Linux 4.9: Reported regressions as of Sunday, 2016-10-30

Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Thorsten Leemhuis]

Hi! Here is my second regression report for Linux 4.9. It lists 14
regressions I'm aware of. 4 of them are new; 3 got fixed since last weeks report.

As always: Are you aware of any other regressions? Then please let me
know (simply CC regressions@leemhuis.info). And please tell me if there
is anything in the report that shouldn't be there.

Ciao, Thorsten

== Current regressions ==

Desc: tpm0: TPM self test failed
Repo: 2016-10-28 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1259943.html
Stat: 2016-10-28 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1260452.html https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1256705.html 
Note: Fix hopefully soon heading upstream

Desc: Radeon Oops on shutdown
Repo: 2016-10-19 https://bugzilla.kernel.org/show_bug.cgi?id=178421
Stat: 2016-10-30 https://bugzilla.kernel.org/show_bug.cgi?id=178421#c6
Note: WIP

Desc: module loadling broken due to kbuild changes
Repo: 2016-10-15 http://www.gossamer-threads.com/lists/linux/kernel/2544734
Stat: 2016-10-27 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1259418.html
Note: Fix available, waiting for Michal to get back from vacation; wondering if those will fix https://bugzilla.kernel.org/show_bug.cgi?id=185581 and https://lkml.org/lkml/2016/10/27/471 and https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1250105.html as well

Desc: pci: artpec-6: imprecise external abort
Repo: 2016-10-14 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1249646.html
Stat: 2016-10-14 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1249922.html
Note: Patch available

Desc: PPC32: fails to boot on my PowerBook G4 Aluminum; bisected to commit 05fd007e4629
Repo: 2016-10-20 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1253391.html
Stat: 2016-10-22 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1255516.html https://www.linux-mips.org/archives/linux-mips/2016-10/msg00176.html https://lkml.org/lkml/2016/10/18/142
Note: Larry made a hack that works for him.

Desc: "Nokia N900 (omap3-n900) with ""VDD1: ramp_delay not set"" string in printk frequently"
Repo: 2016-10-19 https://bugzilla.kernel.org/show_bug.cgi?id=178371
Stat: n/a https://lkml.org/lkml/2016/10/27/527
Note: discussion on lkml ongoing

Desc: radeon performance drop from 4.8 to 4.9-rc1 in Shadow of Mordor
Repo: 2016-10-14 https://bugzilla.kernel.org/show_bug.cgi?id=178221 https://lists.freedesktop.org/archives/dri-devel/2016-October/120693.html
Stat: 2016-10-20 https://lists.freedesktop.org/archives/dri-devel/2016-October/121427.html
Note: Stuck? Poked bugzilla

Desc: unable to handle kernel NULL pointer dereference at fuse_setattr
Repo: 2016-10-17 https://bugzilla.kernel.org/show_bug.cgi?id=177801
Stat: 2016-10-18 https://bugzilla.kernel.org/show_bug.cgi?id=177801#c5
Note: Fix heading upstream

Desc: Skylake gen6 suspend/resume video regression
Repo: 2016-10-16 https://bugzilla.kernel.org/show_bug.cgi?id=177731
Stat: 2016-10-25 https://bugzilla.kernel.org/show_bug.cgi?id=177731#c3
Note: WIP

Desc: warning in intel_dp_aux_transfer: CPU: 0 PID: 4 at drivers/gpu/drm/i915/intel_dp.c:1062 intel_dp_aux_transfer+0x1ed/0x230#
Repo: 2016-10-16 https://bugzilla.kernel.org/show_bug.cgi?id=177701
Stat: 2016-10-27 https://bugs.freedesktop.org/show_bug.cgi?id=97344
Note: Poked Janni to give a statement

Desc: """Failed to find cpu0 device node"" in dmesg"
Repo: 2016-10-16 https://bugzilla.kernel.org/show_bug.cgi?id=177681 https://bugzilla.kernel.org/show_bug.cgi?id=180031
Stat: 2016-10-28 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1260133.html
Note: Patches afaics in the work

Desc: boot failure of Intel Mobile Internet Devices due to a change in the PCI subsystem that appeared in v4.9-rc1.
Repo: 2016-10-23 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1255643.html
Stat: 2016-10-26 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1258579.html
Note: Fix proposed

Desc: 4.9-rc1 boot regression, ambiguous bisect result
Repo: 2016-10-19 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1253369.html
Stat: 2016-10-21 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1255296.html
Note: Poked list, as this looks stuck or was is discussed (or even fixed) somewhere else


== Stalled, waiting for feedback from reporter ==

Desc: can't boot with root fs on md raid 0; mdadm: no devices listed in conf file were found. 
Repo: 2016-10-17 https://bugzilla.kernel.org/show_bug.cgi?id=178211
Stat: 2016-10-18 https://bugzilla.kernel.org/show_bug.cgi?id=178211#c1
Note: Root cause unknown; might be a controller driver issue


== Going to be removed from the list ==

Desc: some gpio drivers broken by commit 762c2e46
Repo: 2016-10-18 https://www.spinics.net/lists/linux-gpio/msg17283.html
Stat: 2016-10-21 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1253906.html
Note: Not a 4.9 regression


== Fixed since last report ==

Desc: "Regression with 0b9e2988ab22 (""ahci: use pci_alloc_irq_vectors"")"
Repo: 2016-10-21 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1254825.html
Fix:  https://git.kernel.org/torvalds/c/17a51f12cfbd2814fd35966a069b242569c53e27 https://git.kernel.org/torvalds/c/0ce57f8af1782fd12d3a81872a4ab97244989802

