* [PATCH] crypto:padata: Fix return err for PADATA_RESET
@ 2023-08-09 12:50 Lu Jialin
2023-08-18 6:02 ` Herbert Xu
0 siblings, 1 reply; 3+ messages in thread
From: Lu Jialin @ 2023-08-09 12:50 UTC (permalink / raw)
To: Steffen Klassert, Daniel Jordan; +Cc: Lu Jialin, linux-crypto
We found a hungtask bug in test_aead_vec_cfg as follows:
INFO: task cryptomgr_test:391009 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Call trace:
__switch_to+0x98/0xe0
__schedule+0x6c4/0xf40
schedule+0xd8/0x1b4
schedule_timeout+0x474/0x560
wait_for_common+0x368/0x4e0
wait_for_completion+0x20/0x30
test_aead_vec_cfg+0xab4/0xd50
test_aead+0x144/0x1f0
alg_test_aead+0xd8/0x1e0
alg_test+0x634/0x890
cryptomgr_test+0x40/0x70
kthread+0x1e0/0x220
ret_from_fork+0x10/0x18
Kernel panic - not syncing: hung_task: blocked tasks
For padata_do_parallel, when the return err is 0 or -EBUSY, it will call
wait_for_completion(&wait->completion) in test_aead_vec_cfg. In normal
case, aead_request_complete() will be called in pcrypt_aead_serial and the
return err is 0 for padata_do_parallel. But, when pinst->flags is
PADATA_RESET, the return err is -EBUSY for padata_do_parallel, and it
won't call aead_request_complete(). Therefore, test_aead_vec_cfg will
hung at wait_for_completion(&wait->completion), which will cause
hungtask.
The problem comes as following:
(padata_do_parallel) |
rcu_read_lock_bh(); |
err = -EINVAL; | (padata_replace)
| pinst->flags |= PADATA_RESET;
err = -EBUSY |
if (pinst->flags & PADATA_RESET) |
rcu_read_unlock_bh() |
return err |
In order to resolve the problem, change the return err to -EINVAL when
pinst->flags is set PADATA_RESET.
Signed-off-by: Lu Jialin <lujialin4@huawei.com>
---
kernel/padata.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/kernel/padata.c b/kernel/padata.c
index 222d60195de6..cc326ffb209a 100644
--- a/kernel/padata.c
+++ b/kernel/padata.c
@@ -202,7 +202,6 @@ int padata_do_parallel(struct padata_shell *ps,
*cb_cpu = cpu;
}
- err = -EBUSY;
if ((pinst->flags & PADATA_RESET))
goto out;
--
2.34.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] crypto:padata: Fix return err for PADATA_RESET
2023-08-09 12:50 [PATCH] crypto:padata: Fix return err for PADATA_RESET Lu Jialin
@ 2023-08-18 6:02 ` Herbert Xu
2023-08-21 8:41 ` Lu Jialin
0 siblings, 1 reply; 3+ messages in thread
From: Herbert Xu @ 2023-08-18 6:02 UTC (permalink / raw)
To: Lu Jialin; +Cc: steffen.klassert, daniel.m.jordan, lujialin4, linux-crypto
Lu Jialin <lujialin4@huawei.com> wrote:
> We found a hungtask bug in test_aead_vec_cfg as follows:
>
> INFO: task cryptomgr_test:391009 blocked for more than 120 seconds.
> "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> Call trace:
> __switch_to+0x98/0xe0
> __schedule+0x6c4/0xf40
> schedule+0xd8/0x1b4
> schedule_timeout+0x474/0x560
> wait_for_common+0x368/0x4e0
> wait_for_completion+0x20/0x30
> test_aead_vec_cfg+0xab4/0xd50
> test_aead+0x144/0x1f0
> alg_test_aead+0xd8/0x1e0
> alg_test+0x634/0x890
> cryptomgr_test+0x40/0x70
> kthread+0x1e0/0x220
> ret_from_fork+0x10/0x18
> Kernel panic - not syncing: hung_task: blocked tasks
>
> For padata_do_parallel, when the return err is 0 or -EBUSY, it will call
> wait_for_completion(&wait->completion) in test_aead_vec_cfg. In normal
> case, aead_request_complete() will be called in pcrypt_aead_serial and the
> return err is 0 for padata_do_parallel. But, when pinst->flags is
> PADATA_RESET, the return err is -EBUSY for padata_do_parallel, and it
> won't call aead_request_complete(). Therefore, test_aead_vec_cfg will
> hung at wait_for_completion(&wait->completion), which will cause
> hungtask.
