pmu-chain-promotion kvm-unit-test failures in mainline

pmu-chain-promotion kvm-unit-test failures in mainline

[Mark Brown]

[-- Attachment #1.1: Type: text/plain, Size: 3680 bytes --]

Arm's internal test infrastructure has started seeing failures for the
pmu-chain-promption test in kvm-unit-tests on TX2 when running mainline
and kvm-unit-tests commit 2480430a36102f8ea276b3bfb1d64d5dacc23b8f
("configure: Show the option in case it is not known").  The log I'm
seeing from the test runner is:

TESTNAME=pmu-chain-promotion TIMEOUT=90s ACCEL= ./arm/run arm/pmu.flat -smp 1 -append 'pmu-chain-promotion'
FAIL pmu-chain-promotion (7 tests, 2 unexpected failures)

Unfortunately I don't have direct access to a TX2 so can't readily
alter the test run other than replacing the kernel.  I believe the
specific failures were:

FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: CHAIN counter was incremented and overflow
FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: 32b->64b CHAIN counter was incremented and overflow

A bisect (log below) identified 7af9da8ce8f9a16221e ("thunderbolt: Add
quirk to disable CLx") as the breaking commit which doesn't seem
entirely credible unless there's some image size thing going on, the
issue manifests with v6.3-rc3 and not with v6.3-rc1.

git bisect start
# bad: [009795d283d1f9f043e5a4ff97d4140cde17e2d3] Add linux-next specific files for 20230411
git bisect bad 009795d283d1f9f043e5a4ff97d4140cde17e2d3
# bad: [0d3eb744aed40ffce820cded61d7eac515199165] Merge tag 'urgent-rcu.2023.04.07a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu
git bisect bad 0d3eb744aed40ffce820cded61d7eac515199165
# good: [fe15c26ee26efa11741a7b632e9f23b01aca4cc6] Linux 6.3-rc1
git bisect good fe15c26ee26efa11741a7b632e9f23b01aca4cc6
# good: [d0072ca529674c36421023ffe90837a7de9387f3] mm: mmap: remove newline at the end of the trace
git bisect good d0072ca529674c36421023ffe90837a7de9387f3
# bad: [b527ac44eb1782cf179d4e08ceda7d2a9643aff5] Merge tag 'for-6.3/dm-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
git bisect bad b527ac44eb1782cf179d4e08ceda7d2a9643aff5
# good: [d7b5c942f0752894a91a925c7249116333d4ab25] Merge tag 'for-v6.3-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply
git bisect good d7b5c942f0752894a91a925c7249116333d4ab25
# good: [18940c888c85fd7527375343bd4fcc94a540c69c] Merge tag 'sched_urgent_for_v6.3_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
git bisect good 18940c888c85fd7527375343bd4fcc94a540c69c
# bad: [69620d5c030553e81562f6f6f490c1fdd7f798fd] Merge tag 'drm-misc-fixes-2023-03-30' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes
git bisect bad 69620d5c030553e81562f6f6f490c1fdd7f798fd
# bad: [451b15ed138ec15bffbebb58a00ebdd884c3e659] usb: chipidea: core: fix possible concurrent when switch role
git bisect bad 451b15ed138ec15bffbebb58a00ebdd884c3e659
# good: [1272fd652a226ccb34e9f47371b6121948048438] usb: cdns3: Fix issue with using incorrect PCI device function
git bisect good 1272fd652a226ccb34e9f47371b6121948048438
# bad: [1716efdb07938bd6510e1127d02012799112c433] thunderbolt: Use const qualifier for `ring_interrupt_index`
git bisect bad 1716efdb07938bd6510e1127d02012799112c433
# bad: [cd0c1e582b055dea615001b8bd8eccaf6f69f7ce] thunderbolt: Add missing UNSET_INBOUND_SBTX for retimer access
git bisect bad cd0c1e582b055dea615001b8bd8eccaf6f69f7ce
# bad: [acec726473822bc6b585961f4ca2a11fa7f28341] thunderbolt: Fix memory leak in margining
git bisect bad acec726473822bc6b585961f4ca2a11fa7f28341
# bad: [7af9da8ce8f9a16221ecd8ba4280582f5bd452fc] thunderbolt: Add quirk to disable CLx
git bisect bad 7af9da8ce8f9a16221ecd8ba4280582f5bd452fc
# first bad commit: [7af9da8ce8f9a16221ecd8ba4280582f5bd452fc] thunderbolt: Add quirk to disable CLx

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: pmu-chain-promotion kvm-unit-test failures in mainline

[Marc Zyngier]

On Tue, 11 Apr 2023 22:15:02 +0100,
Mark Brown <broonie@kernel.org> wrote:
> 
> [1  <text/plain; us-ascii (7bit)>]
> Arm's internal test infrastructure has started seeing failures for the
> pmu-chain-promption test in kvm-unit-tests on TX2 when running mainline
> and kvm-unit-tests commit 2480430a36102f8ea276b3bfb1d64d5dacc23b8f
> ("configure: Show the option in case it is not known").  The log I'm
> seeing from the test runner is:
> 
> TESTNAME=pmu-chain-promotion TIMEOUT=90s ACCEL= ./arm/run arm/pmu.flat -smp 1 -append 'pmu-chain-promotion'
> FAIL pmu-chain-promotion (7 tests, 2 unexpected failures)
> 
> Unfortunately I don't have direct access to a TX2 so can't readily
> alter the test run other than replacing the kernel.  I believe the
> specific failures were:
> 
> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: CHAIN counter was incremented and overflow
> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: 32b->64b CHAIN counter was incremented and overflow

Is this related to [1]?

	M.

[1] https://lore.kernel.org/r/20230315110725.1215523-1-eric.auger@redhat.com

-- 
Without deviation from the norm, progress is not possible.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: pmu-chain-promotion kvm-unit-test failures in mainline

[Eric Auger]

Hi,
On 4/12/23 09:39, Marc Zyngier wrote:
> On Tue, 11 Apr 2023 22:15:02 +0100,
> Mark Brown <broonie@kernel.org> wrote:
>>
>> [1  <text/plain; us-ascii (7bit)>]
>> Arm's internal test infrastructure has started seeing failures for the
>> pmu-chain-promption test in kvm-unit-tests on TX2 when running mainline
>> and kvm-unit-tests commit 2480430a36102f8ea276b3bfb1d64d5dacc23b8f
>> ("configure: Show the option in case it is not known").  The log I'm
>> seeing from the test runner is:
>>
>> TESTNAME=pmu-chain-promotion TIMEOUT=90s ACCEL= ./arm/run arm/pmu.flat -smp 1 -append 'pmu-chain-promotion'
>> FAIL pmu-chain-promotion (7 tests, 2 unexpected failures)
>>
>> Unfortunately I don't have direct access to a TX2 so can't readily
>> alter the test run other than replacing the kernel.  I believe the
>> specific failures were:
>>
>> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: CHAIN counter was incremented and overflow
>> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: 32b->64b CHAIN counter was incremented and overflow
> 
> Is this related to [1]?
> 
> 	M.
> 
> [1] https://lore.kernel.org/r/20230315110725.1215523-1-eric.auger@redhat.com

Yes please let me know if it fixes your issue. I also tried bisection
but it did not lead to anything.

Thanks

Eric
> 


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: pmu-chain-promotion kvm-unit-test failures in mainline

[Mark Rutland]

On Wed, Apr 12, 2023 at 10:00:40AM +0200, Eric Auger wrote:
> Hi,
> On 4/12/23 09:39, Marc Zyngier wrote:
> > On Tue, 11 Apr 2023 22:15:02 +0100,
> > Mark Brown <broonie@kernel.org> wrote:
> >>
> >> [1  <text/plain; us-ascii (7bit)>]
> >> Arm's internal test infrastructure has started seeing failures for the
> >> pmu-chain-promption test in kvm-unit-tests on TX2 when running mainline
> >> and kvm-unit-tests commit 2480430a36102f8ea276b3bfb1d64d5dacc23b8f
> >> ("configure: Show the option in case it is not known").  The log I'm
> >> seeing from the test runner is:
> >>
> >> TESTNAME=pmu-chain-promotion TIMEOUT=90s ACCEL= ./arm/run arm/pmu.flat -smp 1 -append 'pmu-chain-promotion'
> >> FAIL pmu-chain-promotion (7 tests, 2 unexpected failures)
> >>
> >> Unfortunately I don't have direct access to a TX2 so can't readily
> >> alter the test run other than replacing the kernel.  I believe the
> >> specific failures were:
> >>
> >> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: CHAIN counter was incremented and overflow
> >> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: 32b->64b CHAIN counter was incremented and overflow
> > 
> > Is this related to [1]?
> > 
> > 	M.
> > 
> > [1] https://lore.kernel.org/r/20230315110725.1215523-1-eric.auger@redhat.com
> 
> Yes please let me know if it fixes your issue. I also tried bisection
> but it did not lead to anything.

FWIW, that worked for me; I replied there with a Tested-by (and test results).

Thanks,
Mark.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: pmu-chain-promotion kvm-unit-test failures in mainline

[Eric Auger]

Hi Mark,

On 4/12/23 10:48, Mark Rutland wrote:
> On Wed, Apr 12, 2023 at 10:00:40AM +0200, Eric Auger wrote:
>> Hi,
>> On 4/12/23 09:39, Marc Zyngier wrote:
>>> On Tue, 11 Apr 2023 22:15:02 +0100,
>>> Mark Brown <broonie@kernel.org> wrote:
>>>>
>>>> [1  <text/plain; us-ascii (7bit)>]
>>>> Arm's internal test infrastructure has started seeing failures for the
>>>> pmu-chain-promption test in kvm-unit-tests on TX2 when running mainline
>>>> and kvm-unit-tests commit 2480430a36102f8ea276b3bfb1d64d5dacc23b8f
>>>> ("configure: Show the option in case it is not known").  The log I'm
>>>> seeing from the test runner is:
>>>>
>>>> TESTNAME=pmu-chain-promotion TIMEOUT=90s ACCEL= ./arm/run arm/pmu.flat -smp 1 -append 'pmu-chain-promotion'
>>>> FAIL pmu-chain-promotion (7 tests, 2 unexpected failures)
>>>>
>>>> Unfortunately I don't have direct access to a TX2 so can't readily
>>>> alter the test run other than replacing the kernel.  I believe the
>>>> specific failures were:
>>>>
>>>> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: CHAIN counter was incremented and overflow
>>>> FAIL: pmu: pmu-chain-promotion: 32-bit overflows: CHAIN counter enabled: 32b->64b CHAIN counter was incremented and overflow
>>>
>>> Is this related to [1]?
>>>
>>> 	M.
>>>
>>> [1] https://lore.kernel.org/r/20230315110725.1215523-1-eric.auger@redhat.com
>>
>> Yes please let me know if it fixes your issue. I also tried bisection
>> but it did not lead to anything.
> 
> FWIW, that worked for me; I replied there with a Tested-by (and test results).

Thank you for testing!

Eric
> 
> Thanks,
> Mark.
> 


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel