Package Details: catalyst 15.9-13

Git Clone URL: https://aur.archlinux.org/catalyst.git (read-only)
Package Base: catalyst
Description: AMD/ATI Catalyst drivers for linux. fglrx kernel module only. Radeons HD 2 3 4 xxx ARE NOT SUPPORTED
Upstream URL: http://www.amd.com
Licenses: custom
Conflicts: catalyst-test
Submitter: wonder
Maintainer: Vi0L0
Last Packager: Vi0L0
Votes: 338
Popularity: 0.776848
First Submitted: 2009-08-09 08:02
Last Updated: 2016-10-19 19:48

Required by (3)

Sources (14)

Latest Comments

Crabsi commented on 2016-11-23 16:44

@pavanky what are the right mirrors now? I still can't install catalyst-hook

Edit: nevermind got it now... for anyone wondering you have to add this to /etc/pacman.conf:
[xorg117]
Server = http://mirror.hactar.xyz/Vi0L0/xorg117/$arch
[catalyst]
Server = http://mirror.hactar.xyz/Vi0L0/catalyst/$arch

pavanky commented on 2016-11-22 06:36

Not sure if this is the right place to ask, but did you forget to update the repos on wirephire ? The catalyst-hook package there still wants linux < 4.7

**EDIT:** Nevermind I just noticed that the mirrors changed. It's all good.

mikaelbrun commented on 2016-10-20 13:42

I forgot to give a feedback ViOLO. Never too late, I hope.

It was my mistake. I forgot to put the xorg117 on the top :-)
Repos are alright!

Thanks for the reminder!

Vi0L0 commented on 2016-09-18 19:40

If I would update this dependency and you will start catalyst on xserver 1.18 you will only see black screen of death.
It's because catalyst doesn't support xserver 1.18 and (for 99%) never will.

You can build xorg-server 1.17 from arch's git: https://git.archlinux.org/svntogit/packages.git/snapshot/packages-e02b7fe044e46037ad80325fb9aebe3e7c39fa05.tar.gz

or use [xorg117] repo: https://wiki.archlinux.org/index.php/AMD_Catalyst#Xorg_repositories

socrat commented on 2016-09-18 04:04

trying to install from your repo gives error cannot resolve "xorg-server<1.18.0"
plz upgrade xorg-server dependency

Vi0L0 commented on 2016-08-30 15:07

Repos are fine:

# LANG=C pacman -Sl xorg117 | grep xorg-server
xorg117 xorg-server 1.17.4-2 [installed]
xorg117 xorg-server-common 1.17.4-2 [installed]
xorg117 xorg-server-devel 1.17.4-2
xorg117 xorg-server-utils 7.6-4
xorg117 xorg-server-xdmx 1.17.4-2
xorg117 xorg-server-xephyr 1.17.4-2
xorg117 xorg-server-xnest 1.17.4-2
xorg117 xorg-server-xvfb 1.17.4-2 [installed]
xorg117 xorg-server-xwayland 1.17.4-2 [installed]

Make sure that you added [xorg117] repo above every other repo in /etc/pacman.conf

I also recommand using hactar.bz mirror as I cannot access wirephire anymore (xorg117 on wirephire works fine as for now but won't get any update):
[xorg117]
Server = http://mirror.hactar.bz/Vi0L0/xorg117/$arch

mikaelbrun commented on 2016-08-30 09:00

I have problems getting the xorg-server 1.17 work. It used to, but after a reinstall, it only installs the xorg-server 1.18.4.

Is there something wrong with your repos, ViOLO? I have tried both servers, without any luck.
I want my catalyst back!

Vi0L0 commented on 2016-08-24 11:41

hd 7770 is gcn so should work fine in both catalyst versions 15.9 and 15.12

timxx commented on 2016-08-23 15:35

Is this driver support for HD 7770? Since I downloaded one from official website, it's name is radeon-crimson-15.12-15.302-151217a-297685e.zip. :(

Vi0L0 commented on 2016-06-11 16:57

Some files get corrupted, I reuploaded all packages of [catalyst] and [catalyst-stable] once again. Looks better now.

shahin commented on 2016-06-10 23:52

thanks for the other mirror as the original one still has errors.

pavanky commented on 2016-06-10 15:22

The original repo still seems to have issues with the key

Vi0L0 commented on 2016-06-10 05:28

Thanks guys and sorry for inconvinience.
My key is fine, more like a problems with space, please check out now.
If it didn't help look at hactar mirror:
[catalyst]
Server = http://mirror.hactar.bz/Vi0L0/catalyst/$arch

pavanky commented on 2016-06-09 22:20

Did your key change? Your repos no longer seem to work.

wendradon commented on 2016-06-09 14:10

catalyst: signature from "Vi0L0 <vi0l093@gmail.com>" is invalid

i tried adding SigLevel = Optional TrustAll and pacman-key --recv-keys 0xabed422d653c3094, but nothing worked for me :-/

Vi0L0 commented on 2016-06-08 20:43

just a sidenote: valve fixed this cs:go wallhack some time ago. works fine now on 4.5 and 4.6 kernels

Vi0L0 commented on 2016-03-23 21:24

been using 4.5 kernel since rc5 on catalyst, but decided to make some game testing before commiting 4.5 support.
There was no problems except for cs:go which generates some glitches - missing walls :), it could be used as a wallhack, literally, not sure if it is bug or a feature :P. Occures on catalyst 15.9 and 15.12, catalyst-firepro works well.
So if you dont like this bug (could be that im the only one touched by problem) use older linux 4.4 or linux-lts or mentioned catalyst-firepro

Vi0L0 commented on 2016-02-12 10:29

@USA-RedDragon:
https://wiki.archlinux.org/index.php/Arch_User_Repository#Prerequisites

USA-RedDragon commented on 2016-02-11 04:30

You might think about adding patch and fakeroot to the dependencies. I'm installing a new system and had to restart the install twice. Just a quick suggestion. Same with catalyst-utils

Vi0L0 commented on 2016-01-20 14:26

15.9-9:
- added 4.4 kernel support

Vi0L0 commented on 2016-01-04 18:27

I don't like this dirty, kinda 'illegal', patch and that's why I was always trying to avoid it, but yes - sometimes it has to be applied.
I will change it in the next update (which should also bring 4.4 support), it should happen within days.

nous commented on 2016-01-04 17:53

I'm using catalyst-hook-15.12 from the [catalyst] repo. Trying to build the fglrx module in an older LTS kernel (x86_64, latest 3.14.58-lts), I'm getting the following:

LD [M] /tmp/catalyst.PWoUZ4/fglrx.o
Building modules, stage 2.
MODPOST 1 modules
FATAL: modpost: GPL-incompatible module fglrx.ko uses GPL-only symbol 'task_nice'
/usr/lib/modules/3.14.58-pf-lts/build/scripts/Makefile.modpost:90: recipe for target '__modpost' failed
make[1]: *** [__modpost] Error 1
Makefile:1281: recipe for target 'modules' failed
make: *** [modules] Error 2

Please, remove the $CARCH before the fglrx_gpl_symbol.patch:
test "${CARCH}" = "i686" && patch -Np1 -i ../fglrx_gpl_symbol.patch

Same error also with catalyst-hook 15.9 from [catalyst-stable]. Manually applying the patch allows modpost to succeed in 64bit machines. Thanks.

Vi0L0 commented on 2015-12-22 08:22

This message comes on catalyst >= 15.11. For now its available only on catalyst-test@AUR and [catalyst] repo

15.9 officially supports 5 and 6 k. [catalyst-stable] repo and most AUR packages still contains it. It will be updated in some future (maybe month) and then special legacy repo will arrive.

Markismus commented on 2015-12-21 13:51

I got this message:

------------ DROPPED SUPPORT FOR RADEONS HD 5 6 k --------------
Please note that this release SADLY :( is
dropping support for pre-GCN cards so mainly
Radeons of series 5000 and 6000
----------------------------------------------------------------

What could I try to install catalyst on a 5k card with kernel 4.2?

Vi0L0 commented on 2015-12-13 19:55

15.9-8:
just like commit says:
- adding new patch from crimson
- removing gcc49 dependency
- adding initial grsec kernel support

should be safe to remove gcc49 pkg :)

Vi0L0 commented on 2015-10-28 18:45

catalyst-test now supports 4.2 kernel. You have to install gcc49, thought.
Feel free to test and give feedback - especially if your system is i686 or when its not working.

https://aur.archlinux.org/packages/catalyst-test

mouseman commented on 2015-10-25 19:54

Thanks for the Gnome fix!

Vi0L0 commented on 2015-10-23 08:25

It would be most easy for you if you would use catalyst-test from AUR or [catalyst] repository. Look at instalation and configuration on wiki, it should be simple:
https://wiki.archlinux.org/index.php/AMD_Catalyst#Installing_the_driver

catalyst doesn't support 4.2 kernel, so boot your system from linux-lts (configure the bootloader to use it).

Also since you are using gnome 3.18 read my comment below and use my patched cogl package

herycp commented on 2015-10-23 07:34

im use notebook:
a8 4500
radeon 7640g + 7470m

Sistem :
kernel 4.2.3-1 + lts
gnome 3.18
gdm

is save if im install it??
or must from unofficial??
need configure?

Vi0L0 commented on 2015-10-18 09:36

GNOME 3.18 fix
In [catalyst] and [catalyst-stable] you may find a new package cogl 1.22.0-2. It brings fix for recent problems with new GNOME 3.18 and new GDM (it baheves strange with plasma though, after choosing plasma and entering password I have to switch to tty and then back to X to see plasma desktop, probably another bug).
Source and a patch is here: http://mirror.rts-informatique.fr/archlinux-catalyst/tarball/gnome318-problems/cogl-1.22.0-2.src.tar.gz

Vi0L0 commented on 2015-09-29 20:30

updated to 15.9
4.2 kernel is in [core] and we still don't have support for it.
So for now please use linux-lts kernel.

be aware that I removed linux from dependency array and so it will let linux to update (while it's not supported tongue), if you will not configure bootloader to use kernel older than 4.2, like linux-lts, it will probably boot linux and you will probably gonna see a freeze or blank screen. System should work though, you can ssh to it and config bootloader from here, or use single mode if you can't ssh, or well there's possibility that linux-lts was added before, boot it and config as default kernel to boot.

Bumble-Bee commented on 2015-09-16 08:22

15.9 is now out (see http://support.amd.com/en-us/kb-articles/Pages/AMDCatalyst14-9LINReleaseNotes.aspx) still no 4.x kernel support tho :(

olive commented on 2015-09-01 15:23

I suffer terrible performance issue (Radeon HD6570) when I try to resize windows (particularly chromium) under xfce. The windows flinker and take a lot of the to resize. It may be the same bug as the chromium glitch mentioned in the comment below. For now, I use the opensource ati driver that does not have this problem and seems to work relatively well.

Vi0L0 commented on 2015-07-13 18:29

here's bug report of chromium glitch : http://ati.cchtml.com/show_bug.cgi?id=1182
feel free to add comment - this should make it more "visible"

Vi0L0 commented on 2015-07-13 17:30

iirc chrome was always glitching on plasma DE, with one catalyst release i was able to fight it with systemsettings5, but it wasnt general solution.
As for sublime not long ago ekr was writing about sublimes terrible performance on catalyst 15.5 release.
The one and only real workaround for both chromium and sublime glitches/performance issues is to disable compositing.

dkreuter commented on 2015-07-13 17:03

I believe the chromium glitching started one release before 15.7.
With 15.7 Sublime Text becomes super slow (press Ctrl+K B to trigger sidebar hide/reveal animation to demonstrate).

Vi0L0 commented on 2015-07-09 18:57

15.7::
OpenGL is now in 4.5 version, it brings fixes for some titles like borderlands 2 which now works well, still looks like crap smile but at least its fluid. Even warsow seems snappier, but it's not mindblowing. Dota 2 is kinda laggy in some situations, plus it crashed on exit (sic!) and I wasn't able to kill its process (sadface, at least it wasnt eating resources, only ate some ram and sit)...
They fixed vaapi which latelly wasn't working with -va vaapi option.
It does support xserver 1.17, but tty - just like with ubuntu 15.4 version - is all black - another sadpanda.
And ofcourse glitching (on plasma DE) chromium - cherry on the top.


I don't think it deserves [catalyst-stable] or catalyst@AUR...

I will put it into [catalyst] for now, catalyst-test@AUR already is updated

Bumble-Bee commented on 2015-07-09 08:20

Catalyst 15.7 is out, release notes state "Xorg/Xserver 7.4 and above (up to 1.17)"

ekr commented on 2015-06-24 14:43

Anyone else having problems with terrible scrolling performance in sublime text with catalyst? Other apps don't seem to have this problem, although it's clearly a 2D acceleration issue.

I'm back to catalyst after the xorg-server crashed several times with the open source driver.

Vi0L0 commented on 2015-06-04 13:23

updated to 15.5-1 - AKA "nothing interesting":
- it doesn't support xserver 1.17
- looks older than ubuntu vivid driver which we are using ie in catalyst-test@AUR
- i've added initial support for 4.1 kernel (using kolasa's patch, tested on 4.1rc5)
- rel notes: http://support.amd.com/en-us/kb-articles/Pages/AMDCatalyst15-5LINReleaseNotes.aspx

Vi0L0 commented on 2015-05-04 19:36

14.12-3:
- 4.0 kernel support (thanks to kolasa for patch)

Vi0L0 commented on 2015-03-08 14:06

14.12-2:
- using kolasa's patch to cover 3.19 kernel support

optomus commented on 2015-01-29 23:30

I tired to install today after running pacman -Syu, which brought down linux-3.18.4-1. This threw an error "make *** /usr/lib/modules/3.18.2-2-ARCH cannot be found: Directory or file does not exist" (or something like that). I believe the kernel has been updated to 3.18.4-1-ARCH though this (PKGBUILD?) is expecting 3.18.2-2-ARCH.
To get around this problem until an update is produced run:
sudo mv /usr/lib/modules/3.18.4-1-ARCH /usr/lib/modules/3.18.2-2-ARCH
Then install with "yaourt -S catalyst" and before rebooting run:
sudo mv /usr/lib/modules/3.18.2-2-ARCH /usr/lib/modules/3.18.4-1-ARCH
Otherwise your kernel will fail to load.

Vi0L0 commented on 2014-12-09 08:00

Uuuuu finaly same new release.
Im afk and should be back tomorrows evening, then I should be able to test and update

Vi0L0 commented on 2014-12-09 07:59

Uuuuu finaly same new release.
Im afk and should be back tomorrows evening.

Vi0L0 commented on 2014-10-03 22:33

don't look at the amd's versioning, it's pure craziness ;P.
I'm always versioning in a old-fasion way "year.month" to keep pacman sane - theres only one problem with it - whenever amd decide to put two separate versions in a same month, where newer is actually worse/older...
catalyst-test brings beta driver which was prepared for ubuntu 14.10 to support xserver 1.16.
And this stable driver, versioned by amd as 14.9, is just another mile-stone - probably old because getting certification from microsoft is taking some time (remember that catalyst shares code between linux and windows). And it doesn't support xserver 1.16 - but its working, amd marked is as stable, and so am I

GreenRaccoon23 commented on 2014-10-03 22:19

Right now, this one is newer than the catalyst-test package, right? Looking at the pkgbuild's, this one is version 14.301.1001 and catalyst-test is 14.201. I'm just a little confused why the catalyst-test package supports xserver 1.17 and this one doesn't.

Vi0L0 commented on 2014-10-03 20:54

updated to 14.9-1:
- official, stable release
http://support.amd.com/en-us/kb-articles/Pages/AMDCatalyst14-9LINReleaseNotes.aspx

- it doesn't support xserver 1.16

- I added my fresh patch for kernel 3.17 rc6.
Notes about the patch:
1. it probably wont work on 3.17 kernels older than rc6 because I don't know how to tell preprocessor about rc version :P
2. im unable to test resume/suspend
3. im not the developer, just pressing random keys :P so the patch could be less/more noobish :-)

Vi0L0 commented on 2014-10-03 19:56

well yes, catalyst-test @AUR or packages from [catalyst] are basing on driver for ubuntu 14.10 and does support xserver 1.16.

Its just that official and stable AMD's 14.9 doesn't support xserv 1.16.

TheNiceGuy commented on 2014-10-03 16:36

Someone on Reddit claimed that the next version works with xorg 1.16.
http://www.reddit.com/r/linux/comments/2hzwo9/amd_catalyst_149_officially_released_for_linux/ckyab6g?context=3

Vi0L0 commented on 2014-09-30 18:48

I will be back at Friday, and then I will update to the new stable release (14.9).
Just note that it doesn't support xserver 1.16 (it should be better than 14.4, though)

cooperrj commented on 2014-09-20 12:17

Need to remove Catalyst to update Linux to 3:16, then reinstalled Catalyst.

Vi0L0 commented on 2014-09-16 19:32

how? depends list is saying 'linux<3.17'

cooperrj commented on 2014-09-16 15:07

Error on update, Requires linux<3.16.

Vi0L0 commented on 2014-08-18 16:37

Nope, xserver 1.16 still isn't supported

Neije commented on 2014-08-18 12:07

Just to be sure : xorg 1.16 ist not yet supported ?

Vi0L0 commented on 2014-08-17 14:53

fixed, thanks for the head up

LeonardK commented on 2014-08-17 14:23

"linux<3.16"
shouldnt it atelast be '<=' ? or is catalyst not yet 3.16 ready?

Vi0L0 commented on 2014-08-16 09:10

It's embedded driver, useless for us. But keep the head up, new catalyst should be released soon

NeoAcheron commented on 2014-08-16 08:23

Catalyst driver 14.8 has been "quietly" made available...

Not on the main driver selection pages on AMD's site though.

Saw this via Phoronix: http://www.phoronix.com/scan.php?page=news_item&px=MTc2MzM

But here's a link to AMD's site for the driver: http://support.amd.com/en-us/download/embedded?os=Linux+x86&rev=13.151

marmotta commented on 2014-07-02 22:03

No way to use gdm? What is the latest working version with gdm?

Vi0L0 commented on 2014-06-09 20:44

14.4-21: just a pkgrel bump to confirm 3.15 kernel support

Vi0L0 commented on 2014-05-25 11:33

updated to 14.4-20 (catalyst 14.4 rev2)

ChemBro commented on 2014-05-24 10:28

New version is out: 14.4 rev 2

A one game bugfix-only release for The Witcher II.

Aerion commented on 2014-05-17 15:46

Yes, I'm experiencing exactly the same issue. The second screen (the primary display in my setup) is completely unusable, with a totally garbled screen. Disabling Tear Free does indeed fix the issue, so thanks for that tip!

Aerion commented on 2014-05-17 15:45

@bheinks No, I'm experiencing exactly the same isse. The second screen (the primary display in my setup) is completely unusable, with a totally garbled screen. Disabling Tear Free does indeed fix the issue, so thanks for that tip!

Why is development of the Linux Catalyst driver so incredibly slow and so buggy? The Windows drivers just work, no hassle.

bheinks commented on 2014-05-12 08:08

Is anyone else experiencing graphical glitches with a dual head setup? It makes my second monitor completely unusable. A workaround has been to disable Tear Free at the cost of some obvious tearing in video playback and games.

Vi0L0 commented on 2014-05-01 19:22

There's a bug in linux 3.14.2:
https://bugs.archlinux.org/index.php?do=details&action=details.addvote&task_id=40171

Temporary workaround is to downgrade linux and linux-headers to 3.14.1 or use gcc from [testing]. The first solution should be imho safer. You can find 3.14.1 at arm:

x86_64:
ftp://seblu.net/archlinux/arm/2014/04/25/core/os/x86_64/linux-3.14.1-1-x86_64.pkg.tar.xz
ftp://seblu.net/archlinux/arm/2014/04/25/core/os/x86_64/linux-headers-3.14.1-1-x86_64.pkg.tar.xz

i686:
ftp://seblu.net/archlinux/arm/2014/04/25/core/os/i686/linux-3.14.1-1-i686.pkg.tar.xz
ftp://seblu.net/archlinux/arm/2014/04/25/core/os/i686/linux-headers-3.14.1-1-i686.pkg.tar.xz

Vi0L0 commented on 2014-04-28 20:32

updated to 14.4-10, catalyst 14.4 stable:
- xserver 1.15 support
(gotjazz wrote: little warning though - if you use xbmc, get sdl-nobackingstore from the AUR or you'll have tearing during video playback with xorg 1.15)
- many fixes here and there

Vi0L0 commented on 2014-04-05 11:20

13.12-2:
added Cold's 3.14 patch

Vi0L0 commented on 2014-03-22 21:54

from official, xorg-server package contains additional security patch, sources here: http://catalyst.wirephire.com/tarball/xorg114/

olive commented on 2014-03-22 16:02

@ginoboy in pacman.conf

[xorg114]
SigLevel = Never

or manually add the Vi0l0 key.

By the where can I get the PKGBUILD of xorg14. Is it simply older official archlinux PKGBUILD? (that we can get with svn) or are there other tweaks?

ginoboy commented on 2014-03-20 17:34

I can't use xorg114 repository:

pacman error:
xorg114: signature from "Vi0L0 <vi0l093@gmail.com>" is unknown trust

Thanks.

Vi0L0 commented on 2014-02-02 21:25

14.1 beta 1.3 is availabe here:
https://aur.archlinux.org/packages/catalyst-test

unflagging...

lordxarus commented on 2014-01-30 14:07

Thank you Vi0L0

Vi0L0 commented on 2014-01-30 07:46

I wrote about this on catalyst-utils@AUR:

catalyst doesn't support xorg-server 1.15.
Use [xorg114] repo:
https://wiki.archlinux.org/index.php/AMD_Catalyst#Xorg_repositories

lordxarus commented on 2014-01-29 17:53

I apologize if I am missing something but catalyst requires catalyst-utils, which has xorg-server<1.15.0 as a dependency, and that doesn't exist. Could this be fixed if it's an issue? Or tell me off if I'm being slow :P.

Vi0L0 commented on 2014-01-23 19:13

Ok, i've updated all 13.12 packages without pkgrel bump.

AMD has those updated files:
/etc/ati/amdpcsdb.default
/etc/ati/contro
/etc/ati/signature

After some investigation I will bump pkgrel of catalyst-utils/-total packages.

Vi0L0 commented on 2014-01-23 07:29

Thanks, I should update later today

Shelvak commented on 2014-01-23 00:56

The new catalyst (from amd) have another md5 (I download that 5 times =P)

6b16cad1acf7aa4851fa00e43b9e3c1f amd-catalyst-13.12-linux-x86.x86_64.zip

A hug!

dequis commented on 2014-01-02 12:28

Personally I've added this to /etc/hosts, since the hosting is still alive.

193.107.69.230 catalyst.wirephire.com

(got the IP using the DNS listed here http://who.is/domain-history/wirephire.com )

Vi0L0 commented on 2013-12-30 08:38

It's actually not my domain. Many years ago Apocalypsus gave me the space for the [catalyst] repo project. But I'm unable to do anything with it (except for throwing files there ;)). I've write to Apocalypsus about this problem, be patient. Meanwhile you can use the mirrors, lets quote wiki:

repository mirrors are provided by yanom at [1] (USA) and rtsinformatique at [2] (France). These mirrors however come with no warranty and are not guaranteed to always be operational:
[catalyst]
#Server = http://70.239.162.206/catalyst-mirror/repo/catalyst/$arch
#Server = http://mirror.rts-informatique.fr/archlinux-catalyst/repo/catalyst/$arch

ase1590 commented on 2013-12-29 15:42

looks like your domain name at catalyst.wirephire.com expired today.

Vi0L0 commented on 2013-12-21 22:19

@dun32: good to hear it :)

Vi0L0 commented on 2013-12-21 22:18

General note:
I'm AFK and will so be till next saturday.
In case of problems please write on bar and grill thread. If no one will help and there's some serious bug - please write the bug report on bugzilla.

Meanwhile you can still use archive repos like ones for 13.4 and 13.11-beta94:
[catalyst-stable-13.4] and [catalyst-13.11-beta94]
Server's URLs are like on the wiki for [catalyst].

Or build pkgs on your own basing on archive tarballs (use newest ver):
http://catalyst.wirephire.com/tarball/13.4/
http://catalyst.wirephire.com/tarball/13.11-beta94/

Thanks and merry christmass ;)

dun32 commented on 2013-12-21 13:24

Update to 13.12 solved my problem with low power GPU. Thank ViOLO for your work !

lembang commented on 2013-12-21 06:15

hmm i get this error, i use linux-lqx kernel.

make: Entering directory '/usr/src/linux-3.12.5-1-lqx'
CC [M] /home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.o
/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.c: In function 'KCL_GetEffectiveUid':
/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.c:1758:5: error: incompatible types when returning type 'kuid_t' but 'KCL_TYPE_Uid' was expected
return current_euid();
^
/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.c:1762:1: warning: control reaches end of non-void function [-Wreturn-type]
}
^
scripts/Makefile.build:308: recipe for target '/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.o' failed
make[1]: *** [/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/firegl_public.o] Error 1
Makefile:1228: recipe for target '_module_/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod' failed
make: *** [_module_/home/the/aur/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod] Error 2
make: Leaving directory '/usr/src/linux-3.12.5-1-lqx'

Vi0L0 commented on 2013-12-19 21:46

ahh... + supports up to 3.13 kernel (last test on -rc4)

Vi0L0 commented on 2013-12-19 21:45

13.12-1:
- AMD marked it as stable, so am I...
- http://support.amd.com/en-us/kb-articles/Pages/AMDCatalyst13-12LINReleaseNotes.aspx

dun32 commented on 2013-12-04 07:55

I've created a topic at : https://bbs.archlinux.org/viewtopic.php?pid=1356515#p1356515

You can have a look at what I tried to do but I've still the same problem.

dun32 commented on 2013-12-03 20:21

My virtual machines don't boot because I miss 3D acceleration.
Tried to run amdcccle but it doesn't start anymore.
I don't know when it brooke but I don't have any error appearing when I run it from cli.
What's happening ?

Vi0L0 commented on 2013-11-18 19:35

silent update, no pkgrel bump:
- thanks to zoopp and his referer we no longer have to download installer manually :)

Vi0L0 commented on 2013-11-17 19:55

amd's site is not allowing direct download anymore...
read the top of PKGBUILD for instructions.

Vi0L0 commented on 2013-11-17 19:55

amd's site is not allowing direct download anymore...
read the top of PKGBUILD for instructions.

i also added symlink as sugested by lano1106:
/usr/lib32/libXvBAW.so -> libXvBAW.so.1.0

Vi0L0 commented on 2013-11-08 19:12

13.4-16:
- added workaround to bring support of kernel 3.12, by Krzysztof Kolasa

mojtabazn commented on 2013-10-21 11:35

bettr if you put xorg version the driver support for example this one support until 1.13

idecasso commented on 2013-10-09 04:25

Apparently the driver 13.10.1 Beta has been corrected and no longer presents the memory leak. But more evidence is needed to confirm this.

David-UPCT commented on 2013-09-29 08:41

goviedo, currently stable catalyst driver (13.4) doesn't work with xorg 1.14, check the wiki https://wiki.archlinux.org/index.php/AMD_Catalyst. You can downgrade xorg to 1.13, or you can install catalyst-test instead of catalyst. The 13.8 beta driver in catalyst-test is prepared for xorg 1.14.

goviedo commented on 2013-09-26 23:25

error: target not found: catalyst-utils
==> ERROR: 'pacman' failed to install missing dependencies.

Catalyst-utils from Aur need Xorg < 1.14, and the most recent release has X.Org X Server 1.14.3
Release Date: 2013-09-12

David-UPCT commented on 2013-09-19 20:15

AMD has just downgraded the stable version to 13.4, and beta version remains 13.8

Vi0L0 commented on 2013-09-19 19:22

guys, 13.9 has memory leak, it's not stable at all.
AMD could say otherwise but don't listen to them :P

you will find 13.9 at catalyst-test and catalyst-test-pxp @ AUR,
and maybe tomorrow on the [catalyst] repo

bakgwailo commented on 2013-09-18 19:33

13.9 officially released: http://www2.ati.com/drivers/linux/amd-catalyst-13.9-linux-x86.x86_64.zip

Vi0L0 commented on 2013-09-08 14:59

bumped pkgrel to 13.4-14:
just to change dependencies, because 3.11 kernel is also supported

Vi0L0 commented on 2013-08-17 10:34

Why flagging out of date? This package is aimed for stable releases only, and there was no stable release since 13.4.

Vi0L0 commented on 2013-08-03 13:52

updated to 13.4-13 :
- switching to 3.10 kernel patch of Emil Karlson from gentoo which is genereting correct proc entries

life_enjoyer commented on 2013-07-25 20:12

works now...

life_enjoyer commented on 2013-07-25 16:55

youre the best Vi0L0

life_enjoyer commented on 2013-07-25 16:52

youre the best Vi0L0

Vi0L0 commented on 2013-07-25 15:48

yes, gimme some time

life_enjoyer commented on 2013-07-25 15:46

why doesnt it work with kernel 3.10? does it just need to be updated?

Vi0L0 commented on 2013-06-30 11:54

Yes, I've mailed Apocalypsus who was kind to serve catalyst repositories for all these years.
Right now I'm waiting for his answer. Please be patient and use AUR when necessary

Anonymous comment on 2013-06-29 15:49

Hello,

i receive server out of service because of bandwidth exceed of the owner.

greets

Vi0L0 commented on 2013-06-01 12:18

Please do not flag this out of date.
Beta driver is available on catalyst-test@AUR and [catalyst] repo.
This is a place for the stable driver.

Localizator commented on 2013-05-29 00:13

AMD Catalyst 13.6 Linux Graphics Driver Beta Released http://www.phoronix.com/scan.php?page=news_item&px=MTM4MDQ

Vi0L0 commented on 2013-05-21 15:54

updated to 13.4-3:
- fixes efi support and so ttys bug (thanks to Krzysztof Kolasa :: https://github.com/kolasa)
- fixes support of intel's iommu (thanks to lano1106)

Anonymous comment on 2013-04-27 23:49

I take it we'll have to wait for a version that is compatible with the newer version of xorg-server (1.14.1-1)

Vi0L0 commented on 2013-04-27 22:05

updated to 13.4-1:
http://support.amd.com/us/kbarticles/Pages/AMDCatalyst13-4LINReleaseNotes.aspx

@Tumi:
I've just wrote an email about this, you can always use catalyst-total/catalyst-total-pxp or look at the tarball archive:
http://catalyst.wirephire.com/tarball/

Here's the direct link to the catalyst-utils 13.4 source files:
http://catalyst.wirephire.com/tarball/13.4/catalyst-utils-13.4-1.src.tar.gz

Tumi commented on 2013-04-27 11:08

Hi, guys, catalyst-utils pkg is missing in AUR, where else can I find it?

solarwind commented on 2013-04-25 10:27

Catalyst 13.4 released.

Vi0L0 commented on 2013-03-25 19:04

package has been once again updated, but silently, no pkgrel bump
because AMD has changed files that don't belong to this package, only catalyst-utils/total*

Vi0L0 commented on 2013-03-21 18:50

ahhh... i also cleaned up PKGBUILD

Vi0L0 commented on 2013-03-21 18:49

package has been updated, but silently, no pkgrel
because AMD has changed files that don't belong to this package, only catalyst-utils/total*

Vi0L0 commented on 2013-03-21 10:51

yes, the source has changed recently, amd has updated it, i will check and update after gettin back home.

andmars commented on 2013-03-21 09:06

md5sums is not correct in the PKGBUILD. Here's the right one;

md5sums=('a6f5dc89f7ffa29ee0aec90e665f34bd'
'e27945e920f06aad087a110687b04499'
'3e1b82bd69774ea808da69c983d6a43b'
'ff60c162b46e21e9810a722718023451'
'52a79bddac813f19f72fcb32acebb1b9')

Anonymous comment on 2013-03-11 16:44

Solved. It's amazing how well things work when I use the current, valid repository!

Thank you for your help and thank you for maintaining Catalyst and alternate Xorg, Vi0L0.

Vi0L0 commented on 2013-03-09 00:44

https://bbs.archlinux.org/viewtopic.php?pid=1229999#p1229999

catalyst and xorg repos changed domain/server: was: catalyst.apocalypsus.net , is: catalyst.wirephire.com

after changing pacman.conf do:
pacman -Syy

Anonymous comment on 2013-03-08 23:56

I have been happily using these /etc/pacman.conf repository directives for many months:
[xorg112]
Server = http://catalyst.apocalypsus.net/repo/xorg112/$arch
[catalyst]
Server = http://catalyst.apocalypsus.net/repo/catalyst/$arch

Today, the apocalypsus.net domain appears to be "parked" and the repository inaccessible.

Anyone else? Fixes? Workarounds?

weirddan455 commented on 2013-02-01 23:31

Woops nevermind, found the 13.2 Beta here:

https://aur.archlinux.org/packages/catalyst-test/

weirddan455 commented on 2013-02-01 16:03

Is this going to be updated to 13.2 Beta? Not trying to rush you or anything since it just came out recently but it seems kind of silly to have 13.1 here since it's in the offical [community] repo.

Vi0L0 commented on 2013-01-26 12:51

True for catalyst legacy (also called catalyst-hd234k here), not for this catalyst on which we are commenting

mityukov commented on 2013-01-25 12:50

There's the following warning message on the Wiki page (Catalyst):
--
Warning: The Legacy driver does not support Xorg 1.13 for now. Support should be added by AMD in a century or two, but this could be sooner. Should you want to use this driver, see #Xorg repositories for instructions on how to roll back to or hold back Xorg 1.12.
--

Is it still true for AMD Catalyst™ 13.1?

Vi0L0 commented on 2013-01-18 14:50

updated to 13.1-1:
- changelog is here: http://support.amd.com/us/kbarticles/Pages/AMDCatalyst131ProprietaryLinuxGraphicsDriverReleaseNotes.aspx
- i've added "testing" support for 3.8 kernel (through own patch)

lwjef commented on 2013-01-18 14:14

http://www2.ati.com/drivers/linux/amd-driver-installer-catalyst-13.1-linux-x86.x86_64.zip

ali commented on 2013-01-16 21:25

These are not the newer version? beta11
http://support.amd.com/us/kbarticles/Pages/AMDCatalyst1211betadriver.aspx

btw, last catalyst support amd apu e350?

ali commented on 2013-01-16 21:24

These are not the newer version? beta11
http://support.amd.com/us/kbarticles/Pages/AMDCatalyst1211betadriver.aspx

ali commented on 2013-01-16 21:23

These are not the newer version?
http://support.amd.com/us/kbarticles/Pages/AMDCatalyst1211betadriver.aspx

Vi0L0 commented on 2012-12-15 11:20

silent update (no pkg rel bump):
- added patch that brings 3.7 kernel support

@flrichar: but it's for embedded devices, since now emb devs drivers were useless to us

@Det: indeed, fixed, thanks :>

flrichar commented on 2012-12-12 22:47

As of yesterday, 12.12 is out:

http://support.amd.com/us/gpudownload/embedded/Pages/embedded_linux.aspx

Det commented on 2012-12-02 17:00

There still is no http://www.ati.amd.com.

Anonymous comment on 2012-11-08 23:42

Confirming 12.11 works, even a nifty testing use only image on bottom right. thanks Vi0L0.

Vi0L0 commented on 2012-10-24 15:57

updated to 12.11 beta:
but it's stable,
it's working better with wine and
supports xserver 1.13 (like previous 12.9 ubuntu version)

m1ck3y commented on 2012-10-24 01:28

12.10 is out

Anonymous comment on 2012-10-19 03:45

blank screen exiting X (keyboard works) 3.6.2-1/Radeon 6770 HD -> fglrx(0): Restoring Recent Mode via PCS is not supported in RANDR 1.2 capable environments

Vi0L0 commented on 2012-09-16 17:34

I decided to put 12.9 into catalyst-test @ AUR, becase it's targeted for embedded devices, not well-tested etc. I'm also afraid that the list of supported gpus is short

thirtythreeforty commented on 2012-09-14 15:10

Catalyst 12.9 is out, see http://www.phoronix.com/scan.php?page=news_item&px=MTE4NDk

dom96 commented on 2012-09-03 10:43

I have an AMD 7750 and using the latest Catalyst and Gnome3 goes into fallback mode? Why is that?
https://bbs.archlinux.org/viewtopic.php?id=148174

Vi0L0 commented on 2012-08-26 09:12

HD 540V is a rebrand of the HD4500(55nm) series, so unfortunatelly this driver won't work :-/
check out legacy

JoveYu commented on 2012-08-26 07:22

is this can be installed on HD 540v ??

Vi0L0 commented on 2012-08-19 18:19

indeed :)
will fix it with next release
thanks!

Det commented on 2012-08-19 15:46

There is no http://www.ati.amd.com.

agapito commented on 2012-08-16 11:47

12.8 is out

http://www2.ati.com/drivers/linux/amd-driver-installer-12.8-x86.x86_64.zip

agapito commented on 2012-08-16 09:14

12.8 is out.

antiuser commented on 2012-07-23 02:57

yeah, i found that in ver -4 it was fixed :) thanks

Vi0L0 commented on 2012-07-22 18:38

12.6-4:
updating ati_make.sh, so from now on this package doesn't need /lib at all

Vi0L0 commented on 2012-07-22 18:36

@antiuser:
rebuild package

antiuser commented on 2012-07-22 16:43

[root@lenarch ~]# modprobe -v fglrx
insmod /usr/lib/modules/3.4.6-1-ARCH/extramodules/fglrx.ko.gz
ERROR: could not insert 'fglrx': Exec format error
[root@lenarch ~]# uname -a
Linux lenarch 3.4.6-1-ARCH #1 SMP PREEMPT Fri Jul 20 08:21:26 CEST 2012 x86_64 GNU/Linux
[root@lenarch ~]# pacman -Ss catalyst
catalyst/catalyst 12.6-3 [установлен]
AMD/ATI Catalyst drivers for linux. fglrx kernel module only. Radeons HD 2 3
4 xxx ARE NOT SUPPORTED

Vi0L0 commented on 2012-07-14 19:55

update:
12.6-3 should work fine with glibc 2.16.0-2 and kernel 3.5

Vi0L0 commented on 2012-07-08 20:13

@joveyu:
thanks :)
here's the info where i put legacy: https://bbs.archlinux.org/viewtopic.php?pid=1126975#p1126975

JoveYu commented on 2012-07-07 03:07

AMD Releases New Catalyst Linux Legacy Driver Catalyst 12.6 support Radeons HD 2 3 4 xxx ,support linux3.4 and xorg1.12!

http://www.phoronix.com/scan.php?page=news_item&px=MTEzNDA

Vi0L0,the address :http://www2.ati.com/drivers/legacy/amd-driver-installer-12.6-legacy-x86.x86_64.zip

Anonymous comment on 2012-07-06 20:11

Jomi's fix worked for me. I did not need to extract the .tar.gz archive. Thanks!

Anonymous comment on 2012-07-06 12:12

Here's what I did to get this running with the latest kernel:

# mkdir -p /lib/modules/3.4.4-3-ARCH
# ln -s /usr/lib/modules/3.4.4-3-ARCH/build /lib/modules/3.4.4-3-ARCH/build

$ yaourt -S catalyst
....
....
# mv /lib/modules/3.4.4-3-ARCH/video/fglrx.ko.gz /usr/lib/modules/3.4.4-3-ARCH/
Not sure if you can keep it in it's gz archive or not. I extracted it however.
# depmod
# modprobe fglrx (to see if it works)
# reboot

after reboot check to see if it's loaded:
$ lsmod | grep fglrx

alucryd commented on 2012-07-05 07:39

Building fglrx module no longer works, at least when using testing, due to arch's removal of /lib, now modules are to be compiled in /usr/lib/modules. The catalyst_build_module will have to be patched to look into this directory. Just tried changinge every /lib to /usr/lib but to no avail, and I don't have time to look into it right now. I'll try again later unless someone comes up with a good solution!

dplayle commented on 2012-07-05 07:04

I had the same 'unsupported' error with my Radeon 6310. Fixed it by extracting the /etc/ati/control and /etc/ati/signature files from the previous catalyst package I had built. From this link:

http://forum.xbmc.org/showthread.php?tid=116996&pid=1138227#pid1138227

antiuser commented on 2012-07-03 05:20

Unsupported Hardware Watermark

This can happen if your card's PCI ID wasn't officially certified to work with a particular version of Catalyst. It does not necessarily mean that your card is unsupported, but it does mean that you shouldn't file bugs with that particular card/driver combination. If you installed the driver by downloading it from AMD/ATI, installing a newer version of Catalyst will probably help.

If you installed the proprietary driver included with Ubuntu or you do not want to upgrade to a newer version, it may be possible to work around the issue by using a control file from a newer version of Catalyst than the one you're running.

cd ~/; mkdir catalyst12.6; cd catalyst12.6/
wget http://www2.ati.com/drivers/linux/amd-driver-installer-12-6-x86.x86_64.run
chmod +x amd-driver-installer-12-6-x86.x86_64.run
sh amd-driver-installer-12-6-x86.x86_64.run --extract driver
sudo mv /etc/ati/control ~/control.bak
sudo cp driver/common/etc/ati/control /etc/ati

antiuser commented on 2012-07-03 05:16

Unsupported Hardware Watermark

This can happen if your card's PCI ID wasn't officially certified to work with a particular version of Catalyst. It does not necessarily mean that your card is unsupported, but it does mean that you shouldn't file bugs with that particular card/driver combination. If you installed the driver by downloading it from AMD/ATI, installing a newer version of Catalyst will probably help.

If you installed the proprietary driver included with Ubuntu or you do not want to upgrade to a newer version, it may be possible to work around the issue by using a control file from a newer version of Catalyst than the one you're running.

cd ~/; mkdir catalyst12.6; cd catalyst12.6/
wget http://www2.ati.com/drivers/linux/amd-driver-installer-12-6-x86.x86_64.run
chmod +x amd-driver-installer-12-6-x86.x86_64.run
sh amd-driver-installer-12-6-x86.x86_64.run --extract driver
sudo mv /etc/ati/control ~/control.bak
sudo cp driver/common/etc/ati/control /etc/ati

antiuser commented on 2012-07-03 05:11

# lspci | grep VGA
00:01.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI Wrestler [Radeon HD 6310]

# aticonfig
/opt/bin/aticonfig: No supported adapters detected

sudokode commented on 2012-07-03 04:04

[ 27.523] (EE) fglrx(0): incompatible kernel module detected - HW accelerated OpenGL will not work

Cannot get HW acceleration with 12.6. I'm on a 5770.

Vi0L0 commented on 2012-07-01 18:00

Updated to 12.6:

- supports Radeons HD >= 5xxx ONLY (hd 2 3 4k users please use catalyst-total-hd234k @AUR or [catalyst-hd234k] repo)

- supports 3.4 kernel (through patches);

- supports xorg-server 1.12

Radioactiveman commented on 2012-06-29 01:08

The final version of 12.6 is now available.

Vi0L0 commented on 2012-06-16 12:13

updated to 12.4-6:
got now nice clean patch, 3.4 support for both architectures :)

thx to 3V0LU710N for finding it, and Sebastian Siebert from OpenSUSE for a patch

Vi0L0 commented on 2012-06-16 12:11

Here: https://wiki.archlinux.org/index.php/Pacman-key#Unofficial_keys

Vi0L0 commented on 2012-06-16 12:10

Here: https://wiki.archlinux.org/index.php/Pacman-key#Unofficial_keys

Neije commented on 2012-06-16 11:49

Hey Violo,

Am confused, I just tried to update my system and I get the following message :
Erreur : catalyst-utils: signature from "Vi0L0 <vi0l093@gmail.com>" is unknown trust
Erreur : catalyst: signature from "Vi0L0 <vi0l093@gmail.com>" is unknown trust
Erreur : lib32-catalyst-utils: signature from "Vi0L0 <vi0l093@gmail.com>" is unknown trust

And I am quite sure I have seen pacman dowloading your key.

Did I miss anything and I ahve to configure something manually ?

Cheers !

Vi0L0 commented on 2012-06-12 17:04

All radeon hd 2xxx 3xxx 4xxx owners please read this:
https://bbs.archlinux.org/viewtopic.php?pid=1115031#p1115031

Vi0L0 commented on 2012-06-10 20:33

yes, but unfortunatelly i don't think that we can apply kernel header' patch in fakeroot... so i'm unable to provide this patch with this package


i686 users please use generator or hook or daemon for (kinda dirty) 3.4 kernel support

thanks again nob :)

Vi0L0 commented on 2012-06-10 20:00

yes, but unfortunatelly i don't think that we can apply kernel header' patch in fakeroot... so i'm unable to provide this patch with this package


i686 users please use generator or hook or daemon for (kinda dirty) 3.4 kernel support

thanks again nob :)

Treens commented on 2012-06-08 21:26

For the 3.4 compilation, i gave patches in catalyst-test package.

Vi0L0 commented on 2012-06-03 22:38

12.4-4:
- because of a compilation error on i686 with 3.4 kernel i decided to stay with linux<3.4 dependency

x86_64 is able to work with 3.4 kernel and provided 3.4.patch

Vi0L0 commented on 2012-06-03 22:02

updated to 12.4-3:
- support of 3.4 kernel is now provided by other patch found here: http://devgurus.amd.com/thread/159241
clean and kinda official :)

Vi0L0 commented on 2012-06-03 19:51

To provide dirty dirty 3.4 kernel support we have to patch kernel's source file (and restore original source file after fglrx module compilation).

It's impossible to do this inside PKGBUILD, chroot etc - that's why this package won't get 3.4 kernel support, please use generator or hook or daemon

Vi0L0 commented on 2012-06-02 01:30

@markuznw: look here: https://aur.archlinux.org/packages.php?ID=34773

Anonymous comment on 2012-05-31 21:06

Since amd is not providing anymore monthly catalyst (they will follow nvidia model of X betas) could you please start track the betas? :) http://support.amd.com/us/kbarticles/Pages/AMDCatalyst126beta.aspx out today :)

Vi0L0 commented on 2012-05-26 17:54

BTW: since your request doesn't refer to this catalyst package i suggest to continue our conversation on aur's catalyst-hook page, thank you

Vi0L0 commented on 2012-05-26 17:50

"Nobody expect spanish inquisition!" :) http://www.youtube.com/watch?v=Tym0MObFpTI

Hundreds of AUR and repo users are using catalyst-hook and catalyst-daemon packages and they like it.
They can use catalyst-generator which is AUR-way compatible but still they preffer packages mentioned above.

Afaik till now only i was mentioning that those packages aren't compatible with ARCH philosophy, i modified hook some time ago so it would not add itself into system's files but then users asked me to make it automagic. And i though "why not"? Users are informed about that fact in package's postinstall infos and on the wiki page. Till now no user had problems with it.

Till now (so after more than 2 years) noone except you was asking me to change it, and since i believe that we should be as much democratic as we can - i'm not gonna remove this functionality till more users would ask me to do so.

seblu commented on 2012-05-22 17:55

As explain in virtualbox-hook package, install script must not update mkinitcpio.conf and pacman.conf! Please fix this your package.