>
> The problem comes as following:
> (padata_do_parallel) |
> rcu_read_lock_bh(); |
> err = -EINVAL; | (padata_replace)
> | pinst->flags |= PADATA_RESET;
> err = -EBUSY |
> if (pinst->flags & PADATA_RESET) |
> rcu_read_unlock_bh() |
> return err |
>
> In order to resolve the problem, change the return err to -EINVAL when
> pinst->flags is set PADATA_RESET.
>
> Signed-off-by: Lu Jialin <lujialin4@huawei.com>
> ---
> kernel/padata.c | 1 -
> 1 file changed, 1 deletion(-)
Thanks for the patch.
So the issue here is that the Crypto API uses EBUSY for a specific
purpose but padata uses it too and they're getting confused with
each other.
I think what we should do is get pcrypt to check the error value from
padata_do_parallel, and if it's EBUSY then change it to something
else.
Thanks,
--
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] crypto:padata: Fix return err for PADATA_RESET
2023-08-18 6:02 ` Herbert Xu
@ 2023-08-21 8:41 ` Lu Jialin
0 siblings, 0 replies; 3+ messages in thread
From: Lu Jialin @ 2023-08-21 8:41 UTC (permalink / raw)
To: Herbert Xu; +Cc: steffen.klassert, daniel.m.jordan, linux-crypto
Thanks for your suggestions. I will update my patch in v2
On 2023/8/18 14:02, Herbert Xu wrote:
> Lu Jialin <lujialin4@huawei.com> wrote:
>> We found a hungtask bug in test_aead_vec_cfg as follows:
>>
>> INFO: task cryptomgr_test:391009 blocked for more than 120 seconds.
>> "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
>> Call trace:
>> __switch_to+0x98/0xe0
>> __schedule+0x6c4/0xf40
>> schedule+0xd8/0x1b4
>> schedule_timeout+0x474/0x560
>> wait_for_common+0x368/0x4e0
>> wait_for_completion+0x20/0x30
>> test_aead_vec_cfg+0xab4/0xd50
>> test_aead+0x144/0x1f0
>> alg_test_aead+0xd8/0x1e0
>> alg_test+0x634/0x890
>> cryptomgr_test+0x40/0x70
>> kthread+0x1e0/0x220
>> ret_from_fork+0x10/0x18
>> Kernel panic - not syncing: hung_task: blocked tasks
>>
>> For padata_do_parallel, when the return err is 0 or -EBUSY, it will call
>> wait_for_completion(&wait->completion) in test_aead_vec_cfg. In normal
>> case, aead_request_complete() will be called in pcrypt_aead_serial and the
>> return err is 0 for padata_do_parallel. But, when pinst->flags is
>> PADATA_RESET, the return err is -EBUSY for padata_do_parallel, and it
>> won't call aead_request_complete(). Therefore, test_aead_vec_cfg will
>> hung at wait_for_completion(&wait->completion), which will cause
>> hungtask.
>>
>> The problem comes as following:
>> (padata_do_parallel) |
>> rcu_read_lock_bh(); |
>> err = -EINVAL; | (padata_replace)
>> | pinst->flags |= PADATA_RESET;
>> err = -EBUSY |
>> if (pinst->flags & PADATA_RESET) |
>> rcu_read_unlock_bh() |
>> return err |
>>
>> In order to resolve the problem, change the return err to -EINVAL when
>> pinst->flags is set PADATA_RESET.
>>
>> Signed-off-by: Lu Jialin <lujialin4@huawei.com>
>> ---
>> kernel/padata.c | 1 -
>> 1 file changed, 1 deletion(-)
>
> Thanks for the patch.
>
> So the issue here is that the Crypto API uses EBUSY for a specific
> purpose but padata uses it too and they're getting confused with
> each other.
>
> I think what we should do is get pcrypt to check the error value from
> padata_do_parallel, and if it's EBUSY then change it to something
> else.
>
> Thanks,
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-08-21 8:41 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-09 12:50 [PATCH] crypto:padata: Fix return err for PADATA_RESET Lu Jialin
2023-08-18 6:02 ` Herbert Xu
2023-08-21 8:41 ` Lu Jialin
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.