Ordered work functions aren't guaranteed to be handled by the same thread
which executed the normal work functions. The only way execution between
normal/ordered functions is synchronized is via the WORK_DONE_BIT,
unfortunately the used bitops don't guarantee any ordering whatsoever.
This manifested as seemingly inexplicable crashes on ARM64, where
async_chunk::inode is seen as non-null in async_cow_submit which causes
submit_compressed_extents to be called and crash occurs because
async_chunk::inode suddenly became NULL. The call trace was similar to:
pc : submit_compressed_extents+0x38/0x3d0
lr : async_cow_submit+0x50/0xd0
sp : ffff800015d4bc20
<registers omitted for brevity>
Call trace:
submit_compressed_extents+0x38/0x3d0
async_cow_submit+0x50/0xd0
run_ordered_work+0xc8/0x280
btrfs_work_helper+0x98/0x250
process_one_work+0x1f0/0x4ac
worker_thread+0x188/0x504
kthread+0x110/0x114
ret_from_fork+0x10/0x18
Fix this by adding respective barrier calls which ensure that all
accesses preceding setting of WORK_DONE_BIT are strictly ordered before
setting the flag. At the same time add a read barrier after reading of
WORK_DONE_BIT in run_ordered_work which ensures all subsequent loads
would be strictly ordered after reading the bit. This in turn ensures
are all accesses before WORK_DONE_BIT are going to be strictly ordered
before any access that can occur in ordered_func.
Reported-by: Chris Murphy <lists@colorremedies.com>
Fixes: 08a9ff3264 ("btrfs: Added btrfs_workqueue_struct implemented ordered execution based on kernel workqueue")
CC: stable@vger.kernel.org # 4.4+
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2011928
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Tested-by: Chris Murphy <chris@colorremedies.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
During unmount we can have a job from the delayed inode items work queue
still running, that can lead to at least two bad things:
1) A crash, because the worker can try to create a transaction just
after the fs roots were freed;
2) A transaction leak, because the worker can create a transaction
before the fs roots are freed and just after we committed the last
transaction and after we stopped the transaction kthread.
A stack trace example of the crash:
[79011.691214] kernel BUG at lib/radix-tree.c:982!
[79011.692056] invalid opcode: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI
[79011.693180] CPU: 3 PID: 1394 Comm: kworker/u8:2 Tainted: G W 5.6.0-rc2-btrfs-next-54 #2
(...)
[79011.696789] Workqueue: btrfs-delayed-meta btrfs_work_helper [btrfs]
[79011.697904] RIP: 0010:radix_tree_tag_set+0xe7/0x170
(...)
[79011.702014] RSP: 0018:ffffb3c84a317ca0 EFLAGS: 00010293
[79011.702949] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[79011.704202] RDX: ffffb3c84a317cb0 RSI: ffffb3c84a317ca8 RDI: ffff8db3931340a0
[79011.705463] RBP: 0000000000000005 R08: 0000000000000005 R09: ffffffff974629d0
[79011.706756] R10: ffffb3c84a317bc0 R11: 0000000000000001 R12: ffff8db393134000
[79011.708010] R13: ffff8db3931340a0 R14: ffff8db393134068 R15: 0000000000000001
[79011.709270] FS: 0000000000000000(0000) GS:ffff8db3b6a00000(0000) knlGS:0000000000000000
[79011.710699] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[79011.711710] CR2: 00007f22c2a0a000 CR3: 0000000232ad4005 CR4: 00000000003606e0
[79011.712958] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[79011.714205] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[79011.715448] Call Trace:
[79011.715925] record_root_in_trans+0x72/0xf0 [btrfs]
[79011.716819] btrfs_record_root_in_trans+0x4b/0x70 [btrfs]
[79011.717925] start_transaction+0xdd/0x5c0 [btrfs]
[79011.718829] btrfs_async_run_delayed_root+0x17e/0x2b0 [btrfs]
[79011.719915] btrfs_work_helper+0xaa/0x720 [btrfs]
[79011.720773] process_one_work+0x26d/0x6a0
[79011.721497] worker_thread+0x4f/0x3e0
[79011.722153] ? process_one_work+0x6a0/0x6a0
[79011.722901] kthread+0x103/0x140
[79011.723481] ? kthread_create_worker_on_cpu+0x70/0x70
[79011.724379] ret_from_fork+0x3a/0x50
(...)
