Re: suspend regression in 4.1-rc1

[Omar Sandoval <osandov@osandov.com>]

On Sun, May 17, 2015 at 08:50:41PM +0200, Michal Hocko wrote:
> Hi,
> s2ram broke after 4.1-rc1 for me. The second s2ram simply doesn't wake
> up (fans turn on but the screen is off). I have even noticed fans
> starting also while suspended in some instances (which was especially
> annoying when it happened on the way home from work).
> I've tried /sys/power/pm_test and the issue starts at processors mode.
> Nothing really interesting shows up in the netconsole but I didn't get
> to a more detailed testing there.
> 
> I've tried to bisect this as 4.0 works reliably. This was tricky though
> because the first bad commit is a merge:
> 
> commit 1dcf58d6e6e6eb7ec10e9abc56887b040205b06f
> Merge: 80dcc31fbe55 e4b0db72be24
> Author: Linus Torvalds <torvalds@linux-foundation.org>
> Date:   Tue Apr 14 16:49:17 2015 -0700
> 
>     Merge branch 'akpm' (patches from Andrew)
> 
> The merge commit is empty and both 80dcc31fbe55 and e4b0db72be24 work
> properly but the merge is bad. So it seems like some of the commits in
> either branch has a side effect which needs other branch in order to
> reproduce.
> 
> So've tried to bisect ^80dcc31fbe55 e4b0db72be24 and merged 80dcc31fbe55
> in each step. This lead to:
> 
> commit 195daf665a6299de98a4da3843fed2dd9de19d3a
> Author: Ulrich Obergfell <uobergfe@redhat.com>
> Date:   Tue Apr 14 15:44:13 2015 -0700
> 
>     watchdog: enable the new user interface of the watchdog mechanism
> 
> The patch doesn't revert because of follow up changes so I have reverted
> all three:
> 692297d8f968 ("watchdog: introduce the hardlockup_detector_disable() function")
> b2f57c3a0df9 ("watchdog: clean up some function names and arguments")
> 195daf665a62 ("watchdog: enable the new user interface of the watchdog mechanism")
> 
> on top of my current Linus tree (4cfceaf0c087f47033f5e61a801f4136d6fb68c6)
> and the issue is gone. I have hard time to understand what these 3 could have
> to do with suspend path, though.
> 
> Then I've tried to bisect the other branch and merge 195daf665a62 during
> each step to find out which patch starts failing. This lead to an even
> weirder commit a1e12da4796a ("perf tools: Add 'I' event modifier for
> exclude_idle bit") but maybe I've just screwed something on the way.
> 
> I will continue debugging tomorrow but any hints would be helpful.
> -- 
> Michal Hocko
> SUSE Labs

The symptoms here are different, and the bisect indicates that it's
completely unrelated, but just for kicks you could also try:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=533445c6e53368569e50ab3fb712230c03d523f3
-- 
Omar