Bug 109221 - GM965 plays Chrome/Chromium videos slowly on VGA head
Summary: GM965 plays Chrome/Chromium videos slowly on VGA head
Status: RESOLVED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard: Triaged
Keywords:
Depends on:
Blocks:
 
Reported: 2019-01-04 05:07 UTC by Ryan Underwood
Modified: 2019-10-02 19:01 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments

Description Ryan Underwood 2019-01-04 05:07:23 UTC
I do not know the technical details but can reproduce on demand.

My displays are configured with 1280x800 on the eDP panel and 1280x1024 on the VGA head.

This has been the case for all 4.x kernel versions and at least three recent Ubuntu distributions.

In Chrome/Chromium (all versions recent and latest), YouTube and videos from similar video sites (it really doesn't matter which) drop frames significantly.  

Mitigation: If I configure the VGA head to 1024x768, the Chrome video frame drops are reduced by about 1/2.  If I configure the VGA head to 800x600, any Chrome video frame drops are then imperceptible.

I had long accepted this as a hardware limitation.

However! I just noticed today that if I use Firefox in the exact same configuration, there are no frame drops and completely smooth video on the same websites in the same 1280x800/1280x1024 CRTC configurations.

Clearly Chrome/Chromium is doing something different than Firefox, but I have no idea what it could be.  Could I help determine what it is?
Comment 1 Ryan Underwood 2019-01-04 05:11:39 UTC
For the sake of clarity, there are also no perceptible frame drops in Chrome/Chromium on the LVDS head (sorry, I said eDP earlier - this is an older laptop).  The frame drops only occur 1) in Chrome/Chromium, 2) on the VGA head, and 3) only when the VGA head CRTC is configured larger than 800x600.
Comment 2 Lakshmi 2019-07-03 08:31:33 UTC
Ryan, My apologies for responding on this issue so late.
Can you please verify this issue on drmtip (https://cgit.freedesktop.org/drm-tip) and kernel parameters drm.debug=0x1e log_buf_len=4M, and if the problem persists attach the full dmesg from boot.
Comment 3 Lakshmi 2019-07-29 12:06:39 UTC
Ryan, Do you still have the issue with drmtip?
Comment 4 Lakshmi 2019-08-28 10:27:05 UTC
Ryan, any updates here?
Comment 5 Lakshmi 2019-10-02 19:01:58 UTC
No feedback from more than 2 months. I assume this issue has been fixed, resolving this issue as WORKSFORME.


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.