The following diagram shows a sequence of steps that lead to the crash
during ummount of the filesystem:
CPU 1 CPU 2 CPU 3
btrfs_punch_hole()
btrfs_btree_balance_dirty()
btrfs_balance_delayed_items()
--> sees
fs_info->delayed_root->items
with value 200, which is greater
than
BTRFS_DELAYED_BACKGROUND (128)
and smaller than
BTRFS_DELAYED_WRITEBACK (512)
btrfs_wq_run_delayed_node()
--> queues a job for
fs_info->delayed_workers to run
btrfs_async_run_delayed_root()
btrfs_async_run_delayed_root()
--> job queued by CPU 1
--> starts picking and running
delayed nodes from the
prepare_list list
close_ctree()
btrfs_delete_unused_bgs()
btrfs_commit_super()
btrfs_join_transaction()
--> gets transaction N
btrfs_commit_transaction(N)
--> set transaction state
to TRANTS_STATE_COMMIT_START
btrfs_first_prepared_delayed_node()
--> picks delayed node X through
the prepared_list list
btrfs_run_delayed_items()
btrfs_first_delayed_node()
--> also picks delayed node X
but through the node_list
list
__btrfs_commit_inode_delayed_items()
--> runs all delayed items from
this node and drops the
node's item count to 0
through call to
btrfs_release_delayed_inode()
--> finishes running any remaining
delayed nodes
--> finishes transaction commit
--> stops cleaner and transaction threads
btrfs_free_fs_roots()
--> frees all roots and removes them
from the radix tree
fs_info->fs_roots_radix
btrfs_join_transaction()
start_transaction()
btrfs_record_root_in_trans()
record_root_in_trans()
radix_tree_tag_set()
--> crashes because
the root is not in
the radix tree
anymore
If the worker is able to call btrfs_join_transaction() before the unmount
task frees the fs roots, we end up leaking a transaction and all its
resources, since after the call to btrfs_commit_super() and stopping the
transaction kthread, we don't expect to have any transaction open anymore.
When this situation happens the worker has a delayed node that has no
more items to run, since the task calling btrfs_run_delayed_items(),
which is doing a transaction commit, picks the same node and runs all
its items first.
We can not wait for the worker to complete when running delayed items
through btrfs_run_delayed_items(), because we call that function in
several phases of a transaction commit, and that could cause a deadlock
because the worker calls btrfs_join_transaction() and the task doing the
transaction commit may have already set the transaction state to
TRANS_STATE_COMMIT_DOING.
Also it's not possible to get into a situation where only some of the
items of a delayed node are added to the fs/subvolume tree in the current
transaction and the remaining ones in the next transaction, because when
running the items of a delayed inode we lock its mutex, effectively
waiting for the worker if the worker is running the items of the delayed
node already.
Since this can only cause issues when unmounting a filesystem, fix it in
a simple way by waiting for any jobs on the delayed workers queue before
calling btrfs_commit_supper() at close_ctree(). This works because at this
point no one can call btrfs_btree_balance_dirty() or
btrfs_balance_delayed_items(), and if we end up waiting for any worker to
complete, btrfs_commit_super() will commit the transaction created by the
worker.
CC: stable@vger.kernel.org # 4.4+
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The attribute is more relaxed than const and the functions could
dereference pointers, as long as the observable state is not changed. We
do have such functions, based on -Wsuggest-attribute=pure .
The visible effects of this patch are negligible, there are differences
in the assembly but hard to summarize.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit ac0c7cf8be ("btrfs: fix crash when tracepoint arguments are
freed by wq callbacks") added a void pointer, wtag, which is passed into
trace_btrfs_all_work_done() instead of the freed work item. This is
silly for a few reasons:
1. The freed work item still has the same address.
2. work is still in scope after it's freed, so assigning wtag doesn't
stop anyone from using it.
3. The tracepoint has always taken a void * argument, so assigning wtag
doesn't actually make things any more type-safe. (Note that the
original bug in commit bc074524e1 ("btrfs: prefix fsid to all trace
events") was that the void * was implicitly casted when it was passed
to btrfs_work_owner() in the trace point itself).
Instead, let's add some clearer warnings as comments.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Commit 9e0af23764 ("Btrfs: fix task hang under heavy compressed
write") worked around the issue that a recycled work item could get a
false dependency on the original work item due to how the workqueue code
guarantees non-reentrancy. It did so by giving different work functions
to different types of work.
