We store the progress (root and inode numbers) of the extent map shrinker
in fs_info without any synchronization but we can have multiple tasks
calling into the shrinker during memory allocations when there's enough
memory pressure for example.
This can result in a task A reading fs_info->extent_map_shrinker_last_ino
after another task B updates it, and task A reading
fs_info->extent_map_shrinker_last_root before task B updates it, making
task A see an odd state that isn't necessarily harmful but may make it
skip certain inode ranges or do more work than necessary by going over
the same inodes again. These unprotected accesses would also trigger
warnings from tools like KCSAN.
So add a lock to protect access to these progress fields.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The extent map shrinker can be called in a variety of contexts where we
are under memory pressure, and of them is when a task is trying to
allocate memory. For this reason the shrinker is typically called with a
value of struct shrink_control::nr_to_scan that is much smaller than what
we return in the nr_cached_objects callback of struct super_operations
(fs/btrfs/super.c:btrfs_nr_cached_objects()), so that the shrinker does
not take a long time and cause high latencies. However we can still take
a lot of time in the shrinker even for a limited amount of nr_to_scan:
1) When traversing the red black tree that tracks open inodes in a root,
as for example with millions of open inodes we get a deep tree which
takes time searching for an inode;
2) Iterating over the extent map tree, which is a red black tree, of an
inode when doing the rb_next() calls and when removing an extent map
from the tree, since often that requires rebalancing the red black
tree;
3) When trying to write lock an inode's extent map tree we may wait for a
significant amount of time, because there's either another task about
to do IO and searching for an extent map in the tree or inserting an
extent map in the tree, and we can have thousands or even millions of
extent maps for an inode. Furthermore, there can be concurrent calls
to the shrinker so the lock might be busy simply because there is
already another task shrinking extent maps for the same inode;
4) We often reschedule if we need to, which further increases latency.
So improve on this by stopping the extent map shrinking code whenever we
need to reschedule and make it skip an inode if we can't immediately lock
its extent map tree.
Reported-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Reported-by: Andrea Gelmini <andrea.gelmini@gmail.com>
Link: https://lore.kernel.org/linux-btrfs/CABXGCsMmmb36ym8hVNGTiU8yfUS_cGvoUmGCcBrGWq9OxTrs+A@mail.gmail.com/
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Another improper use of __folio_put() in an error path after freshly
allocating pages/folios which returns them with the refcount initialized
to 1. The refactor from __free_pages() -> __folio_put() (instead of
folio_put) removed a refcount decrement found in __free_pages() and
folio_put but absent from __folio_put().
Fixes: 13df3775ef ("btrfs: cleanup metadata page pointer usage")
CC: stable@vger.kernel.org # 6.8+
Tested-by: Ed Tomlinson <edtoml@gmail.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The conversion to folios switched __free_page() to __folio_put() in the
error path in btrfs_do_encoded_write().
However, this gets the page refcounting wrong. If we do hit that error
path (I reproduced by modifying btrfs_do_encoded_write to pretend to
always fail in a way that jumps to out_folios and running the fstests
case btrfs/281), then we always hit the following BUG freeing the folio:
BUG: Bad page state in process btrfs pfn:40ab0b
page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x61be5 pfn:0x40ab0b
flags: 0x5ffff0000000000(node=0|zone=2|lastcpupid=0x1ffff)
raw: 05ffff0000000000 0000000000000000 dead000000000122 0000000000000000
raw: 0000000000061be5 0000000000000000 00000001ffffffff 0000000000000000
page dumped because: nonzero _refcount
Call Trace:
<TASK>
dump_stack_lvl+0x3d/0xe0
bad_page+0xea/0xf0
free_unref_page+0x8e1/0x900
? __mem_cgroup_uncharge+0x69/0x90
__folio_put+0xe6/0x190
btrfs_do_encoded_write+0x445/0x780
? current_time+0x25/0xd0
btrfs_do_write_iter+0x2cc/0x4b0
btrfs_ioctl_encoded_write+0x2b6/0x340
It turns out __free_page() decreases the page reference count while
__folio_put() does not. Switch __folio_put() to folio_put() which
decreases the folio reference count first.
Fixes: 400b172b8c ("btrfs: compression: migrate compression/decompression paths to folios")
Tested-by: Ed Tomlinson <edtoml@gmail.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
In the ref-verify tool, when processing the inline references of an extent
item, we may end up returning with uninitialized return value, because:
1) The 'ret' variable is not initialized if there are no inline extent
references ('ptr' == 'end' before the while loop starts);
2) If we find an extent owner inline reference we don't initialize 'ret'.
So fix these cases by initializing 'ret' to 0 when declaring the variable
and set it to -EINVAL if we find an extent owner inline references and
simple quotas are not enabled (as well as print an error message).
Reported-by: Mirsad Todorovac <mtodorovac69@gmail.com>
Link: https://lore.kernel.org/linux-btrfs/59b40ebe-c824-457d-8b24-0bbca69d472b@gmail.com/
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Syzbot reports the following regression detected by KASAN:
BUG: KASAN: slab-out-of-bounds in btrfs_qgroup_inherit+0x42e/0x2e20 fs/btrfs/qgroup.c:3277
Read of size 8 at addr ffff88814628ca50 by task syz-executor318/5171
CPU: 0 PID: 5171 Comm: syz-executor318 Not tainted 6.10.0-rc2-syzkaller-00010-g2ab795141095 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/02/2024
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
print_address_description mm/kasan/report.c:377 [inline]
print_report+0x169/0x550 mm/kasan/report.c:488
kasan_report+0x143/0x180 mm/kasan/report.c:601
btrfs_qgroup_inherit+0x42e/0x2e20 fs/btrfs/qgroup.c:3277
create_pending_snapshot+0x1359/0x29b0 fs/btrfs/transaction.c:1854
create_pending_snapshots+0x195/0x1d0 fs/btrfs/transaction.c:1922
btrfs_commit_transaction+0xf20/0x3740 fs/btrfs/transaction.c:2382
create_snapshot+0x6a1/0x9e0 fs/btrfs/ioctl.c:875
btrfs_mksubvol+0x58f/0x710 fs/btrfs/ioctl.c:1029
btrfs_mksnapshot+0xb5/0xf0 fs/btrfs/ioctl.c:1075
__btrfs_ioctl_snap_create+0x387/0x4b0 fs/btrfs/ioctl.c:1340
btrfs_ioctl_snap_create_v2+0x1f2/0x3a0 fs/btrfs/ioctl.c:1422
btrfs_ioctl+0x99e/0xc60
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:907 [inline]
__se_sys_ioctl+0xfc/0x170 fs/ioctl.c:893
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7fcbf1992509
RSP: 002b:00007fcbf1928218 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 00007fcbf1a1f618 RCX: 00007fcbf1992509
RDX: 0000000020000280 RSI: 0000000050009417 RDI: 0000000000000003
RBP: 00007fcbf1a1f610 R08: 00007ffea1298e97 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007fcbf19eb660
R13: 00000000200002b8 R14: 00007fcbf19e60c0 R15: 0030656c69662f2e
</TASK>
And it also pinned it down to commit b5357cb268 ("btrfs: qgroup: do not
check qgroup inherit if qgroup is disabled").
