Bug 8697 - can not enable screen 2 or 3 on eyefinity card, worked in mga2.
Summary: can not enable screen 2 or 3 on eyefinity card, worked in mga2.
Status: RESOLVED OLD
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: x86_64 Linux
Priority: High normal
Target Milestone: ---
Assignee: Mageia Bug Squad
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-01-15 18:10 CET by Simon Putt
Modified: 2015-12-24 07:26 CET (History)
5 users (show)

See Also:
Source RPM: x11-server-xorg
CVE:
Status comment:


Attachments
backtrace xorg after switching to 3head mode (5888x1152) (11.80 KB, application/octet-stream)
2013-01-15 21:28 CET, Simon Putt
Details
fedora patch to expand virtual screens to hold physical ones. (2.15 KB, patch)
2013-04-04 09:14 CEST, Simon Putt
Details | Diff

Description Simon Putt 2013-01-15 18:10:08 CET
I am using a Radeon HD 6770 Eyefinity card (3 head support), and in Mga2 it worked great either with the gnome3 displays option, or with xrandr scripts which i will list at the end.

When using the same scripts or attempting to enable the heads in mga3 beta 2 i get screen corruption and a mouse pointer.

I think this may be a problem with xorg-driver-ati, xorg or xrandr

I made the following in ARandR

enable all screens (3)

#!/bin/sh
xrandr --output DisplayPort-0 --mode 1920x1080 --pos 3968x0 --rotate normal --output DVI-1 --mode 1920x1080 --pos 0x0 --rotate normal --output DVI-0 --mode 2048x1152 --pos 1920x0 --rotate normal

just the right head

#!/bin/sh
xrandr --output DisplayPort-0 --mode 1920x1080 --pos 2048x0 --rotate normal --output DVI-1 --off --output DVI-0 --mode 2048x1152 --pos 0x0 --rotate normal


going back to single output is no possible as i can not navigate the screen.
Comment 1 Simon Putt 2013-01-15 21:02:45 CET
[ 2987.084461] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.085366] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.086123] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.086914] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.087643] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.088274] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.089365] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.090145] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.090624] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.091538] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.092787] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.093847] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.094817] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.095834] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.096545] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.097542] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.098442] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.099981] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.100888] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
[ 2987.101831] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!

this is what is printed in the dmesg when it happens.

running glxgears and full screening it on a single display causes the same corruption
Comment 2 Simon Putt 2013-01-15 21:28:13 CET
Created attachment 3379 [details]
backtrace xorg after switching to 3head mode (5888x1152)

back trace from xorg
Simon Putt 2013-01-16 20:09:04 CET

Priority: Normal => release_blocker
Hardware: i586 => x86_64

Comment 3 Simon Putt 2013-01-17 02:20:12 CET
also if i change to a vt on switch back the screen is also corrupted
Comment 4 Sander Lepik 2013-01-17 09:52:32 CET
Bugs that can be fixed with update after release are not release blockers.

Priority: release_blocker => High
CC: (none) => sander.lepik

Comment 5 Simon Putt 2013-01-17 18:23:11 CET
(In reply to comment #3)
> also if i change to a vt on switch back the screen is also corrupted

this seems to have been fixed with the new kernel and libdrm updates for cauldron.

only problem persisting now is the cannot enable 2nd head
Comment 6 Simon Putt 2013-01-17 18:36:30 CET
(In reply to comment #1)
> [ 2987.084461] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.085366] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.086123] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.086914] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.087643] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.088274] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.089365] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.090145] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.090624] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.091538] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.092787] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.093847] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.094817] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.095834] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.096545] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.097542] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.098442] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.099981] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.100888] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> [ 2987.101831] [drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!
> 
> this is what is printed in the dmesg when it happens.
> 
> running glxgears and full screening it on a single display causes the same
> corruption

Fixed with mesa/kernel
Comment 7 Simon Putt 2013-01-30 13:56:26 CET
now that mesa 9.1 has branched and looking at a late feb release, could I get a new build so i can see if my corruption bug has been fixed? many thanks.
Comment 8 Simon Putt 2013-02-24 12:08:39 CET
things are a little better now with the newer mesa builds the corruption has gone and 2nd head is usable now, 3rd monitor though (on display-port) still kicks back to gdm. according to the people on #radeon this is a mesa problem, so i will get back to them.
Comment 9 Simon Putt 2013-04-04 09:14:21 CEST
Created attachment 3692 [details]
fedora patch to expand virtual screens to hold physical ones.
Comment 10 Simon Putt 2013-04-04 09:15:53 CEST
I have done additional testing and in fedora i can enable a wide spanned desktop over all three screens, with ARandR in fedora there is a wide virtual display where you can arrange your screens

in mageia it just covers the primary one, leaving the other 2 screens with no place to be arranged

i believe gnome-shell crashes because xorg can not expand the virtual screens to hold the physcal ones, however, i did find a patch in fedoras xorg-server that fixes this, could we include it in ours? i would be willing to test, as other wise it would mean changing distro which i would rather not do as i am more than happy with mageia.
Comment 11 Simon Putt 2013-04-04 10:47:51 CEST
I have also tested this with my 3 cards, a 5670 (redwood) a 6670 (Turks) and a 6770 (Juniper) radeon cards, all 3 show the same bug.
Simon Putt 2013-04-04 10:55:16 CEST

CC: (none) => thierry.vignaud
Source RPM: (none) => x11-server-xorg

Simon Putt 2013-04-04 10:55:36 CEST

CC: (none) => tmb

Simon Putt 2013-04-04 10:57:32 CEST

CC: (none) => luigiwalser

Comment 12 David Walser 2013-05-02 15:14:06 CEST
Dual-head is working for me in current Cauldron with a Radeon HD 5450.
Comment 13 Simon Putt 2013-05-02 19:47:00 CEST
(In reply to David Walser from comment #12)
> Dual-head is working for me in current Cauldron with a Radeon HD 5450.

What DE?
Comment 14 David Walser 2013-05-02 20:38:14 CEST
(In reply to Simon Putt from comment #13)
> (In reply to David Walser from comment #12)
> > Dual-head is working for me in current Cauldron with a Radeon HD 5450.
> 
> What DE?

KDE.  I created an xorg.conf for doing dual head, so it also works in KDM.

I haven't tried using randr/krandrtray to do it just for a login session, but I assume that still works.
Comment 15 David Walser 2014-01-23 19:26:58 CET
Simon, have you tested this for Mageia 4 (i.e., current Cauldron)?
Comment 16 Nic Baxter 2015-12-24 07:26:21 CET
Since there are insufficient details provided in this report for us to investigate the issue further, and we have not received feedback to the information we have requested above, we will assume the problem was not reproducible, or has been fixed in one of the updates we have released for the reporter's distribution.

Users who have experienced this problem are encouraged to upgrade to the latest update of their distribution, and if this issue turns out to still be reproducible in the latest update, please reopen this bug with additional information.

Closing as OLD.

Status: NEW => RESOLVED
CC: (none) => nic
Resolution: (none) => OLD


Note You need to log in before you can comment on or make changes to this bug.