However, the fixes in the previous few patches are more complete, as
they prevent a work item from being recycled at all (except for a tiny
window that the kernel workqueue code handles for us). This obsoletes
the previous fix, so we don't need the unique helpers for correctness.
The only other reason to keep them would be so they show up in stack
traces, but they always seem to be optimized to a tail call, so they
don't show up anyways. So, let's just get rid of the extra indirection.
While we're here, rename normal_work_helper() to the more informative
btrfs_work_helper().
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
We hit the following very strange deadlock on a system with Btrfs on a
loop device backed by another Btrfs filesystem:
1. The top (loop device) filesystem queues an async_cow work item from
cow_file_range_async(). We'll call this work X.
2. Worker thread A starts work X (normal_work_helper()).
3. Worker thread A executes the ordered work for the top filesystem
(run_ordered_work()).
4. Worker thread A finishes the ordered work for work X and frees X
(work->ordered_free()).
5. Worker thread A executes another ordered work and gets blocked on I/O
to the bottom filesystem (still in run_ordered_work()).
6. Meanwhile, the bottom filesystem allocates and queues an async_cow
work item which happens to be the recently-freed X.
7. The workqueue code sees that X is already being executed by worker
thread A, so it schedules X to be executed _after_ worker thread A
finishes (see the find_worker_executing_work() call in
process_one_work()).
Now, the top filesystem is waiting for I/O on the bottom filesystem, but
the bottom filesystem is waiting for the top filesystem to finish, so we
deadlock.
This happens because we are breaking the workqueue assumption that a
work item cannot be recycled while it still depends on other work. Fix
it by waiting to free the work item until we are done with all of the
related ordered work.
P.S.:
One might ask why the workqueue code doesn't try to detect a recycled
work item. It actually does try by checking whether the work item has
the same work function (find_worker_executing_work()), but in our case
the function is the same. This is the only key that the workqueue code
has available to compare, short of adding an additional, layer-violating
"custom key". Considering that we're the only ones that have ever hit
this, we should just play by the rules.
Unfortunately, we haven't been able to create a minimal reproducer other
than our full container setup using a compress-force=zstd filesystem on
top of another compress-force=zstd filesystem.
Suggested-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
The workqueue name is constructed from a format string but the prefix
does not need to be set by %s.
Reviewed-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Remove GPL boilerplate text (long, short, one-line) and keep the rest,
ie. personal, company or original source copyright statements. Add the
SPDX header.
Signed-off-by: David Sterba <dsterba@suse.com>
We've seen the following backtrace stack in ftrace or dmesg log,
kworker/u16:10-4244 [000] 241942.480955: function: btrfs_put_ordered_extent
kworker/u16:10-4244 [000] 241942.480956: kernel_stack: <stack trace>
=> finish_ordered_fn (ffffffffa0384475)
=> btrfs_scrubparity_helper (ffffffffa03ca577) <-----"incorrect"
=> btrfs_freespace_write_helper (ffffffffa03ca98e) <-----"correct"
=> process_one_work (ffffffff81117b2f)
=> worker_thread (ffffffff81118c2a)
=> kthread (ffffffff81121de0)
=> ret_from_fork (ffffffff81d7087a)
btrfs_freespace_write_helper is actually calling normal_worker_helper
instead of btrfs_scrubparity_helper, so somehow kernel has parsed the
incorrect function address while unwinding the stack,
btrfs_scrubparity_helper really shouldn't be shown up.
It's caused by compiler doing inline for our helper function, adding a
noinline tag can fix that.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ use noinline_for_stack ]
Signed-off-by: David Sterba <dsterba@suse.com>
Tracepoint arguments are all read-only. If we mark the arguments
as const, we're able to keep or convert those arguments to const
where appropriate.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Enabling btrfs tracepoints leads to instant crash, as reported. The wq
callbacks could free the memory and the tracepoints started to
dereference the members to get to fs_info.
The proposed fix https://marc.info/?l=linux-btrfs&m=148172436722606&w=2
removed the tracepoints but we could preserve them by passing only the
required data in a safe way.
