Bug 90212 - [BDW-Y bisected] Screen flicker when start X with FBC enable
Summary: [BDW-Y bisected] Screen flicker when start X with FBC enable
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: medium normal
Assignee: Paulo Zanoni
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-04-28 09:22 UTC by ye.tian
Modified: 2017-07-24 22:47 UTC (History)
3 users (show)

See Also:
i915 platform: BDW
i915 features: display/FBC


Attachments
dmesg info (119.85 KB, text/plain)
2015-04-28 09:24 UTC, ye.tian
no flags Details
Xorg.0.log info (15.27 KB, text/plain)
2015-04-28 09:24 UTC, ye.tian
no flags Details

Description ye.tian 2015-04-28 09:22:19 UTC
==System Environment==       
-----------------------------------------------------
Regression: Yes 
Bad  2015_04_28/drm-intel-next-queued/de4de566f84b87f4d5f8265b11932088adb19896
Good 2015_01_30/drm-intel-next-queued/b8d24a06568368076ebd5a858a011699a97bfa42
Non-working platforms: BDW-Y  Lenovo Yoga 3 pro-1370

==Kernel==
--------------------------------------------------
commit de4de566f84b87f4d5f8265b11932088adb19896
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Apr 23 22:02:54 2015 +0200

    drm/i915: Update DRIVER_DATE to 20150423

    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

==Bug detailed description==
--------------------------------------------------
Screen flicker when start X with FBC enable, next-queued and nightly branch is bad, fixes branch is good.
==Reproduce steps==
----------------------------
1, boot machine
2, start X
Comment 1 ye.tian 2015-04-28 09:24:24 UTC
Created attachment 115395 [details]
dmesg info
Comment 2 ye.tian 2015-04-28 09:24:42 UTC
Created attachment 115396 [details]
Xorg.0.log info
Comment 3 Ander Conselvan de Oliveira 2015-04-28 10:59:34 UTC
Could you double check the bisect? That commit only changes DRIVER_DATE, which shouldn't really change the driver behavior.
Comment 4 ye.tian 2015-04-29 02:54:24 UTC
Bisect shows, the first bad commit is dbef0f1.

commit dbef0f15b5c83231dacb214dbf9a6dba063ca21c
Author:     Paulo Zanoni <paulo.r.zanoni@intel.com>
AuthorDate: Fri Feb 13 17:23:46 2015 -0200
Commit:     Daniel Vetter <daniel.vetter@ffwll.ch>
CommitDate: Tue Mar 17 22:29:56 2015 +0100

    drm/i915: add frontbuffer tracking to FBC

    Kill the blt/render tracking we currently have and use the frontbuffer
    tracking infrastructure.

    Don't enable things by default yet.

    v2: (Rodrigo) Fix small conflict on rebase and typo at subject.
    v3: (Paulo) Rebase on RENDER_CS change.
    v4: (Paulo) Rebase.
    v5: (Paulo) Simplify: flushes don't have origin (Daniel).
                Also rebase due to patch order changes.

    Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
    Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
    Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Comment 5 Paulo Zanoni 2015-07-16 20:38:57 UTC
If you're using one of the SDVs, please boot the machine and go to the BIOS

Intel Advanced Menu ->
System Agent (SA) Configuration ->
Graphics Configuration ->

Then please change the "DVMT pre-allocated" size to something big (it's probably 32mb, and setting to 64mb should be enough). Then boot and see if the problem is gone.

Don't forget to set the value back to 32mb after testing :)
This isn't really a "fix", but just a way to confirm that something is using the top of stolen memory.
Comment 6 Jani Nikula 2015-10-23 12:25:35 UTC
Timeout, closing. Please reopen if the problem persists with latest kernels and the config changes from comment #5.


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.