Me to. Cursor movements triggers this. Screen will flicker and even go black. Started right after the update. Revert to stable and all is fixed. Toshiba  Google_Swanky.5216.238.5

marcheu@ can you please take a look and reassign as necessary.

 Issue 711995  has been merged into this issue.

From a duplicate bug:

UserAgent: Mozilla/5.0 (X11; CrOS x86_64 9449.0.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3065.0 Safari/537.36
Platform: 9449.0.0 (Official Build) dev-channel swanky

Steps to reproduce the problem:
1. Update to latest Chrome Dev

What is the expected behavior?

What went wrong?
Screen flickers and system becomes unresponsive.

Did this work before? N/A 

Chrome version: 59.0.3065.0  Channel: dev
OS Version: 9449.0.0
Flash Version: 25.0.0.156 

Running dmesg in shell produces a flood of these:

[  251.923327] [drm:intel_psr_work] *ERROR* Timed out waiting for PSR Idle for re-enable

Also:
[    8.854150] byt-max98090 byt-max98090: ASoC: CPU DAI baytrail-pcm-audio not registered
[    8.854161] byt-max98090 byt-max98090: snd_soc_register_card failed -517

also:
   6.960479] [drm:intel_cpu_fifo_underrun_irq_handler] *ERROR* CPU pipe A FIFO underrun

also:
[    3.959402] frecon(311): Frecon using drm driver i915, version 1.6, date(20160411), desc(Intel Graphics)
 

I've got same issue on my personal Toshiba CB2. Device is almost unusable with this flickering so bumping to P1.

This is a regression from kernel 4.4 switch

I believe I had the same issue.  Screen flashing after update and contiued to become unresponsive.  Forced to change channel back to stable.

I am not sure if we ever tested (or enabled) PSR on BYT devices. Can we double-check and then disable PSR for BYT devices if that is the delta that is exposing this issue?

No, we did not test PSR on BYT devices.  afaik, Swanky was the only BYT with the panel that supports PSR but the kernel wasn't fully ready or validated to support PSR at the time.

I just got a swanky from Chromestop running kernel 4.4 (R58, 9334.52.0) and I do not see this issue on that unit. Does swanky have PSR and non-PSR SKU(s)?

Either case I think it makes sense to disable PSR on baytrail and I'm working on that.

Also issue doesn't reproduce with my latest image on ToT.

Re comment #11, Swanky had two panel options, a 1080p and a 768p, I suspect only the higher end 1080p had the (unused) PSR option.

[+Zach]

Ok, I was able to reproduce it on my system. (I had missed that the change to enable PSR was a fairly recent one - 90d7b8d9df07 "CHROMIUM: drm/i915: Enable PSR"). I have a CL that fixes it on my system:

https://chromium-review.googlesource.com/c/486964/

The drm code differentiates the graphics devices based on the PCI-ids, and on Baytrail devices that I had (Swanky, Candy, Kip), identifies the graphics device as a "Valleyview-M (mobile)" device. Other than the PCI-id we have on our baytrails, there are a bunch of other PCI device IDs that can also be recognized as Valleyview-M(s):

#define INTEL_VLV_M_IDS(info) \
        INTEL_VGA_DEVICE(0x0f30, info), \
        INTEL_VGA_DEVICE(0x0f31, info), \ <--------- This is on swanky!
        INTEL_VGA_DEVICE(0x0f32, info), \
        INTEL_VGA_DEVICE(0x0f33, info), \
        INTEL_VGA_DEVICE(0x0157, info)


#define VLV_FEATURES  \
        .gen = 7, .num_pipes = 2, \
        .need_gfx_hws = 1, .has_hotplug = 1, \
        .ring_mask = RENDER_RING | BSD_RING | BLT_RING, \
        .display_mmio_offset = VLV_DISPLAY_BASE, \
        GEN_DEFAULT_PIPEOFFSETS, \
        CURSOR_OFFSETS

static const struct intel_device_info intel_valleyview_m_info = {
        VLV_FEATURES,
        .is_valleyview = 1,
        .is_mobile = 1,
};

There's also a Valleyview-D device that is not a mobile CPU.