Fixes: bc074524e1 ("btrfs: prefix fsid to all trace events")
CC: stable@vger.kernel.org # 4.8+
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Problem statement: unprivileged user who has read-write access to more than
one btrfs subvolume may easily consume all kernel memory (eventually
triggering oom-killer).
Reproducer (./mkrmdir below essentially loops over mkdir/rmdir):
[root@kteam1 ~]# cat prep.sh
DEV=/dev/sdb
mkfs.btrfs -f $DEV
mount $DEV /mnt
for i in `seq 1 16`
do
mkdir /mnt/$i
btrfs subvolume create /mnt/SV_$i
ID=`btrfs subvolume list /mnt |grep "SV_$i$" |cut -d ' ' -f 2`
mount -t btrfs -o subvolid=$ID $DEV /mnt/$i
chmod a+rwx /mnt/$i
done
[root@kteam1 ~]# sh prep.sh
[maxim@kteam1 ~]$ for i in `seq 1 16`; do ./mkrmdir /mnt/$i 2000 2000 & done
[root@kteam1 ~]# for i in `seq 1 4`; do grep "kmalloc-128" /proc/slabinfo | grep -v dma; sleep 60; done
kmalloc-128 10144 10144 128 32 1 : tunables 0 0 0 : slabdata 317 317 0
kmalloc-128 9992352 9992352 128 32 1 : tunables 0 0 0 : slabdata 312261 312261 0
kmalloc-128 24226752 24226752 128 32 1 : tunables 0 0 0 : slabdata 757086 757086 0
kmalloc-128 42754240 42754240 128 32 1 : tunables 0 0 0 : slabdata 1336070 1336070 0
The huge numbers above come from insane number of async_work-s allocated
and queued by btrfs_wq_run_delayed_node.
The problem is caused by btrfs_wq_run_delayed_node() queuing more and more
works if the number of delayed items is above BTRFS_DELAYED_BACKGROUND. The
worker func (btrfs_async_run_delayed_root) processes at least
BTRFS_DELAYED_BATCH items (if they are present in the list). So, the machinery
works as expected while the list is almost empty. As soon as it is getting
bigger, worker func starts to process more than one item at a time, it takes
longer, and the chances to have async_works queued more than needed is getting
higher.
The problem above is worsened by another flaw of delayed-inode implementation:
if async_work was queued in a throttling branch (number of items >=
BTRFS_DELAYED_WRITEBACK), corresponding worker func won't quit until
the number of items < BTRFS_DELAYED_BACKGROUND / 2. So, it is possible that
the func occupies CPU infinitely (up to 30sec in my experiments): while the
func is trying to drain the list, the user activity may add more and more
items to the list.
The patch fixes both problems in straightforward way: refuse queuing too
many works in btrfs_wq_run_delayed_node and bail out of worker func if
at least BTRFS_DELAYED_WRITEBACK items are processed.
Changed in v2: remove support of thresh == NO_THRESHOLD.
Signed-off-by: Maxim Patlasov <mpatlasov@virtuozzo.com>
Signed-off-by: Chris Mason <clm@fb.com>
Cc: stable@vger.kernel.org # v3.15+
In order to provide an fsid for trace events, we'll need a btrfs_fs_info
pointer. The most lightweight way to do that for btrfs_work structures
is to associate it with the __btrfs_workqueue structure. Each queued
btrfs_work structure has a workqueue associated with it, so that's
a natural fit. It's a privately defined structures, so we add accessors
to retrieve the fs_info pointer.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Parameter of trace_btrfs_work_queued() can be freed in its workqueue.
So no one use use that pointer after queue_work().
Fix the user-after-free bug by move the trace line before queue_work().
Reported-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
We don't have to use GFP_NOFS to allocate workqueue structures, this is
done from mount context or potentially scrub start context, safe to fail
in both cases.
Signed-off-by: David Sterba <dsterba@suse.com>
At initializing time, for threshold-able workqueue, it's max_active
of kernel workqueue should be 1 and grow if it hits threshold.
But due to the bad naming, there is both 'max_active' for kernel
workqueue and btrfs workqueue.
So wrong value is given at workqueue initialization.
This patch fixes it, and to avoid further misunderstanding, change the
member name of btrfs_workqueue to 'current_active' and 'limit_active'.
Also corresponding comment is added for readability.