[CAUSE]
That offending commit skips the whole qgroup inherit check if qgroup is
not enabled.
But that also skips the very basic checks like
num_ref_copies/num_excl_copies and the structure size checks.
Meaning if a qgroup enable/disable race is happening at the background,
and we pass a btrfs_qgroup_inherit structure when the qgroup is
disabled, the check would be completely skipped.
Then at the time of transaction commitment, qgroup is re-enabled and
btrfs_qgroup_inherit() is going to use the incorrect structure and
causing the above KASAN error.
[FIX]
Make btrfs_qgroup_check_inherit() only skip the source qgroup checks.
So that even if invalid btrfs_qgroup_inherit structure is passed in, we
can still reject invalid ones no matter if qgroup is enabled or not.
Furthermore we do already have an extra safety inside
btrfs_qgroup_inherit(), which would just ignore invalid qgroup sources,
so even if we only skip the qgroup source check we're still safe.
Reported-by: syzbot+a0d1f7e26910be4dc171@syzkaller.appspotmail.com
Fixes: b5357cb268 ("btrfs: qgroup: do not check qgroup inherit if qgroup is disabled")
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Jeongjun Park <aha310510@gmail.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
calc_available_free_space() returns the total size of metadata (or
system) block groups, which can be allocated from unallocated disk
space. The logic is wrong on zoned mode in two places.
First, the calculation of data_chunk_size is wrong. We always allocate
one zone as one chunk, and no partial allocation of a zone. So, we
should use zone_size (= data_sinfo->chunk_size) as it is.
Second, the result "avail" may not be zone aligned. Since we always
allocate one zone as one chunk on zoned mode, returning non-zone size
aligned bytes will result in less pressure on the async metadata reclaim
process.
This is serious for the nearly full state with a large zone size device.
Allowing over-commit too much will result in less async reclaim work and
end up in ENOSPC. We can align down to the zone size to avoid that.
Fixes: cb6cbab790 ("btrfs: adjust overcommit logic when very close to full")
CC: stable@vger.kernel.org # 6.9
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: David Sterba <dsterba@suse.com>
There is a potential parallel list adding for retrying in
btrfs_reclaim_bgs_work and adding to the unused list. Since the block
group is removed from the reclaim list and it is on a relocation work,
it can be added into the unused list in parallel. When that happens,
adding it to the reclaim list will corrupt the list head and trigger
list corruption like below.
Fix it by taking fs_info->unused_bgs_lock.
[177.504][T2585409] BTRFS error (device nullb1): error relocating ch= unk 2415919104
[177.514][T2585409] list_del corruption. next->prev should be ff1100= 0344b119c0, but was ff11000377e87c70. (next=3Dff110002390cd9c0)
[177.529][T2585409] ------------[ cut here ]------------
[177.537][T2585409] kernel BUG at lib/list_debug.c:65!
[177.545][T2585409] Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN NOPTI
[177.555][T2585409] CPU: 9 PID: 2585409 Comm: kworker/u128:2 Tainted: G W 6.10.0-rc5-kts #1
[177.568][T2585409] Hardware name: Supermicro SYS-520P-WTR/X12SPW-TF, BIOS 1.2 02/14/2022
[177.579][T2585409] Workqueue: events_unbound btrfs_reclaim_bgs_work[btrfs]
[177.589][T2585409] RIP: 0010:__list_del_entry_valid_or_report.cold+0x70/0x72
[177.624][T2585409] RSP: 0018:ff11000377e87a70 EFLAGS: 00010286
[177.633][T2585409] RAX: 000000000000006d RBX: ff11000344b119c0 RCX:0000000000000000
[177.644][T2585409] RDX: 000000000000006d RSI: 0000000000000008 RDI:ffe21c006efd0f40
[177.655][T2585409] RBP: ff110002e0509f78 R08: 0000000000000001 R09:ffe21c006efd0f08
[177.665][T2585409] R10: ff11000377e87847 R11: 0000000000000000 R12:ff110002390cd9c0
[177.676][T2585409] R13: ff11000344b119c0 R14: ff110002e0508000 R15:dffffc0000000000
[177.687][T2585409] FS: 0000000000000000(0000) GS:ff11000fec880000(0000) knlGS:0000000000000000
[177.700][T2585409] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[177.709][T2585409] CR2: 00007f06bc7b1978 CR3: 0000001021e86005 CR4:0000000000771ef0
[177.720][T2585409] DR0: 0000000000000000 DR1: 0000000000000000 DR2:0000000000000000
[177.731][T2585409] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:0000000000000400
[177.742][T2585409] PKRU: 55555554
[177.748][T2585409] Call Trace:
[177.753][T2585409] <TASK>
[177.759][T2585409] ? __die_body.cold+0x19/0x27
[177.766][T2585409] ? die+0x2e/0x50
[177.772][T2585409] ? do_trap+0x1ea/0x2d0
[177.779][T2585409] ? __list_del_entry_valid_or_report.cold+0x70/0x72
[177.788][T2585409] ? do_error_trap+0xa3/0x160
[177.795][T2585409] ? __list_del_entry_valid_or_report.cold+0x70/0x72
[177.805][T2585409] ? handle_invalid_op+0x2c/0x40
[177.812][T2585409] ? __list_del_entry_valid_or_report.cold+0x70/0x72
[177.820][T2585409] ? exc_invalid_op+0x2d/0x40
[177.827][T2585409] ? asm_exc_invalid_op+0x1a/0x20
[177.834][T2585409] ? __list_del_entry_valid_or_report.cold+0x70/0x72
[177.843][T2585409] btrfs_delete_unused_bgs+0x3d9/0x14c0 [btrfs]
There is a similar retry_list code in btrfs_delete_unused_bgs(), but it is
safe, AFAICS. Since the block group was in the unused list, the used bytes
should be 0 when it was added to the unused list. Then, it checks
block_group->{used,reserved,pinned} are still 0 under the
block_group->lock. So, they should be still eligible for the unused list,
not the reclaim list.
The reason it is safe there it's because because we're holding
space_info->groups_sem in write mode.
That means no other task can allocate from the block group, so while we
are at deleted_unused_bgs() it's not possible for other tasks to
allocate and deallocate extents from the block group, so it can't be
added to the unused list or the reclaim list by anyone else.
The bug can be reproduced by btrfs/166 after a few rounds. In practice
this can be hit when relocation cannot find more chunk space and ends
with ENOSPC.
Reported-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Suggested-by: Johannes Thumshirn <Johannes.Thumshirn@wdc.com>
Fixes: 4eb4e85c4f ("btrfs: retry block group reclaim without infinite loop")
CC: stable@vger.kernel.org # 5.15+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If during the quota disable we fail when cleaning the quota tree or when
deleting the root from the root tree, we jump to the 'out' label without
ever dropping the reference on the quota root, resulting in a leak of the
root since fs_info->quota_root is no longer pointing to the root (we have
set it to NULL just before those steps).
Fix this by always doing a btrfs_put_root() call under the 'out' label.
This is a problem that exists since qgroups were first added in 2012 by
commit bed92eae26 ("Btrfs: qgroup implementation and prototypes"), but
back then we missed a kfree on the quota root and free_extent_buffer()
calls on its root and commit root nodes, since back then roots were not
yet reference counted.
Reviewed-by: Boris Burkov <boris@bur.io>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
When running btrfs/060 with forced RST feature, it would crash the
following ASSERT() inside scrub_read_endio():
ASSERT(sector_nr < stripe->nr_sectors);
Before that, we would have tree dump from
btrfs_get_raid_extent_offset(), as we failed to find the RST entry for
the range.
[CAUSE]
Inside scrub_submit_extent_sector_read() every time we allocated a new
bbio we immediately called btrfs_map_block() to make sure there was some
RST range covering the scrub target.
But if btrfs_map_block() fails, we immediately call endio for the bbio,
while the bbio is newly allocated, it's completely empty.
Then inside scrub_read_endio(), we go through the bvecs to find
the sector number (as bi_sector is no longer reliable if the bio is
submitted to lower layers).
And since the bio is empty, such bvecs iteration would not find any
sector matching the sector, and return sector_nr == stripe->nr_sectors,
triggering the ASSERT().
[FIX]
Instead of calling btrfs_map_block() after allocating a new bbio, call
btrfs_map_block() first.
Since our only objective of calling btrfs_map_block() is only to update
stripe_len, there is really no need to do that after btrfs_alloc_bio().
This new timing would avoid the problem of handling empty bbio
completely, and in fact fixes a possible race window for the old code,
where if the submission thread is the only owner of the pending_io, the
scrub would never finish (since we didn't decrease the pending_io
counter).
Although the root cause of RST lookup failure still needs to be
addressed.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When creating a new block group, it calls btrfs_add_new_free_space() to add
the entire block group range into the free space accounting.
__btrfs_add_free_space_zoned() checks if size == block_group->length to
detect the initial free space adding, and proceed that case properly.
However, if the zone_capacity == zone_size and the over-write speed is fast
enough, the entire zone can be over-written within one transaction. That
confuses __btrfs_add_free_space_zoned() to handle it as an initial free
space accounting. As a result, that block group becomes a strange state: 0
used bytes, 0 zone_unusable bytes, but alloc_offset == zone_capacity (no
allocation anymore).
The initial free space accounting can properly be checked by checking
alloc_offset too.
Fixes: 98173255bd ("btrfs: zoned: calculate free space from zone capacity")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If inc_block_group_ro systematically fails (e.g. due to ETXTBUSY from
swap) or btrfs_relocate_chunk systematically fails (from lack of
space), then this worker becomes an infinite loop.
At the very least, this strands the cleaner thread, but can also result
in hung tasks/RCU stalls on PREEMPT_NONE kernels and if the
reclaim_bgs_lock mutex is not contended.
I believe the best long term fix is to manage reclaim via work queue,
where we queue up a relocation on the triggering condition and re-queue
on failure. In the meantime, this is an easy fix to apply to avoid the
immediate pain.
Fixes: 7e27180994 ("btrfs: reinsert BGs failed to reclaim")
CC: stable@vger.kernel.org # 6.6+
Signed-off-by: Boris Burkov <boris@bur.io>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
[BUG]
Since v6.8 there are rare kernel crashes reported by various people,
the common factor is bad page status error messages like this:
BUG: Bad page state in process kswapd0 pfn:d6e840
page: refcount:0 mapcount:0 mapping:000000007512f4f2 index:0x2796c2c7c
pfn:0xd6e840
aops:btree_aops ino:1
flags: 0x17ffffe0000008(uptodate|node=0|zone=2|lastcpupid=0x3fffff)
page_type: 0xffffffff()
raw: 0017ffffe0000008 dead000000000100 dead000000000122 ffff88826d0be4c0
raw: 00000002796c2c7c 0000000000000000 00000000ffffffff 0000000000000000
page dumped because: non-NULL mapping
[CAUSE]
Commit 09e6cef19c ("btrfs: refactor alloc_extent_buffer() to
allocate-then-attach method") changes the sequence when allocating a new
extent buffer.
