Bug 44549

Summary: "Ring buffer test failed" on Radeon 5450
Product: DRI Reporter: Eric Bischoff <ebischoff>
Component: DRM/RadeonAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact:
Severity: normal    
Priority: medium    
Version: XOrg git   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
cat /proc/cmdline
none
cat /proc/meminfo
none
dmesg
none
glxinfo
none
lspci -v
none
uname -a
none
correct version of dmesg
none
correct version of cat /proc/cmdline none

Description Eric Bischoff 2012-01-07 03:42:02 UTC 
linux kernel : 3.2.0
libdrm-radeon1 : 2.4.29
xserver-xorg-video-radeon : 6.14.99
motherboard : ASUS M2A-VM HDMI (*no* VIA chipset)
Graphics card : msi R5450 with 1 Gb onboard
RAM memory size : 4 Gb

Description : when booting, the message "Ring buffer test failed" appears in dmesg log. A a consequence, the software rasterizer is chosen, and after X is started, there is no 3D acceleration.

Workaround : when I add "mem=3G" to the kernel parameters, the message disappears, and I get 3D acceleration. Of course, 1 Gb of RAM memory is lost for the applications.

(log files will get attached to this bug report)
Comment 1 Eric Bischoff 2012-01-07 03:45:18 UTC 
It's very similar to the problem described in http://lkml.indiana.edu/hypermail/linux/kernel/1105.2/03496.html

Nevertheless, it's not exactly the same hardware conditions, and the radeon driver patch that did solve that problem does not work for me.
Comment 2 Eric Bischoff 2012-01-07 04:23:14 UTC 
Created attachment 55245 [details]
cat /proc/cmdline
Comment 3 Eric Bischoff 2012-01-07 04:24:01 UTC 
Created attachment 55246 [details]
cat /proc/meminfo
Comment 4 Eric Bischoff 2012-01-07 04:24:30 UTC 
Created attachment 55247 [details]
dmesg
Comment 5 Eric Bischoff 2012-01-07 04:25:04 UTC 
Created attachment 55248 [details]
glxinfo
Comment 6 Eric Bischoff 2012-01-07 04:25:29 UTC 
Created attachment 55249 [details]
lspci -v
Comment 7 Eric Bischoff 2012-01-07 04:25:57 UTC 
Created attachment 55250 [details]
uname -a
Comment 8 Eric Bischoff 2012-01-07 04:45:33 UTC 
Ouch, I just realized there was a "nomodeset" option left from an old test. I'll see if that changes anything to the dmesg etc. Sorry.
Comment 9 Eric Bischoff 2012-01-07 04:57:32 UTC 
Created attachment 55257 [details]
correct version of dmesg
Comment 10 Eric Bischoff 2012-01-07 04:58:18 UTC 
Created attachment 55258 [details]
correct version of cat /proc/cmdline
Comment 11 Eric Bischoff 2012-01-07 05:00:16 UTC 
Rebooted with correct kernel parameters and attached corresponding versions of dmesg and cat /proc/cmdline. The other files do not change significantly. Sorry for the mistake.
Comment 12 Martin Peres 2019-11-19 08:24:58 UTC 
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/245.

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.