Anonymous comment on 2012-05-18 00:03

not sure if it's worth it from a stability standpoint, was just hoping to get Xorg 1.12 support :)

Vi0L0 commented on 2012-05-14 22:04

patch is a member of base-devel group which have to be installed by AUR users, that's why there's no need to add this particular dependency.

Is leaked 12.5 worth it? I heard not

Anonymous comment on 2012-05-12 10:58

Please add 'patch' to MAKEDEPENDS. And yes, you still need [xorg111] to be able to use proprietary Catalyst driver.

Anonymous comment on 2012-05-08 19:33

Hi Vi0L0, any chance we might see the leaked Catalyst 12.5 Beta on here?

walkindude commented on 2012-05-07 08:27

no, it's not possible

hsn commented on 2012-05-05 17:20

Is it possible to build this with xorg-server 1.12.1?

kallisti5 commented on 2012-04-30 14:06

Hmm... 12.4 results in barf-tastic black screen flashes on screen update on my Radeon HD 7970. I wouldn't update if you have a 7970 series card.

Vi0L0 commented on 2012-04-25 23:26

Unfortunatelly no

Anonymous comment on 2012-04-25 17:57

12.4 is out, but works it with xserver 1.12?

Anonymous comment on 2012-04-25 11:51

The 12.4 driver is released. :)

http://support.amd.com/us/gpudownload/embedded/Pages/embedded_linux.aspx

Inkaine commented on 2012-04-18 16:28

No, xorg 1.12 support most likely won't arrive before catalyst 12.5.

My comment from 11 Apr was on a (apparently now deleted) comment by someone: "It doesn't work with latest kernel".

Neije commented on 2012-04-17 21:02

Well, I tried. It doesn't work cause catalyst-utils requires x.org < 1.12

abique commented on 2012-04-17 06:48

I believe that yes.

Neije commented on 2012-04-17 06:26

I'm lost now; So can I (or not) remove my xorgg111 repo ? does the latest version works on 1.12 ?

Inkaine commented on 2012-04-11 12:52

It does!

Anonymous comment on 2012-03-31 06:40

Here the same. :(

Anonymous comment on 2012-03-31 05:06

It doesn't work on xorg-server 1.12.0

Anonymous comment on 2012-03-28 15:45

Catalyst 12.3 is out: http://www.computerbase.de/downloads/treiber/grafik/amd-catalyst-proprietary-linux-driver/

Huulivoide commented on 2012-03-28 15:44

exaos: Read my last comment and do the same

Anonymous comment on 2012-03-28 15:43

Try to install catalyst to solve hybrid graphics problem as mentioned here:
http://ubuntuforums.org/showthread.php?t=1930450
Unfortunately, I got:
catalyst needs xorg-server <1.12.0
hybrid graphics needs catalyst 12.2 + xf86-video-intel > 2.17 (current: 2.18.0-1) with SNA
xf86-video-intel needs xorg-server >= 1.12.0
How to solve such an issue?

Huulivoide commented on 2012-03-28 14:07

exaos you need xorg 1.11. Go to the catalyst archwikipage and
enable the [xorg111] repo as instructed on there.

Roken commented on 2012-03-27 01:00

Thanks guys - pretty much fully updated and working now - just need to wait for linux-ck to hit 3.3 and for libaudiofile to get fixed, but at least I've got the big ones (catalyst, kernel and VB) working together again :)

Vi0L0 commented on 2012-03-25 19:58

@Roken: im usually bumping version whenever new kernel is hitting [core] since i'm almost always stable user, but here you go :)
yes, it's safe

Huulivoide commented on 2012-03-25 18:23

Roken: Use the [xorg111] repo, no need to hold back xorg updates.
You can also update virtualbox, it includes vboxbuild command,
which builds the virtualbox host modules for yourt current kernel.

Roken commented on 2012-03-25 18:21

12.2- is it safe to change the PKGBUILD to remove the dependency on linux<3.3, or will 12.3 be fixed to support the newer kernel. I'm currently holding back a bucket load of updates now (xorg, kernel 3.3, virtualbox that depends on 3.3 kernel etc) waiting for catalyst to catch up.

OvsInc commented on 2012-03-25 07:05

2 caasi:
Perhaps the problem with this?
...
1626.072] (--) fglrx(0): Chipset: "AMD Radeon HD 6500M/5600/5700 Series" (Chipset = 0x68c1)
[ 1626.072] (--) fglrx(0): (PciSubVendor = 0x17aa, PciSubDevice = 0x3951)
[ 1626.072] (==) fglrx(0): board vendor info: third party graphics adapter - NOT original ATI
...
and we have this
[ 1626.124] (II) fglrx(0): Invalid ATI BIOS from int10, the adapter is not VGA-enabled

Anonymous comment on 2012-03-23 21:37

Has anyone tried the 12.04 beta drivers with OpenCL?
http://developer.amd.com/Downloads/OpenCL1.2betadriversLinux.tgz

Anonymous comment on 2012-03-21 14:33

when I run `sudo aticonfig --initial`
I got this error:

PowerXpress error: Cannot stat '/usr/lib/fglrx/switchlibGL': No such file or directory
Failed to initialize libglx for discrete GPU

and here ( http://en.gentoo-wiki.com/wiki/Fglrx-hybrid-graphics ) I see the likely problem in Gentoo. But in Arch, there's a little difference because when I install the `catalyst-utils`, the `libGL`must be removed first. So the script may not be useful. Anyone can help me ?

This is my Xorg.0.log http://hpaste.org/65657
This is my xorg.conf http://hpaste.org/65655
BTW. Now, If the Xorg.conf exists. I can't boot successfully ( ERROR: Unable to find root device '/dev/disk/by-uid/03619a2f-908f-4724-ba4b-1fb3985e382b' You are being dropped to a recovery shell Type 'exit' to try and continue booting sh: can't access tty; job control turned off ). I have to manually run `modprobe ahci`,`modprobe sd_mod` Only in this way can I boot.
Thanks.

Vi0L0 commented on 2012-03-07 19:25

~~> 12.2:
- xv is fixed when using xserver 1.11

alucryd commented on 2012-03-04 20:31

Downgraded to 1.11.4 and now it works fine, thx for the help guys.

alucryd commented on 2012-03-04 19:58

Hmm you must be right, I'll try downgrading. Thx!

walkindude commented on 2012-03-04 19:34

to Alucryd:
I think that it's because of xserver 1.11.99.903 (as it says in your Xorg.0.log). Tried with it when it appeared in testing, got blacksreen, stayed with 1.11.4.

walkindude commented on 2012-03-04 19:25

to Alucryd:
I think, that it's because of xserver 1.11.99.903 (as it says in your Xorg.0.log). Tried with it when it appeared in testing, got blacksreen, stayed with 1.11.4.

alucryd commented on 2012-03-04 18:19

Manually removing and rebuilding didn't work for me. Here are my xorg.conf (generated by aticonfig --initial) and my Xorg.0.log.
http://pastebin.com/XEAaTURZ
http://pastebin.com/R3GMLEtn
ArchLinux is up to date, with testing enabled, and I'm using linux-ck-corex. It would seem the xorg.conf is at fault, aticonfig does not do it right. I'll try changing the BusID.

walkindude commented on 2012-03-04 15:15

now it's working on 3.2.9 after manually removing fglrx from extramodules directory and compiling again

walkindude commented on 2012-03-04 15:07

now it's working on 3.2.9 after manually removing fglrx from extramodules directory and compiling again

Vi0L0 commented on 2012-03-04 14:08

on 3.2.9 x86_64 from testing fglrx is compiling and working fine on both kdm and gdm here

can you guys put your Xorg.0.log and xorg.conf on some pastebin server?

walkindude commented on 2012-03-04 07:38

Yeah, it's compiling on both 3.2.8 and 3.2.9, but actually works only with the first one, with 3.2.9, when trying to load fglrx error happens: 'Exec format error' or in dmesg: 'disagrees about version of symbol module_layout'.
Using linux-ck, gnome-shell

walkindude commented on 2012-03-04 07:24

Yeah, it's compiling on both 3.2.8 and 3.2.9, but actually works only with the first one, with 3.2.9, when trying to load fglrx error happens: 'Exec format error' or in dmesg: 'disagrees about version of symbol module_layout'.
Using linux-ck, gnome-shell

alucryd commented on 2012-03-04 00:57

Well it is compiling here on x86_64 (package and kernel module), but when X tries to load gdm, I've got nothing but a black screen (happens with 3.2.8 too, tested on 2 computers after fresh install). I may still have 3.2.7 on my laptop, but I guess I'll wait for catalyst 12.2 and hope it works.

Vi0L0 commented on 2012-03-03 10:53

@Alucryd: on virtual i686 machine i can only see that it's compiling on 3.2.9.
Isn't rebuilding of package a solution?

alucryd commented on 2012-03-03 01:11

Just wondering, does the 3.2.8 patch also fixes 3.2.9? (Doesn't seem to work here)

Huulivoide commented on 2012-02-24 19:40

A custom kernel is a even more better reasn to use catalyst-hook

ttux commented on 2012-02-24 19:39

I use a custom kernel so that was not the problem. I just re-switched from radeon to fglrx now.

Huulivoide commented on 2012-02-24 19:28

XD I tought this was a mailinglist conversation :D so I mailed you ttux there :D
Il repost my answe here too anyway :DD

You need to rebuild the fglrx module again evrynow and then (after each kernel
update, well only after 3.x.y x-version changes, but if you wanna play safe
after each update)

To easy this job install catalyst-hook pkg that will automatically rebuild the
module every time you update your initrd (this gets done automatically after
every kernel update)

ttux commented on 2012-02-24 19:26

problem solved, I ran src/archive_files/common/lib/modules/fglrx/build_mod/make.sh and then src/archive_files/common/lib/modules/fglrx/make_install.sh and copied the fglrx.ko manually in /lib/modules/3.2.6/kernel/drivers/video/ and problem is sorted.

ttux commented on 2012-02-24 19:13

I thought I would give catalyst another try but I'm facing a strange problem. The install of catalyst and catalyst-utils went just fine but I can't load fglrx module (FATAL: Module fglrx not found) so I checked /lib/modules/3.2.6/kernel/drivers/video/ and I don't have fglrx.ko in there. I have it in src/archive_files/common/lib/modules/fglrx/build_mod/ where I built catalyst with makepkg. Any idea what went wrong?

Anonymous comment on 2012-02-08 19:56

does somebody know where to get the /usr/lib/fglrx/switchlibGl scripts ?

Anonymous comment on 2012-02-07 21:13

@Vi0L0:
We are considering removing /lib/modprobe.d/framebuffer_blacklist.conf from the udev package. This might cause problems for catalyst. Please consider adding /lib/modprobe.d/catalyst.conf, with "blacklist" entries for whatever modules you rely on being blacklisted.

Anonymous comment on 2012-01-29 00:16

No problem, only clinfo segmentation fault

Vi0L0 commented on 2012-01-27 16:53

Laggs/slow windows movement in GNOME3 ? Check out "export CLUTTER_VBLANK=none":
https://wiki.archlinux.org/index.php/ATI_Catalyst#Laggs.2Fslow_windows_movement_in_GNOME3

Vi0L0 commented on 2012-01-27 16:33

@mrspin: thanks, catalyst-utils was updated, although without pkgrel bump

Vi0L0 commented on 2012-01-27 16:24

yes, xv is fine with [xorg110] repo, also pls read lib32-catalyst-utils comment :>

dotmagic commented on 2012-01-27 11:08

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=649346

> This is OT, but to be exactly, you may thank X11, which breaks again
> again and again the ABI (where fglrx has to be adjusted) and now this
> bug, which only happens with this new ABI bling bling driver.

That mean with Xorg 1.10 and catalyst 12.1 xvideo works? Someone tried it? I still search for the best way to downgrade.

mrspin commented on 2012-01-27 00:12

/usr/bin/amdupdaterandrconfig seems to be a script that gets installed but isn't given execute permissions

Anonymous comment on 2012-01-26 13:38

With 12.1 gnome-shell works flawlessly (so far no artifacts). But it is much slower than fallback.

agapito commented on 2012-01-26 04:52

Downgrade to 1.10 xserver is the solution.

CPUnltd commented on 2012-01-26 00:13

is there a workaround? I can't use minitube, Cheese, xawtv or anything with my webcam and it's the most annoying thing in the world right now...

Vi0L0 commented on 2012-01-25 23:18

I would tell you if it would be fixed, so yup - its still crashing X. You can find confirmations from other people @ phoronix forums

sudokode commented on 2012-01-25 23:13

Can anyone confirm if xvideo is fixed? It was broken in 11.11 and 11.12. Anything will work, tvtime or mplayer -vo xv. If still broken, it should just crash X.

andmars commented on 2012-01-25 22:14

same sh... different day...x crashes when playing videos (xv-output), flash-videos in fullscreen stutter, hang, lag (but ran smoothly with 11.11)...how can AMD keep theses things unfixed for so long? I'm really disapointed because I thought spending 200€ on my HD 6950 was a good descission...well, not with Linux

Vi0L0 commented on 2012-01-25 22:00

updated to 12.1:
- no changes.............
- well just a little ones, but not so important
http://www.phoronix.com/scan.php?page=news_item&px=MTA0ODM

agapito commented on 2012-01-25 19:43

http://www2.ati.com/drivers/linux/amd-driver-installer-12-1-x86.x86_64.run

Huulivoide commented on 2012-01-14 12:29

ali: there was a message in mailing list about 3.1.9-1 bring broken
Upgrae to 3.1.9-2, to fix that.

ali commented on 2012-01-14 12:13

I have problem http://pastebin.com/XMjg6Abb

ttux commented on 2011-12-13 18:54

me too, especially after seeing http://code.google.com/p/chromium/issues/detail?id=101225...

mrspin commented on 2011-12-13 18:52

I get a conflict with /usr/bin/clinfo so it conflicts with the amdstream package (at least 2.5)

Vi0L0 commented on 2011-12-13 18:24

updated to 11.12:
- nothing new, just nothing, 80 files were changed when comparing to 11.11, but it feels EXACTLY the same...
- GNOME3 support sux,
- xv video output causes X to crash...
and where is GL_ARB_robustness extension?

Darn, i'm sooo disappointed with these release :/

Vi0L0 commented on 2011-12-05 19:08

update 11.11-1 => 11.11-2:
recommended, https://bbs.archlinux.org/viewtopic.php?pid=1025077#p1025077

theodore commented on 2011-11-20 12:35

yeap there is a problem with xv output, you can use instead gl, gl2 or vaapi

https://bbs.archlinux.org/viewtopic.php?pid=1018387#p1018387

theodore commented on 2011-11-20 12:35

yeap there is a problem with xv output, you can use instead gl, gl2 or vaapi

https://bbs.archlinux.org/viewtopic.php?pid=1018387#p1018387

Anonymous comment on 2011-11-20 06:37

it looks like xorg-server 1.11 problem.

when downgrade to 1.10, video output of xv is nomal

Anonymous comment on 2011-11-20 05:53

bug: use mplayer could make screen black and restart kdm

theodore commented on 2011-11-19 15:46

@update

with catalyst-utils 11.11-3 and the removal of libcl problem solved ;-)
thanks @ViOLO

theodore commented on 2011-11-19 15:28

i used the package from the repository
actually when i tried to update pacman was complaining that

/usr/bin/clinfo
/usr/lib/libOpenCL.so
/usr/lib/libOpenCL.so.1

were already in my system so i removed them and then update proceeded successfully but at the end i took ldconfig complaining

ldconfig: /usr/lib/libOpenCL.so.1 is not a symbolic link

then i uninstall both catalyst-daemon and catalyst-utils and i install them from scratch, when i uninstalled the previous packages i checked out and i saw also that clinfo, libOpenCL.so and libOpenCL.so.1 were not in my system anymore

however after the reinstallation ldconfig was still complaining, any idea?

theodore commented on 2011-11-19 15:21

i used the package from the repository
actually when i tried to update pacman was complaining that

/usr/bin/clinfo
/usr/lib/libOpenCL.so
/usr/lib/libOpenCL.so.1

were already in my system so i removed them and then update proceeded successfully but at the end i took ldconfig complaining

ldconfig: /usr/lib/libOpenCL.so.1 is not a symbolic link

then i uninstall both catalyst-daemon and catalyst-utils and i install them from scratch, when i uninstalled the previous packages i checked and i saw also that clinfo, libOpenCL.so and libOpenCL.so.1 were not in my system anymore

however after the reinstallation ldconfig was still complaining, any idea?

theodore commented on 2011-11-19 15:15

yeap i did it with makepkg

actually when i tried to update pacman was complaining that

/usr/bin/clinfo
/usr/lib/libOpenCL.so
/usr/lib/libOpenCL.so.1

were already in my system so i removed them and then i took ldconfig complaining

ldconfig: /usr/lib/libOpenCL.so.1 is not a symbolic link

then i uninstall both catalyst-daemon and catalyst-utils and i install them from scratch, when i uninstalled the previous packages i checked and i saw also that clinfo, libOpenCL.so and libOpenCL.so.1 were not in my system anymore

however after the reinstallation ldconfig was still complaining, any idea?

Vi0L0 commented on 2011-11-19 12:22

@theodore: strange error, have you used makepkg or some aur manager like yaourt to build a package?

theodore commented on 2011-11-19 00:21

ldconfig is complaining about libOpenCl

ldconfig: /usr/lib/libOpenCL.so.1 is not a symbolic link

should i change something manually or it is normal

theodore commented on 2011-11-18 23:45

ldconfig is complaining about libOpenCl

ldconfig: /usr/lib/libOpenCL.so.1 is not a symbolic link

should i change something manually or it is normal

CPUnltd commented on 2011-11-18 00:20

will be trying out on my laptops today!

Vi0L0 commented on 2011-11-18 00:11

updated to 11.11:
more info here: https://bbs.archlinux.org/viewtopic.php?pid=1018105#p1018105

mrspin commented on 2011-11-17 12:40

I had an issue with libOpenCL conflicting since it's included now.

Anonymous comment on 2011-11-17 06:41

I managed to install Catalyst 11.11 and Catalyst-utils 11.11 by downloading the AUR tarballs and editing PKGBUILDs.

All I had to do was change the pkgver variable from 11.10 to 11.11 and amend the MD5 hashes in the array (first entry) to 7196077aacfbaf7234a705f5352a1111

Hope this helps some people wanting to move to the latest xorg server while we patiently wait for Vi0L0's return :)

Vi0L0 commented on 2011-11-16 21:46

aaaaah such a fast update! sorry guys and lads - i'm on company's delegation, hundreds miles away from my pc... hope to get soon like tomorrow, till then there will be no update :(

sleepforlife commented on 2011-11-16 11:00

update?

ttux commented on 2011-11-16 08:16

I confirm 11.11 works with kernel 3.2 and xorg 1.11. Performance is better on gnome-shell as well and the problem I had with the cursor not visible is solved.

Anonymous comment on 2011-11-16 04:07

AMD Catalyst 11.11 was released.

Anonymous comment on 2011-11-15 21:32

That leads to a 404. Below is the correct one

http://www2.ati.com/drivers/linux/ati-driver-installer-11-11-x86.x86_64.run

ChemBro commented on 2011-11-15 20:28

New driver. Some say, it has xorg 1.11 support and kernel 3.1 and 3.2 support.

http://www2.ati.com/drivers/linux/amd-driver-installer-11-11-x86.x86_64.run

Vi0L0 commented on 2011-11-12 01:53

another update:
https://bbs.archlinux.org/viewtopic.php?pid=1015554#p1015554

Vi0L0 commented on 2011-11-09 19:10

important update:
https://bbs.archlinux.org/viewtopic.php?pid=1014475#p1014475

mbostwick commented on 2011-11-09 02:31

http://henryhermawan.blogspot.com/2011/06/ati-catalyst-proprietary-display-driver.html
seems like this needs the patches from above so that linux 3.0 can run

CPUnltd commented on 2011-11-06 20:08

also, wouldn't you be better off just adding xorg to the ignore list in pacman.conf? Seems like a lot less work than telling your pkgmgr of choice to ignore a list of specific files... That's what I did, and every manager I've used respects that ignore list.

BlackIkeEagle commented on 2011-11-06 10:24

@Ccy :: maybe you could read the comments before yours,
indeed you cant use xorg-server > 1.10, but there were some solutions offered which make your life a lot easier
and you still can use the latest catalyst and its utils

CyrIng commented on 2011-11-06 10:10

Still can't upgrade Xorg with this release.
To full upgrade others packages, think about :
yaourt -Syu --ignore xorg-server --ignore xorg-server-common --ignore xorg-xdm --ignore xorg-xinit --ignore xf86-input-evdev --ignore catalyst --ignore catalyst-utils

Neije commented on 2011-11-05 20:50

Thanks violo : this was exactly my pb. I didn't notice the order did matter and added the line at the end of pacman.conf.

Update is ongoing but 2.4 Gb is heavy :-(

Vi0L0 commented on 2011-11-05 14:32

just a side note: packages from [xorg*] repositories were not build by me, they are just archived from [extra] repo

Vi0L0 commented on 2011-11-05 14:17

@Neije: be sure that you added [xorg110] above all other repositories in pacman.conf, then pacman -Syu

transmutated commented on 2011-11-05 11:54

@Neije,
You need to downgrade Xorg to 1.10. Read this:
https://wiki.archlinux.org/index.php/Downgrading_Packages

I had to do the same thing.

Neije commented on 2011-11-05 10:05

Well it doesn't work : did I miss something ?

# pacman -Su
:: Début de la mise à jour complète du système...
résolution des dépendances...
Avertissement: cycle de dépendances détecté:
Avertissement: lib32-gcc-libs sera installé avant sa dépendance gcc-libs-multilib
recherche des conflits entre paquets...
Erreur: la préparation de la transaction a échoué (la satisfaction des dépendances a échoué)
:: catalyst-utils: requiert xorg-server<1.11.0

BlackIkeEagle commented on 2011-11-05 07:36

@Neije:
add the following to your pacman.conf, packages are provided by ViOLO and served to us by atanas

[xorg110]
Server = http://catalyst.apocalypsus.net/repo/xorg110/$arch

BlackIkeEagle commented on 2011-11-05 07:35

@Neije:
add the following to your pacman.conf, packages are provided by ViOLO and served to us by atanas

[xorg110]
Server = http://catalyst.apocalypsus.net/repo/xorg110/i686

Neije commented on 2011-11-05 00:41

So I'm a bit lost now. I was waiting for this version to launch a pacman -Syu but it seems it still not compatible ? So how to make the update correctly ? I need now to update arch but I also need my catalyst working.

Vi0L0 commented on 2011-11-05 00:06

but that's not GNOME3, only Unity which is basing on gnome 2.x iirc

Anonymous comment on 2011-11-04 23:56

I have uploaded a lil video about performance of catalyst 11.10 in Ubuntu 11.10.This is the link:http://youtu.be/3TE_F5BKAsM.

Vi0L0 commented on 2011-11-03 22:35

There's no ubuntu's patch for fglrx neither here http://phoronix.com/forums/forumdisplay.php?53-Catalyst-Packaging-Script-Changes/page2&order=desc
nor inside fresh 11.10 catalyst. At least quick search didn't refuse any result.
Could it be that ubuntu is somehow patching their GNOME3? I will look at it when i get some more time (i'm and was very busy lately).
Maybe someone could look at it in a spare-time?

CPUnltd commented on 2011-11-03 15:50

curious, gigi17/dongfanglu... was this with the newest xorg, or a previous version?

CPUnltd commented on 2011-11-03 15:49

curious, gigi17/dongfanglu... was this with the newest xorg, or a previous version?

Anonymous comment on 2011-11-03 08:52

I can confirm also that catalyst 11.10 works in Ubuntu 11.10 without any problem.Experience with arch wasnt even close to that.

Anonymous comment on 2011-11-03 03:15

I finded at here:http://forum.ubuntu.org.cn/viewtopic.php?f=42&t=350772

I tried ubuntu 11.10, install gnome-shell and catalyst 11.10. The effect was much better than archlinux.

ttux commented on 2011-11-02 08:14

Sorry but I don't see where do you see the catalyst driver works perfectly with gnome-shell in the link you gave?

I talked with a gnome developer and the problem with gnome-shell/catalyst is clearly in AMD hands so I don't see what ubuntu could do there. With the way it looks with catalyst 11.10, I think it will be fine with 11.11.

You can see how gnome-shell works for me with the catalyst driver here: http://inebium.com/post/catalyst-11-10-gnome-shell-still-not-usable

gnome-shell works flawlessly with the open source one (radeon) but the open source driver has less good performance and more importantly, it doesn't support opencl.

Anonymous comment on 2011-11-02 06:05

In ubuntu, the gnome3 problem was solved almost perfect.Please see as follows:
http://wiki.cchtml.com/index.php/Ubuntu_Oneiric_Installation_Guide#The_Options

So when archlinux?

le_fuzzy commented on 2011-10-31 21:25

So is 11.10 worth upgrading over 11.8?

Vi0L0 commented on 2011-10-31 19:55

updated to 11.10,

you already know everyting, summary:
- no xserver 1.11 support;
- GNOME3 is better, but still lagging.

ttux commented on 2011-10-31 19:11

much better, no glitches for me but still not usable as I can't see my mouse and it is not smooth (to not say sluggish). it didn't crash also which is another + compared to 11.9.

kidoz commented on 2011-10-31 19:06

@ali
No.

Huulivoide commented on 2011-10-31 18:31

Gnome is usable in 11.10, but still has some graphical glitches
, in scroll bar and tooltips. Some times some quick ending full-
screen corruption. In my opininion gnome is much faster with
the open driver on hd4850.

ali commented on 2011-10-31 16:42

works on xorg 1.11?

sonny commented on 2011-10-31 16:34

catalyst 11.10 is out

Huulivoide commented on 2011-10-27 14:44

windows driver will be avaiable on 31th, dunno about
linux one tought.

Anonymous comment on 2011-10-27 13:00

Hi, When the new Catalyst 11.10 will be released ?
I hope they fix gnome-shell bugs !

Anonymous comment on 2011-10-21 10:59

Oh, indeed I downgraded recently and window resizing stopped making kde crash. I didn't pay attention to that. Now I understand and agree - 11.9 sux :)

ttux commented on 2011-10-21 10:11

Because as Vi0l0 said in the comments:
Comment by: Vi0L0 on Thu, 29 Sep 2011 18:24:20 +0000
11.9 sux ;P
downgraded to 11.8 :)

He also said how you can get it if you want:
11.9 is available:
- as PKGBUILDs here: http://catalyst.apocalypsus.net/tarball/11.9/
- and in a repo, with pacman's entry:
[catalyst-11.9]
Server = http://catalyst.apocalypsus.net/repo/catalyst/$arch

Anonymous comment on 2011-10-21 10:09

Why 11.9 isn't here?

Vi0L0 commented on 2011-10-12 15:38

catalyst isn't working with fresh xorg-server 1.11
use [xorg110] repo: https://wiki.archlinux.org/index.php/ATI_Catalyst#Xorg-server_repositories

Vi0L0 commented on 2011-10-03 21:33

just a side note:

11.9 is available:
- as PKGBUILDs here: http://catalyst.apocalypsus.net/tarball/11.9/
- and in a repo, with pacman's entry:
[catalyst-11.9]
Server = http://catalyst.apocalypsus.net/repo/catalyst/$arch

silvernode commented on 2011-10-03 21:20

So does this mean 11.8 is going to be the version available in this page? I know people are saying 11.9 sucks but I'd rather not try it out using the amd driver support site.

Anonymous comment on 2011-10-03 05:32

@shaurz, if you just had checked the comments..

shaurz commented on 2011-10-03 02:41

11.9 is out.

Anonymous comment on 2011-10-02 05:52

AMD promised better gnome-shell support in 11.9

Anonymous comment on 2011-09-30 08:09

You are all being too hard on AMD, they have been working hard on this for 3 months!

ttux commented on 2011-09-30 07:17

I wrote a rant here http://inebium.com/post/catalyst-11-9-gnome-shell-broken, feel free to pass your anger in the comments. I followed the "Linux Crew Driver Feedback" also but I think it's for nothing. Did someone ever get feedback from this?

Deusdies commented on 2011-09-30 05:53

Downgrading to 11.8 as well. I had contstant random xorg crashes with 11.9. ATI, is it MAYBE possible that you release drivers that do fix a thing, but don't break another thing?!

Vi0L0 commented on 2011-09-29 18:24

11.9 sux ;P
downgraded to 11.8 :)

ttux commented on 2011-09-29 18:21

they didn't fix gnome3 support, it's still all broken: slow, flickering, garbage and crash after while.

Fabioamd87 commented on 2011-09-29 06:24

when I started the pc after installed this driver I saw litte characters in archlinux loading procedure, this happen usually when I'm on OpenSource driver because of the KMS.

This mean that 10.9 support KERNEL MODE SETTING?

Anonymous comment on 2011-09-28 18:28

I updated to 11.9-1 and all of a sudden my screen starts tearing. I can't enable "Tear Free Desktop" anymore because I have insufficient video RAM... I have 1GB of video ram so that can't really be the case. Anyone else experiencing this problem?

Vi0L0 commented on 2011-09-28 15:45

updated to 11.9

haaah - i don't see rainbow bar in GNOME3 although other bad things stayed, like slow window moving or strange flickering/garbages rendered whenever i will run kde app with systray icon...

_odd thing_ - the URL has changed - was: ati-driver-installer-11-9-x86.x86_64.run , is ati-driver-installer-11.9.x86.x86_64.run, so - is now . in version, i actually found this 11.9 driver because of misstype :) LOL

Vi0L0 commented on 2011-09-28 14:43

11.9 is online, downloading now, testing and aur update later
im curious does they fixed GNOME3 support :>

Vi0L0 commented on 2011-09-24 23:31

@Neije: if you think that some information could be usefull then feel free to write it down, thats what wiki is for, i believe :)

Neije commented on 2011-09-24 07:55

Thanks for this explanation. I will try the repo method then for my next update !

@all : I have added a dual screen entry in the french wiki for catalyst page. Do you think it is also needed in the english one ? => it's a very nice way to recycle an old 1280x1024 screen ;-)

Vi0L0 commented on 2011-09-23 20:44

Well... im preparing all catalyst' packages for both AUR and [catalyst] repo so, umm... im not using neither AUR nor repo nor installer ;)
But if i were a user i would pick up [catalyst] repo - just because it's rly fastest way and provides packages with same functionality as AUR ones.

BTW: Security is a primary thing for me - so as a maintainer im always trying to prepare repo packages in most secure way - ie. im compiling those on virtual machines which i power on only when i need to. So yeah - i believe the repo is fine.

Neije commented on 2011-09-23 20:03

Wow, you did rework the wiki quite a lot. I read it through and it's really better now. The introduction makes all the rest clearer.

Just out of curiosity : what method are you using ?

Vi0L0 commented on 2011-09-23 19:41

@Neije: thanks for your 2 cents, those have been very helpfull. I just changed wiki, believe it is better now :)

Anonymous comment on 2011-09-22 14:47

I always use catalyst repo i have 0 problem with it. Just add repo install catalyst,catalyst-utils,add nomodeset to /boot/grub/menu.lst reboot ur done.

Neije commented on 2011-09-20 13:13

@violo

I don't know about the <code> tags but some things bother me on the wiki:
- With every Catalyst update you should remove amdpcsdb file => why ? I never do it and I experience no pb. if you remove you catalyst package, is this file removed also ? this could explain why ...
- Note: Catalyst >= 10.10 build supports only xorg-server 1.9 => not true. Latest version build with xorg 10.10. Sentence is not clear.
- Whether or not you install from AUR, you will have to reinstall/update fglrx every time the kernel is updated. Otherwise, X will fail to start. => here it means, you have to reboot after install and install from tty ... or you have to use the auto-compile and/or generator tools. It should be explained clearly from here.
- Note: Since version 10.9, Catalyst is split into catalyst and catalyst-utils ==> and lib-32-catalyst ... is important and not so visible afterwards in the text.
- finally, from the 3 install methods (AUR, Repo, Official), which one for which situation? + which differences ? A hint would be great for beginners.

my 2 cents.

Vi0L0 commented on 2011-09-15 17:30

@josephgbr: done, thanks.

@all: do you also think that <code> tags on Archwiki page are looking ugly? i'm thinking about removing them...

rafaelff commented on 2011-09-12 23:39

I don't have an AMD/ATI card, but this to give my 2 cents:
- pkg description should be updated to "for linux 3.0" or something like that.
- Archwiki page for ATI Catalyst still mentions kernel26. I suggestion someone interested start fixing that. ;)

tsachev commented on 2011-09-08 07:25

I am having trouble with 11.8 too. I am with Radeon HD 6850 and my system freezes shortly after login (gnome 3). Downgrading to 11.7 works for me too. Any idea what would be the problem?

shaurz commented on 2011-09-05 20:59

Version 11.8 causes my system to freeze several times a day on my Radeon HD 6950. I've downgraded to 11.7 and it's working fine.

Anonymous comment on 2011-08-22 13:37

maybe I'll start thinking of switching to KDE :)

Vi0L0 commented on 2011-08-21 11:03

@majlo202: i don't live with rainbow bar - i'm KDE user for ages now :)
Although i got GNOME installed - mainly in case that my KDE would brake or something and i would need to use some X enviroment immidiately. The other reason is to check how catalyst works with it ;).
Anyway u got the point - catalyst sux in gnome-shell support, and oss ati sux in power management. And this IS horrible!
Guess that catalyst will gain gnome-shell support when one of supported distribution will have it by default...

Anonymous comment on 2011-08-21 07:44

@ViOLO: and you are still using catalyst even with the f****g rainbow bar ? How can you live with it ? :) I'm staying on the radeon, but the missing power management kicked down my battery life to 1.5 hours !!! horrible.

Anonymous comment on 2011-08-20 11:04

I searched, found and removed it. -Rdd'ed 'catalyst-total' and re-installed it. This time I symlinked /usr/lib/catalystpxp/libGL.so.1 to /usr/lib/libGL.so.1 and ran aticonfig again, without anything to complain about it did what it should.
A reboot later I tried 'glxgears', let it run for some seconds without problems and <CTRL>-C'ed it. Thinking of that being THE CURE I tried MineCraft and *baam*: Freeze.

In the meantime I checked the catalyst thread in the forums and found someone with seemingly the exact same problem AND card: Panzerino with a HD 6450 ... therefore I'm switching to the forum.

Vi0L0 commented on 2011-08-20 10:40

i wasn't aware of this aticonfig bug, gotta check it, thanks for da info.

yes, search for whatever file you copied and remove it, then install fresh packages (cuz of mentioned bug i recommand clean catalyst, not -total)

Anonymous comment on 2011-08-20 09:55

Right, when I tried catalyst-total I wanted to re-generate xorg.conf (to see if THAT helps) with 'aticonfig --initial', but it complained about not finding 'libGL.so.1' and I didn't find one either, so I copied the most familiar looking file and hoped it would be the one it wants ... I guess I guessed wrong.
I think it was a symlink I copied, so that would explain why it survived the re-installs (removed it with 'pacman -Rdd'), is that possible?
(That day, I wasn't really sane, I also tried to use 'xf86-video-nouveau' because I thought it was the open-source ati driver ...)

I should most probably try to remove it totally, that is also searching for my corrupt copy and install it cleanly, right?

(I'm also getting a [FAIL] at bott when it tries to load modules ... but I checked: it is using 'fglrx'. Must be something else.)

Vi0L0 commented on 2011-08-20 09:28

Wait wait wait - copying 'libGLU.so.1' to 'libGL.so.1' in order to get aticonfig to shut up?
You mean when using catalyst-total? I need some more info here.

Anonymous comment on 2011-08-19 21:32

I appreciate all the work put into the catalyst-packages!

Well, yesterday I installed the 11.8 update and was VERY happy about the new bug which is the disappearance of the mouse cursor freeze ... ;)
But seriously, since the update I'm unable to do anything 3D accelerated, the system freezes up totally and I have to reboot.
I tried pretty much every package (this, -hook and -total) and the problem staid. Could it be related to copying 'libGLU.so.1' to 'libGL.so.1' in order to get aticonfig to shut up? I have no idea, it could be very well some totally wrong library ...
That would be my last guess, but I re-installed catalyst-total twice.

Any hints are welcome, or ... needed (how do I remove the minecraft dependency from my life? xD)!

(If I get too desperate, I might as well just re-install arch and see how it looks like on a fresh install ... I'm unsure: can I install it formatting only my root partition leaving the home one untouched?)

Any warnings about that the written should not be here are supposedly valid. Shall I post it in the forums?

Thank you in advance!

Vi0L0 commented on 2011-08-19 18:02

for me it's the same

Anonymous comment on 2011-08-19 17:33

what about the rainbow bar/panel ? is it still the same ?

Vi0L0 commented on 2011-08-18 16:10

@shaurz:
also there's catalyst-total with all you need: http://aur.archlinux.org/packages.php?ID=31914
It has got not-working powerXpress support, which could be annoying. i'm trying to fix it but if i will fail - catalyst-total will become powerXpress support free.
Stay tuned.

Anonymous comment on 2011-08-18 12:56

@shaurz
http://www.archlinux.org/pacman/makepkg.conf.5.html

shaurz commented on 2011-08-18 12:13

Why not merge catalyst and catalyst-utils in to the same package (like catalyst-test)? It's silly that I have to download the same 72MB file twice, and quite time-consuming.

Anonymous comment on 2011-08-18 09:43

Thank YOU! Finally, after upgraded catalyst-daemon (11.7-1 -> 11.8-1) the cursor bug didnt catch me anymore! :-)
My hardware is ATI PowerColor PCIe 1024MB HD5750 Passiv

Anonymous comment on 2011-08-18 09:36

whats the status on the gnome3 panel bug?

dotmagic commented on 2011-08-18 06:45

I can NOT confirm the cursor Bug is fixed! I have a dual-head setup and tried to remove the SWCursor option, but if I'm doing this there's no mouse cursor on my seconds screen.

Ok, changed back to SWCursor and now the mouse cursor flicker again (especially with slow movements visible). Its so frustrating because the Hardware Cursor worked well for many years with old relases :(

eyenx commented on 2011-08-17 21:52

I can confirm. The bug was fixed.

Vi0L0 commented on 2011-08-17 20:16

on phoronix forums there's and info that mouse freeze bug has been fixed in this release

Vi0L0 commented on 2011-08-17 20:11

updated to 11.8

i don't see any improvements in GNOME3 :(

kidoz commented on 2011-08-17 17:43

Notice an improvement in the gnome-shell.

kidoz commented on 2011-08-17 16:46

11.8 is released

nobody44 commented on 2011-08-12 12:56

Is someone else experiencing problems with two monitors and kernel 3.0?
X doesnt start sometimes at all. I am trying to find the log-entries...

Anonymous comment on 2011-08-11 03:08

I still face the mouse lag issue and SWcursor + tear free does not solve the flickering cursor either.
catalyst 11.7
linux 3.0.1

CPUnltd commented on 2011-08-09 15:02

got it working late last night... was too sleepy to tell if it was before or after a reboot...

Vi0L0 commented on 2011-08-09 14:54

you should reboot after installing new kernel and then build this package - it's because `kernel -r` command is defining what's kernel is in use to build fglrx module for it.
If you want to build fglrx before reboot - change _kernver= line of PKGBUILD - if you are updating into fresh linux 3.0 kernel then it should be: _kernver=3.0-ARCH

CPUnltd commented on 2011-08-09 04:47

make: *** /lib/modules/2.6.39-ARCH/build: No such file or directory. Stop.
==> ERROR: A failure occurred in build().
Aborting...

Vi0L0 commented on 2011-08-07 21:30

@Synthead: done, with pkgrel bump.
If somebody wants to use it with older kernel then (s)he will have to remove/change linux from deps and linux-headers from makedeps.


@ungoliant:
yes, it looks like AMD didn't fix this bug well... you will find confirmation on phoronix forums.

Synthead commented on 2011-08-07 20:29

Please update to work with package 'linux' instead of 'kernel26'

Kev commented on 2011-08-03 22:11

OsVersion says: SMP=1
file /proc/kallsyms says: SMP=1
file /lib/modules/2.6.39-ARCH/build/include/generated/autoconf.h says: SMP=1
file /lib/modules/2.6.39-ARCH/build/include/generated/autoconf.h says: MODVERSIONS=
file /lib/modules/2.6.39-ARCH/build/arch/x86/include/asm/compat.h says: COMPAT_ALLOC_USER_SPACE=arch_compat_alloc_user_space
==> Unpacking archive ...
Creating directory archive_files
Verifying archive integrity... All good.
Uncompressing ATI Catalyst(TM) Proprietary Driver-8.872........................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................Extraction failed.

Anonymous comment on 2011-08-02 13:43

Just installed 11.7. It was said that with 11.7, mouse lag issue on corner would be solved, but im STILL having this problem. Can anyone else confirm? (alredy tried cleaning up everything and reinstalling 11.7)

Also, tried using SWcursor and enabling tear free desktop, but I still get the mouse flickering.

kidoz commented on 2011-07-30 06:17

@MaTachi
In exactly the same way.

MaTachi commented on 2011-07-30 01:02

Anyone who have tried CCC 11.7 on Gnome 3? 11.6 was a horrible experience and almost totally unusable.

Vi0L0 commented on 2011-07-28 13:26

ok, done, no pkgrel bump

Vi0L0 commented on 2011-07-28 13:24

well... on the other hand since catalyst is working well with all kernel versions - i guess i can remove version from kernel26 in catalyst's deps...
yeah this sounds good, if it will work i will update asap

Vi0L0 commented on 2011-07-28 13:11

yes, it works
but adding linux into deps will make only mess on non-testing users systems.
linux package provides kernel26, but it's version is not set (ie. kernel26=3.0) and that's making deps problem.
I will wait until linux will land in [core] with such deps change.

+ thanks for the good news in pxp case

csslayer commented on 2011-07-28 12:44

From phronix, it says linux 3.0 also works... so would you add such depends?

csslayer commented on 2011-07-28 06:25

Good news, I already test 11.7 and the powerxpress works very well. amdcccle provide a new option to choose the display card, for me (KDE 4.7 and chakra), only logout will switch the card. The bad news is this might only works on 2 ati card, not ati/index. (Have hd 4250/5470)

Only keep makefile_compat.patch will make it compile.

sonny commented on 2011-07-27 19:34

11.7 has been released

Anonymous comment on 2011-07-26 13:29

When the next version of catalyst will be released?

csslayer commented on 2011-07-26 06:23

Yes, I have already test with chakra linux 3.0, seems to work well.

Anonymous comment on 2011-07-23 07:54

By the way: even though the package says 'kernel26<2.6.40' it seems to work just fine against my custom 3.0 vanilla kernel.

Vi0L0 commented on 2011-07-08 11:27

udpated, thanks dasjoe

Anonymous comment on 2011-07-07 23:06

New md5sum for http://www2.ati.com/drivers/linux/ati-driver-installer-11.6/./-}-x86.x86_64.run: 645e3b06f802e3edaa52244ff550c4f7

kjslag commented on 2011-07-04 03:32

mrspin: That has happened to me 3 times.

Anonymous comment on 2011-06-27 20:52

Adding this to my xorg.conf:

Section "Device"
...
Option "SWCursor" "true"
EndSection

and then enabling Tear Free Desktop in the Catalyst Control Center both fixed the mouse freezing and flickering.

mrspin commented on 2011-06-24 04:23

Does anyone else have a problem where suddenly the screen goes black, and only a hard shutdown fixes it? The display dies, then I can still ssh in. Restarting X doesn't fix it. If I reboot, there's no display even at POST. I need to actually shut down and power up again to restore display.

Vi0L0 commented on 2011-06-16 10:54

@ungoliant: true, mouse lag bug is still present - it's confirmed on phoronix forums.

CatalystCreator @ twitter said: "For Mouse cursor issue: 11.6 should fix for just about all situations. 11.7 will 100% resolve for every case (including Eyefinity configs)"

Dunno is that about linux, thou

Anonymous comment on 2011-06-16 10:17

Just installed, the mouse lag issue doesn't seem to be fixed. I find it hard to belive, since it is alredy fixed in windows release. Can someone confirm?

Vi0L0 commented on 2011-06-15 16:34

updated to 11.6

Whats new:
- some bugfixes?
- GNOME3 support still sux :(
- mageia linux distro support ;P

@kidoz: thanks

kidoz commented on 2011-06-15 15:06

catalyst 11.6 released:
http://www2.ati.com/drivers/linux/ati-driver-installer-11-6-x86.x86_64.run

chheien commented on 2011-06-13 10:20

aur/zukitwo-themes fixes the rainbow bar, but diagonal renderings are still occuring

Synthead commented on 2011-06-11 23:47

Is this driver making Minecraft crash for anyone else?

sonny commented on 2011-06-11 13:11

@Vi0l0 I dind't meant the package was not working I was just saying that it would be better to use a udev rule.
But as you said it may cause problems with specific configuration (I'm not aware on how this is handled).

Vi0L0 commented on 2011-06-11 13:03

@sonny: catalyst package from repo is working fine here on x86_64, what arch you got there?

Vi0L0 commented on 2011-06-11 12:56

@sonny: i was writing about 20-fglrx.conf more than month ago:
"It looks like Xserver is taking 20-fglrx.conf file in 1st place, even if xorg.conf file is present... It's just ignoring it..."
So it will make problems for multi-screens setups and other settings stored in xorg.conf by amdcccle

+ i will look at catalyst package - i'm compiling it on virtual machine (safe way + im not using stock kernel)

sonny commented on 2011-06-11 11:48

Same problem on the catalyst package from the catalyst repo by the way.

sonny commented on 2011-06-11 11:47

Hi, stupid question why don't you use a udev rule instead of asking the user to set up an xorg.conf file?
For those who are interested, remove /etc/X11/xorg.conf and put this in /etc/X11/xorg.conf.d/20-fglrx.conf

Section "Monitor"
Identifier "aticonfig-Monitor[0]-0"
Option "VendorName" "ATI Proprietary Driver"
Option "ModelName" "Generic Autodetecting Monitor"
Option "DPMS" "true"
EndSection

Section "Device"
Identifier "aticonfig-Device[0]-0"
Driver "fglrx"
EndSection

Vi0L0 commented on 2011-06-01 19:35

no

Anonymous comment on 2011-06-01 16:35

has the rainbow bar been fixed already ?

Anonymous comment on 2011-06-01 11:11

It did help. The mouse still does some strange flicker when I move it through that area, and some other times, but it doesnt freeze anymore.
Thanks!

Vi0L0 commented on 2011-06-01 10:19

as for mouse lag maybe this could help (xorg.conf):

Section "Device"
...
Option "SWCursor" "true"
EndSection

Vi0L0 commented on 2011-05-31 20:59

I know no workaround for this, sorry :/


all archive PKGBUILDs with other needed files are here: http://catalyst.apocalypsus.net/tarball
so you can build it with makepkg

there is also archive repository:
[catalyst-11.4]
Server = http://catalyst.apocalypsus.net/repo/catalyst/$arch

Anonymous comment on 2011-05-31 18:12

I upgraded to catalyst 11.5, and now im suffering the mouse lag bug when moving the pointer to the bottom right corner (It's a known bug that affects lots of users, and windows got a patch alredy). It's supposedly caused by the pointer's shadow. Do you know any way to disable the pointer's shadow, or any other workarround for this? if not, how can I fall back to catalyst 11.4? I used yaourt to build

Thanks

Vi0L0 commented on 2011-05-13 11:04

ok, md5sums were updated, only catalyst-utils' and catalyst-total' pkgver was changed

CPUnltd commented on 2011-05-12 23:35

ah, I see... for the record, the md5 seems to be off, I had to edit the PKGBUILD and drop in a new one for both this and catalyst-utils.

Vi0L0 commented on 2011-05-12 23:32

@CPUnltd:
yes, there's only fglrx module in catalyst package so it's small. Don't forget to install catalyst-utils, and do everything that is written in post install messages.

CPUnltd commented on 2011-05-12 21:02

is the resulting package supposed to be less than 4MB? just seems off to me. Also, I'm sure I'm running this driver already, but OpenGL programs are telling me I am not running a supported driver... kinda lost atm...

niqingliang2003 commented on 2011-05-11 11:12

11.5 still rainbow on dataland 6870

Vi0L0 commented on 2011-05-10 14:53

LOL, if you want it that much i can make new package, like... catalyst-hook + catalyst-utils + lib32-catalyst-utils in 1 package...
In fact i just taken catalyst-total [1], the name is good, i will update it later.

[1]http://aur.archlinux.org/packages.php?ID=31914

dotmagic commented on 2011-05-10 14:43

If you do install the the lib32-catalyst-utils package too, each upgrade download the drivers 3 times!

Vi0L0 commented on 2011-05-10 14:33

There was a time when it was in one package, but i really don't want to explain one more time why it's splitted again, it's all on Bar & Grill forum.
It's always better and safer to build packages manually with makepkg.

Even wonder said that we should stop using yaourt, mainly because it doesn't detect replaces.

Neije commented on 2011-05-10 12:22

First of all thx for providing this package.

But yes, I agree with Dotmagic, downloading the package twice is really weird. I always do "yaourt -S catalyst" and got the dependency on catalyst-utils (normal) and 2 downloads (weird).

dotmagic commented on 2011-05-10 11:44

@Vi0L0
Thanks for remove the stupid dependency check! I always had to download & build the pacakges manual because of this.... now i can just do yaourt -Syua
I still dont understand why catalyst is split in 2 packages. Nobody will only install driver OR utils alone, its pointless. yaourt download the driver twice because of this! :(

Vi0L0 commented on 2011-05-10 10:03

ok, so with silent update (no pkgrel bump) i've removed checking version dependency of catalyst-utils.
This should get a rid over that "error", but please remember to always use catalyst and catalyst-utils in same version.

Ps. if you got catalyst package built earlier than today then this "error" will persist also with next update, ofcourse.

Anonymous comment on 2011-05-10 01:06

Every time I'm going to update I have to remove catalyst, catalyst-utils andlib32-catalyst-utils and them install them again.

I'm using "yaourt -Syu --devel --aur"

Is there any way to update them without erros like these:
error: failed to prepare transaction (could not satisfy dependencies)
:: catalyst: requires catalyst-utils=11.4
==> WARNING: Your packages are saved in /tmp/yaourt-tmp-jteixeira
==> ERROR: unable to update


Thank a lot

Vi0L0 commented on 2011-05-09 21:30

udpated to 11.5
more info here: https://bbs.archlinux.org/viewtopic.php?pid=931705#p931705

btw: looks like GNOME3 still got some troubles

Anonymous comment on 2011-05-09 16:16

Catalyst 11.5 is out
http://www2.ati.com/drivers/linux/ati-driver-installer-11-5-x86.x86_64.run
Why have they stop producing release notes?

Anonymous comment on 2011-05-03 18:18

good to hear that I'm not the only one with rainbow :) I went back to radeon and will wait for the fixed catalyst driver ...

niqingliang2003 commented on 2011-05-03 11:22

yes I got the "rainbow" panel when using gnome3

niqingliang2003 commented on 2011-05-03 11:20

when use gnome3, the top panel's color is abnormal and some flash.

Anonymous comment on 2011-04-30 15:55

Same error here laptop w/ ati 4850
failed to link fglrx-libgl.so, please check wether driver is installed correctly

Vi0L0 commented on 2011-04-29 20:24

@majlo202:
yes, it's well know bug and it must be fixed by amd's devs.

Anonymous comment on 2011-04-29 17:14

are you also experiencing problems with "rainbow" bar in gnome-shell ?

Vi0L0 commented on 2011-04-29 11:38

@tealson:
i'm working on it, it should teoretically be easy to fix although i got some troubles...
anyway it will work as expected, except powerXpress switching support - but this feature is amd's early look and it fails in most cases anyway...

Anonymous comment on 2011-04-29 09:45

I installed everything without any errors. lsmod tells me fglrx is loaded but:
aticonfig --initial
tells me:
failed to link fglrx-libgl.so, please check wether driver is installed correctly

nicoadamo commented on 2011-04-28 04:37

We ATI users are very used to it until :)
Or at least until we discover such things as catalyst-daemon or even better, the repo [catalyst] for pacman.conf. Then life's good again :)

CPUnltd commented on 2011-04-28 02:54

would be nice if the install for this stuff was a bit less manual... Would like it to uninstall the old stuff and install the new stuff if not overwrite the old (not sure which Nvidia does, but it's fairly seemless)...

just sayin...

nicoadamo commented on 2011-04-28 01:30

Thanks for the answer ViolO! You've been pretty clear. Since "Bleeding Edge" is what we're all about, I'm buying 11.4 Jajaja!I hope I don't break all my system down when uninstalling xorg 1.9 and catalys 11.3 from the repos. Cheers! :)

Vi0L0 commented on 2011-04-27 21:53

@nicoadamo: well... it's hard to answer that question. It depends on what do you expect from catalyst.
Have you tried vaapi on catalyst-test or 11.4? If not then I think you should - it's not working for me (hd4850), but it surelly is working for some others (sometimes out of the box, sometimes when downgrading mplayer-vaapi to mplayer-vaapi-33025-2 version).
If it's not working and you need it - definitelly use catalyst 11.3 - aspecially when it's so easy to do - you have got [xorg19] and [catalyst-11.3] repositories available, plus sources if you need them: http://catalyst.apocalypsus.net/tarball

You won't gain anything spectacular with newer catalyst 11.4 (or newer xorg-server 1.10) - i have heard about better openCL performance, but this feature is targeted mainly for devs atm.

nicoadamo commented on 2011-04-27 18:11

Is it woth keeping Xorg 1.09 and the older drive until better support? Thanks

nicoadamo commented on 2011-04-27 18:11

Is it woth keeping Xorg 1.09 and the older drive until better support? Thanks

Vi0L0 commented on 2011-04-27 17:28

updated to 11.4:
- surprisingly it supports xserver 1.10
- got officiall 2.6.38 kernel support
- vaapi isn't working for me (just like with catalyst-test)

niqingliang2003 commented on 2011-04-14 06:39

and a curious problem, before update kernel, the catalyst 11.3 is OK, but when I change the original VGA cable(connect with 6870's first DVI through a adaptor) to a 5m VGA cable, my GDM login interface's resolution changed to 1024 * 768 (my display is 1440 * 900), and after login, I changed the resolution to 1440 * 900, but after reboot, the gdm login interface is also 1024 * 768, after login is 1440 * 900

niqingliang2003 commented on 2011-04-14 06:32

and a curious problem, before update kernel, the catalyst 10.3 is OK, but when I change the original VGA cable(connect with 6870's first DVI through a adaptor) to a 5m VGA cable, my GDM login interface's resolution changed to 1024 * 768 (my display is 1440 * 900), and after login, I changed the resolution to 1440 * 900, but after reboot, the gdm login interface is also 1024 * 768, after login is 1440 * 900

niqingliang2003 commented on 2011-04-14 06:27

x server is stay on 1.09, no update to 1.10, because pacman report catalyst dpends on xorg server 1.09

Vi0L0 commented on 2011-04-13 15:05

just a FYI: afaik catalyst 11.4 will not support xserver 1.10

Anonymous comment on 2011-04-13 14:50

As Vi0L0 has said in one of the catalyst-utils comments, here are the possible solutions:

> - add xorg-server-* and xf86-input-* packages which you installed into pacman.conf's IgnorePkg list
> OR
> - use [xorg19] repo - read https://wiki.archlinux.org/index.php/ATI_Catalyst#.5Bxorg19.5D
> OR
> - install catalyst-test package which have such support, but know that it lacks vaapi support : http://aur.archlinux.org/packages.php?ID=34773

I've gone for the third option, and the catalyst-test package works perfectley with the new xorg version.

Anonymous comment on 2011-04-13 13:58

did your update you X server to 1.10, The driver do not support 1.10, and did your black the driver radeon.My computer work fine on kernel 2.6.38,just wait the 11.4

niqingliang2003 commented on 2011-04-13 13:52

I update kernel to 2.6.38, and reinstalled catalyst and utils; when I start gdm, my display display "no signal", like no VGA input.

ptchinster commented on 2011-04-12 23:08

Yup, i just updated and my machine locks up and i have to do a hard reboot. Ill be sticking with 11.2 for a while i guess.

Achterin commented on 2011-04-11 18:25

maybe its a bit late, but it seems that catalyst is broken for my card since 11.2. 11.1 works without problems, but when i try to start the xserver i get a blank screen...
this is the xorg.log output: http://www.privatepaste.com/download/78f54bbe58
i don't know if you are the right person to talk to, but maybe someone else has an idea or the same problem so we can report this to amd.
my kernel version is now 2.6.38 but it was also broken with 2.6.37.
my card is a hd5650 mobility in a sony vaio vpceb1z1e.
if you need more information please contact me.

Vi0L0 commented on 2011-03-29 21:26

updated to 11.3:
- still no xorg-server 1.10 support (please use [xorg19] repo or pacman.conf's IgnorePkg when xorg-server hits [extra] or when using [testing]
- some bugfixes (switching apps from fullscreen when Tear-Free Desktop option is enabled does not cause X server brake on my gfx anymore)
- (possibly) Tear-Free Desktop option does not eat such big amount of memory

Anonymous comment on 2011-03-06 00:02

solved by following the instruction https://wiki.archlinux.org/index.php/ATI_Catalyst.

Nice work.

Anonymous comment on 2011-03-05 23:54

After installing qt, amdcccle tells no driver installed or ATI driver is not functioning properly.

Anonymous comment on 2011-03-05 23:53


amdcccle: error while loading shared libraries: libQtGui.so.4: cannot open shared object file: No such file or directory


Anonymous comment on 2011-03-05 03:22

I would love it if "patch" was a dependency so I don't remember half way through the install every time XD.

Anonymous comment on 2011-03-01 01:25

@Vi0L0
thanks for the really quick reply and the fix worked perfectly :) thanks heaps

Vi0L0 commented on 2011-03-01 00:26

edit /etc/X11/xorg.conf and add those options into Section "ServerLayout":
Option "BlankTime" "0"
Option "StandbyTime" "0"
Option "SuspendTime" "0"
Option "OffTime" "0"

edit: and restart X server ;)

Vi0L0 commented on 2011-03-01 00:25

edit /etc/X11/xorg.conf and add those options into Section "ServerLayout":
Option "BlankTime" "0"
Option "StandbyTime" "0"
Option "SuspendTime" "0"
Option "OffTime" "0"

Anonymous comment on 2011-03-01 00:12

@Vi0L0
not sure if its a bug in catalyst itself but i have disabled any screensaver or suspend options, after 10 mins or so the monitor shuts of and i am unable to get back on my computer have to hard restart for it to work again

Vi0L0 commented on 2011-02-23 22:16

Yes, you can safely remove catalyst and install one of "optional" packages. I can now see that "optional" isn't a good term in situation like this :>.
I will clear this up with next update.

Vi0L0 commented on 2011-02-23 22:15

Yes, you can safely remove catalyst and install one of "optional" packages. I can now see that "optional" isn't a good term in situation like this :>.

Renfield commented on 2011-02-23 17:58

Sorry if this has been answered. I have installed this package, and I'm reading the catalyst Arch Wiki page and I think I should install one of catalyst-generator, catalyst-hook, or catalyst-daemon. In fact when I installed catalyst it mentioned those as optional packages. However, when I go to install any one of them, it says that there is a conflict with catalyst, and asks me if I wish to remove it.

So, my question, can I safely remove catalyst and install one of the other packages? Will that provide both the driver, and some mechanism to rebuild the kernel module as needed?

Thanks for your support.

Vi0L0 commented on 2011-02-20 22:25

@rogersce: so it is, and was - driver's url stay unchanged for ages.
Does it mean that you got some problems with downloading with makepkg or aur manager?
Please don't suggest in url presented on aur site - it was always wrong, instead look into PKGBUILD - a place of miracles ;).

rogersce commented on 2011-02-20 03:07

Correct link for the driver should be

http://www2.ati.com/drivers/linux/ati-driver-installer-11-2-x86.x86_64.run

Anonymous comment on 2011-02-16 09:55

Updated to 11.2 and same error: http://paste.pocoo.org/show/339481/

Will try on the forums suggested by @Vi0L0

Vi0L0 commented on 2011-02-15 23:41

updated to 11.2:
- driver package is now much smaller (83MiB) that's because amd unified some of their binaries for different X.Org Server;
- Radeon HD 6800 series should now be officially supported;
- bugfixes.

BluePeril commented on 2011-02-15 18:48

Catalyst 11.2 seems to be available (you can download it, but it's not posted on the website)
md5sum: 4db782c2a7924b965918275c39157a28

Anonymous comment on 2011-02-15 13:49

(from: http://aur.archlinux.org/packages.php?ID=40833)

Trying to install it with coreutils-8.9 doesn't work either, BUT the log indicates that some more work has been done (although leads to the same error): http://paste.pocoo.org/show/338903/

If I run the Makefile by myself it seems to finish the build successfully, but no fglrx.ko file appears (just *.o files): http://paste.pocoo.org/show/338905/

Thanks

Vi0L0 commented on 2011-02-10 17:51

Silent update.
I didn't like the way how ati's code was cluttering PKGBUILDs or catalyst_build_module scripts, so i turned this code into ati_make.sh file and import it's content with . (source)
New way also fits to ati license better... ;P
Oh and now all AUR catalyst's packages have 2.6.38 patch so they support 38 rc3, rc4 and probably newer rcs (no rc1, rc2 support, but hey rc3 and rc4 are much better).
Change is mainly cosmetical, there's no need to update packages.

johnthekipper commented on 2011-02-06 23:05

@ Vi0L0

all plugins works just not the plugins with shaderss.... see. http://wiki.ubuntuusers.de/KWin/Arbeitsfl%C3%A4cheneffekte

Vi0L0 commented on 2011-02-06 22:29

@johnthekipper: for me explosion was never working, nether with xrendr or opengl. afaik it's well known bug

johnthekipper commented on 2011-02-06 21:18

i have with xrandr a dualhead screen and the kwin effekt plugin explosion don't can be used

.xsession-errors

OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI Radeon HD 4800 Series
OpenGL version string: 1.4 (3.3.10428 Compatibility Profile Context)
Driver: Catalyst
Driver version: 1.4
GPU class: R700
OpenGL version: 1.4
X server version: 1.9.4
Linux kernel version: 2.6.37
Direct rendering: no
Requires strict binding: yes
GLSL shaders: no
Texture NPOT support: yes
kwin(2430) KWin::EffectsHandlerImpl::loadEffect: EffectsHandler::loadEffect : Effect “kwin4_effect_explosion” is not supported



cat /var/log/Xorg.0.log |grep WW
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[ 121.680] (WW) The directory “/usr/share/fonts/OTF/” does not exist.
[ 121.680] (WW) The directory “/usr/share/fonts/OTF/” does not exist.
[ 121.680] (WW) AllowEmptyInput is on, devices using drivers ‘kbd’, ‘mouse’ or ‘vmmouse’ will be disabled.
[ 121.680] (WW) Disabling Mouse0
[ 121.680] (WW) Disabling Keyboard0
[ 121.681] (WW) Open ACPI failed (/var/run/acpid.socket) (No such file or directory)
[ 121.703] (WW) Falling back to old probe method for fglrx
[ 121.707] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:0:0) found
[ 121.707] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:2:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:5:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:11:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:12:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:18:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:1) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:2) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:3) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:4) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:19:5) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:0) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:1) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:2) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:3) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@0:20:4) found
[ 121.708] (WW) fglrx: No matching Device section for instance (BusID PCI:0@1:0:1) found
[ 121.751] (WW) fglrx(0): board is an unknown third party board, chipset is supported
[ 122.837] (WW) fglrx(0): Option “AllowGLXWithComposite” is not used
[ 122.837] (WW) fglrx(0): Option “RenderAccel” is not used


kdm.log
(EE) fglrx(0): Not enough video memory to allocate Shadow surface, shadowFB functionality cannot be enabled..
(EE) fglrx(0): Not enough video memory to allocate Shadow surface, shadowFB functionality cannot be enabled..
Current Operating System: Linux ArchLinux 2.6.36-ARCH #1 SMP PREEMPT Mon Jan 24 18:41:03 CET 2011 x86_64
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
Current Operating System: Linux ArchLinux 2.6.36-ARCH #1 SMP PREEMPT Mon Jan 24 18:41:03 CET 2011 x86_64
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
Current Operating System: Linux ArchLinux 2.6.37-ARCH #1 SMP PREEMPT Sat Jan 29 20:00:33 CET 2011 x86_64
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(EE) fglrx(0): ACPI: DRM connection failed
(EE) fglrx(0): ACPI: DRM connection failed
(EE) fglrx(0): atiddxDriScreenInit failed, GPS not been initialized.
(EE) fglrx(0): XMM failed to open CMMQS connection.(EE) fglrx(0):
(EE) fglrx(0): XMM failed to initialize
(EE) fglrx(0): firegl_SetSuspendResumeState FAILED -9.
Current Operating System: Linux ArchLinux 2.6.37-ARCH #1 SMP PREEMPT Sat Jan 29 20:00:33 CET 2011 x86_64
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(EE) fglrx(0): ACPI: DRM connection failed
(EE) fglrx(0): ACPI: DRM connection failed
(EE) fglrx(0): atiddxDriScreenInit failed, GPS not been initialized.

Anonymous comment on 2011-02-03 18:59

Thanks a lot, also helped me.

Anonymous comment on 2011-02-03 18:22

http://twitter.com/illespal/status/32530753275301888

This helped me

Anonymous comment on 2011-02-03 18:17

I've a bit of a tumble with this package.
I updated to this version, system is booting normally, but when it gets to the login manager I get a black screen of death.
I tried some troubleshooting, but nothing works so far.

I deleted /ati/amdposdb, no effect
If I delete xorg.conf X and KDE starts with the vesa driver...

I dont think this is just a Problem of mine, so if it has been solved already, please give me a little help.

kralyk commented on 2011-01-31 00:31

Ah, I see it's in utils, sry, nevermind...

kralyk commented on 2011-01-31 00:26

Do I understand correctly, that you're no more including CAL .so files?
Not even in utils or any other pkg? If so, I could remove the conflict...

Vi0L0 commented on 2011-01-30 13:34

@amass: this scrpit seems to work fine with 11.1 on other distros so it has to work fine also here ;)
AFAIK you don't need to put it inside daemons directory (rc.d), try to run it once as root.
Maku sure you installed binutils and sed.

Anonymous comment on 2011-01-30 12:26

Hi ViOLO

Latest update works fine, thanks a lot!

Owning a Radeon HD 6870, I just get the "AMD unsupported hardware" watermark back, any idea?

My /etc/rc.d/ati files is:
#!/bin/sh
DRIVER=/usr/lib/xorg/modules/drivers/fglrx_drv.so
for x in $(objdump -d $DRIVER|awk '/call/&&/EnableLogo/{print "\\x"$2"\\x"$3"\\x"$4"\\x"$5"\\x"$6}'); do
sed -i "s/$x/\x90\x90\x90\x90\x90/g" $DRIVER
done

Anonymous comment on 2011-01-28 21:43

Just wondering if anyone else had issues with resuming from suspend. I have a 785GM motherboard that includes a HD4200, the monitor is connected via DVI (also had HDMI and VGA output), but it seems the monitor is not turned back on - I don't think the computer gets back properly either. Does anyone have experience with a situation like that?

divansantana commented on 2011-01-28 13:28

@ViOLO Is actually quite funny. Thanks for the simple and effective solution.
Thanks for the great package :)

Vi0L0 commented on 2011-01-28 12:11

@divansantana: at least it's pretty funny ;) :

--> resolving catalyst-utils=11.1... found in AUR
looking for inter-conflicts...
:: catalyst: requires catalyst-utils=10.12

yeees :)
so maybe do it in ugly way:
pacman -Rd catalyst catalyst-utils
bauerbill --aur -S catalyst catalyst-utils

divansantana commented on 2011-01-28 12:00

#bauerbill --aur -S catalyst
WARNING: catalyst does not match any package or group in sync db
--> Searching for AUR packages...
--> Loading AUR taurball data...
WARNING: about to source /tmp/bauerbill/build/aur/catalyst/PKGBUILD
view [v] backup [b] edit [e] continue [ok] skip [s] abort [x] ok
--> an associated install file was found for catalyst: /tmp/bauerbill/build/aur/catalyst/catalyst.install
view [v] backup [b] edit [e] continue [ok] skip [s] abort [x] ok
--> resolving kernel26>=2.6.29... installed
--> resolving kernel26<2.6.39... installed
--> resolving kernel26-headers... installed
--> resolving catalyst-utils=11.1... found in AUR
--> resolving gcc-libs... installed
--> resolving gcc>4.0.0... installed
--> resolving make... installed
WARNING: about to source /tmp/bauerbill/build/aur/catalyst-utils/PKGBUILD
view [v] backup [b] edit [e] continue [ok] skip [s] abort [x] ok
--> an associated install file was found for catalyst-utils: /tmp/bauerbill/build/aur/catalyst-utils/catalyst-utils.install
view [v] backup [b] edit [e] continue [ok] skip [s] abort [x] ok
--> resolving xorg-server>=1.9.0... installed
--> resolving xorg-server<1.10.0... installed
--> resolving netkit-bsd-finger... installed
--> resolving libxrandr... installed
--> resolving libsm... installed
--> resolving fontconfig... installed
--> resolving libxcursor... installed
--> resolving libxi... installed
--> Determining sizes of source files...
--> Installing /tmp/bauerbill/build/aur/catalyst-utils/catalyst-utils-11.1-2-x86_64.pkg.tar.xz...
--> Running pacman: '/usr/bin/pacman' '--config' '/etc/pacman.conf' '-U' '/tmp/bauerbill/build/aur/catalyst-utils/catalyst-utils-11.1-2-x86_64.pkg.tar.xz' '--asdeps' '--cachedir' '/var/cache/pacman/pkg'
resolving dependencies...
looking for inter-conflicts...
error: failed to prepare transaction (could not satisfy dependencies)
:: catalyst: requires catalyst-utils=10.12
error: pacman exited with an error (256)

Not sure why it says catalyst: requires catalyst-utils=10.12
I also rm -rf /tmp/bauerbill/build/aur/catalyst* and seems to do the same.

I'll keep playing.

Vi0L0 commented on 2011-01-28 11:39

@divansantana: looks like some bauerbill error to me... umm... what will happen if you will: bauerbill --aur -S catalyst
?

divansantana commented on 2011-01-28 02:38

Strange dep error when installing latest version, any ideas?

--> Installing /tmp/bauerbill/build/aur/catalyst-utils/catalyst-utils-11.1-2-x86_64.pkg.tar.xz...
--> Running pacman: '/usr/bin/pacman' '--config' '/etc/pacman.conf' '-U' '/tmp/bauerbill/build/aur/catalyst-utils/catalyst-utils-11.1-2-x86_64.pkg.tar.xz' '--asdeps' '--cachedir' '/var/cache/pacman/pkg'
resolving dependencies...
looking for inter-conflicts...
error: failed to prepare transaction (could not satisfy dependencies)
:: catalyst: requires catalyst-utils=10.12
error: pacman exited with an error (256)

Vi0L0 commented on 2011-01-27 12:32

@OvsInc: ok, i will change that with next update. Thanks.

OvsInc commented on 2011-01-27 09:09

Sorry, I wrote in the wrong place (http://aur.archlinux.org/packages.php?ID=40834).
Perhaps more correctly it would be in the catalyst package replace the
depends=('kernel26>=2.6.29' 'kernel26<2.6.39' 'kernel26-headers' "catalyst-utils=${pkgver}" 'gcc-libs' 'gcc>4.0.0' 'make')
by
depends=('kernel26>=2.6.29' 'kernel26<2.6.39' "catalyst-utils=${pkgver}")
makedepends=('gcc-libs' 'gcc>4.0.0' 'make' 'kernel26-headers')

Vi0L0 commented on 2011-01-26 21:45

ok, updated, i bumped all pkgrels cuz it's different driver than before...

Vi0L0 commented on 2011-01-26 21:26

oh lol lol lol... not only md5sum is different, but its a whole different version of the driver :)
was 8.81,
is 8.812,
a lot of files is different also those used in module compilation...
let me test it, then i will update AUR and repo again :)

divansantana commented on 2011-01-26 21:08

I hope this new driver fixes my wake from suspend issue where my CPU goes crazy and harddisk because all my used/application memory is swapped out to disk. Really wierd bug.

Glaucous commented on 2011-01-26 20:50

Seems like the md5sum for AMD/ATI driver download is wrong, I had to change it to: 'f28f358b3a40a0a06e16f45925d9a875'

Vi0L0 commented on 2011-01-26 20:01

updated to 11.1:
- official kernel support up to 2.6.37 (works fine also on 2.6.38rc2)
- new amdcccle option, let me translate umm... "distortion/deformation elimination" - anyway something like enhanced v-sync ;P

Vi0L0 commented on 2011-01-23 22:48

thanks amass, sums of all catalyst-10.12 packages were updated, no pkgrel bump

only 3 files were updated: etc/ati/amdpcsdb.default + /etc/ati/control + /etc/ati/signature
so this strange and late update is touching _only_ our catalyst-utils package

Anonymous comment on 2011-01-23 14:13

FYI, md5sums validation is KO for ati-driver-installer-10-12-x86.x86_64.run file...

Vi0L0 commented on 2011-01-09 20:37

yep, 2.6.37 seems to work fine as long as kernel is compiled WITH BKL option, so it will NOT work with CONFIG_BKL=n option.

Possibility of turning BKL (big kernel lock) off is one of most important (or most famed) features of newest stable kernel and almost all parts (UDF is exception) of the kernel are now able to run without using it.

Will CONFIG_BKL=n gave performance boost? I don't think so, not now anyway (no-BKL need some optimizations) and if any - it will touch mainly (only?) real time systems.

Will community prepare catalyst's BKL patches? Hard to tell, it's possible that only AMD can do this.

Vi0L0 commented on 2011-01-09 20:09

yep, 2.6.37 seems to work fine as long as kernel is compiled WITH BKL option, so CONFIG_BKL=y is requiered.

Possibility of turning BKL (big kernel lock) off is one of most important (or most famed) features of newest stable kernel and almost all parts (UDF is exception) of the kernel are now able to run without using it.

Will CONFIG_BKL=n gave performance boost? I don't think so, not now anyway (no-BKL need some optimizations) and if any - it will touch mainly (only?) real time systems.

Will community prepare catalyst's BKL patches? Hard to tell, it's possible that only AMD can do this.

Anonymous comment on 2011-01-07 09:21

PrinceAMD: Catalyst (fglrx) from this package works with 2.6.37. I have no problems so far.

Anonymous comment on 2011-01-07 06:45

any patch for kernel 2.6.37 final?

Vi0L0 commented on 2010-12-13 20:20

updated to 10.12

lots of files is different / updated, unfortunatelly i can not see any difference (again?), probably bugfixes

still no official 2.6.36 kernel support :(

i also added 2.6.37-rc* kernel (up to 2.6.37-rc5 for now) support to all catalyst packages

Anonymous comment on 2010-11-26 22:37

Yeah it took me something like 18 tries to download before the checksum matched. just recompiled type thing for 2.36 kernal.

Vi0L0 commented on 2010-11-24 13:07

@anish: are you sure? I just downloaded it, and checksum is the same. Maybe you should re-download it and check again.

anish commented on 2010-11-24 07:01

the checksum for ati-driver-installer-10-11-x86.x86_64.run needs to be updated

Anonymous comment on 2010-11-24 01:26

Patch belongs to the base-devel group. This group is expected to be installed when using ABS or AUR. So it's not needed as dependency. Please, read the ABS and AUR wiki pages.

Anonymous comment on 2010-11-24 01:06

Please add 'patch' to the deps, which could be missing on a clean system :)
Thanks!

Vi0L0 commented on 2010-11-23 15:53

@CPUnltd: this should be helpful: https://wiki.archlinux.org/index.php/ATI_Catalyst#Installation
If any problems occur - write here: https://bbs.archlinux.org/viewtopic.php?id=57084&p=107

CPUnltd commented on 2010-11-23 04:06

was reading the install file and saw instructions for updating to this newest driver, but I have a question... it doesn't give detailed info and says something about needing a kernel module... not sure exactly what I'm supposed to do with a stock kernel (updated to most recent in stable repos)... can someone explain exactly how I'm supposed to install this from AUR so I don't screw anything up?

BlindPenguin commented on 2010-11-18 18:29

Couldn't find any problems. Desktop (Xfce) is useable, blender works nice, urban terror has 20-40 fps. All on a HD 2400 (r6xx). No difference. I think it's ok. :)

So far.. nice work ViOLO. Thank you. :)

BlindPenguin commented on 2010-11-18 16:37

Hope so. I'll try it. :)

Vi0L0 commented on 2010-11-18 00:58

updated to 10.11

What's new? That's a good question ;)
Probably some bugfixes...

m3thodic commented on 2010-11-07 17:03

Sorry about that, aticonfig setup a keyboard/mouse directive in my xorg.conf. Apparently this causes a blank screen in X. I commented out the input sections and X fires up. Getting slow playback (about 50% full speed) but that is probably an issue with Alsa.

m3thodic commented on 2010-11-07 15:55

It sort of works on x86_64.... if I "startx" without a .xinitrc file, X fires up twm (doesn't matter if I do it as root or a regular user). If I try to exec twm out of a .xinitrc file (again, user doesn't matter) I get a blank screen.

Vi0L0 commented on 2010-11-04 13:34

i've added 2.6.37 rc1 testing support to AUR's catalyst-generator

after your confirmation (it's working fine on my i686, i need signoff for x86_64) - i will update others AUR's and repo's packages.

BTW: also here there's no problem with xserver 1.9.2

Vi0L0 commented on 2010-11-04 13:33

i've added 2.6.37 rc1 testing support to AUR's catalyst-generator

after your confirmation (it's working fine on my i686, i need signoff for x86_64) - i will update others AUR's and repo's packages.

Anonymous comment on 2010-11-04 13:02

I've upgraded without problems.

silvernode commented on 2010-11-04 10:29

is it safe to upgrade to xorg-server-1.9.2-1?

Anonymous comment on 2010-11-03 14:26

I couldn't get catalyst-generator to compile the fglrx module for kernel26-mainline 2.6.37-rc1 package.

agapito commented on 2010-11-01 20:30

It works with 2.6.37 rc1?

dotmagic commented on 2010-10-24 09:24

@ptchinster:
Another reason could be you still have the Xorg 1.8 repository in your pacman.conf. It works well after I removed it.

Vi0L0 commented on 2010-10-23 18:32

@ptchinster:
this 10.10 build supports only xorg-server 1.9
you probably added xorg-server into IgnorePkg list of pacman.conf - that's probably why it's building from the AUR

feel free to install xorg-server 1.9 from [extra] repo

ptchinster commented on 2010-10-22 18:22

Why is it having me build xorg-server 1.9 from the AUR?

Vi0L0 commented on 2010-10-22 17:31

updated, seems it's working fine at least on i686.

i'm not able to test it precisely, but there are not so many changes comparing to 10.10 RC1 (catalyst-test 10.10-999) so it should be fine.

agapito commented on 2010-10-22 15:37

10.10 Released

https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/drivers/linux/ati-driver-installer-10-10-x86.x86_64.run

agapito commented on 2010-10-22 15:37

10.10 released

BlindPenguin commented on 2010-10-22 14:49

No problem. I want a working driver, so take your time.

Vi0L0 commented on 2010-10-22 14:29

catalyst 10.10 should be released today, but my main 500GB hdd with arch x86_64 just refuse to any work :(...

so probably i will update it tomorrow, sorry for inconvenience

Anonymous comment on 2010-09-29 16:13

Well, forget it. Was a false alarm.

Vi0L0 commented on 2010-09-29 14:38

xorg-server 1.9 landed in [extra] repo, so either use
- catalyst 10.9 with [xorg18] repo
or
- catalyst-test package

Vi0L0 commented on 2010-09-28 03:03

x86_64 2.6.35.6 kernel problem: https://bbs.archlinux.org/viewtopic.php?pid=832169#p832169

Anonymous comment on 2010-09-22 19:48

@violo: at this point i guess its time to thank you for all your work and support. did not update my driver lately but will do and its good to know guys like you are out there. you sir are awesome. greetings from germany

Vi0L0 commented on 2010-09-22 19:07

@Hamsterkill: indeed

Hamsterkill commented on 2010-09-22 14:17

Is the catalyst_build_module command not included unless using one of the auto recompilation packages?

Vi0L0 commented on 2010-09-19 19:56

fixed, it should now always work, feedback is welcome :)

oh and thanks ptchinster

Vi0L0 commented on 2010-09-19 14:17

@ptchinster:
temporary solution is to remove PKGBUILD's patching lines from 39 to 45:

if [ "${CARCH}" = "x86_64" ]; then
fix=$(grep arch_compat_alloc_user_space /lib/modules/${_kernver}/build/arch/x86/include/asm/compat.h )
if [ "$fix" = "static inline void __user *arch_compat_alloc_user_space(long len)" ]; then
cd "${srcdir}/archive_files/"
patch -Np1 -i ../arch_compat.patch || return 1
fi
fi

ptchinster commented on 2010-09-19 00:19

#uname -a
Linux hostname 2.6.34-ARCH #1 SMP PREEMPT Wed Aug 11 00:23:15 CEST 2010 x86_64 Intel(R) Core(TM) i7 CPU Q 820 @ 1.73GHz GenuineIntel GNU/Linux

ptchinster commented on 2010-09-19 00:18

==> Unpacking archive ...
Warning: target directory exists archive_files
Verifying archive integrity... All good.
Uncompressing ATI Catalyst(TM) Proprietary Driver-8.771.............................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................
Aborting...
==> ERROR: Makepkg was unable to build catalyst.
==> Restart building catalyst ? [y/N]
==> ---------------------------------
==>

Vi0L0 commented on 2010-09-18 16:24

update: maybe my code isn't beautiful (got no time) but it should work now with fresh x86_64 2.6.35.4-2 stock kernel

i686 users don't need to update to 10.9-3, because hole was only in x86_64 kernel

i've found patch here: http://www.phoronix.com/forums/showthread.php?t=25485&page=2
Thanks goes to aidanlinz

Vi0L0 commented on 2010-09-18 16:23

update: maybe my code isn't beautiful (got no time) but it should work now with fresh 2.6.35.4-2 stock kernel

i686 users don't need to update 10.9-3, because hole was only in x86_64 kernel

i've found patch here: http://www.phoronix.com/forums/showthread.php?t=25485&page=2
Thanks goes to aidanlinz

Vi0L0 commented on 2010-09-18 14:32

@kama:
Thanks!

I see that stock kernel was patched with very important patch that fix some critical security hole in kernel, so it's good.

Now i will look at it, and write a fglrx patch / or search the net for such patch.

kama commented on 2010-09-18 14:17

makepkg crashes with new kernel26 2.6.35.4-2:

CC [M] /var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/kcl_ioctl.o
/var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/kcl_ioctl.c: In Funktion »KCL_IOCTL_AllocUserSpace32«:
/var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/kcl_ioctl.c:196:5: Fehler: Implizite Deklaration der Funktion »compat_alloc_user_space«
/var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/kcl_ioctl.c:196:5: Warnung: return erzeugt Zeiger von Ganzzahl ohne Typkonvertierung
make[1]: *** [/var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod/kcl_ioctl.o] Fehler 1
make: *** [_module_/var/abs/local/catalyst/src/archive_files/common/lib/modules/fglrx/build_mod] Fehler 2
make: Leaving directory `/usr/src/linux-2.6.35-ARCH'

netmanny commented on 2010-09-17 20:06

OK sorry ahead for my confusion here but why catalyst conflicts with -hook or -daemon?
aren't they part of the same build system?

Vi0L0 commented on 2010-09-16 12:24

@v6ak:
actually not :) because you don't need xserver to build kernel's module. And a module won't work without catalyst-utils which is the "center" of the catalyst drivers now.
catalyst-utils got xorg-server<1.9 dep

Anonymous comment on 2010-09-16 07:38

'xorg-server<1.9' should be added to depends array, shouldn't it?

Vi0L0 commented on 2010-09-15 20:23

@fabioamd87:
it's because information about version is stored in /etc/ati/amdpcsdb file.
This file is created while first run of X/DE by freshly installed catalyst, and isn't provided by catalyst-utils package - so it's not updated.
It's not my fault actually, it's ati's idea.
I suggest you to kill your X/DE, then remove /etc/ati/amdpcsdb, restart your X/DE, and then set up amdcccle.

Fabioamd87 commented on 2010-09-15 20:15

Why amdcccle still say that I have 10.7?

Vi0L0 commented on 2010-09-15 19:35

@cyberpatrol
heh... ok, so here's direct link to my forum's post: https://bbs.archlinux.org/viewtopic.php?pid=825546#p825546

I suggest you to write your doubts there, on the forum.

Anonymous comment on 2010-09-15 19:28

@Vi0L0: Why should an automated recompilation done by the hook build a different fglrx module than a manual recompilation by manually entering makepkg? Then all the automated recompilation shouldn't have worked, yet. If such an automated recompilation is done at boot time, this can indeed result in a different fglrx.

Vi0L0 commented on 2010-09-15 19:20

@agapito: maybe because lib32-catalyst-utils were (are not atm) optdeps for catalyst-utils... but that's strange you catched such issue, anyway try again

Vi0L0 commented on 2010-09-15 19:15

@cyberpatrol:
I know exactly what i am doing.
I splited catalyst between catalyst and catalyst-utils for same reasons for which nvidia package is splited.

I proved that automatic recompilation is making different fglrx module than makepkg on forum. Also read ragnarok2040 which is even more important.

If you don't think that what i just did is right way - then again ask arch devs why it's right way.
I already know what they think about 'auto re-compilation'.

agapito commented on 2010-09-15 19:09

hmmm i can't update catalyst on my 32 bits installation ... I have issues with lib32-catalyst-utils ... why i have to install lib32-libs?

Anonymous comment on 2010-09-15 18:59

@Vi0L0: What's wrong with automatic re-compilations? If they are done correctly, in the right way and at the right time then there's nothing to say against that. But if they are done at the wrong time (at boot time like it is done with your "daemon") then it is indeed NOT good. And all this stuff has absolutely NOTHING to do with an Arch way of some sort.

Splitting the package is just inconvenient. What do you need this splitting for? Catalyst and catalyst-utils are needed at the same time. So why splitting this into two packages? And I explained enough about the "daemon" method. So you can easily keep catalyst-hook in catalyst. Btw., if you are saying that the package catalyst is only for kernel26 again and not for other kernel packages, what is the sense of this regression? This was, why catalyst-nostock was created and this was why you adopted it to catalyst. If I would need to rebuild and reupload catalyst-nostock or catalyst-fbcondecor to AUR again, this would really be a regression.

And sorry, but currently I doubt that you know what you are doing.

Vi0L0 commented on 2010-09-15 18:49

@cyberpatrol:
automatic re-compilation solutions are NOT good, are NOT safe and are far from Arch Way. fglrx module built with them is DIFFERENT than module built with makepkg.
I'm SURE that split is right way.

I've got far more work with this, but i know that's good.

Vi0L0 commented on 2010-09-15 18:49

@cyberpatrol:
automatic re-compilation solutions are NOT good, are NOT safe and are far from Arch Way. fglrx module built with them is DIFFERENT than module built with makepkg.
I'm SURE that split is right way.

I've got far more work with this, but i know what i am doing.

Vi0L0 commented on 2010-09-15 18:39

@cyberpatrol:
Going back to Arch Way is really a good way.
Plus you have got catalyst-hook package that should work fine, catalyst-hook contains only hook itself, catalyst_build_module script and needed source files, and depends on catalyst-utils package which provides other needed files.

Anonymous comment on 2010-09-15 18:38

@Vi0L0: And what is the Arch way? This package and your thinkings have nothing to do with an Arch way. The sense of catalyst-nostock was in the first place not the automatically rebuilding of a kernel, but the possibility to only need one single pacakge (catalyst resp. catalyst-nostock) for every kernel and for more than one parallel installed kernels and not only for the stock kernel, so that all those packages kernel26-fbcondecor, kernel26-howevertheywerecalled aren't necessary anymore.

This was the sense of this package. And if you shoot this down, than this is just terrible, a regression and just crap. Sorry to say that.

Vi0L0 commented on 2010-09-15 18:33

@cyberpatrol:
Going back to Arch Way is really a good way.
Plus you have got catalyst-hook package that should work fine, catalyst-hook contains only hook itself and catalyst_build_module script, and depends on catalyst-utils package which provides other needed files.

Anonymous comment on 2010-09-15 18:33

@dcelasun: Seriously, these are the comments for catalyst, Vi0L0 has announced several changes which are not comprehensible and make this package inconvenient. So this is the place for writing comments about this package. And if Vi0L0 does things with this package which infringe every Linux policies and waste my and your time and system resources, then I simply have to comment and discuss this.

And if you think, that this discussion is pointless it's your opinion. I got comments in the forums, the comments for other AUR packages and by private e-mail that several users appreciated this discussion. And there are several people who agree with me and who learned about how the system (Linux) is working.

So I won't stop posting factual comments.

Vi0L0 commented on 2010-09-15 18:32

@cyberpatrol:
Going back to Arch Way is really a good way.
Plus you have got catalyst-hook package that should work fine, catalyst-hook contains only hook itself and catalyst_build_module script, and depends on catalyst-utils package which provides other needed files.

dcelasun commented on 2010-09-15 18:17

@cyberpatrol: Seriously?!?!? After all the endless, pointless discussion in the ATI thread, now you are moving that spam here? Please don't. I (and I'm sure many more) receive notification mails for every comment, so PLEASE DO NOT spam my inbox. Thank you.

Anonymous comment on 2010-09-15 18:15

@Vi0L0: What is the sense of this inconvenient and user-unfriendly packaging? And what do users of other kernels than the stock kernel install? That was the sense of the hooks which was adopted from catalyst-nostock. I probably shouldn't have let catalyst-nostock be removed from AUR, if I knew what you are doing with this package. Too bad. You've done a good job so far. But in the meantime I'm not quite sure anymore.

Vi0L0 commented on 2010-09-15 17:54

updated to 10.9:

- v-sync is working fine again
- xorg-server 1.9 is still not supported
- looks like after quick tests theres no artifacts in firefox when using ATi 2D acceleration

Release notes is here: www2.ati.com/drivers/linux/catalyst_108_linux.pdf

IMPORTANT:
SPLIT! SPLIT! SPLIT!
catalyst package has been splited between:
- catalyst (module for stock kernel26)
- catalyst-utils (libs and stuff)

and optional packages that provides automatic re-compilation:
- catalyst-hook (auto rebuilding script and source files)
- catalyst-daemon (auto rebuilding script and source files)

I know it's maybe less "user-friendly" but this split had to be done.

Vi0L0 commented on 2010-09-15 16:10

10.9 is available and i'm downloading it already from www2.ati.com/drivers/linux/ati-driver-installer-10-9-x86.x86_64.run

Release notes isn't available atm.

Vi0L0 commented on 2010-09-08 20:16

Future of catalyst package: New auto-recompilation method - deamon instead of hook.
Request for Comments : https://bbs.archlinux.org/viewtopic.php?pid=823219#p823219

Vi0L0 commented on 2010-09-03 21:06

@pal.illes:
lib32-catalyst-utils works now with [multilib] repository
You _NEED_ to read this:
https://bbs.archlinux.org/viewtopic.php?pid=817136#p817136

Anonymous comment on 2010-09-03 20:11

using 10.8-2, evergreen, 5xxx libva/vaapi still gives garbled output on hw accel decoding. :( Otherwise glitches are less.

Also i had to add /usr/lib32 to /etc/ld.conf.so.d/ to let me run 32-bit games with lib32-catalyst-utils, it seems it was missing or linked someway else.

Vi0L0 commented on 2010-08-28 22:22

little update:
libstdc++5 has been removed from deps array, because it's not required by catalyst - gcc-libs package brings newer libstdc++ lib.
Feel free to remove libstdc++5 package from your system (ofcourse if it's not required by other package).

Vi0L0 commented on 2010-08-28 22:08

little update:
libstdc++5 has been removed from deps array, because it's not required by catalyst - gcc-libs package brings newer libstdc++ lib.
Feel free to remove libstdc++5 package from you system (ofcourse if it's not required by other package).

Vi0L0 commented on 2010-08-28 22:06

little update:
libstdc++5 has been removed from deps array, because it's no longer required by catalyst - gcc-libs package brings newer libstdc++ lib.
Feel free to remove libstdc++5 package from you system (ofcourse if it's not required by other package).

Vi0L0 commented on 2010-08-27 02:01

updated:
catalyst should now work with 2.6.36 rc2 kernel
patch wroten by Spasswolf, i found it here: http://www.phoronix.com/forums/showthread.php?t=25485
repo will be updated later

@LeCrayonVert: like i said - i uploaded libs to http://catalyst.apocalypsus.net/tarball/lib32-catalyst-utils-10.7.tar.gz

LeCrayonVert commented on 2010-08-26 19:52

Vi0L0 >> thanks. I'm on x86_64, so I may need lib32-catalyst-utils for 10.7 or is it safe to keep the 10.8 version ?

Vi0L0 commented on 2010-08-26 19:50

@LeCrayonVert: it could be unsafe, so i've uploaded older 10.7 libs...

LeCrayonVert commented on 2010-08-26 19:46

Vi0L0 >> thanks. I'm on x86_64, so I may need lib32-catalyst-utils for 10.7 or is it safe to keep the 10.8 version ?

Vi0L0 commented on 2010-08-26 19:18

@all: please.. please post about some specific catalyst's problems on ATI Bar & Grill forum: https://bbs.archlinux.org/viewtopic.php?id=57084&p=1
Aur's comments should concern on packaging. Sure it's good to write about advantages/disadvantages of new release here - it'll help a lot of people, but it's really not the place for asking how to solve some specific driver problems :).

@LeCrayonVert: sure, just uploaded 10.7, older PKGBUILDs are in tarball archives here: http://catalyst.apocalypsus.net/tarball/

LeCrayonVert commented on 2010-08-26 19:00

Vi0L0 >> how can I downgrade to 10.7 ? Can you give me a link to the old PKGBUILD ? It seems powerplay is a bit too sensitive with this one...(fan at full speed most of the time)

Anonymous comment on 2010-08-26 10:27

I have Mobility Radeon HD 5430

Anonymous comment on 2010-08-26 10:23

Hi, I have problem with catalyst. With version 10.7 I have problem with colors in firefox(and others browsers - like opera and other) and in 10.8 is like in 10.7. In version 10.6 this problem doesnt exist. Problem is in some .jpg but not all. Colors are converse - blue:pink, grey:black blue

BlindPenguin commented on 2010-08-26 06:54

Yeah i know how to fix it. I did it before because i couldn't use kde or something else without getting black or grey boxes. But now they are gone. Only firefox shows it sometimes, but that's ok. I can use opera or arora too. :)

I don't use compositing.

DarksideEE7 commented on 2010-08-25 23:17

All right scratch that it looks like I 'fixed' it. I had to disable functionality checks under openGL settings AND uncheck the first three boxes under 'Common Settings' in Configure Desktop Effects

DarksideEE7 commented on 2010-08-25 23:14

All right, thanks for the fast update. I've been waiting on this one to test in Starcraft 2.


So Vogelfrey were you having trouble with compositing? I can't enabled advanced effects in KDE 4.5 (4.5 was working fine before I updated Catalyst and lib32-catalyst-utils). I got several failed builds too....I thought it was my CFLAGS being set to -j5 but it turns out it just failing to extract the tar ball.

BTW, you can fix those black boxes in Firefox using this command IIRC:
sudo aticonfig --set-pcs-str=DDX,ForceXAA,TRUE

I also used the patched xorg-server:
xorg-server-1.8-catalyst-maximize-fix

BlindPenguin commented on 2010-08-25 21:28

Forget it.. xD

I deleted catalyst completly and removed all files with

rm -R /etc/ati/

Then i rebooted and reinstalled catalyst. It works fine. KDE is fine, Xfce is fine.. only on firefox are some black boxes, but it is not that annoying as it was before. All other browser works fine.

Nearly great work amd. :D

And thanks to Vi0L0 for the update. :)

BlindPenguin commented on 2010-08-25 19:34

Ok.. is the black/grey boxes bug gone?
I would try it, but i can't delete the workaround. I tried already this:

aticonfig --del-pcs-key=DDX,ForceXAA

but it's still there. No direct2daccel. :(

Vi0L0 commented on 2010-08-25 18:20

catalyst has been updated to 10.8 version. repo will be updated soon.

I can see NO changes except OpenGL ES 2.0 and some not_so_important fixes

WARNING:
__catalyst isn't supporting xorg-server 1.9___ which just landed
in [testing] repo and soon will be pushed to mainline
To avoid black screen use [xorg18] repository:
http://wiki.archlinux.org/index.php/ATI_Catalyst#.5Bxorg18.5D_repository

Vi0L0 commented on 2010-08-25 00:03

indeed, xserver 1.9 is in testing, so for now there is [xorg18] repo - read about it on ATI Catalyst Wiki page.

catalyst 10.8 should arrive tomorrow or today, anyway on Wednesday, so... catalyst! we are waiting for you and your xserver 1.9 support! xD

dcelasun commented on 2010-08-24 16:43

We need to wait for Ubuntu's next version, because most of the time that's when AMD releases a new driver. Well, Phoronix is wrong, catalyst, as of version 10.7, doesn't work with xorg-server 1.9. See the ATI thread for details.

Anonymous comment on 2010-08-24 16:29

Btw., I just read on Phoronix that catalyst 10.7 shall work with xorg-server 1.9.
http://www.phoronix.com/scan.php?page=news_item&px=ODQ4Ng

Anonymous comment on 2010-08-24 16:24

Why do we have to wait for Ubuntu? We have to wait for AMD to release a new version of catalyst.

dcelasun commented on 2010-08-24 15:57

Catalyst is incompatible with xorg 1.9 since it has a new ABI. We'll have to wait until Ubuntu 10.10 for that.

Anonymous comment on 2010-08-24 15:56

xorg 1.9 is in testing. Will a testing package be uploaded for this in catalyst repo?

Vi0L0 commented on 2010-08-20 19:07

@b4283:
wow! Thats very important note! Thank you. As you can see package has been updated, repo will be updated soon (edit: or maybe i will wait with it for new catalyst release).

IMPORTANT NOTE in post_install/post_upgrade:
If you are using libva and xvba-video packages then you need to reinstall them, and type this as root:
rm /usr/lib/xorg/modules/dri/*_drv_video.so

Vi0L0 commented on 2010-08-20 18:48

@b4283:
wow! Thats very important note! Thank you. As you can see package has been updated, repo will be updated soon.

IMPORTANT NOTE in post_install/post_upgrade:
If you are using libva and xvba-video packages then you need to reinstall them, and type this as root:
rm /usr/lib/xorg/modules/dri/*_drv_video.so

Anonymous comment on 2010-08-20 15:31

There's a file conflict with libva because this package puts /usr/lib/dri as a link, but the other library actually has files in it.

Vi0L0 commented on 2010-08-13 13:05

@cyberpatrol: yes, i guess today i'll find some time to do this

Anonymous comment on 2010-08-13 04:28

@Vi0L0: Will you upload the latest version (10.7-3) to the [catalyst] repo? The latest version in the repo is 10.7-2.

baghera commented on 2010-08-11 00:12

Of course I think this is a better solution :)
Thanks Vi0L0 for the awesome work you do in mantaining this package.

Vi0L0 commented on 2010-08-10 21:56

another little update - new useful functionality:
# catalyst_build_module all
- all will try to build fglrx modules for all working system's kernels

Vi0L0 commented on 2010-08-10 21:41

another little update - new useful functionality:
# catalyst_build_module build_all
- build_all will try to build fglrx modules for all working system's kernels

Vi0L0 commented on 2010-08-10 19:20

I could also turn auto on by default, but then users who don't want auto on would have to turn it off with every catalyst update.
(btw there's also catalyst_build_module autooff option to turn auto off)
So it's better to have auto off by default, and turn it on so it could work all the time :)

Vi0L0 commented on 2010-08-10 19:09

:) as you can see i've already uploaded new version - first on aur, then probably at repo. They are right that this solution is kinda bypassing KISS rule.
So now fglrx module will be built after installation, with auto re-compilation turned off by default.
But it's very easy to turn it on, just type as root:
# catalyst_build_module auto
This information is also showed after installing/updating catalyst.
I think that this new solution is the best. What do you think about it?

Anonymous comment on 2010-08-10 18:44

Usually I would agree, but in this case I don't, because catalyst won't work, if it's not rebuilt after a kernel update. And because kernel26 is being split and kernel26-headers are installed separately, what I generally don't like, it's necessary to install kernel26-headers before kernel26. I was or am still thinking about a way to rebuilding the module after a major kernel update, but I still don't know one. But I didn't thought about SyncFirst. It's of course not the ideal, but a working solution. I'd prefer having the kernel headers included in kernel26 instead of the split kernel26-headers. Then we wouldn't have this problem. With the kernels in AUR you don't have this problem, because they still have the headers included.

Anonymous comment on 2010-08-10 18:39

I think baghera is right. There should be the typical note, where to find more info or what to do. But, the automation don't have to leave completely: Put the sed's in a simple shell script "catalyst-modify-conf.sh" and tell the user to fire it up or look up the wiki. So the user can decide to do it manual or scripted.

baghera commented on 2010-08-10 18:00

I don't like a package that modify pacman.conf and mkinitcpio.conf in post install, I think it's against arch philosophy. It's ok to provide catalyst_build_module script and hook-fglrx, and of course doing the first build in post install, but let the user decide to set up automatic updates in mkinitcpio.conf with hooks. The wiki is the good place to store those infos about the optional procedure.

What others think about it?

Vi0L0 commented on 2010-08-10 13:23

Little update. Changes are important so please read about them here: https://bbs.archlinux.org/viewtopic.php?pid=806747#p806747

Vi0L0 commented on 2010-08-09 02:49

updated catalyst 10.7 testing driver with OpenCL 1.1 support is at AUR's catalyst-test: http://aur.archlinux.org/packages.php?ID=34773

Vi0L0 commented on 2010-08-09 02:42

updated catalyst 10.7 driver with OpenCL 1.1 support is at AUR's catalyst-test: http://aur.archlinux.org/packages.php?ID=34773

Anonymous comment on 2010-08-01 14:31

@Inkaine: lib32-catalyst-utils is an optional dependency, because it needs to be installed additionally to catalyst (both packages need to be installed) if someone wants to run 32 bit games on an x86_64 system. Optdepends doesn't mean that you must install these packages. It just means that you have to install them if you need a certain optional feature. This is the case with running i686 software on x86_64 systems. And this is a good way to inform people about this fact.

Inkaine commented on 2010-08-01 08:38

lib32-catalyst-utils is no optional dependency for catalyst. It was if it provided new features for catalyst itself but was not essentially required. That's not the case. It's the other way around that lib32 can't exist without the main package. You should remove that optdep again when you update next.

Vi0L0 commented on 2010-07-26 18:15

Update.
Some quick tests show:
- 2.6.34 kernel support (2.6.35-rc6 also works fine), so no need of patch
- fixed textures in q3a based games
- looks like artifacts with ATi 2D accel still exist
- Eyefinity support (not tested)

Vi0L0 commented on 2010-07-26 17:58

Update.
Some quick tests show:
- 2.6.34 kernel support (2.6.35-rc6 also works fine), so no need of patch
- fixed textures in q3a based games
- looks like artifacts with ATi 2D accel still exist

Vi0L0 commented on 2010-07-26 17:21

10.7 is already online @ http://www2.ati.com/drivers/linux/ati-driver-installer-10-7-x86.x86_64.run
Im downloading it right now. Then as always: build, some tests, and updates :)

Vi0L0 commented on 2010-07-23 21:56

little update = no new funcionality, just cleaning:
# catalyst_build_module remove
after removing /lib/modules/some_kernel/video/fglrx.ko and /lib/modules/some_kernel/video/ will now _check_ if directory /lib/modules/some_kernel still exist and then will remove /lib/modules/some_kernel if its empty or tell that it is unused. That _check_ is needed because sometimes system is automatically removing directory /lib/modules/some_kernel when it's empty (ive just observed that, till now i was thinking system is never removing this dir automagically) - and so in that situation 'catalyst_build_module remove' _was_ echoing (just echoing, no other action) some stupid false things like error and then that it is removing previously removed directory ;P
Still you _don't need_ to update, its just cleaning.

Vi0L0 commented on 2010-07-23 21:46

little update = no new funcionality, just cleaning:
# catalyst_build_module remove
after removing /lib/modules/some_kernel/video/fglrx.ko and /lib/modules/some_kernel/video/ will now _check_ if directory /lib/modules/some_kernel still exist and then will remove /lib/modules/some_kernel or tell that it is unused. That _check_ is needed because sometimes system is automatically removing directory /lib/modules/some_kernel when it's empty (ive just observed that, till now i was thinking system is never removing this dir automagically) - and so in that situation 'catalyst_build_module remove' _was_ echoing (just echoing, no other action) some stupid false things like error and then that it is removing previously removed directory ;P
Still you _don't need_ to update, its just cleaning.

Vi0L0 commented on 2010-07-23 21:40

little update = no new funcionality, just cleaning:
# catalyst_build_module remove
after removing /lib/modules/some_kernel/video/fglrx.ko and /lib/modules/some_kernel/video/ will now _check_ if directory /lib/modules/some_kernel still exist and then will remove /lib/modules/some_kernel or tell that it is unused. That _check_ is needed because sometimes system is automatically removing directory /lib/modules/some_kernel when it's empty (ive just observed that, till now i was thinking system is never removing this dir automagically) - and so in that situation 'catalyst_build_module remove' _was_ echoing some stupid false things ;P
Still you _don't need_ to update, its just cleaning.

Vi0L0 commented on 2010-07-23 21:35

little update = no new funcionality, just cleaning:
# catalyst_build_module remove
after removing /lib/modules/some_kernel/video/fglrx.ko and /lib/modules/some_kernel/video/ will now check if directory /lib/modules/some_kernel still exist and then will remove /lib/modules/some_kernel or tell that it is unused. Thats because sometimes system is automatically removing directory /lib/modules/some_kernel when it's empty (ive just observed that, till now i was thinking system is never removing this dir automagically) - and so in that situation 'catalyst_build_module remove' _was_ echoing some stupid false things ;P

Anonymous comment on 2010-07-23 11:05

Thank you very much kensai, dickeny and ViOLO. :)

Vi0L0 commented on 2010-07-21 13:08

Thank you mishigaki, im glad to hear that, but im only cleaning up here ;). Most of the great job was done by kensai and dickeny - so all our thanks and love xD goes to them.

Anonymous comment on 2010-07-21 04:42

This is an amazing quality product. Thank you very much ViOLO!!!!!!!!!!!!!!!!!!!!!!!!!! ATI should learn from you.

Vi0L0 commented on 2010-07-12 21:45

Read more about new features of catalyst package at http://wiki.archlinux.org/index.php/ATI_Catalyst#Stock_and_Custom_Kernels

Anonymous comment on 2010-07-11 18:50

@skodabenz: I already suggested this. Vi0L0 modified it a bit to this info:
echo "For troubleshooting visit:"
echo "http://wiki.archlinux.org/index.php/ATI_Catalyst"

The aticonfig command can be found there, too.

Anonymous comment on 2010-07-11 18:31

Thanks for this PKGBUILD. I tried catalyst-nostock but somegow the fglrx module was not being built (kernel 2.6.35 rc4 x86_64). You should add this info to the post_install script (from http://aur.archlinux.org/packages/catalyst-nostock/catalyst-nostock/catalyst-nostock.install)

If you experience problems with e.g. black artifacts you can
try: aticonfig --set-pcs-str=DDX,ForceXAA,TRUE

I experienced black artifacts while using firefox.

Vi0L0 commented on 2010-07-11 15:06

little update:
+ 'catalyst_build_module remove' shall now correctly remove empty /lib/modules/kernel_name directories
+ some cleaning

theking2 commented on 2010-07-11 11:55

Thanks ViOLO, Works fantastic on the EeeBox B206 with Radeon HD 3400 and X v1.8.1

theking2 commented on 2010-07-11 10:27

I get
(EE) Failed to load module "fgl.renamed.libglx" (module does not exist. 0)
(WW) fglrx: No matching Device section for instance (BusID PCI:003:0:1) found

have a radeon HD3500

agapito commented on 2010-07-10 15:40

hmmm I think md5sums from ATI driver are wrong... maybe ati changed installer?

agapito commented on 2010-07-10 15:37

hmmm I think md5sums from ATI driver are wrong... maybe ati changed installer?

Vi0L0 commented on 2010-07-10 15:14

Update! Lots of changes! More here: http://bbs.archlinux.org/viewtopic.php?pid=789476#p789476

Anonymous comment on 2010-07-09 19:02

Thanks for the answer ViOLO

Vi0L0 commented on 2010-07-09 18:59

@Cory: 10.5 is working with patched xserver 1.8. You may ie get it from [catalyst-10.5] repo. All needed info is here: http://wiki.archlinux.org/index.php/ATI_Catalyst#Catalyst.27s_repositories Also look at troubleshooting section...

Anonymous comment on 2010-07-09 18:42

Catalyst 10.5 support xorg 1.8? i have strange artifacts (catalyst 10.6 xorg 1.8)

Anonymous comment on 2010-07-09 15:59

i use xorg 1.8.0
so i forced 1.8.0 in PKGBUILD

it's working

Anonymous comment on 2010-07-06 21:10

quote mpausch! Same problem here! i'm switched back to 10.5 and xorg 1.8.

Anonymous comment on 2010-07-06 17:23

in my case 10.6 sucked way too much... too much drawing/display errors (in thunderbird, vlc, konsole, etc)... switched back to 10.5 and Xorg-Server 1.8 (http://aur.archlinux.org/packages.php?ID=36328)... now I can enjoy again :-)

Anonymous comment on 2010-07-06 17:12

I am getting the Black Screen of Death after Xorg launch with these drivers. Could someone help me out here? I am using a 4870HD video card. and I am almost sure it worked before with 10.5 (though not sure which Xorg version I did use)

thnx

Det commented on 2010-07-03 18:28

I don't think at least with the next version as AMD really don't concentrate on the Linux Catalyst driver.
Btw. the "# Previous Maintainer:..." line could just be "Contributor:..." :).

Vi0L0 commented on 2010-07-03 17:50

hah, i builded xorg-server-dev just before reading your post, so i knew it wont work but i was curious what will happen...

dlopen: /usr/lib/xorg/modules/drivers/fglrx_drv.so: undefined symbol: savedScreenInfo

So bug is pretty similar to your.
Im just wondering when will catalyst support new abi and new xserver ;)...

Det commented on 2010-07-03 17:34

Well, actually I don't even have a ATI card :). Also with RC4 the xorg-server 1.9 jumped to use the v8 ABI (Application Binary Interface), meaning video drivers need to be updated from the v7 ABI to v8 to support the new X.

I originally thought it was a different issue, when I asked about it in the xorg mailing list: http://lists.freedesktop.org/archives/xorg/2010-July/050640.html

E: if you wondered my "I never said" part of my last post 'Ollila' is my other account here :).

Det commented on 2010-07-03 17:32

Well, actually I don't even have a ATI card :). Also with RC4 the xorg-server 1.9 jumped to use the v8 ABI (Application Binary Interface), meaning video drivers need to be updated from the v7 ABI to v8 to support the new X.

I originally thought it was a different issue, when I asked about it in the xorg mailing list: http://lists.freedesktop.org/archives/xorg/2010-July/050640.html

Det commented on 2010-07-03 17:27

Well actually I don't even have a ATI card and with RC4 the xorg-server 1.9 jumped to use the v8 ABI, meaning all video drivers need to be updated from v7 to v8.

I asked about it in the xorg mailing list, as X didn't start (thought it was a different issue then): http://lists.freedesktop.org/archives/xorg/2010-July/050640.html

Vi0L0 commented on 2010-07-03 17:14

@Det: did you test catalyst for working with xorg-server 1.9 rc4? :)

Vi0L0 commented on 2010-07-03 14:02

update - added package() function ;)

Det commented on 2010-07-03 08:16

Well, I'm using the package() function with my own packages. I never said there was a _need_ to use it, btw. With single packages it's only benefit is to have fakeroot being used as little as possible.

I never, ever had any problems with fakeroot myself but I've added the function to my packages _just in case_ ;).

Det commented on 2010-07-03 08:15

Well, I'm using the package() function with my own packages. I never said there was a _need_ to use it, btw. With single packages it's only benefit is to have fakeroot being used as little as possible.

I never, ever had any problems with fakeroot myself but I use it _just in case_ ;).

Vi0L0 commented on 2010-07-03 02:54

@Ollila: I asked at irc to clear it out, and theres no need of using package() function.
Are you really using repackaging with catalyst?
Although i will probably add this function with next release.

Anonymous comment on 2010-07-02 20:04

@Vi0l0, wrong: http://bbs.archlinux.org/viewtopic.php?id=100069
Also as you can see from the pacman changelog (http://projects.archlinux.org/pacman.git/tree/NEWS?id=v3.4.0):
makepkg
- repackaging without a package() function is deprecated

Vi0L0 commented on 2010-07-02 15:22

Splitting catalyst and lib32-catalyst-utils? Please read this: http://bbs.archlinux.org/viewtopic.php?pid=785253#p785253

@Ollila: package() function should be use only for splitted packages.

Anonymous comment on 2010-07-02 13:27

At least the package() function should be used. dkxls's PKGBUILD is also a bit cleaner.

Anonymous comment on 2010-06-27 23:48

Btw., even if AUR would support split packages it wasn't a good idea, because the AUR scripts like yaourt and aurbuild would build and install every package by default. And it's really not nice if people were forced to make a multilib system even if they don't need or want it.

And it's really no problem to install an additional package lib32-catalyst-utils if a multilib system is really needed.

Anonymous comment on 2010-06-27 23:33

@dkxls: Including lib32-catalyst-utils into catalyst is not a good idea, because lib32-catalyst-utils is only used for running 32-bit software (mostly games) on x86_64 systems. So lib32-catalyst-utils is not needed by everyone and should not be built by the package catalyst and catalyst shouldn't and doesn't need to depend on lib32-catalyst-utils. And AUR doesn't support split packages. So, please, don't do this.

Anonymous comment on 2010-06-27 23:07

We had in April the discussion about merging the 'lib32-catalyst-utils' to this packages. I had today a couple of hours time to do this merge. It was straight forward, as already mentioned earlier. And just to clarify it: there is no need to change the arch variable in the packages function. I post the PKGBUILD here: http://pastebin.com/HBawLG0g
The PKGBUILD is based on the 'catalyst' and 'lib32-catalyst-utils' one, where did besides merging them only some indentations and comments to make the reading and understanding easier. :-)

With this PKGBUILD the 'catalyst' packages is built for either 'i686' or 'x86_64', where in the case of 'x86_64' (and only in this case) a second package called 'lib32-catalyst' is build. The 'x86_64'-users can than freely chose to install the 'lib32-catalyst' package or not.

I choose the name to 'lib32-catalyst' since it only contains the userspace libs and no utilities!

I would be happy about some comments (maybe in the forum). And if Vi0L0 will pick up this PKGBUILD, than we don't need the 'lib32-catalyst-utils' any more and we have a clean, all-inclusive 'catalyst'-PKGBUILD.

Cheers,
Armin

Anonymous comment on 2010-06-23 17:30

update: with xorg-server 1.8 I still have the very rare graphic errors (for example sometimes when I open KDE-konsole).
Don't need xorg-server with maximize-fix anymore.

Anonymous comment on 2010-06-19 09:12

after installation of 10.6 I had to add ' BusID "PCI:1:0:0" ' to Device-Section in xorg.conf (with 10.5 it worked without it).

in KDE with compositing activated I sometimes have graphic errors. Not reproducable.
Using Xorg-Server 1.7 with Maximize fix from AUR (http://aur.archlinux.org/packages.php?ID=35686 )

Inkaine commented on 2010-06-19 09:03

Guys, for discussions please use the forum. This is for package related issues and notifications not for general chit-chat and discussions about upstream related performance.

dcelasun commented on 2010-06-19 06:56

Define "bad", please?

Inkaine commented on 2010-06-17 21:08

Usually using an old lib32-catalyst version with the latest driver shouldn't be a major problem. I wouldn't recommend the other way round though. Will update in a minute anyways. ;-)

Anonymous comment on 2010-06-17 18:21

@Vi0L0 Alright, thanks for clearing that up for me.

Vi0L0 commented on 2010-06-17 18:20

@AptQuadruped: its not safe. You may always use lib32-catalyst-utils 10.6-1 from my repo, address you will find on Archlinux ATi Catalyst Wiki page. lib32-catalyst-utils 10.6-1 will work on every kernel, so feel free to use it.

Anonymous comment on 2010-06-17 18:09

Is it safe to use "lib32-catalyst-utils 10.5-1" with the 10.6 drivers, or should I wait until they are also updated?

Vi0L0 commented on 2010-06-17 17:24

bad versioning of 10.6 in amdcccle is known bug ;)

Fabioamd87 commented on 2010-06-17 15:59

solved reinstalling it.

dcelasun commented on 2010-06-17 15:35

Same here, 10.6 reports itself as 10.5 on the control center.

Fabioamd87 commented on 2010-06-17 15:16

problem: after update (fine) still 10.5 on amdcccle

Vi0L0 commented on 2010-06-16 20:21

updated to 10.6, repos will be updated soon

Whats new in 10.6?:
- working fine 2d acceleration (before using i did `rm /etc/ati/amdpcsdb`);
- official support for 2.6.33 kernel;
- official support of OGL 3.3/4.

Vi0L0 commented on 2010-06-16 14:04

!!! catalyst's repositories are movin into new home: http://bbs.archlinux.org/viewtopic.php?pid=776039#p776039
Please update your repos addresses.

Vi0L0 commented on 2010-06-15 15:03

Thanks mwc, updated.

Anonymous comment on 2010-06-14 18:07

The package build should explicitly version the libatical provide to the package version, something like:

provides=("libatical=${pkgver}" ...)

This fixes a build issue with the amdstream PKGBUILD which requires libatical>=10.4, either the version shipped in the driver (fglrx) or another version distributed outside fglrx. As it stands now, makepkg reports that the dependency 'libatical>=10.4' is not satisfied. Modifying the provide to libatical=10.5 fixes this for me.

Anonymous comment on 2010-06-11 13:24

If you save the file into the same directory where the PKGBUILD is saved then it is not downloaded. If this file isn't there then it is downloaded.

Anonymous comment on 2010-06-11 12:35

Wait the PKGBUILD doesn't handle the downloads does it? sorry I'm new to Arch

Anonymous comment on 2010-06-11 12:30

Hi, is it possible to make the PKGBULID not download the ati-driver....rum file if it's already downloaded?

Anonymous comment on 2010-06-04 14:05

Well, I saw, that makepkg replaces the `uname -r` by the actual kernel version, but I didn't know yet that it also replaces a static version number.

Vi0L0 commented on 2010-06-04 13:51

@cyberpatrol: yes with 10.5 I forgot to check that line... but it's really not so important - makepkg will always change KERNEL_VERSION line with kernel version you are using ;).
Although ive changed that just to make it cleaner ;)

Vi0L0 commented on 2010-06-04 13:50

@cyberpatrol: yes with 10.5 I forget to check that line... but it's really not so important - makepkg will always change KERNEL_VERSION line with kernel version you are using ;).
Although ive changed that just to make it cleaner ;)

Anonymous comment on 2010-06-04 02:16

In the file catalyst.install you should probably change the line
KERNEL_VERSION=2.6.34-ARCH
to
KERNEL_VERSION=`uname -r`
for people who are using the stable repos instead of [testing].

dcelasun commented on 2010-05-28 18:09

@RoboSK: This question has been asked hundreds of times, literally. Please, next time just search the forum or at least read the comments here before posting.

Anyway you need to this before pacman -S catalyst:

pacman -S xorg-server (if you are installing for the first time)
pacman -Rd libgl (catalyst provides its own gl implementation)
pacman -S catalyst (or pacman -U catalyst-[something].xz if you built it with makepkg)

Anonymous comment on 2010-05-28 17:37

hi, i have ATI 3870 and install aur catalyst 10.4/10.5 fail - clear instalation...only ssh...

-Sy catalyst
...
resolving dependencies...
looking for inter-conflicts...

Targets (2): libgl-7.7.1-1 xorg-server-1.7.6-3
...
loading package data...
checking dependencies...
error: replacing packages with -U is not supported yet
error: you can replace packages manually using -Rd and -U
error: failed to prepare transaction (conflicting dependencies)
:: catalyst: conflicts with libgl

pls help
thx

agapito commented on 2010-05-27 00:52

Works fine here 1.8 / 2.6.34

Anonymous comment on 2010-05-26 23:46

Thx a lot for your hard work Vi0L0. Wine runs better now :D

baghera commented on 2010-05-26 23:45

@ Leonid.I

netkit-bsd-finger contains /usr/bin/finger , which is used in /etc/acpi/ati-powermode.sh

grep finger /etc/acpi/ati-powermode.sh
user=`finger| grep -m1 ":$displaynum " | awk '{print $1}'`
user=`finger| grep -m1 ":$displaynum" | awk '{print $1}'`

Vi0L0 commented on 2010-05-26 23:40

Thanks :) Updated, repos will be updated tomorrow.

This release is pretty nice, 10.5:
opengl 3.3/4 support;
xvba working without bad_colours/trashes on my hd4850 xD
resolved issues pertaining to Hybrid CrossFire, kernel panics on hibernation, and suspend-and-resume fixes;

Ps. wine - its probably working good (without bugs) and faster with wine 1.2 RC

Tested on x86_64, 2.6.34 kernel, patched xserver 1.8

Vi0L0 commented on 2010-05-26 23:40

Updated, repos will be updated tomorrow.

This release is pretty nice, 10.5:
opengl 3.3/4 support;
xvba working without bad_colours/trashes on my hd4850 xD
resolved issues pertaining to Hybrid CrossFire, kernel panics on hibernation, and suspend-and-resume fixes;

Ps. wine - its probably working good (without bugs) and faster with wine 1.2 RC

Tested on x86_64, 2.6.34 kernel, patched xserver 1.8

Anonymous comment on 2010-05-26 22:39

Why is there netkit-bsd-finger dependency? I can build successfully without it, and it is not required in the Release Notes from AMD...

ChemBro commented on 2010-05-26 21:21

Catalyst 10.5 is out. https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/drivers/linux/ati-driver-installer-10-5-x86.x86_64.run

dcelasun commented on 2010-05-26 18:39

See the ATI/AMD Bar&Grill thread on the forum for a patched xorg18 repo by Vi0L0. Alternatively, you can use http://aur.archlinux.org/packages.php?ID=36328

Anonymous comment on 2010-05-26 18:31

Does the Catalyst 10.4 run under xorg server 1.8.x? And if: how?

Anonymous comment on 2010-05-19 19:19

Builds and installs fine (had to install some other stuff to get it installed).
But after startx, whole systems crash and I cant grep any log output, too over ssh nothing is logged.

Maybe a problem with the .33 kernel, will setup this weekend a test installation, maybe I get it working with some tweaks.

Anonymous comment on 2010-05-19 09:21

@Vi0L0

Thanks a lot! I'm gona test it asap. :)

Vi0L0 commented on 2010-05-19 00:18

@magys: nevermind asking, heres tarball needed to build this catalyst 10.3 with openGL4 supp: http://dl.dropbox.com/u/4321633/tarball/10.3_ogl4.tar.gz
It should work fine with 2.6.33/34 kernels, but you need xserver 1.6 - you can find it in [catalyst-10.3] repo, more here: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491

Plus if you are using x86_64 heres tarball of lib32-catalyst-utils: http://dl.dropbox.com/u/4321633/tarball/lib32-catalyst-utils-10.3_ogl4.tar.gz

NOTE: you need to have ati_catalyst_10.3_ogl4_preview_linux_march29.zip file downloaded and locateted next to PKGBUILD, i dunno why but it cannot be downloaded correctly with makepkg

Oh and... i havent really tested it :) but it should work, at least its building correctly ;P

Vi0L0 commented on 2010-05-18 23:47

@magys: nevermind asking, heres tarball needed to build this catalyst 10.3 with openGL4 supp: http://dl.dropbox.com/u/4321633/tarball/10.3_ogl4.tar.gz
It should work fine with 2.6.33/34 kernels, but you need xserver 1.6 - you can find it in [catalyst-10.3] repo, more here: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491

NOTE: you need to have ati_catalyst_10.3_ogl4_preview_linux_march29.zip file downloaded and locateted next to PKGBUILD, i dunno why but it cannot be downloaded correctly with makepkg

Oh and... i havent really tested it :) but it should work, at least its building correctly ;P

Vi0L0 commented on 2010-05-18 23:46

@magys: nevermind asking, heres tarball needed to build this catalyst 10.3 with openGL4 supp: http://dl.dropbox.com/u/4321633/tarball/10.3_ogl4.tar.gz
It should work fine with 2.6.33/34 kernels, but you need xserver 1.6 - you can find it in [catalyst-10.3] repo, more here: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491

NOTE: you need to have ati_catalyst_10.3_ogl4_preview_linux_march29.zip file downloaded and locateted next to PKGBUILD, i dunno why but it cannot be downloaded correctly with makepkg

Vi0L0 commented on 2010-05-18 23:07

@magys: true, but that wasn't 10.4 - it was 10.3 without xserver 1.7/1.8 support... if you still want it - i can prepare pkgbuild

Vi0L0 commented on 2010-05-18 23:02

@magys: true, but that wasn't 10.4 - it was 10.3 without xserver 1.7/1.8 support... if you still want it - leave me a message, i will prepare pkgbuild

Anonymous comment on 2010-05-18 18:53

There was a catalyst 10.4 with OpenGL 4 released, is there a package for it?
It seems that the OpenGL4 driver fixes some null pointer problems with wine so that it is possible again to launch source based games.

I tried to modify the PKGBuild but it was not working. There seems to be a new strucutre in the installer files.


Here is the link for the OpenGL4 Driver: http://support.amd.com/us/kbarticles/Pages/Catalyst-OpenGL-preview-driver.aspx

Inkaine commented on 2010-05-02 13:54

As it says in your quote:

[ 168.047] compiled for 1.7.1, module version = 8.72.5
[ 168.047] Module class: X.Org Video Driver
[ 168.047] [atiddxSetup] X version mismatch - detected X.org 7.1.0.901, required X.org 7.5.1.0

This catalyst is specifically for xorg-server 1.7.1 from extra. For support of xorg 1.8 you will need a patch in xorg-server package as stated by the package maintainer below. Lemme quote:

Comment by: Vi0L0 on Wed, 28 Apr 2010 23:14:03 +0000
> [catalyst] repo (now with xorg-server 1.7) has been updated, fresh [catalyst-xorg18] repo has arrived, [catalyst-test-xorg18] repo got different address now.
> More here: http://bbs.archlinux.org/viewtopic.php?pid=750810#p750810

Reading always helps before complaining. ;-)

P.S.: This package might come handy: http://aur.archlinux.org/packages.php?ID=36328

Inkaine commented on 2010-05-02 13:53

As it says in your quote:

[ 168.047] compiled for 1.7.1, module version = 8.72.5
[ 168.047] Module class: X.Org Video Driver
[ 168.047] [atiddxSetup] X version mismatch - detected X.org 7.1.0.901, required X.org 7.5.1.0

This catalyst is specifically for xorg-server 1.7.1 from extra. For support of xorg 1.8 you will need a patch in xorg-server package as stated by the package maintainer below. Lemme quote:

Comment by: Vi0L0 on Wed, 28 Apr 2010 23:14:03 +0000
> [catalyst] repo (now with xorg-server 1.7) has been updated, fresh [catalyst-xorg18] repo has arrived, [catalyst-test-xorg18] repo got different address now.
> More here: http://bbs.archlinux.org/viewtopic.php?pid=750810#p750810

Reading always helps before complaining. ;-)

Anonymous comment on 2010-05-02 13:02

I had got startx failed.Here is log:
[ 168.037] (II) LoadModule: "fglrx"
[ 168.037] (II) Loading /usr/lib/xorg/modules/drivers/fglrx_drv.so
[ 168.047] (II) Module fglrx: vendor="FireGL - ATI Technologies Inc."
[ 168.047] compiled for 1.7.1, module version = 8.72.5
[ 168.047] Module class: X.Org Video Driver
[ 168.047] [atiddxSetup] X version mismatch - detected X.org 7.1.0.901, required X.org 7.5.1.0
[ 168.047] (II) UnloadModule: "fglrx"
[ 168.047] (II) Unloading /usr/lib/xorg/modules/drivers/fglrx_drv.so
[ 168.047] (EE) Failed to load module "fglrx" (module requirement mismatch, 0)
[ 168.047] (EE) No drivers available.

My xorg is from testing repo.

Anonymous comment on 2010-05-01 14:53

Thanks a lot buddy,
you really saved my day :D

Anonymous comment on 2010-04-30 23:27

@ViOLO: LoL, I didn't read well cuz my chronic lack of time, sorry, but ok all works, really, thank you

Vi0L0 commented on 2010-04-30 21:40

@bfx: cuz dropbox doesnt allow to view packages with the browser... read post carefully and add repo to /etc/pacman.conf

Vi0L0 commented on 2010-04-29 23:00

@bfx: catalyst 10.3 repo is still available: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491
It should work fine until 2.6.34-ARCH appears.
If you are using kernel other than 2.6.33-ARCH heres tarball for 10.3: http://dl.dropbox.com/u/4321633/tarball/catalyst-10.3.tar.gz
For kernel >= 2.6.34-rc4 use fglrx-2.6.34-rc4.patch (along with fglrx-2.6.33.patch)

Vi0L0 commented on 2010-04-29 22:59

@bfx: catalyst 10.3 repo is still available: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491
It should work fine until 2.6.34-ARCH appears.
If you are using other kernel heres tarball for 10.3: http://dl.dropbox.com/u/4321633/tarball/catalyst-10.3.tar.gz
For kernel >= 2.6.34-rc4 use fglrx-2.6.34-rc4.patch (along with fglrx-2.6.33.patch)

Anonymous comment on 2010-04-29 22:45

Catalyst control center, contrast, brightness... doesn't work... please, can anybody make the old package Catalyst 10.3 available? TYA

Vi0L0 commented on 2010-04-29 21:02

@Inkaine, indeed. Updated. Ati has changed those files:
/etc/ati/amdpcsdb.default
/etc/ati/control
/etc/ati/signature

Inkaine commented on 2010-04-29 18:27

==> Validating source files with md5sums...
ati-driver-installer-10-4-x86.x86_64.run ... FAILED
catalyst.sh ... Passed
amdcccle.desktop ... Passed
atieventsd.sh ... Passed
fglrx-2.6.33.patch ... Passed
fglrx-2.6.34-rc4.patch ... Passed
==> ERROR: One or more files did not pass the validity check!

I'm getting a different md5 sum (e5a2343e2a6a8ea2fedd272a97fcadcc) on the driver file, even after redownloading a second time. The file probably changed again after yesterday's upload since it's also giving me the date Apr 29th 2010 19:08.

Vi0L0 commented on 2010-04-29 00:07

[xorg18] has been moved to [testing] :/

Vi0L0 commented on 2010-04-28 23:14

[catalyst] repo (now with xorg-server 1.7) has been updated, fresh [catalyst-xorg18] repo has arrived, [catalyst-test-xorg18] repo got different address now.
More here: http://bbs.archlinux.org/viewtopic.php?pid=750810#p750810

Vi0L0 commented on 2010-04-28 20:09

Thanks agapito
Updated.
Not much change between this and catalyst-test...
xserver 1.7 support, and patched xserver 1.8 support.
Version of 2d driver = 8.72.5
Version of opengl lib = 3.2.9756
For kernel newer than 2.6.32 we still need to use (included) patches.
Works also with 2.6.34-rc5

agapito commented on 2010-04-28 18:32

10.4 released

https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/drivers/linux/catalyst_104_linux.pdf

Vi0L0 commented on 2010-04-28 13:01

@kullfar: catalyst 10.4 should support xserver 1.7
plus there is a package called catalyst-test http://aur.archlinux.org/packages.php?ID=34773 which supports xserver 1.7
There is also a repo with catalyst-test and patched for it xserver 1.8 which works pretty nice, read here: http://bbs.archlinux.org/viewtopic.php?pid=740353#p740353

Vi0L0 commented on 2010-04-28 11:57

@kullfar: catalyst 10.4 should support xserver 1.7
plus there is a package called catalyst-test http://aur.archlinux.org/packages.php?ID=34773 which supports xserver 1.7
There is also a repo with catalyst-test and patched for it xserver 1.8 with works pretty nice, read here: http://bbs.archlinux.org/viewtopic.php?pid=740353#p740353

kullfar commented on 2010-04-28 05:41

xorg 1.8 is coming.
the worsest thing, that can happens - xorg 18 in extra before catalyst supporting xorg 17...
any news on xorg 1.7 support?

Vi0L0 commented on 2010-04-13 11:18

Yes, both !radeon in MODULES and nomodeset in menu.lst got same effect. Although i believe nomodeset is more appropriate way.

Vi0L0 commented on 2010-04-13 11:10

Yes, both !radeon in MODULES and nomodeset in menu.lst got same effect.

dcelasun commented on 2010-04-13 08:42

Well, modesetting is turned on if an ATI card is detected, so I guess !radeon in rc.conf and nomodeset in menu.lst have the same effect :)

Anonymous comment on 2010-04-13 08:36

@kullfar: had the black-screen-problem dcelasun wrote about. added !radeon to rc.conf and it works now. nomodeset is not set. maybe this is of interest.

dcelasun commented on 2010-04-13 08:06

Maybe !radeon in rc.conf is doing the same job. Try removing it and see if fglrx still works :)

For me, I have no ati-related stuff in MODULES array and without nomodeset X freezes at startup. Just a black screen, nothing else.

kullfar commented on 2010-04-13 07:54

sounds familiar... KMS...
but.. I have 2.6.33 and this package installed, and didn't add "nomodeset" option to grub.
and all (incl. compiz) is working. What am I doing wrong? ;-)
maybe I have default options in grub somewhere?
or this option just makes smth better, but we can live wo it as well?
or am I just too lucky? ;-)

another point
I have !drm !ttm !radeon fglrx in MODULES in rc.conf.
maybe this makes me lucky?

dcelasun commented on 2010-04-12 08:29

Kernel 2.6.33 has KMS on by default and fglrx doesn't support KMS yet. Hence the "nomodeset" line.

kullfar commented on 2010-04-12 07:12

Vi0L0, why 'nomodeset'? what for? It's working for me wo that.

Vi0L0 commented on 2010-04-10 13:10

Nope, theres no way to override architecture in split packages... and lib32-catalyst-utils is only for x86_64 as we all know. So the whole splitting idea here is pointless.

Anonymous comment on 2010-04-10 07:17

I have to disagree with Inkaine. You can build multiple packages with one PKGBUILD (as former mentioned by gdamjan). This would enable us to have all the knowledge about the catalyst driver (where to put the files, symlinks, ...) in one PKGBUILD. It would even help reducing bugs! You can define for each package separate variables like "depends", "conflicts", "provides" or "makedepends", so the outcome would be two packages, where you only have to install the driver package!

This leads to a "clean" system for everyone who wants to. All the others that need to have the compatibility libs have them right away in the second package!

So I don't see any reason from the technical point of view why the two packages couldn't be merged. (In fact I would like it.) :-)

Inkaine commented on 2010-04-09 22:36

Regarding integrating lib32 into this package, it's not about being fair with me (I am the maintainer) but it's about the arch structure. It would force everyone to install an unclean system that they may not wish. The lib32-system are compatability libraries - installation of 32bit files into a clean 64bit environment. By including them here you would force every 64bit user to install them. There has always been a reason for this split.

If you're using yaourt and want it easier (complaining about the double download), its there where you should complain. Yaourt did download all source files locally before and stored that them at a central location but they removed it. Its a yaourt feature request and nothing that out of your personaly easiness should change archs package structure. Despite Violo told the way to go. ;-)

Vi0L0 commented on 2010-04-09 22:27

! Add nomodeset to your kernel line in /boot/grub/menu.lst , so it may looks like this:
kernel /boot/vmlinuz26 root=/dev/disk/by-uuid/1aa731c0-f626-4d01-a551-8eed3c2a35f3 ro nomodeset vga=794 splash quiet

Vi0L0 commented on 2010-04-09 14:36

update - added 'openssl-compatibility' to depends array

Vi0L0 commented on 2010-04-09 14:20

!!!
to get it work with old xserver 1.6 you need to have /usr/lib/libssl.so.0.9.8 and /usr/lib/libcrypto.so.0.9.8
you may get it with AUR's openssl-compatibility http://aur.archlinux.org/packages.php?ID=36308
(read comments to get PKGBUILD for x86_64 version)

Vi0L0 commented on 2010-04-02 21:35

@gdamjan: i think it would be unfair for lib32-catalyst-utils maintainer, i also believe hes got some knowledge about creating that package. Besides its not so hard to copy ati-driver-installer into lib32-catalyst-utils directory and makepkg there...

Vi0L0 commented on 2010-04-02 20:27

@gdamjan: good idea, i will think about that.

gdamjan commented on 2010-04-02 17:24

how about integrating lib32-catalyst-utils package into this one?
so that the 'source' archive is not downloaded twice, the Arch split packages support can be used for this.

sovok_kpss commented on 2010-03-29 01:59

Yes, I'm talking about realtime kernel. I've tried some patches that worked with for older kernels but 2.6.33.

sovok_kpss commented on 2010-03-29 01:58

@Vi0L0: Yes, I'm talking about realtime kernel. I've tried some patches that worked with for older kernels but 2.6.33

Vi0L0 commented on 2010-03-29 01:33

@sovok_kpss: you mean realtime kernel? - never heard about any patches :/.
btw Thats a good qestion :)

Vi0L0 commented on 2010-03-28 23:49

@sovok_kpss: you mean realtime kernel? - never heard.
btw Thats a good qestion :)

sovok_kpss commented on 2010-03-28 16:17

How can I install this package with on rt-kernel? Maybe some patches?

Anonymous comment on 2010-03-25 18:16

And another interesting release came:
http://support.amd.com/us/kbarticles/Pages/Catalyst-OpenGL-preview-driver.aspx

https://a248.e.akamai.net/f/674/9206/0/www2.ati.com/drivers/hotfix/ATI_Catalyst_10.3_OGL4_Preview_Linux.zip

Vi0L0 commented on 2010-03-25 17:25

main [catalyst] repo has been updated with catalyst 10.3
more here: http://bbs.archlinux.org/viewtopic.php?pid=712491#p712491

still catalyst 10.4 beta with xserver 1.7 support ownz http://aur.archlinux.org/packages.php?ID=34773

Vi0L0 commented on 2010-03-25 15:39

update
10.3 still without xserver 1.7 support
rel notes: http://www2.ati.com/drivers/linux/catalyst_103_linux.pdf

tested on hd4850 i686 2.6.33.1 kwin - fine

Vi0L0 commented on 2010-03-25 14:02

10.3 is out, only make some test and update...

kralyk commented on 2010-03-25 10:57

@Vi0L0: Perfect, thank you...

sati commented on 2010-03-25 02:36

so it is..amd being sloppy..apologies, will delete these comments shortly to keep comments clean..as i imagine thats part of the reason for the AUR downtime & new pastebin links :)

Vi0L0 commented on 2010-03-25 01:56

@sati: no, its not available - try to hit download button and you will see deadlink

sati commented on 2010-03-25 01:26

10.3 is available - though I know it doesn't support xserver 1.7
http://support.amd.com/us/gpudownload/linux/Pages/radeon_linux.aspx?type=2.7&product=2.7.4.3.3.3.1&lang=English

Vi0L0 commented on 2010-03-24 23:54

catalyst 10.4 beta ownz http://aur.archlinux.org/packages.php?ID=34773

Vi0L0 commented on 2010-03-24 23:53

why out of date? theres no 10.3, besides 10.3 wont support xserver 1.7 so... i wouldnt wait for it anyway ;P

Vi0L0 commented on 2010-03-24 12:31

@kralyk: ok, done

kralyk commented on 2010-03-23 23:15

Hi,
would you please add 'libatical' to provides array?
I need that for package amdstream, which depends on these libs but
should be preferably catalyst-independent
Thank you!