Reported-by: Alex Lyakas <alex.btrfs@zadarastorage.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
lockdep report following warning in test:
[25176.843958] =================================
[25176.844519] [ INFO: inconsistent lock state ]
[25176.845047] 4.1.0-rc3 #22 Tainted: G W
[25176.845591] ---------------------------------
[25176.846153] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
[25176.846713] fsstress/26661 [HC0[0]:SC1[1]:HE1:SE0] takes:
[25176.847246] (&wr_ctx->wr_lock){+.?...}, at: [<ffffffffa04cdc6d>] scrub_free_ctx+0x2d/0xf0 [btrfs]
[25176.847838] {SOFTIRQ-ON-W} state was registered at:
[25176.848396] [<ffffffff810bf460>] __lock_acquire+0x6a0/0xe10
[25176.848955] [<ffffffff810bfd1e>] lock_acquire+0xce/0x2c0
[25176.849491] [<ffffffff816489af>] mutex_lock_nested+0x7f/0x410
[25176.850029] [<ffffffffa04d04ff>] scrub_stripe+0x4df/0x1080 [btrfs]
[25176.850575] [<ffffffffa04d11b1>] scrub_chunk.isra.19+0x111/0x130 [btrfs]
[25176.851110] [<ffffffffa04d144c>] scrub_enumerate_chunks+0x27c/0x510 [btrfs]
[25176.851660] [<ffffffffa04d3b87>] btrfs_scrub_dev+0x1c7/0x6c0 [btrfs]
[25176.852189] [<ffffffffa04e918e>] btrfs_dev_replace_start+0x36e/0x450 [btrfs]
[25176.852771] [<ffffffffa04a98e0>] btrfs_ioctl+0x1e10/0x2d20 [btrfs]
[25176.853315] [<ffffffff8121c5b8>] do_vfs_ioctl+0x318/0x570
[25176.853868] [<ffffffff8121c851>] SyS_ioctl+0x41/0x80
[25176.854406] [<ffffffff8164da17>] system_call_fastpath+0x12/0x6f
[25176.854935] irq event stamp: 51506
[25176.855511] hardirqs last enabled at (51506): [<ffffffff810d4ce5>] vprintk_emit+0x225/0x5e0
[25176.856059] hardirqs last disabled at (51505): [<ffffffff810d4b77>] vprintk_emit+0xb7/0x5e0
[25176.856642] softirqs last enabled at (50886): [<ffffffff81067a23>] __do_softirq+0x363/0x640
[25176.857184] softirqs last disabled at (50949): [<ffffffff8106804d>] irq_exit+0x10d/0x120
[25176.857746]
other info that might help us debug this:
[25176.858845] Possible unsafe locking scenario:
[25176.859981] CPU0
[25176.860537] ----
[25176.861059] lock(&wr_ctx->wr_lock);
[25176.861705] <Interrupt>
[25176.862272] lock(&wr_ctx->wr_lock);
[25176.862881]
*** DEADLOCK ***
Reason:
Above warning is caused by:
Interrupt
-> bio_endio()
-> ...
-> scrub_put_ctx()
-> scrub_free_ctx() *1
-> ...
-> mutex_lock(&wr_ctx->wr_lock);
scrub_put_ctx() is allowed to be called in end_bio interrupt, but
in code design, it will never call scrub_free_ctx(sctx) in interrupe
context(above *1), because btrfs_scrub_dev() get one additional
reference of sctx->refs, which makes scrub_free_ctx() only called
withine btrfs_scrub_dev().
Now the code runs out of our wish, because free sequence in
scrub_pending_bio_dec() have a gap.
Current code:
-----------------------------------+-----------------------------------
scrub_pending_bio_dec() | btrfs_scrub_dev
-----------------------------------+-----------------------------------
atomic_dec(&sctx->bios_in_flight); |
wake_up(&sctx->list_wait); |
| scrub_put_ctx()
| -> atomic_dec_and_test(&sctx->refs)
scrub_put_ctx(sctx); |
-> atomic_dec_and_test(&sctx->refs)|
-> scrub_free_ctx() |
-----------------------------------+-----------------------------------
We expected:
-----------------------------------+-----------------------------------
scrub_pending_bio_dec() | btrfs_scrub_dev
-----------------------------------+-----------------------------------
atomic_dec(&sctx->bios_in_flight); |
wake_up(&sctx->list_wait); |
scrub_put_ctx(sctx); |
-> atomic_dec_and_test(&sctx->refs)|
| scrub_put_ctx()
| -> atomic_dec_and_test(&sctx->refs)
| -> scrub_free_ctx()
-----------------------------------+-----------------------------------
Fix:
Move scrub_pending_bio_dec() to a workqueue, to avoid this function run
in interrupt context.
Tested by check tracelog in debug.
Changelog v1->v2:
Use workqueue instead of adjust function call sequence in v1,
because v1 will introduce a bug pointed out by:
Filipe David Manana <fdmanana@gmail.com>
Reported-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
This patch implement data repair function when direct read fails.
The detail of the implementation is:
- When we find the data is not right, we try to read the data from the other
mirror.
- When the io on the mirror ends, we will insert the endio work into the
dedicated btrfs workqueue, not common read endio workqueue, because the
original endio work is still blocked in the btrfs endio workqueue, if we
insert the endio work of the io on the mirror into that workqueue, deadlock
would happen.
- After we get right data, we write it back to the corrupted mirror.
- And if the data on the new mirror is still corrupted, we will try next
mirror until we read right data or all the mirrors are traversed.
- After the above work, we set the uptodate flag according to the result.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
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>
Reproducer:
mount /dev/ubda /mnt
mount -oremount,thread_pool=42 /mnt
Gives a crash:
? btrfs_workqueue_set_max+0x0/0x70
btrfs_resize_thread_pool+0xe3/0xf0
? sync_filesystem+0x0/0xc0
? btrfs_resize_thread_pool+0x0/0xf0
btrfs_remount+0x1d2/0x570
? kern_path+0x0/0x80
do_remount_sb+0xd9/0x1c0
do_mount+0x26a/0xbf0
? kfree+0x0/0x1b0
SyS_mount+0xc4/0x110
It's a call
btrfs_workqueue_set_max(fs_info->scrub_wr_completion_workers, new_pool_size);
with
fs_info->scrub_wr_completion_workers = NULL;
as scrub wqs get created only on user's demand.
Patch skips not-created-yet workqueues.
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
CC: Qu Wenruo <quwenruo@cn.fujitsu.com>
CC: Chris Mason <clm@fb.com>
CC: Josef Bacik <jbacik@fb.com>
CC: linux-btrfs@vger.kernel.org
Signed-off-by: Chris Mason <clm@fb.com>
Since most of the btrfs_workqueue is printed as pointer address,
for easier analysis, add trace for btrfs_workqueue alloc/destroy.
So it is possible to determine the workqueue that a given work belongs
to(by comparing the wq pointer address with alloc trace event).
Signed-off-by: Qu Wenruo <quenruo@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>
Add ftrace for btrfs_workqueue for further workqueue tunning.
This patch needs to applied after the workqueue replace patchset.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Josef Bacik <jbacik@fb.com>
The new btrfs_workqueue still use open-coded function defition,
this patch will change them into btrfs_func_t type which is much the
same as kernel workqueue.
Signed-off-by: Qu Wenruo <quwenruo@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>
Since all the btrfs_worker is replaced with the newly created
btrfs_workqueue, the old codes can be easily remove.
Signed-off-by: Quwenruo <quwenruo@cn.fujitsu.com>
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
The original btrfs_workers has thresholding functions to dynamically
create or destroy kthreads.
Though there is no such function in kernel workqueue because the worker
is not created manually, we can still use the workqueue_set_max_active
to simulated the behavior, mainly to achieve a better HDD performance by
setting a high threshold on submit_workers.
(Sadly, no resource can be saved)
So in this patch, extra workqueue pending counters are introduced to
dynamically change the max active of each btrfs_workqueue_struct, hoping
to restore the behavior of the original thresholding function.
Also, workqueue_set_max_active use a mutex to protect workqueue_struct,
which is not meant to be called too frequently, so a new interval
mechanism is applied, that will only call workqueue_set_max_active after
a count of work is queued. Hoping to balance both the random and
sequence performance on HDD.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Add high priority function to btrfs_workqueue.
This is implemented by embedding a btrfs_workqueue into a
btrfs_workqueue and use some helper functions to differ the normal
priority wq and high priority wq.
So the high priority wq is completely independent from the normal
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>
Use kernel workqueue to implement a new btrfs_workqueue_struct, which
has the ordering execution feature like the btrfs_worker.