Previously we always called grab_extent_buffer() under
mapping->i_private_lock, to ensure the safety on modification on
folio::private (which is a pointer to extent buffer for regular
sectorsize).
This can lead to the following race:
Thread A is trying to allocate an extent buffer at bytenr X, with 4
4K pages, meanwhile thread B is trying to release the page at X + 4K
(the second page of the extent buffer at X).
Thread A | Thread B
-----------------------------------+-------------------------------------
| btree_release_folio()
| | This is for the page at X + 4K,
| | Not page X.
| |
alloc_extent_buffer() | |- release_extent_buffer()
|- filemap_add_folio() for the | | |- atomic_dec_and_test(eb->refs)
| page at bytenr X (the first | | |
| page). | | |
| Which returned -EEXIST. | | |
| | | |
|- filemap_lock_folio() | | |
| Returned the first page locked. | | |
| | | |
|- grab_extent_buffer() | | |
| |- atomic_inc_not_zero() | | |
| | Returned false | | |
| |- folio_detach_private() | | |- folio_detach_private() for X
| |- folio_test_private() | | |- folio_test_private()
| Returned true | | | Returned true
|- folio_put() | |- folio_put()
Now there are two puts on the same folio at folio X, leading to refcount
underflow of the folio X, and eventually causing the BUG_ON() on the
page->mapping.
The condition is not that easy to hit:
- The release must be triggered for the middle page of an eb
If the release is on the same first page of an eb, page lock would kick
in and prevent the race.
- folio_detach_private() has a very small race window
It's only between folio_test_private() and folio_clear_private().
That's exactly when mapping->i_private_lock is used to prevent such race,
and commit 09e6cef19c ("btrfs: refactor alloc_extent_buffer() to
allocate-then-attach method") screwed that up.
At that time, I thought the page lock would kick in as
filemap_release_folio() also requires the page to be locked, but forgot
the filemap_release_folio() only locks one page, not all pages of an
extent buffer.
[FIX]
Move all the code requiring i_private_lock into
attach_eb_folio_to_filemap(), so that everything is done with proper
lock protection.
Furthermore to prevent future problems, add an extra
lockdep_assert_locked() to ensure we're holding the proper lock.
To reproducer that is able to hit the race (takes a few minutes with
instrumented code inserting delays to alloc_extent_buffer()):
#!/bin/sh
drop_caches () {
while(true); do
echo 3 > /proc/sys/vm/drop_caches
echo 1 > /proc/sys/vm/compact_memory
done
}
run_tar () {
while(true); do
for x in `seq 1 80` ; do
tar cf /dev/zero /mnt > /dev/null &
done
wait
done
}
mkfs.btrfs -f -d single -m single /dev/vda
mount -o noatime /dev/vda /mnt
# create 200,000 files, 1K each
./simoop -n 200000 -E -f 1k /mnt
drop_caches &
(run_tar)
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/linux-btrfs/CAHk-=wgt362nGfScVOOii8cgKn2LVVHeOvOA7OBwg1OwbuJQcw@mail.gmail.com/
Reported-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Link: https://lore.kernel.org/lkml/CABXGCsPktcHQOvKTbPaTwegMExije=Gpgci5NW=hqORo-s7diA@mail.gmail.com/
Reported-by: Toralf Förster <toralf.foerster@gmx.de>
Link: https://lore.kernel.org/linux-btrfs/e8b3311c-9a75-4903-907f-fc0f7a3fe423@gmx.de/
Reported-by: syzbot+f80b066392366b4af85e@syzkaller.appspotmail.com
Fixes: 09e6cef19c ("btrfs: refactor alloc_extent_buffer() to allocate-then-attach method")
CC: stable@vger.kernel.org # 6.8+
CC: Chris Mason <clm@fb.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Qgroup extent records are created when delayed ref heads are created and
then released after accounting extents at btrfs_qgroup_account_extents(),
called during the transaction commit path.
If a transaction is aborted we free the qgroup records by calling
btrfs_qgroup_destroy_extent_records() at btrfs_destroy_delayed_refs(),
unless we don't have delayed references. We are incorrectly assuming
that no delayed references means we don't have qgroup extents records.
We can currently have no delayed references because we ran them all
during a transaction commit and the transaction was aborted after that
due to some error in the commit path.
So fix this by ensuring we btrfs_qgroup_destroy_extent_records() at
btrfs_destroy_delayed_refs() even if we don't have any delayed references.
Reported-by: syzbot+0fecc032fa134afd49df@syzkaller.appspotmail.com
Link: https://lore.kernel.org/linux-btrfs/0000000000004e7f980619f91835@google.com/
Fixes: 81f7eb00ff ("btrfs: destroy qgroup extent records on transaction abort")
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have been seeing crashes on duplicate keys in
btrfs_set_item_key_safe():
BTRFS critical (device vdb): slot 4 key (450 108 8192) new key (450 108 8192)
------------[ cut here ]------------
kernel BUG at fs/btrfs/ctree.c:2620!
