* [dm-devel] [PATCH] dm-crypt: fix softlockup in dmcrypt_write
@ 2023-02-23 3:19 yangerkun
2023-02-26 2:01 ` Bart Van Assche
0 siblings, 1 reply; 12+ messages in thread
From: yangerkun @ 2023-02-23 3:19 UTC (permalink / raw)
To: agk, snitzer, dm-devel; +Cc: yangerkun, yangerkun
From: yangerkun <yangerkun@huawei.com>
We meet a softlockup:
localhost login: [ 3391.153255][ C12] watchdog: BUG: soft lockup - CPU#12 stuck for 23s! [dmcrypt_write/2:2897]
...
[ 3391.258372][ C12] CPU: 12 PID: 2897 Comm: dmcrypt_write/2 Tainted: G L 5.10.0+ #8
[ 3391.267288][ C12] Hardware name: Huawei TaiShan 2280 V2/BC82AMDDA, BIOS 1.75 04/26/2021
[ 3391.275428][ C12] pstate: 80400009 (Nzcv daif +PAN -UAO -TCO BTYPE=--)
[ 3391.282102][ C12] pc : blk_attempt_bio_merge.part.6+0x38/0x158
[ 3391.288080][ C12] lr : blk_attempt_plug_merge+0xc0/0x1b0
[ 3391.293540][ C12] sp : ffff8000718abc30
[ 3391.297530][ C12] x29: ffff8000718abc30 x28: 0000000000000000
[ 3391.303509][ C12] x27: ffff2020361d9ac0 x26: 0000000000000001
[ 3391.309488][ C12] x25: 0000000000000001 x24: ffff8000718abd08
[ 3391.315467][ C12] x23: ffff0020a10dbb00 x22: 0000000000000001
[ 3391.321445][ C12] x21: ffff8000718abe20 x20: ffff0020a10dbb00
[ 3391.327424][ C12] x19: ffff0020aed7b040 x18: 0000000000000000
[ 3391.333402][ C12] x17: 0000000000000000 x16: fffffdffffe00000
[ 3391.339381][ C12] x15: 0000000000001000 x14: 0000000000000000
[ 3391.345359][ C12] x13: 0000000000002000 x12: ffff2060011f9070
[ 3391.351338][ C12] x11: 0000000000000001 x10: 0000000000000002
[ 3391.357316][ C12] x9 : ffff800010586c38 x8 : ffff202009a7f200
[ 3391.363294][ C12] x7 : ffff8000718abd00 x6 : ffff20200df33750
[ 3391.369274][ C12] x5 : 0000000000000001 x4 : 0000000000000000
[ 3391.375252][ C12] x3 : 0000000000000001 x2 : ffff0020a10dbb00
[ 3391.381230][ C12] x1 : ffff0020aed7b040 x0 : 0000000000001000
[ 3391.387210][ C12] Call trace:
[ 3391.390338][ C12] blk_attempt_bio_merge.part.6+0x38/0x158
[ 3391.395970][ C12] blk_attempt_plug_merge+0xc0/0x1b0
[ 3391.401085][ C12] blk_mq_submit_bio+0x398/0x550
[ 3391.405856][ C12] submit_bio_noacct+0x308/0x380
[ 3391.410630][ C12] dmcrypt_write+0x1e4/0x208 [dm_crypt]
[ 3391.416005][ C12] kthread+0x130/0x138
[ 3391.419911][ C12] ret_from_fork+0x10/0x18
dmcrypt_write may see not empty write_tree for a long time if we have a
heavy load of write, and submit_bio for high-performance disk like nvme
can always get a valid tag, so the bio routine will not yield cpu too.
So the softlockup seems reasonably.
Fix it by schedule periodically.
