[29322:29325:0331/175025.857174:196052064503:VERBOSE2:drm_device.cc(407)] Cannot query for resources for '/sys/devices/platform/vgem/drm/card1'
[29322:29325:0331/175025.857223:196052064550:VERBOSE1:drm_device_manager.cc(54)] Could not initialize DRM device for /sys/devices/platform/vgem/drm/card1
[29322:29322:0331/175025.857362:196052064690:VERBOSE1:display_configurator.cc(951)] Display snapshots invalidated.
[29322:29322:0331/175025.857493:196052064820:VERBOSE1:update_display_configuration_task.cc(74)] OnDisplaysUpdated: new_display_state=SINGLE new_power_state=ALL_ON flags=1 force_configure=1 display_count=1
[29322:29322:0331/175025.857518:196052064843:VERBOSE2:update_display_configuration_task.cc(97)] EnterState
[29322:29322:0331/175025.857541:196052064867:VERBOSE1:display_configurator.cc(217)] EnterState: display=SINGLE power=ALL_ON
[29322:29325:0331/175025.857733:196052065061:VERBOSE1:drm_display.cc(102)] DRM configuring: device=/sys/devices/pci0000:00/0000:00:02.0/drm/card0 crtc=20 connector=29 origin=0,0 size=2560x1700
[29322:29325:0331/175025.857921:196052065248:VERBOSE2:screen_manager.cc(44)] Crtc has no saved state or wasn't modeset
[29322:29322:0331/175026.315303:196052522637:VERBOSE2:configure_displays_task.cc(96)] Configured status=1 display=13761487533244416 origin=0,0 mode=[2560x1700 rate=60.004852]
[29322:29322:0331/175026.315383:196052522710:VERBOSE1:display_configurator.cc(1061)] OnConfigured: success=1 new_display_state=SINGLE new_power_state=ALL_ON
[29322:29323:0331/175026.315938:196052523268:VERBOSE1:gl_context_egl.cc(106)] EGL_EXT_create_context_robustness supported.
[29322:29325:0331/175026.568199:196052775531:ERROR:hardware_display_plane_manager.cc(249)] Failed to find a free plane for crtc 20
[29322:29325:0331/175026.568284:196052775611:ERROR:crtc_controller.cc(102)] Failed to assign overlay planes for crtc 20: Invalid argument

This is on samus.

Similar fail messages on Auron_yuna as #2:
[2609:2612:0331/163855.301711:256402308:ERROR:hardware_display_plane_manager.cc(249)] Failed to find a free plane for crtc 20
[2609:2612:0331/163855.301758:256402348:ERROR:crtc_controller.cc(102)] Failed to assign overlay planes for crtc 20: Invalid argument

It is hanging in RenderingHelper::Initialize()
https://cs.chromium.org/chromium/src/media/gpu/rendering_helper.cc?type=cs&q=rendering_help&l=518
vsync_provider->GetVSyncParameters(base::Bind(
        &RenderingHelper::UpdateVSyncParameters, base::Unretained(this), done));

whuch |done| is not signaled.

Hi Daniele,

By bisect, I found the your CL is the first change that make VDA hang
https://chromium.googlesource.com/chromium/src/+/9d7a8aef95959ab06d2b565fe129fd27ed42b963
on many boards
https://wmatrix.googleplex.com/unfiltered?hide_missing=True&releases=tot&tests=video_VideoDecodeAccelerator&days_back=30
(My bisect ran on samus)

This is my faster command line to repro:
/usr/local/autotest/deps/chrome_test/test_src/out/Release/video_decode_accelerator_unittest \
  --test_video_data="/usr/local/autotest/deps/chrome_test/test_src/media/test/data/test-25fps.h264:320:240:250:258:35:150:1" \
  --ozone-platform=gbm \
  --gtest_filter=ResourceExhaustion/VideoDecodeAcceleratorParamTest.TestSimpleDecode/1


p.s. detail for repro & bisect:
In range 002b2ee07c58..b720a94625b2, need to cherry-pick b720a94625b2 due to another regression.

That CL introduced a leak that has been addressed with crrev.com/2790553002

Can you reproduce the problem after that CL?

1. Yes, I cherry-picked crrev.com/2790553002 for the necessary range and the bisect result is identical.
2. ToT can still reproduce this issue.

dcastagna, ping

Thank you for the command line repro. It was really helpful.

The problem seems to be that samus doesn't support planes with alpha and we're trying to schedule one (for the primary plane).
Before my change we would always remove the alpha from the format.

crrev.com/2825933002 is out for review.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5d308f5270f55633e074bdaa2b2e06d303057d53

commit 5d308f5270f55633e074bdaa2b2e06d303057d53
Author: dcastagna <dcastagna@chromium.org>
Date: Wed Apr 19 16:58:38 2017

ozone: Create RGBX (not RGBA) buffers in GbmSurface.

We used to always create opaque DRM framebuffers.
After crrev.com/2790553002 we started allowing non-opaque fb.

GbmSurface, that was used in some tests, would allocate a BGRA
buffer and use it as primary plane. On some devices formats
with alpha are not supported on primary plane.

This CL makes sure that GbmSurface allocates RGBX buffers.

BUG= 707108 
TEST=video_decode_accelerator_unittest now passes.

Review-Url: https://codereview.chromium.org/2825933002
Cr-Commit-Position: refs/heads/master@{#465640}

[modify] https://crrev.com/5d308f5270f55633e074bdaa2b2e06d303057d53/ui/ozone/platform/drm/gpu/gbm_surface.cc

Do we need to merge this to 59? crrev.com/2790553002 seems to have gotten into 59, but the fix above not?

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

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/7d1c67cbf9a27757cc682787d6559c1c60447513

commit 7d1c67cbf9a27757cc682787d6559c1c60447513
Author: Aaron Gable <agable@chromium.org>
Date: Thu Apr 27 21:43:57 2017

ozone: Create RGBX (not RGBA) buffers in GbmSurface.

We used to always create opaque DRM framebuffers.
After crrev.com/2790553002 we started allowing non-opaque fb.

GbmSurface, that was used in some tests, would allocate a BGRA
buffer and use it as primary plane. On some devices formats
with alpha are not supported on primary plane.

This CL makes sure that GbmSurface allocates RGBX buffers.

BUG= 707108 
TEST=video_decode_accelerator_unittest now passes.

Review-Url: https://codereview.chromium.org/2825933002
Cr-Commit-Position: refs/heads/master@{#465640}
(cherry picked from commit 5d308f5270f55633e074bdaa2b2e06d303057d53)

TBR=dcastagna@chromium.org

Change-Id: I8643150de8afe776f8e11dae14d028c2e02abdb8
Reviewed-on: https://chromium-review.googlesource.com/489490
Reviewed-by: Aaron Gable <agable@chromium.org>
Cr-Commit-Position: refs/branch-heads/3071@{#270}
Cr-Branched-From: a106f0abbf69dad349d4aaf4bcc4f5d376dd2377-refs/heads/master@{#464641}
[modify] https://crrev.com/7d1c67cbf9a27757cc682787d6559c1c60447513/ui/ozone/platform/drm/gpu/gbm_surface.cc

closing as per https://wmatrix.googleplex.com/unfiltered?hide_missing=True&tests=video_VideoDecodeAccelerator&builds=R59-9460.57.0