The func is executed in a concurrency way, and the
ordred_func/ordered_free is executed in the sequence them are queued
after the corresponding func is done.
The new btrfs_workqueue works much like the original one, one workqueue
for normal work and a list for ordered work.
When a work is queued, ordered work will be added to the list and helper
function will be queued into the workqueue.
The helper function will execute a normal work and then check and execute as many
ordered work as possible in the sequence they were queued.
At this patch, high priority work queue or thresholding is not added yet.
The high priority feature and thresholding will be added in the following patches.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
In case we do not refill, we can overwrite cur pointer from prio_head
by one from not prioritized head, what looks as something that was
not intended.
This change make we always take works from prio_head first until it's
not empty.
Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
Signed-off-by: Josef Bacik <jbacik@fb.com>
__btrfs_start_workers returns 0 in case it raced with
btrfs_stop_workers and lost the race. This is wrong because worker in
this case is not allowed to start and is in fact destroyed. Return
-EINVAL instead.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Fix spacing issues detected via checkpatch.pl in accordance with the
kernel style guidelines.
Signed-off-by: Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
The current implementation of worker threads in Btrfs has races in
worker stopping code, which cause all kinds of panics and lockups when
running btrfs/011 xfstest in a loop. The problem is that
btrfs_stop_workers is unsynchronized with respect to check_idle_worker,
check_busy_worker and __btrfs_start_workers.
E.g., check_idle_worker race flow:
btrfs_stop_workers(): check_idle_worker(aworker):
- grabs the lock
- splices the idle list into the
working list
- removes the first worker from the
working list
- releases the lock to wait for
its kthread's completion
- grabs the lock
- if aworker is on the working list,
moves aworker from the working list
to the idle list
- releases the lock
- grabs the lock
- puts the worker
- removes the second worker from the
working list
......
btrfs_stop_workers returns, aworker is on the idle list
FS is umounted, memory is freed
......
aworker is waken up, fireworks ensue
With this applied, I wasn't able to trigger the problem in 48 hours,
whereas previously I could reliably reproduce at least one of these
races within an hour.
Reported-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Each ordered operation has a free callback, and this was called with the
worker spinlock held. Josef made the free callback also call iput,
which we can't do with the spinlock.
This drops the spinlock for the free operation and grabs it again before
moving through the rest of the list. We'll circle back around to this
and find a cleaner way that doesn't bounce the lock around so much.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
cc: stable@kernel.org
* pm-sleep: (51 commits)
PM: Drop generic_subsys_pm_ops
PM / Sleep: Remove forward-only callbacks from AMBA bus type
PM / Sleep: Remove forward-only callbacks from platform bus type
PM: Run the driver callback directly if the subsystem one is not there
PM / Sleep: Make pm_op() and pm_noirq_op() return callback pointers
PM / Sleep: Merge internal functions in generic_ops.c
PM / Sleep: Simplify generic system suspend callbacks
PM / Hibernate: Remove deprecated hibernation snapshot ioctls
PM / Sleep: Fix freezer failures due to racy usermodehelper_is_disabled()
PM / Sleep: Recommend [un]lock_system_sleep() over using pm_mutex directly
PM / Sleep: Replace mutex_[un]lock(&pm_mutex) with [un]lock_system_sleep()
PM / Sleep: Make [un]lock_system_sleep() generic
PM / Sleep: Use the freezer_count() functions in [un]lock_system_sleep() APIs
PM / Freezer: Remove the "userspace only" constraint from freezer[_do_not]_count()
PM / Hibernate: Replace unintuitive 'if' condition in kernel/power/user.c with 'else'
Freezer / sunrpc / NFS: don't allow TASK_KILLABLE sleeps to block the freezer
PM / Sleep: Unify diagnostic messages from device suspend/resume
ACPI / PM: Do not save/restore NVS on Asus K54C/K54HR
PM / Hibernate: Remove deprecated hibernation test modes
PM / Hibernate: Thaw processes in SNAPSHOT_CREATE_IMAGE ioctl test path
...
Conflicts:
kernel/kmod.c
Dan Carpenter noticed that we were doing a double unlock on the worker
lock, and sometimes picking a worker thread without the lock held.
This fixes both errors.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
* master: (848 commits)
SELinux: Fix RCU deref check warning in sel_netport_insert()
binary_sysctl(): fix memory leak
mm/vmalloc.c: remove static declaration of va from __get_vm_area_node
ipmi_watchdog: restore settings when BMC reset
oom: fix integer overflow of points in oom_badness
memcg: keep root group unchanged if creation fails
nilfs2: potential integer overflow in nilfs_ioctl_clean_segments()
nilfs2: unbreak compat ioctl
cpusets: stall when updating mems_allowed for mempolicy or disjoint nodemask
evm: prevent racing during tfm allocation
evm: key must be set once during initialization
mmc: vub300: fix type of firmware_rom_wait_states module parameter
Revert "mmc: enable runtime PM by default"
mmc: sdhci: remove "state" argument from sdhci_suspend_host
x86, dumpstack: Fix code bytes breakage due to missing KERN_CONT
IB/qib: Correct sense on freectxts increment and decrement
RDMA/cma: Verify private data length
cgroups: fix a css_set not found bug in cgroup_attach_proc
oprofile: Fix uninitialized memory access when writing to writing to oprofilefs
Revert "xen/pv-on-hvm kexec: add xs_reset_watches to shutdown watches from old kernel"
...
Conflicts:
kernel/cgroup_freezer.c
Al pointed out we have some random problems with the way we account for
num_workers_starting in the async thread stuff. First of all we need to make
sure to decrement num_workers_starting if we fail to start the worker, so make
__btrfs_start_workers do this. Also fix __btrfs_start_workers so that it
doesn't call btrfs_stop_workers(), there is no point in stopping everybody if we
failed to create a worker. Also check_pending_worker_creates needs to call
__btrfs_start_work in it's work function since it already increments
num_workers_starting.
People only start one worker at a time, so get rid of the num_workers argument
everywhere, and make btrfs_queue_worker a void since it will always succeed.
Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
If we have a constant stream of end_io completions or crc work,
we can hit softlockup messages from the async helper threads. This
adds a cond_resched() into the loop to avoid them.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There is no reason to export two functions for entering the
refrigerator. Calling refrigerator() instead of try_to_freeze()
doesn't save anything noticeable or removes any race condition.
* Rename refrigerator() to __refrigerator() and make it return bool
indicating whether it scheduled out for freezing.
* Update try_to_freeze() to return bool and relay the return value of
__refrigerator() if freezing().
* Convert all refrigerator() users to try_to_freeze().
* Update documentation accordingly.
* While at it, add might_sleep() to try_to_freeze().
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Samuel Ortiz <samuel@sortiz.org>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jan Kara <jack@suse.cz>
Cc: KONISHI Ryusuke <konishi.ryusuke@lab.ntt.co.jp>
Cc: Christoph Hellwig <hch@infradead.org>
Yan Zheng noticed two places we were doing a lot of work
without task->state set to TASK_RUNNING. This sets the state
properly after we get ready to sleep but decide not to.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
The btrfs async worker threads are used for a wide variety of things,
including processing bio end_io functions. This means that when
the endio threads aren't running, the rest of the FS isn't
able to do the final processing required to clear PageWriteback.
The endio threads also try to exit as they become idle and
start more as the work piles up. The problem is that starting more
threads means kthreadd may need to allocate ram, and that allocation
may wait until the global number of writeback pages on the system is
below a certain limit.
The result of that throttling is that end IO threads wait on
kthreadd, who is waiting on IO to end, which will never happen.
This commit fixes the deadlock by handing off thread startup to a
dedicated thread. It also fixes a bug where the on-demand thread
creation was creating far too many threads because it didn't take into
account threads being started by other procs.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
It was possible for an async worker thread to be selected to
receive a new work item, but exit before the work item was
actually placed into that thread's work list.
This commit fixes the race by incrementing the num_pending
counter earlier, and making sure to check the number of pending
work items before a thread exits.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
The exit-on-idle code for async worker threads was incorrectly
calling spin_lock_irq with interrupts already off.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Nikolay Borisov
Filipe Manana
David Sterba
Omar Sandoval
Liu Bo
Jeff Mahoney
Maxim Patlasov
Qu Wenruo
Zhao Lei
David Sterba
Miao Xie
Sergei Trofimovich
Filipe Manana
Stanislaw Gruszka
Ilya Dryomov
Dulshani Gunawardhana
Chris Mason
Rafael J. Wysocki
Chris Mason
Josef Bacik
Tejun Heo