The AMD/ATI Bar & Grill

savalaserg1 · 2013-07-01 14:08:51

New kernel was pushed 3.9.8.1 it seems to not load your hook ? I could do the workaround but till than staying on old kernel any news ?

snack · 2013-07-01 14:13:42

savalaserg1 wrote:

New kernel was pushed 3.9.8.1 it seems to not load your hook ? I could do the workaround but till than staying on old kernel any news ?

I'm on linux-ck 3.9.8-1 since some days and I had no problem with catalyst-hook from [catalyst-hd234k]...

savalaserg1 · 2013-07-01 14:15:22

I am using the repos ..... and using catalyst beta/stable

Last edited by savalaserg1 (2013-07-01 14:18:06)

snack · 2013-07-01 14:17:00

savalaserg1 wrote:

I am using the repos .....

I guess you mean Vi0l0's catalyst repos on apocalypsus... well, in this case I'm using the repos too, the repo for legacy radeon hardware to be precise.

savalaserg1 · 2013-07-01 14:24:53

Warning: Catalyst from Vi0L0's repository is compiled against one kernel and one kernel only. For Vi0L0, it's getting more and more difficult to keep up with kernel's pace and as such, we have seen a lot of users asking for what to do in the forums. Often, the solution is to a) wait for Vi0L0 to rebuilt the Catalyst package against those newer versions, b) do so yourself or c) hold back the Linux kernel update. The easiest solution is to do so yourself, not manually, but by using Vi0L0's excellent #Catalyst-hook.

This is basically why I am not updating my kernel.... Guess I should just install the catalyst-hook package and be done with it....

snack · 2013-07-01 14:28:02

savalaserg1 wrote:

This is basically why I am not updating my kernel.... Guess I should just install the catalyst-hook package and be done with it....

Yes, you're right. I'm using catalyst-hook since a long time and I never had problems with kernel upgrades nor had to wait to update my kernel. Plus, most of the times you don't have to download a new build of catalyst because the kernel module is automatically rebuilt on your machine.
Go with it

savalaserg1 · 2013-07-01 14:45:24

Well that was pretty straight forward lol.... All set and running 3.9.8.1 ....

snack · 2013-07-01 14:46:10

savalaserg1 wrote:

Well that was pretty straight forward lol.... All set and running 3.9.8.1 ....

Welcome to another fan of Vi0l0

savalaserg1 · 2013-07-01 15:02:33

Seems like alot of new updates coming for the catalyst driver it seems or for open source driver. Currently the driver is not that great....

Roken · 2013-07-01 19:21:16

Personally - I have never had an issue with Vi0l0's builds (though I have had the odd issue with catalyst - which aren't his fault). I honestly hope he lives forever, because he takes the pain out of catalyst for me.

lano1106 · 2013-07-04 04:23:24

Hi guys,

AMD staff did look at the bug 816:
http://ati.cchtml.com/show_bug.cgi?id=816

The bug title is 'catalyst 13.4 generates tons of useless file system I/O'

They are not able to reproduce the problem so to help them I have tried to reproduce it with catalyst 13.6 beta. On my system the problem is pretty much attenuated, the amd configs files recreation has been reduced from 25 recreation to exactly 2 recreation which is a big improvement.

I don't know if the improvement is from catalyst upgrade or from xorg or something else but I invite you to check again and add a comment in the bug page if you have different results than I do.

It is very easy to check and the details on how to check are described in the bug description.

thank you,

lano1106 · 2013-07-05 03:45:39

Vi0l0,

do you remember when sometime ago, I asked help on AUR about my problem with loading fglrx.ko.gz created with catalyst-generator?

My only workaround for this was to manually unzip the ko file after having built it.

I have figured out today what was missing.

Once the driver built, I had to run

# depmod -a

Ftroop · 2013-07-05 18:26:49

@Roken Ditto there !