invalid opcode: 0000 [#1] PREEMPT SMP PTI
CPU: 0 PID: 3139 Comm: xfs_io Kdump: loaded Not tainted 6.9.0 #6
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-2.fc40 04/01/2014
RIP: 0010:btrfs_set_item_key_safe+0x11f/0x290 [btrfs]
With the following stack trace:
#0 btrfs_set_item_key_safe (fs/btrfs/ctree.c:2620:4)
#1 btrfs_drop_extents (fs/btrfs/file.c:411:4)
#2 log_one_extent (fs/btrfs/tree-log.c:4732:9)
#3 btrfs_log_changed_extents (fs/btrfs/tree-log.c:4955:9)
#4 btrfs_log_inode (fs/btrfs/tree-log.c:6626:9)
#5 btrfs_log_inode_parent (fs/btrfs/tree-log.c:7070:8)
#6 btrfs_log_dentry_safe (fs/btrfs/tree-log.c:7171:8)
#7 btrfs_sync_file (fs/btrfs/file.c:1933:8)
#8 vfs_fsync_range (fs/sync.c:188:9)
#9 vfs_fsync (fs/sync.c:202:9)
#10 do_fsync (fs/sync.c:212:9)
#11 __do_sys_fdatasync (fs/sync.c:225:9)
#12 __se_sys_fdatasync (fs/sync.c:223:1)
#13 __x64_sys_fdatasync (fs/sync.c:223:1)
#14 do_syscall_x64 (arch/x86/entry/common.c:52:14)
#15 do_syscall_64 (arch/x86/entry/common.c:83:7)
#16 entry_SYSCALL_64+0xaf/0x14c (arch/x86/entry/entry_64.S:121)
So we're logging a changed extent from fsync, which is splitting an
extent in the log tree. But this split part already exists in the tree,
triggering the BUG().
This is the state of the log tree at the time of the crash, dumped with
drgn (https://github.com/osandov/drgn/blob/main/contrib/btrfs_tree.py)
to get more details than btrfs_print_leaf() gives us:
>>> print_extent_buffer(prog.crashed_thread().stack_trace()[0]["eb"])
leaf 33439744 level 0 items 72 generation 9 owner 18446744073709551610
leaf 33439744 flags 0x100000000000000
fs uuid e5bd3946-400c-4223-8923-190ef1f18677
chunk uuid d58cb17e-6d02-494a-829a-18b7d8a399da
item 0 key (450 INODE_ITEM 0) itemoff 16123 itemsize 160
generation 7 transid 9 size 8192 nbytes 8473563889606862198
block group 0 mode 100600 links 1 uid 0 gid 0 rdev 0
sequence 204 flags 0x10(PREALLOC)
atime 1716417703.220000000 (2024-05-22 15:41:43)
ctime 1716417704.983333333 (2024-05-22 15:41:44)
mtime 1716417704.983333333 (2024-05-22 15:41:44)
otime 17592186044416.000000000 (559444-03-08 01:40:16)
item 1 key (450 INODE_REF 256) itemoff 16110 itemsize 13
index 195 namelen 3 name: 193
item 2 key (450 XATTR_ITEM 1640047104) itemoff 16073 itemsize 37
location key (0 UNKNOWN.0 0) type XATTR
transid 7 data_len 1 name_len 6
name: user.a
data a
item 3 key (450 EXTENT_DATA 0) itemoff 16020 itemsize 53
generation 9 type 1 (regular)
extent data disk byte 303144960 nr 12288
extent data offset 0 nr 4096 ram 12288
extent compression 0 (none)
item 4 key (450 EXTENT_DATA 4096) itemoff 15967 itemsize 53
generation 9 type 2 (prealloc)
prealloc data disk byte 303144960 nr 12288
prealloc data offset 4096 nr 8192
item 5 key (450 EXTENT_DATA 8192) itemoff 15914 itemsize 53
generation 9 type 2 (prealloc)
prealloc data disk byte 303144960 nr 12288
prealloc data offset 8192 nr 4096
...
So the real problem happened earlier: notice that items 4 (4k-12k) and 5
(8k-12k) overlap. Both are prealloc extents. Item 4 straddles i_size and
item 5 starts at i_size.
Here is the state of the filesystem tree at the time of the crash:
>>> root = prog.crashed_thread().stack_trace()[2]["inode"].root
>>> ret, nodes, slots = btrfs_search_slot(root, BtrfsKey(450, 0, 0))
>>> print_extent_buffer(nodes[0])
leaf 30425088 level 0 items 184 generation 9 owner 5
leaf 30425088 flags 0x100000000000000
fs uuid e5bd3946-400c-4223-8923-190ef1f18677
chunk uuid d58cb17e-6d02-494a-829a-18b7d8a399da
...
item 179 key (450 INODE_ITEM 0) itemoff 4907 itemsize 160
generation 7 transid 7 size 4096 nbytes 12288
block group 0 mode 100600 links 1 uid 0 gid 0 rdev 0
sequence 6 flags 0x10(PREALLOC)
atime 1716417703.220000000 (2024-05-22 15:41:43)
ctime 1716417703.220000000 (2024-05-22 15:41:43)
mtime 1716417703.220000000 (2024-05-22 15:41:43)
otime 1716417703.220000000 (2024-05-22 15:41:43)
item 180 key (450 INODE_REF 256) itemoff 4894 itemsize 13
index 195 namelen 3 name: 193
item 181 key (450 XATTR_ITEM 1640047104) itemoff 4857 itemsize 37
location key (0 UNKNOWN.0 0) type XATTR
transid 7 data_len 1 name_len 6
name: user.a
data a
item 182 key (450 EXTENT_DATA 0) itemoff 4804 itemsize 53
generation 9 type 1 (regular)
extent data disk byte 303144960 nr 12288
extent data offset 0 nr 8192 ram 12288
extent compression 0 (none)
item 183 key (450 EXTENT_DATA 8192) itemoff 4751 itemsize 53
generation 9 type 2 (prealloc)
prealloc data disk byte 303144960 nr 12288
prealloc data offset 8192 nr 4096
Item 5 in the log tree corresponds to item 183 in the filesystem tree,
but nothing matches item 4. Furthermore, item 183 is the last item in
the leaf.
btrfs_log_prealloc_extents() is responsible for logging prealloc extents
beyond i_size. It first truncates any previously logged prealloc extents
that start beyond i_size. Then, it walks the filesystem tree and copies
the prealloc extent items to the log tree.
If it hits the end of a leaf, then it calls btrfs_next_leaf(), which
unlocks the tree and does another search. However, while the filesystem
tree is unlocked, an ordered extent completion may modify the tree. In
particular, it may insert an extent item that overlaps with an extent
item that was already copied to the log tree.
