.  Fix this by _always_ resetting the generation on the block group
inode, this way we only ever have valid or invalid cache.

With this patch I was no longer able to reproduce cache corruption with
dm-log-writes and my bpf error injection tool.

Signed-off-by: Josef Bacik 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

Btrfs: fix early ENOSPC due to delalloc

2017-08-11T15:49:31+00:00

commit 17024ad0a0fdfcfe53043afb969b813d3e020c21 upstream.

If a lot of metadata is reserved for outstanding delayed allocations, we
rely on shrink_delalloc() to reclaim metadata space in order to fulfill
reservation tickets. However, shrink_delalloc() has a shortcut where if
it determines that space can be overcommitted, it will stop early. This
made sense before the ticketed enospc system, but now it means that
shrink_delalloc() will often not reclaim enough space to fulfill any
tickets, leading to an early ENOSPC. (Reservation tickets don't care
about being able to overcommit, they need every byte accounted for.)

Fix it by getting rid of the shortcut so that shrink_delalloc() reclaims
all of the metadata it is supposed to. This fixes early ENOSPCs we were
seeing when doing a btrfs receive to populate a new filesystem, as well
as early ENOSPCs Christoph saw when doing a big cp -r onto Btrfs.

Fixes: 957780eb2788 ("Btrfs: introduce ticketed enospc infrastructure")
Tested-by: Christoph Anton Mitterer 
Reviewed-by: Josef Bacik 
Signed-off-by: Omar Sandoval 
Signed-off-by: David Sterba 
Signed-off-by: Nikolay Borisov 
Signed-off-by: Greg Kroah-Hartman

Btrfs: use down_read_nested to make lockdep silent

2017-08-07T01:59:47+00:00

[ Upstream commit e321f8a801d7b4c40da8005257b05b9c2b51b072 ]

If @block_group is not @used_bg, it'll try to get @used_bg's lock without
droping @block_group 's lock and lockdep has throwed a scary deadlock warning
about it.
Fix it by using down_read_nested.

Signed-off-by: Liu Bo 
Reviewed-by: David Sterba 
Signed-off-by: David Sterba 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix memory leak in update_space_info failure path

2017-06-14T13:06:02+00:00

commit 896533a7da929136d0432713f02a3edffece2826 upstream.

If we fail to add the space_info kobject, we'll leak the memory
for the percpu counter.

Fixes: 6ab0a2029c (btrfs: publish allocation data in sysfs)
Signed-off-by: Jeff Mahoney 
Reviewed-by: Liu Bo 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix error handling when run_delayed_extent_op fails

2017-01-19T19:18:05+00:00

commit aa7c8da35d1905d80e840d075f07d26ec90144b5 upstream.

In __btrfs_run_delayed_refs, the error path when run_delayed_extent_op
fails sets locked_ref->processing = 0 but doesn't re-increment
delayed_refs->num_heads_ready.  As a result, we end up triggering
the WARN_ON in btrfs_select_ref_head.

Fixes: d7df2c796d7 (Btrfs: attach delayed ref updates to delayed ref heads)
Reported-by: Jon Nelson 
Signed-off-by: Jeff Mahoney 
Reviewed-by: Liu Bo 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix locking when we put back a delayed ref that's too new

2017-01-19T19:18:05+00:00

commit d0280996437081dd12ed1e982ac8aeaa62835ec4 upstream.

In __btrfs_run_delayed_refs, when we put back a delayed ref that's too
new, we have already dropped the lock on locked_ref when we set
->processing = 0.

This patch keeps the lock to cover that assignment.

Fixes: d7df2c796d7 (Btrfs: attach delayed ref updates to delayed ref heads)
Signed-off-by: Jeff Mahoney 
Reviewed-by: Liu Bo 
Signed-off-by: David Sterba 
Signed-off-by: Greg Kroah-Hartman

btrfs: fix WARNING in btrfs_select_ref_head()

2016-10-24T16:20:29+00:00

This issue was found when testing in-band dedupe enospc behaviour,
sometimes run_one_delayed_ref() may fail for enospc reason, then
__btrfs_run_delayed_refs（）will return, but forget to add num_heads_read
back, which will trigger "WARN_ON(delayed_refs->num_heads_ready == 0)" in
btrfs_select_ref_head().

Signed-off-by: Wang Xiaoguang 
Signed-off-by: David Sterba

Revert "btrfs: let btrfs_delete_unused_bgs() to clean relocated bgs"

2016-10-10T20:43:31+00:00

This reverts commit 5d8eb6fe517583f9c6d5b94faf2254a0207a45c9.

When we remove devices, we free the device structures.  Delaying
btfs_remove_chunk() ends up hitting a use-after-free on them.

Signed-off-by: Chris Mason