Bug 493332

Summary: Radeon Test Day: KMS breaks startup on HD3200 [1002:9610]
Product: [Fedora] Fedora Reporter: Stefan Becker <chemobejk>
Component: xorg-x11-drv-atiAssignee: Dave Airlie <airlied>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: low    
Version: 11CC: awilliam, fdc, fedora, jigorou3, kdekorte, michael.madore, nphilipp, opensource, remm, tcallawa, xgl-maint, yaneti, yulrottmann
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
URL: http://www.smolts.org/client/show/pub_aca10945-08f6-4f63-8b44-3a4892d089fa
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-09-12 08:10:33 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
X Log file from default boot (modesetting disabled)
none
Requested dmesg from loading radeon with modeset=1
none
Requested lspci -nn output
none
Xorg.0.log without modesetting
none
dmesg.log from radeon crash on HD3200 none

Description Stefan Becker 2009-04-01 13:35:44 UTC
Description of problem:

When KMS is enabled with radeon.modeset=1 bootup hangs when plymouth starts up. Non-KMS startup works OK.

Version-Release number of selected component (if applicable):
kernel-2.6.29.1-35.rc1.fc11.i586
xorg-x11-drv-ati-6.12.1-2.fc11.i586
libdrm-2.4.5-3.fc11.i586
mesa-libGL-7.5-0.6.fc11.i586

How reproducible:
always

Additional info:

Smolt profile attached as URL.

Comment 1 Tom "spot" Callaway 2009-04-01 13:44:43 UTC
I reproduced this on my system as well. 
http://www.smolts.org/show?uuid=pub_dd3d0c48-9a13-4ecd-a8a7-70b3de694bf2

Comment 2 Stefan Becker 2009-04-01 13:53:33 UTC
Sorry, forgot to mention: connected to Samsung DLP TV with HDMI. It's a VDR, so there is no other monitor :-)

Comment 3 Tom "spot" Callaway 2009-04-01 14:03:32 UTC
Mine is on a Philips 170S LCD, connected via VGA.

Comment 4 Tom "spot" Callaway 2009-04-01 14:04:22 UTC
Created attachment 337535 [details]
X Log file from default boot (modesetting disabled)

Comment 5 Remy Maucherat 2009-04-01 16:11:03 UTC
It also hangs on RV730 and RV770 when doing the modesetting, but did Dave post anywhere KMS was functional on R600+ ? Personally, I don't think so.

Comment 6 Michael Madore 2009-04-01 21:43:11 UTC
According to the basic test case instructions:

If you have an r600 or r700 chipset card (Radeon HD 2xxx, HD 3xxx or HD 4xxx),
add radeon.modeset=1 as a kernel parameter: in the live CD, hit any key when
the 'automatic boot in X seconds' prompt appears, then hit tab to edit boot
options

I tried this with the live CD and the kernel complained that it didn't
understand the radeon.modeset option.  The screen then went blank and I
eventually rebooted the system.  If I don't provide the radeon.modeset option,
then it will boot up in text mode and then start X successfully. I will test it
with a rawhide install next.

My smolt profile:

http://www.smolts.org/client/show/pub_d0215872-38fe-4625-8f82-ced21a8e012f

Comment 7 Tom "spot" Callaway 2009-04-01 21:56:49 UTC
Created attachment 337672 [details]
Requested dmesg from loading radeon with modeset=1

Comment 8 Tom "spot" Callaway 2009-04-01 21:58:14 UTC
Created attachment 337673 [details]
Requested lspci -nn output

Comment 9 Kevin DeKorte 2009-04-02 19:10:14 UTC
ATI 3650 shows a blankscreen when radeon.modeset=1 is used, without this everything is fine.

Comment 10 François Cami 2009-04-02 19:38:58 UTC
*** Bug 493388 has been marked as a duplicate of this bug. ***

Comment 11 Jeandre du Toit 2009-04-02 19:39:44 UTC
Created attachment 337884 [details]
Xorg.0.log without modesetting

Comment 12 Jeandre du Toit 2009-04-02 19:42:36 UTC
Also have a blank screen when radeon.modeset=1 is used. 

lspci -nn:

01:05.0 VGA compatible controller [0300]: ATI Technologies Inc RS780M/RS780MN [Radeon HD 3200 Graphics] [1002:9612]

Comment 13 François Cami 2009-04-02 19:42:36 UTC
Thanks for the logs and information, switching severity
to high because of crash, keeping priority low because
we have a workaround.

Switching to ASSIGNED.

---
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 14 Stefan Becker 2009-04-03 18:08:00 UTC
Created attachment 338096 [details]
dmesg.log from radeon crash on HD3200

You probably have this information already, but I thought I add the log from my system too:

  boot with nomodeset 3
  modprobe -r radeon drm
  modprobe drm debug=1
  modprobe radeon modeset=1

Comment 15 Adam Williamson 2009-04-15 19:20:30 UTC
kevin dekorte, can you file a separate bug? KMS failures on different chipsets don't necessarily have the same cause, so we need separate reports for each failing chipset.

-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers

Comment 16 Bug Zapper 2009-06-09 12:56:45 UTC
This bug appears to have been reported against 'rawhide' during the Fedora 11 development cycle.
Changing version to '11'.

More information and reason for this action is here:
http://fedoraproject.org/wiki/BugZappers/HouseKeeping

Comment 17 Stefan Becker 2009-09-12 08:10:33 UTC
Fedora 12 Radeon Test Day: KMS works now on this HW.

I doubt this will ever be fixed for F11, because it still uses the old radeon code base.