* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
[not found] <20260607095727.647295505@linuxfoundation.org>
@ 2026-06-07 17:32 ` Miguel Ojeda
2026-06-08 9:55 ` Jamal Hadi Salim
0 siblings, 1 reply; 6+ messages in thread
From: Miguel Ojeda @ 2026-06-07 17:32 UTC (permalink / raw)
To: gregkh
Cc: achill, akpm, broonie, conor, f.fainelli, hargar, jonathanh,
linux-kernel, linux, lkft-triage, patches, patches, pavel,
rwarsow, shuah, sr, stable, sudipm.mukherjee, torvalds,
Miguel Ojeda, Kito Xu (veritas501), Victor Nogueira, Paolo Abeni,
Jamal Hadi Salim, Jiri Pirko, netdev
On Sun, 07 Jun 2026 11:56:37 +0200 Greg Kroah-Hartman <gregkh@linuxfoundation•org> wrote:
>
> This is the start of the stable review cycle for the 6.12.93 release.
> There are 307 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses should be made by Tue, 09 Jun 2026 09:56:47 +0000.
> Anything received after that time might be too late.
Boot-tested under QEMU for Rust x86_64, arm64 and riscv64; built-tested
for loongarch64:
Tested-by: Miguel Ojeda <ojeda@kernel•org>
arm32 also builds fine.
On arm64, I am seeing:
net/sched/act_mirred.c:451:43: warning: variable 'm_eaction' is uninitialized when used here [-Wuninitialized]
451 | is_redirect = tcf_mirred_is_act_redirect(m_eaction);
| ^~~~~~~~~
net/sched/act_mirred.c:429:18: note: initialize the variable 'm_eaction' to silence this warning
429 | int i, m_eaction;
| ^
| = 0
due to commit a01fbdecc3a2 ("net/sched: act_mirred: Fix return code in
early mirred redirect error paths") here.
And that one seems to be missing at least the assignment to the variable
that happened in commit a005fa5d7502 ("net/sched: act_mirred: Fix
blockcast recursion bypass leading to stack overflow").
I hope this helps!
Cc: Kito Xu (veritas501) <hxzene@gmail•com>
Cc: Victor Nogueira <victor@mojatatu•com>
Cc: Paolo Abeni <pabeni@redhat•com>
Cc: Jamal Hadi Salim <jhs@mojatatu•com>
Cc: Jiri Pirko <jiri@resnulli•us>
Cc: netdev@vger•kernel.org
Cheers,
Miguel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
2026-06-07 17:32 ` [PATCH 6.12 000/307] 6.12.93-rc1 review Miguel Ojeda
@ 2026-06-08 9:55 ` Jamal Hadi Salim
2026-06-08 10:08 ` Miguel Ojeda
0 siblings, 1 reply; 6+ messages in thread
From: Jamal Hadi Salim @ 2026-06-08 9:55 UTC (permalink / raw)
To: Miguel Ojeda
Cc: gregkh, achill, akpm, broonie, conor, f.fainelli, hargar,
jonathanh, linux-kernel, linux, lkft-triage, patches, patches,
pavel, rwarsow, shuah, sr, stable, sudipm.mukherjee, torvalds,
Kito Xu (veritas501), Victor Nogueira, Paolo Abeni, Jiri Pirko,
netdev
Hi Miguel,
Thanks for reporting this.
On Sun, Jun 7, 2026 at 1:32 PM Miguel Ojeda <ojeda@kernel•org> wrote:
>
> On Sun, 07 Jun 2026 11:56:37 +0200 Greg Kroah-Hartman <gregkh@linuxfoundation•org> wrote:
> >
> > This is the start of the stable review cycle for the 6.12.93 release.
> > There are 307 patches in this series, all will be posted as a response
> > to this one. If anyone has any issues with these being applied, please
> > let me know.
> >
> > Responses should be made by Tue, 09 Jun 2026 09:56:47 +0000.
> > Anything received after that time might be too late.
>
> Boot-tested under QEMU for Rust x86_64, arm64 and riscv64; built-tested
> for loongarch64:
>
> Tested-by: Miguel Ojeda <ojeda@kernel•org>
>
> arm32 also builds fine.
>
> On arm64, I am seeing:
>
> net/sched/act_mirred.c:451:43: warning: variable 'm_eaction' is uninitialized when used here [-Wuninitialized]
> 451 | is_redirect = tcf_mirred_is_act_redirect(m_eaction);
> | ^~~~~~~~~
> net/sched/act_mirred.c:429:18: note: initialize the variable 'm_eaction' to silence this warning
> 429 | int i, m_eaction;
> | ^
> | = 0
>
> due to commit a01fbdecc3a2 ("net/sched: act_mirred: Fix return code in
> early mirred redirect error paths") here.
