LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (http://www.linuxquestions.org/questions/slackware-14/)
-   -   Amd/Ati proprietary driver mesa 9 xorg 1.13 issue (http://www.linuxquestions.org/questions/slackware-14/amd-ati-proprietary-driver-mesa-9-xorg-1-13-issue-4175451657/)

Amrod78 02-25-2013 02:24 PM

Amd/Ati proprietary driver mesa 9 xorg 1.13 issue
 
Hi everybody, I have a strange issue with the catalyst driver after the recent upgrade of mesa and xorg; I have upgraded following the usual steps (disinstall fglrx,
upgrade x packages and then reinstall fglrx) and all worked fine, after a reboot X still continued to work but if I run fglrxinfo I have the following output

Code:

libGL error: failed to load driver: swrast
libGL error: Try again with LIBGL_DEBUG=verbose for more details.
display: :0.0  screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: AMD Radeon HD 7800 Series
OpenGL version string: 1.4 (2.1 (4.2.12172 Compatibility Profile Context 12.10.17))

by reinstalling fglrx the problem is solved till a reboot/shutdown after which the problem reappears.

I made some tests with a fresh install and the results were the same; after a couple of days of headache trying to figure out what would be the problem I found nothing; have someone the same problem? Any idea to resolve the problem?

Thanks in advance.

wildwizard 02-25-2013 03:16 PM

Not familiar with the AMD drivers (nvidia owner), but does it have kernel modules that need to load and perhaps they aren't getting loaded on boot?

TobiSGD 02-25-2013 03:21 PM

Have you patched the driver? It shouldn't install at all on the kernel 3.7 used by -current without being patched.
Where there any errors during the patching or the driver install?

Amrod78 02-25-2013 04:54 PM

@wildwizard: AMD drivers have a kernel module too and it is loaded, I checked it now by rebooting the pc

@TobiSGD: Yes the driver are patched, there were no errors during patching and building the kernel module.

I forgot to mention that I'm using the catalyst's internal slackbuild to build the package.

The thing that I don't understand is why only after a reboot this issue is coming out, if the driver weren't working for some reason they
have not to work even if I reinstall them, but by reinstalling all works like a charm.

adamk75 02-25-2013 07:19 PM

Show us your /var/log/Xorg.0.log file from when it's not working, and then from when it is working.

Beelzebud 02-25-2013 07:41 PM

Quote:

Originally Posted by Amrod78 (Post 4899602)
@wildwizard: AMD drivers have a kernel module too and it is loaded, I checked it now by rebooting the pc

@TobiSGD: Yes the driver are patched, there were no errors during patching and building the kernel module.

I forgot to mention that I'm using the catalyst's internal slackbuild to build the package.

The thing that I don't understand is why only after a reboot this issue is coming out, if the driver weren't working for some reason they
have not to work even if I reinstall them, but by reinstalling all works like a charm.

If you're running 64-bit with multilibs enabled, AMD's slackbuild will not build a proper package. The only way I've gotten everything to work on 64-bit multilib with an AMD graphics driver is to load up the installer and choose the generic install selection, instead of building a slackpkg.

Amrod78 02-26-2013 12:11 PM

2 Attachment(s)
Quote:

Originally Posted by adamk75 (Post 4899690)
Show us your /var/log/Xorg.0.log file from when it's not working, and then from when it is working.

This is the log when there is the issue Attachment 11938

and this when all is ok Attachment 11939

I'm using catalyst 13.2beta6 in the log somewhere is written that are unsupported but the same problem appeared with the 13.1

Quote:

If you're running 64-bit with multilibs enabled, AMD's slackbuild will not build a proper package. The only way I've gotten everything to work on 64-bit multilib with an AMD graphics driver is to load up the installer and choose the generic install selection, instead of building a slackpkg.
The problem is present both in a multilib and only 64-bit installation, tomorrow after work I'll try a 32 bit installation to check all the variables.


All times are GMT -5. The time now is 01:30 AM.