Discussion: the ati/radeon open source drivers & radeon repository

korpenkraxar · 2010-01-20 22:03:22

yimm wrote:

You have to enable it using dynpm parameter (dynpm=1). You can set this on loading module time (modprobe radeon dynpm=1), or in GRUB (radeon.dynpm=1) or in /etc/modprobe.d/some.file

I'll try that out.

Although I have to say that I have stabilitiy issues with this kernel. Every now and then my CPU cores gets stuck at 800MHz and cpufreq and its various tools reports errors. In KDE the whole screen goes white at times if I switch windows while having compositing running, forcing me to power off the machine. Finally, suspend/resume is broken, as reported by others above... so not the best experience so far :-)

I am actually having a better time running rc4 + zen kernel from AUR. Maybe we could apply the patches to that one?

yimm · 2010-01-21 13:53:54

korpenkraxar wrote:

Although I have to say that I have stabilitiy issues with this kernel. Every now and then my CPU cores gets stuck at 800MHz and cpufreq and its various tools reports errors. In KDE the whole screen goes white at times if I switch windows while having compositing running, forcing me to power off the machine. Finally, suspend/resume is broken, as reported by others above... so not the best experience so far :-)

Which kernel? Stock kernel or 2.6.33-rc4?

korpenkraxar wrote:

I am actually having a better time running rc4 + zen kernel from AUR. Maybe we could apply the patches to that one?

Theses patches is only for 2.6.33-rc(?) no ? I don't know really.

Last edited by yimm (2010-01-21 13:54:12)

Black Mage · 2010-01-21 17:01:07

A nice update.

all graphic options on nexuiz work with the 21/01/10 drivers, it used to show crappy graphics with glsl 2.0, and work extremely slow with bloor effects (all other effects worked fine).

also alien arena compiled (it failed before complaining that there is no glsl1.2), unfortunatly even on lowest settings it drops fps quite often :\

Last edited by Black Mage (2010-01-21 18:27:05)

hunterthomson · 2010-01-22 08:23:04

handy wrote:

Is anyone playing Regnum with the open-source GPU driver solution?
I just installed Regnum but get the following error, which I put down to my using the stable open-source packages for my HD2600Pro:
Unsupported video card!
There are three possible causes for this error:
1. Your video card is too old
2. You haven't installed the latest available drivers
3. You haven't installed the latest DirectX version

Hum, ya I'm loving the Open driver. However, I have a lvl 42 Warlock or was it 32.... Well a couple levels past the point that the Quests stop. That game is fairly graphically intense. I think it may be some time until the Open driver can run it.

I bet you could play Eternal Lands though. I have like a lvl30 in that game. I like Regnum better though. Eternal Lands is not vary group orientated. The economy is the coolest thing is that game.

Last edited by hunterthomson (2010-01-22 11:17:33)

Neuro · 2010-01-22 18:40:00

Black Mage wrote:

A nice update.
all graphic options on nexuiz work with the 21/01/10 drivers, it used to show crappy graphics with glsl 2.0, and work extremely slow with bloor effects (all other effects worked fine).
also alien arena compiled (it failed before complaining that there is no glsl1.2), unfortunatly even on lowest settings it drops fps quite often :\

Are you talking about the mesa or the kernel modules?

rs · 2010-01-22 21:04:26

Neuro wrote:

Perry3D wrote:
Neuro wrote:
Ok, I'm using 2.6.33rc4 with Ati PM patches for KMS and mesa full from 20100117. KMS is on and I'm using KDE 4.4. Sometimes the dropdown shadow around the windwos in Kwin turns transparent-yellow. Anyone experienced anything similar?
Maybe it is fixed now: http://cgit.freedesktop.org/mesa/mesa/c … f5db7391df
Nah, it's not that. My chipset is r600 so I didn't think it would work.
Also I have a problem with my kernel. Suspend/resume doesn't work on Ati. Either from X or from console. It just hangs in the framebuffer while suspending. As I have a separate Intel card, I can check, and everything works there. I thought it was the fault of the PM patches, but without them I only have the fan up constantly, but suspend/resume doesn't work.
Has anyone had a similar bug? Where can I report it?

I also see the annoying bug with yellow KDE4.4 kwin shadows on a Radeon HD 4650, OpenGL compositing desktop, xf86-video-ati master branch, Mesa master branch, Xorg server 1.7.4, libdrm 2.4.17, KMS enabled (modprobe radeon modeset=1 dynclks=1).
Finally, a similar report.

For me, it happens since 2.6.33-rc4 (and including rc5). 2.6.33-rc3 was the last functional release.
So something must have happened in between rc3 and rc4.

Last edited by rs (2010-01-22 21:08:21)

dan.boff · 2010-01-22 21:39:14

i'm having some issues with stock mesa/kernel/...

I added radeon in MODULUS in mkinitcpio.conf and enable kms during boot with the boot parameter radeon.modeset=1

however if I do so my screen gets blank as soon as X starts and stays that way: So I can't use X with kms enabled. this also happens when I only enable kms with the boot parameter, but not change mkinitcpio.conf

booting to runlevel 3 works, only starting X causes the problem. It's not like the time when kms produced a blank screen as soon as radeon is loaded...

any ideas?

apollokk · 2010-01-22 22:01:32

Has anyone tried powersaving (with kernel26-git of course) ?

yimm · 2010-01-22 22:39:47

Yes, we are 2-3 people who have tested powersaving (but with kernel26-rc )
For me by example :
http://bbs.archlinux.org/viewtopic.php? … 68#p691668

Last edited by yimm (2010-01-23 08:31:59)

Perry3D · 2010-01-23 01:27:32

dan.boff wrote:

i'm having some issues with stock mesa/kernel/...
I added radeon in MODULUS in mkinitcpio.conf and enable kms during boot with the boot parameter radeon.modeset=1
however if I do so my screen gets blank as soon as X starts and stays that way: So I can't use X with kms enabled. this also happens when I only enable kms with the boot parameter, but not change mkinitcpio.conf
booting to runlevel 3 works, only starting X causes the problem. It's not like the time when kms produced a blank screen as soon as radeon is loaded...
any ideas?

Are you using the git versions of libdrm and xf86-video-ati? KMS doesn't work with the driver in the repo.

mdias · 2010-01-23 02:01:02

dan.boff wrote:

i'm having some issues with stock mesa/kernel/...
I added radeon in MODULUS in mkinitcpio.conf and enable kms during boot with the boot parameter radeon.modeset=1
however if I do so my screen gets blank as soon as X starts and stays that way: So I can't use X with kms enabled. this also happens when I only enable kms with the boot parameter, but not change mkinitcpio.conf
booting to runlevel 3 works, only starting X causes the problem. It's not like the time when kms produced a blank screen as soon as radeon is loaded...
any ideas?

Exacly the same problem here.
I have the rc5 kernel, latest xf86-video-ati and libdrm from git.

I do have a high-res console that displays only at my 2nd monitor (god knows why). Everything seems to work just fine, only my 1st monitor isn't displaying anything at all, however it does seem to detect some signal since the blinking light stops blinking, but I can't even access the monitor's menu, which leads me to believe that the card isn't just sending a blank frame to the screen, it's also somehow unable to modeset it?

Switching the heads of the monitors doesn't change a thing. My primary monitor still displays a blank screen.

Xorg log of such a failed session here: http://pastebin.com/m25e339f9

Info:
primary monitor: 1920x1200 max
secondary monitor: 1280x1024 max

mdias · 2010-01-23 13:38:29

Ok, after some more testing, I figured something else;
Connecting my main monitor using a VGA cable works. Previously its been connected using a DVI-HDMI cable.

Neuro · 2010-01-23 15:37:10

rs wrote:

I also see the annoying bug with yellow KDE4.4 kwin shadows on a Radeon HD 4650, OpenGL compositing desktop, xf86-video-ati master branch, Mesa master branch, Xorg server 1.7.4, libdrm 2.4.17, KMS enabled (modprobe radeon modeset=1 dynclks=1).
Finally, a similar report.
For me, it happens since 2.6.33-rc4 (and including rc5). 2.6.33-rc3 was the last functional release.
So something must have happened in between rc3 and rc4.

Yea, well this ca't be a bug in mesa. I've just switched back to kernel 2.6.32.3 (from 2.6.33rc4 and rc5), because suspend/resume didn't work and under Intel graphics, my iwlagn WiFI card stopped resuming. In 2.6.32.3 and KMS, I get these awful yellow strips. After switching away from ATI KMS back to UMS, everything works fine, no artifacts.
Although I miss dm patches and the opengl vsync, I see no reason to have 2.6.33. It only breaks a lot of things

apollokk · 2010-01-23 16:27:01

mesa-full finishes with this error:

nv04_context.c: In function 'nv04_init_hwctx':
nv04_context.c:44: error: 'NV04_TEXTURED_TRIANGLE_NOP' undeclared (first use in this function)
nv04_context.c:44: error: (Each undeclared identifier is reported only once
nv04_context.c:44: error: for each function it appears in.)
nv04_context.c:47: error: 'NV04_TEXTURED_TRIANGLE_CONTROL' undeclared (first use in this function)
nv04_context.c:50: error: 'NV04_TEXTURED_TRIANGLE_BLEND' undeclared (first use in this function)
nv04_context.c:53: error: 'NV04_TEXTURED_TRIANGLE_FORMAT' undeclared (first use in this function)
nv04_context.c:55: error: 'NV04_TEXTURED_TRIANGLE_FILTER' undeclared (first use in this function)
nv04_context.c:57: error: 'NV04_TEXTURED_TRIANGLE_COLORKEY' undeclared (first use in this function)
nv04_context.c:61: error: 'NV04_TEXTURED_TRIANGLE_FOGCOLOR' undeclared (first use in this function)
nv04_context.c: In function 'nv04_create':
nv04_context.c:91: warning: assignment from incompatible pointer type
make[4]: *** [nv04_context.o] Error 1
make[4]: Leaving directory `/home/gtklocker/works/mesa-full/src/mesa-build/src/gallium/drivers/nv04'
make[3]: *** [default] Error 1
make[3]: Leaving directory `/home/gtklocker/works/mesa-full/src/mesa-build/src/gallium/drivers'
make[2]: *** [default] Error 1
make[2]: Leaving directory `/home/gtklocker/works/mesa-full/src/mesa-build/src/gallium'
make[1]: *** [subdirs] Error 1
make[1]: Leaving directory `/home/gtklocker/works/mesa-full/src/mesa-build/src'
make: *** [default] Error 1
==> ERROR: Build Failed.
    Aborting...

What's up?

ChemBro · 2010-01-23 16:33:13

apollokk wrote:

mesa-full finishes with this error:
something something
What's up?

Try again? It helps sometimes. (That's how I handle it)

Or maybe it is not buildable at the moment.

apollokk · 2010-01-23 16:35:54

Has anyone build this?

Can someone upload it?

Edit: Tried to build it again, same problem.

Last edited by apollokk (2010-01-23 16:47:45)

lman · 2010-01-23 17:08:47

try removing the --enable-gallium-nouveau option in the PKGBUILD. I think it's unnecesarry to build it with this option on since you have an ati card.

Last edited by lman (2010-01-23 17:10:33)

apollokk · 2010-01-23 17:20:09

Tried mesa-split, exits with this error:

mklib: Making Linux static library:  libxorgtracker.a
ar: creating libxorgtracker.a
make[4]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/state_trackers/xorg'
make[3]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/state_trackers'
make[2]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium'
make[2]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys'
make[3]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm'
make[4]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon'
make[5]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
rm -f depend
touch depend
fdepend -I/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.2/include -I/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.2/include-fixed -I. -I../../../../../../src/gallium/include -I../../../../../../src/gallium/auxiliary -I../../../../../../src/gallium/drivers -I/src/gallium/include -I/src/gallium/auxiliary -I/src/gallium/drivers -I../../../../../../src/gallium/drivers/r300 -I/usr/include/drm   radeon_buffer.c radeon_drm.c radeon_r300.c   2> /dev/null
make[5]: [depend] Error 127 (ignored)
make[5]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
make[5]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
gcc -c -I. -I../../../../../../src/gallium/include -I../../../../../../src/gallium/auxiliary -I../../../../../../src/gallium/drivers -I/src/gallium/include -I/src/gallium/auxiliary -I/src/gallium/drivers -I../../../../../../src/gallium/drivers/r300 -I/usr/include/drm   -D_GNU_SOURCE -DPTHREADS -DHAVE_POSIX_MEMALIGN -DUSE_XCB -DGLX_USE_TLS -DPTHREADS -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_DIRECT_RENDERING -DGLX_INDIRECT_RENDERING -DHAVE_ALIAS -march=x86-64 -mtune=generic -O2 -pipe -Wall -Wmissing-prototypes -std=c99 -ffast-math -fvisibility=hidden -fno-strict-aliasing  -fPIC  -DUSE_X86_64_ASM -D_GNU_SOURCE -DPTHREADS -DHAVE_POSIX_MEMALIGN -DUSE_XCB -DGLX_USE_TLS -DPTHREADS -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_DIRECT_RENDERING -DGLX_INDIRECT_RENDERING -DHAVE_ALIAS  radeon_buffer.c -o radeon_buffer.o
In file included from radeon_buffer.c:33:
radeon_buffer.h:43:23: error: radeon_bo.h: No such file or directory
radeon_buffer.h:44:23: error: radeon_cs.h: No such file or directory
In file included from radeon_drm.h:46,
                 from radeon_buffer.h:46,
                 from radeon_buffer.c:33:
radeon_r300.h:31:27: error: radeon_cs_gem.h: No such file or directory
radeon_buffer.c:35:27: error: radeon_bo_gem.h: No such file or directory
radeon_buffer.c: In function 'radeon_buffer_create':
radeon_buffer.c:85: error: 'RADEON_GEM_DOMAIN_VRAM' undeclared (first use in this function)
radeon_buffer.c:85: error: (Each undeclared identifier is reported only once
radeon_buffer.c:85: error: for each function it appears in.)
radeon_buffer.c:88: error: 'RADEON_GEM_DOMAIN_GTT' undeclared (first use in this function)
radeon_buffer.c:94: warning: implicit declaration of function 'radeon_bo_open'
radeon_buffer.c:95: warning: assignment makes pointer from integer without a cast
radeon_buffer.c: In function 'radeon_buffer_user_create':
radeon_buffer.c:114: warning: implicit declaration of function 'radeon_bo_map'
radeon_buffer.c:115: error: dereferencing pointer to incomplete type
radeon_buffer.c:116: warning: implicit declaration of function 'radeon_bo_unmap'
radeon_buffer.c: In function 'radeon_buffer_del':
radeon_buffer.c:151: warning: implicit declaration of function 'radeon_bo_unref'
radeon_buffer.c: In function 'radeon_buffer_map':
radeon_buffer.c:173: warning: implicit declaration of function 'radeon_bo_is_busy'
radeon_buffer.c:177: warning: implicit declaration of function 'radeon_bo_is_referenced_by_cs'
radeon_buffer.c:189: error: dereferencing pointer to incomplete type
radeon_buffer.c: In function 'radeon_pipe_winsys':
radeon_buffer.c:290: warning: implicit declaration of function 'radeon_bo_manager_gem_ctor'
radeon_buffer.c:290: warning: assignment makes pointer from integer without a cast
make[5]: *** [radeon_buffer.o] Error 1
make[5]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
make[4]: *** [default] Error 1
make[4]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon'
make[3]: *** [default] Error 1
make[3]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm'
make[2]: *** [default] Error 1
make[2]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys'
make[1]: *** [subdirs] Error 1
make[1]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src'
make: *** [default] Error 1
==> ERROR: Build Failed.
    Aborting...

EDIT: Finally I managed to get mesa-full compiled, but I don't have 3D.

(also, with the kernel26-rc, option dynpm is not recognised at boot)

Last edited by apollokk (2010-01-23 18:40:26)

csslayer · 2010-01-23 18:43:43

apollokk wrote:

Tried mesa-split, exits with this error:

mklib: Making Linux static library:  libxorgtracker.a
ar: creating libxorgtracker.a
make[4]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/state_trackers/xorg'
make[3]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/state_trackers'
make[2]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium'
make[2]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys'
make[3]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm'
make[4]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon'
make[5]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
rm -f depend
touch depend
fdepend -I/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.2/include -I/usr/lib/gcc/x86_64-unknown-linux-gnu/4.4.2/include-fixed -I. -I../../../../../../src/gallium/include -I../../../../../../src/gallium/auxiliary -I../../../../../../src/gallium/drivers -I/src/gallium/include -I/src/gallium/auxiliary -I/src/gallium/drivers -I../../../../../../src/gallium/drivers/r300 -I/usr/include/drm   radeon_buffer.c radeon_drm.c radeon_r300.c   2> /dev/null
make[5]: [depend] Error 127 (ignored)
make[5]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
make[5]: Entering directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
gcc -c -I. -I../../../../../../src/gallium/include -I../../../../../../src/gallium/auxiliary -I../../../../../../src/gallium/drivers -I/src/gallium/include -I/src/gallium/auxiliary -I/src/gallium/drivers -I../../../../../../src/gallium/drivers/r300 -I/usr/include/drm   -D_GNU_SOURCE -DPTHREADS -DHAVE_POSIX_MEMALIGN -DUSE_XCB -DGLX_USE_TLS -DPTHREADS -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_DIRECT_RENDERING -DGLX_INDIRECT_RENDERING -DHAVE_ALIAS -march=x86-64 -mtune=generic -O2 -pipe -Wall -Wmissing-prototypes -std=c99 -ffast-math -fvisibility=hidden -fno-strict-aliasing  -fPIC  -DUSE_X86_64_ASM -D_GNU_SOURCE -DPTHREADS -DHAVE_POSIX_MEMALIGN -DUSE_XCB -DGLX_USE_TLS -DPTHREADS -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_DIRECT_RENDERING -DGLX_INDIRECT_RENDERING -DHAVE_ALIAS  radeon_buffer.c -o radeon_buffer.o
In file included from radeon_buffer.c:33:
radeon_buffer.h:43:23: error: radeon_bo.h: No such file or directory
radeon_buffer.h:44:23: error: radeon_cs.h: No such file or directory
In file included from radeon_drm.h:46,
                 from radeon_buffer.h:46,
                 from radeon_buffer.c:33:
radeon_r300.h:31:27: error: radeon_cs_gem.h: No such file or directory
radeon_buffer.c:35:27: error: radeon_bo_gem.h: No such file or directory
radeon_buffer.c: In function 'radeon_buffer_create':
radeon_buffer.c:85: error: 'RADEON_GEM_DOMAIN_VRAM' undeclared (first use in this function)
radeon_buffer.c:85: error: (Each undeclared identifier is reported only once
radeon_buffer.c:85: error: for each function it appears in.)
radeon_buffer.c:88: error: 'RADEON_GEM_DOMAIN_GTT' undeclared (first use in this function)
radeon_buffer.c:94: warning: implicit declaration of function 'radeon_bo_open'
radeon_buffer.c:95: warning: assignment makes pointer from integer without a cast
radeon_buffer.c: In function 'radeon_buffer_user_create':
radeon_buffer.c:114: warning: implicit declaration of function 'radeon_bo_map'
radeon_buffer.c:115: error: dereferencing pointer to incomplete type
radeon_buffer.c:116: warning: implicit declaration of function 'radeon_bo_unmap'
radeon_buffer.c: In function 'radeon_buffer_del':
radeon_buffer.c:151: warning: implicit declaration of function 'radeon_bo_unref'
radeon_buffer.c: In function 'radeon_buffer_map':
radeon_buffer.c:173: warning: implicit declaration of function 'radeon_bo_is_busy'
radeon_buffer.c:177: warning: implicit declaration of function 'radeon_bo_is_referenced_by_cs'
radeon_buffer.c:189: error: dereferencing pointer to incomplete type
radeon_buffer.c: In function 'radeon_pipe_winsys':
radeon_buffer.c:290: warning: implicit declaration of function 'radeon_bo_manager_gem_ctor'
radeon_buffer.c:290: warning: assignment makes pointer from integer without a cast
make[5]: *** [radeon_buffer.o] Error 1
make[5]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon/core'
make[4]: *** [default] Error 1
make[4]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm/radeon'
make[3]: *** [default] Error 1
make[3]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys/drm'
make[2]: *** [default] Error 1
make[2]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src/gallium/winsys'
make[1]: *** [subdirs] Error 1
make[1]: Leaving directory `/home/gtklocker/works/mesa-split/src/mesa-build/src'
make: *** [default] Error 1
==> ERROR: Build Failed.
    Aborting...

I think you need to recompile libdrm with exprimental radeon enabled, check the #1 and you can get more information.

edit the official PKGBUILD and uncomment the line then rebuild libdrm.

apollokk · 2010-01-23 18:48:50

I've already done it

Cdh · 2010-01-23 19:13:15

Can someone give a summary what works now and what does not, please?

I tried to enable KMS with my HD4650 and I could not start xorg: Even X -configure made the system completely freeze and not responding to sysrq keys.
I also tried the packages

 libdrm-git mesa-git libgl-git ati-dri-git xf86-video-ati-git glproto-git dri2proto-git

from the nightly repo and with KMS I had the same issue with X. With disabled KMS I could start X but with git packages I did not have proper 3D (only software rasterizer).
KMS itself works good with all setups now.
I use kernel 2.6.32-4 from the repos.

Questions:
Is KMS supposed to work with X with the packages from the repos?
Is KMS supposed to work with X with the packages from git? Do you need any patches for that?
Is KMS supposed to work with X under kernel 2.6.32 at all? Do you need 2.6.33? Do you need anything patched?

yimm · 2010-01-23 19:22:55

Cdh wrote:

did not have proper 3D (only software rasterizer).

Try with mesa-full, it helps me a lot.

apollokk · 2010-01-23 19:29:39

I don't have 3D here, even with mesa-full

If it continues, I'll install Fedora to the laptop.

Cdh · 2010-01-23 19:35:11

yimm wrote:

Try with mesa-full, it helps me a lot.

Definitely will. All the guys above are patching things.
Do I need a patch with mesa-full?

mdias · 2010-01-23 20:32:04

mesa-full here: http://dl.dropbox.com/u/87650/forum_pos … pkg.tar.gz
libdrm-git here: http://dl.dropbox.com/u/87650/forum_pos … pkg.tar.gz

No patches applied or anything. Straight from AUR.

[edit] added libdrm-git package.

Last edited by mdias (2010-01-23 22:25:46)

Arch Linux

#676 2010-01-20 22:03:22

Re: Discussion: the ati/radeon open source drivers & radeon repository

#677 2010-01-21 13:53:54

Re: Discussion: the ati/radeon open source drivers & radeon repository

#678 2010-01-21 17:01:07

Re: Discussion: the ati/radeon open source drivers & radeon repository

#679 2010-01-22 08:23:04

Re: Discussion: the ati/radeon open source drivers & radeon repository

#680 2010-01-22 18:40:00

Re: Discussion: the ati/radeon open source drivers & radeon repository

#681 2010-01-22 21:04:26

Re: Discussion: the ati/radeon open source drivers & radeon repository

#682 2010-01-22 21:39:14

Re: Discussion: the ati/radeon open source drivers & radeon repository

#683 2010-01-22 22:01:32

Re: Discussion: the ati/radeon open source drivers & radeon repository

#684 2010-01-22 22:39:47

Re: Discussion: the ati/radeon open source drivers & radeon repository

#685 2010-01-23 01:27:32

Re: Discussion: the ati/radeon open source drivers & radeon repository

#686 2010-01-23 02:01:02

Re: Discussion: the ati/radeon open source drivers & radeon repository

#687 2010-01-23 13:38:29

Re: Discussion: the ati/radeon open source drivers & radeon repository

#688 2010-01-23 15:37:10

Re: Discussion: the ati/radeon open source drivers & radeon repository

#689 2010-01-23 16:27:01

Re: Discussion: the ati/radeon open source drivers & radeon repository

#690 2010-01-23 16:33:13

Re: Discussion: the ati/radeon open source drivers & radeon repository

#691 2010-01-23 16:35:54

Re: Discussion: the ati/radeon open source drivers & radeon repository

#692 2010-01-23 17:08:47

Re: Discussion: the ati/radeon open source drivers & radeon repository

#693 2010-01-23 17:20:09

Re: Discussion: the ati/radeon open source drivers & radeon repository

#694 2010-01-23 18:43:43

Re: Discussion: the ati/radeon open source drivers & radeon repository

#695 2010-01-23 18:48:50

Re: Discussion: the ati/radeon open source drivers & radeon repository

#696 2010-01-23 19:13:15

Re: Discussion: the ati/radeon open source drivers & radeon repository

#697 2010-01-23 19:22:55

Re: Discussion: the ati/radeon open source drivers & radeon repository

#698 2010-01-23 19:29:39

Re: Discussion: the ati/radeon open source drivers & radeon repository

#699 2010-01-23 19:35:11

Re: Discussion: the ati/radeon open source drivers & radeon repository

#700 2010-01-23 20:32:04

Re: Discussion: the ati/radeon open source drivers & radeon repository

Board footer