The AMD/ATI Bar & Grill

devadittya · 2011-01-31 14:45:45

Everything updated smoothly today

Vi0L0 · 2011-01-31 18:01:11

repo:
along with catalyst and xorg-server-catalyst-maximize-fix i've updated xvba-video - i versioned it as 0.7.8 but it's actually in version 0.7.8.pre1 at the moment. I did this to avoid problems with updating because if i remember correctly 0.7.8.pre1 will be newer than 0.7.8 for pacman. Please correct me if i'm wrong here or if that problem was fixed

Last edited by Vi0L0 (2011-02-01 00:46:24)

foutrelis · 2011-01-31 18:49:15

The screen corruption problem I mention in the previous page actually goes away if I just enable vertical synchronization and leave "Tear Free Desktop" disabled.

Atondwal · 2011-02-01 01:11:35

foutrelis wrote:

The screen corruption problem I mention in the previous page actually goes away if I just enable vertical synchronization and leave "Tear Free Desktop" disabled.

If you want to leave on tear-free, switching to another tty and then back to X fixes the screen corruption for me whenever it occurs.

Vi0L0 · 2011-02-01 17:09:48

2.6.38 rc3:
i made a patch that should support this kernel well and put it into catalyst-generator 11.1-3 AUR package (it's my testing ground).
So for now from 38 line catalyst-generator supports only rc3 edit: and rc4 (no rc1 and rc2 support).

Last edited by Vi0L0 (2011-02-09 00:00:27)

Vi0L0 · 2011-02-03 14:29:42

repo clean up:
there's a lot of old archive catalyst repos on http://catalyst.apocalypsus.net/repo, all of them supports old kernels and old xservers and i want to remove them to make space for newer one. Preciselly i want to remove catalyst-archive-old, catalyst-archive and new-xvba directories which contains catalysts from 10.3 to 10.8.
If there's a need you can always build those packages with tarballs from http://catalyst.apocalypsus.net/tarball using makepkg or write me to reup oldie package.
If somebody want to save some package from annihilation then let me know, i will wait with clean up for next 3 days.

repo stats:
repo got now more than 2 000 users, this number has doubled over the last half year. In january we had 46.66Gb traffic. Nice
BIG THANKS to Apocalypsus for hosting! :* ;P

Edit: i made repo and wiki clean up today.
+ updated wiki with some info about Tear Free Desktop and useful troubleshooting discovered by pal.illes.

Last edited by Vi0L0 (2011-02-08 22:12:03)

Apollo Costa · 2011-02-10 16:15:59

Hi ViOLO,

with the first version 11.1 my HD6850 works fine. I have update to version 11.1-2 and xorg-server-catalyst-maximize-fix 1.9.4-1 now i got a "Unsupported hardware" watermarke like befor with 10.XX. Have you a backup of the first version from 11.1 (x64)?

Last edited by Apollo Costa (2011-02-10 16:24:41)

Vi0L0 · 2011-02-10 17:46:04

Silently i modified AUR's catalyst and catalyst-{daemon,hook,generator} packages. 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)
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.

@Apollo Costa:
i reup packages (built with new way mentioned above) and made new repo:

[catalyst-11.1-6xxx]
Server = http://catalyst.apocalypsus.net/repo/catalyst/x86_64

If there is i686 user who also want such repo - please let me know.

Ps. I could upload old ati-driver-installer-11-1-x86.x86_64.run file but it's too big for my connection (uploading is rly slow, around 256kbps), so i prefered to upload much smaller packages.

Off-topic: if you got compressed btrfs volume - please downgrade coreutils 8.10 to 8.9 otherwise your system may get damaged.

Last edited by Vi0L0 (2011-02-11 18:15:27)

Apollo Costa · 2011-02-10 18:08:03

thx ViOLO, with Catalyst 11.1-1 and Catalyst-utils 11.1-1 it works fine.

Apocalypsus · 2011-02-11 07:44:53

-

Last edited by Apocalypsus (2012-09-24 22:08:53)

xephinx · 2011-02-11 11:37:37

Is there no way to move the [catalyst] repo in the official [core] repo? /:

Vi0L0 · 2011-02-11 12:08:15

xephinx wrote:

Is there no way to move the [catalyst] repo in the official [core] repo? /:

LOL, ofcourse not, and the reason is still the same as it was 2 years ago:
ATI sucks in supporting freshly released xorg-servers (and also freshly released kernels but problems with kernel are in most cases easy to fix with patches).
It is nice coincidence that last xserver 1.9 was released just before ubuntu 10.10 which was also having it so ati made support for it.
And it was nice coincidence that xserver 1.9 was laing in [testing] for 1,5 month before release (strangely nvidia didn't ship support in time).
So we (catalyst users) could just be happy to have support right in time.
But remember that just 10 months ago we gain xserver 1.7 support for which we have been waiting for almost 6 months
And remember patched xserver 1.8 with a versioning trick?
Catalyst is still away from Arch Linux Way.

