Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

[Naresh Kamboju]

On Wed, 23 Apr 2025 at 20:16, Greg Kroah-Hartman
<gregkh@linuxfoundation.org> wrote:
>
> This is the start of the stable review cycle for the 6.1.135 release.
> There are 291 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 Fri, 25 Apr 2025 14:25:27 +0000.
> Anything received after that time might be too late.
>
> The whole patch series can be found in one patch at:
>         https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.135-rc1.gz
> or in the git tree and branch at:
>         git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
> and the diffstat can be found below.
>
> thanks,
>
> greg k-h

Regressions on arm, riscv and x86_64 with following kernel configs with
clang-20 and clang-nightly toolchains on stable-rc 6.1.135-rc1.

Build regressions:
* arm, build
  - clang-20-allmodconfig

* i386, build
  - clang-nightly-lkftconfig-kselftest

* riscv, build
  - clang-20-allmodconfig

* x86_64, build
  - clang-20-allmodconfig
  - clang-nightly-lkftconfig-kselftest

Regression Analysis:
 - New regression? Yes
 - Reproducibility? Yes

Build regression: arm allmodconfig variable 'is_redirect' is used
uninitialized whenever 'if' condition is true

Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>

## Build error:
net/sched/act_mirred.c:265:6: error: variable 'is_redirect' is used
uninitialized whenever 'if' condition is true
[-Werror,-Wsometimes-uninitialized]
  265 |         if (unlikely(!(dev->flags & IFF_UP)) ||
!netif_carrier_ok(dev)) {
      |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

## Source
* Kernel version: 6.1.135-rc1
* Git tree: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
* Git sha: b8b5da130779a27bf7d189bd3b40ebf59ee0e0e9
* Git describe: v6.1.134-292-gb8b5da130779
* Project details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.134-292-gb8b5da130779
* Architectures: arm riscv x86_64
* Toolchains: clang-20, clang-nightly
* Kconfigs: allmodconfig

## Build
* Build log: https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.134-292-gb8b5da130779/testrun/28211709/suite/build/test/clang-20-allmodconfig/log
* Build history:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.134-292-gb8b5da130779/testrun/28211709/suite/build/test/clang-20-allmodconfig/history/
* Build details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.134-292-gb8b5da130779/testrun/28211709/suite/build/test/clang-20-allmodconfig/details/
* Build link: https://storage.tuxsuite.com/public/linaro/lkft/builds/2w8QtyR8377WMNLrbtsmGi2lkaI/
* Kernel config:
https://storage.tuxsuite.com/public/linaro/lkft/builds/2w8QtyR8377WMNLrbtsmGi2lkaI/config

## Steps to reproduce on riscv
 - tuxmake --runtime podman --target-arch arm --toolchain clang-20
--kconfig allmodconfig LLVM=1 LLVM_IAS=1

--
Linaro LKFT
https://lkft.linaro.org

Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

[Greg Kroah-Hartman]

On Thu, Apr 24, 2025 at 07:01:02PM +0530, Naresh Kamboju wrote:
> On Wed, 23 Apr 2025 at 20:16, Greg Kroah-Hartman
> <gregkh@linuxfoundation.org> wrote:
> >
> > This is the start of the stable review cycle for the 6.1.135 release.
> > There are 291 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 Fri, 25 Apr 2025 14:25:27 +0000.
> > Anything received after that time might be too late.
> >
> > The whole patch series can be found in one patch at:
> >         https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.135-rc1.gz
> > or in the git tree and branch at:
> >         git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
> > and the diffstat can be found below.
> >
> > thanks,
> >
> > greg k-h
> 
> Regressions on arm, riscv and x86_64 with following kernel configs with
> clang-20 and clang-nightly toolchains on stable-rc 6.1.135-rc1.
> 
> Build regressions:
> * arm, build
>   - clang-20-allmodconfig
> 
> * i386, build
>   - clang-nightly-lkftconfig-kselftest
> 
> * riscv, build
>   - clang-20-allmodconfig
> 
> * x86_64, build
>   - clang-20-allmodconfig
>   - clang-nightly-lkftconfig-kselftest
> 
> Regression Analysis:
>  - New regression? Yes
>  - Reproducibility? Yes
> 
> Build regression: arm allmodconfig variable 'is_redirect' is used
> uninitialized whenever 'if' condition is true
> 
> Reported-by: Linux Kernel Functional Testing <lkft@linaro.org>
> 
> ## Build error:
> net/sched/act_mirred.c:265:6: error: variable 'is_redirect' is used
> uninitialized whenever 'if' condition is true
> [-Werror,-Wsometimes-uninitialized]
>   265 |         if (unlikely(!(dev->flags & IFF_UP)) ||
> !netif_carrier_ok(dev)) {
>       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Odd this isn't showing up in newer releases, as this is an old commit
and nothing has changed in this file since then (it showed up in 6.8.)

Is there some follow-up commit somewhere that I'm missing that resolved
this issue?

thanks,

greg k-h

Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

[Arnd Bergmann]

On Thu, Apr 24, 2025, at 15:41, Greg Kroah-Hartman wrote:
> On Thu, Apr 24, 2025 at 07:01:02PM +0530, Naresh Kamboju wrote:
>> 
>> ## Build error:
>> net/sched/act_mirred.c:265:6: error: variable 'is_redirect' is used
>> uninitialized whenever 'if' condition is true
>> [-Werror,-Wsometimes-uninitialized]
>>   265 |         if (unlikely(!(dev->flags & IFF_UP)) ||
>> !netif_carrier_ok(dev)) {
>>       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> Odd this isn't showing up in newer releases, as this is an old commit
> and nothing has changed in this file since then (it showed up in 6.8.)
>
> Is there some follow-up commit somewhere that I'm missing that resolved
> this issue?

I think the difference is commit 16085e48cb48 ("net/sched: act_mirred:
Create function tcf_mirred_to_dev and improve readability") from 
v6.8, which adds the initialization that 166c2c8a6a4d ("net/sched:
act_mirred: don't override retval if we already lost the skb")
relies on.

      Arnd

Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

[Greg Kroah-Hartman]

On Thu, Apr 24, 2025 at 04:34:04PM +0200, Arnd Bergmann wrote:
> On Thu, Apr 24, 2025, at 15:41, Greg Kroah-Hartman wrote:
> > On Thu, Apr 24, 2025 at 07:01:02PM +0530, Naresh Kamboju wrote:
> >> 
> >> ## Build error:
> >> net/sched/act_mirred.c:265:6: error: variable 'is_redirect' is used
> >> uninitialized whenever 'if' condition is true
> >> [-Werror,-Wsometimes-uninitialized]
> >>   265 |         if (unlikely(!(dev->flags & IFF_UP)) ||
> >> !netif_carrier_ok(dev)) {
> >>       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >
> > Odd this isn't showing up in newer releases, as this is an old commit
> > and nothing has changed in this file since then (it showed up in 6.8.)
> >
> > Is there some follow-up commit somewhere that I'm missing that resolved
> > this issue?
> 
> I think the difference is commit 16085e48cb48 ("net/sched: act_mirred:
> Create function tcf_mirred_to_dev and improve readability") from 
> v6.8, which adds the initialization that 166c2c8a6a4d ("net/sched:
> act_mirred: don't override retval if we already lost the skb")
> relies on.

Ok, that didn't apply cleanly either, so I'm just going to drop this
backported patch and wait for the submitter to fix it up and resend it.

thanks,

greg k-h

Re: [PATCH 6.1 000/291] 6.1.135-rc1 review

[He Zhe]

On 2025/4/24 23:28, Greg Kroah-Hartman wrote:
> On Thu, Apr 24, 2025 at 04:34:04PM +0200, Arnd Bergmann wrote:
>> On Thu, Apr 24, 2025, at 15:41, Greg Kroah-Hartman wrote:
>>> On Thu, Apr 24, 2025 at 07:01:02PM +0530, Naresh Kamboju wrote:
>>>> ## Build error:
>>>> net/sched/act_mirred.c:265:6: error: variable 'is_redirect' is used
>>>> uninitialized whenever 'if' condition is true
>>>> [-Werror,-Wsometimes-uninitialized]
>>>>   265 |         if (unlikely(!(dev->flags & IFF_UP)) ||
>>>> !netif_carrier_ok(dev)) {
>>>>       |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>> Odd this isn't showing up in newer releases, as this is an old commit
>>> and nothing has changed in this file since then (it showed up in 6.8.)
>>>
>>> Is there some follow-up commit somewhere that I'm missing that resolved
>>> this issue?
>> I think the difference is commit 16085e48cb48 ("net/sched: act_mirred:
>> Create function tcf_mirred_to_dev and improve readability") from 
>> v6.8, which adds the initialization that 166c2c8a6a4d ("net/sched:
>> act_mirred: don't override retval if we already lost the skb")
>> relies on.
> Ok, that didn't apply cleanly either, so I'm just going to drop this
> backported patch and wait for the submitter to fix it up and resend it.

Sorry for the inconvenience. We have fixed the above issue found by clang-20
and will sent out v2 soon.

Thanks,
Zhe

>
> thanks,
>
> greg k-h