Re: Linux 6.1-rc2

Re: Linux 6.1-rc2

[Guenter Roeck]

On Sun, Oct 23, 2022 at 04:03:45PM -0700, Linus Torvalds wrote:
> Hmm. Usually rc2 is a pretty quiet week, and it mostly started out
> that way too, but then things took a turn for the strange. End result:
> 6.1-rc2 ended up being unusually large.
> 
> The main reason is fairly benign, though: Mauro had screwed up the
> media tree pull request during the merge window, so rc2 ends up having
> a "oops, here's the part that was missing" moment. Since it had all
> been in linux-next (yes, I checked, so nobody else should try that
> trick), I ended up pulling that missing part during the rc2 week.
> 
> But if you ignore that media tree portion, things look pretty normal for an rc2.
> 
> Anyway, ignoring those media changes, we have a little bit of
> everything in here - arch updates, drivers (gpu, device mapper,
> networking), EFI, some core kernel fixes (mm, scheduler, cgroup,
> networking). The full shortlog is appended (and that shortlog does
> include the media pieces).
> 
> Please do go test,

Build results:
	total: 152 pass: 152 fail: 0
Qemu test results:
	total: 499 pass: 499 fail: 0

================

Runtime warnings

powerpc
-------

BUG: using smp_processor_id() in preemptible [00000000] code: swapper/0/1

Not a new problem (seen as far back as v5.15.y), but fixed by:
    "powerpc/64s: Disable preemption in hash lazy mmu mode"
    "powerpc/64s: Fix hash__change_memory_range preemption warning"
    "powerpc: fix reschedule bug in KUAP-unlocked user copy"

at:
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013151647.1857994-1-npiggin@gmail.com/
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013151647.1857994-2-npiggin@gmail.com/
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013151647.1857994-3-npiggin@gmail.com/

================================
WARNING: inconsistent lock state
inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.

Fixed by:
    "powerpc/64s: Add lockdep for HPTE lock"
    "powerpc/64s: make HPTE lock and native_tlbie_lock irq-safe"
    "powerpc/64s: make linear_map_hash_lock a raw spinlock:

at:
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013230710.1987253-1-npiggin@gmail.com/
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013230710.1987253-2-npiggin@gmail.com/
https://patchwork.ozlabs.org/project/linuxppc-dev/patch/20221013230710.1987253-3-npiggin@gmail.com/

Guenter

Re: Linux 6.1-rc2

[Linus Torvalds]

On Tue, Oct 25, 2022 at 9:24 AM Guenter Roeck <linux@roeck-us.net> wrote:
>
> Build results:
>         total: 152 pass: 152 fail: 0
> Qemu test results:
>         total: 499 pass: 499 fail: 0

Woo-hoo!

That was quick, for once.

> Runtime warnings

Oh.

Well, close enough, and those fixes are presumably pending too.

              Linus

Re: Linux 6.1-rc2

[Guenter Roeck]

On Tue, Oct 25, 2022 at 09:41:50AM -0700, Linus Torvalds wrote:
> On Tue, Oct 25, 2022 at 9:24 AM Guenter Roeck <linux@roeck-us.net> wrote:
> >
> > Build results:
> >         total: 152 pass: 152 fail: 0
> > Qemu test results:
> >         total: 499 pass: 499 fail: 0
> 
> Woo-hoo!
> 
> That was quick, for once.
> 

Yes, it was. I now have an even better one, with a new personal
milestone reached (500 qemu boot tests):

Build results:
	total: 152 pass: 152 fail: 0
Qemu test results:
	total: 500 pass: 500 fail: 0
	       ^^^

... after getting yet another qemu machine to boot.

> > Runtime warnings
> 
> Oh.
> 
> Well, close enough, and those fixes are presumably pending too.
> 

Let's hope so. I think I forgot to copy the ppc maintainers last week,
so we'll see if we get some feedback on the status of those problems.

Thanks,
Guenter

Re: Linux 6.1-rc2

[Michael Ellerman]

Guenter Roeck <linux@roeck-us.net> writes:
> On Tue, Oct 25, 2022 at 09:41:50AM -0700, Linus Torvalds wrote:
>> On Tue, Oct 25, 2022 at 9:24 AM Guenter Roeck <linux@roeck-us.net> wrote:
>> >
>> > Build results:
>> >         total: 152 pass: 152 fail: 0
>> > Qemu test results:
>> >         total: 499 pass: 499 fail: 0
>> 
>> Woo-hoo!
>> 
>> That was quick, for once.
>> 
>
> Yes, it was. I now have an even better one, with a new personal
> milestone reached (500 qemu boot tests):
>
> Build results:
> 	total: 152 pass: 152 fail: 0
> Qemu test results:
> 	total: 500 pass: 500 fail: 0
> 	       ^^^
>
> ... after getting yet another qemu machine to boot.
>
>> > Runtime warnings
>> 
>> Oh.
>> 
>> Well, close enough, and those fixes are presumably pending too.
>> 
>
> Let's hope so. I think I forgot to copy the ppc maintainers last week,
> so we'll see if we get some feedback on the status of those problems.

I have those in my fixes-test branch. While testing one of them I
stumbled on some issues that I've been trying to confirm are existing
and not regressions.

But I plan for those fixes to hit rc3.

cheers