The free space cache has turned out to be a scalability bottleneck on
large, busy filesystems. When the cache for a lot of block groups needs
to be written out, we can get extremely long commit times; if this
happens in the critical section, things are especially bad because we
block new transactions from happening.
The main problem with the free space cache is that it has to be written
out in its entirety and is managed in an ad hoc fashion. Using a B-tree
to store free space fixes this: updates can be done as needed and we get
all of the benefits of using a B-tree: checksumming, RAID handling,
well-understood behavior.
With the free space tree, we get commit times that are about the same as
the no cache case with load times slower than the free space cache case
but still much faster than the no cache case. Free space is represented
with extents until it becomes more space-efficient to use bitmaps,
giving us similar space overhead to the free space cache.
The operations on the free space tree are: adding and removing free
space, handling the creation and deletion of block groups, and loading
the free space for a block group. We can also create the free space tree
by walking the extent tree and clear the free space tree.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We're also going to load the free space tree from caching_thread(), so
we should refactor some of the common code.
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
As of my previous change titled "Btrfs: fix scrub preventing unused block
groups from being deleted", the following warning at
extent-tree.c:btrfs_delete_unused_bgs() can be hit when we mount the a
filesysten with "-o discard":
10263 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
10264 {
(...)
10405 if (trimming) {
10406 WARN_ON(!list_empty(&block_group->bg_list));
10407 spin_lock(&trans->transaction->deleted_bgs_lock);
10408 list_move(&block_group->bg_list,
10409 &trans->transaction->deleted_bgs);
10410 spin_unlock(&trans->transaction->deleted_bgs_lock);
10411 btrfs_get_block_group(block_group);
10412 }
(...)
This happens because scrub can now add back the block group to the list of
unused block groups (fs_info->unused_bgs). This is dangerous because we
are moving the block group from the unused block groups list to the list
of deleted block groups without holding the lock that protects the source
list (fs_info->unused_bgs_lock).
The following diagram illustrates how this happens:
CPU 1 CPU 2
cleaner_kthread()
btrfs_delete_unused_bgs()
sees bg X in list
fs_info->unused_bgs
deletes bg X from list
fs_info->unused_bgs
scrub_enumerate_chunks()
searches device tree using
its commit root
finds device extent for
block group X
gets block group X from the tree
fs_info->block_group_cache_tree
(via btrfs_lookup_block_group())
sets bg X to RO (again)
scrub_chunk(bg X)
sets bg X back to RW mode
adds bg X to the list
fs_info->unused_bgs again,
since it's still unused and
currently not in that list
sets bg X to RO mode
btrfs_remove_chunk(bg X)
--> discard is enabled and bg X
is in the fs_info->unused_bgs
list again so the warning is
triggered
--> we move it from that list into
the transaction's delete_bgs
list, but we can have another
task currently manipulating
the first list (fs_info->unused_bgs)
Fix this by using the same lock (fs_info->unused_bgs_lock) to protect both
the list of unused block groups and the list of deleted block groups. This
makes it safe and there's not much worry for more lock contention, as this
lock is seldom used and only the cleaner kthread adds elements to the list
of deleted block groups. The warning goes away too, as this was previously
an impossible case (and would have been better a BUG_ON/ASSERT) but it's
not impossible anymore.
Reproduced with fstest btrfs/073 (using MOUNT_OPTIONS="-o discard").
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Commit 0ed4792 ('btrfs: qgroup: Switch to new extent-oriented qgroup
mechanism.') removed our qgroup accounting during
btrfs_drop_snapshot(). Predictably, this results in qgroup numbers
going bad shortly after a snapshot is removed.
Fix this by adding a dirty extent record when we encounter extents during
our shared subtree walk. This effectively restores the functionality we had
with the original shared subtree walking code in 1152651 (btrfs: qgroup:
account shared subtrees during snapshot delete).
The idea with the original patch (and this one) is that shared subtrees can
get skipped during drop_snapshot. The shared subtree walk then allows us a
chance to visit those extents and add them to the qgroup work for later
processing. This ultimately makes the accounting for drop snapshot work.
The new qgroup code nicely handles all the other extents during the tree
walk via the ref dec/inc functions so we don't have to add actions beyond
what we had originally.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Signed-off-by: Chris Mason <clm@fb.com>
When a block group becomes unused and the cleaner kthread is currently
running, we can end up getting the current transaction aborted with error
-ENOENT when we try to commit the transaction, leading to the following
trace:
[59779.258768] WARNING: CPU: 3 PID: 5990 at fs/btrfs/extent-tree.c:3740 btrfs_write_dirty_block_groups+0x17c/0x214 [btrfs]()
[59779.272594] BTRFS: Transaction aborted (error -2)
(...)
[59779.291137] Call Trace:
[59779.291621] [<ffffffff812566f4>] dump_stack+0x4e/0x79
[59779.292543] [<ffffffff8104d0a6>] warn_slowpath_common+0x9f/0xb8
[59779.293435] [<ffffffffa04cb81f>] ? btrfs_write_dirty_block_groups+0x17c/0x214 [btrfs]
[59779.295000] [<ffffffff8104d107>] warn_slowpath_fmt+0x48/0x50
[59779.296138] [<ffffffffa04c2721>] ? write_one_cache_group.isra.32+0x77/0x82 [btrfs]
[59779.297663] [<ffffffffa04cb81f>] btrfs_write_dirty_block_groups+0x17c/0x214 [btrfs]
[59779.299141] [<ffffffffa0549b0d>] commit_cowonly_roots+0x1de/0x261 [btrfs]
[59779.300359] [<ffffffffa04dd5b6>] btrfs_commit_transaction+0x4c4/0x99c [btrfs]
[59779.301805] [<ffffffffa04b5df4>] btrfs_sync_fs+0x145/0x1ad [btrfs]
[59779.302893] [<ffffffff81196634>] sync_filesystem+0x7f/0x93
(...)
[59779.318186] ---[ end trace 577e2daff90da33a ]---
The following diagram illustrates a sequence of steps leading to this
problem:
CPU 1 CPU 2
<at transaction N>
adds bg A to list
fs_info->unused_bgs
adds bg B to list
fs_info->unused_bgs
<transaction kthread
commits transaction N
and wakes up the
cleaner kthread>
cleaner kthread
delete_unused_bgs()
sees bg A in list
fs_info->unused_bgs
btrfs_start_transaction()
<transaction N + 1 starts>
deletes bg A
update_block_group(bg C)
--> adds bg C to list
fs_info->unused_bgs
deletes bg B
sees bg C in the list
fs_info->unused_bgs
btrfs_remove_chunk(bg C)
btrfs_remove_block_group(bg C)
--> checks if the block group
is in a dirty list, and
because it isn't now, it
does nothing
--> the block group item
is deleted from the
extent tree
--> adds bg C to list
transaction->dirty_bgs
some task calls
btrfs_commit_transaction(t N + 1)
commit_cowonly_roots()
btrfs_write_dirty_block_groups()
--> sees bg C in cur_trans->dirty_bgs
--> calls write_one_cache_group()
which returns -ENOENT because
it did not find the block group
item in the extent tree
--> transaction aborte with -ENOENT
because write_one_cache_group()
returned that error
So fix this by adding a block group to the list of dirty block groups
before adding it to the list of unused block groups.
This happened on a stress test using fsstress plus concurrent calls to
fallocate 20G and truncate (releasing part of the space allocated with
fallocate).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently scrub can race with the cleaner kthread when the later attempts
to delete an unused block group, and the result is preventing the cleaner
kthread from ever deleting later the block group - unless the block group
becomes used and unused again. The following diagram illustrates that
race:
CPU 1 CPU 2
cleaner kthread
btrfs_delete_unused_bgs()
gets block group X from
fs_info->unused_bgs and
removes it from that list
scrub_enumerate_chunks()
searches device tree using
its commit root
finds device extent for
block group X
gets block group X from the tree
fs_info->block_group_cache_tree
(via btrfs_lookup_block_group())
sets bg X to RO
sees the block group is
already RO and therefore
doesn't delete it nor adds
it back to unused list
So fix this by making scrub add the block group again to the list of
unused block groups if the block group is still unused when it finished
scrubbing it and it hasn't been removed already.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We were using only 1 transaction unit when attempting to delete an unused
block group but in reality we need 3 + N units, where N corresponds to the
number of stripes. We were accounting only for the addition of the orphan
item (for the block group's free space cache inode) but we were not
accounting that we need to delete one block group item from the extent
tree, one free space item from the tree of tree roots and N device extent
items from the device tree.
While one unit is not enough, it worked most of the time because for each
single unit we are too pessimistic and assume an entire tree path, with
the highest possible heigth (8), needs to be COWed with eventual node
splits at every possible level in the tree, so there was usually enough
reserved space for removing all the items and adding the orphan item.
However after adding the orphan item, writepages() can by called by the VM
subsystem against the btree inode when we are under memory pressure, which
causes writeback to start for the nodes we COWed before, this forces the
operation to remove the free space item to COW again some (or all of) the
same nodes (in the tree of tree roots). Even without writepages() being
called, we could fail with ENOSPC because these items are located in
multiple trees and one of them might have a higher heigth and require
node/leaf splits at many levels, exhausting all the reserved space before
removing all the items and adding the orphan.
In the kernel 4.0 release, commit 3d84be7991 ("Btrfs: fix BUG_ON in
btrfs_orphan_add() when delete unused block group"), we attempted to fix
a BUG_ON due to ENOSPC when trying to add the orphan item by making the
cleaner kthread reserve one transaction unit before attempting to remove
the block group, but this was not enough. We had a couple user reports
still hitting the same BUG_ON after 4.0, like Stefan Priebe's report on
a 4.2-rc6 kernel for example:
http://www.spinics.net/lists/linux-btrfs/msg46070.html
So fix this by reserving all the necessary units of metadata.
Reported-by: Stefan Priebe <s.priebe@profihost.ag>
Fixes: 3d84be7991 ("Btrfs: fix BUG_ON in btrfs_orphan_add() when delete unused block group")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
It's possible to reach a state where the cleaner kthread isn't able to
start a transaction to delete an unused block group due to lack of enough
free metadata space and due to lack of unallocated device space to allocate
a new metadata block group as well. If this happens try to use space from
the global block group reserve just like we do for unlink operations, so
that we don't reach a permanent state where starting a transaction for
filesystem operations (file creation, renames, etc) keeps failing with
-ENOSPC. Such an unfortunate state was observed on a machine where over
a dozen unused data block groups existed and the cleaner kthread was
failing to delete them due to ENOSPC error when attempting to start a
transaction, and even running balance with a -dusage=0 filter failed with
ENOSPC as well. Also unmounting and mounting again the filesystem didn't
help. Allowing the cleaner kthread to use the global block reserve to
delete the unused data block groups fixed the problem.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
No need to use root->fs_info in btrfs_delete_unused_bgs(),
use fs_info directly instead.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Reproduce:
(In integration-4.3 branch)
TEST_DEV=(/dev/vdg /dev/vdh)
TEST_DIR=/mnt/tmp
umount "$TEST_DEV" >/dev/null
mkfs.btrfs -f -d raid1 "${TEST_DEV[@]}"
mount -o nospace_cache "$TEST_DEV" "$TEST_DIR"
umount "$TEST_DEV"
mount -o nospace_cache "$TEST_DEV" "$TEST_DIR"
btrfs filesystem usage $TEST_DIR
We can see the data chunk changed from raid1 to single:
# btrfs filesystem usage $TEST_DIR
Data,single: Size:8.00MiB, Used:0.00B
/dev/vdg 8.00MiB
#
Reason:
When a empty filesystem mount with -o nospace_cache, the last
data blockgroup will be auto-removed in umount.
Then if we mount it again, there is no data chunk in the
filesystem, so the only available data profile is 0x0, result
is all new chunks are created as single type.
Fix:
Don't auto-delete last blockgroup for a raid type.
Test:
Test by above script, and confirmed the logic by debug output.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
When executing generic/001 in a loop on a ppc64 machine (with both sectorsize
and nodesize set to 64k), the following call trace is observed,
WARNING: at /root/repos/linux/fs/btrfs/locking.c:253
Modules linked in:
CPU: 2 PID: 8353 Comm: umount Not tainted 4.3.0-rc5-13676-ga5e681d #54
task: c0000000f2b1f560 ti: c0000000f6008000 task.ti: c0000000f6008000
NIP: c000000000520c88 LR: c0000000004a3b34 CTR: 0000000000000000
REGS: c0000000f600a820 TRAP: 0700 Not tainted (4.3.0-rc5-13676-ga5e681d)
MSR: 8000000102029032 <SF,VEC,EE,ME,IR,DR,RI> CR: 24444884 XER: 00000000
CFAR: c0000000004a3b30 SOFTE: 1
GPR00: c0000000004a3b34 c0000000f600aaa0 c00000000108ac00 c0000000f5a808c0
GPR04: 0000000000000000 c0000000f600ae60 0000000000000000 0000000000000005
GPR08: 00000000000020a1 0000000000000001 c0000000f2b1f560 0000000000000030
GPR12: 0000000084842882 c00000000fdc0900 c0000000f600ae60 c0000000f070b800
GPR16: 0000000000000000 c0000000f3c8a000 0000000000000000 0000000000000049
GPR20: 0000000000000001 0000000000000001 c0000000f5aa01f8 0000000000000000
GPR24: 0f83e0f83e0f83e1 c0000000f5a808c0 c0000000f3c8d000 c000000000000000
GPR28: c0000000f600ae74 0000000000000001 c0000000f3c8d000 c0000000f5a808c0
NIP [c000000000520c88] .btrfs_tree_lock+0x48/0x2a0
LR [c0000000004a3b34] .btrfs_lock_root_node+0x44/0x80
Call Trace:
[c0000000f600aaa0] [c0000000f600ab80] 0xc0000000f600ab80 (unreliable)
[c0000000f600ab80] [c0000000004a3b34] .btrfs_lock_root_node+0x44/0x80
[c0000000f600ac00] [c0000000004a99dc] .btrfs_search_slot+0xa8c/0xc00
[c0000000f600ad40] [c0000000004ab878] .btrfs_insert_empty_items+0x98/0x120
[c0000000f600adf0] [c00000000050da44] .btrfs_finish_chunk_alloc+0x1d4/0x620
[c0000000f600af20] [c0000000004be854] .btrfs_create_pending_block_groups+0x1d4/0x2c0
[c0000000f600b020] [c0000000004bf188] .do_chunk_alloc+0x3c8/0x420
[c0000000f600b100] [c0000000004c27cc] .find_free_extent+0xbfc/0x1030
[c0000000f600b260] [c0000000004c2ce8] .btrfs_reserve_extent+0xe8/0x250
[c0000000f600b330] [c0000000004c2f90] .btrfs_alloc_tree_block+0x140/0x590
[c0000000f600b440] [c0000000004a47b4] .__btrfs_cow_block+0x124/0x780
[c0000000f600b530] [c0000000004a4fc0] .btrfs_cow_block+0xf0/0x250
[c0000000f600b5e0] [c0000000004a917c] .btrfs_search_slot+0x22c/0xc00
[c0000000f600b720] [c00000000050aa40] .btrfs_remove_chunk+0x1b0/0x9f0
[c0000000f600b850] [c0000000004c4e04] .btrfs_delete_unused_bgs+0x434/0x570
[c0000000f600b950] [c0000000004d3cb8] .close_ctree+0x2e8/0x3b0
[c0000000f600ba20] [c00000000049d178] .btrfs_put_super+0x18/0x30
[c0000000f600ba90] [c000000000243cd4] .generic_shutdown_super+0xa4/0x1a0
[c0000000f600bb10] [c0000000002441d8] .kill_anon_super+0x18/0x30
[c0000000f600bb90] [c00000000049c898] .btrfs_kill_super+0x18/0xc0
[c0000000f600bc10] [c0000000002444f8] .deactivate_locked_super+0x98/0xe0
[c0000000f600bc90] [c000000000269f94] .cleanup_mnt+0x54/0xa0
[c0000000f600bd10] [c0000000000bd744] .task_work_run+0xc4/0x100
[c0000000f600bdb0] [c000000000016334] .do_notify_resume+0x74/0x80
[c0000000f600be30] [c0000000000098b8] .ret_from_except_lite+0x64/0x68
Instruction dump:
fba1ffe8 fbc1fff0 fbe1fff8 7c791b78 f8010010 f821ff21 e94d0290 81030040
812a04e8 7d094a78 7d290034 5529d97e <0b090000> 3b400000 3be30050 3bc3004c
The above call trace is seen even on x86_64; albeit very rarely and that too
with nodesize set to 64k and with nospace_cache mount option being used.
The reason for the above call trace is,
btrfs_remove_chunk
check_system_chunk
Allocate chunk if required
For each physical stripe on underlying device,
btrfs_free_dev_extent
...
Take lock on Device tree's root node
btrfs_cow_block("dev tree's root node");
btrfs_reserve_extent
find_free_extent
index = BTRFS_RAID_DUP;
have_caching_bg = false;
When in LOOP_CACHING_NOWAIT state, Assume we find a block group
which is being cached; Hence have_caching_bg is set to true
When repeating the search for the next RAID index, we set
have_caching_bg to false.
Hence right after completing the LOOP_CACHING_NOWAIT state, we incorrectly
skip LOOP_CACHING_WAIT state and move to LOOP_ALLOC_CHUNK state where we
allocate a chunk and try to add entries corresponding to the chunk's physical
stripe into the device tree. When doing so the task deadlocks itself waiting
for the blocking lock on the root node of the device tree.
This commit fixes the issue by introducing a new local variable to help
indicate as to whether a block group of any RAID type is being cached.
Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Between btrfs_allocerved_file_extent() and
btrfs_add_delayed_qgroup_reserve(), there is a window that delayed_refs
are run and delayed ref head maybe freed before
btrfs_add_delayed_qgroup_reserve().
This will cause btrfs_dad_delayed_qgroup_reserve() to return -ENOENT,
and cause transaction to be aborted.
This patch will record qgroup reserve space info into delayed_ref_head
at btrfs_add_delayed_ref(), to eliminate the race window.
Reported-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we are extremely fragmented then we won't be able to create a free_cluster.
So if this happens set last_ptr->fragmented so that all future allcations will
give up trying to create a cluster. When we unpin extents we will unset
->fragmented if we free up a sufficient amount of space in a block group.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We try really really hard to make allocations, but sometimes it is just not
going to happen, especially when free space is extremely fragmented. So add a
few short cuts through the looping states. For example if we couldn't allocate
a chunk, just go straight to the NO_EMPTY_SIZE loop. If there are no uncached
block groups and we've done a full search, go straight to the ALLOC_CHUNK stage.
And finally if we already have empty_size and empty_cluster set to 0 go ahead
and return -ENOSPC. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we hit ENOSPC when setting up a space cache don't bother setting up any of
the other space cache's in this transaction, it'll just induce unnecessary
latency. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we are heavily fragmented we can induce a lot of latency trying to make an
allocation happen that is simply not going to happen. Thankfully we keep track
of our max_extent_size when going through the allocator, so if we get to the
point where we are exiting find_free_extent with ENOSPC then set our
space_info->max_extent_size so we can keep future allocations from having to pay
this cost. We reset the max_extent_size whenever we release pinned bytes back
into this space info so we can redo all the work. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The space cache needs to have contiguous allocations, and the allocator tries to
make allocations by reducing the amount of bytes requested and re-searching.
But this just makes us waste time when we are very fragmented, so if we can't
find our space just exit, don't bother trying to search again. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
I want to set some per transaction flags, so instead of adding yet another int
lets just convert the current two int indicators to flags and add a flags field
for future use. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
In tracking down these weird bitmap problems it was helpful to artificially
create an extremely fragmented file system. These mount options let us either
fragment data or metadata or both. With these options I could reproduce all
sorts of weird latencies and hangs that occur under extreme fragmentation and
get them fixed. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
In clear_bit_hook, qgroup reserved data is already handled quite well,
either released by finish_ordered_io or invalidatepage.
So calling btrfs_qgroup_free_data() here is completely meaningless, and
since btrfs_qgroup_free_data() will lock io_tree, so it can't be called
with io_tree lock hold.
This patch will add a new function
btrfs_free_reserved_data_space_noquota() for clear_bit_hook() to cease
the lockdep warning.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
For NOCOW and inline case, there will be no delayed_ref created for
them, so we should free their reserved data space at proper
time(finish_ordered_io for NOCOW and cow_file_inline for inline).
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Cleanup the old facilities which use old btrfs_qgroup_reserve() function
call, replace them with the newer version, and remove the "__" prefix in
them.
Also, make btrfs_qgroup_reserve/free() functions private, as they are
now only used inside qgroup codes.
Now, the whole btrfs qgroup is swithed to use the new reserve facilities.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add new version of btrfs_delalloc_reserve_space() and
btrfs_delalloc_release_space() functions, which supports accurate qgroup
reserve.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Use new reserve/free for buffered write and inode cache.
For buffered write case, as nodatacow write won't increase quota account,
so unlike old behavior which does reserve before check nocow, now we
check nocow first and then only reserve data if we can't do nocow write.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add new functions __btrfs_check_data_free_space() and
__btrfs_free_reserved_data_space() to work with new accurate qgroup
reserved space framework.
The new function will replace old btrfs_check_data_free_space() and
btrfs_free_reserved_data_space() respectively, but until all the change
is done, let's just use the new name.
Also, export internal use function btrfs_alloc_data_chunk_ondemand(), as
now qgroup reserve requires precious bytes, some operation can't get the
accurate number in advance(like fallocate).
But data space info check and data chunk allocate doesn't need to be
that accurate, and can be called at the beginning.
So export it for later operations.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
As we have the new metadata reservation functions, use them to replace
the old btrfs_qgroup_reserve() call for metadata.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Qgroup reserved space needs to be released from inode dirty map and get
freed at different timing:
1) Release when the metadata is written into tree
After corresponding metadata is written into tree, any newer write will
be COWed(don't include NOCOW case yet).
So we must release its range from inode dirty range map, or we will
forget to reserve needed range, causing accounting exceeding the limit.
2) Free reserved bytes when delayed ref is run
When delayed refs are run, qgroup accounting will follow soon and turn
the reserved bytes into rfer/excl numbers.
As run_delayed_refs and qgroup accounting are all done at
commit_transaction() time, we are safe to free reserved space in
run_delayed_ref time().
With these timing to release/free reserved space, we should be able to
resolve the long existing qgroup reserve space leak problem.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
btrfs_raid_array[] holds attributes of all raid types.
Use btrfs_raid_array[].devs_min is best way for request
in btrfs_reduce_alloc_profile(), instead of use complex
condition of each raid types.
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Rather than have three separate if() statements for the same outcome
we should just OR them together in the same if() statement.
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Alexandru Moise <00moses.alexander00@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Josef ran into a deadlock while a transaction handle was finalizing the
creation of its block groups, which produced the following trace:
[260445.593112] fio D ffff88022a9df468 0 8924 4518 0x00000084
[260445.593119] ffff88022a9df468 ffffffff81c134c0 ffff880429693c00 ffff88022a9df488
[260445.593126] ffff88022a9e0000 ffff8803490d7b00 ffff8803490d7b18 ffff88022a9df4b0
[260445.593132] ffff8803490d7af8 ffff88022a9df488 ffffffff8175a437 ffff8803490d7b00
[260445.593137] Call Trace:
[260445.593145] [<ffffffff8175a437>] schedule+0x37/0x80
[260445.593189] [<ffffffffa0850f37>] btrfs_tree_lock+0xa7/0x1f0 [btrfs]
[260445.593197] [<ffffffff810db7c0>] ? prepare_to_wait_event+0xf0/0xf0
[260445.593225] [<ffffffffa07eac44>] btrfs_lock_root_node+0x34/0x50 [btrfs]
[260445.593253] [<ffffffffa07eff6b>] btrfs_search_slot+0x88b/0xa00 [btrfs]
[260445.593295] [<ffffffffa08389df>] ? free_extent_buffer+0x4f/0x90 [btrfs]
[260445.593324] [<ffffffffa07f1a06>] btrfs_insert_empty_items+0x66/0xc0 [btrfs]
[260445.593351] [<ffffffffa07ea94a>] ? btrfs_alloc_path+0x1a/0x20 [btrfs]
[260445.593394] [<ffffffffa08403b9>] btrfs_finish_chunk_alloc+0x1c9/0x570 [btrfs]
[260445.593427] [<ffffffffa08002ab>] btrfs_create_pending_block_groups+0x11b/0x200 [btrfs]
[260445.593459] [<ffffffffa0800964>] do_chunk_alloc+0x2a4/0x2e0 [btrfs]
[260445.593491] [<ffffffffa0803815>] find_free_extent+0xa55/0xd90 [btrfs]
[260445.593524] [<ffffffffa0803c22>] btrfs_reserve_extent+0xd2/0x220 [btrfs]
[260445.593532] [<ffffffff8119fe5d>] ? account_page_dirtied+0xdd/0x170
[260445.593564] [<ffffffffa0803e78>] btrfs_alloc_tree_block+0x108/0x4a0 [btrfs]
[260445.593597] [<ffffffffa080c9de>] ? btree_set_page_dirty+0xe/0x10 [btrfs]
[260445.593626] [<ffffffffa07eb5cd>] __btrfs_cow_block+0x12d/0x5b0 [btrfs]
[260445.593654] [<ffffffffa07ebbff>] btrfs_cow_block+0x11f/0x1c0 [btrfs]
[260445.593682] [<ffffffffa07ef8c7>] btrfs_search_slot+0x1e7/0xa00 [btrfs]
[260445.593724] [<ffffffffa08389df>] ? free_extent_buffer+0x4f/0x90 [btrfs]
[260445.593752] [<ffffffffa07f1a06>] btrfs_insert_empty_items+0x66/0xc0 [btrfs]
[260445.593830] [<ffffffffa07ea94a>] ? btrfs_alloc_path+0x1a/0x20 [btrfs]
[260445.593905] [<ffffffffa08403b9>] btrfs_finish_chunk_alloc+0x1c9/0x570 [btrfs]
[260445.593946] [<ffffffffa08002ab>] btrfs_create_pending_block_groups+0x11b/0x200 [btrfs]
[260445.593990] [<ffffffffa0815798>] btrfs_commit_transaction+0xa8/0xb40 [btrfs]
[260445.594042] [<ffffffffa085abcd>] ? btrfs_log_dentry_safe+0x6d/0x80 [btrfs]
[260445.594089] [<ffffffffa082bc84>] btrfs_sync_file+0x294/0x350 [btrfs]
[260445.594115] [<ffffffff8123e29b>] vfs_fsync_range+0x3b/0xa0
[260445.594133] [<ffffffff81023891>] ? syscall_trace_enter_phase1+0x131/0x180
[260445.594149] [<ffffffff8123e35d>] do_fsync+0x3d/0x70
[260445.594169] [<ffffffff81023bb8>] ? syscall_trace_leave+0xb8/0x110
[260445.594187] [<ffffffff8123e600>] SyS_fsync+0x10/0x20
[260445.594204] [<ffffffff8175de6e>] entry_SYSCALL_64_fastpath+0x12/0x71
This happened because the same transaction handle created a large number
of block groups and while finalizing their creation (inserting new items
and updating existing items in the chunk and device trees) a new metadata
extent had to be allocated and no free space was found in the current
metadata block groups, which made find_free_extent() attempt to allocate
a new block group via do_chunk_alloc(). However at do_chunk_alloc() we
ended up allocating a new system chunk too and exceeded the threshold
of 2Mb of reserved chunk bytes, which makes do_chunk_alloc() enter the
final part of block group creation again (at
btrfs_create_pending_block_groups()) and attempt to lock again the root
of the chunk tree when it's already write locked by the same task.
Similarly we can deadlock on extent tree nodes/leafs if while we are
running delayed references we end up creating a new metadata block group
in order to allocate a new node/leaf for the extent tree (as part of
a CoW operation or growing the tree), as btrfs_create_pending_block_groups
inserts items into the extent tree as well. In this case we get the
following trace:
[14242.773581] fio D ffff880428ca3418 0 3615 3100 0x00000084
[14242.773588] ffff880428ca3418 ffff88042d66b000 ffff88042a03c800 ffff880428ca3438
[14242.773594] ffff880428ca4000 ffff8803e4b20190 ffff8803e4b201a8 ffff880428ca3460
[14242.773600] ffff8803e4b20188 ffff880428ca3438 ffffffff8175a437 ffff8803e4b20190
[14242.773606] Call Trace:
[14242.773613] [<ffffffff8175a437>] schedule+0x37/0x80
[14242.773656] [<ffffffffa057ff07>] btrfs_tree_lock+0xa7/0x1f0 [btrfs]
[14242.773664] [<ffffffff810db7c0>] ? prepare_to_wait_event+0xf0/0xf0
[14242.773692] [<ffffffffa0519c44>] btrfs_lock_root_node+0x34/0x50 [btrfs]
[14242.773720] [<ffffffffa051ef6b>] btrfs_search_slot+0x88b/0xa00 [btrfs]
[14242.773750] [<ffffffffa0520a06>] btrfs_insert_empty_items+0x66/0xc0 [btrfs]
[14242.773758] [<ffffffff811ef4a2>] ? kmem_cache_alloc+0x1d2/0x200
[14242.773786] [<ffffffffa0520ad1>] btrfs_insert_item+0x71/0xf0 [btrfs]
[14242.773818] [<ffffffffa052f292>] btrfs_create_pending_block_groups+0x102/0x200 [btrfs]
[14242.773850] [<ffffffffa052f96e>] do_chunk_alloc+0x2ae/0x2f0 [btrfs]
[14242.773934] [<ffffffffa0532825>] find_free_extent+0xa55/0xd90 [btrfs]
[14242.773998] [<ffffffffa0532c22>] btrfs_reserve_extent+0xc2/0x1d0 [btrfs]
[14242.774041] [<ffffffffa0532e38>] btrfs_alloc_tree_block+0x108/0x4a0 [btrfs]
[14242.774078] [<ffffffffa051a5cd>] __btrfs_cow_block+0x12d/0x5b0 [btrfs]
[14242.774118] [<ffffffffa051abff>] btrfs_cow_block+0x11f/0x1c0 [btrfs]
[14242.774155] [<ffffffffa051e8c7>] btrfs_search_slot+0x1e7/0xa00 [btrfs]
[14242.774194] [<ffffffffa0528021>] ? __btrfs_free_extent.isra.70+0x2e1/0xcb0 [btrfs]
[14242.774235] [<ffffffffa0520a06>] btrfs_insert_empty_items+0x66/0xc0 [btrfs]
[14242.774274] [<ffffffffa051994a>] ? btrfs_alloc_path+0x1a/0x20 [btrfs]
[14242.774318] [<ffffffffa052c433>] __btrfs_run_delayed_refs+0xbb3/0x1020 [btrfs]
[14242.774358] [<ffffffffa052f404>] btrfs_run_delayed_refs.part.78+0x74/0x280 [btrfs]
[14242.774391] [<ffffffffa052f627>] btrfs_run_delayed_refs+0x17/0x20 [btrfs]
[14242.774432] [<ffffffffa05be236>] commit_cowonly_roots+0x8d/0x2bd [btrfs]
[14242.774474] [<ffffffffa059d07f>] ? __btrfs_run_delayed_items+0x1cf/0x210 [btrfs]
[14242.774516] [<ffffffffa05adac3>] ? btrfs_qgroup_account_extents+0x83/0x130 [btrfs]
[14242.774558] [<ffffffffa0544c40>] btrfs_commit_transaction+0x590/0xb40 [btrfs]
[14242.774599] [<ffffffffa0589b9d>] ? btrfs_log_dentry_safe+0x6d/0x80 [btrfs]
[14242.774642] [<ffffffffa055ac54>] btrfs_sync_file+0x294/0x350 [btrfs]
[14242.774650] [<ffffffff8123e29b>] vfs_fsync_range+0x3b/0xa0
[14242.774657] [<ffffffff81023891>] ? syscall_trace_enter_phase1+0x131/0x180
[14242.774663] [<ffffffff8123e35d>] do_fsync+0x3d/0x70
[14242.774669] [<ffffffff81023bb8>] ? syscall_trace_leave+0xb8/0x110
[14242.774675] [<ffffffff8123e600>] SyS_fsync+0x10/0x20
[14242.774681] [<ffffffff8175de6e>] entry_SYSCALL_64_fastpath+0x12/0x71
Fix this by never recursing into the finalization phase of block group
creation and making sure we never trigger the finalization of block group
creation while running delayed references.
Reported-by: Josef Bacik <jbacik@fb.com>
Fixes: 00d80e342c ("Btrfs: fix quick exhaustion of the system array in the superblock")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
btrfs_error() and btrfs_std_error() does the same thing
and calls _btrfs_std_error(), so consolidate them together.
And the main motivation is that btrfs_error() is closely
named with btrfs_err(), one handles error action the other
is to log the error, so don't closely name them.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Suggested-by: David Sterba <dsterba@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
When dropping a snapshot we need to account for the qgroup changes. If we drop
the snapshot in all one go then the backref code will fail to find blocks from
the snapshot we dropped since it won't be able to find the root in the fs root
cache. This can lead to us failing to find refs from other roots that pointed
at blocks in the now deleted root. To handle this we need to not remove the fs
roots from the cache until after we process the qgroup operations. Do this by
adding dropped roots to a list on the transaction, and letting the transaction
remove the roots at the same time it drops the commit roots. This will keep all
of the backref searching code in sync properly, and fixes a problem Mark was
seeing with snapshot delete and qgroups. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Tested-by: Holger Hoffstätte <holger.hoffstaette@googlemail.com>
Signed-off-by: Chris Mason <clm@fb.com>
Commit 2e6e518335 ("Btrfs: fix block group ->space_info null pointer
dereference") accidently marked a space info as full when initializing
it with a value of 0 total bytes. This introduces an ENOSPC problem when
writing file data if we mount a filesystem that has no data block groups
allocated, because the data space info is initialized with 0 total bytes,
marked as full, and it never gets its total bytes incremented by a
(positive) value to unmark it as full (because there are no data block
groups loaded when the fs is mounted).
For metadata and system spaces this issue can never happen since we always
have at least one metadata block group and one system block group (even
for an empty filesystem).
So fix this by just not initializing a space info as full, reverting the
offending part of the commit mentioned above.
The following test case for fstests reproduces the issue:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
tmp=/tmp/$$
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
# real QA test starts here
_need_to_be_root
_supported_fs btrfs
_supported_os Linux
_require_scratch
rm -f $seqres.full
_scratch_mkfs >>$seqres.full 2>&1
# Mount our filesystem without space caches enabled so that we do not
# get any space used from the initial data block group that mkfs creates
# (space caches used space from data block groups).
_scratch_mount "-o nospace_cache"
# Need an fs with at least 2Gb to make sure mkfs.btrfs does not create
# an fs using mixed block groups (used both for data and metadata). We
# really need to have dedicated block groups for data to reproduce the
# issue and mkfs.btrfs defaults to mixed block groups only for small
# filesystems (up to 1Gb).
_require_fs_space $SCRATCH_MNT $((2 * 1024 * 1024))
# Run balance with the purpose of deleting the unused data block group
# that mkfs created. We could also wait for the background kthread to
# automatically delete the unused block group, but we do not have a way
# to make it run and wait for it to complete, so just do a balance
# instead of some unreliable sleep
_run_btrfs_util_prog balance start -dusage=0 $SCRATCH_MNT
# Now unmount the filesystem, mount it again (either with or with space
# caches enabled, it does not matter to trigger the problem) and attempt
# to create a file with some data - this used to fail with ENOSPC
# because there were no data block groups when the filesystem was
# mounted and the data space info object was marked as full when
# initialized (because it had 0 total bytes), which prevented the file
# write path from attempting to allocate a data block group and fail
# immediately with ENOSPC.
_scratch_remount
echo "hello world" > $SCRATCH_MNT/foobar
echo "Silence is golden"
status=0
exit
Signed-off-by: Filipe Manana <fdmanana@suse.com>
These wrong comment was copyed from another function(expired) from
init, this patch fixed them.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
More than one code call set_block_group_ro() and restore rw in fail.
Old code use bool bit to save blockgroup's ro state, it can not
support parallel case(it is confirmd exist in my debug log).
This patch use ref count to store ro state, and rename
set_block_group_ro/set_block_group_rw
to
inc_block_group_ro/dec_block_group_ro.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we clear the dirty bits in btrfs_delete_unused_bgs for extents
in the empty block group, it results in btrfs_finish_extent_commit being
unable to discard the freed extents.
The block group removal patch added an alternate path to forget extents
other than btrfs_finish_extent_commit. As a result, any extents that
would be freed when the block group is removed aren't discarded. In my
test run, with a large copy of mixed sized files followed by removal, it
left nearly 2/3 of extents undiscarded.
To clean up the block groups, we add the removed block group onto a list
that will be discarded after transaction commit.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Since we now clean up block groups automatically as they become
empty, iterating over block groups is no longer sufficient to discard
unused space.
This patch iterates over the unused chunk space and discards any regions
that are unallocated, regardless of whether they were ever used. This is
a change for btrfs but is consistent with other file systems.
We do this in a transactionless manner since the discard process can take
a substantial amount of time and a transaction would need to be started
before the acquisition of the device list lock. That would mean a
transaction would be held open across /all/ of the discards collectively.
In order to prevent other threads from allocating or freeing chunks, we
hold the chunks lock across the search and discard calls. We release it
between searches to allow the file system to perform more-or-less
normally. Since the running transaction can commit and disappear while
we're using the transaction pointer, we take a reference to it and
release it after the search. This is safe since it would happen normally
at the end of the transaction commit after any locks are released anyway.
We also take the commit_root_sem to protect against a transaction starting
and committing while we're running.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Btrfs doesn't track superblocks with extent records so there is nothing
persistent on-disk to indicate that those blocks are in use. We track
the superblocks in memory to ensure they don't get used by removing them
from the free space cache when we load a block group from disk. Prior
to 47ab2a6c6a (Btrfs: remove empty block groups automatically), that
was fine since the block group would never be reclaimed so the superblock
was always safe. Once we started removing the empty block groups, we
were protected by the fact that discards weren't being properly issued
for unused space either via FITRIM or -odiscard. The block groups were
still being released, but the blocks remained on disk.
In order to properly discard unused block groups, we need to filter out
the superblocks from the discard range. Superblocks are located at fixed
locations on each device, so it makes sense to filter them out in
btrfs_issue_discard, which is used by both -odiscard and FITRIM.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
It's possible, though unexpected, to pass unaligned offsets and lengths
to btrfs_issue_discard. We then shift the offset/length values to sector
units. If an unaligned offset has been passed, it will result in the
entire sector being discarded, possibly losing data. An unaligned
length is safe but we'll end up returning an inaccurate number of
discarded bytes.
This patch aligns the offset to the 512B boundary, adjusts the length,
and warns, since we shouldn't be discarding on an offset that isn't
aligned with our sector size.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Initially this will just be the length argument passed to it,
but the following patches will adjust that to reflect re-alignment
and skipped blocks.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Omar reported that after commit 4fbcdf6694 ("Btrfs: fix -ENOSPC when
finishing block group creation"), introduced in 4.2-rc1, the following
test was failing due to exhaustion of the system array in the superblock:
#!/bin/bash
truncate -s 100T big.img
mkfs.btrfs big.img
mount -o loop big.img /mnt/loop
num=5
sz=10T
for ((i = 0; i < $num; i++)); do
echo fallocate $i $sz
fallocate -l $sz /mnt/loop/testfile$i
done
btrfs filesystem sync /mnt/loop
for ((i = 0; i < $num; i++)); do
echo rm $i
rm /mnt/loop/testfile$i
btrfs filesystem sync /mnt/loop
done
umount /mnt/loop
This made btrfs_add_system_chunk() fail with -EFBIG due to excessive
allocation of system block groups. This happened because the test creates
a large number of data block groups per transaction and when committing
the transaction we start the writeout of the block group caches for all
the new new (dirty) block groups, which results in pre-allocating space
for each block group's free space cache using the same transaction handle.
That in turn often leads to creation of more block groups, and all get
attached to the new_bgs list of the same transaction handle to the point
of getting a list with over 1500 elements, and creation of new block groups
leads to the need of reserving space in the chunk block reserve and often
creating a new system block group too.
So that made us quickly exhaust the chunk block reserve/system space info,
because as of the commit mentioned before, we do reserve space for each
new block group in the chunk block reserve, unlike before where we would
not and would at most allocate one new system block group and therefore
would only ensure that there was enough space in the system space info to
allocate 1 new block group even if we ended up allocating thousands of
new block groups using the same transaction handle. That worked most of
the time because the computed required space at check_system_chunk() is
very pessimistic (assumes a chunk tree height of BTRFS_MAX_LEVEL/8 and
that all nodes/leafs in a path will be COWed and split) and since the
updates to the chunk tree all happen at btrfs_create_pending_block_groups
it is unlikely that a path needs to be COWed more than once (unless
writepages() for the btree inode is called by mm in between) and that
compensated for the need of creating any new nodes/leads in the chunk
tree.
So fix this by ensuring we don't accumulate a too large list of new block
groups in a transaction's handles new_bgs list, inserting/updating the
chunk tree for all accumulated new block groups and releasing the unused
space from the chunk block reserve whenever the list becomes sufficiently
large. This is a generic solution even though the problem currently can
only happen when starting the writeout of the free space caches for all
dirty block groups (btrfs_start_dirty_block_groups()).
Reported-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Tested-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we have an extent that got N references removed and N new references
added in the same transaction, we must run the insertion of the references
first because otherwise the last removed reference will remove the extent
item from the extent tree, resulting in a failure for the insertions.
This is a regression introduced in the 4.2-rc1 release and this fix just
brings back the behaviour of selecting reference additions before any
reference removals.
The following test case for fstests reproduces the issue:
seq=`basename $0`
seqres=$RESULT_DIR/$seq
echo "QA output created by $seq"
tmp=/tmp/$$
status=1 # failure is the default!
trap "_cleanup; exit \$status" 0 1 2 3 15
_cleanup()
{
_cleanup_flakey
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
. ./common/dmflakey
# real QA test starts here
_need_to_be_root
_supported_fs btrfs
_supported_os Linux
_require_scratch
_require_dm_flakey
_require_cloner
_require_metadata_journaling $SCRATCH_DEV
rm -f $seqres.full
_scratch_mkfs >>$seqres.full 2>&1
_init_flakey
_mount_flakey
# Create prealloc extent covering range [160K, 620K[
$XFS_IO_PROG -f -c "falloc 160K 460K" $SCRATCH_MNT/foo
# Now write to the last 80K of the prealloc extent plus 40K to the unallocated
# space that immediately follows it. This creates a new extent of 40K that spans
# the range [620K, 660K[.
$XFS_IO_PROG -c "pwrite -S 0xaa 540K 120K" $SCRATCH_MNT/foo | _filter_xfs_io
# At this point, there are now 2 back references to the prealloc extent in our
# extent tree. Both are for our file offset 160K and one relates to a file
# extent item with a data offset of 0 and a length of 380K, while the other
# relates to a file extent item with a data offset of 380K and a length of 80K.
# Make sure everything done so far is durably persisted (all back references are
# in the extent tree, etc).
sync
# Now clone all extents of our file that cover the offset 160K up to its eof
# (660K at this point) into itself at offset 2M. This leaves a hole in the file
# covering the range [660K, 2M[. The prealloc extent will now be referenced by
# the file twice, once for offset 160K and once for offset 2M. The 40K extent
# that follows the prealloc extent will also be referenced twice by our file,
# once for offset 620K and once for offset 2M + 460K.
$CLONER_PROG -s $((160 * 1024)) -d $((2 * 1024 * 1024)) -l 0 $SCRATCH_MNT/foo \
$SCRATCH_MNT/foo
# Now create one new extent in our file with a size of 100Kb. It will span the
# range [3M, 3M + 100K[. It also will cause creation of a hole spanning the
# range [2M + 460K, 3M[. Our new file size is 3M + 100K.
$XFS_IO_PROG -c "pwrite -S 0xbb 3M 100K" $SCRATCH_MNT/foo | _filter_xfs_io
# At this point, there are now (in memory) 4 back references to the prealloc
# extent.
#
# Two of them are for file offset 160K, related to file extent items
# matching the file offsets 160K and 540K respectively, with data offsets of
# 0 and 380K respectively, and with lengths of 380K and 80K respectively.
#
# The other two references are for file offset 2M, related to file extent items
# matching the file offsets 2M and 2M + 380K respectively, with data offsets of
# 0 and 380K respectively, and with lengths of 389K and 80K respectively.
#
# The 40K extent has 2 back references, one for file offset 620K and the other
# for file offset 2M + 460K.
#
# The 100K extent has a single back reference and it relates to file offset 3M.
# Now clone our 100K extent into offset 600K. That offset covers the last 20K
# of the prealloc extent, the whole 40K extent and 40K of the hole starting at
# offset 660K.
$CLONER_PROG -s $((3 * 1024 * 1024)) -d $((600 * 1024)) -l $((100 * 1024)) \
$SCRATCH_MNT/foo $SCRATCH_MNT/foo
# At this point there's only one reference to the 40K extent, at file offset
# 2M + 460K, we have 4 references for the prealloc extent (2 for file offset
# 160K and 2 for file offset 2M) and 2 references for the 100K extent (1 for
# file offset 3M and a new one for file offset 600K).
# Now fsync our file to make all its new data and metadata updates are durably
# persisted and present if a power failure/crash happens after a successful
# fsync and before the next transaction commit.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
echo "File digest before power failure:"
md5sum $SCRATCH_MNT/foo | _filter_scratch
# Silently drop all writes and ummount to simulate a crash/power failure.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# Allow writes again, mount to trigger log replay and validate file contents.
# During log replay, the btrfs delayed references implementation used to run the
# deletion of back references before the addition of new back references, which
# made the addition fail as it didn't find the key in the extent tree that it
# was looking for. The failure triggered by this test was related to the 40K
# extent, which got 1 reference dropped and 1 reference added during the fsync
# log replay - when running the delayed references at transaction commit time,
# btrfs was applying the deletion before the insertion, resulting in a failure
# of the insertion that ended up turning the fs into read-only mode.
_load_flakey_table $FLAKEY_ALLOW_WRITES
_mount_flakey
echo "File digest after log replay:"
md5sum $SCRATCH_MNT/foo | _filter_scratch
_unmount_flakey
status=0
exit
This issue turned the filesystem into read-only mode (current transaction
aborted) and produced the following traces:
[ 8247.578385] ------------[ cut here ]------------
[ 8247.579947] WARNING: CPU: 0 PID: 11341 at fs/btrfs/extent-tree.c:1547 lookup_inline_extent_backref+0x17d/0x45d [btrfs]()
(...)
[ 8247.601697] Call Trace:
[ 8247.602222] [<ffffffff8145f077>] dump_stack+0x4f/0x7b
[ 8247.604320] [<ffffffff8104b3b0>] warn_slowpath_common+0xa1/0xbb
[ 8247.605488] [<ffffffffa0506c8d>] ? lookup_inline_extent_backref+0x17d/0x45d [btrfs]
[ 8247.608226] [<ffffffffa0506c8d>] lookup_inline_extent_backref+0x17d/0x45d [btrfs]
[ 8247.617061] [<ffffffffa0507957>] insert_inline_extent_backref+0x41/0xb2 [btrfs]
[ 8247.621856] [<ffffffffa0507c4f>] __btrfs_inc_extent_ref+0x8c/0x20a [btrfs]
[ 8247.624366] [<ffffffffa050ee60>] __btrfs_run_delayed_refs+0xb0c/0xd49 [btrfs]
[ 8247.626176] [<ffffffffa0510dcd>] btrfs_run_delayed_refs+0x6d/0x1d4 [btrfs]
[ 8247.627435] [<ffffffff81155c9b>] ? __cache_free+0x4a7/0x4b6
[ 8247.628531] [<ffffffffa0520482>] btrfs_commit_transaction+0x4c/0xa20 [btrfs]
(...)
[ 8247.648430] ---[ end trace 2461e55f92c2ac2d ]---
[ 8247.727263] WARNING: CPU: 3 PID: 11341 at fs/btrfs/extent-tree.c:2771 btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs]()
[ 8247.728954] BTRFS: Transaction aborted (error -5)
(...)
[ 8247.760866] Call Trace:
[ 8247.761534] [<ffffffff8145f077>] dump_stack+0x4f/0x7b
[ 8247.764271] [<ffffffff8104b3b0>] warn_slowpath_common+0xa1/0xbb
[ 8247.767582] [<ffffffffa0510e04>] ? btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs]
[ 8247.769373] [<ffffffff8104b410>] warn_slowpath_fmt+0x46/0x48
[ 8247.770836] [<ffffffffa0510e04>] btrfs_run_delayed_refs+0xa4/0x1d4 [btrfs]
[ 8247.772532] [<ffffffff81155c9b>] ? __cache_free+0x4a7/0x4b6
[ 8247.773664] [<ffffffffa0520482>] btrfs_commit_transaction+0x4c/0xa20 [btrfs]
[ 8247.775047] [<ffffffff81087310>] ? trace_hardirqs_on+0xd/0xf
[ 8247.776176] [<ffffffff81155dd5>] ? kmem_cache_free+0x12b/0x189
[ 8247.777427] [<ffffffffa055a920>] btrfs_recover_log_trees+0x2da/0x33d [btrfs]
[ 8247.778575] [<ffffffffa055898e>] ? replay_one_extent+0x4fc/0x4fc [btrfs]
[ 8247.779838] [<ffffffffa051e265>] open_ctree+0x1cc0/0x201a [btrfs]
[ 8247.781020] [<ffffffff81120f48>] ? register_shrinker+0x56/0x81
[ 8247.782285] [<ffffffffa04fb12c>] btrfs_mount+0x5f0/0x734 [btrfs]
(...)
[ 8247.793394] ---[ end trace 2461e55f92c2ac2e ]---
[ 8247.794276] BTRFS: error (device dm-0) in btrfs_run_delayed_refs:2771: errno=-5 IO failure
[ 8247.797335] BTRFS: error (device dm-0) in btrfs_replay_log:2375: errno=-5 IO failure (Failed to recover log tree)
Fixes: c6fc245499 ("btrfs: delayed-ref: Use list to replace the ref_root in ref_head.")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Acked-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
btrfs will report no_space when we run following write and delete
file loop:
# FILE_SIZE_M=[ 75% of fs space ]
# DEV=[ some dev ]
# MNT=[ some dir ]
#
# mkfs.btrfs -f "$DEV"
# mount -o nodatacow "$DEV" "$MNT"
# for ((i = 0; i < 100; i++)); do dd if=/dev/zero of="$MNT"/file0 bs=1M count="$FILE_SIZE_M"; rm -f "$MNT"/file0; done
#
Reason:
iput() and evict() is run after write pages to block device, if
write pages work is not finished before next write, the "rm"ed space
is not freed, and caused above bug.
Fix:
We can add "-o flushoncommit" mount option to avoid above bug, but
it have performance problem. Actually, we can to wait for on-the-fly
writes only when no-space happened, it is which this patch do.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Switch from old ref_node based qgroup to extent based qgroup mechanism
for normal operations.
The new mechanism should hugely reduce the overhead of btrfs quota
system, and further more, the codes and logic should be more clean and
easier to maintain.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
__btrfs_inc_extent_ref() and __btrfs_free_extent() have already had too
many parameters, but three of them can be extracted from
btrfs_delayed_ref_node struct.
So use btrfs_delayed_ref_node struct as a single parameter to replace
the bytenr/num_byte/no_quota parameters.
The real objective of this patch is to allow btrfs_qgroup_record_ref()
get the delayed_ref_node in incoming qgroup patches.
Other functions calling btrfs_qgroup_record_ref() are not affected since
the rest will only add/sub exclusive extents, where node is not used.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
This patch replace the rbtree used in ref_head to list.
This has the following advantage:
1) Easier merge logic.
With the new list implement, we only need to care merging the tail
ref_node with the new ref_node.
And this can be done quite easy at insert time, no need to do a
indicated merge at run_delayed_refs().
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
When allocating a new chunk or removing one we need to update num_devs
device items and insert or remove a chunk item in the chunk tree, so
in the worst case the space needed in the chunk space_info is:
btrfs_calc_trunc_metadata_size(chunk_root, num_devs) +
btrfs_calc_trans_metadata_size(chunk_root, 1)
That is, in the worst case we need to cow num_devs paths and cow 1 other
path that can result in splitting every node and leaf, and each path
consisting of BTRFS_MAX_LEVEL - 1 nodes and 1 leaf. We were requiring
some additional chunk_root->nodesize * BTRFS_MAX_LEVEL * num_devs bytes,
which were unnecessary since updating the existing device items does
not result in splitting the nodes and leaf since after updating them
they remain with the same size.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
The return value of read_tree_block() can confuse callers as it always
returns NULL for either -ENOMEM or -EIO, so it's likely that callers
parse it to a wrong error, for instance, in btrfs_read_tree_root().
This fixes the above issue.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
read_tree_block may take a reference on the 'eb', a following
free_extent_buffer is necessary.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
Unlike when attempting to allocate a new block group, where we check
that we have enough space in the system space_info to update the device
items and insert a new chunk item in the chunk tree, we were not checking
if the system space_info had enough space for updating the device items
and deleting the chunk item in the chunk tree. This often lead to -ENOSPC
error when attempting to allocate blocks for the chunk tree (during btree
node/leaf COW operations) while updating the device items or deleting the
chunk item, which resulted in the current transaction being aborted and
turning the filesystem into read-only mode.
While running fstests generic/038, which stresses allocation of block
groups and removal of unused block groups, with a large scratch device
(750Gb) this happened often, despite more than enough unallocated space,
and resulted in the following trace:
[68663.586604] WARNING: CPU: 3 PID: 1521 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x114 [btrfs]()
[68663.600407] BTRFS: Transaction aborted (error -28)
(...)
[68663.730829] Call Trace:
[68663.732585] [<ffffffff8142fa46>] dump_stack+0x4f/0x7b
[68663.734334] [<ffffffff8108b6a2>] ? console_unlock+0x361/0x3ad
[68663.739980] [<ffffffff81045ea5>] warn_slowpath_common+0xa1/0xbb
[68663.757153] [<ffffffffa036ca6d>] ? __btrfs_abort_transaction+0x52/0x114 [btrfs]
[68663.760925] [<ffffffff81045f05>] warn_slowpath_fmt+0x46/0x48
[68663.762854] [<ffffffffa03b159d>] ? btrfs_update_device+0x15a/0x16c [btrfs]
[68663.764073] [<ffffffffa036ca6d>] __btrfs_abort_transaction+0x52/0x114 [btrfs]
[68663.765130] [<ffffffffa03b3638>] btrfs_remove_chunk+0x597/0x5ee [btrfs]
[68663.765998] [<ffffffffa0384663>] ? btrfs_delete_unused_bgs+0x245/0x296 [btrfs]
[68663.767068] [<ffffffffa0384676>] btrfs_delete_unused_bgs+0x258/0x296 [btrfs]
[68663.768227] [<ffffffff8143527f>] ? _raw_spin_unlock_irq+0x2d/0x4c
[68663.769081] [<ffffffffa038b109>] cleaner_kthread+0x13d/0x16c [btrfs]
[68663.799485] [<ffffffffa038afcc>] ? btrfs_alloc_root+0x28/0x28 [btrfs]
[68663.809208] [<ffffffff8105f367>] kthread+0xef/0xf7
[68663.828795] [<ffffffff810e603f>] ? time_hardirqs_on+0x15/0x28
[68663.844942] [<ffffffff8105f278>] ? __kthread_parkme+0xad/0xad
[68663.846486] [<ffffffff81435a88>] ret_from_fork+0x58/0x90
[68663.847760] [<ffffffff8105f278>] ? __kthread_parkme+0xad/0xad
[68663.849503] ---[ end trace 798477c6d6dbaad6 ]---
[68663.850525] BTRFS: error (device sdc) in btrfs_remove_chunk:2652: errno=-28 No space left
So fix this by verifying that enough space exists in system space_info,
and reserving the space in the chunk block reserve, before attempting to
delete the block group and allocate a new system chunk if we don't have
enough space to perform the necessary updates and delete in the chunk
tree. Like for the block group creation case, we don't error our if we
fail to allocate a new system chunk, since we might end up not needing
it (no node/leaf splits happen during the COW operations and/or we end
up not needing to COW any btree nodes or leafs because they were already
COWed in the current transaction and their writeback didn't start yet).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
While creating a block group, we often end up getting ENOSPC while updating
the chunk tree, which leads to a transaction abortion that produces a trace
like the following:
[30670.116368] WARNING: CPU: 4 PID: 20735 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x106 [btrfs]()
[30670.117777] BTRFS: Transaction aborted (error -28)
(...)
