Re: [PATCH 5.15 000/106] 5.15.49-rc1 review

From: Naresh Kamboju
Date: Tue Jun 21 2022 - 07:36:26 EST


On Tue, 21 Jun 2022 at 16:05, Greg Kroah-Hartman
<gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Tue, Jun 21, 2022 at 11:45:51AM +0200, Jason A. Donenfeld wrote:
> > On Tue, Jun 21, 2022 at 10:55:33AM +0200, Greg Kroah-Hartman wrote:
> > > On Tue, Jun 21, 2022 at 02:06:06PM +0530, Naresh Kamboju wrote:
> > > > On Mon, 20 Jun 2022 at 18:36, Greg Kroah-Hartman
> > > > <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> > > > >
> > > > > This is the start of the stable review cycle for the 5.15.49 release.
> > > > > There are 106 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 Wed, 22 Jun 2022 12:47:02 +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/v5.x/stable-review/patch-5.15.49-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-5.15.y
> > > > > and the diffstat can be found below.
> > > > >
> > > > > thanks,
> > > > >
> > > > > greg k-h
> > > >
> > > > Following commit causing regression while building allmodconfig for clang-13
> > > > on arm64, riscv and x86_64.
> > > >
> > > > > Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
> > > > > netfs: gcc-12: temporarily disable '-Wattribute-warning' for now
> > > >
> > > > fs/afs/inode.c:29:32: error: unknown warning group
> > > > '-Wattribute-warning', ignored [-Werror,-Wunknown-warning-option]
> > > > #pragma GCC diagnostic ignored "-Wattribute-warning"
> > > > ^
> > > > 1 error generated.
> > > >
> > > > Regressions:
> > > > - arm64/build/clang-13-allmodconfig - Failed
> > > > - riscv/build/clang-13-allmodconfig - Failed
> > > > - x86_64/build/clang-13-allmodconfig - Failed
> > >
> > > Does Linus's tree also show this issue?
> >
> > Linus' tree got rid of the pragma with:
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/fs/afs/inode.c?id=874c8ca1e60b2c564a48f7e7acc40d328d5c8733
>
> That isn't going to work on 5.15.y at all without a lot of hand tweaking :(
>
> > and then a subsequent cleanup:
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/fs/afs/inode.c?id=e81fb4198e27925b151aad1450e0fd607d6733f
>
> That doesn't work on 5.18 either.
>
> I guess I'll go revert the 5.15 patch here for now and let others sort
> it all out when they want to build 5.15 with gcc-12.

Whereas clang-14 builds pass. I am considering this as a waiver.
The problem is with clang-13. you may keep the above patch which is
adding support for gcc-12.

We will stop building with clang-13 and we will upgrade our toolchains to
clang-15 (when released and current clang-nightly builds are in production)
and gcc-12 builds are running in staging. We would like to bump versions soon.

- Naresh