This may manifest in several ways depending on the exact scenario,
including an EEXIST error that is silently translated to a full sync,
overlapping items in the log tree, or this crash. This particular crash
is triggered by the following sequence of events:
- Initially, the file has i_size=4k, a regular extent from 0-4k, and a
prealloc extent beyond i_size from 4k-12k. The prealloc extent item is
the last item in its B-tree leaf.
- The file is fsync'd, which copies its inode item and both extent items
to the log tree.
- An xattr is set on the file, which sets the
BTRFS_INODE_COPY_EVERYTHING flag.
- The range 4k-8k in the file is written using direct I/O. i_size is
extended to 8k, but the ordered extent is still in flight.
- The file is fsync'd. Since BTRFS_INODE_COPY_EVERYTHING is set, this
calls copy_inode_items_to_log(), which calls
btrfs_log_prealloc_extents().
- btrfs_log_prealloc_extents() finds the 4k-12k prealloc extent in the
filesystem tree. Since it starts before i_size, it skips it. Since it
is the last item in its B-tree leaf, it calls btrfs_next_leaf().
- btrfs_next_leaf() unlocks the path.
- The ordered extent completion runs, which converts the 4k-8k part of
the prealloc extent to written and inserts the remaining prealloc part
from 8k-12k.
- btrfs_next_leaf() does a search and finds the new prealloc extent
8k-12k.
- btrfs_log_prealloc_extents() copies the 8k-12k prealloc extent into
the log tree. Note that it overlaps with the 4k-12k prealloc extent
that was copied to the log tree by the first fsync.
- fsync calls btrfs_log_changed_extents(), which tries to log the 4k-8k
extent that was written.
- This tries to drop the range 4k-8k in the log tree, which requires
adjusting the start of the 4k-12k prealloc extent in the log tree to
8k.
- btrfs_set_item_key_safe() sees that there is already an extent
starting at 8k in the log tree and calls BUG().
Fix this by detecting when we're about to insert an overlapping file
extent item in the log tree and truncating the part that would overlap.
CC: stable@vger.kernel.org # 6.1+
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If a write path in COW mode fails, either before submitting a bio for the
new extents or an actual IO error happens, we can end up allowing a fast
fsync to log file extent items that point to unwritten extents.
This is because dropping the extent maps happens when completing ordered
extents, at btrfs_finish_one_ordered(), and the completion of an ordered
extent is executed in a work queue.
This can result in a fast fsync to start logging file extent items based
on existing extent maps before the ordered extents complete, therefore
resulting in a log that has file extent items that point to unwritten
extents, resulting in a corrupt file if a crash happens after and the log
tree is replayed the next time the fs is mounted.
This can happen for both direct IO writes and buffered writes.
For example consider a direct IO write, in COW mode, that fails at
btrfs_dio_submit_io() because btrfs_extract_ordered_extent() returned an
error:
1) We call btrfs_finish_ordered_extent() with the 'uptodate' parameter
set to false, meaning an error happened;
2) That results in marking the ordered extent with the BTRFS_ORDERED_IOERR
flag;
3) btrfs_finish_ordered_extent() queues the completion of the ordered
extent - so that btrfs_finish_one_ordered() will be executed later in
a work queue. That function will drop extent maps in the range when
it's executed, since the extent maps point to unwritten locations
(signaled by the BTRFS_ORDERED_IOERR flag);
4) After calling btrfs_finish_ordered_extent() we keep going down the
write path and unlock the inode;
5) After that a fast fsync starts and locks the inode;
6) Before the work queue executes btrfs_finish_one_ordered(), the fsync
task sees the extent maps that point to the unwritten locations and
logs file extent items based on them - it does not know they are
unwritten, and the fast fsync path does not wait for ordered extents
to complete, which is an intentional behaviour in order to reduce
latency.
For the buffered write case, here's one example:
1) A fast fsync begins, and it starts by flushing delalloc and waiting for
the writeback to complete by calling filemap_fdatawait_range();
2) Flushing the dellaloc created a new extent map X;
3) During the writeback some IO error happened, and at the end io callback
(end_bbio_data_write()) we call btrfs_finish_ordered_extent(), which
sets the BTRFS_ORDERED_IOERR flag in the ordered extent and queues its
completion;
4) After queuing the ordered extent completion, the end io callback clears
the writeback flag from all pages (or folios), and from that moment the
fast fsync can proceed;
5) The fast fsync proceeds sees extent map X and logs a file extent item
based on extent map X, resulting in a log that points to an unwritten
data extent - because the ordered extent completion hasn't run yet, it
happens only after the logging.
To fix this make btrfs_finish_ordered_extent() set the inode flag
BTRFS_INODE_NEEDS_FULL_SYNC in case an error happened for a COW write,
so that a fast fsync will wait for ordered extent completion.
Note that this issues of using extent maps that point to unwritten
locations can not happen for reads, because in read paths we start by
locking the extent range and wait for any ordered extents in the range
to complete before looking for extent maps.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Although 'norecovery' mount option was marked as deprecated for a long
time and a warning message was printed during the deprecation window,
it's still actively utilized by several projects that need a safer way
to mount a btrfs without any writes.
Furthermore this 'norecovery' mount option is supported by other major
filesystems, which makes it less clear what's our motivation to remove
it.
Re-introduce the 'norecovery' mount option, and output a message to recommend
'rescue=nologreplay' option.
Link: https://lore.kernel.org/linux-btrfs/ZkxZT0J-z0GYvfy8@gardel-login/#t
Link: https://github.com/systemd/systemd/pull/32892
Link: https://bugzilla.suse.com/show_bug.cgi?id=1222429
Reported-by: Lennart Poettering <lennart@poettering.net>
Reported-by: Jiri Slaby <jslaby@suse.com>
Fixes: a1912f7121 ("btrfs: remove code for inode_cache and recovery mount options")
CC: stable@vger.kernel.org # 6.8+
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
At lookup_extent_data_ref() we are incorrectly checking if we are at the
last slot of the last leaf in the extent tree. We are returning -ENOENT
if btrfs_next_leaf() returns a value greater than 1, but btrfs_next_leaf()
never returns anything greater than 1:
1) It returns < 0 on error;
2) 0 if there is a next leaf (or a new item was added to the end of the
current leaf after releasing the path);
3) 1 if there are no more leaves (and no new items were added to the last
leaf after releasing the path).
So fix this by checking if the return value is greater than zero instead
of being greater than one.
Fixes: 1618aa3c2e ("btrfs: simplify return variables in lookup_extent_data_ref()")
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The following error message is displayed:
../fs/btrfs/scrub.c:2152:9: error: ‘ret’ may be used uninitialized
in this function [-Werror=maybe-uninitialized]"
Compiler version: gcc version: (Debian 10.2.1-6) 10.2.1 20210110
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: Lu Yao <yaolu@kylinos.cn>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
While loading a zone's info during creation of a block group, we can race
with a device replace operation and then trigger a use-after-free on the
device that was just replaced (source device of the replace operation).
This happens because at btrfs_load_zone_info() we extract a device from
the chunk map into a local variable and then use the device while not
under the protection of the device replace rwsem. So if there's a device
replace operation happening when we extract the device and that device
is the source of the replace operation, we will trigger a use-after-free
if before we finish using the device the replace operation finishes and
frees the device.
Fix this by enlarging the critical section under the protection of the
device replace rwsem so that all uses of the device are done inside the
critical section.
CC: stable@vger.kernel.org # 6.1.x: 15c12fcc50: btrfs: zoned: introduce a zone_info struct in btrfs_load_block_group_zone_info
CC: stable@vger.kernel.org # 6.1.x: 09a46725cc: btrfs: zoned: factor out per-zone logic from btrfs_load_block_group_zone_info
CC: stable@vger.kernel.org # 6.1.x: 9e0e3e74dc: btrfs: zoned: factor out single bg handling from btrfs_load_block_group_zone_info
CC: stable@vger.kernel.org # 6.1.x: 87463f7e02: btrfs: zoned: factor out DUP bg handling from btrfs_load_block_group_zone_info
CC: stable@vger.kernel.org # 6.1.x
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
If we delete subvolumes whose ID is the largest in the filesystem, then
unmount and mount again, then btrfs_init_root_free_objectid on the
tree_root will select a subvolid smaller than that one and thus allow
reusing it.
If we are also using qgroups (and particularly squotas) it is possible
to delete the subvol without deleting the qgroup. In that case, we will
be able to create a new subvol whose id already has a level 0 qgroup.
This will result in re-using that qgroup which would then lead to
incorrect accounting.
Fixes: 6ed05643dd ("btrfs: create qgroup earlier in snapshot creation")
CC: stable@vger.kernel.org # 6.7+
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Boris Burkov <boris@bur.io>
Signed-off-by: David Sterba <dsterba@suse.com>
On filesystems without enabled quotas there's still a warning message in
the logs when rescan is called. In that case it's not a problem that
should be reported, rescan can be called unconditionally. Change the
error code to ENOTCONN which is used for 'quotas not enabled' elsewhere.
Remove message (also a warning) when rescan is called during an ongoing
rescan, this brings no useful information and the error code is
sufficient.
Change message levels to debug for now, they can be removed eventually.
CC: stable@vger.kernel.org # 6.6+
Reviewed-by: Boris Burkov <boris@bur.io>
Signed-off-by: David Sterba <dsterba@suse.com>
The "i++" was accidentally left out so it just sets qgids[0] over and
over.
This can lead to unexpected problems, as the groups[1:] would be all 0,
leading to later find_qgroup_rb() unable to find a qgroup and cause
snapshot creation failure.
Fixes: 5343cd9364 ("btrfs: qgroup: simple quota auto hierarchy for nested subvolumes")
CC: stable@vger.kernel.org # 6.7+
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently the error status of super block write is tracked in page/folio
status bit Error. For that we need to keep the reference for the whole
duration of write and wait.
Count the number of superblock writeback errors in the btrfs_device.
That means we don't need the folio to stay around until it's waited for,
and can avoid the extra call to folio_get/put.
Also remove a mention of PageError in a comment as it's the last mention
of the page Error state.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Iterate over folios instead of bvecs. Switch the order of unlock and put
to be the usual order; we know this folio can't be put until it's been
waited for, but that's fragile. Remove the calls to ClearPageUptodate /
SetPageUptodate -- if PAGE_SIZE is larger than BTRFS_SUPER_INFO_SIZE,
we'd be marking the entire folio uptodate without having actually
initialised all the bytes in the page.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is a direct conversion from pages to folios, assuming single page
folio. Also removes some calls to obsolete APIs and some hidden calls to
compound_head().
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is a direct conversion from pages to folios, assuming single page
folio. Also removes a few calls to compound_head() and calls to
obsolete APIs.
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we have the lock_extent tightly coupled with
extent_clear_unlock_delalloc we can add a cached state to
extent_clear_unlock_delalloc and benefit from skipping the extra lookup
when we're doing cow.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We don't need to include the time we spend in the allocator under our
extent lock protection, move it after the allocator and make sure we
lock the extent in the error case to ensure we're not clearing these
bits without the extent lock held.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we've got the extent lock pushed into cow_file_range() we can
push it further down into the allocation loop. This allows us to only
hold the extent lock during the dropping of the extent map range and
inserting the ordered extent.
This makes the error case a little trickier as we'll now have to lock
the range before clearing any of the other extent bits for the range,
but this is the error path so is less performance critical.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
These checks aren't reliant on the extent lock. Move this up into
cow_file_range_inline(), and then update encoded writes to call this
check before calling __cow_file_range_inline(). This will allow us to
skip the extent lock if we're not able to inline the given extent.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that we've pushed the lock_extent() into cow_file_range() we can
push the extent locking into cow_file_range_inline() and move the
lock_extent in cow_file_range() to after we call
cow_file_range_inline().