#define INTEL_VLV_D_IDS(info) \
        INTEL_VGA_DEVICE(0x0155, info)

static const struct intel_device_info intel_valleyview_d_info = {
        VLV_FEATURES,
        .is_valleyview = 1,
};

My patch disables PSR on all devices identified as valleyview (M/D). Please let me know if that does not sound right.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/db5ba98e3e91570cc0abd60408e02e7367787837

commit db5ba98e3e91570cc0abd60408e02e7367787837
Author: Rajat Jain <rajatja@google.com>
Date: Tue Apr 25 23:32:49 2017

CHROMIUM: drm/i915: DIsable PSR on baytrails / valleyview

We've never tested PSR on baytrail systems, so no need to enable it.

BUG= chromium:711522 
TEST=Verify screen does not flicker on swanky

Change-Id: Ib5b2ff16862e7ace462abfb66bf8791ade4e4d84
Reviewed-on: https://chromium-review.googlesource.com/486964
Commit-Ready: Rajat Jain <rajatja@chromium.org>
Tested-by: Rajat Jain <rajatja@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/db5ba98e3e91570cc0abd60408e02e7367787837/drivers/gpu/drm/i915/intel_psr.c

OK, the CL has been submitted, I'd now proceed to request merge for R58 and R59.

This bug requires manual review: Request affecting a post-stable build
Please contact the milestone owner if you have questions.
Owners: amineer@(Android), cmasso@(iOS), bhthompson@(ChromeOS), govind@(Desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

LGTM for 58, this is a simple change just disabling a feature.

Actually, I realized that the original commit that enabled PSR did not land in M58, so we need this fixed only in M59 and ToT.

https://chromium-review.googlesource.com/c/487830

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/e6dfb7f2137d27ab66436ecb256514cd4c41f558

commit e6dfb7f2137d27ab66436ecb256514cd4c41f558
Author: Rajat Jain <rajatja@google.com>
Date: Wed Apr 26 22:29:18 2017

CHROMIUM: drm/i915: DIsable PSR on baytrails / valleyview

We've never tested PSR on baytrail systems, so no need to enable it.

BUG= chromium:711522 
TEST=Verify screen does not flicker on swanky

Change-Id: Ib5b2ff16862e7ace462abfb66bf8791ade4e4d84
Reviewed-on: https://chromium-review.googlesource.com/486964
Commit-Ready: Rajat Jain <rajatja@chromium.org>
Tested-by: Rajat Jain <rajatja@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
(cherry picked from commit db5ba98e3e91570cc0abd60408e02e7367787837)
Reviewed-on: https://chromium-review.googlesource.com/487830
Commit-Queue: Rajat Jain <rajatja@chromium.org>
Trybot-Ready: Rajat Jain <rajatja@chromium.org>
Reviewed-by: Dmitry Torokhov <dtor@chromium.org>

[modify] https://crrev.com/e6dfb7f2137d27ab66436ecb256514cd4c41f558/drivers/gpu/drm/i915/intel_psr.c

Your change meets the bar and is auto-approved for M59. Please go ahead and merge the CL to branch 3071 manually. Please contact milestone owner if you have questions.
Owners: amineer@(Android), cmasso@(iOS), gkihumba@(ChromeOS), Abdul Syed@(Desktop)

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

I verified this with R59-9460.13.0 today on my swanky. 

Closing now. Please re-open if needed.

This issue has been approved for a merge. Please merge the fix to any appropriate branches as soon as possible!

If all merges have been completed, please remove any remaining Merge-Approved labels from this issue.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Uh, I don't know which release does the sheriffbot wants me to merge in. This has been merged into ToT and 59, which is all that is needed. Thanks.

This issue has been approved for a merge. Please merge the fix to any appropriate branches as soon as possible!

If all merges have been completed, please remove any remaining Merge-Approved labels from this issue.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot