Re: Linux 7.1-rc2

[Jani Nikula <jani.nikula@linux.intel.com>]

On Mon, 04 May 2026, Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:
> On Mon, May 04, 2026 at 06:52:02AM +0200, Luna Jernberg wrote:
>> Hey!
>> 
>> Having the same backlight problem as in rc1
>
> I think you can try to find a time and bisect [1] to the problematic commit.
> Note, using `ccache` (your distro should have that) may speed up a lot the
> build process, hence drastically decrease the time needed to find the culprit.
> Taking into account the amount of changes in the kernel it usually takes less
> than 14 steps (basically amount of how many times you need to build a kernel).

The bisect is likely the quickest way to root cause this.

But if that's too much to ask, please file a bug as described at [1],
with the full dmesg attached, with the proper drm.debugs enabled.

I've seen [2] but it's not the full dmesg, and it doesn't have the
appropriate drm.debugs enabled (it's a bit mask).

The replies to Linus' -rc mails give an air of urgency, but there's just
not enough actionable information here. There's not going to be a fix in
-rc3 either if we don't know what the problem is!


BR,
Jani.


[1] https://drm.pages.freedesktop.org/intel-docs/how-to-file-i915-bugs.html
[2] https://lore.kernel.org/r/CADo9pHiFE6fX_XxYWL+0UEqDLsXQYGzfvnFF5NK_NWbvMsz25Q@mail.gmail.com

>
>> Den sön 3 maj 2026 kl 23:41 skrev Linus Torvalds
>> <torvalds@linux-foundation.org>:
>> >
>> > I bring you tidings of another regular rc release - 7.1-rc2 is out,
>> > and looks fairly normal.
>> >
>> > Now, the raw diffstat doesn't _look_ normal, with half of the diff
>> > being to the kvm selftests, but that's pretty much entirely due to
>> > just renaming in the selftests to have the same naming conventions as
>> > the kernel code has (wrt both variable and type names), so it all
>> > looks big and strange, but you should just ignore that oddity.
>> >
>> > And when that part is masked out, the rest looks very normal - about
>> > half various driver fixes (gpu and networking dominating as usual),
>> > with the rest being spread all over the place.
>> >
>> > It's not small, and while it's a bit early to say for sure, I do
>> > suspect we're seeing the same continued pattern of more patches than
>> > usual - probably due to AI tooling - that we saw in 7.0.
>> >
>> > Let's keep testing,
>
> [1]:
> https://docs.kernel.org/admin-guide/bug-bisect.html
> https://wiki.ubuntu.com/Kernel/KernelBisection
> https://wiki.gentoo.org/wiki/Kernel_git-bisect

-- 
Jani Nikula, Intel