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Now that cow_file_range is the only function that is called with the
range locked, push this call into cow_file_range so we can further
narrow the scope.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
This is used by zoned but also as the fallback for uncompressed extents
when we fail to compress the ranges. Push the extent lock into
run_dealloc_cow(), and adjust the compression case to take the extent
lock after calling run_delalloc_cow().
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since we immediately unlock the extent range when we enter
run_delalloc_compressed() simply move the lock_extent() down to cover
cow_file_range() and then remove the unlock_extent() from
run_delalloc_compressed.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
run_delalloc_nocow is a little special because we use the file extents
to see if we can nocow a range. We don't actually need the protection
of the extent lock to look at the file extents at this point however.
We are currently holding the page lock for this range, so we are
protected from anybody who would simultaneously be modifying the file
extent items for this range.
* mmap() - we're holding the page lock.
* buffered writes - we're holding the page lock.
* direct writes - we're holding the page lock and direct IO has to flush
page cache before it's able to continue.
* fallocate() - all callers flush the range and wait on ordered extents
while holding the inode lock and the mmap lock, so we are again saved
by the page lock.
We want to use the extent lock to protect
1) The mapping tree for the given range.
2) The ordered extents for the given range.
3) The io_tree for the given range.
Push the extent lock down to cover these operations. In the
fallback_to_cow() case we simply lock before doing anything and rely on
the cow_file_range() helper to handle it's range properly.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We have the following pattern
while (1) {
if (cur_offset > end)
break;
}
Which is just
while (cur_offset <= end) {
...
}
so adjust the code to be more clear.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
run_delalloc_nocow is a bit special as it walks through the file extents
for the inode and determines what it can nocow and what it can't. This
is the more complicated area for extent locking, so start with this
function.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We want to limit the scope of the extent lock to be around operations
that can change in flight. Currently we hold the extent lock through
the entire writepage operation, which isn't really necessary.
We want to protect to make sure nobody has updated DELALLOC. In
find_lock_delalloc_range we must lock the range in order to validate the
contents of our io_tree. However once we've done that we're safe to
unlock the range and continue, as we have the page lock already held for
the range.
We are protected from all operations at this point.
* mmap() - we're holding the page lock, thus are protected.
* buffered writes - again, we're protected because we take the page lock
for the first and last page in our range for buffered writes so we
won't create new delalloc ranges in this area.
* direct IO - we invalidate pagecache before attempting to write a new
area, which requires the page lock, so again are protected once we're
holding the page lock on this range.
Additionally this behavior actually already exists for compressed, we
unlock the range as soon as we start to process the async extents, and
re-lock it during compression. So this is completely safe, and makes
the locking more consistent.
Make this simple by just pushing the extent lock into
btrfs_run_delalloc_range. From there followup patches will push the
lock further down into its users.
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We currently don't lock the extent when we're doing a
cow_file_range_inline() for a compressed extent. This isn't a problem
necessarily, but it's inconsistent with the rest of our usage of
cow_file_range_inline(). This also leads to some extra weird logic
around whether the extent is locked or not. Fix this to lock the extent
before calling cow_file_range_inline() in compression to make it
consistent with the rest of the inline users. In future patches this
will be pushed down into the cow_file_range_inline() helper, so we're
fine with the quick and dirty locking here. This patch exists to make
the behavior change obvious.
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We duplicate the extent cleanup for cow_file_range_inline() in the cow
and compressed case. The encoded case doesn't need to do cleanup the
same way, so rename cow_file_range_inline to __cow_file_range_inline and
then make cow_file_range_inline handle the extent cleanup appropriately,
and update the callers.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Since 4750af3bbe ("btrfs: prevent extent_clear_unlock_delalloc() to
unlock page not locked by __process_pages_contig()") we have been
unlocking the locked page manually instead of via
extent_clear_unlock_delalloc() because of subpage blocksize support.
However we actually disable inline extent creation for subpage blocksize
support, so this behavior isn't necessary. Remove this code and
comment, if at some point the subpage blocksize code grows support for
inline extents this can be re-evaluated.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Currently we have a lot of duplicated checks of
if (start == 0 && fs_info->sectorsize == PAGE_SIZE)
cow_file_range_inline();
Instead of duplicating this check everywhere, consolidate all of the
inline extent logic into a helper which documents all of the checks and
then use that helper inside of cow_file_range_inline(). With this we
can clean up all of the calls to either unconditionally call
cow_file_range_inline(), or at least reduce the checks we're doing
before we call cow_file_range_inline();
Reviewed-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
In the cow path we will clone the reloc csums for relocated data
extents, and if there's an error we already have an ordered extent and
rely on the ordered extent finishing to clean everything up.
There's a problem however, we don't mark the ordered extent with an
error, we pretend like everything was just fine. If we were at the end
of our range we won't actually bubble up this error anywhere, and we
could end up inserting an extent that doesn't have csums where it should
have them.
Fix this by adding a helper to mark the ordered extent with an error,
and then use this when we fail to lookup the csums in
btrfs_reloc_clone_csums. Use this helper in the other place where we
use the same pattern while we're here.
This will prevent us from erroneously inserting the extent that doesn't
have the required checksums.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The function create_io_em() is called before we submit an IO, to update
the in-memory extent map for the involved range.
This patch changes the following aspects:
- Does not allow BTRFS_ORDERED_NOCOW type
For real NOCOW (excluding NOCOW writes into preallocated ranges)
writes, we never call create_io_em(), as we does not need to update
the extent map at all.
So remove the sanity check allowing BTRFS_ORDERED_NOCOW type.
- Add extra sanity checks
* PREALLOC
- @block_len == len
For uncompressed writes.
* REGULAR
- @block_len == @orig_block_len == @ram_bytes == @len
We're creating a new uncompressed extent, and referring all of it.
- @orig_start == @start
We haven no offset inside the extent.
* COMPRESSED
- valid @compress_type
- @len <= @ram_bytes
This is to co-operate with encoded writes, which can cause a new
file extent referring only part of a uncompressed extent.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
With the tree-checker ensuring all inline file extents starts at file
offset 0 and has a length no larger than sectorsize, we can simplify the
calculation to assigned those fixes values directly.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