Desc: 761ed4a94582 tty: serial_core: convert uart_close to use tty_port_close
Repo: 2016-10-21 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1254753.html
Fix:  https://git.kernel.org/torvalds/c/d0f4bce2bce7e998abc906f3590e9032af7a41ba

Desc: [selinux/audit/netlink, regression?] Warning at kernel/softirq.c:161
Repo: 2016-10-20 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1254148.html
Fix:  https://git.kernel.org/torvalds/c/2a73306b6096fafd5c2ae06ded1f92bbacb39df2

Re: Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Benjamin Herrenschmidt]

On Sun, 2016-10-30 at 14:20 +0100, Thorsten Leemhuis wrote:
> 
> Desc: PPC32: fails to boot on my PowerBook G4 Aluminum; bisected to
> commit 05fd007e4629
> Repo: 2016-10-20 https://www.mail-archive.com/linux-kernel@vger.kerne
> l.org/msg1253391.html
> Stat: 2016-10-22 https://www.mail-archive.com/linux-kernel@vger.kerne
> l.org/msg1255516.html https://www.linux-mips.org/archives/linux-mips/
> 2016-10/msg00176.html https://lkml.org/lkml/2016/10/18/142
> Note: Larry made a hack that works for him.

This breaks framebuffer console on ppc64 I've been told as well, I
heard...

I'm at KS now, hard to get you more details, but there's something
fishy here either with the commit or with something we do on ppc with
fbdev that this commit breaks.

Cheers,
Ben.

Re: Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Linus Torvalds]

On Sun, Oct 30, 2016 at 6:20 AM, Thorsten Leemhuis
<regressions@leemhuis.info> wrote:
>
> Desc: tpm0: TPM self test failed
> Repo: 2016-10-28 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1259943.html
> Stat: 2016-10-28 https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1260452.html https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1256705.html
> Note: Fix hopefully soon heading upstream
>
> Desc: "Nokia N900 (omap3-n900) with ""VDD1: ramp_delay not set"" string in printk frequently"
> Repo: 2016-10-19 https://bugzilla.kernel.org/show_bug.cgi?id=178371
> Stat: n/a https://lkml.org/lkml/2016/10/27/527
> Note: discussion on lkml ongoing

These two should hopefully be fixed as of today in current -git.

> Desc: radeon performance drop from 4.8 to 4.9-rc1 in Shadow of Mordor
> Repo: 2016-10-14 https://bugzilla.kernel.org/show_bug.cgi?id=178221 https://lists.freedesktop.org/archives/dri-devel/2016-October/120693.html
> Stat: 2016-10-20 https://lists.freedesktop.org/archives/dri-devel/2016-October/121427.html
> Note: Stuck? Poked bugzilla

I think this was fixed Friday by the merge of the PAT regression fixes
from DaveA (merge commit .bdb520845b57)

               Linus

Re: Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Andreas Schwab]

On Okt 30 2016, Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:

> On Sun, 2016-10-30 at 14:20 +0100, Thorsten Leemhuis wrote:
>> 
>> Desc: PPC32: fails to boot on my PowerBook G4 Aluminum; bisected to
>> commit 05fd007e4629
>> Repo: 2016-10-20 https://www.mail-archive.com/linux-kernel@vger.kerne
>> l.org/msg1253391.html
>> Stat: 2016-10-22 https://www.mail-archive.com/linux-kernel@vger.kerne
>> l.org/msg1255516.html https://www.linux-mips.org/archives/linux-mips/
>> 2016-10/msg00176.html https://lkml.org/lkml/2016/10/18/142
>> Note: Larry made a hack that works for him.
>
> This breaks framebuffer console on ppc64 I've been told as well, I
> heard...
>
> I'm at KS now, hard to get you more details, but there's something
> fishy here either with the commit or with something we do on ppc with
> fbdev that this commit breaks.

See the thread at <http://marc.info/?t=147678252000003&r=1&w=2>.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Paul Bolle]

On Sun, 2016-10-30 at 14:20 +0100, Thorsten Leemhuis wrote:
> As always: Are you aware of any other regressions? Then please let me
> know (simply CC regressions@leemhuis.info).

Do build regressions count?

Because I was trying to fix an obscure build issue in arch/mips, choose
a random configuration that should hit that issue, and promptly ran
into
https://lkml.kernel.org/r/<201610301405.K82KqqW0%25fengguang.wu@intel.com>
The same configuration does build under v4.8, I tested that of course.

(Side note: I had to manually insert "25" after "%" to get this to
work. Should Intel fix its mail setup, or should lkml.kernel.org learn
to escape "%"?)

Thanks,


Paul Bolle

Re: Linux 4.9: Reported regressions as of Sunday, 2016-10-30

[Thorsten Leemhuis]

Lo! On 01.11.2016 09:18, Paul Bolle wrote:
> On Sun, 2016-10-30 at 14:20 +0100, Thorsten Leemhuis wrote:
>> As always: Are you aware of any other regressions? Then please let me
>> know (simply CC regressions@leemhuis.info).
> Do build regressions count?

That's a good question.

> Because I was trying to fix an obscure build issue in arch/mips, choose
> a random configuration that should hit that issue, and promptly ran
> into
> https://lkml.kernel.org/r/<201610301405.K82KqqW0%25fengguang.wu@intel.com>
> The same configuration does build under v4.8, I tested that of course.

I'd say it's a practical problem that users run into and hence it's a
regression. Sure, in this case it hits only those that compile kernels
themselves; but those are users, too, and we don't want to scare them
away with things that suddenly stop working.

IOW: I'll include it in this weeks report.

Ciao, Thorsten