Bug 100651

Summary:	[BAT][BSW] CI run hit a scheduler warning (rq->clock_update_flags < RQCF_ACT_SKIP)
Product:	DRI	Reporter:	Martin Peres <martin.peres>
Component:	DRM/Intel	Assignee:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Status:	CLOSED NOTOURBUG	QA Contact:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity:	critical
Priority:	highest	CC:	intel-gfx-bugs
Version:	DRI git
Hardware:	Other
OS:	All
Whiteboard:	ReadyForDev
i915 platform:	BSW/CHT	i915 features:	power/suspend-resume

Description Martin Peres 2017-04-11 14:03:13 UTC

Here is the warning in question:

[  568.160295] ------------[ cut here ]------------
[  568.160327] WARNING: CPU: 1 PID: 0 at kernel/sched/sched.h:833 pick_next_task_rt+0x250/0x260
[  568.160335] rq->clock_update_flags < RQCF_ACT_SKIP
[  568.160343] Modules linked in: vgem snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel r8169 mii lpc_ich i2c_hid i915 snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core snd_pcm prime_numbers i2c_designware_platform i2c_designware_core
[  568.160529] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G     U  W       4.11.0-rc6-CI-CI_DRM_2509+ #1
[  568.160540] Hardware name:                  /NUC5CPYB, BIOS PYBSWCEL.86A.0058.2016.1102.1842 11/02/2016
[  568.160548] Call Trace:
[  568.160571]  dump_stack+0x67/0x92
[  568.160591]  __warn+0xc6/0xe0
[  568.160611]  warn_slowpath_fmt+0x46/0x50
[  568.160637]  pick_next_task_rt+0x250/0x260
[  568.160658]  __schedule+0x160/0xbc0
[  568.160673]  ? ktime_get+0x89/0x120
[  568.160696]  schedule+0x3b/0x90
[  568.160712]  schedule_preempt_disabled+0x13/0x20
[  568.160726]  do_idle+0x145/0x1f0
[  568.160745]  cpu_startup_entry+0x18/0x20
[  568.160760]  start_secondary+0x102/0x120
[  568.160778]  start_cpu+0x14/0x14
[  568.160810] ---[ end trace eb40e71579767b8a ]---

Full logs: https://intel-gfx-ci.01.org/CI/CI_DRM_2509/fi-bsw-n3050/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-c.html

Comment 1 Ricardo 2017-05-09 17:00:09 UTC

Adding tag into "Whiteboard" field - ReadyForDev
The bug still active
*Status is correct
*Platform is included
*Feature is included
*Priority and Severity correctly set

Comment 2 Maarten Lankhorst 2017-05-10 08:41:26 UTC

Seems to be a bug in the kernel, not i915..

https://patchwork.kernel.org/patch/9586351/

Comment 3 Martin Peres 2017-05-10 16:38:59 UTC

(In reply to Maarten Lankhorst from comment #2)
> Seems to be a bug in the kernel, not i915..
> 
> https://patchwork.kernel.org/patch/9586351/

Yes, it is. I filed it here because that's what we agreed on before, that we should file the bugs with i915 first since it impacts our testing, and then move it to the right owner. Of course, we never got to do that and I hope that we will soon have the man power to handle these bugs appropriately...

Anyway, this is becoming funnier. The patch you mentioned made it to drm-tip but 2 days ago, we got this result: https://intel-gfx-ci.01.org/CI/CI_DRM_2590/fi-bsw-n3050/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-c.html

Comment 4 Jani Saarinen 2017-06-02 08:12:38 UTC

Martin, what are next steps here?

Comment 5 Martin Peres 2017-06-02 10:16:13 UTC

I guess we really need to get someone to communicate with upstream so as they could propose a patch for it.

Comment 6 Martin Peres 2017-06-15 11:15:13 UTC

(In reply to Martin Peres from comment #5)
> I guess we really need to get someone to communicate with upstream so as
> they could propose a patch for it.

I don't see this happening anymore. Let's whitelist this test.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.