Fixes: dc2676210c42 ("dm crypt: offload writes to thread")
Signed-off-by: yangerkun <yangerkun@huawei.com>
---
drivers/md/dm-crypt.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 2653516bcdef..755a01d72cdb 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1891,6 +1891,7 @@ static int dmcrypt_write(void *data)
{
struct crypt_config *cc = data;
struct dm_crypt_io *io;
+ unsigned long start_time = jiffies;
while (1) {
struct rb_root write_tree;
@@ -1913,6 +1914,7 @@ static int dmcrypt_write(void *data)
schedule();
+ start_time = jiffies;
set_current_state(TASK_RUNNING);
spin_lock_irq(&cc->write_thread_lock);
goto continue_locked;
@@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
BUG_ON(rb_parent(write_tree.rb_node));
+ if (time_is_before_jiffies(start_time + HZ)) {
+ schedule();
+ start_time = jiffies;
+ }
/*
* Note: we cannot walk the tree here with rb_next because
* the structures may be freed when kcryptd_io_write is called.
--
2.31.1
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [dm-devel] [PATCH] dm-crypt: fix softlockup in dmcrypt_write
2023-02-23 3:19 [dm-devel] [PATCH] dm-crypt: fix softlockup in dmcrypt_write yangerkun
@ 2023-02-26 2:01 ` Bart Van Assche
2023-02-27 1:31 ` yangerkun
0 siblings, 1 reply; 12+ messages in thread
From: Bart Van Assche @ 2023-02-26 2:01 UTC (permalink / raw)
To: yangerkun, agk, snitzer, dm-devel; +Cc: yangerkun
On 2/22/23 19:19, yangerkun wrote:
> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
>
> BUG_ON(rb_parent(write_tree.rb_node));
>
> + if (time_is_before_jiffies(start_time + HZ)) {
> + schedule();
> + start_time = jiffies;
> + }
Why schedule() instead of cond_resched()?
Thanks,
Bart.
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] [PATCH] dm-crypt: fix softlockup in dmcrypt_write
2023-02-26 2:01 ` Bart Van Assche
@ 2023-02-27 1:31 ` yangerkun
2023-02-27 17:55 ` [dm-devel] " Mike Snitzer
0 siblings, 1 reply; 12+ messages in thread
From: yangerkun @ 2023-02-27 1:31 UTC (permalink / raw)
To: Bart Van Assche, agk, snitzer, dm-devel; +Cc: yangerkun
在 2023/2/26 10:01, Bart Van Assche 写道:
> On 2/22/23 19:19, yangerkun wrote:
>> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
>> BUG_ON(rb_parent(write_tree.rb_node));
>> + if (time_is_before_jiffies(start_time + HZ)) {
>> + schedule();
>> + start_time = jiffies;
>> + }
>
> Why schedule() instead of cond_resched()?
cond_resched may not really schedule, which may trigger the problem too,
but it seems after 1 second, it may never happend?
Thanks,
Kun.
>
> Thanks,
>
> Bart.
>
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-27 1:31 ` yangerkun
@ 2023-02-27 17:55 ` Mike Snitzer
2023-02-27 18:03 ` Mike Snitzer
2023-02-28 1:25 ` yangerkun
0 siblings, 2 replies; 12+ messages in thread
From: Mike Snitzer @ 2023-02-27 17:55 UTC (permalink / raw)
To: yangerkun; +Cc: yangerkun, dm-devel, Bart Van Assche, agk
On Sun, Feb 26 2023 at 8:31P -0500,
yangerkun <yangerkun@huaweicloud.com> wrote:
>
>
> 在 2023/2/26 10:01, Bart Van Assche 写道:
> > On 2/22/23 19:19, yangerkun wrote:
> > > @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
> > > BUG_ON(rb_parent(write_tree.rb_node));
> > > + if (time_is_before_jiffies(start_time + HZ)) {
> > > + schedule();
> > > + start_time = jiffies;
> > > + }
> >
> > Why schedule() instead of cond_resched()?
>
> cond_resched may not really schedule, which may trigger the problem too, but
> it seems after 1 second, it may never happend?
I had the same question as Bart when reviewing your homegrown
conditional schedule(). Hopefully you can reproduce this issue? If
so, please see if simply using cond_resched() fixes the issue.
Thanks,
Mike
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-27 17:55 ` [dm-devel] " Mike Snitzer
@ 2023-02-27 18:03 ` Mike Snitzer
2023-02-27 18:06 ` Mike Snitzer
2023-02-28 1:25 ` yangerkun
1 sibling, 1 reply; 12+ messages in thread
From: Mike Snitzer @ 2023-02-27 18:03 UTC (permalink / raw)
To: yangerkun; +Cc: yangerkun, dm-devel, Bart Van Assche, agk
On Mon, Feb 27 2023 at 12:55P -0500,
Mike Snitzer <snitzer@kernel.org> wrote:
> On Sun, Feb 26 2023 at 8:31P -0500,
> yangerkun <yangerkun@huaweicloud.com> wrote:
>
> >
> >
> > 在 2023/2/26 10:01, Bart Van Assche 写道:
> > > On 2/22/23 19:19, yangerkun wrote:
> > > > @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
> > > > BUG_ON(rb_parent(write_tree.rb_node));
> > > > + if (time_is_before_jiffies(start_time + HZ)) {
> > > > + schedule();
> > > > + start_time = jiffies;
> > > > + }
> > >
> > > Why schedule() instead of cond_resched()?
> >
> > cond_resched may not really schedule, which may trigger the problem too, but
> > it seems after 1 second, it may never happend?
>
> I had the same question as Bart when reviewing your homegrown
> conditional schedule(). Hopefully you can reproduce this issue? If
> so, please see if simply using cond_resched() fixes the issue.
This seems like a more appropriate patch:
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 87c5706131f2..faba1be572f9 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1937,6 +1937,7 @@ static int dmcrypt_write(void *data)
io = crypt_io_from_node(rb_first(&write_tree));
rb_erase(&io->rb_node, &write_tree);
kcryptd_io_write(io);
+ cond_resched();
} while (!RB_EMPTY_ROOT(&write_tree));
blk_finish_plug(&plug);
}
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-27 18:03 ` Mike Snitzer
@ 2023-02-27 18:06 ` Mike Snitzer
2023-02-28 1:40 ` yangerkun
0 siblings, 1 reply; 12+ messages in thread
From: Mike Snitzer @ 2023-02-27 18:06 UTC (permalink / raw)
To: yangerkun; +Cc: yangerkun, dm-devel, Bart Van Assche, agk
On Mon, Feb 27 2023 at 1:03P -0500,
Mike Snitzer <snitzer@kernel.org> wrote:
> On Mon, Feb 27 2023 at 12:55P -0500,
> Mike Snitzer <snitzer@kernel.org> wrote:
>
> > On Sun, Feb 26 2023 at 8:31P -0500,
> > yangerkun <yangerkun@huaweicloud.com> wrote:
> >
> > >
> > >
> > > 在 2023/2/26 10:01, Bart Van Assche 写道:
> > > > On 2/22/23 19:19, yangerkun wrote:
> > > > > @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
> > > > > BUG_ON(rb_parent(write_tree.rb_node));
> > > > > + if (time_is_before_jiffies(start_time + HZ)) {
> > > > > + schedule();
> > > > > + start_time = jiffies;
> > > > > + }
> > > >
> > > > Why schedule() instead of cond_resched()?
> > >
> > > cond_resched may not really schedule, which may trigger the problem too, but
> > > it seems after 1 second, it may never happend?
> >
> > I had the same question as Bart when reviewing your homegrown
> > conditional schedule(). Hopefully you can reproduce this issue? If
> > so, please see if simply using cond_resched() fixes the issue.
>
> This seems like a more appropriate patch:
>
> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> index 87c5706131f2..faba1be572f9 100644
> --- a/drivers/md/dm-crypt.c
> +++ b/drivers/md/dm-crypt.c
> @@ -1937,6 +1937,7 @@ static int dmcrypt_write(void *data)
> io = crypt_io_from_node(rb_first(&write_tree));
> rb_erase(&io->rb_node, &write_tree);
> kcryptd_io_write(io);
> + cond_resched();
> } while (!RB_EMPTY_ROOT(&write_tree));
> blk_finish_plug(&plug);
> }
or:
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 87c5706131f2..3ba2fd3e4358 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1934,6 +1934,7 @@ static int dmcrypt_write(void *data)
*/
blk_start_plug(&plug);
do {
+ cond_resched();
io = crypt_io_from_node(rb_first(&write_tree));
rb_erase(&io->rb_node, &write_tree);
kcryptd_io_write(io);
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-27 17:55 ` [dm-devel] " Mike Snitzer
2023-02-27 18:03 ` Mike Snitzer
@ 2023-02-28 1:25 ` yangerkun
2023-02-28 7:22 ` yangerkun
1 sibling, 1 reply; 12+ messages in thread
From: yangerkun @ 2023-02-28 1:25 UTC (permalink / raw)
To: Mike Snitzer; +Cc: yangerkun, dm-devel, Bart Van Assche, agk
在 2023/2/28 1:55, Mike Snitzer 写道:
> On Sun, Feb 26 2023 at 8:31P -0500,
> yangerkun <yangerkun@huaweicloud.com> wrote:
>
>>
>>
>> 在 2023/2/26 10:01, Bart Van Assche 写道:
>>> On 2/22/23 19:19, yangerkun wrote:
>>>> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
>>>> BUG_ON(rb_parent(write_tree.rb_node));
>>>> + if (time_is_before_jiffies(start_time + HZ)) {
>>>> + schedule();
>>>> + start_time = jiffies;
>>>> + }
>>>
>>> Why schedule() instead of cond_resched()?
>>
>> cond_resched may not really schedule, which may trigger the problem too, but
>> it seems after 1 second, it may never happend?
>
> I had the same question as Bart when reviewing your homegrown
> conditional schedule(). Hopefully you can reproduce this issue? If
> so, please see if simply using cond_resched() fixes the issue.
Yes, our testcase can trigger the issue, will do it with cond_resched.
>
> Thanks,
> Mike
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-27 18:06 ` Mike Snitzer
@ 2023-02-28 1:40 ` yangerkun
2023-03-06 18:02 ` Mikulas Patocka
0 siblings, 1 reply; 12+ messages in thread
From: yangerkun @ 2023-02-28 1:40 UTC (permalink / raw)
To: Mike Snitzer; +Cc: yangerkun, dm-devel, Bart Van Assche, agk
在 2023/2/28 2:06, Mike Snitzer 写道:
> On Mon, Feb 27 2023 at 1:03P -0500,
> Mike Snitzer <snitzer@kernel.org> wrote:
>
>> On Mon, Feb 27 2023 at 12:55P -0500,
>> Mike Snitzer <snitzer@kernel.org> wrote:
>>
>>> On Sun, Feb 26 2023 at 8:31P -0500,
>>> yangerkun <yangerkun@huaweicloud.com> wrote:
>>>
>>>>
>>>>
>>>> 在 2023/2/26 10:01, Bart Van Assche 写道:
>>>>> On 2/22/23 19:19, yangerkun wrote:
>>>>>> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
>>>>>> BUG_ON(rb_parent(write_tree.rb_node));
>>>>>> + if (time_is_before_jiffies(start_time + HZ)) {
>>>>>> + schedule();
>>>>>> + start_time = jiffies;
>>>>>> + }
>>>>>
>>>>> Why schedule() instead of cond_resched()?
>>>>
>>>> cond_resched may not really schedule, which may trigger the problem too, but
>>>> it seems after 1 second, it may never happend?
>>>
>>> I had the same question as Bart when reviewing your homegrown
>>> conditional schedule(). Hopefully you can reproduce this issue? If
>>> so, please see if simply using cond_resched() fixes the issue.
