Linux 4.8: Reported regressions as of Sunday, 2016-09-18

Linux 4.8: Reported regressions as of Sunday, 2016-09-18

[Thorsten Leemhuis]

Hi! Here is my fourth regression report for Linux 4.8. It lists 14
regressions I'm aware of. 5 of them are new; 1 mentioned in last 
weeks report got fixed.

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

Ciao, Thorsten

== Current regressions ==

Desc: genirq: Flags mismatch irq 8, 00000088 (mmc0) vs. 00000080 (rtc0). mmc0: Failed to request irq 8: -16
Repo: 2016-08-01 https://bugzilla.kernel.org/show_bug.cgi?id=150881
Stat: 2016-09-09 https://bugzilla.kernel.org/show_bug.cgi?id=150881#c34
Note: stalled; root cause somewhere in the main gpio merge for 4.8, but problematic commit still unknown

Desc: [lkp] [xfs] 68a9f5e700: aim7.jobs-per-min -13.6% regression
Repo: 2016-08-09 http://www.spinics.net/lists/kernel/msg2317052.html
Stat: 2016-09-09 https://marc.info/?t=147341519500003&r=1&w=2
Note: looks like post-4.8 material at this point: Mel working on it in his spare time, but "The progression of this series has been unsatisfactory."

Desc: scsi host6: runtime PM trying to activate child device host6 but parent (2-2:1.0) is not active
Repo: 2016-08-15 https://bugzilla.kernel.org/show_bug.cgi?id=153171
Stat: 2016-09-14 https://bugzilla.kernel.org/show_bug.cgi?id=153171#c5
Note: patch available; mkp "I would like the change to get a bit of soak time before we backport it."

Desc: DT/OCTEON driver probing broken
Repo: 2016-08-16 http://www.spinics.net/lists/devicetree/msg138990.html
Stat: 2016-08-30 http://www.spinics.net/lists/devicetree/msg140682.html
Note: stalled, poked Rob a week ago, got no reply

Desc: gpio-leds broken on OCTEON
Repo: 2016-08-23 http://www.spinics.net/lists/devicetree/msg139863.html
Stat: 2016-09-12 http://www.spinics.net/lists/devicetree/msg140179.html
Note: patch in Linux MIPS patchwork https://patchwork.linux-mips.org/patch/14091/

Desc: Skylake graphics regression: projector failure with 4.8-rc3
Repo: 2016-08-26 http://www.spinics.net/lists/intel-gfx/msg105478.html
Stat: 2016-09-01 https://lkml.org/lkml/2016/8/31/946
Note: stalled, poked lists

Desc: lk 4.8 + !CONFIG_SHMEM + shmat() = oops
Repo: 2016-08-30 http://www.spinics.net/lists/linux-mm/msg112920.html
Stat: 2016-09-07 http://www.spinics.net/lists/linux-mm/msg113177.html
Note: just like last week: patch "ipc/shm: fix crash if CONFIG_SHMEM is not set" is going to fix this, but has not hit mainline yet

Desc: regression in re-read operation by iozone ~10%
Repo: 2016-09-02 https://bugzilla.kernel.org/show_bug.cgi?id=155821
Stat: n/a 
Note: stalled; told reporter he might be better of posting about the issue to some mailing list

Desc: brcmfmac is preventing suspend (Dell XPS 13 9350 / Skylake)/
Repo: 2016-09-13 https://bugzilla.kernel.org/show_bug.cgi?id=156631
Stat: n/a 
Note: used to be https://bugzilla.kernel.org/show_bug.cgi?id=156361 in last weeks report

Desc: CPU speed set very low
Repo: 2016-09-09 https://lkml.org/lkml/2016/9/9/608
Stat: 2016-09-14 https://lkml.org/lkml/2016/9/14/588
Note: Larry: "Testing continues."

Desc: pinctrl: qcom: Clear all function selection bits introduced a regression by not properly masking the calculated value.
Repo: 2016-09-12 http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1229055.html
Stat: n/a 
Note: Patch available in initial report

Desc: WARN in vb2_dma_sg_alloc when makes dvbstream fail with with AverMedia Hybrid+FM
Repo: 2016-09-13 https://bugzilla.kernel.org/show_bug.cgi?id=156751
Stat: n/a 
Note: told reporter he might be better of posting the issue to linux-media

Desc: it is now common that machine needs re-run of xrandr after resume
Repo: 2016-09-13 http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1230782.html
Stat: 2016-09-15 http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1232834.html http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1231104.html
Note: Seems Martin isn't seeing the problem anymore; Pavel's issue might be a different problem anyway; and he now has a userland-problem that complicates testing

Desc: usb: gadget: udc: atmel: fix endpoint name
Repo: 2016-09-15 http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg1232309.html http://www.mail-archive.com/linux-usb@vger.kernel.org/msg80645.html 
Stat: n/a 
Note: Patch available: "Felipe, Greg,It is clearly a regression and material for 4.8-fixes."

== Fixed since last report ==

Desc: ath9k: bring back direction setting in ath9k_{start_stop}
Repo: 2016-09-01 https://marc.info/?l=linux-wireless&m=147292415030585&w=2
Fix:  https://git.kernel.org/torvalds/c/e34f2ff40e0339f6a379e1ecf49e8f2759056453

Re: Linux 4.8: Reported regressions as of Sunday, 2016-09-18

[Dave Chinner]

On Sun, Sep 18, 2016 at 03:20:53PM +0200, Thorsten Leemhuis wrote:
> Hi! Here is my fourth regression report for Linux 4.8. It lists 14
> regressions I'm aware of. 5 of them are new; 1 mentioned in last 
> weeks report got fixed.
> 
> As always: Are you aware of any other regressions? Then please let me
> know (simply CC regressions@leemhuis.info). And pls tell me if there
> is anything in the report that shouldn't be there.
> 
> Ciao, Thorsten
> 
> == Current regressions ==
> 
> Desc: genirq: Flags mismatch irq 8, 00000088 (mmc0) vs. 00000080 (rtc0). mmc0: Failed to request irq 8: -16
> Repo: 2016-08-01 https://bugzilla.kernel.org/show_bug.cgi?id=150881
> Stat: 2016-09-09 https://bugzilla.kernel.org/show_bug.cgi?id=150881#c34
> Note: stalled; root cause somewhere in the main gpio merge for 4.8, but problematic commit still unknown
> 
> Desc: [lkp] [xfs] 68a9f5e700: aim7.jobs-per-min -13.6% regression
> Repo: 2016-08-09 http://www.spinics.net/lists/kernel/msg2317052.html
> Stat: 2016-09-09 https://marc.info/?t=147341519500003&r=1&w=2
> Note: looks like post-4.8 material at this point: Mel working on it in his spare time, but "The progression of this series has been unsatisfactory."

Actually, what Mel was working on (mapping lock contention) was not
related to the reported XFS regression. The regression was an XFS
sub-page write issue introduced by the new iomap infrastructure,
and nobody has been able to reproduce it exactly
outside of the reaim benchmark. We've reproduced other, similar
issues, and the fixes for those are queued for the 4.9 window.

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com