The AMD/ATI Bar & Grill

scjet · 2013-12-28 10:34:46

although this question is no surprise, but, with respect to the "newer" amd-HD-5/6/7/... gpu's,
any news when Arch goes to xorg 1.15, that latest amd/ati catalyst drivers will be ready for it ?

Last edited by scjet (2013-12-28 10:59:42)

ivdok · 2013-12-28 18:02:04

scjet wrote:

although this question is no surprise, but, with respect to the "newer" amd-HD-5/6/7/... gpu's,
any news when Arch goes to xorg 1.15, that latest amd/ati catalyst drivers will be ready for it ?

Hell, we don't even have .14 support in [stable], so don't even start to think about .15. New software versions drop from [testing] rather frequently, most of it - under a month or 3 weeks. Typically from 3 days to a week.

JoshQuake · 2013-12-28 22:52:32

Hi, I just installed catalyst following the guide on arch wiki, but it seems X isn't starting now. When I try to do fglrxinfo I get "Error: Unable to open display (null)" and when I try to do aticonfig --adapter=all --odgt I get "ERROR - X needs to be running to perform AMD Overdrive(TM) commands"

rulex · 2013-12-29 12:05:35

Hi, the catalyst.wirephire.com domain seems to no longer be active. Is there a new domain?

tehgeek · 2013-12-29 15:34:06

rulex wrote:

Hi, the catalyst.wirephire.com domain seems to no longer be active. Is there a new domain?

Going to the site in a browser you get "This Domain Name Has Expired". =/

Steef435 · 2013-12-29 16:35:50

Tip: Because catalyst.wirephire.com will go down if a certain bandwidth limit is exceeded (this happened in the past) or may be too slow at your location, 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:

I find this in the wiki.

s0l1dsnak3123 · 2013-12-29 22:23:25

I can confirm that rts-informatique.fr mirrors correctly, but that wirephire repo being down has broken the xorg113 repository too

Last edited by s0l1dsnak3123 (2013-12-29 22:23:48)

Vi0L0 · 2013-12-29 22:30:01

s0l1dsnak3123 wrote:

I can confirm that rts-informatique.fr mirrors correctly, but that wirephire repo being down has broken the xorg113 repository too

Mirrors are also serving xorg repos. While waiting for Apocalypsus' answer I've created [xorg114] repo, you can find out all repos ie. here:
rtsinformatique, France: http://mirror.rts-informatique.fr/archl … lyst/repo/
yanom, USA: http://70.239.162.206/catalyst-mirror/repo/

Last edited by Vi0L0 (2013-12-30 08:41:15)

s0l1dsnak3123 · 2013-12-29 22:32:50

Beautiful, thanks Vi0l0

scjet · 2013-12-30 16:40:45

ivdok wrote:

scjet wrote:
although this question is no surprise, but, with respect to the "newer" amd-HD-5/6/7/... gpu's,
any news when Arch goes to xorg 1.15, that latest amd/ati catalyst drivers will be ready for it ?
Hell, we don't even have .14 support in [stable], so don't even start to think about .15. New software versions drop from [testing] rather frequently, most of it - under a month or 3 weeks. Typically from 3 days to a week.

, yes I agree regarding [stable] possibly?, ,.., sry, I should have specified I'm using (AUR)catalyst-test 13.12-1, with "X.Org X Server 1.14.5", and have no problems so far,
on my Laptop -> AMD/A10-APU/GPU/7600M.

-just wondered about Xorg 1.15 eventually comin down the pipe.
but OK thx, I guess it is what it is, when it is, with regards to AMD/ATI, still.

Last edited by scjet (2013-12-30 16:47:50)

brittyazel · 2013-12-30 19:11:20

I am getting an error message this morning saying that your have an invalid or corrupted PGP signature.

"error: database 'catalyst' is not valid (invalid or corrupted database (PGP signature))"

I have tried clearing my pgp keys and repulling yours down, but it is still broken. Any thoughts?

ilusi0n · 2013-12-30 19:29:14

brittyazel wrote:

I am getting an error message this morning saying that your have an invalid or corrupted PGP signature.
"error: database 'catalyst' is not valid (invalid or corrupted database (PGP signature))"
I have tried clearing my pgp keys and repulling yours down, but it is still broken. Any thoughts?

use the servers provide by Violo 2 posts above

clfarron4 · 2013-12-30 21:58:16

Any reference to Catalyst is the catalyst-hook package from Vi0L0's repository.

i'm trying to build the kernel modules for the 3.12.4 and 3.12.6 linux-lqx kernels from the AUR, and it will quite happily do with Catalyst 13.11-10 but not with Catalyst 13.12.