>>
>> This seems like a more appropriate patch:
>>
>> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
>> index 87c5706131f2..faba1be572f9 100644
>> --- a/drivers/md/dm-crypt.c
>> +++ b/drivers/md/dm-crypt.c
>> @@ -1937,6 +1937,7 @@ static int dmcrypt_write(void *data)
>> io = crypt_io_from_node(rb_first(&write_tree));
>> rb_erase(&io->rb_node, &write_tree);
>> kcryptd_io_write(io);
>> + cond_resched();
>> } while (!RB_EMPTY_ROOT(&write_tree));
>> blk_finish_plug(&plug);
>> }
>
>
> or:
>
> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> index 87c5706131f2..3ba2fd3e4358 100644
> --- a/drivers/md/dm-crypt.c
> +++ b/drivers/md/dm-crypt.c
> @@ -1934,6 +1934,7 @@ static int dmcrypt_write(void *data)
> */
> blk_start_plug(&plug);
> do {
> + cond_resched();
> io = crypt_io_from_node(rb_first(&write_tree));
> rb_erase(&io->rb_node, &write_tree);
> kcryptd_io_write(io);
Hi,
Thanks a lot for your review!
It's ok to fix the softlockup, but for async write encrypt,
kcryptd_crypt_write_io_submit will add bio to write_tree, and once we
call cond_resched before every kcryptd_io_write, the write performance
may be poor while we meet a high cpu usage scene.
kcryptd_crypt_write_io_submit will wakeup write_thread once there is a
empty write_tree, and dmcrypt_write will peel the old write_tree to
submit bio, so there can not exist too many bio in write_tree. Then I
choose yield cpu before the 'while' that submit bio...
Thanks,
Kun.
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-28 1:25 ` yangerkun
@ 2023-02-28 7:22 ` yangerkun
0 siblings, 0 replies; 12+ messages in thread
From: yangerkun @ 2023-02-28 7:22 UTC (permalink / raw)
To: yangerkun, Mike Snitzer; +Cc: dm-devel, Bart Van Assche, agk
在 2023/2/28 9:25, yangerkun 写道:
>
>
> 在 2023/2/28 1:55, Mike Snitzer 写道:
>> On Sun, Feb 26 2023 at 8:31P -0500,
>> yangerkun <yangerkun@huaweicloud.com> wrote:
>>
>>>
>>>
>>> 在 2023/2/26 10:01, Bart Van Assche 写道:
>>>> On 2/22/23 19:19, yangerkun wrote:
>>>>> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
>>>>> BUG_ON(rb_parent(write_tree.rb_node));
>>>>> + if (time_is_before_jiffies(start_time + HZ)) {
>>>>> + schedule();
>>>>> + start_time = jiffies;
>>>>> + }
>>>>
>>>> Why schedule() instead of cond_resched()?
>>>
>>> cond_resched may not really schedule, which may trigger the problem
>>> too, but
>>> it seems after 1 second, it may never happend?
>>
>> I had the same question as Bart when reviewing your homegrown
>> conditional schedule(). Hopefully you can reproduce this issue? If
>> so, please see if simply using cond_resched() fixes the issue.
>
> Yes, our testcase can trigger the issue, will do it with cond_resched.
Without this patch, the softlockup may trigger soon, after this patch no
matter cond_resched or schedule, softlockup won't trigger after two hour
test.
Thanks,
Kun.
