linux-kernel - Re: [Regression] [Radeon] Stream command failures and render problems in Evergreen (Radeon HD 5600) in kernel 3.4

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 7 Jun 2012 10:25:02 -0400
From:	Jerome Glisse <j.glisse@...il.com>
To:	Brian Gitonga Marete <marete@...hnix.com>
Cc:	LKML <linux-kernel@...r.kernel.org>,
	David Airlie <airlied@...ux.ie>,
	Alex Deucher <alexander.deucher@....com>,
	Jerome Glisse <jglisse@...hat.com>
Subject: Re: [Regression] [Radeon] Stream command failures and render problems
 in Evergreen (Radeon HD 5600) in kernel 3.4

On Wed, Jun 6, 2012 at 7:01 PM, Brian Gitonga Marete <marete@...hnix.com> wrote:
> On Thu, Jun 7, 2012 at 1:59 AM, Brian Gitonga Marete <marete@...hnix.com> wrote:
>> On Sun, Jun 3, 2012 at 8:26 PM, Brian Gitonga Marete <marete@...hnix.com> wrote:
>>> Hello,
>>>
>>> 3.4 breaks rendering on my Radeon evergreen card. Immediately after
>>> logging in into the desktop, I get the following errors in the kernel
>>> log:
>>>
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:439 cb[0] bo too
>>> small (layer size 147456, offset 0, max layer 1, bo size 143360, slice
>>> 575)
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:443 problematic
>>> surf: (192 192) (0 4 1 0 0 0 0)
>>> radeon 0000:01:00.0: evergreen_packet3_check:2055 invalid cmd stream 461
>>> [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream !
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:439 cb[0] bo too
>>> small (layer size 147456, offset 0, max layer 1, bo size 143360, slice
>>> 575)
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:443 problematic
>>> surf: (192 192) (0 4 1 0 0 0 0)
>>> radeon 0000:01:00.0: evergreen_packet3_check:2055 invalid cmd stream 461
>>> [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream !
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:439 cb[0] bo too
>>> small (layer size 798720, offset 0, max layer 1, bo size 794624, slice
>>> 3119)
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:443 problematic
>>> surf: (384 520) (0 4 1 0 0 0 0)
>>> radeon 0000:01:00.0: evergreen_packet3_check:2055 invalid cmd stream 461
>>> [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream !
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:439 cb[0] bo too
>>> small (layer size 458752, offset 0, max layer 1, bo size 446464, slice
>>> 1791)
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:443 problematic
>>> surf: (512 224) (0 4 1 0 0 0 0)
>>> radeon 0000:01:00.0: evergreen_packet3_check:2055 invalid cmd stream 7170
>>> [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream !
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:439 cb[0] bo too
>>> small (layer size 798720, offset 0, max layer 1, bo size 790528, slice
>>> 3119)
>>> radeon 0000:01:00.0: evergreen_cs_track_validate_cb:443 problematic
>>> surf: (384 520) (0 4 1 0 0 0 0)
>>> radeon 0000:01:00.0: evergreen_packet3_check:2055 invalid cmd stream 461
>>> [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream !
>>>
>>> Meanwhile, windows on the desktop are rendered incorrectly (with
>>> certain regions covered by strange patterns or smudged) or not at all.
>>> This makes the desktop unusable.
>>>
>>> I can always reproduce this problem on 3.4
>>>
>>> Kernels prior to 3.4 work fine.
>>>
>>> I will try and find the time to bisect this issue but meanwhile here
>>> are some further details:
>>>
>>
>> I have git-bisected this to find that the first bad commit is:
>> 285484e2d55 - drm/radeon: add support for evergreen/ni tiling
>> informations v11.
>>
>> So I have also added Jerome to the CC list. Please see the attached
>> bisect.log for the git-bisect log.
>
> Sorry, forgot to attach. Now attached.
>
> --
> Brian Gitonga Marete
> Toshnix Systems - http://toshnix.com

Update your libdrm to libdrm from git should fix your issue.

Cheers,
Jerome
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/