You are not logged in.

#2076 2010-07-01 15:30:48

agapito
Member
From: Spain
Registered: 2008-11-13
Posts: 495

Re: The AMD/ATI Bar & Grill

Section "ServerLayout"
        Identifier     "aticonfig Layout"
        Screen      0  "aticonfig-Screen[0]-0" 0 0
EndSection

Section "Files"
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   "0-CRT1"
        Option      "VendorName" "ATI Proprietary Driver"
        Option      "ModelName" "Generic Autodetecting Monitor"
        Option      "DPMS" "true"
        Option      "PreferredMode" "1280x1024"
        Option      "TargetRefresh" "75"
        Option      "Position" "0 0"
        Option      "Rotate" "normal"
        Option      "Disable" "false"
EndSection

Section "Device"
        Identifier  "aticonfig-Device[0]-0"
        Driver      "fglrx"
        Option      "Monitor-CRT1" "0-CRT1"
        BusID       "PCI:1:0:0"
EndSection

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

Excuse my poor English.

Offline

#2077 2010-07-01 15:47:40

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

Thanks agapito. Seems I had it messed up... Let's if that works.

Offline

#2078 2010-07-01 15:48:07

snack
Member
From: Italy
Registered: 2009-01-13
Posts: 672

Re: The AMD/ATI Bar & Grill

Here's my single screen xorg.conf:

Section "ServerLayout"
        Identifier     "aticonfig Layout"
        Screen      0  "Screen0" 0 0
EndSection

Section "Device"
    Identifier  "HD3470"
    Driver      "fglrx"
    BusID       "PCI:1:0:0"
EndSection

Section "Monitor"
    Identifier   "PANEL"
    Option       "DPMS" "true"
EndSection

Section "Screen"
    Identifier "Screen0"
    Device     "HD3470"
    Monitor    "PANEL"
    DefaultDepth     24
    SubSection "Display"
        Viewport   0 0
        Depth     24
        Virtual 2560 1024
    EndSubSection
EndSection

and here's for dual screen:

Section "ServerLayout"
    Identifier     "X.org Configured"
    Screen      0  "Screen0" 0 0
EndSection

Section "Monitor"
        Identifier      "LAPTOP"
        Option          "PreferredMode" "1280x800"
    #Option          "DPMS" "true"
EndSection

Section "Monitor"
        Identifier      "VGA"
        Option          "PreferredMode" "1280x1024"
    Option          "RightOf"  "LAPTOP"
EndSection

Section "Device"
        Identifier      "HD3470"
        Driver          "fglrx"
    BusID           "PCI:1:0:0"
        # Monitor options:   radeon    radeonhd    catalyst
        #          laptop     LVDS      PANEL        LVDS
        #         external    VGA-0     VGA_1        CRT1
        Option         "monitor-CRT1" "VGA"  
        Option         "monitor-LVDS" "LAPTOP"
EndSection

Section "DRI"
    Mode 0666
    EndSection

Section "Screen"
        Identifier     "Screen0"
        Device         "HD3470"
        Monitor        "LAPTOP"
        DefaultDepth 24
        SubSection     "Display"
                Modes           "1280x800"
                Depth           24
                Virtual         2560 1024
        EndSubSection
EndSection

They work beautifully on my Mobility HD3470.

Offline

#2079 2010-07-02 15:19:28

Vi0L0
Member
From: Poland
Registered: 2009-06-24
Posts: 1,306
Website

Re: The AMD/ATI Bar & Grill

Splitting catalyst and lib32-catalyst-utils?

Im writing this because of comments and actions at AUR's catalyst and lib32-catalyst-utils pages.
dkxls made some good PKGBUILD for splitting those packages http://pastebin.com/HBawLG0g.
And ofcourse with splitting those package there are some advantages - like saving time of wine users. And ofcourse disadvantages like all wroten by cyberpatrol.

cyberpatrol wrote:

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.

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.

First thing - those packages wont be splitted.
BUT.