>
>
>
>>
>> Thanks,
>> Mike
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-02-28 1:40 ` yangerkun
@ 2023-03-06 18:02 ` Mikulas Patocka
2023-03-06 18:16 ` Bart Van Assche
0 siblings, 1 reply; 12+ messages in thread
From: Mikulas Patocka @ 2023-03-06 18:02 UTC (permalink / raw)
To: yangerkun; +Cc: yangerkun, Mike Snitzer, dm-devel, agk, Bart Van Assche
[-- Attachment #1: Type: text/plain, Size: 3451 bytes --]
On Tue, 28 Feb 2023, yangerkun wrote:
>
>
> 在 2023/2/28 2:06, Mike Snitzer 写道:
> > On Mon, Feb 27 2023 at 1:03P -0500,
> > Mike Snitzer <snitzer@kernel.org> wrote:
> >
> >> On Mon, Feb 27 2023 at 12:55P -0500,
> >> Mike Snitzer <snitzer@kernel.org> wrote:
> >>
> >>> On Sun, Feb 26 2023 at 8:31P -0500,
> >>> yangerkun <yangerkun@huaweicloud.com> wrote:
> >>>
> >>>>
> >>>>
> >>>> 在 2023/2/26 10:01, Bart Van Assche 写道:
> >>>>> On 2/22/23 19:19, yangerkun wrote:
> >>>>>> @@ -1924,6 +1926,10 @@ static int dmcrypt_write(void *data)
> >>>>>> BUG_ON(rb_parent(write_tree.rb_node));
> >>>>>> + if (time_is_before_jiffies(start_time + HZ)) {
> >>>>>> + schedule();
> >>>>>> + start_time = jiffies;
> >>>>>> + }
> >>>>>
> >>>>> Why schedule() instead of cond_resched()?
> >>>>
> >>>> cond_resched may not really schedule, which may trigger the problem too,
> but
> >>>> it seems after 1 second, it may never happend?
> >>>
> >>> I had the same question as Bart when reviewing your homegrown
> >>> conditional schedule(). Hopefully you can reproduce this issue? If
> >>> so, please see if simply using cond_resched() fixes the issue.
> >>
> >> This seems like a more appropriate patch:
> >>
> >> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> >> index 87c5706131f2..faba1be572f9 100644
> >> --- a/drivers/md/dm-crypt.c
> >> +++ b/drivers/md/dm-crypt.c
> >> @@ -1937,6 +1937,7 @@ static int dmcrypt_write(void *data)
> >> io = crypt_io_from_node(rb_first(&write_tree));
> >> rb_erase(&io->rb_node, &write_tree);
> >> kcryptd_io_write(io);
> >> + cond_resched();
> >> } while (!RB_EMPTY_ROOT(&write_tree));
> >> blk_finish_plug(&plug);
> >> }
> >
> >
> > or:
> >
> > diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> > index 87c5706131f2..3ba2fd3e4358 100644
> > --- a/drivers/md/dm-crypt.c
> > +++ b/drivers/md/dm-crypt.c
> > @@ -1934,6 +1934,7 @@ static int dmcrypt_write(void *data)
> > */
> > blk_start_plug(&plug);
> > do {
> > + cond_resched();
> > io = crypt_io_from_node(rb_first(&write_tree));
> > rb_erase(&io->rb_node, &write_tree);
> > kcryptd_io_write(io);
>
> Hi,
>
> Thanks a lot for your review!
>
> It's ok to fix the softlockup, but for async write encrypt,
> kcryptd_crypt_write_io_submit will add bio to write_tree, and once we
> call cond_resched before every kcryptd_io_write, the write performance
> may be poor while we meet a high cpu usage scene.
Hi
To fix this problem, find the PID of the process "dmcrypt_write" and
change its priority to -20, for example "renice -n -20 -p 34748".
This is the proper way how to fix it; locking up the process for one
second is not.
We used to have high-priority workqueues by default, but it caused audio
playback skipping, so we had to revert it - see
f612b2132db529feac4f965f28a1b9258ea7c22b.
Perhaps we should add an option to have high-priority kernel threads?
Mikulas
> kcryptd_crypt_write_io_submit will wakeup write_thread once there is a
> empty write_tree, and dmcrypt_write will peel the old write_tree to
> submit bio, so there can not exist too many bio in write_tree. Then I
> choose yield cpu before the 'while' that submit bio...
>
> Thanks,
> Kun.
>
> --
> dm-devel mailing list
> dm-devel@redhat.com
> https://listman.redhat.com/mailman/listinfo/dm-devel
>
[-- Attachment #2: Type: text/plain, Size: 98 bytes --]
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-03-06 18:02 ` Mikulas Patocka
@ 2023-03-06 18:16 ` Bart Van Assche
2023-03-06 19:03 ` Mikulas Patocka
0 siblings, 1 reply; 12+ messages in thread
From: Bart Van Assche @ 2023-03-06 18:16 UTC (permalink / raw)
To: Mikulas Patocka, yangerkun; +Cc: yangerkun, Mike Snitzer, dm-devel, agk
On 3/6/23 10:02, Mikulas Patocka wrote:
> On Tue, 28 Feb 2023, yangerkun wrote:
>> It's ok to fix the softlockup, but for async write encrypt,
>> kcryptd_crypt_write_io_submit will add bio to write_tree, and once we
>> call cond_resched before every kcryptd_io_write, the write performance
>> may be poor while we meet a high cpu usage scene.
>
> Hi
>
> To fix this problem, find the PID of the process "dmcrypt_write" and
> change its priority to -20, for example "renice -n -20 -p 34748".
>
> This is the proper way how to fix it; locking up the process for one
> second is not.
>
> We used to have high-priority workqueues by default, but it caused audio
> playback skipping, so we had to revert it - see
> f612b2132db529feac4f965f28a1b9258ea7c22b.
>
> Perhaps we should add an option to have high-priority kernel threads?
Would calling cond_resched() every n iterations instead of every
iteration help? From mm/swapfile.c:
if (unlikely(--latency_ration < 0)) {
cond_resched();
latency_ration = LATENCY_LIMIT;
}
Thanks,
Bart.
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [dm-devel] dm-crypt: fix softlockup in dmcrypt_write
2023-03-06 18:16 ` Bart Van Assche
@ 2023-03-06 19:03 ` Mikulas Patocka
0 siblings, 0 replies; 12+ messages in thread
From: Mikulas Patocka @ 2023-03-06 19:03 UTC (permalink / raw)
To: Bart Van Assche; +Cc: yangerkun, agk, Mike Snitzer, dm-devel, yangerkun
On Mon, 6 Mar 2023, Bart Van Assche wrote:
> On 3/6/23 10:02, Mikulas Patocka wrote:
> > On Tue, 28 Feb 2023, yangerkun wrote:
> > > It's ok to fix the softlockup, but for async write encrypt,
> > > kcryptd_crypt_write_io_submit will add bio to write_tree, and once we
> > > call cond_resched before every kcryptd_io_write, the write performance
> > > may be poor while we meet a high cpu usage scene.
> >
> > Hi
> >
> > To fix this problem, find the PID of the process "dmcrypt_write" and
> > change its priority to -20, for example "renice -n -20 -p 34748".
> >
> > This is the proper way how to fix it; locking up the process for one
> > second is not.
> >
> > We used to have high-priority workqueues by default, but it caused audio
> > playback skipping, so we had to revert it - see
> > f612b2132db529feac4f965f28a1b9258ea7c22b.
> >
> > Perhaps we should add an option to have high-priority kernel threads?
>
> Would calling cond_resched() every n iterations instead of every iteration
> help? From mm/swapfile.c:
>
> if (unlikely(--latency_ration < 0)) {
> cond_resched();
> latency_ration = LATENCY_LIMIT;
> }
>
> Thanks,
>
> Bart.
I think that if this helps, it is really a bug in the scheduler... It
shouldn't switch tasks so often.
Mikulas
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2023-03-06 19:04 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-02-23 3:19 [dm-devel] [PATCH] dm-crypt: fix softlockup in dmcrypt_write yangerkun
2023-02-26 2:01 ` Bart Van Assche
2023-02-27 1:31 ` yangerkun
2023-02-27 17:55 ` [dm-devel] " Mike Snitzer
2023-02-27 18:03 ` Mike Snitzer
2023-02-27 18:06 ` Mike Snitzer
2023-02-28 1:40 ` yangerkun
2023-03-06 18:02 ` Mikulas Patocka
2023-03-06 18:16 ` Bart Van Assche
2023-03-06 19:03 ` Mikulas Patocka
2023-02-28 1:25 ` yangerkun
2023-02-28 7:22 ` yangerkun
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).