>
> And that one seems to be missing at least the assignment to the variable
> that happened in commit a005fa5d7502 ("net/sched: act_mirred: Fix
> blockcast recursion bypass leading to stack overflow").
>
Unless i am looking at the wrong code version, in the current code i
see m_eaction is always initialized before being used.
m_eaction = READ_ONCE(m->tcfm_eaction);
Probably a compiler false positive?
As per suggestion emmited, initializing m_eaction to 0 should help.
Not sure what the process is.
Perhaps, Miguel you want to send the fix? commit should say "appease
compiler ..."
cheers,
jamal
> I hope this helps!
>
> Cc: Kito Xu (veritas501) <hxzene@gmail•com>
> Cc: Victor Nogueira <victor@mojatatu•com>
> Cc: Paolo Abeni <pabeni@redhat•com>
> Cc: Jamal Hadi Salim <jhs@mojatatu•com>
> Cc: Jiri Pirko <jiri@resnulli•us>
> Cc: netdev@vger•kernel.org
>
> Cheers,
> Miguel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
2026-06-08 9:55 ` Jamal Hadi Salim
@ 2026-06-08 10:08 ` Miguel Ojeda
2026-06-08 10:57 ` Jamal Hadi Salim
0 siblings, 1 reply; 6+ messages in thread
From: Miguel Ojeda @ 2026-06-08 10:08 UTC (permalink / raw)
To: Jamal Hadi Salim
Cc: Miguel Ojeda, gregkh, achill, akpm, broonie, conor, f.fainelli,
hargar, jonathanh, linux-kernel, linux, lkft-triage, patches,
patches, pavel, rwarsow, shuah, sr, stable, sudipm.mukherjee,
torvalds, Kito Xu (veritas501), Victor Nogueira, Paolo Abeni,
Jiri Pirko, netdev
On Mon, Jun 8, 2026 at 11:55 AM Jamal Hadi Salim <jhs@mojatatu•com> wrote:
>
> Thanks for reporting this.
You're welcome!
> Unless i am looking at the wrong code version, in the current code i
> see m_eaction is always initialized before being used.
> m_eaction = READ_ONCE(m->tcfm_eaction);
> Probably a compiler false positive?
No, it is an actual bug -- you have:
int i, m_eaction;
...
is_redirect = tcf_mirred_is_act_redirect(m_eaction);
...
m_eaction = READ_ONCE(m->tcfm_eaction);
i.e. the assignment goes later.
Maybe you are looking at mainline? Please note that this is a 6.12.y
-rc thread -- are you looking at the commit hash I mentioned?
Or maybe you are looking at `tcf_blockcast()` instead of `tcf_mirred_act()`?
I hope that helps.
Cheers,
Miguel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
2026-06-08 10:08 ` Miguel Ojeda
@ 2026-06-08 10:57 ` Jamal Hadi Salim
2026-06-08 11:15 ` Miguel Ojeda
0 siblings, 1 reply; 6+ messages in thread
From: Jamal Hadi Salim @ 2026-06-08 10:57 UTC (permalink / raw)
To: Miguel Ojeda
Cc: Miguel Ojeda, gregkh, achill, akpm, broonie, conor, f.fainelli,
hargar, jonathanh, linux-kernel, linux, lkft-triage, patches,
patches, pavel, rwarsow, shuah, sr, stable, sudipm.mukherjee,
torvalds, Kito Xu (veritas501), Victor Nogueira, Paolo Abeni,
Jiri Pirko, netdev
On Mon, Jun 8, 2026 at 6:08 AM Miguel Ojeda
<miguel.ojeda.sandonis@gmail•com> wrote:
>
> On Mon, Jun 8, 2026 at 11:55 AM Jamal Hadi Salim <jhs@mojatatu•com> wrote:
> >
> > Thanks for reporting this.
>
> You're welcome!
>
> > Unless i am looking at the wrong code version, in the current code i
> > see m_eaction is always initialized before being used.
> > m_eaction = READ_ONCE(m->tcfm_eaction);
> > Probably a compiler false positive?
>
> No, it is an actual bug -- you have:
>
> int i, m_eaction;
> ...
> is_redirect = tcf_mirred_is_act_redirect(m_eaction);
> ...
> m_eaction = READ_ONCE(m->tcfm_eaction);
>
> i.e. the assignment goes later.
>
> Maybe you are looking at mainline? Please note that this is a 6.12.y
> -rc thread -- are you looking at the commit hash I mentioned?
>
> Or maybe you are looking at `tcf_blockcast()` instead of `tcf_mirred_act()`?
>
I was certainly looking at the wrong version.