[30670.163567] Call Trace:
[30670.163906] [<ffffffff8142fa46>] dump_stack+0x4f/0x7b
[30670.164522] [<ffffffff8108b6a2>] ? console_unlock+0x361/0x3ad
[30670.165171] [<ffffffff81045ea5>] warn_slowpath_common+0xa1/0xbb
[30670.166323] [<ffffffffa035daa7>] ? __btrfs_abort_transaction+0x52/0x106 [btrfs]
[30670.167213] [<ffffffff81045f05>] warn_slowpath_fmt+0x46/0x48
[30670.167862] [<ffffffffa035daa7>] __btrfs_abort_transaction+0x52/0x106 [btrfs]
[30670.169116] [<ffffffffa03743d7>] btrfs_create_pending_block_groups+0x101/0x130 [btrfs]
[30670.170593] [<ffffffffa038426a>] __btrfs_end_transaction+0x84/0x366 [btrfs]
[30670.171960] [<ffffffffa038455c>] btrfs_end_transaction+0x10/0x12 [btrfs]
[30670.174649] [<ffffffffa036eb6b>] btrfs_check_data_free_space+0x11f/0x27c [btrfs]
[30670.176092] [<ffffffffa039450d>] btrfs_fallocate+0x7c8/0xb96 [btrfs]
[30670.177218] [<ffffffff812459f2>] ? __this_cpu_preempt_check+0x13/0x15
[30670.178622] [<ffffffff81152447>] vfs_fallocate+0x14c/0x1de
[30670.179642] [<ffffffff8116b915>] ? __fget_light+0x2d/0x4f
[30670.180692] [<ffffffff81152863>] SyS_fallocate+0x47/0x62
[30670.186737] [<ffffffff81435b32>] system_call_fastpath+0x12/0x17
[30670.187792] ---[ end trace 0373e6b491c4a8cc ]---
This is because we don't do proper space reservation for the chunk block
reserve when we have multiple tasks allocating chunks in parallel.
So block group creation has 2 phases, and the first phase essentially
checks if there is enough space in the system space_info, allocating a
new system chunk if there isn't, while the second phase updates the
device, extent and chunk trees. However, because the updates to the
chunk tree happen in the second phase, if we have N tasks, each with
its own transaction handle, allocating new chunks in parallel and if
there is only enough space in the system space_info to allocate M chunks,
where M < N, none of the tasks ends up allocating a new system chunk in
the first phase and N - M tasks will get -ENOSPC when attempting to
update the chunk tree in phase 2 if they need to COW any nodes/leafs
from the chunk tree.
Fix this by doing proper reservation in the chunk block reserve.
The issue could be reproduced by running fstests generic/038 in a loop,
which eventually triggered the problem.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Commit 2f0810880f changed
btrfs_set_block_group_ro to avoid trying to allocate new chunks with the
new raid profile during conversion. This fixed failures when there was
no space on the drive to allocate a new chunk, but the metadata
reserves were sufficient to continue the conversion.
But this ended up causing a regression when the drive had plenty of
space to allocate new chunks, mostly because reduce_alloc_profile isn't
using the new raid profile.
Fixing btrfs_reduce_alloc_profile is a bigger patch. For now, do a
partial revert of 2f0810880, and don't error out if we hit ENOSPC.
Signed-off-by: Chris Mason <clm@fb.com>
Tested-by: Dave Sterba <dsterba@suse.cz>
Reported-by: Holger Hoffstaette <holger.hoffstaette@googlemail.com>
If while setting a block group read-only we end up allocating a system
chunk, through check_system_chunk(), we were not doing it while holding
the chunk mutex which is a problem if a concurrent chunk allocation is
happening, through do_chunk_alloc(), as it means both block groups can
end up using the same logical addresses and physical regions in the
device(s). So make sure we hold the chunk mutex.
Cc: stable@vger.kernel.org # 4.0+
Fixes: 2f0810880f ("btrfs: delete chunk allocation attemp when
setting block group ro")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
So creating a block group has 2 distinct phases:
Phase 1 - creates the btrfs_block_group_cache item and adds it to the
rbtree fs_info->block_group_cache_tree and to the corresponding list
space_info->block_groups[];
Phase 2 - adds the block group item to the extent tree and corresponding
items to the chunk tree.
The first phase adds the block_group_cache_item to a list of pending block
groups in the transaction handle, and phase 2 happens when
btrfs_end_transaction() is called against the transaction handle.
It happens that once phase 1 completes, other concurrent tasks that use
their own transaction handle, but points to the same running transaction
(struct btrfs_trans_handle->transaction), can use this block group for
space allocations and therefore mark it dirty. Dirty block groups are
tracked in a list belonging to the currently running transaction (struct
btrfs_transaction) and not in the transaction handle (btrfs_trans_handle).
This is a problem because once a task calls btrfs_commit_transaction(),
it calls btrfs_start_dirty_block_groups() which will see all dirty block
groups and attempt to start their writeout, including those that are
still attached to the transaction handle of some concurrent task that
hasn't called btrfs_end_transaction() yet - which means those block
groups haven't gone through phase 2 yet and therefore when
write_one_cache_group() is called, it won't find the block group items
in the extent tree and abort the current transaction with -ENOENT,
turning the fs into readonly mode and require a remount.
Fix this by ignoring -ENOENT when looking for block group items in the
extent tree when we attempt to start the writeout of the block group
caches outside the critical section of the transaction commit. We will
try again later during the critical section and if there we still don't
find the block group item in the extent tree, we then abort the current
transaction.
This issue happened twice, once while running fstests btrfs/067 and once
for btrfs/078, which produced the following trace:
[ 3278.703014] WARNING: CPU: 7 PID: 18499 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x114 [btrfs]()
[ 3278.707329] BTRFS: Transaction aborted (error -2)
(...)
[ 3278.731555] Call Trace:
[ 3278.732396] [<ffffffff8142fa46>] dump_stack+0x4f/0x7b
[ 3278.733860] [<ffffffff8108b6a2>] ? console_unlock+0x361/0x3ad
[ 3278.735312] [<ffffffff81045ea5>] warn_slowpath_common+0xa1/0xbb
[ 3278.736874] [<ffffffffa03ada6d>] ? __btrfs_abort_transaction+0x52/0x114 [btrfs]
[ 3278.738302] [<ffffffff81045f05>] warn_slowpath_fmt+0x46/0x48
[ 3278.739520] [<ffffffffa03ada6d>] __btrfs_abort_transaction+0x52/0x114 [btrfs]
[ 3278.741222] [<ffffffffa03b9e56>] write_one_cache_group+0xae/0xbf [btrfs]
[ 3278.742797] [<ffffffffa03c487b>] btrfs_start_dirty_block_groups+0x170/0x2b2 [btrfs]
[ 3278.744492] [<ffffffffa03d309c>] btrfs_commit_transaction+0x130/0x9c9 [btrfs]
[ 3278.746084] [<ffffffff8107d33d>] ? trace_hardirqs_on+0xd/0xf
[ 3278.747249] [<ffffffffa03e5660>] btrfs_sync_file+0x313/0x387 [btrfs]
[ 3278.748744] [<ffffffff8117acad>] vfs_fsync_range+0x95/0xa4
[ 3278.749958] [<ffffffff81435b54>] ? ret_from_sys_call+0x1d/0x58
[ 3278.751218] [<ffffffff8117acd8>] vfs_fsync+0x1c/0x1e
[ 3278.754197] [<ffffffff8117ae54>] do_fsync+0x34/0x4e
[ 3278.755192] [<ffffffff8117b07c>] SyS_fsync+0x10/0x14
[ 3278.756236] [<ffffffff81435b32>] system_call_fastpath+0x12/0x17
[ 3278.757366] ---[ end trace 9a4d4df4969709aa ]---
Fixes: 1bbc621ef2 ("Btrfs: allow block group cache writeout
outside critical section in commit")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
This is one of the first places to give out when memory is tight. Handle
it properly rather than with a BUG_ON.
Also fix the comment about the return value, which is an ERR_PTR, not
NULL, on error.
Signed-off-by: Omar Sandoval <osandov@osandov.com>
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
Now that we're doing free space cache writeback outside the critical
section in the commit, there is a bigger window for delalloc_bytes to
be added after a cache has been written. find_free_extent may do this
without putting the block group back into the dirty list, and also
without a transaction running.
Checking for delalloc_bytes in cache_save_setup means we might leave the
cache marked as written without invalidating it. Consistency checks
during mount will toss the cache, but it's better to get rid of the
check in cache_save_setup and let it get invalidated by the checks
already done during cache write out.
Signed-off-by: Chris Mason <clm@fb.com>
While running xfstests I ran into the following:
[20892.242791] ------------[ cut here ]------------
[20892.243776] WARNING: CPU: 0 PID: 13299 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x52/0x114 [btrfs]()
[20892.245874] BTRFS: Transaction aborted (error -2)
[20892.247329] Modules linked in: btrfs dm_snapshot dm_bufio dm_flakey dm_mod crc32c_generic xor raid6_pq nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace fscache sunrpc loop fuse$
[20892.258488] CPU: 0 PID: 13299 Comm: fsstress Tainted: G W 4.0.0-rc5-btrfs-next-9+ #2
[20892.262011] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014
[20892.264738] 0000000000000009 ffff880427f8bc18 ffffffff8142fa46 ffffffff8108b6a2
[20892.266244] ffff880427f8bc68 ffff880427f8bc58 ffffffff81045ea5 ffff880427f8bc48
[20892.267761] ffffffffa0509a6d 00000000fffffffe ffff8803545d6f40 ffffffffa05a15a0
[20892.269378] Call Trace:
[20892.269915] [<ffffffff8142fa46>] dump_stack+0x4f/0x7b
[20892.271097] [<ffffffff8108b6a2>] ? console_unlock+0x361/0x3ad
[20892.272173] [<ffffffff81045ea5>] warn_slowpath_common+0xa1/0xbb
[20892.273386] [<ffffffffa0509a6d>] ? __btrfs_abort_transaction+0x52/0x114 [btrfs]
[20892.274857] [<ffffffff81045f05>] warn_slowpath_fmt+0x46/0x48
[20892.275851] [<ffffffffa0509a6d>] __btrfs_abort_transaction+0x52/0x114 [btrfs]
[20892.277341] [<ffffffffa0515e10>] write_one_cache_group+0x68/0xaf [btrfs]
[20892.278628] [<ffffffffa052088a>] btrfs_start_dirty_block_groups+0x18d/0x29b [btrfs]
[20892.280191] [<ffffffffa052f077>] btrfs_commit_transaction+0x130/0x9c9 [btrfs]
[20892.281781] [<ffffffff8107d33d>] ? trace_hardirqs_on+0xd/0xf
[20892.282873] [<ffffffffa054163b>] btrfs_sync_file+0x313/0x387 [btrfs]
[20892.284111] [<ffffffff8117acad>] vfs_fsync_range+0x95/0xa4
[20892.285203] [<ffffffff810e603f>] ? time_hardirqs_on+0x15/0x28
[20892.286290] [<ffffffff8123960b>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[20892.287469] [<ffffffff8117acd8>] vfs_fsync+0x1c/0x1e
[20892.288412] [<ffffffff8117ae54>] do_fsync+0x34/0x4e
[20892.289348] [<ffffffff8117b07c>] SyS_fsync+0x10/0x14
[20892.290255] [<ffffffff81435b32>] system_call_fastpath+0x12/0x17
[20892.291316] ---[ end trace 597f77e664245373 ]---
[20892.293955] BTRFS: error (device sdg) in write_one_cache_group:3184: errno=-2 No such entry
[20892.297390] BTRFS info (device sdg): forced readonly
This happens because in btrfs_start_dirty_block_groups() we splice the
transaction's list of dirty block groups into a local list and then we
keep extracting the first element of the list without holding the
cache_write_mutex mutex. This means that before we acquire that mutex
the first block group on the list might be removed by a conurrent task
running btrfs_remove_block_group(). So make sure we extract the first
element (and test the list emptyness) while holding that mutex.
Fixes: 1bbc621ef2 ("Btrfs: allow block group cache writeout
outside critical section in commit")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
There are two problems in qgroup:
a). The PAGE_CACHE is 4K, even when we are writing a data of 1K,
qgroup will reserve a 4K size. It will cause the last 3K in a qgroup
is not available to user.
b). When user is writing a inline data, qgroup will not reserve it,
it means this is a window we can exceed the limit of a qgroup.
The main idea of this patch is reserving the data size of write_bytes
rather than the reserve_bytes. It means qgroup will not care about
the data size btrfs will reserve for user, but only care about the
data size user is going to write. Then reserve it when user want to
write and release it in transaction committed.
In this way, qgroup can be released from the complex procedure in
btrfs and only do the reserve when user want to write and account
when the data is written in commit_transaction().
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currenly, in data writing, ->reserved is accounted in
fill_delalloc(), but ->may_use is released in clear_bit_hook()
which is called by btrfs_finish_ordered_io(). That's too late,
that said, between fill_delalloc() and btrfs_finish_ordered_io(),
the data is doublely accounted by qgroup. It will cause some
unexpected -EDQUOT.
Example:
# btrfs quota enable /root/btrfs-auto-test/
# btrfs subvolume create /root/btrfs-auto-test//sub
Create subvolume '/root/btrfs-auto-test/sub'
# btrfs qgroup limit 1G /root/btrfs-auto-test//sub
dd if=/dev/zero of=/root/btrfs-auto-test//sub/file bs=1024 count=1500000
dd: error writing '/root/btrfs-auto-test//sub/file': Disk quota exceeded
681353+0 records in
681352+0 records out
697704448 bytes (698 MB) copied, 8.15563 s, 85.5 MB/s
It's (698 MB) when we got an -EDQUOT, but we limit it by 1G.
This patch move the btrfs_qgroup_reserve/free() for data from
btrfs_delalloc_reserve/release_metadata() to btrfs_check_data_free_space()
and btrfs_free_reserved_data_space(). Then the accounter in qgroup
will be updated at the same time with the accounter in space_info updated.
In this way, the unexpected -EDQUOT will be killed.
Reported-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently, for pre_alloc or delay_alloc, the bytes will be accounted
in space_info by the three guys.
space_info->bytes_may_use --- space_info->reserved --- space_info->used.
But on the other hand, in qgroup, there are only two counters to account the
bytes, qgroup->reserved and qgroup->excl. And qg->reserved accounts
bytes in space_info->bytes_may_use and qg->excl accounts bytes in
space_info->used. So the bytes in space_info->reserved is not accounted
in qgroup. If so, there is a window we can exceed the quota limit when
bytes is in space_info->reserved.
Example:
# btrfs quota enable /mnt
# btrfs qgroup limit -e 10M /mnt
# for((i=0;i<20;i++));do fallocate -l 1M /mnt/data$i; done
# sync
# btrfs qgroup show -pcre /mnt
qgroupid rfer excl max_rfer max_excl parent child
-------- ---- ---- -------- -------- ------ -----
0/5 20987904 20987904 0 10485760 --- ---
qg->excl is 20987904 larger than max_excl 10485760.
This patch introduce a new counter named may_use to qgroup, then
there are three counters in qgroup to account bytes in space_info
as below.
space_info->bytes_may_use --- space_info->reserved --- space_info->used.
qgroup->may_use --- qgroup->reserved --- qgroup->excl
With this patch applied:
# btrfs quota enable /mnt
# btrfs qgroup limit -e 10M /mnt
# for((i=0;i<20;i++));do fallocate -l 1M /mnt/data$i; done
fallocate: /mnt/data9: fallocate failed: Disk quota exceeded
fallocate: /mnt/data10: fallocate failed: Disk quota exceeded
fallocate: /mnt/data11: fallocate failed: Disk quota exceeded
fallocate: /mnt/data12: fallocate failed: Disk quota exceeded
fallocate: /mnt/data13: fallocate failed: Disk quota exceeded
fallocate: /mnt/data14: fallocate failed: Disk quota exceeded
fallocate: /mnt/data15: fallocate failed: Disk quota exceeded
fallocate: /mnt/data16: fallocate failed: Disk quota exceeded
fallocate: /mnt/data17: fallocate failed: Disk quota exceeded
fallocate: /mnt/data18: fallocate failed: Disk quota exceeded
fallocate: /mnt/data19: fallocate failed: Disk quota exceeded
# sync
# btrfs qgroup show -pcre /mnt
qgroupid rfer excl max_rfer max_excl parent child
-------- ---- ---- -------- -------- ------ -----
0/5 9453568 9453568 0 10485760 --- ---
Reported-by: Cyril SCETBON <cyril.scetbon@free.fr>
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we exceed quota limit in writing, we will free
some reserved extent when we need to drop but not free
account in qgroup. It means, each time we exceed quota
in writing, there will be some remain space in qg->reserved
we can not use any more. If things go on like this, the
all space will be ate up.
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Reproduce:
while true; do
dd if=/dev/zero of=/mnt/btrfs/file count=[75% fs_size]
rm /mnt/btrfs/file
done
Then we can see above loop failed on NO_SPACE.
It it long-term problem since very beginning, because delayed-iput
after rm are not run.
We already have commit_transaction() in alloc_space code, but it is
not triggered in above case.
This patch trigger commit_transaction() to run delayed-iput and
reflash pinned-space to to make write success.
It is based on previous fix of delayed-iput in commit_transaction(),
need to be applied on top of:
btrfs: Fix NO_SPACE bug caused by delayed-iput
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Steps to reproduce:
while true; do
dd if=/dev/zero of=/btrfs_dir/file count=[fs_size * 75%]
rm /btrfs_dir/file
sync
done
And we'll see dd failed because btrfs return NO_SPACE.
Reason:
Normally, btrfs_commit_transaction() call btrfs_run_delayed_iputs()
in end to free fs space for next write, but sometimes it hadn't
done work on time, because btrfs-cleaner thread get delayed-iputs
from list before, but do iput() after next write.
This is log:
[ 2569.050776] comm=btrfs-cleaner func=btrfs_evict_inode() begin
[ 2569.084280] comm=sync func=btrfs_commit_transaction() call btrfs_run_delayed_iputs()
[ 2569.085418] comm=sync func=btrfs_commit_transaction() done btrfs_run_delayed_iputs()
[ 2569.087554] comm=sync func=btrfs_commit_transaction() end
[ 2569.191081] comm=dd begin
[ 2569.790112] comm=dd func=__btrfs_buffered_write() ret=-28
[ 2569.847479] comm=btrfs-cleaner func=add_pinned_bytes() 0 + 32677888 = 32677888
[ 2569.849530] comm=btrfs-cleaner func=add_pinned_bytes() 32677888 + 23834624 = 56512512
...
[ 2569.903893] comm=btrfs-cleaner func=add_pinned_bytes() 943976448 + 21762048 = 965738496
[ 2569.908270] comm=btrfs-cleaner func=btrfs_evict_inode() end
Fix:
Make btrfs_commit_transaction() wait current running btrfs-cleaner's
delayed-iputs() done in end.
Test:
Use script similar to above(more complex),
before patch:
7 failed in 100 * 20 loop.
after patch:
0 failed in 100 * 20 loop.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
space_info's value calculation is some complex and easy to cause
bug, add WARN_ON() to help debug.
Changelog v1->v2:
Put WARN_ON()s under the ENOSPC_DEBUG mount option.
Suggested by: David Sterba <dsterba@suse.cz>
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Bug1:
space_info->bytes_readonly was set to very large(negative) value in
btrfs_remove_block_group().
Reason:
Current code set block_group_cache->pinned = 0 in btrfs_delete_unused_bgs(),
but above space was not counted to space_info->bytes_readonly.
Then in btrfs_remove_block_group():
block_group->space_info->bytes_readonly -= block_group->key.offset;
We can see following value in trace:
btrfs_remove_block_group: pid=2677 comm=btrfs-cleaner WARNING: bytes_readonly=12582912, key.offset=134217728
Bug2:
space_info->total_bytes_pinned grow to value larger than fs size.
In a 1.2G fs, we can get following trace log:
at first:
ZL_DEBUG: add_pinned_bytes: pid=2710 comm=sync change total_bytes_pinned flags=1 869793792 + 95944704 = 965738496
after some op:
ZL_DEBUG: add_pinned_bytes: pid=2770 comm=sync change total_bytes_pinned flags=1 1780178944 + 95944704 = 1876123648
after some op:
ZL_DEBUG: add_pinned_bytes: pid=3193 comm=sync change total_bytes_pinned flags=1 2924568576 + 95551488 = 3020120064
...
Reason:
Similar to bug1, we also need to adjust space_info->total_bytes_pinned
in above code block.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we have any chance to make a successful write, we should not give up.
This patch adjust commit-transaction condition from:
pinned >= wanted
to
left + pinned >= wanted
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Old code bypass commit transaction when we don't have enough
pinned space, but another case is there exist freed bgs in current
transction, it have possibility to make alloc_chunk success.
This patch modify the condition to:
if (have_free_bg || have_pinned_space) commit_transaction()
Confirmed above action by printk before and after patch.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Near the end of close_ctree, we're calling btrfs_free_block_rsv
to free up the orphan rsv. The problem is this call updates the
space_info, which has already been freed.
This adds a new __ function that directly calls kfree instead of trying
to update the space infos.
Signed-off-by: Chris Mason <clm@fb.com>
We loop through all of the dirty block groups during commit and write
the free space cache. In order to make sure the cache is currect, we do
this while no other writers are allowed in the commit.
If a large number of block groups are dirty, this can introduce long
stalls during the final stages of the commit, which can block new procs
trying to change the filesystem.
This commit changes the block group cache writeout to take appropriate
locks and allow it to run earlier in the commit. We'll still have to
redo some of the block groups, but it means we can get most of the work
out of the way without blocking the entire FS.
Signed-off-by: Chris Mason <clm@fb.com>
Block group cache writeout is currently waiting on the pages for each
block group cache before moving on to writing the next one. This commit
switches things around to send down all the caches and then wait on them
in batches.
The end result is much faster, since we're keeping the disk pipeline
full.
Signed-off-by: Chris Mason <clm@fb.com>
We're triggering a huge number of commits from
btrfs_async_reclaim_metadata_space. These aren't really requried,
because everyone calling the async reclaim code is going to end up
triggering a commit on their own.
Signed-off-by: Chris Mason <clm@fb.com>
When truncate starts, it allocates some space in the block reserves so
that we'll have enough to update metadata along the way.
For very large files, we can easily go through all of that space as we
loop through the extents. This changes truncate to refill the space
reservation as it progresses through the file.
Signed-off-by: Chris Mason <clm@fb.com>
As we delete large extents, we end up doing huge amounts of COW in order
to delete the corresponding crcs. This adds accounting so that we keep
track of that space and flushing of delayed refs so that we don't build
up too much delayed crc work.
This helps limit the delayed work that must be done at commit time and
tries to avoid ENOSPC aborts because the crcs eat all the global
reserves.
Signed-off-by: Chris Mason <clm@fb.com>
While committing a transaction we free the log roots before we write the
new super block. Freeing the log roots implies marking the disk location
of every node/leaf (metadata extent) as pinned before the new super block
is written. This is to prevent the disk location of log metadata extents
from being reused before the new super block is written, otherwise we
would have a corrupted log tree if before the new super block is written
a crash/reboot happens and the location of any log tree metadata extent
ended up being reused and rewritten.
Even though we pinned the log tree's metadata extents, we were issuing a
discard against them if the fs was mounted with the -o discard option,
resulting in corruption of the log tree if a crash/reboot happened before
writing the new super block - the next time the fs was mounted, during
the log replay process we would find nodes/leafs of the log btree with
a content full of zeroes, causing the process to fail and require the
use of the tool btrfs-zero-log to wipeout the log tree (and all data
previously fsynced becoming lost forever).
Fix this by not doing a discard when pinning an extent. The discard will
be done later when it's safe (after the new super block is committed) at
extent-tree.c:btrfs_finish_extent_commit().
Fixes: e688b7252f (Btrfs: fix extent pinning bugs in the tree log)
CC: <stable@vger.kernel.org>
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs fixes from Chris Mason:
"Most of these are fixing extent reservation accounting, or corners
with tree writeback during commit.
Josef's set does add a test, which isn't strictly a fix, but it'll
keep us from making this same mistake again"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix outstanding_extents accounting in DIO
Btrfs: add sanity test for outstanding_extents accounting
Btrfs: just free dummy extent buffers
Btrfs: account merges/splits properly
Btrfs: prepare block group cache before writing
Btrfs: fix ASSERT(list_empty(&cur_trans->dirty_bgs_list)
Btrfs: account for the correct number of extents for delalloc reservations
Btrfs: fix merge delalloc logic
Btrfs: fix comp_oper to get right order
Btrfs: catch transaction abortion after waiting for it
btrfs: fix sizeof format specifier in btrfs_check_super_valid()
I introduced a regression wrt outstanding_extents accounting. These are tricky
areas that aren't easily covered by xfstests as we could change MAX_EXTENT_SIZE
at any time. So add sanity tests to cover the various conditions that are
tricky in order to make sure we don't introduce regressions in the future.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Writing the block group cache will modify the extent tree quite a bit because it
truncates the old space cache and pre-allocates new stuff. To try and cut down
on the churn lets do the setup dance first, then later on hopefully we can avoid
looping with newly dirtied roots. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Direct IO can easily pass in an buffer that is greater than
BTRFS_MAX_EXTENT_SIZE, so take this into account when reserving extents in the
delalloc reservation code. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs fixes from Chris Mason:
"Outside of misc fixes, Filipe has a few fsync corners and we're
pulling in one more of Josef's fixes from production use here"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs:__add_inode_ref: out of bounds memory read when looking for extended ref.
Btrfs: fix data loss in the fast fsync path
Btrfs: remove extra run_delayed_refs in update_cowonly_root
Btrfs: incremental send, don't rename a directory too soon
btrfs: fix lost return value due to variable shadowing
Btrfs: do not ignore errors from btrfs_lookup_xattr in do_setxattr
Btrfs: fix off-by-one logic error in btrfs_realloc_node
Btrfs: add missing inode update when punching hole
Btrfs: abort the transaction if we fail to update the free space cache inode
Btrfs: fix fsync race leading to ordered extent memory leaks
The divisor is derived from nodesize or PAGE_SIZE, fits into 32bit type.
Get rid of a few more do_div instances.
Signed-off-by: David Sterba <dsterba@suse.cz>
Switch to div_u64 if the divisor is a numeric constant or sum of
sizeof()s. We can remove a few instances of do_div that has the hidden
semtantics of changing the 1st argument.
Small power-of-two divisors are converted to bitshifts, large values are
kept intact for clarity.
Signed-off-by: David Sterba <dsterba@suse.cz>
Our gluster boxes were hitting a problem where they'd run out of space when
updating the block group cache and therefore wouldn't be able to update the free
space inode. This is a problem because this is how we invalidate the cache and
protect ourselves from errors further down the stack, so if this fails we have
to abort the transaction so we make sure we don't end up with stale free space
cache. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs updates from Chris Mason:
"This pull is mostly cleanups and fixes:
- The raid5/6 cleanups from Zhao Lei fixup some long standing warts
in the code and add improvements on top of the scrubbing support
from 3.19.
- Josef has round one of our ENOSPC fixes coming from large btrfs
clusters here at FB.
- Dave Sterba continues a long series of cleanups (thanks Dave), and
Filipe continues hammering on corner cases in fsync and others
This all was held up a little trying to track down a use-after-free in
btrfs raid5/6. It's not clear yet if this is just made easier to
trigger with this pull or if its a new bug from the raid5/6 cleanups.
Dave Sterba is the only one to trigger it so far, but he has a
consistent way to reproduce, so we'll get it nailed shortly"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (68 commits)
Btrfs: don't remove extents and xattrs when logging new names
Btrfs: fix fsync data loss after adding hard link to inode
Btrfs: fix BUG_ON in btrfs_orphan_add() when delete unused block group
Btrfs: account for large extents with enospc
Btrfs: don't set and clear delalloc for O_DIRECT writes
Btrfs: only adjust outstanding_extents when we do a short write
btrfs: Fix out-of-space bug
Btrfs: scrub, fix sleep in atomic context
Btrfs: fix scheduler warning when syncing log
Btrfs: Remove unnecessary placeholder in btrfs_err_code
btrfs: cleanup init for list in free-space-cache
btrfs: delete chunk allocation attemp when setting block group ro
btrfs: clear bio reference after submit_one_bio()
Btrfs: fix scrub race leading to use-after-free
Btrfs: add missing cleanup on sysfs init failure
Btrfs: fix race between transaction commit and empty block group removal
btrfs: add more checks to btrfs_read_sys_array
btrfs: cleanup, rename a few variables in btrfs_read_sys_array
btrfs: add checks for sys_chunk_array sizes
btrfs: more superblock checks, lower bounds on devices and sectorsize/nodesize
...
int alloc_chunk is never used in this function, remove it.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.cz>
This is the 3rd independent patch of a larger project to cleanup btrfs's
internal usage of btrfs_root. Many functions take btrfs_root only to
grab the fs_info struct.
By requiring a root these functions cause programmer overhead. That
these functions can accept any valid root is not obvious until
inspection.
This patch reduces the specificity of such functions to accept the
fs_info directly.
These patches can be applied independently and thus are not being
submitted as a patch series. There should be about 26 patches by the
project's completion. Each patch will cleanup between 1 and 34 functions
apiece. Each patch covers a single file's functions.
This patch affects the following function(s):
1) csum_tree_block
2) csum_dirty_buffer
3) check_tree_block_fsid
4) btrfs_find_tree_block
5) clean_tree_block
Signed-off-by: Daniel Dressler <danieru.dressler@gmail.com>
Signed-off-by: David Sterba <dsterba@suse.cz>
On our gluster boxes we stream large tar balls of backups onto our fses. With
160gb of ram this means we get really large contiguous ranges of dirty data, but
the way our ENOSPC stuff works is that as long as it's contiguous we only hold
metadata reservation for one extent. The problem is we limit our extents to
128mb, so we'll end up with at least 800 extents so our enospc accounting is
quite a bit lower than what we need. To keep track of this make sure we
increase outstanding_extents for every multiple of the max extent size so we can
be sure to have enough reserved metadata space. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Below test will fail currently:
mkfs.ext4 -F /dev/sda
btrfs-convert /dev/sda
mount /dev/sda /mnt
btrfs device add -f /dev/sdb /mnt
btrfs balance start -v -dconvert=raid1 -mconvert=raid1 /mnt
The reason is there are some block groups with usage 0, but the whole
disk hasn't free space to allocate new chunk, so we even can't set such
block group readonly. This patch deletes the chunk allocation when
setting block group ro. For META, we already have reserve. But for
SYSTEM, we don't have, so the check_system_chunk is still required.
Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
"run_most" is not used anymore.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
1: ref_count is simple than current RBIO_HOLD_BBIO_MAP_BIT flag
to keep btrfs_bio's memory in raid56 recovery implement.
2: free function for bbio will make code clean and flexible, plus
forced data type checking in compile.
Changelog v1->v2:
Rename following by David Sterba's suggestion:
put_btrfs_bio() -> btrfs_put_bio()
get_btrfs_bio() -> btrfs_get_bio()
bbio->ref_count -> bbio->refs
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Very often our extent buffer's header generation doesn't match the current
transaction's id or it is also referenced by other trees (snapshots), so
we don't need the corresponding block group cache object. Therefore only
search for it if we are going to use it, so we avoid an unnecessary search
in the block groups rbtree (and acquiring and releasing its spinlock).
Freeing a tree block is performed when COWing or deleting a node/leaf,
which implies we are holding the node/leaf's parent node lock, therefore
reducing the amount of time spent when freeing a tree block helps reducing
the amount of time we are holding the parent node's lock.
For example, for a run of xfstests/generic/083, the block group cache
object was needed only 682 times for a total of 226691 calls to free
a tree block.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently any time we try to update the block groups on disk we will walk _all_
block groups and check for the ->dirty flag to see if it is set. This function
can get called several times during a commit. So if you have several terabytes
of data you will be a very sad panda as we will loop through _all_ of the block
groups several times, which makes the commit take a while which slows down the
rest of the file system operations.
This patch introduces a dirty list for the block groups that we get added to
when we dirty the block group for the first time. Then we simply update any
block groups that have been dirtied since the last time we called
btrfs_write_dirty_block_groups. This allows us to clean up how we write the
free space cache out so it is much cleaner. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When removing a block group we were deleting it from its space_info's
ro_bgs list without the correct protection - the space info's spinlock.
Fix this by doing the list delete while holding the spinlock of the
corresponding space info, which is the correct lock for any operation
on that list.
This issue was introduced in the 3.19 kernel by the following change:
Btrfs: move read only block groups onto their own list V2
commit 633c0aad4c
I ran into a kernel crash while a task was running statfs, which iterates
the space_info->ro_bgs list while holding the space info's spinlock,
and another task was deleting it from the same list, without holding that
spinlock, as part of the block group remove operation (while running the
function btrfs_remove_block_group). This happened often when running the
stress test xfstests/generic/038 I recently made.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We shouldn't BUG_ON() if there is corruption. I hit this while testing my block
group patch and the abort worked properly. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Finally it's clear that the requested blocksize is always equal to
nodesize, with one exception, the superblock.
Superblock has fixed size regardless of the metadata block size, but
uses the same helpers to initialize sys array/chunk tree and to work
with the chunk items. So it pretends to be an extent_buffer for a
moment, btrfs_read_sys_array is full of special cases, we're adding one
more.
Signed-off-by: David Sterba <dsterba@suse.cz>
It doesn't do anything special, it just calls btrfs_discard_extent(),
so just remove it.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we abort a transaction we iterate over all the ranges marked as dirty
in fs_info->freed_extents[0] and fs_info->freed_extents[1], clear them
from those trees, add them back (unpin) to the free space caches and, if
the fs was mounted with "-o discard", perform a discard on those regions.
Also, after adding the regions to the free space caches, a fitrim ioctl call
can see those ranges in a block group's free space cache and perform a discard
on the ranges, so the same issue can happen without "-o discard" as well.
This causes corruption, affecting one or multiple btree nodes (in the worst
case leaving the fs unmountable) because some of those ranges (the ones in
the fs_info->pinned_extents tree) correspond to btree nodes/leafs that are
referred by the last committed super block - breaking the rule that anything
that was committed by a transaction is untouched until the next transaction
commits successfully.
I ran into this while running in a loop (for several hours) the fstest that
I recently submitted:
[PATCH] fstests: add btrfs test to stress chunk allocation/removal and fstrim
The corruption always happened when a transaction aborted and then fsck complained
like this:
_check_btrfs_filesystem: filesystem on /dev/sdc is inconsistent
*** fsck.btrfs output ***
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
read block failed check_tree_block
Couldn't open file system
In this case 94945280 corresponded to the root of a tree.
Using frace what I observed was the following sequence of steps happened:
1) transaction N started, fs_info->pinned_extents pointed to
fs_info->freed_extents[0];
2) node/eb 94945280 is created;
3) eb is persisted to disk;
4) transaction N commit starts, fs_info->pinned_extents now points to
fs_info->freed_extents[1], and transaction N completes;
5) transaction N + 1 starts;
6) eb is COWed, and btrfs_free_tree_block() called for this eb;
7) eb range (94945280 to 94945280 + 16Kb) is added to
fs_info->pinned_extents (fs_info->freed_extents[1]);
8) Something goes wrong in transaction N + 1, like hitting ENOSPC
for example, and the transaction is aborted, turning the fs into
readonly mode. The stack trace I got for example:
[112065.253935] [<ffffffff8140c7b6>] dump_stack+0x4d/0x66
[112065.254271] [<ffffffff81042984>] warn_slowpath_common+0x7f/0x98
[112065.254567] [<ffffffffa0325990>] ? __btrfs_abort_transaction+0x50/0x10b [btrfs]
[112065.261674] [<ffffffff810429e5>] warn_slowpath_fmt+0x48/0x50
[112065.261922] [<ffffffffa032949e>] ? btrfs_free_path+0x26/0x29 [btrfs]
[112065.262211] [<ffffffffa0325990>] __btrfs_abort_transaction+0x50/0x10b [btrfs]
[112065.262545] [<ffffffffa036b1d6>] btrfs_remove_chunk+0x537/0x58b [btrfs]
[112065.262771] [<ffffffffa033840f>] btrfs_delete_unused_bgs+0x1de/0x21b [btrfs]
[112065.263105] [<ffffffffa0343106>] cleaner_kthread+0x100/0x12f [btrfs]
(...)
[112065.264493] ---[ end trace dd7903a975a31a08 ]---
[112065.264673] BTRFS: error (device sdc) in btrfs_remove_chunk:2625: errno=-28 No space left
[112065.264997] BTRFS info (device sdc): forced readonly
9) The clear kthread sees that the BTRFS_FS_STATE_ERROR bit is set in
fs_info->fs_state and calls btrfs_cleanup_transaction(), which in
turn calls btrfs_destroy_pinned_extent();
10) Then btrfs_destroy_pinned_extent() iterates over all the ranges
marked as dirty in fs_info->freed_extents[], and for each one
it calls discard, if the fs was mounted with "-o discard", and
adds the range to the free space cache of the respective block
group;
11) btrfs_trim_block_group(), invoked from the fitrim ioctl code path,
sees the free space entries and performs a discard;
12) After an umount and mount (or fsck), our eb's location on disk was full
of zeroes, and it should have been untouched, because it was marked as
dirty in the fs_info->pinned_extents tree, and therefore used by the
trees that the last committed superblock points to.
Fix this by not performing a discard and not adding the ranges to the free space
caches - it's useless from this point since the fs is now in readonly mode and
we won't write free space caches to disk anymore (otherwise we would leak space)
nor any new superblock. By not adding the ranges to the free space caches, it
prevents other code paths from allocating that space and write to it as well,
therefore being safer and simpler.
This isn't a new problem, as it's been present since 2011 (git commit
acce952b02).
Cc: stable@vger.kernel.org # any kernel released after 2011-01-06
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Always clear a block group's rbnode after removing it from the rbtree to
ensure that any tasks that might be holding a reference on the block group
don't end up accessing stale rbnode left and right child pointers through
next_block_group().
This is a leftover from the change titled:
"Btrfs: fix invalid block group rbtree access after bg is removed"
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
This was written when we didn't do a caching control for the fast free space
cache loading. However we started doing that a long time ago, and there is
still a small window of time that we could be caching the block group the fast
way, so if there is a caching_ctl at all on the block group just return it, the
callers all wait properly for what they want. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
On block group remove if the corresponding extent map was on the
transaction->pending_chunks list, we were deleting the extent map
from that list, through remove_extent_mapping(), without any
synchronization with chunk allocation (which iterates that list
and adds new elements to it). Fix this by ensure that this is done
while the chunk mutex is held, since that's the mutex that protects
the list in the chunk allocation code path.
This applies on top (depends on) of my previous patch titled:
"Btrfs: fix race between fs trimming and block group remove/allocation"
But the issue in fact was already present before that change, it only
became easier to hit after Josef's 3.18 patch that added automatic
removal of empty block groups.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
On chunk allocation error (label "error_del_extent"), after adding the
extent map to the tree and to the pending chunks list, we would leave
decrementing the extent map's refcount by 2 instead of 3 (our allocation
+ tree reference + list reference).
Also, on chunk/block group removal, if the block group was on the list
pending_chunks we weren't decrementing the respective list reference.
Detected by 'rmmod btrfs':
[20770.105881] kmem_cache_destroy btrfs_extent_map: Slab cache still has objects
[20770.106127] CPU: 2 PID: 11093 Comm: rmmod Tainted: G W L 3.17.0-rc5-btrfs-next-1+ #1
[20770.106128] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014
[20770.106130] 0000000000000000 ffff8800ba867eb8 ffffffff813e7a13 ffff8800a2e11040
[20770.106132] ffff8800ba867ed0 ffffffff81105d0c 0000000000000000 ffff8800ba867ee0
[20770.106134] ffffffffa035d65e ffff8800ba867ef0 ffffffffa03b0654 ffff8800ba867f78
[20770.106136] Call Trace:
[20770.106142] [<ffffffff813e7a13>] dump_stack+0x45/0x56
[20770.106145] [<ffffffff81105d0c>] kmem_cache_destroy+0x4b/0x90
[20770.106164] [<ffffffffa035d65e>] extent_map_exit+0x1a/0x1c [btrfs]
[20770.106176] [<ffffffffa03b0654>] exit_btrfs_fs+0x27/0x9d3 [btrfs]
[20770.106179] [<ffffffff8109dc97>] SyS_delete_module+0x153/0x1c4
[20770.106182] [<ffffffff8121261b>] ? trace_hardirqs_on_thunk+0x3a/0x3c
[20770.106184] [<ffffffff813ebf52>] system_call_fastpath+0x16/0x1b
This applies on top (depends on) of my previous patch titled:
"Btrfs: fix race between fs trimming and block group remove/allocation"
But the issue in fact was already present before that change, it only
became easier to hit after Josef's 3.18 patch that added automatic
removal of empty block groups.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
If the transaction handle doesn't have used blocks but has created new block
groups make sure we turn the fs into readonly mode too. This is because the
new block groups didn't get all their metadata persisted into the chunk and
device trees, and therefore if a subsequent transaction starts, allocates
space from the new block groups, writes data or metadata into that space,
commits successfully and then after we unmount and mount the filesystem
again, the same space can be allocated again for a new block group,
resulting in file data or metadata corruption.
Example where we don't abort the transaction when we fail to finish the
chunk allocation (add items to the chunk and device trees) and later a
future transaction where the block group is removed fails because it can't
find the chunk item in the chunk tree:
[25230.404300] WARNING: CPU: 0 PID: 7721 at fs/btrfs/super.c:260 __btrfs_abort_transaction+0x50/0xfc [btrfs]()
[25230.404301] BTRFS: Transaction aborted (error -28)
[25230.404302] Modules linked in: btrfs dm_flakey nls_utf8 fuse xor raid6_pq ntfs vfat msdos fat xfs crc32c_generic libcrc32c ext3 jbd ext2 dm_mod nfsd auth_rpcgss oid_registry nfs_acl nfs lockd fscache sunrpc loop psmouse i2c_piix4 i2ccore parport_pc parport processor button pcspkr serio_raw thermal_sys evdev microcode ext4 crc16 jbd2 mbcache sr_mod cdrom ata_generic sg sd_mod crc_t10dif crct10dif_generic crct10dif_common virtio_scsi floppy e1000 ata_piix libata virtio_pci virtio_ring scsi_mod virtio [last unloaded: btrfs]
[25230.404325] CPU: 0 PID: 7721 Comm: xfs_io Not tainted 3.17.0-rc5-btrfs-next-1+ #1
[25230.404326] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014
[25230.404328] 0000000000000000 ffff88004581bb08 ffffffff813e7a13 ffff88004581bb50
[25230.404330] ffff88004581bb40 ffffffff810423aa ffffffffa049386a 00000000ffffffe4
[25230.404332] ffffffffa05214c0 000000000000240c ffff88010fc8f800 ffff88004581bba8
[25230.404334] Call Trace:
[25230.404338] [<ffffffff813e7a13>] dump_stack+0x45/0x56
[25230.404342] [<ffffffff810423aa>] warn_slowpath_common+0x7f/0x98
[25230.404351] [<ffffffffa049386a>] ? __btrfs_abort_transaction+0x50/0xfc [btrfs]
[25230.404353] [<ffffffff8104240b>] warn_slowpath_fmt+0x48/0x50
[25230.404362] [<ffffffffa049386a>] __btrfs_abort_transaction+0x50/0xfc [btrfs]
[25230.404374] [<ffffffffa04a8c43>] btrfs_create_pending_block_groups+0x10c/0x135 [btrfs]
[25230.404387] [<ffffffffa04b77fd>] __btrfs_end_transaction+0x7e/0x2de [btrfs]
[25230.404398] [<ffffffffa04b7a6d>] btrfs_end_transaction+0x10/0x12 [btrfs]
[25230.404408] [<ffffffffa04a3d64>] btrfs_check_data_free_space+0x111/0x1f0 [btrfs]
[25230.404421] [<ffffffffa04c53bd>] __btrfs_buffered_write+0x160/0x48d [btrfs]
[25230.404425] [<ffffffff811a9268>] ? cap_inode_need_killpriv+0x2d/0x37
[25230.404429] [<ffffffff810f6501>] ? get_page+0x1a/0x2b
[25230.404441] [<ffffffffa04c7c95>] btrfs_file_write_iter+0x321/0x42f [btrfs]
[25230.404443] [<ffffffff8110f5d9>] ? handle_mm_fault+0x7f3/0x846
[25230.404446] [<ffffffff813e98c5>] ? mutex_unlock+0x16/0x18
[25230.404449] [<ffffffff81138d68>] new_sync_write+0x7c/0xa0
[25230.404450] [<ffffffff81139401>] vfs_write+0xb0/0x112
[25230.404452] [<ffffffff81139c9d>] SyS_pwrite64+0x66/0x84
[25230.404454] [<ffffffff813ebf52>] system_call_fastpath+0x16/0x1b
[25230.404455] ---[ end trace 5aa5684fdf47ab38 ]---
[25230.404458] BTRFS warning (device sdc): btrfs_create_pending_block_groups:9228: Aborting unused transaction(No space left).
[25288.084814] BTRFS: error (device sdc) in btrfs_free_chunk:2509: errno=-2 No such entry (Failed lookup while freeing chunk.)
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Our fs trim operation, which is completely transactionless (doesn't start
or joins an existing transaction) consists of visiting all block groups
and then for each one to iterate its free space entries and perform a
discard operation against the space range represented by the free space
entries. However before performing a discard, the corresponding free space
entry is removed from the free space rbtree, and when the discard completes
it is added back to the free space rbtree.
If a block group remove operation happens while the discard is ongoing (or
before it starts and after a free space entry is hidden), we end up not
waiting for the discard to complete, remove the extent map that maps
logical address to physical addresses and the corresponding chunk metadata
from the the chunk and device trees. After that and before the discard
completes, the current running transaction can finish and a new one start,
allowing for new block groups that map to the same physical addresses to
be allocated and written to.
So fix this by keeping the extent map in memory until the discard completes
so that the same physical addresses aren't reused before it completes.
If the physical locations that are under a discard operation end up being
used for a new metadata block group for example, and dirty metadata extents
are written before the discard finishes (the VM might call writepages() of
our btree inode's i_mapping for example, or an fsync log commit happens) we
end up overwriting metadata with zeroes, which leads to errors from fsck
like the following:
checking extents
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
read block failed check_tree_block
owner ref check failed [833912832 16384]
Errors found in extent allocation tree or chunk allocation
checking free space cache
checking fs roots
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
read block failed check_tree_block
root 5 root dir 256 error
root 5 inode 260 errors 2001, no inode item, link count wrong
unresolved ref dir 256 index 0 namelen 8 name foobar_3 filetype 1 errors 6, no dir index, no inode ref
root 5 inode 262 errors 2001, no inode item, link count wrong
unresolved ref dir 256 index 0 namelen 8 name foobar_5 filetype 1 errors 6, no dir index, no inode ref
root 5 inode 263 errors 2001, no inode item, link count wrong
(...)
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
There's a race between adding a block group to the list of the unused
block groups and removing an unused block group (cleaner kthread) that
leads to freeing extents that are in use or a crash during transaction
commmit. Basically the cleaner kthread, when executing
btrfs_delete_unused_bgs(), might catch the newly added block group to
the list fs_info->unused_bgs and clear the range representing the whole
group from fs_info->freed_extents[] before the task that added the block
group to the list (running update_block_group()) marked the last freed
extent as dirty in fs_info->freed_extents (pinned_extents).
That is:
CPU 1 CPU 2
btrfs_delete_unused_bgs()
update_block_group()
add block group to
fs_info->unused_bgs
got block group from the list
clear_extent_bits for the whole
block group range in freed_extents[]
set_extent_dirty for the
range covering the freed
extent in freed_extents[]
(fs_info->pinned_extents)
block group deleted, and a new block
group with the same logical address is
created
reserve space from the new block group
for new data or metadata - the reserved
space overlaps the range specified by
CPU 1 for set_extent_dirty()
commit transaction
find all ranges marked as dirty in
fs_info->pinned_extents, clear them
and add them to the free space cache
Alternatively, if CPU 2 doesn't create a new block group with the same
logical address, we get a crash/BUG_ON at transaction commit when unpining
extent ranges because we can't find a block group for the range marked as
dirty by CPU 1. Sample trace:
[ 2163.426462] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC
[ 2163.426640] Modules linked in: btrfs xor raid6_pq dm_thin_pool dm_persistent_data dm_bio_prison dm_bufio crc32c_generic libcrc32c dm_mod nfsd auth_rpc
gss oid_registry nfs_acl nfs lockd fscache sunrpc loop psmouse parport_pc parport i2c_piix4 processor thermal_sys i2ccore evdev button pcspkr microcode serio_raw ext4 crc16 jbd2 mbcache
sg sr_mod cdrom sd_mod crc_t10dif crct10dif_generic crct10dif_common ata_generic virtio_scsi floppy ata_piix libata e1000 scsi_mod virtio_pci virtio_ring virtio
[ 2163.428209] CPU: 0 PID: 11858 Comm: btrfs-transacti Tainted: G W 3.17.0-rc5-btrfs-next-1+ #1
[ 2163.428519] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014
[ 2163.428875] task: ffff88009f2c0650 ti: ffff8801356bc000 task.ti: ffff8801356bc000
[ 2163.429157] RIP: 0010:[<ffffffffa037728e>] [<ffffffffa037728e>] unpin_extent_range.isra.58+0x62/0x192 [btrfs]
[ 2163.429562] RSP: 0018:ffff8801356bfda8 EFLAGS: 00010246
[ 2163.429802] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[ 2163.429990] RDX: 0000000041bfffff RSI: 0000000001c00000 RDI: ffff880024307080
[ 2163.430042] RBP: ffff8801356bfde8 R08: 0000000000000068 R09: ffff88003734f118
[ 2163.430042] R10: ffff8801356bfcb8 R11: fffffffffffffb69 R12: ffff8800243070d0
[ 2163.430042] R13: 0000000083c04000 R14: ffff8800751b0f00 R15: ffff880024307000
[ 2163.430042] FS: 0000000000000000(0000) GS:ffff88013f400000(0000) knlGS:0000000000000000
[ 2163.430042] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 2163.430042] CR2: 00007ff10eb43fc0 CR3: 0000000004cb8000 CR4: 00000000000006f0
[ 2163.430042] Stack:
[ 2163.430042] ffff8800243070d0 0000000083c08000 0000000083c07fff ffff88012d6bc800
[ 2163.430042] ffff8800243070d0 ffff8800751b0f18 ffff8800751b0f00 0000000000000000
[ 2163.430042] ffff8801356bfe18 ffffffffa037a481 0000000083c04000 0000000083c07fff
[ 2163.430042] Call Trace:
[ 2163.430042] [<ffffffffa037a481>] btrfs_finish_extent_commit+0xac/0xbf [btrfs]
[ 2163.430042] [<ffffffffa038c06d>] btrfs_commit_transaction+0x6ee/0x882 [btrfs]
[ 2163.430042] [<ffffffffa03881f1>] transaction_kthread+0xf2/0x1a4 [btrfs]
[ 2163.430042] [<ffffffffa03880ff>] ? btrfs_cleanup_transaction+0x3d8/0x3d8 [btrfs]
[ 2163.430042] [<ffffffff8105966b>] kthread+0xb7/0xbf
[ 2163.430042] [<ffffffff810595b4>] ? __kthread_parkme+0x67/0x67
[ 2163.430042] [<ffffffff813ebeac>] ret_from_fork+0x7c/0xb0
[ 2163.430042] [<ffffffff810595b4>] ? __kthread_parkme+0x67/0x67
So fix this by making update_block_group() first set the range as dirty
in pinned_extents before adding the block group to the unused_bgs list.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we grab a block group, for example in btrfs_trim_fs(), we will be holding
a reference on it but the block group can be removed after we got it (via
btrfs_remove_block_group), which means it will no longer be part of the
rbtree.
However, btrfs_remove_block_group() was only calling rb_erase() which leaves
the block group's rb_node left and right child pointers with the same content
they had before calling rb_erase. This was dangerous because a call to
next_block_group() would access the node's left and right child pointers (via
rb_next), which can be no longer valid.
Fix this by clearing a block group's node after removing it from the tree,
and have next_block_group() do a tree search to get the next block group
instead of using rb_next() if our block group was removed.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
If right after starting the snapshot creation ioctl we perform a write against a
file followed by a truncate, with both operations increasing the file's size, we
can get a snapshot tree that reflects a state of the source subvolume's tree where
the file truncation happened but the write operation didn't. This leaves a gap
between 2 file extent items of the inode, which makes btrfs' fsck complain about it.
For example, if we perform the following file operations:
$ mkfs.btrfs -f /dev/vdd
$ mount /dev/vdd /mnt
$ xfs_io -f \
-c "pwrite -S 0xaa -b 32K 0 32K" \
-c "fsync" \
-c "pwrite -S 0xbb -b 32770 16K 32770" \
-c "truncate 90123" \
/mnt/foobar
and the snapshot creation ioctl was just called before the second write, we often
can get the following inode items in the snapshot's btree:
item 120 key (257 INODE_ITEM 0) itemoff 7987 itemsize 160
inode generation 146 transid 7 size 90123 block group 0 mode 100600 links 1 uid 0 gid 0 rdev 0 flags 0x0
item 121 key (257 INODE_REF 256) itemoff 7967 itemsize 20
inode ref index 282 namelen 10 name: foobar
item 122 key (257 EXTENT_DATA 0) itemoff 7914 itemsize 53
extent data disk byte 1104855040 nr 32768
extent data offset 0 nr 32768 ram 32768
extent compression 0
item 123 key (257 EXTENT_DATA 53248) itemoff 7861 itemsize 53
extent data disk byte 0 nr 0
extent data offset 0 nr 40960 ram 40960
extent compression 0
There's a file range, corresponding to the interval [32K; ALIGN(16K + 32770, 4096)[
for which there's no file extent item covering it. This is because the file write
and file truncate operations happened both right after the snapshot creation ioctl
called btrfs_start_delalloc_inodes(), which means we didn't start and wait for the
ordered extent that matches the write and, in btrfs_setsize(), we were able to call
btrfs_cont_expand() before being able to commit the current transaction in the
snapshot creation ioctl. So this made it possibe to insert the hole file extent
item in the source subvolume (which represents the region added by the truncate)
right before the transaction commit from the snapshot creation ioctl.
Btrfs' fsck tool complains about such cases with a message like the following:
"root 331 inode 257 errors 100, file extent discount"
>From a user perspective, the expectation when a snapshot is created while those
file operations are being performed is that the snapshot will have a file that
either:
1) is empty
2) only the first write was captured
3) only the 2 writes were captured
4) both writes and the truncation were captured
But never capture a state where only the first write and the truncation were
captured (since the second write was performed before the truncation).
A test case for xfstests follows.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Due to ignoring errors returned by clear_extent_bits (at the moment only
-ENOMEM is possible), we can end up freeing an extent that is actually in
use (i.e. return the extent to the free space cache).
The sequence of steps that lead to this:
1) Cleaner thread starts execution and calls btrfs_delete_unused_bgs(), with
the goal of freeing empty block groups;
2) btrfs_delete_unused_bgs() finds an empty block group, joins the current
transaction (or starts a new one if none is running) and attempts to
clear the EXTENT_DIRTY bit for the block group's range from freed_extents[0]
and freed_extents[1] (of which one corresponds to fs_info->pinned_extents);
3) Clearing the EXTENT_DIRTY bit (via clear_extent_bits()) fails with
-ENOMEM, but such error is ignored and btrfs_delete_unused_bgs() proceeds
to delete the block group and the respective chunk, while pinned_extents
remains with that bit set for the whole (or a part of the) range covered
by the block group;
4) Later while the transaction is still running, the chunk ends up being reused
for a new block group (maybe for different purpose, data or metadata), and
extents belonging to the new block group are allocated for file data or btree
nodes/leafs;
5) The current transaction is committed, meaning that we unpinned one or more
extents from the new block group (through btrfs_finish_extent_commit() and
unpin_extent_range()) which are now being used for new file data or new
metadata (through btrfs_finish_extent_commit() and unpin_extent_range()).
And unpinning means we returned the extents to the free space cache of the
new block group, which implies those extents can be used for future allocations
while they're still in use.
Alternatively, we can hit a BUG_ON() when doing a lookup for a block group's cache
object in unpin_extent_range() if a new block group didn't end up being allocated for
the same chunk (step 4 above).
Fix this by not freeing the block group and chunk if we fail to clear the dirty bit.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Our gluster boxes were spending lots of time in statfs because our fs'es are
huge. The problem is statfs loops through all of the block groups looking for
read only block groups, and when you have several terabytes worth of data that
ends up being a lot of block groups. Move the read only block groups onto a
read only list and only proces that list in
btrfs_account_ro_block_groups_free_space to reduce the amount of churn. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs fixes from Chris Mason:
"Filipe is nailing down some problems with our skinny extent variation,
and Dave's patch fixes endian problems in the new super block checks"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: fix race that makes btrfs_lookup_extent_info miss skinny extent items
Btrfs: properly clean up btrfs_end_io_wq_cache
Btrfs: fix invalid leaf slot access in btrfs_lookup_extent()
btrfs: use macro accessors in superblock validation checks
We have a race that can lead us to miss skinny extent items in the function
btrfs_lookup_extent_info() when the skinny metadata feature is enabled.
So basically the sequence of steps is:
1) We search in the extent tree for the skinny extent, which returns > 0
(not found);
2) We check the previous item in the returned leaf for a non-skinny extent,
and we don't find it;
3) Because we didn't find the non-skinny extent in step 2), we release our
path to search the extent tree again, but this time for a non-skinny
extent key;
4) Right after we released our path in step 3), a skinny extent was inserted
in the extent tree (delayed refs were run) - our second extent tree search
will miss it, because it's not looking for a skinny extent;
5) After the second search returned (with ret > 0), we look for any delayed
ref for our extent's bytenr (and we do it while holding a read lock on the
leaf), but we won't find any, as such delayed ref had just run and completed
after we released out path in step 3) before doing the second search.
Fix this by removing completely the path release and re-search logic. This is
safe, because if we seach for a metadata item and we don't find it, we have the
guarantee that the returned leaf is the one where the item would be inserted,
and so path->slots[0] > 0 and path->slots[0] - 1 must be the slot where the
non-skinny extent item is if it exists. The only case where path->slots[0] is
zero is when there are no smaller keys in the tree (i.e. no left siblings for
our leaf), in which case the re-search logic isn't needed as well.
This race has been present since the introduction of skinny metadata (change
3173a18f70).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we couldn't find our extent item, we accessed the current slot
(path->slots[0]) to check if it corresponds to an equivalent skinny
metadata item. However this slot could be beyond our last item in the
leaf (i.e. path->slots[0] >= btrfs_header_nritems(leaf)), in which case
we shouldn't process it.
Since btrfs_lookup_extent() is only used to find extent items for data
extents, fix this by removing completely the logic that looks up for an
equivalent skinny metadata item, since it can not exist.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Pull btrfs updates from Chris Mason:
"The largest set of changes here come from Miao Xie. He's cleaning up
and improving read recovery/repair for raid, and has a number of
related fixes.
I've merged another set of fsync fixes from Filipe, and he's also
improved the way we handle metadata write errors to make sure we force
the FS readonly if things go wrong.
Otherwise we have a collection of fixes and cleanups. Dave Sterba
gets a cookie for removing the most lines (thanks Dave)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (139 commits)
btrfs: Fix compile error when CONFIG_SECURITY is not set.
Btrfs: fix compiles when CONFIG_BTRFS_FS_RUN_SANITY_TESTS is off
btrfs: Make btrfs handle security mount options internally to avoid losing security label.
Btrfs: send, don't delay dir move if there's a new parent inode
btrfs: add more superblock checks
Btrfs: fix race in WAIT_SYNC ioctl
Btrfs: be aware of btree inode write errors to avoid fs corruption
Btrfs: remove redundant btrfs_verify_qgroup_counts declaration.
btrfs: fix shadow warning on cmp
Btrfs: fix compilation errors under DEBUG
Btrfs: fix crash of btrfs_release_extent_buffer_page
Btrfs: add missing end_page_writeback on submit_extent_page failure
btrfs: Fix the wrong condition judgment about subset extent map
Btrfs: fix build_backref_tree issue with multiple shared blocks
Btrfs: cleanup error handling in build_backref_tree
btrfs: move checks for DUMMY_ROOT into a helper
btrfs: new define for the inline extent data start
btrfs: kill extent_buffer_page helper
btrfs: drop constant param from btrfs_release_extent_buffer_page
btrfs: hide typecast to definition of BTRFS_SEND_TRANS_STUB
...
While we have a transaction ongoing, the VM might decide at any time
to call btree_inode->i_mapping->a_ops->writepages(), which will start
writeback of dirty pages belonging to btree nodes/leafs. This call
might return an error or the writeback might finish with an error
before we attempt to commit the running transaction. If this happens,
we might have no way of knowing that such error happened when we are
committing the transaction - because the pages might no longer be
marked dirty nor tagged for writeback (if a subsequent modification
to the extent buffer didn't happen before the transaction commit) which
makes filemap_fdata[write|wait]_range unable to find such pages (even
if they're marked with SetPageError).
So if this happens we must abort the transaction, otherwise we commit
a super block with btree roots that point to btree nodes/leafs whose
content on disk is invalid - either garbage or the content of some
node/leaf from a past generation that got cowed or deleted and is no
longer valid (for this later case we end up getting error messages like
"parent transid verify failed on 10826481664 wanted 25748 found 29562"
when reading btree nodes/leafs from disk).
Note that setting and checking AS_EIO/AS_ENOSPC in the btree inode's
i_mapping would not be enough because we need to distinguish between
log tree extents (not fatal) vs non-log tree extents (fatal) and
because the next call to filemap_fdatawait_range() will catch and clear
such errors in the mapping - and that call might be from a log sync and
not from a transaction commit, which means we would not know about the
error at transaction commit time. Also, checking for the eb flag
EXTENT_BUFFER_IOERR at transaction commit time isn't done and would
not be completely reliable, as the eb might be removed from memory and
read back when trying to get it, which clears that flag right before
reading the eb's pages from disk, making us not know about the previous
write error.
Using the new 3 flags for the btree inode also makes us achieve the
goal of AS_EIO/AS_ENOSPC when writepages() returns success, started
writeback for all dirty pages and before filemap_fdatawait_range() is
called, the writeback for all dirty pages had already finished with
errors - because we were not using AS_EIO/AS_ENOSPC,
filemap_fdatawait_range() would return success, as it could not know
that writeback errors happened (the pages were no longer tagged for
writeback).
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Rename to btrfs_alloc_tree_block as it fits to the alloc/find/free +
_tree_block family. The parameter blocksize was set to the metadata
block size, directly or indirectly.
Signed-off-by: David Sterba <dsterba@suse.cz>
The parent_transid parameter has been unused since its introduction in
ca7a79ad8d ("Pass down the expected generation number when reading
tree blocks"). In reada_tree_block, it was even wrongly set to leafsize.
Transid check is done in the proper read and readahead ignores errors.
Signed-off-by: David Sterba <dsterba@suse.cz>
There are the branch hints that obviously depend on the data being
processed, the CPU predictor will do better job according to the actual
load. It also does not make sense to use the hints in slow paths that do
a lot of other operations like locking, waiting or IO.
Signed-off-by: David Sterba <dsterba@suse.cz>
This is to receive 0a30288da1 ("blk-mq, percpu_ref: implement a
kludge for SCSI blk-mq stall during probe") which implements
__percpu_ref_kill_expedited() to work around SCSI blk-mq stall. The
commit reverted and patches to implement proper fix will be added.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
Trying to reproduce a log enospc bug I hit a panic in the async reclaim code
during log replay. This is because we use fs_info->fs_root as our root for
shrinking and such. Technically we can use whatever root we want, but let's
just not allow async reclaim while we're doing log replay. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
One problem that has plagued us is that a user will use up all of his space with
data, remove a bunch of that data, and then try to create a bunch of small files
and run out of space. This happens because all the chunks were allocated for
data since the metadata requirements were so low. But now there's a bunch of
empty data block groups and not enough metadata space to do anything. This
patch solves this problem by automatically deleting empty block groups. If we
notice the used count go down to 0 when deleting or on mount notice that a block
group has a used count of 0 then we will queue it to be deleted.
When the cleaner thread runs we will double check to make sure the block group
is still empty and then we will delete it. This patch has the side effect of no
longer having a bunch of BUG_ON()'s in the chunk delete code, which will be
helpful for both this and relocate. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
There were several problems about chunk mutex usage:
- Lock chunk mutex when updating metadata. It would cause the nested
deadlock because updating metadata might need allocate new chunks
that need acquire chunk mutex. We remove chunk mutex at this case,
because b-tree lock and other lock mechanism can help us.
- ABBA deadlock occured between device_list_mutex and chunk_mutex.
When we update device status, we must acquire device_list_mutex at the
beginning, and then we might get chunk_mutex during the device status
update because we need allocate new chunks for metadata COW. But at
most place, we acquire chunk_mutex at first and then acquire device list
mutex. We need change the lock order.
- Some place we needn't acquire chunk_mutex. For example we needn't get
chunk_mutex when we free a empty seed fs_devices structure.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
One of my tests shows that when we really don't have space to reclaim via
flush_space and also run out of space, this async reclaim work loops on adding
itself into the workqueue and keeps writing something to disk according to
iostat's results, and these writes mainly comes from commit_transaction which
writes super_block. This's unacceptable as it can be bad to disks, especially
memeory storages.
This adds a check to avoid the above situation.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
The nodesize and leafsize were never of different values. Unify the
usage and make nodesize the one. Cleanup the redundant checks and
helpers.
Shaves a few bytes from .text:
text data bss dec hex filename
852418 24560 23112 900090 dbbfa btrfs.ko.before
851074 24584 23112 898770 db6d2 btrfs.ko.after
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
btrfs_set_key_type and btrfs_key_type are used inconsistently along with
open coded variants. Other members of btrfs_key are accessed directly
without any helpers anyway.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
Percpu allocator now supports allocation mask. Add @gfp to
percpu_counter_init() so that !GFP_KERNEL allocation masks can be used
with percpu_counters too.
We could have left percpu_counter_init() alone and added
percpu_counter_init_gfp(); however, the number of users isn't that
high and introducing _gfp variants to all percpu data structures would
be quite ugly, so let's just do the conversion. This is the one with
the most users. Other percpu data structures are a lot easier to
convert.
This patch doesn't make any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jan Kara <jack@suse.cz>
Acked-by: "David S. Miller" <davem@davemloft.net>
Cc: x86@kernel.org
Cc: Jens Axboe <axboe@kernel.dk>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
This has been reported and discussed for a long time, and this hang occurs in
both 3.15 and 3.16.
Btrfs now migrates to use kernel workqueue, but it introduces this hang problem.
Btrfs has a kind of work queued as an ordered way, which means that its
ordered_func() must be processed in the way of FIFO, so it usually looks like --
normal_work_helper(arg)
work = container_of(arg, struct btrfs_work, normal_work);
work->func() <---- (we name it work X)
for ordered_work in wq->ordered_list
ordered_work->ordered_func()
ordered_work->ordered_free()
The hang is a rare case, first when we find free space, we get an uncached block
group, then we go to read its free space cache inode for free space information,
so it will
file a readahead request
btrfs_readpages()
for page that is not in page cache
__do_readpage()
submit_extent_page()
btrfs_submit_bio_hook()
btrfs_bio_wq_end_io()
submit_bio()
end_workqueue_bio() <--(ret by the 1st endio)
queue a work(named work Y) for the 2nd
also the real endio()
So the hang occurs when work Y's work_struct and work X's work_struct happens
to share the same address.
A bit more explanation,
A,B,C -- struct btrfs_work
arg -- struct work_struct
kthread:
worker_thread()
pick up a work_struct from @worklist
process_one_work(arg)
worker->current_work = arg; <-- arg is A->normal_work
worker->current_func(arg)
normal_work_helper(arg)
A = container_of(arg, struct btrfs_work, normal_work);
A->func()
A->ordered_func()
A->ordered_free() <-- A gets freed
B->ordered_func()
submit_compressed_extents()
find_free_extent()
load_free_space_inode()
... <-- (the above readhead stack)
end_workqueue_bio()
btrfs_queue_work(work C)
B->ordered_free()
As if work A has a high priority in wq->ordered_list and there are more ordered
works queued after it, such as B->ordered_func(), its memory could have been
freed before normal_work_helper() returns, which means that kernel workqueue
code worker_thread() still has worker->current_work pointer to be work
A->normal_work's, ie. arg's address.
Meanwhile, work C is allocated after work A is freed, work C->normal_work
and work A->normal_work are likely to share the same address(I confirmed this
with ftrace output, so I'm not just guessing, it's rare though).
When another kthread picks up work C->normal_work to process, and finds our
kthread is processing it(see find_worker_executing_work()), it'll think
work C as a collision and skip then, which ends up nobody processing work C.
So the situation is that our kthread is waiting forever on work C.
Besides, there're other cases that can lead to deadlock, but the real problem
is that all btrfs workqueue shares one work->func, -- normal_work_helper,
so this makes each workqueue to have its own helper function, but only a
wraper pf normal_work_helper.
With this patch, I no long hit the above hang.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Chris Mason <clm@fb.com>
The original code allocated new chunks by the number of the writable devices
and missing devices to make sure that any RAID levels on a degraded FS continue
to be honored, but it introduced a problem that it stopped us to allocating
new chunks, the steps to reproduce is following:
# mkfs.btrfs -m raid1 -d raid1 -f <dev0> <dev1>
# mkfs.btrfs -f <dev1> //Removing <dev1> from the original fs
# mount -o degraded <dev0> <mnt>
# dd if=/dev/null of=<mnt>/tmpfile bs=1M
It is because we allocate new chunks only on the writable devices, if we take
the number of missing devices into account, and want to allocate new chunks
with higher RAID level, we will fail becaue we don't have enough writable
device. Fix it by ignoring the number of missing devices when allocating
new chunks.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
During its tree walk, btrfs_drop_snapshot() will skip any shared
subtrees it encounters. This is incorrect when we have qgroups
turned on as those subtrees need to have their contents
accounted. In particular, the case we're concerned with is when
removing our snapshot root leaves the subtree with only one root
reference.
In those cases we need to find the last remaining root and add
each extent in the subtree to the corresponding qgroup exclusive
counts.
This patch implements the shared subtree walk and a new qgroup
operation, BTRFS_QGROUP_OPER_SUB_SUBTREE. When an operation of
this type is encountered during qgroup accounting, we search for
any root references to that extent and in the case that we find
only one reference left, we go ahead and do the math on it's
exclusive counts.
Signed-off-by: Mark Fasheh <mfasheh@suse.de>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Before I extended the no_quota arg to btrfs_dec/inc_ref because I didn't
understand how snapshot delete was using it and assumed that we needed the
quota operations there. With Mark's work this has turned out to be not the
case, we _always_ need to use no_quota for btrfs_dec/inc_ref, so just drop the
argument and make __btrfs_mod_ref call it's process function with no_quota set
always. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
This percpu counter @total_bytes_pinned is introduced to skip unnecessary
operations of 'commit transaction', it accounts for those space we may free
but are stuck in delayed refs.
And we zero out @space_info->total_bytes_pinned every transaction period so
we have a better idea of how much space we'll actually free up by committing
this transaction. However, we do the 'zero out' part a little earlier, before
we actually unpin space, so we end up returning ENOSPC when we actually have
free space that's just unpinned from committing transaction.
xfstests/generic/074 complained then.
This fixes it by actually accounting the percpu pinned number when 'unpin',
and since it's protected by space_info->lock, the race is gone now.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
When we mounted the filesystem after the crash, we got the following
message:
BTRFS error (device xxx): block group xxxx has wrong amount of free space
BTRFS error (device xxx): failed to load free space cache for block group xxx
It is because we didn't update the metadata of the allocated space (in extent
tree) until the file data was written into the disk. During this time, there was
no information about the allocated spaces in either the extent tree nor the
free space cache. when we wrote out the free space cache at this time (commit
transaction), those spaces were lost. In fact, only the free space that is
used to store the file data had this problem, the others didn't because
the metadata of them is updated in the same transaction context.
There are many methods which can fix the above problem
- track the allocated space, and write it out when we write out the free
space cache
- account the size of the allocated space that is used to store the file
data, if the size is not zero, don't write out the free space cache.
The first one is complex and may make the performance drop down.
This patch chose the second method, we use a per-block-group variant to
account the size of that allocated space. Besides that, we also introduce
a per-block-group read-write semaphore to avoid the race between
the allocation and the free space cache write out.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
We are currently allocating space_info objects in an array when we
allocate space_info. When a user does something like:
# btrfs balance start -mconvert=raid1 -dconvert=raid1 /mnt
# btrfs balance start -mconvert=single -dconvert=single /mnt -f
# btrfs balance start -mconvert=raid1 -dconvert=raid1 /
We can end up with memory corruption since the kobject hasn't
been reinitialized properly and the name pointer was left set.
The rationale behind allocating them statically was to avoid
creating a separate kobject container that just contained the
raid type. It used the index in the array to determine the index.
Ultimately, though, this wastes more memory than it saves in all
but the most complex scenarios and introduces kobject lifetime
questions.
This patch allocates the kobjects dynamically instead. Note that
we also remove the kobject_get/put of the parent kobject since
kobject_add and kobject_del do that internally.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reported-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
Delayed extent operations are triggered during transaction commits.
The goal is to queue up a healthly batch of changes to the extent
allocation tree and run through them in bulk.
This farms them off to async helper threads. The goal is to have the
bulk of the delayed operations being done in the background, but this is
also important to limit our stack footprint.
Signed-off-by: Chris Mason <clm@fb.com>
I've noticed an extra line after "use no compression", but search
revealed much more in messages of more critical levels and rare errors.
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <clm@fb.com>
This exercises the various parts of the new qgroup accounting code. We do some
basic stuff and do some things with the shared refs to make sure all that code
works. I had to add a bunch of infrastructure because I needed to be able to
insert items into a fake tree without having to do all the hard work myself,
hopefully this will be usefull in the future. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently qgroups account for space by intercepting delayed ref updates to fs
trees. It does this by adding sequence numbers to delayed ref updates so that
it can figure out how the tree looked before the update so we can adjust the
counters properly. The problem with this is that it does not allow delayed refs
to be merged, so if you say are defragging an extent with 5k snapshots pointing
to it we will thrash the delayed ref lock because we need to go back and
manually merge these things together. Instead we want to process quota changes
when we know they are going to happen, like when we first allocate an extent, we
free a reference for an extent, we add new references etc. This patch
accomplishes this by only adding qgroup operations for real ref changes. We
only modify the sequence number when we need to lookup roots for bytenrs, this
reduces the amount of churn on the sequence number and allows us to merge
delayed refs as we add them most of the time. This patch encompasses a bunch of
architectural changes
1) qgroup ref operations: instead of tracking qgroup operations through the
delayed refs we simply add new ref operations whenever we notice that we need to
when we've modified the refs themselves.
2) tree mod seq: we no longer have this separation of major/minor counters.
this makes the sequence number stuff much more sane and we can remove some
locking that was needed to protect the counter.
3) delayed ref seq: we now read the tree mod seq number and use that as our
sequence. This means each new delayed ref doesn't have it's own unique sequence
number, rather whenever we go to lookup backrefs we inc the sequence number so
we can make sure to keep any new operations from screwing up our world view at
that given point. This allows us to merge delayed refs during runtime.
With all of these changes the delayed ref stuff is a little saner and the qgroup
accounting stuff no longer goes negative in some cases like it was before.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Before applying this patch, the task had to reclaim the metadata space
by itself if the metadata space was not enough. And When the task started
the space reclamation, all the other tasks which wanted to reserve the
metadata space were blocked. At some cases, they would be blocked for
a long time, it made the performance fluctuate wildly.
So we introduce the background metadata space reclamation, when the space
is about to be exhausted, we insert a reclaim work into the workqueue, the
worker of the workqueue helps us to reclaim the reserved space at the
background. By this way, the tasks needn't reclaim the space by themselves at
most cases, and even if the tasks have to reclaim the space or are blocked
for the space reclamation, they will get enough space more quickly.
Here is my test result(Tested by compilebench):
Memory: 2GB
CPU: 2Cores * 1CPU
Partition: 40GB(SSD)
Test command:
# compilebench -D <mnt> -m
Without this patch:
intial create total runs 30 avg 54.36 MB/s (user 0.52s sys 2.44s)
compile total runs 30 avg 123.72 MB/s (user 0.13s sys 1.17s)
read compiled tree total runs 3 avg 81.15 MB/s (user 0.74s sys 4.89s)
delete compiled tree total runs 30 avg 5.32 seconds (user 0.35s sys 4.37s)
With this patch:
intial create total runs 30 avg 59.80 MB/s (user 0.52s sys 2.53s)
compile total runs 30 avg 151.44 MB/s (user 0.13s sys 1.11s)
read compiled tree total runs 3 avg 83.25 MB/s (user 0.76s sys 4.91s)
delete compiled tree total runs 30 avg 5.29 seconds (user 0.34s sys 4.34s)
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
If we had to retry on the profiles seqlock (due to a concurrent write), we
would set bits on the input flags that corresponded both to the current
profile and to previous values of the profile.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Chris Mason <clm@fb.com>
If skinny metadata is enabled and our first tree search fails to find a
skinny extent item, we may repeat a tree search for a "fat" extent item
(if the previous item in the leaf is not the "fat" extent we're looking
for). However we were not setting the new key's objectid to the right
value, as we previously used the same key variable to peek at the previous
item in the leaf, which has a different objectid. So just set the right
objectid to avoid modifying/deleting a wrong item if we repeat the tree
search.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Chris Mason <clm@fb.com>
I'm not sure why we weren't aborting here in the first place, it is obviously a
bad time from the fact that we print the leaf and yell loudly about it. Fix
this up, otherwise we panic because our path could be pointing into oblivion.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
When encountering memory pressure, testers have run into the following
lockdep warning. It was caused by __link_block_group calling kobject_add
with the groups_sem held. kobject_add calls kvasprintf with GFP_KERNEL,
which gets us into reclaim context. The kobject doesn't actually need
to be added under the lock -- it just needs to ensure that it's only
added for the first block group to be linked.
=========================================================
[ INFO: possible irq lock inversion dependency detected ]
3.14.0-rc8-default #1 Not tainted
---------------------------------------------------------
kswapd0/169 just changed the state of lock:
(&delayed_node->mutex){+.+.-.}, at: [<ffffffffa018baea>] __btrfs_release_delayed_node+0x3a/0x200 [btrfs]
but this lock took another, RECLAIM_FS-unsafe lock in the past:
(&found->groups_sem){+++++.}
and interrupts could create inverse lock ordering between them.
other info that might help us debug this:
Possible interrupt unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&found->groups_sem);
local_irq_disable();
lock(&delayed_node->mutex);
lock(&found->groups_sem);
<Interrupt>
lock(&delayed_node->mutex);
*** DEADLOCK ***
2 locks held by kswapd0/169:
#0: (shrinker_rwsem){++++..}, at: [<ffffffff81159e8a>] shrink_slab+0x3a/0x160
#1: (&type->s_umount_key#27){++++..}, at: [<ffffffff811bac6f>] grab_super_passive+0x3f/0x90
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
Lets try this again. We can deadlock the box if we send on a box and try to
write onto the same fs with the app that is trying to listen to the send pipe.
This is because the writer could get stuck waiting for a transaction commit
which is being blocked by the send. So fix this by making sure looking at the
commit roots is always going to be consistent. We do this by keeping track of
which roots need to have their commit roots swapped during commit, and then
taking the commit_root_sem and swapping them all at once. Then make sure we
take a read lock on the commit_root_sem in cases where we search the commit root
to make sure we're always looking at a consistent view of the commit roots.
Previously we had problems with this because we would swap a fs tree commit root
and then swap the extent tree commit root independently which would cause the
backref walking code to screw up sometimes. With this patch we no longer
deadlock and pass all the weird send/receive corner cases. Thanks,
Reportedy-by: Hugo Mills <hugo@carfax.org.uk>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We could have possibly added an extent_op to the locked_ref while we dropped
locked_ref->lock, so check for this case as well and loop around. Otherwise we
could lose flag updates which would lead to extent tree corruption. Thanks,
cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We needn't flush all delalloc inodes when we doesn't get s_umount lock,
or we would make the tasks wait for a long time.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
generic/074 in xfstests failed sometimes because of the enospc error,
the reason of this problem is that we just reclaimed the space we need
from the reserved space for delalloc, and then tried to reserve the space,
but if some task did no-flush reservation between the above reclamation
and reservation,
Task1 Task2
shrink_delalloc()
reclaim 1 block
(The space that can
be reserved now is 1
block)
do no-flush reservation
reserve 1 block
(The space that can
be reserved now is 0
block)
reserving 1 block failed
the reservation of Task1 failed, but in fact, there was enough space to
reserve if we could reclaim more space before.
Fix this problem by the aggressive reclamation of the reserved delalloc
metadata space.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
The reason is:
- The per-cpu counter has its own lock to protect itself.
- Here we needn't get a exact value.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
If the snapshot creation happened after the nocow write but before the dirty
data flush, we would fail to flush the dirty data because of no space.
So we must keep track of when those nocow write operations start and when they
end, if there are nocow writers, the snapshot creators must wait. In order
to implement this function, I introduce btrfs_{start, end}_nocow_write(),
which is similar to mnt_{want,drop}_write().
These two functions are only used for nocow file write operations.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Since the "_struct" suffix is mainly used for distinguish the differnt
btrfs_work between the original and the newly created one,
there is no need using the suffix since all btrfs_workers are changed
into btrfs_workqueue.
Also this patch fixed some codes whose code style is changed due to the
too long "_struct" suffix.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Replace the fs_info->cache_workers with the newly created
btrfs_workqueue.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
A user reported a 100% cpu hang with my new delayed ref code. Turns out I
forgot to increase the count check when we can't run a delayed ref because of
the tree mod log. If we can't run any delayed refs during this there is no
point in continuing to look, and we need to break out. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We allocate the free space from the former block group, not the current
one, so should use the former one to output the trace information.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
used_block_group is just used for the space cluster which doesn't
belong to the current block group, the other place needn't use it.
Or the logic of code seems unclear.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
After the change titled "Btrfs: add support for inode properties", if
btrfs was built-in the kernel (i.e. not as a module), it would cause a
kernel panic, as reported recently by Fengguang:
[ 2.024722] BUG: unable to handle kernel NULL pointer dereference at (null)
[ 2.027814] IP: [<ffffffff81501594>] crc32c+0xc/0x6b
[ 2.028684] PGD 0
[ 2.028684] Oops: 0000 [#1] SMP
[ 2.028684] Modules linked in:
[ 2.028684] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.13.0-rc7-04795-ga7b57c2 #1
[ 2.028684] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[ 2.028684] task: ffff88000edba100 ti: ffff88000edd6000 task.ti: ffff88000edd6000
[ 2.028684] RIP: 0010:[<ffffffff81501594>] [<ffffffff81501594>] crc32c+0xc/0x6b
[ 2.028684] RSP: 0000:ffff88000edd7e58 EFLAGS: 00010246
[ 2.028684] RAX: 0000000000000000 RBX: ffffffff82295550 RCX: 0000000000000000
[ 2.028684] RDX: 0000000000000011 RSI: ffffffff81efe393 RDI: 00000000fffffffe
[ 2.028684] RBP: ffff88000edd7e60 R08: 0000000000000003 R09: 0000000000015d20
[ 2.028684] R10: ffffffff81ef225e R11: ffffffff811b0222 R12: ffffffffffffffff
[ 2.028684] R13: 0000000000000239 R14: 0000000000000000 R15: 0000000000000000
[ 2.028684] FS: 0000000000000000(0000) GS:ffff88000fa00000(0000) knlGS:0000000000000000
[ 2.028684] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 2.028684] CR2: 0000000000000000 CR3: 000000000220c000 CR4: 00000000000006f0
[ 2.028684] Stack:
[ 2.028684] ffffffff82295550 ffff88000edd7e80 ffffffff8238af62 ffffffff8238ac05
[ 2.028684] 0000000000000000 ffff88000edd7e98 ffffffff8238ac0f ffffffff8238ac05
[ 2.028684] ffff88000edd7f08 ffffffff810002ba ffff88000edd7f00 ffffffff810e2404
[ 2.028684] Call Trace:
[ 2.028684] [<ffffffff8238af62>] btrfs_props_init+0x4f/0x96
[ 2.028684] [<ffffffff8238ac05>] ? ftrace_define_fields_btrfs_space_reservation+0x145/0x145
[ 2.028684] [<ffffffff8238ac0f>] init_btrfs_fs+0xa/0xf0
[ 2.028684] [<ffffffff8238ac05>] ? ftrace_define_fields_btrfs_space_reservation+0x145/0x145
[ 2.028684] [<ffffffff810002ba>] do_one_initcall+0xa4/0x13a
[ 2.028684] [<ffffffff810e2404>] ? parse_args+0x25f/0x33d
[ 2.028684] [<ffffffff8234cf75>] kernel_init_freeable+0x1aa/0x230
[ 2.028684] [<ffffffff8234c785>] ? do_early_param+0x88/0x88
[ 2.028684] [<ffffffff819f61b5>] ? rest_init+0x89/0x89
[ 2.028684] [<ffffffff819f61c3>] kernel_init+0xe/0x109
The issue here is that the initialization function of btrfs (super.c:init_btrfs_fs)
started using crc32c (from lib/libcrc32c.c). But when it needs to call crc32c (as
part of the properties initialization routine), the libcrc32c is not yet initialized,
so crc32c derreferenced a NULL pointer (lib/libcrc32c.c:tfm), causing the kernel
panic on boot.
The approach to fix this is to use crypto component directly to use its crc32c (which
is basically what lib/libcrc32c.c is, a wrapper around crypto). This is what ext4 is
doing as well, it uses crypto directly to get crc32c functionality.
Verified this works both when btrfs is built-in and when it's loadable kernel module.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
On one of our gluster clusters we noticed some pretty big lag spikes. This
turned out to be because our transaction commit was taking like 3 minutes to
complete. This is because we have like 30 gigs of metadata, so our global
reserve would end up being the max which is like 512 mb. So our throttling code
would allow a ridiculous amount of delayed refs to build up and then they'd all
get run at transaction commit time, and for a cold mounted file system that
could take up to 3 minutes to run. So fix the throttling to be based on both
the size of the global reserve and how long it takes us to run delayed refs.
This patch tracks the time it takes to run delayed refs and then only allows 1
seconds worth of outstanding delayed refs at a time. This way it will auto-tune
itself from cold cache up to when everything is in memory and it no longer has
to go to disk. This makes our transaction commits take much less time to run.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Currently we have two rb-trees, one for delayed ref heads and one for all of the
delayed refs, including the delayed ref heads. When we process the delayed refs
we have to hold onto the delayed ref lock for all of the selecting and merging
and such, which results in quite a bit of lock contention. This was solved by
having a waitqueue and only one flusher at a time, however this hurts if we get
a lot of delayed refs queued up.
So instead just have an rb tree for the delayed ref heads, and then attach the
delayed ref updates to an rb tree that is per delayed ref head. Then we only
need to take the delayed ref lock when adding new delayed refs and when
selecting a delayed ref head to process, all the rest of the time we deal with a
per delayed ref head lock which will be much less contentious.
The locking rules for this get a little more complicated since we have to lock
up to 3 things to properly process delayed refs, but I will address that problem
later. For now this passes all of xfstests and my overnight stress tests.
Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We may return early in btrfs_drop_snapshot(), we shouldn't
call btrfs_std_err() for this case, fix it.
Cc: stable@vger.kernel.org
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
@full is not protected within global_rsv.lock, so we may think global_rsv
is already full but in fact it's not, so we miss the opportunity to return
free space to global_rsv directly when we release other block_rsvs.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We can starve out the transaction commit with a bunch of caching threads all
running at the same time. This is because we will only drop the
extent_commit_sem if we need_resched(), which isn't likely to happen since we
will be reading a lot from the disk so have already schedule()'ed plenty. Alex
observed that he could starve out a transaction commit for up to a minute with
32 caching threads all running at once. This will allow us to drop the
extent_commit_sem to allow the transaction commit to swap the commit_root out
and then all the cachers will start back up. Here is an explanation provided by
Igno
So, just to fill in what happens in this loop:
mutex_unlock(&caching_ctl->mutex);
cond_resched();
goto again;
where 'again:' takes caching_ctl->mutex and fs_info->extent_commit_sem
again:
again:
mutex_lock(&caching_ctl->mutex);
/* need to make sure the commit_root doesn't disappear */
down_read(&fs_info->extent_commit_sem);
So, if I'm reading the code correct, there can be a fair amount of
concurrency here: there may be multiple 'caching kthreads' per filesystem
active, while there's one fs_info->extent_commit_sem per filesystem
AFAICS.
So, what happens if there are a lot of CPUs all busy holding the
->extent_commit_sem rwsem read-locked and a writer arrives? They'd all
rush to try to release the fs_info->extent_commit_sem, and they'd block in
the down_read() because there's a writer waiting.
So there's a guarantee of forward progress. This should answer akpm's
concern I think.
Thanks,
Acked-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <clm@fb.com>
Convert all applicable cases of printk and pr_* to the btrfs_* macros.
Fix all uses of the BTRFS prefix.
Signed-off-by: Frank Holton <fholton@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
We met the following oops when doing space balance:
kobject (ffff88081b590278): tried to init an initialized object, something is seriously wrong.
...
Call Trace:
[<ffffffff81937262>] dump_stack+0x49/0x5f
[<ffffffff8137d259>] kobject_init+0x89/0xa0
[<ffffffff8137d36a>] kobject_init_and_add+0x2a/0x70
[<ffffffffa009bd79>] ? clear_extent_bit+0x199/0x470 [btrfs]
[<ffffffffa005e82c>] __link_block_group+0xfc/0x120 [btrfs]
[<ffffffffa006b9db>] btrfs_make_block_group+0x24b/0x370 [btrfs]
[<ffffffffa00a899b>] __btrfs_alloc_chunk+0x54b/0x7e0 [btrfs]
[<ffffffffa00a8c6f>] btrfs_alloc_chunk+0x3f/0x50 [btrfs]
[<ffffffffa0060123>] do_chunk_alloc+0x363/0x440 [btrfs]
[<ffffffffa00633d4>] btrfs_check_data_free_space+0x104/0x310 [btrfs]
[<ffffffffa0069f4d>] btrfs_write_dirty_block_groups+0x48d/0x600 [btrfs]
[<ffffffffa007aad4>] commit_cowonly_roots+0x184/0x250 [btrfs]
...
Steps to reproduce:
# mkfs.btrfs -f <dev>
# mount -o nospace_cache <dev> <mnt>
# btrfs balance start <mnt>
# dd if=/dev/zero of=<mnt>/tmpfile bs=1M count=1
The reason of this problem is that we initialized the raid kobject when we added
a block group into a empty raid list. As we know, when we mounted a btrfs filesystem,
the raid list was empty, we would initialize the raid kobject when we added the first
block group. But if there was not data stored in the block group, the block group
would be freed when doing balance, and the raid list would be empty. And then if we
allocated a new block group and added it into the raid list, we would initialize
the raid kobject again, the oops happened.
Fix this problem by initializing the raid kobject just when mounting the fs.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Reported-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
This patch fixes the following warnings:
fs/btrfs/extent-tree.c:6201:12: sparse: symbol 'get_raid_name' was not declared. Should it be static?
fs/btrfs/extent-tree.c:8430:9: error: format not a string literal and no format arguments [-Werror=format-security] get_raid_name(index));
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
The variable found_uncached_bg in find_free_extent is not used since commit
285ff5af6c
(Btrfs: remove the ideal caching code)
Signed-off-by: Valentina Giusti <valentina.giusti@microon.de>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <clm@fb.com>