I just thought that maybe it's a good idea to make possible (but not by default) to build such splitted package in catalyst PKGBUILD. That can be achieved easilly, look:

if [ "${CARCH}" = "x86_64" ]; then
  pkgname=('catalyst') #comment this line if want to build catalyst and lib32-catalyst-utils
#  pkgname=('catalyst' 'lib32-catalyst-utils') #uncomment this line if want to build catalyst and lib32-catalyst-utils
fi

That way all users should be sattisfied. Those who want can build those two packages by simply changing two lines. I think that it can help many users.
Theres only one thing - lib32-catalyst-utils package dont have maintainer right now. lib32-catalyst-utils is important package, which need to exist in aur, theres no other option. And i wont take that one. It need to be maintained by some advanced wine user, which im not. Unfortunatelly i believe that if i will make possibility of building both catalyst and lib32-catalyst-utils in one catalyst's PKGBUILD - i will deter everyone of maintaing lib32-catalyst-utils...

Thats why im asking here - who want to maintain lib32-catalyst-utils? If so - will he/she share his/her code with me (?) so possibility (one more time - only possibility) of splitting catalyst and lib32-catalyst-utils in catalyst package could be done.
If he/she dont want to share his/her code with me - nothing happens - there will be no such split possibility.
Still maintainer of lib32-catalyst-utils is much more needed and much more important than split possibility.

Till this is unclear i will not make such split possible in aur's catalyst.

Offline

#2080 2010-07-02 19:18:47

cyberpatrol
Member
From: Germany
Registered: 2006-11-22
Posts: 68

Re: The AMD/ATI Bar & Grill

Vi0L0 wrote:
if [ "${CARCH}" = "x86_64" ]; then
  pkgname=('catalyst') #comment this line if want to build catalyst and lib32-catalyst-utils
#  pkgname=('catalyst' 'lib32-catalyst-utils') #uncomment this line if want to build catalyst and lib32-catalyst-utils
fi

This won't work because single and split packages are completely different. Look at /usr/share/pacman/PKGBUILD.proto and /usr/share/pacman/PKGBUILD-split.proto and at the PKGBUILDs of kernel26 and kernel26-fbcondecor.

It's the easiest and best maintaining two single packages as it is done now. Btw., for people who generally don't want to install lib32-catalyst-utils but want to try a 32 bit game once, such mixed single-split PKGBUILD would make it even harder, because they won't be able to just install lib32-catalyst-utils additionally and deinstall it, if they don't want it anymore, instead they will need to reinstall catalyst and modify this PKGBUILD every time. And think about people who aren't able to writing shell scripts.

So this idea is a bad idea, too, not to mention the fact that it's technically not possible or it would make editing the PKGBUILD really hard.

Where is the problem to maintain and/or install two separate packages if needed? I really don't see any. Btw., it's much less work and a lot easier to install two packages (run pacman, makepkg, yaourt or aurbuild twice with two easy commands), than to install one package and need to heavily edit it before. What's the sense in this?

As I said before, better don't do this.

Offline

#2081 2010-07-03 02:25:18

Vi0L0
Member
From: Poland
Registered: 2009-06-24
Posts: 1,306
Website

Re: The AMD/ATI Bar & Grill

cyberpatrol wrote:

This won't work because single and split packages are completely different. Look at /usr/share/pacman/PKGBUILD.proto and /usr/share/pacman/PKGBUILD-split.proto and at the PKGBUILDs of kernel26 and kernel26-fbcondecor.

It's the easiest and best maintaining two single packages as it is done now. Btw., for people who generally don't want to install lib32-catalyst-utils but want to try a 32 bit game once, such mixed single-split PKGBUILD would make it even harder, because they won't be able to just install lib32-catalyst-utils additionally and deinstall it, if they don't want it anymore, instead they will need to reinstall catalyst and modify this PKGBUILD every time. And think about people who aren't able to writing shell scripts.

This code i prepared will build one package by default - catalyst. Which work just like original one. No additional dependencies etc etc.
Now if you will comment 1st line and uncomment 2nd line - it will build 2 packages - catalyst and lib32-catalyst-utils. Those packages arent connected. You must install them separately ie with pacman -U. Now if you will install both of them you can now ie remove builded that way lib32-catalyst-utils and install original aur's lib32-catalyst-utils from aur. And it will still work as it should.

cyberpatrol wrote:

So this idea is a bad idea, too, not to mention the fact that it's technically not possible or it would make editing the PKGBUILD really hard.