For example what you described as commit a01fbdecc3a2 ("net/sched:
act_mirred: Fix return code in early mirred redirect error paths")
shows up for me as commit e80ad525fc7e
I believe this bug slipped in during a small window but was fixed very
quickly. Probably some fix never trickled to stable.
If you can point me to the exact tree where this happens i can take a look.
Still curious: So only the arm compiler catches this?
cheers,
jamal
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
2026-06-08 10:57 ` Jamal Hadi Salim
@ 2026-06-08 11:15 ` Miguel Ojeda
2026-06-08 20:35 ` Jamal Hadi Salim
0 siblings, 1 reply; 6+ messages in thread
From: Miguel Ojeda @ 2026-06-08 11:15 UTC (permalink / raw)
To: Jamal Hadi Salim
Cc: Miguel Ojeda, gregkh, achill, akpm, broonie, conor, f.fainelli,
hargar, jonathanh, linux-kernel, linux, lkft-triage, patches,
patches, pavel, rwarsow, shuah, sr, stable, sudipm.mukherjee,
torvalds, Kito Xu (veritas501), Victor Nogueira, Paolo Abeni,
Jiri Pirko, netdev
On Mon, Jun 8, 2026 at 12:57 PM Jamal Hadi Salim <jhs@mojatatu•com> wrote:
>
> I believe this bug slipped in during a small window but was fixed very
> quickly. Probably some fix never trickled to stable.
Yeah, as I mentioned above, I think commit a005fa5d7502 ("net/sched:
act_mirred: Fix blockcast recursion bypass leading to stack overflow")
is missing (at least).
I would suggest reviewing the entire chain to see what needs to be backported.
> If you can point me to the exact tree where this happens i can take a look.
This is the 6.12.y -rc tree:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
linux-6.12.y
The stable kernel team puts the link to the repository at the top of
the thread too:
https://lore.kernel.org/stable/20260607095727.647295505@linuxfoundation.org/
> Still curious: So only the arm compiler catches this?
It likely gets caught by other arches too, i.e. I just happened to
catch it in my arm64 build.
This is Clang, not GCC, by the way.
I hope this helps!
Cheers,
Miguel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 6.12 000/307] 6.12.93-rc1 review
2026-06-08 11:15 ` Miguel Ojeda
@ 2026-06-08 20:35 ` Jamal Hadi Salim
0 siblings, 0 replies; 6+ messages in thread
From: Jamal Hadi Salim @ 2026-06-08 20:35 UTC (permalink / raw)
To: Miguel Ojeda
Cc: Miguel Ojeda, gregkh, achill, akpm, broonie, conor, f.fainelli,
hargar, jonathanh, linux-kernel, linux, lkft-triage, patches,
patches, pavel, rwarsow, shuah, sr, stable, sudipm.mukherjee,
torvalds, Kito Xu (veritas501), Victor Nogueira, Paolo Abeni,
Jiri Pirko, netdev
Hi Miguel,
On Mon, Jun 8, 2026 at 7:16 AM Miguel Ojeda
<miguel.ojeda.sandonis@gmail•com> wrote:
>
> On Mon, Jun 8, 2026 at 12:57 PM Jamal Hadi Salim <jhs@mojatatu•com> wrote:
> >
> > I believe this bug slipped in during a small window but was fixed very
> > quickly. Probably some fix never trickled to stable.
>
> Yeah, as I mentioned above, I think commit a005fa5d7502 ("net/sched:
> act_mirred: Fix blockcast recursion bypass leading to stack overflow")
> is missing (at least).
>
> I would suggest reviewing the entire chain to see what needs to be backported.
>
> > If you can point me to the exact tree where this happens i can take a look.
>
> This is the 6.12.y -rc tree:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
> linux-6.12.y
>
> The stable kernel team puts the link to the repository at the top of
> the thread too:
>
> https://lore.kernel.org/stable/20260607095727.647295505@linuxfoundation.org/
>
> > Still curious: So only the arm compiler catches this?
>
> It likely gets caught by other arches too, i.e. I just happened to
> catch it in my arm64 build.
>
> This is Clang, not GCC, by the way.
>
I think it was too early AM here when i was looking at this.
The answer was right there all along in what you said: The missing
piece is commit a005fa5d7502
cheers,
jamal
> I hope this helps!
>
> Cheers,
> Miguel
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-06-08 20:36 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20260607095727.647295505@linuxfoundation.org>
2026-06-07 17:32 ` [PATCH 6.12 000/307] 6.12.93-rc1 review Miguel Ojeda
2026-06-08 9:55 ` Jamal Hadi Salim
2026-06-08 10:08 ` Miguel Ojeda
2026-06-08 10:57 ` Jamal Hadi Salim
2026-06-08 11:15 ` Miguel Ojeda
2026-06-08 20:35 ` Jamal Hadi Salim
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox