Manual Linux regressions report for mainline [2022-09-30]

Manual Linux regressions report for mainline [2022-09-30]

[Thorsten Leemhuis]

Hi Linus! A quick manual regression report because we're close to the
end of this devel cycle:


Graphical issues on Lenovo Yoga 7 14ARB7 laptop since v6.0-rc1
--------------------------------------------------------------
https://lore.kernel.org/lkml/c1f8886a-5624-8f49-31b1-e42b6d20dcf5@augustwikerfors.se/

Unlikely to hit many people afaics. A fix is posted and somewhat
reviewed, but not yet merged by the maintainers afaics:
https://lore.kernel.org/amd-gfx/20220928172726.128863-1-sunpeng.li@amd.com/


arm64 / rock960 : kernel panic (NULL pointer dereference)
---------------------------------------------------------
https://lore.kernel.org/lkml/4f398e4c-973b-7843-2f75-27324d109277@linaro.org/

A fix for that one is in next and might be heading your way:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=master&id=f340ed8664a55a467850ec1689996e63d9ee971a


qemu/KVM boot failures after commit 'iomap: add support for dma aligned
direct-io'
---------------------------------------------------------------------
https://lore.kernel.org/all/fb869c88bd48ea9018e1cc349918aa7cdd524931.camel@redhat.com/

Reported yesterday, caused by bf8d08532bc1. A kernel change exposed a
bug in Qemu:
https://lore.kernel.org/all/32db4f89-a83f-aac4-5d27-0801bdca60bf@redhat.com/

I wonder if that commit should be reverted for now.


efi_pstore and ramoops backends are unable to properly decompress dmesg
sometimes
-----------------------------------------------------------------------
https://lore.kernel.org/all/20220929215515.276486-1-gpiccoli@igalia.com/

Reported in the past 24 hours, includes a revert. Kees wasn't able to
reproduce the problem. Guess developers will look into this today.


system hangs on Flatcar Container Linux
---------------------------------------
https://lore.kernel.org/all/c77bf00f-4618-7149-56f1-b8d1664b9d07@linux.microsoft.com/

Reported two days ago against stable (and thus not strictly a mainline
regression), where it's caused by a backport of 51ae846cff5 (merged this
cycle). The latter had also triggered a syzbot problem earlier (see
thread). Investigation ongoing.


---
That's all I'm aware of, HTH, Thorsten

Re: Manual Linux regressions report for mainline [2022-09-30]

[Linus Torvalds]

On Fri, Sep 30, 2022 at 4:44 AM Thorsten Leemhuis
<regressions@leemhuis.info> wrote:
>
> Graphical issues on Lenovo Yoga 7 14ARB7 laptop since v6.0-rc1
> --------------------------------------------------------------
> https://lore.kernel.org/lkml/c1f8886a-5624-8f49-31b1-e42b6d20dcf5@augustwikerfors.se/
>
> Unlikely to hit many people afaics. A fix is posted and somewhat
> reviewed, but not yet merged by the maintainers afaics:
> https://lore.kernel.org/amd-gfx/20220928172726.128863-1-sunpeng.li@amd.com/

Actually, it's already in my tree because it came in with the drm
fixes pull this morning.

> efi_pstore and ramoops backends are unable to properly decompress dmesg
> sometimes
> -----------------------------------------------------------------------
> https://lore.kernel.org/all/20220929215515.276486-1-gpiccoli@igalia.com/
>
> Reported in the past 24 hours, includes a revert. Kees wasn't able to
> reproduce the problem. Guess developers will look into this today.

The revert also made it to me this morning, so that people can look
into it after 6.0.

The others are still open.

                   Linus