Where is the problem to maintain and/or install two separate packages if needed? I really don't see any. Btw., it's much less work and a lot easier to install two packages (run pacman, makepkg, yaourt or aurbuild twice with two easy commands), than to install one package and need to heavily edit it before. What's the sense in this?

As I said before, better don't do this.

Theres no problem in maintaining two packages. And lib32-catalyst-utils need to be maintained separately. This is just some helpful addition - cuz i believe (for people which build most of packages with makepkg) its much easier to uncomment one line and comment another line (its not hard editing) and have two packages in one directory than building it separately. Its easier than downloading two PKGBUILD, copying ati-driver-installer and makepkg in two directories. Not to mention updating via yaourt where you need to download ati-driver-installer two times (and still with pkgbuild from dkxls and this default code i proposed yaourt will build only catalyst package from aur's catalyst and lib32-catalyst-utils from aur's lib32-catalyst-utils separately) !
(btw - if i had to choose between yaourt and bauerbill i will take bauerbill - its better, and with it you can easilly choose which packages to update and which one to not)

Offline

#2082 2010-07-03 03:41:46

cyberpatrol
Member
From: Germany
Registered: 2006-11-22
Posts: 68

Re: The AMD/ATI Bar & Grill

Vi0L0 wrote:

Now if you will comment 1st line and uncomment 2nd line - it will build 2 packages - catalyst and lib32-catalyst-utils. Those packages arent connected. You must install them separately ie with pacman -U. Now if you will install both of them you can now ie remove builded that way lib32-catalyst-utils and install original aur's lib32-catalyst-utils from aur. And it will still work as it should.

Where is the sense of this? Keep two packages and you can install both packages, too.

Vi0L0 wrote:

Theres no problem in maintaining two packages. And lib32-catalyst-utils need to be maintained separately.

Then two packages have to be maintained for the same feature lib32-catalyst-utils as its own package, which is currently orphaned, btw., and additionally in catalyst. This is everything else than KISS. I'd say keep it simple and don't put it into catalyst neither optionally. It's just too confusing and too much work.

Vi0L0 wrote:

This is just some helpful addition - cuz i believe (for people which build most of packages with makepkg) its much easier to uncomment one line and comment another line (its not hard editing) and have two packages in one directory than building it separately.

Have you looked at the PKGBUILDs and the protos I've mentioned before? It's not just commenting or uncommenting one line. In single packages you need one package() function. In split packages you need several package_*() functions for every resulting package, which have to be called package_name1(), package_name2(), etc. while name1, name2, etc. must be the same as the package names in the array $pkgname. So you would need to put a package() function for the default single package and several other package_*() functions for the split package into the PKGBUILD. So you need to put the same package() function for the resulting package catalyst into this PKGBUILD twice, first as the function package() for the default behaviour and second as package_catalyst() for the split package. And you would need to comment every function and the users who want the split package need to comment the package() function for the single package and uncomment every other package_*() function. And don't forget that you need the variable $pkgbase in split packages, which probably also needs to be commented by default and uncommented by split package users.

One other thing is that there's only one build() function in a split package. So also in the default mode you would force the people who don't want lib32-catalyst-utils to build it anyway.

So you see, this is a lot more than just one line. So you really need to maintain lib32-catalyst-utils twice, in its own package and in catalyst. And typing `aurbuild -s catalyst` and `aurbuild -s lib32-catalyst-utils` could easily be a lot less to typing than `nano PKGBUILD` and commenting and uncommenting all that stuff.

As I said before, single packages and split packages are pretty different.

Btw., with such a stuff the PKGBUILD is much harder to read and to verify. And it's too confusing for many people who are new to Arch and not familiar with PKGBUILDs and scripting.

Vi0L0 wrote:

Its easier than downloading two PKGBUILD, copying ati-driver-installer and makepkg in two directories.

Where is this easier? I really don't see any reason why it should be so hard to download and install two separate packages. And you don't need to copy the file manually, because makepkg downloads it automatically. Well if you don't copy it, you need to download it twice, but that's done automatically.

And waiting for two downloads or copying one downloaded file into another directory could easily be faster, too, than editing and commenting and uncommenting several long functions in the PKGBUILD.

Vi0L0 wrote:

Not to mention updating via yaourt where you need to download ati-driver-installer two times (and still with pkgbuild from dkxls and this default code i proposed yaourt will build only catalyst package from aur's catalyst and lib32-catalyst-utils from aur's lib32-catalyst-utils separately) !

Well, this is generally true, but still it's everything else than KISS. It's better keeping one PKGBUILD for one package, if those packages don't belong together.

Vi0L0 wrote:

(btw - if i had to choose between yaourt and bauerbill i will take bauerbill - its better, and with it you can easilly choose which packages to update and which one to not)

I don't know bauerbill. I'm currently using aurbuild. Nevertheless AUR doesn't support split packages.

If this PKGBUILD would be written for including into the official binary repos then I would say, such a split package could make sense, but not for AUR. It would probably make sense for AUR, if AUR would support split packages and the AUR scripts like yaourt, aurbuild, bauerbill, etc. would also give the choice which of these packages to build and to install. But then there would be the problem how the users can find the various packages and how they know which PKGBUILD contains which package. This probably would be a big task for the AUR developers.

Not to mention, that the PKGBUILD and the .tar.gz archive will get bigger due to the additional lines and the additional source file lib32-catalyst.sh, and people who don't need or want lib32-catalyst-utils will need to download more than needed, even if it's only a few KB.

I had another suggestion. Keep the packages catalyst and lib32-catalyst-utils as they are and keep maintaining them. You probably should adobt lib32-catalyst-utils. And build a new split package called e.g. catalyst-complete or the like which then installs both catalyst and lib32-catalyst-utils. This would be the much better way. Then people who want this split package can just install catalyst-complete, and people who want to be more flexible can still install catalyst and lib32-catalyst-utils separately.

But then you still have the problem that you won't be able to upload this package to AUR, because AUR still can't handle split packages and doesn't accept them. Btw., faking such a PKGBUILD and lead AUR to believe that a split package is a single package doesn't work, too. Several people have already tried this.

So currently the best is keeping both packages as they are.

Last edited by cyberpatrol (2010-07-03 13:58:25)

Offline

#2083 2010-07-03 06:11:35

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

cyberpatrol wrote:

If this PKGBUILD would be written for including into the official binary repos then I would say, such a split package could make sense, but not for AUR.

And this is where I agree. It's true that new users don't pay attention to the PKGBUILD , even ignoring its existence. That's the way I used to be, too. Now, i'm building all the packages from aur locally. That being said I agree with cyberpatrol.

Vi0L0 one question for you. Why the catalyst 10.6 in the repo is built with the dependency of kernel26>=2.6.34, whilst the aur's is kernel26>=2.6.30? I hadn't noticed it until yesterday. The way it is now, users who want to downgrade to .33 have to downgrade catalyst as well, then come xorg, then xf86-input-* . I've been having some freezings lately which led me to downgrade to .33 and the catalyst made me do all the changes described above. With the catalyst 10.6 I built from aur I only had to downgrade one package, the kernel. Will soon open a wiki discussion about the repos, I'll let u know.

Offline

#2084 2010-07-03 13:10:54

Vi0L0
Member
From: Poland
Registered: 2009-06-24
Posts: 1,306
Website

Re: The AMD/ATI Bar & Grill

@cyberpatrol: ok, you convinced me - so there will be no split smile

@maevius: cuz repo's catalyst comes with prebuilded fglrx module which works only with current stack kernel from [core] repo (and it wont work with any other kernel than this specific kernel). So currently its builded for 2.6.34-ARCH kernel. Now with this kernel26>=2.6.34 and kernel26<2.6.35 dependency pacman wont let you to downgrade or upgrade kernel26 package simply cuz of dependency conflict and because repo's catalyst wont work with other kernel. And yes - if you want to use kernel .33 or any kernel other than 2.6.34-ARCH - you need to use aur's catalyst to build fglrx module specifically for kernel you are using.

Offline

#2085 2010-07-03 13:15:43

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

Crap, now I understood that what I wrote was nonsense.... In aur the package is being built the time you download it locally, in repo is already built... Silly me.
Anyway, regarding the glitches in browsers, I got none with 10.6 so I suppose it's sth related to the kernel. I'm using .33.

Last edited by maevius (2010-07-03 13:20:14)

Offline

#2086 2010-07-03 20:24:51

druk
Member
Registered: 2009-10-17
Posts: 24

Re: The AMD/ATI Bar & Grill

Updated my system today and removed the open source radeon driver and installed catalyst-10.6 and xorg 1.8. Overall I'm pretty happy with the change, had some big improvements, like the notebook now runs much cooler (a decrease of 5 degrees) however the 2d performance is far from great. Moving windows, especially firefox is not smooth, does anyone has/had the same problem?

I tried

aticonfig --set-pcs-str=DDX,ForceXAA,TRUE

but no improvements.

Offline

#2087 2010-07-03 20:38:31

dcc24
Member
Registered: 2009-10-31
Posts: 723

Re: The AMD/ATI Bar & Grill

Laggy window movement needs a patched xorg-server. Either use Vi0L0's backclear-patched version from his repo, or use my backfill-patched version (xorg-server-1.8-catalyst-maximize-fix) from AUR.


It is better to keep your mouth shut and be thought a fool than to open it and remove all doubt. (Mark Twain)

My AUR packages

Offline

#2088 2010-07-04 08:24:17

o'neill
Member
Registered: 2010-07-01
Posts: 11

Re: The AMD/ATI Bar & Grill

hey,

It's me again. I've found a problem and don't know if it is a ati-problem or something else.

When I watch videos on youtube or mplayer or something like that, there are very laggy when I switch it to fullscreen-mode.
Under Ubuntu there were very smooth and now there is a „lag party".

Here is my xorg.conf:

Section "ServerLayout"
    Identifier     "aticonfig Layout"
    Screen      0  "aticonfig-Screen[0]-0" 0 0
EndSection

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

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

I download all the packages from the catalyst-repo, but nothing changes.
Please help, I don't wanna change to Ubuntu hmm

Offline

#2089 2010-07-04 08:58:09

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

Youtube videos have to do with flash player. As for mplayer, what videos are you trying to watch? I'm with 10.6 and haven't noticed any problems, either with SD or HD videos. Try to install smplayer and in the video section change the video output driver to gl.

Offline

#2090 2010-07-04 09:23:21

o'neill
Member
Registered: 2010-07-01
Posts: 11

Re: The AMD/ATI Bar & Grill

maevius wrote:

Youtube videos have to do with flash player. As for mplayer, what videos are you trying to watch? I'm with 10.6 and haven't noticed any problems, either with SD or HD videos. Try to install smplayer and in the video section change the video output driver to gl.

Oh, with mplayer is everything alright, sry my mistake.
But with all flash-videos it is horrible slow and laggy when i switched it to fullscreen. (Also low quality videos)

Should I change something in my xorg.conf or is it ok?

Offline

#2091 2010-07-04 09:46:43

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

xorg has nothing to do with this issue. Unfortunately it's flash related. If you have 64bit system consider installing this http://aur.archlinux.org/packages.php?ID=6232
Make a search in the forum wink

Offline

#2092 2010-07-04 09:53:05

o'neill
Member
Registered: 2010-07-01
Posts: 11

Re: The AMD/ATI Bar & Grill

maevius wrote:

xorg has nothing to do with this issue. Unfortunately it's flash related. If you have 64bit system consider installing this http://aur.archlinux.org/packages.php?ID=6232
Make a search in the forum wink

Ohh, but nspluginwrapper is just for mozilla right or is it universal? (I use chromium)

(Sorry for spaming this wonderful thread sad )

Offline

#2093 2010-07-04 09:55:24

maevius
Member
From: Greece
Registered: 2009-09-10
Posts: 134
Website

Re: The AMD/ATI Bar & Grill

I use chromium too! If you haven't firefox installed, make sure you create the folder ~/.mozilla/plugins wink

Offline

#2094 2010-07-04 10:20:03

o'neill
Member
Registered: 2010-07-01
Posts: 11

Re: The AMD/ATI Bar & Grill

maevius wrote:

I use chromium too! If you haven't firefox installed, make sure you create the folder ~/.mozilla/plugins wink

Hahahaha, it works! and it's soooo smooth, like a baby ass.
Thanks!!!! big_smile

Offline

#2095 2010-07-05 06:23:55

druk
Member
Registered: 2009-10-17
Posts: 24

Re: The AMD/ATI Bar & Grill

dcc24 wrote:

Laggy window movement needs a patched xorg-server. Either use Vi0L0's backclear-patched version from his repo, or use my backfill-patched version (xorg-server-1.8-catalyst-maximize-fix) from AUR.

It didn't fix. I tried both patches.

This is what I did.
- Removed xorg-server
- Installed xorg-server-backclear.
- Terminated xorg and ran aticonfig --set-pcs-str=DDX,ForceXAA,TRUE
- Started xorg.

(Did the same thing for maximize-fix patch)

Today I also noticed some firefox glitches while playing a flash game. (Normal usage works fine, even with flash sites)

Offline

#2096 2010-07-07 04:03:59

thetoxikavenger
Member
Registered: 2010-06-05
Posts: 4

Re: The AMD/ATI Bar & Grill

Druk, I think I have the same issue as you do. Can you get any 3D programs to run? All my windows are laggy, and I also have tried both patched versions of xorg-server. When I try to run 3D I get:

X Error of failed request:   BadRequest (invalid request code or no such operation)
Major opcode of failed request:   136 (GLX)
Minor opcode of failed request:   19 (X_GLXQueryServerString)
Serial number of failed request:   15
Current serial number in output stream:   15

EDIT: Well it appears it is an issue with kernel 2.6.35-rc4. When I booted up under 2.6.34, everything was fine.

Last edited by thetoxikavenger (2010-07-07 04:14:34)

Offline

#2097 2010-07-07 04:24:42

dcc24
Member
Registered: 2009-10-31
Posts: 723

Re: The AMD/ATI Bar & Grill

thetoxikavenger wrote:

Druk, I think I have the same issue as you do. Can you get any 3D programs to run? All my windows are laggy, and I also have tried both patched versions of xorg-server. When I try to run 3D I get:

X Error of failed request:   BadRequest (invalid request code or no such operation)
Major opcode of failed request:   136 (GLX)
Minor opcode of failed request:   19 (X_GLXQueryServerString)
Serial number of failed request:   15
Current serial number in output stream:   15

EDIT: Well it appears it is an issue with kernel 2.6.35-rc4. When I booted up under 2.6.34, everything was fine.

Kernel 2.6.35-rc3 works fine for me. I'm using the patched xorg-server from AUR.


It is better to keep your mouth shut and be thought a fool than to open it and remove all doubt. (Mark Twain)

My AUR packages

Offline

#2098 2010-07-07 15:05:03

Vi0L0
Member
From: Poland
Registered: 2009-06-24
Posts: 1,306
Website

Re: The AMD/ATI Bar & Grill

and 2.6.35-rc4 works fine for me

Offline

#2099 2010-07-07 17:26:02

thetoxikavenger
Member
Registered: 2010-06-05
Posts: 4

Re: The AMD/ATI Bar & Grill

dcc24 wrote:

Kernel 2.6.35-rc3 works fine for me. I'm using the patched xorg-server from AUR.

Vi0L0 wrote:

and 2.6.35-rc4 works fine for me

I'm using the patched xorg-server as well. I wonder what I did wrong! tongue

Offline

#2100 2010-07-07 19:34:54

druk
Member
Registered: 2009-10-17
Posts: 24

Re: The AMD/ATI Bar & Grill

thetoxikavenger wrote:

Druk, I think I have the same issue as you do. Can you get any 3D programs to run? All my windows are laggy, and I also have tried both patched versions of xorg-server. When I try to run 3D I get:

X Error of failed request:   BadRequest (invalid request code or no such operation)
Major opcode of failed request:   136 (GLX)
Minor opcode of failed request:   19 (X_GLXQueryServerString)
Serial number of failed request:   15
Current serial number in output stream:   15

EDIT: Well it appears it is an issue with kernel 2.6.35-rc4. When I booted up under 2.6.34, everything was fine.

I'm using kernel 2.6.34.  In my case when moving large windows the performance degrades but not to point of being unusable. I haven't noticed any other problems with 3d applications, everything works fine.

The flash issue that I mentioned before is not related to this.

Last edited by druk (2010-07-07 21:06:09)

Offline

Board footer

Powered by FluxBB