Here is the output from /var/log/catalyst-install.log when I try to build the modules for the 3.12.6 linux-lqx kernel with Catalyst 13.12 (for two separate runs). A reinstall yesterday means I can't create logs for 3.12.4 lqx with Catalyst 13.12 without compiling it again:

 --------
2013-12-30 21:34:20
Building fglrx module for 3.12.6-1-lqx kernel ...
--------
OsVersion says: SMP=1
file /proc/kallsyms says: SMP=1
file /usr/lib/modules/3.12.6-1-lqx/build/include/generated/autoconf.h says: SMP=1
file /usr/lib/modules/3.12.6-1-lqx/build/include/generated/autoconf.h says: MODVERSIONS=1
file /usr/lib/modules/3.12.6-1-lqx/build/arch/x86/include/asm/compat.h says: COMPAT_ALLOC_USER_SPACE=arch_compat_alloc_user_space
make: Entering directory '/usr/src/linux-3.12.6-1-lqx'
  CC [M]  /tmp/catalyst.fdSBSk/firegl_public.o
/tmp/catalyst.fdSBSk/firegl_public.c: In function ‘KCL_GetEffectiveUid’:
/tmp/catalyst.fdSBSk/firegl_public.c:1759:5: error: incompatible types when returning type ‘kuid_t’ but ‘KCL_TYPE_Uid’ was expected
     return current_euid();
     ^
/tmp/catalyst.fdSBSk/firegl_public.c:1763:1: warning: control reaches end of non-void function [-Wreturn-type]
 }
 ^
scripts/Makefile.build:308: recipe for target '/tmp/catalyst.fdSBSk/firegl_public.o' failed
make[1]: *** [/tmp/catalyst.fdSBSk/firegl_public.o] Error 1
Makefile:1228: recipe for target '_module_/tmp/catalyst.fdSBSk' failed
make: *** [_module_/tmp/catalyst.fdSBSk] Error 2
make: Leaving directory '/usr/src/linux-3.12.6-1-lqx'

--------
2013-12-30 21:35:34
Building fglrx module for 3.12.6-1-lqx kernel ...
--------
OsVersion says: SMP=1
file /proc/kallsyms says: SMP=1
file /usr/lib/modules/3.12.6-1-lqx/build/include/generated/autoconf.h says: SMP=1
file /usr/lib/modules/3.12.6-1-lqx/build/include/generated/autoconf.h says: MODVERSIONS=1
file /usr/lib/modules/3.12.6-1-lqx/build/arch/x86/include/asm/compat.h says: COMPAT_ALLOC_USER_SPACE=arch_compat_alloc_user_space
make: Entering directory '/usr/src/linux-3.12.6-1-lqx'
  CC [M]  /tmp/catalyst.zoJyx0/firegl_public.o
/tmp/catalyst.zoJyx0/firegl_public.c: In function ‘KCL_GetEffectiveUid’:
/tmp/catalyst.zoJyx0/firegl_public.c:1759:5: error: incompatible types when returning type ‘kuid_t’ but ‘KCL_TYPE_Uid’ was expected
     return current_euid();
     ^
/tmp/catalyst.zoJyx0/firegl_public.c:1763:1: warning: control reaches end of non-void function [-Wreturn-type]
 }
 ^
scripts/Makefile.build:308: recipe for target '/tmp/catalyst.zoJyx0/firegl_public.o' failed
make[1]: *** [/tmp/catalyst.zoJyx0/firegl_public.o] Error 1
Makefile:1228: recipe for target '_module_/tmp/catalyst.zoJyx0' failed
make: *** [_module_/tmp/catalyst.zoJyx0] Error 2
make: Leaving directory '/usr/src/linux-3.12.6-1-lqx'

Note that Catalyst 13.12 is working fine with my CK-K10 and stock ARCH kernels.

Last edited by clfarron4 (2013-12-30 22:32:30)

Vi0L0 · 2013-12-31 07:22:04

@clfarron4:
have you tried applying the patch provided by Cold @ https://aur.archlinux.org/packages/catalyst-test/ ?

clfarron4 · 2013-12-31 11:13:26

Vi0L0 wrote:

@clfarron4:
have you tried applying the patch provided by Cold @ https://aur.archlinux.org/packages/catalyst-test/ ?

I will try it and report back, but it reads like it fixes the problem for lqx 3.12.4.

EDIT: It works, and all the other kernel modules seem to work work too

Last edited by clfarron4 (2013-12-31 12:06:26)

l0vot · 2013-12-31 21:57:54

I am using the latest Arch, catalyst-total from the AUR, and the xorg packaged from xorg114.
I have a problem, when I try to start x using "startx", I get an error that is similar to this (not exactly cuz i have no screenshots to copy from):

Error: failed to load shared object file: libgcrypt.so.11, No such file or directory.

Due to this error, X refuses to start.

I used locate to find libgcrypt.so, and I found libgcrypt.so, and libgcrypt.so.20 inside of /usr/lib
I also found libgcrypt.so.11 inside lib32, but that doesn't count

I symlinked libgcrypt.so.20 to libgcrypt.so.11, but that did not work (I didn't expect it to, but it was worth a shot).

I am using my backup install until the problem is resolved.(it uses the FOSS drivers to avoide any problems encountered while using the proprietary ones).

Thank you for your time.

EDIT: Why does xorg need what appears to be a cryptography library in order to start anyways?

Last edited by l0vot (2014-01-02 20:57:39)

PhotonX · 2014-01-01 11:29:05

The server seems to be up again, but something is not working. Doing " pacman -Syu" I get

error: GPGME error: No data
error: failed to update catalyst (invalid or corrupted database (PGP signature))
error: database 'catalyst' is not valid (invalid or corrupted database (PGP signature))

So I disabled the signature check with "SigLevel = Never" in /etc/pacman.conf. After "pacman -Syy" there are no errors, but when "pacman -Su" I get many of

error: could not open file /var/lib/pacman/sync/catalyst.db: Unrecognized archive format

l0vot · 2014-01-01 18:04:59

@PhotonX: I have the same problem, but I use the AUR insead of the catalyst repo, so I left that out of my post.

Last edited by l0vot (2014-01-01 18:08:01)

acgtyrant · 2014-01-02 12:10:30

I find a strange phenomenon, it may be a big bug.

I have a ThinkPad and a DELL monitor, and I installed catalyst-utils-pxp and catalyst-hook while I use i3 as my windows manager.

When I use Intel driver, xrandr works very well with the dual screen.

However, when I use AMD driver and xrandr to implement dual screen, the strange phenomenon will happens.

If I am active at the laptop's window, or take a screenshot, it looks very normal, here is the screenshot: https://lh6.googleusercontent.com/-AWKs … enShot.png

But if I am active at the DELL screen, <b>the second screen will be copied at the laptop screen and cover it!</b>, here is the image: https://lh4.googleusercontent.com/-0lpP … 214911.jpg. However the screenshot is still normal, or I active or fresh the laptop screen again then the strange phenomenon will disappears.

I think there must be something copy the second screen to the laptop screen and cover it when the second screen is active. Do you have any idea? Thank you!

Forgive my poor English please.

l0vot · 2014-01-02 20:53:52

@acgtyrant: protip, HTML tags have no power here, you would use

[b][/b]

in order to get bold text.
Have you tried using the AMD control center* to set up dual screens?

*it should be in your main menue somewhere, if that fails, the ArchWiki page is here: https://wiki.archlinux.org/index.php/AM … figuring_X

clfarron4 · 2014-01-03 11:39:08

So, I'm in the middle of re-doing the dual-boot setup on one of my rigs, when I went to the AMD site to get the stable AMD Catalyst drivers for Windows when I was greeted by this. That is, it appears that 13.12 is the latest stable driver and 13.11.9-6 is the latest beta driver. Here is the corresponding Linux page.

Does anyone know what version is the latest stable version of Catalyst?

ChemBro · 2014-01-03 12:40:38

clfarron4 wrote:

Does anyone know what version is the latest stable version of Catalyst?

13.12., as you mentioned.

acgtyrant · 2014-01-03 13:47:17

@l0vot Thank you. But I think you misunderstand my problem, I said,I have set the dual screen successfully, but there are something always copy the second screen to the laptop screen when it is active.

snowblind · 2014-01-03 16:16:34

acgtyrant wrote:

@l0vot Thank you. But I think you misunderstand my problem, I said,I have set the dual screen successfully, but there are something always copy the second screen to the laptop screen when it is active.

This is a known issue with the Catalyst drivers on laptops with PowerXpress graphics -- when using the AMD card, if you use xrandr to expand the desktop to an external monitor, you get strange corruption. AMD has yet to fix this.

In order to deal with this, you have a few options:

1. You can scale your laptop monitor, as such:

xrandr --output HDMI1 --left-of LVDS1 --primary --scale 1x1 --output LVDS1 --scale 1.0001x1.0001

This should make the artifacting go away, with little noticeable difference to the screen output

2. You can use Bumblebee, specifically with YoRyan's AMD hacks: https://github.com/YoRyan/bumblebee-amd-hacks (note: some games don't really like this, and the performance is not as high as just using the AMD driver directly, mostly because Primus still doesn't work with AMD+Bumblebee)

3. You can ditch the Catalyst driver and use the Mesa driver -- from git -- which has PRIME offloading. Performance is not as good on Mesa (I'd peg it around 60-70% of Catalyst for radeonsi, 80% or higher for r600g) and Mesa doesn't have the OpenGL compliance that Catalyst does, but overall it presents a higher quality of life, especially on PowerXpress laptops.

Last edited by snowblind (2014-01-03 16:17:37)

zyth88 · 2014-01-03 21:23:01

Is it recommended to use CLUTTER_VBLANK=none in Gnome if Tear Free is enabled in the catalyst options?

Arch Linux

#6026 2013-12-28 10:34:46

Re: The AMD/ATI Bar & Grill

#6027 2013-12-28 18:02:04

Re: The AMD/ATI Bar & Grill

#6028 2013-12-28 22:52:32

Re: The AMD/ATI Bar & Grill

#6029 2013-12-29 12:05:35

Re: The AMD/ATI Bar & Grill

#6030 2013-12-29 15:34:06

Re: The AMD/ATI Bar & Grill

#6031 2013-12-29 16:35:50

Re: The AMD/ATI Bar & Grill

#6032 2013-12-29 22:23:25

Re: The AMD/ATI Bar & Grill

#6033 2013-12-29 22:30:01

Re: The AMD/ATI Bar & Grill

#6034 2013-12-29 22:32:50

Re: The AMD/ATI Bar & Grill

#6035 2013-12-30 16:40:45

Re: The AMD/ATI Bar & Grill

#6036 2013-12-30 19:11:20

Re: The AMD/ATI Bar & Grill

#6037 2013-12-30 19:29:14

Re: The AMD/ATI Bar & Grill

#6038 2013-12-30 21:58:16

Re: The AMD/ATI Bar & Grill

#6039 2013-12-31 07:22:04

Re: The AMD/ATI Bar & Grill

#6040 2013-12-31 11:13:26

Re: The AMD/ATI Bar & Grill

#6041 2013-12-31 21:57:54

Re: The AMD/ATI Bar & Grill

#6042 2014-01-01 11:29:05

Re: The AMD/ATI Bar & Grill

#6043 2014-01-01 18:04:59

Re: The AMD/ATI Bar & Grill

#6044 2014-01-02 12:10:30

Re: The AMD/ATI Bar & Grill

#6045 2014-01-02 20:53:52

Re: The AMD/ATI Bar & Grill

#6046 2014-01-03 11:39:08

Re: The AMD/ATI Bar & Grill

#6047 2014-01-03 12:40:38

Re: The AMD/ATI Bar & Grill

#6048 2014-01-03 13:47:17

Re: The AMD/ATI Bar & Grill

#6049 2014-01-03 16:16:34

Re: The AMD/ATI Bar & Grill

#6050 2014-01-03 21:23:01

Re: The AMD/ATI Bar & Grill

Board footer