| View previous topic :: View next topic |
| Author |
Message |
coriolan
Apprentice
Joined: 21 Apr 2005
Posts: 273
|
 Posted: Sun Oct 30, 2005 1:04 pm Post subject: Gentoo-sources-2.6.14 breaks fglrx |
 |
|
I upgraded my kernel to 2.6.14 Gentoo, but after rebooting I have, once again, no direct rendering with ati-drivers. I did run module-rebuild which re-emerged the ati-drivers. What I get is:
| Code: | $ dmesg | grep fglrx
fglrx: Unknown symbol register ioctl32_conversion
fglrx: Unknown symbol unregister ioctl32_conversion
|
|
|
| Back to top |
|
 |
sian
Apprentice
Joined: 26 Jan 2004
Posts: 288
|
| Posted: Sun Oct 30, 2005 3:07 pm Post subject: |
|
|
| Did you reemerged ati-driver after the kernel change ? |
|
| Back to top |
|
|
telesphore
n00b
Joined: 17 Jul 2005
Posts: 10
|
| Posted: Sun Oct 30, 2005 4:10 pm Post subject: |
|
|
| sian wrote: | | Did you reemerged ati-driver after the kernel change ? |
I'm experiencing the same symtoms (on an amd64) & yes, In fact I've emerged several versions of the drivers from 8.14.13 to 8.18.8 with the same results. They work under kernel 2.6.13 & not 2.6.14. I did this because I automatically assumed that the drivers were the problem... & they may still be. My assumption was based on the fact that ATI has been a pain to set up. (I should mention that things have gotten better with the ATI drivers in the last few months now that I don't have to do the binary hack any more.)
Things work just fine under kernel 2.6.13-r* so for the time being I'll run kernel 2.6.13-r5 but i'd like to run 2.6.14 if i could. |
|
| Back to top |
|
|
coriolan
Apprentice
Joined: 21 Apr 2005
Posts: 273
|
| Posted: Sun Oct 30, 2005 4:37 pm Post subject: |
|
|
| sian wrote: | | Did you reemerged ati-driver after the kernel change ? |
Yes, I did, and I tried to point that out in the original mail as well. |
|
| Back to top |
|
|
MorLipf
Apprentice
Joined: 09 Nov 2004
Posts: 226
Location: Solingen, Germany
|
| Posted: Sun Oct 30, 2005 4:47 pm Post subject: |
|
|
| Same here. Can someone make a patch and a patched ebuild? |
|
| Back to top |
|
|
RaZoR1394
Guru
Joined: 09 Jan 2005
Posts: 356
|
| Posted: Sun Oct 30, 2005 4:53 pm Post subject: |
|
|
Yeah, I'm having this problem to and vmware won't configure for me either so I'm sticking with 2.6.13-r5. There's more info on the problem over here at the unofficial ati bugzilla.
_________________
http://en.wikipedia.org/wiki/Cell_(microprocessor) |
|
| Back to top |
|
|
Blutkind
n00b
Joined: 11 Apr 2004
Posts: 35
|
| Posted: Mon Oct 31, 2005 3:44 am Post subject: |
|
|
Checked into that bug, (as the ati.cchtml.com link states) the [un]register_ioctl32_conversion is deprecated so you have to define it properly as it is now 100% deprecated as of 2.6.14 the drivers will no longer work. (Fix is in store) I was working on a fix myself but i'm not familiar with the functions to re-produce the same functions.
_________________
ArchLinux 64 Dev | 2.6.14 AMD64 |
|
| Back to top |
|
|
blackwhite
Apprentice
Joined: 24 Jun 2004
Posts: 250
|
| Posted: Sun Nov 27, 2005 6:54 pm Post subject: |
|
|
| Because ati-drivers-8.19.10.ebuild works well with direct rendering under 2.6.14 kernel in my case, I recommand you use the version ebuild. |
|
| Back to top |
|
|
coriolan
Apprentice
Joined: 21 Apr 2005
Posts: 273
|
| Posted: Sun Nov 27, 2005 7:20 pm Post subject: |
|
|
| Yes, well, been there done that. I had not updated this old thread. |
|
| Back to top |
|
|
|
Kernel & Hardware