Vi0L0 · 2013-07-05 20:04:31

@all:
thanks for the support and kind words

@lano1106:
tomorrow I will check is 13.4 still spamming and add the comment into bug report, for now I can only say that also firepro driver in version 12.104.2 got this bug.

As for the depmod I am surprised that catalyst-generator didn't do that for you because it has got depmod inside /usr/share/ati/build_mod/catalyst.install file which is used by catalyst_build_module's makepkg.
Can you comment rm -rf "${workdir}" line on /usr/bin/catalyst_build_module and after running catalyst_build_module check what's inside /tmp/catalyst.XXXXXX/catalyst.install (XXXXXX is random)?

foutrelis · 2013-07-06 07:14:57

I had to add the following to /usr/share/ati/build_mod/firegl_public.c near the top in order for catalyst-generator 13.1-6 (from [catalyst-hd234k]) to work with Linux 3.10.

#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,10,0)                                
#include <linux/seq_file.h>                                                     
#endif

Berseker · 2013-07-06 11:38:07

thank you foutrelis

andmars · 2013-07-07 02:51:39

foutrelis wrote:

I had to add the following to /usr/share/ati/build_mod/firegl_public.c near the top in order for catalyst-generator 13.1-6 (from [catalyst-hd234k]) to work with Linux 3.10.
#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,10,0)                                
#include <linux/seq_file.h>                                                     
#endif

Tried this with catalyst-hook from [catalyst] but that didn't work well.

Roken · 2013-07-07 08:13:18

andmars wrote:

foutrelis wrote:
I had to add the following to /usr/share/ati/build_mod/firegl_public.c near the top in order for catalyst-generator 13.1-6 (from [catalyst-hd234k]) to work with Linux 3.10.
#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,10,0)                                
#include <linux/seq_file.h>                                                     
#endif
Tried this with catalyst-hook from [catalyst] but that didn't work well.

Didn't work with catalyst-test either.

Vi0L0 · 2013-07-07 09:53:06

It's because foutrelis' patch is a patch for the patch
I didn't actually test hd234k driver, just throw into its catalyst-generator the 3.10 kernel patch which is working well with catalyst 13.4 - in most cases catalyst patches are working well that way, and since no one noticed it's not working I was thinking it's fine. But it wasn't - sorry fot that - next time I will be more careful.
Thanks foutrelis, I will surely use your patch .

Today I should be able to apply 3.10 patches into other than catalyst-generator packages, like mentioned -hook or -test. Just give me some time, there are some bugs here and there (like pxp's swrast thing) which I would like to work on at first.

I have found a workaround into "13.6 lag":
The lag it's caused by Tear-Free feature initialization when X is starting.
If I will turn Tear-Free off and restart X / relogin both KDE 4 and GNOME3 are much more responsive, almost as smooth as in catalyst 13.4.
Then, during the X session, I can turn Tear-Free on and the performance is still good - and it's working good until I will restart X...
So the Tear-Free must be disabled between the logons/restarts/shutdowns.

@lano1106:
as for the i/o bug, you probably noticed my answer on the bugzilla:
- catalyst 13.4 generates 6 rewrites of amdpcsdb, 2 opens of atiapfxx.blb / per gl app
- catalyst 13.6 beta generates 2 rewrites of amdpcsdb, 1 open of atiapfxx.blb / per gl app
So it's getting better. Better without our help... LOL

Last edited by Vi0L0 (2013-07-07 09:55:16)

andmars · 2013-07-08 18:46:08

Vi0L0 wrote:

Today I should be able to apply 3.10 patches into other than catalyst-generator packages, like mentioned -hook or -test. Just give me some time, there are some bugs here and there (like pxp's swrast thing) which I would like to work on at first.

I think I speak for all of us catalyst users if I say that you're doing a great job and just take your time. And even if it takes a few more days then so be it.

Vi0L0 · 2013-07-08 20:36:07

Hah, there was a time when I was as fast as Spanish Inquisition! I miss that time...

anyway, yesterday I've updated AUR with patches for 3.10. Today came time for the repos - [catalyst], [catalyst-stable], [catalyst-hd234k] and [catalyst-13.3-beta3] for now.

There's also a nice, shiny repos mirror - a gift from Emmanuel Leroy and hes rts-informatique - all major repos have their mirrors there, you can use it ie. in this way:

[catalyst-stable]
Server = http://mirror.rts-informatique.fr/archlinux/repo/catalyst/$arch

There are still some odd bugs in xauth and codexl+pxp which I'm working on...

rts-informatique · 2013-07-08 21:00:40

The more mirrors out there, the better it is

rsync://mirror.rts-informatique.fr::archlinux-catalyst

A good old "rsync -aP --del" does the trick. Make sure to tell Vi0L0 once your mirror is up and running !

savalaserg1 · 2013-07-09 02:58:00

Got a question im using the manual hook generator instead of Vi0L0's one now how would I go about ignoring the one every time I try to do a pacman -Syu it wants to install the new hook ....

mannemerak · 2013-07-09 14:43:22

Hi all,
I had pxp-catalyst working fine up to my last update (kernel and KDE mostly). Now broken, will try again later to fix it. Back on Radeon to for stable desktop.

My question; I have a hybrid Dell Vostro (Intel+AMD), do I HAVE to use pxp-catalyst or can I just use the native AMD catalyst. I am assuming (help me if I am wrong) that going on Intel with pxp-catalyst does not give me the ability to disable AMD GPU to save power. If I cant save power I am happy to just live in AMD GPU and only go to hybrid when X or Catalyst get round to implementing power saving on Linux.

Luye · 2013-07-10 16:03:43

Hi. I've been having an issue since i updated to catalyst (-daemon) 13.4 (although it might not be related). My 2nd monitor is not acknowledged by xrandr it seems, only one monitor shows upon query.

DFP1 disconnected (normal left inverted right x axis y axis)
...
...
DFP9 disconnected (normal left inverted right x axis y axis)
DFP10 disconnected (normal left inverted right x axis y axis)
DFP11 connected 1920x1080+0+0 (normal left inverted right x axis y axis) 510mm x 290mm
   1920x1080      60.0*+   50.0     59.9     60.1     50.0     60.0  
   1776x1000      50.0     59.9     50.0     60.0  
   1680x1050      60.0     60.1  
   ...
   ...
   ...

Some observations that might help narrow it down.
DFP10 and 11 should both be active, but I can only have one of them active at a time (I can alter between the two by moving orientation, but that is rather arbitary). In the i3 WM, the 'dead' monitor shows a black screen, while in xfce4 it shows the desktop with clickable shortcuts, but no possibility of boxing. For both window managers, attempting to drag a window to the 2nd monitor (moving left to right) works to some degree, but the window appears on the left of the 1st monitor, instead of the 2nd. Cursor behaves as normal.

xorg.conf
-------
Section "ServerLayout"
	Identifier     "aticonfig Layout"
	Screen      0  "aticonfig-Screen[0]-0" 0 0
	Screen         "aticonfig-Screen[0]-1" 1920 0
EndSection

Section "Module"
EndSection

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

Section "Monitor"
	Identifier   "aticonfig-Monitor[0]-1"
	Option	    "VendorName" "ATI Proprietary Driver"
	Option	    "ModelName" "Generic Autodetecting Monitor"
	Option	    "DPMS" "true"
EndSection

Section "Monitor"
	Identifier   "0-DFP10"
	Option	    "VendorName" "ATI Proprietary Driver"
	Option	    "ModelName" "Generic Autodetecting Monitor"
	Option	    "DPMS" "true"
	Option	    "PreferredMode" "1920x1080"
	Option	    "TargetRefresh" "60"
	Option	    "Position" "0 0"
	Option	    "Rotate" "normal"
	Option	    "Disable" "false"
EndSection

Section "Monitor"
	Identifier   "0-DFP11"
	Option	    "VendorName" "ATI Proprietary Driver"
	Option	    "ModelName" "Generic Autodetecting Monitor"
	Option	    "DPMS" "true"
	Option	    "TargetRefresh" "60"
	Option	    "Position" "0 0"
	Option	    "Rotate" "normal"
	Option	    "Disable" "false"
	Option	    "PreferredMode" "1920x1080"
EndSection

Section "Device"
	Identifier  "aticonfig-Device[0]-0"
	Driver      "fglrx"
	Option	    "Monitor-DFP11" "0-DFP11"
	BusID       "PCI:1:0:0"
EndSection

Section "Device"
	Identifier  "aticonfig-Device[0]-1"
	Driver      "fglrx"
	Option	    "Monitor-DFP10" "0-DFP10"
	BusID       "PCI:1:0:0"
	Screen      1
EndSection

Section "Screen"
	Identifier "aticonfig-Screen[0]-0"
	Device     "aticonfig-Device[0]-0"
	DefaultDepth     24
	SubSection "Display"
		Viewport   0 0
		Depth     24
	EndSubSection
EndSection

Section "Screen"
	Identifier "aticonfig-Screen[0]-1"
	Device     "aticonfig-Device[0]-1"
	DefaultDepth     24
	SubSection "Display"
		Viewport   0 0
		Depth     24
	EndSubSection
EndSection

Apologies if the post is massive, figured it would be cleaner this way as someone would probably ask for xorg.conf eventually.

Arch Linux

#5401 2013-07-01 14:08:51

Re: The AMD/ATI Bar & Grill

#5402 2013-07-01 14:13:42

Re: The AMD/ATI Bar & Grill

#5403 2013-07-01 14:15:22

Re: The AMD/ATI Bar & Grill

#5404 2013-07-01 14:17:00

Re: The AMD/ATI Bar & Grill

#5405 2013-07-01 14:24:53

Re: The AMD/ATI Bar & Grill

#5406 2013-07-01 14:28:02

Re: The AMD/ATI Bar & Grill

#5407 2013-07-01 14:45:24

Re: The AMD/ATI Bar & Grill

#5408 2013-07-01 14:46:10

Re: The AMD/ATI Bar & Grill

#5409 2013-07-01 15:02:33

Re: The AMD/ATI Bar & Grill

#5410 2013-07-01 19:21:16

Re: The AMD/ATI Bar & Grill

#5411 2013-07-04 04:23:24

Re: The AMD/ATI Bar & Grill

#5412 2013-07-05 03:45:39

Re: The AMD/ATI Bar & Grill

#5413 2013-07-05 18:26:49

Re: The AMD/ATI Bar & Grill

#5414 2013-07-05 20:04:31

Re: The AMD/ATI Bar & Grill

#5415 2013-07-06 07:14:57

Re: The AMD/ATI Bar & Grill

#5416 2013-07-06 11:38:07

Re: The AMD/ATI Bar & Grill

#5417 2013-07-07 02:51:39

Re: The AMD/ATI Bar & Grill

#5418 2013-07-07 08:13:18

Re: The AMD/ATI Bar & Grill

#5419 2013-07-07 09:53:06

Re: The AMD/ATI Bar & Grill

#5420 2013-07-08 18:46:08

Re: The AMD/ATI Bar & Grill

#5421 2013-07-08 20:36:07

Re: The AMD/ATI Bar & Grill

#5422 2013-07-08 21:00:40

Re: The AMD/ATI Bar & Grill

#5423 2013-07-09 02:58:00

Re: The AMD/ATI Bar & Grill

#5424 2013-07-09 14:43:22

Re: The AMD/ATI Bar & Grill

#5425 2013-07-10 16:03:43

Re: The AMD/ATI Bar & Grill

Board footer