Last edited by Vi0L0 (2011-02-11 12:09:04)

xephinx · 2011-02-11 12:10:56

Sorry, didn't know it... (new to Arch)

Vi0L0 · 2011-02-11 12:43:15

xephinx wrote:

Sorry, didn't know it... (new to Arch)

Arch Linux Revolts Against ATI Catalyst Driver
you know it was on top of magazines some time ago
While i still think it was a good move, im a little bit dissapointed in speed of foss driver developement, just a little bit thou.
I just want to have open catalyst witk kms and other good open features but without catalyst's bugs and defects... and i want it now! xP
ekhem! just kiddin

Last edited by Vi0L0 (2011-02-11 18:14:56)

Vi0L0 · 2011-02-11 20:59:14

Well i'll be damned! Eternally!

What force obligated me to look at ATI_LICENSE.TXT today! What a bad, bad force!
I haven't done this for ages...

Look:

   vii) no rights are granted to distribute the binary form of the AMD Linux
   Kernel Module made by linking the AMD Proprietary  Kernel Library and the 
   AMD  Kernel  Compatibility  Layer  binary  compiled  using  Linux  kernel 
   headers;

What a nice "new" entry...
Darn!
Sorry guys, but i cannot leave catalyst package inside repo after reading this...
I will remove it immidiately, please switch to catalyst-{daemon,hook,generator} package.
In wiki's repo section i will propose to use hook because its got most votes.
I wrote to amd, but i doubt they give me permission...

Yet another reason not to put catalyst into official repo! crappy, crappy license!

Last edited by Vi0L0 (2011-02-11 21:02:45)

Mr.Elendig · 2011-02-11 21:20:27

As said on irc, contact amd and see what they have to say about it.

Apocalypsus · 2011-02-11 22:23:54

I don't really get them ... first this is not official repository so its like building a package for yourself and sharing it with some friends, isn't it? Can't see anything wrong here.

Vi0L0 · 2011-02-12 01:26:10

Apocalypsus wrote:

I don't really get them ... first this is not official repository so its like building a package for yourself and sharing it with some friends, isn't it? Can't see anything wrong here.

Well yes, in that way it sounds good, but still i'm not a lawyer, i don't know if we can bypass license with such nice and friendly talk.
I don't want any kind of problems with it...
I mean we don't want to be pirates here on the linux sea, do we?
The answer is obvious... everyone was dreaming about becoming pirate! Arrrrr!

Nah i want this situation to be as clear as possible. I dont have any contact to amd, so i wrote to bridgman - he's a nice amd guy, maybe he can help us.
It's a weekend so we might have to wait for the answer for a couple of days.

Apocalypsus · 2011-02-12 08:33:35

Hey guys,
I want to public thanks to David, who donated to support our domain renew!
Thanks David, thanks from the name of whole Arch and ATi users!

Domain is now renewed and will expire after 2 years, on 26.02.2013.

Thanks again!

xephinx · 2011-02-12 10:04:18

Thanks David!

agapito · 2011-02-12 10:44:50

Thanks David!

Vi0L0 · 2011-02-12 12:00:20

Thanks David!

d.duponchel · 2011-02-12 18:31:03

you're welcome :-)
But a few dollars are nothing compared to your hard work.

Thank you !

Last edited by d.duponchel (2011-02-12 20:58:54)

borislavl · 2011-02-14 07:41:19

Hi Vi0L0, this error is generated after today's update:

[root@dragon bobby]# pacman -Su
:: Starting full system upgrade...
:: Replace catalyst with catalyst/catalyst-daemon? [Y/n] y
:: Replace catalyst with catalyst/catalyst-generator? [Y/n] y
:: Replace catalyst with catalyst/catalyst-hook? [Y/n] y
resolving dependencies...
looking for inter-conflicts...
error: unresolvable package conflicts detected
error: failed to prepare transaction (conflicting dependencies)
:: catalyst-daemon and catalyst-hook are in conflict

How should I proceed in this situation?

Edit: Ooops, I just read the wiki, now I know how to proceed. Thanks for this wonderful documentation!

Last edited by borislavl (2011-02-14 08:01:18)

hopper · 2011-02-14 12:25:55

I tried the AUR version 2 of 11.1 on a kernel that I built, this is what happens

  gcc -Wp,-MD,/home/manoa/catalyst-11.1/.kcl_ioctl.o.d  -nostdinc -isystem /usr/lib/gcc/x86_64-unknown-linux-gnu/4.5.2/include -I/home/manoa/linux-2.6.37/arch/x86/include -Iinclude  -include include/generated/autoconf.h -D__KERNEL__ -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -O2 -m64 -march=core2 -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -DCONFIG_AS_CFI_SECTIONS=1 -DCONFIG_AS_FXSAVEQ=1 -pipe -Wno-sign-compare -fno-asynchronous-unwind-tables -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -Wframe-larger-than=2048 -fno-stack-protector -fomit-frame-pointer -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -DCC_HAVE_ASM_GOTO -I/home/manoa/catalyst-11.1 -DFGL -DFGL_LINUX -DFGL_GART_RESERVED_SLOT -DFGL_LINUX253P1_VMA_API -DPAGE_ATTR_FIX=0 -DCOMPAT_ALLOC_USER_SPACE=arch_compat_alloc_user_space  -DMODULE  -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(kcl_ioctl)"  -D"KBUILD_MODNAME=KBUILD_STR(fglrx)" -c -o /home/manoa/catalyst-11.1/kcl_ioctl.o /home/manoa/catalyst-11.1/kcl_ioctl.c
/home/manoa/catalyst-11.1/kcl_ioctl.c: In function ?KCL_IOCTL_RegisterConversion32?:
/home/manoa/catalyst-11.1/kcl_ioctl.c:174:5: error: implicit declaration of function ?register_ioctl32_conversion?
/home/manoa/catalyst-11.1/kcl_ioctl.c:174:46: error: ?ioctl_trans_handler_t? undeclared (first use in this function)
/home/manoa/catalyst-11.1/kcl_ioctl.c:174:46: note: each undeclared identifier is reported only once for each function it appears in
/home/manoa/catalyst-11.1/kcl_ioctl.c:174:68: error: expected ?)? before ?handler?
/home/manoa/catalyst-11.1/kcl_ioctl.c: In function ?KCL_IOCTL_UnregisterConversion32?:
/home/manoa/catalyst-11.1/kcl_ioctl.c:186:5: error: implicit declaration of function ?unregister_ioctl32_conversion?
/home/manoa/catalyst-11.1/kcl_ioctl.c: In function ?KCL_IOCTL_RegisterConversion32?:
/home/manoa/catalyst-11.1/kcl_ioctl.c:176:1: warning: control reaches end of non-void function
make[1]: *** [/home/manoa/catalyst-11.1/kcl_ioctl.o] Error 1
make: *** [_module_/home/manoa/catalyst-11.1] Error 2
make: Leaving directory `/home/manoa/linux-2.6.37'

it builds fine with the stock kernel (is also 2.6.37), I enable DRI but none of the drivers inside, what kernel options are required for this module ?

it appears to be in the decision here:

#ifdef KCL_OSCONFIG_IOCTL_COMPAT
    return 0;
#else
    return register_ioctl32_conversion(cmd, (ioctl_trans_handler_t)handler);
#endif

the stock kernel goes to return 0; but the compiled kernel goes to else, but what does that mean ? I couldn't find anything in the kernel .config file with IOCTL in it's name

I found what seems to be the source of problem, the kernel option 32-bit emulation when turned on allows the module to compile. but that is very strange because I am 100% sure that I copied the x64 bit versions

Last edited by hopper (2011-02-14 15:13:12)

Arch Linux

#2776 2011-01-31 14:45:45

Re: The AMD/ATI Bar & Grill

#2777 2011-01-31 18:01:11

Re: The AMD/ATI Bar & Grill

#2778 2011-01-31 18:49:15

Re: The AMD/ATI Bar & Grill

#2779 2011-02-01 01:11:35

Re: The AMD/ATI Bar & Grill

#2780 2011-02-01 17:09:48

Re: The AMD/ATI Bar & Grill

#2781 2011-02-03 14:29:42

Re: The AMD/ATI Bar & Grill

#2782 2011-02-10 16:15:59

Re: The AMD/ATI Bar & Grill

#2783 2011-02-10 17:46:04

Re: The AMD/ATI Bar & Grill

#2784 2011-02-10 18:08:03

Re: The AMD/ATI Bar & Grill

#2785 2011-02-11 07:44:53

Re: The AMD/ATI Bar & Grill

#2786 2011-02-11 11:37:37

Re: The AMD/ATI Bar & Grill

#2787 2011-02-11 12:08:15

Re: The AMD/ATI Bar & Grill

#2788 2011-02-11 12:10:56

Re: The AMD/ATI Bar & Grill

#2789 2011-02-11 12:43:15

Re: The AMD/ATI Bar & Grill

#2790 2011-02-11 20:59:14

Re: The AMD/ATI Bar & Grill

#2791 2011-02-11 21:20:27

Re: The AMD/ATI Bar & Grill

#2792 2011-02-11 22:23:54

Re: The AMD/ATI Bar & Grill

#2793 2011-02-12 01:26:10

Re: The AMD/ATI Bar & Grill

#2794 2011-02-12 08:33:35

Re: The AMD/ATI Bar & Grill

#2795 2011-02-12 10:04:18

Re: The AMD/ATI Bar & Grill

#2796 2011-02-12 10:44:50

Re: The AMD/ATI Bar & Grill

#2797 2011-02-12 12:00:20

Re: The AMD/ATI Bar & Grill

#2798 2011-02-12 18:31:03

Re: The AMD/ATI Bar & Grill

#2799 2011-02-14 07:41:19

Re: The AMD/ATI Bar & Grill

#2800 2011-02-14 12:25:55

Re: The AMD/ATI Bar & Grill

Board footer