[Bug 98289] [BDW] kms_flip / vblank-vs-suspend-interruptible is failing

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=98289

--- Comment #9 from Tomeu Vizoso <tomeu at tomeuvizoso.net> ---
(In reply to Tomeu Vizoso from comment #7)
> Running this test on an APL machine here, I see this in dmesg:
> 
> [10968.960490] [drm:drm_handle_vblank] vblank event on 58648, current 58648
> [10968.960502] [drm:vblank_disable_fn] disabling vblank on crtc 0
> [10971.312512] [drm:edp_panel_vdd_off_sync] Turning eDP port A VDD off
> [10971.312529] [drm:edp_panel_vdd_off_sync] PP_STATUS: 0x80000008
> PP_CONTROL: 0x00000067
> [11031.602616] perf: interrupt took too long (3200 > 3178), lowering
> kernel.perf_event_max_sample_rate to 62000
> [11121.589837] perf: interrupt took too long (4038 > 4000), lowering
> kernel.perf_event_max_sample_rate to 49000
> [11211.908147] PM: Syncing filesystems ... done.
> ...
> [11212.918426] [drm:drm_handle_vblank] vblank event on 73245, current 73246
> 
> 
> So after getting the vblank 58648 and right before suspending there's a big
> delay of 244 seconds (11212 − 10968) that is accompanied of complaints by
> perf.
> 
> After resuming, we get a vblank of 73245 which roughly matches the previous
> vblank count plus 244 * 60.

The delay seems to be caused by receiving RTC interrupts at a rate of 8kHz from
the moment the alarm is programmed.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20170208/1ebd0106/attachment.html>