Bug 111772 - [CI][DRMTIP] igt@* - dmesg-warn - Delta way too big!
Summary: [CI][DRMTIP] igt@* - dmesg-warn - Delta way too big!
Status: NEW
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: DRI git
Hardware: Other All
: low minor
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-09-23 11:02 UTC by Lakshmi
Modified: 2019-10-05 01:03 UTC (History)
3 users (show)

See Also:
i915 platform: CML
i915 features: power/suspend-resume


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Lakshmi 2019-09-23 11:02:02 UTC
https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_375/fi-cml-s/igt@kms_cursor_crc@pipe-c-cursor-suspend.html

<4> [134.028799] ------------[ cut here ]------------
<4> [134.028801] Delta way too big! 18446743950851936958 ts=18446744055928880550 write stamp = 105076943592
If you just came from a suspend/resume,
please switch to the trace global clock:
  echo global > /sys/kernel/debug/tracing/trace_clock
or add trace_clock=global to the kernel command line
<4> [134.028808] WARNING: CPU: 4 PID: 1207 at kernel/trace/ring_buffer.c:2810 rb_handle_timestamp.isra.30+0x6c/0x70
<4> [134.028809] Modules linked in: i2c_dev vgem snd_hda_codec_hdmi snd_hda_codec_generic i915 x86_pkg_temp_thermal coretemp cdc_ether crct10dif_pclmul usbnet snd_hda_intel mii crc32_pclmul snd_intel_nhlt snd_hda_codec ghash_clmulni_intel e1000e thunderbolt snd_hwdep snd_hda_core ptp pps_core snd_pcm prime_numbers
<4> [134.028820] CPU: 4 PID: 1207 Comm: kworker/u24:19 Tainted: G     U            5.3.0-g437bf3f53fa4-drmtip_375+ #1
<4> [134.028821] Hardware name: Intel Corporation CometLake Client Platform/CometLake S UDIMM (ERB/CRB), BIOS CMLSFWR1.R00.1263.D00.1906260926 06/26/2019
<4> [134.028823] Workqueue: events_unbound async_run_entry_fn
<4> [134.028825] RIP: 0010:rb_handle_timestamp.isra.30+0x6c/0x70
<4> [134.028827] Code: 48 8b 73 08 85 c0 48 8b 13 48 c7 c0 e8 fc 07 aa 49 c7 c0 a5 da 08 aa 4c 0f 44 c0 48 8b 0f 48 c7 c7 a8 fd 07 aa e8 b4 14 f2 ff <0f> 0b eb a8 41 54 89 f0 55 53 48 0f a3 47 10 bb 01 00 00 00 72 07
<4> [134.028828] RSP: 0000:ffff9ec840a17cd0 EFLAGS: 00010282
<4> [134.028829] RAX: 0000000000000000 RBX: ffff9ec840a17ce8 RCX: 0000000000000002
<4> [134.028830] RDX: 0000000080000002 RSI: ffff8bbb515158f8 RDI: 00000000ffffffff
<4> [134.028831] RBP: ffff8bbb89b41b78 R08: ffff8bbb515158f8 R09: 0000000000000000
<4> [134.028832] R10: ffff9ec840a17ce0 R11: 0000000000000000 R12: ffff8bbb89888358
<4> [134.028833] R13: ffff8bbb89b41bb8 R14: 00000000000003e8 R15: 0000000000000000
<4> [134.028834] FS:  0000000000000000(0000) GS:ffff8bbb8b800000(0000) knlGS:0000000000000000
<4> [134.028835] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [134.028836] CR2: 0000000000000000 CR3: 00000001bd210001 CR4: 00000000003606e0
<4> [134.028837] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
<4> [134.028838] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
<4> [134.028839] Call Trace:
<4> [134.028842]  ring_buffer_lock_reserve+0x217/0x420
<4> [134.028868]  ? intel_gt_pm_init_early+0x50/0x50 [i915]
<4> [134.028870]  __trace_bputs+0x5b/0x120
<4> [134.028896]  intel_gt_sanitize+0x1c/0x90 [i915]
<4> [134.028916]  i915_pm_resume_early+0xc1/0xf0 [i915]
<4> [134.028936]  ? vlv_resume_prepare+0x690/0x690 [i915]
<4> [134.028939]  dpm_run_callback+0x64/0x280
<4> [134.028942]  device_resume_early+0xa7/0xe0
<4> [134.028945]  async_resume_early+0x14/0x40
<4> [134.028946]  async_run_entry_fn+0x34/0x160
<4> [134.028949]  process_one_work+0x245/0x610
<4> [134.028954]  worker_thread+0x37/0x380
<4> [134.028957]  ? process_one_work+0x610/0x610
<4> [134.028958]  kthread+0x119/0x130
<4> [134.028960]  ? kthread_park+0xa0/0xa0
<4> [134.028963]  ret_from_fork+0x24/0x50
<4> [134.028969] irq event stamp: 972
<4> [134.028971] hardirqs last  enabled at (971): [<ffffffffa9133455>] vprintk_emit+0x315/0x340
<4> [134.028973] hardirqs last disabled at (972): [<ffffffffa9001a8a>] trace_hardirqs_off_thunk+0x1a/0x20
<4> [134.028975] softirqs last  enabled at (916): [<ffffffffa9c0033e>] __do_softirq+0x33e/0x4ae
<4> [134.028976] softirqs last disabled at (909): [<ffffffffa90bcf29>] irq_exit+0xa9/0xc0
<4> [134.028978] WARNING: CPU: 4 PID: 1207 at kernel/trace/ring_buffer.c:2810 rb_handle_timestamp.isra.30+0x6c/0x70
<4> [134.028979] ---[ end trace aecf1495519c4077 ]---
Comment 3 CI Bug Log 2019-09-24 09:23:45 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_375/fi-cml-s/igt@runner@aborted.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_375/fi-cml-s/igt@runner@aborted.html
Comment 4 CI Bug Log 2019-09-24 09:27:34 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@* - dmesg-warn -  Delta way too big! -}
{+ fi-cml-s: igt@* - dmesg-warn -  Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_377/fi-cml-s/igt@kms_vblank@pipe-a-ts-continuation-dpms-suspend.html
Comment 5 CI Bug Log 2019-09-24 09:28:55 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_377/fi-cml-s/igt@runner@aborted.html
Comment 6 CI Bug Log 2019-09-30 09:16:13 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@* - dmesg-warn -  Delta way too big! -}
{+ fi-cml-s: igt@* - dmesg-warn -  Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@kms_fbcon_fbt@psr-suspend.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@gem_ctx_isolation@rcs0-s3.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@i915_pm_rpm@system-suspend-execbuf.html
Comment 7 CI Bug Log 2019-09-30 09:17:51 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@runner@aborted.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@runner@aborted.html
  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_379/fi-cml-s/igt@runner@aborted.html
Comment 8 CI Bug Log 2019-09-30 10:01:43 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@* - dmesg-warn -  Delta way too big! -}
{+ fi-cml-s: igt@* - dmesg-warn -  Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_380/fi-cml-s/igt@kms_frontbuffer_tracking@fbcpsr-suspend.html
Comment 9 CI Bug Log 2019-09-30 10:36:25 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_380/fi-cml-s/igt@runner@aborted.html
Comment 10 Matt Roper 2019-10-02 20:11:56 UTC
All of the warnings here seem to arise following suspend-resume, which is a known system clock issue on some platforms.  The warning message itself suggests a rememdy:

    If you just came from a suspend/resume,
    please switch to the trace global clock:
      echo global > /sys/kernel/debug/tracing/trace_clock
    or add trace_clock=global to the kernel command line

This isn't related to the graphics driver and is a platform-specific quirk, so we should probably take the warning's advice and update the boot command line for fi-cml-s to include the "trace_clock=global" to avoid hitting this.
Comment 11 Lakshmi 2019-10-03 06:06:55 UTC
(In reply to Matt Roper from comment #10)
> All of the warnings here seem to arise following suspend-resume, which is a
> known system clock issue on some platforms.  The warning message itself
> suggests a rememdy:
> 
>     If you just came from a suspend/resume,
>     please switch to the trace global clock:
>       echo global > /sys/kernel/debug/tracing/trace_clock
>     or add trace_clock=global to the kernel command line
> 
> This isn't related to the graphics driver and is a platform-specific quirk,
> so we should probably take the warning's advice and update the boot command
> line for fi-cml-s to include the "trace_clock=global" to avoid hitting this.

@Matt, How about Bug 111883? Above comment is valid for bug 111883 as well?

@Tomi, Need your help in fixing this bug.
Comment 12 Tomi Sarvela 2019-10-03 07:07:14 UTC
Added trace_clock=global WA for fi-cml-s. Will show up in run CI_DRM_6995 and later
Comment 13 CI Bug Log 2019-10-03 11:39:00 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@* - dmesg-warn -  Delta way too big! -}
{+ fi-cml-s: igt@* - dmesg-warn -  Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_381/fi-cml-s/igt@kms_frontbuffer_tracking@fbc-suspend.html
Comment 14 CI Bug Log 2019-10-03 11:44:53 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_381/fi-cml-s/igt@runner@aborted.html
Comment 15 CI Bug Log 2019-10-03 11:47:25 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@* - dmesg-warn -  Delta way too big! -}
{+ fi-cml-s: igt@* - dmesg-warn -  Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_381/fi-cml-s/igt@kms_fbcon_fbt@fbc-suspend.html
Comment 16 CI Bug Log 2019-10-03 11:49:55 UTC
A CI Bug Log filter associated to this bug has been updated:

{- fi-cml-s: igt@runner@aborted - fail - Delta way too big! -}
{+ fi-cml-s: igt@runner@aborted - fail - Delta way too big! +}

New failures caught by the filter:

  * https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_381/fi-cml-s/igt@runner@aborted.html
Comment 17 Matt Roper 2019-10-05 01:03:17 UTC
(In reply to Lakshmi from comment #11)
> (In reply to Matt Roper from comment #10)
> > All of the warnings here seem to arise following suspend-resume, which is a
> > known system clock issue on some platforms.  The warning message itself
> > suggests a rememdy:
> > 
> >     If you just came from a suspend/resume,
> >     please switch to the trace global clock:
> >       echo global > /sys/kernel/debug/tracing/trace_clock
> >     or add trace_clock=global to the kernel command line
> > 
> > This isn't related to the graphics driver and is a platform-specific quirk,
> > so we should probably take the warning's advice and update the boot command
> > line for fi-cml-s to include the "trace_clock=global" to avoid hitting this.
> 
> @Matt, How about Bug 111883? Above comment is valid for bug 111883 as well?
> 
> @Tomi, Need your help in fixing this bug.

Yes, thanks for pointing that one out.  It looks like the same platform clock problem is present on that snb platform.  I can't verify for sure that that machine doesn't already have the trace_clock option on the kernel command line since the boot.txt link over there is broken, but assuming it doesn't we should also add it to that machine.


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.