public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
* Backporting 3a7956e25e1d to LTS kernels?
@ 2021-06-24  2:03 Jakub Jankowski
  2021-06-25 10:12 ` Greg KH
  0 siblings, 1 reply; 2+ messages in thread
From: Jakub Jankowski @ 2021-06-24  2:03 UTC (permalink / raw)
  To: sashal, peterz; +Cc: stable

Hi,

We've recently hit this on 5.4.124:
   WARNING: CPU: 5 PID: 35 at kernel/kthread.c:75 
kthread_is_per_cpu+0x21/0x30

After some digging, I figured out that this should be fixed by commit 
3a7956e25e1d ("kthread: Fix PF_KTHREAD vs to_kthread() race") in mainline.
It Fixes: ac687e6e8c26 ("kthread: Extract KTHREAD_IS_PER_CPU")

3a7956e25e1d has been backported to 5.12.x and 5.11.x:
- 5.12.4:  as 163dd7fa459faebe055bcbebf9e088bae04f8e33
- 5.11.21: as 42e4caa93282ade6a143ff320d71062874dd3a5a

Makes sense, as ac687e6e8c26 originally landed in 5.11.

But it appears that since then ac687e6e8c26 has also been backported to 
LTS trees:
- 5.10.14:  as b20475a80b4bd2c7bc720c3a9a8337c36b20dd8c
- 5.4.96:   as 5b1e4fc2984eab30c5fffbf2bd1f016577cadab6
- 4.19.174: as fbad32181af9c8c76698b16115c28a493c26a2ee
- 4.14.220: as a9b2e4a94eee352e0b3e863045e7379d83131ee2

Maybe those trees could use a backport of 3a7956e25e1d as well then?


Cheers,
  Jakub.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-06-25 10:12 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-06-24  2:03 Backporting 3a7956e25e1d to LTS kernels? Jakub Jankowski
2021-06-25 10:12 ` Greg KH

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox