You are not logged in.
I have been trying to find a PKGBUILD so I can get this on my system so I can have Direct Rendering. I tried to modify the one in ABS to no avail.
Does anyone know of a working one, or how I should modify the existing for this?
this is where the file is:
Thanks in advance!:D
Offline
What hardware do you have?
Offline
ati radeon x1650(rv535) agp... I need mesa 7.1 to get direct rendering. As well, I believe I will need xserver-1.4.99(1.5).
I am using ati-6.9, and have 2.3.1 drm already;
See here for why:
[url]http://www.phoronix.com/scan.php?page=a … &num=1[/url]
http://www.phoronix.com/forums/showthread.php?t=9951
I am running:
athlon64 3000+
gigabyte k8vnxp
1024 ocz performance
37GB raptor
280GB storage
I was never able to get this card setup w/ fglrx (black screens) when I had ubuntu. The card is not officially supported by ATI, because it is AGP not PCI-E. As well, after reviewing benchmarks and seeing for myself, this card runs great on the new ati drivers... I switched to ARCH from that other distro where it was pretty straight forward to set this up.
Offline
with just the DRM and driver I et this in my Xorg.0.log:
cat /var/log/Xorg.0.log |grep EE
Current Operating System: Linux ephraim 2.6.25-ARCH #1 SMP PREEMPT Mon Jul 14 15:25:51 UTC 2008 i686
(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
(II) Loading extension MIT-SCREEN-SAVER
(EE) AIGLX error: Calling driver entry point failed(EE) AIGLX: reverting to software rendering
Due to running version 7.0.3 of mesa
On Wed, May 7, 2008 at 4:01 AM, Alex Rades <[EMAIL PROTECTED]> wrote:
> Hi,
> I'm having this error during X startup on my x1250:
>
> (EE) AIGLX error: Calling driver entry point failed(EE) AIGLX: reverting to
> software rendering
>
> I'm using:
>
> xserver 1.4
> Mesa 7.0.3
> drm (lib and radeon module) from git master (I've also tried r345-cleanup
> but makes no difference)
>
> do you have ideas?
>
yes 7.0.3 doesn't support the rs cards.
Dave.
Offline
I'm interested in this too!
I've tried to compile, but without success.
Offline
I've compiled libdrm, xorg-server, libgl, mesa, intel-dri from git and it seems to choke up with this error concerning
_glapi_tls_Context.
My chipset is the intel gm965.
Need to get the log file....
Offline
Forgot to enable TLS on the xorg server, it worked after that.
However I got graphical corruption when anything 3D was used.... so i reverted back to he repository version of the drivers.
Offline
There's PKGBUILDs for these in SVN trunk now, I'm working on xorg-server 1.4.99.906 at this moment.
Offline
I downloaded them and spend 1.5 hours compiling (including kernel) and it works! thanks .
Offline
I downloaded them and spend 1.5 hours compiling (including kernel) and it works! thanks .
Would anyone be so kind as to detail the order in which to do this.... I'm still a newb and don't want to break it.
Offline
Execute_Method, i would but it appears to be broken again.
was working with mesa7.1rc3, did mesa7.1rc4 upgrade from testing, now opengl for some reason does not work:
$ glxgears
Error: couldn't get an RGB, Double-buffered visual
glxinfo shows it to be installed and everything else but 3D isn't working.
Cheers .
Offline
Yes, 7.1rc4 is broken, that's why I didn't run db-testing yet.
7.1rc4 merged the new GEM stuff, which is either incompatible with the DRI module, or the xf86-video-intel driver we have. For GEM support, our kernel needs patching, for xf86-video-intel support, we should use code from git master.
I wonder why they call these things Release Candidates while the final feature set of the drivers isn't stable.
Offline
I wonder why they call these things Release Candidates while the final feature set of the drivers isn't stable.
for the same reason the kernel dev's do the same thing. Just look at kernel-2.6.27-rc1. I agree with you, a release candidate should have stable features.
Offline
I notice the GEM stuff too, however, I thought that if GEM wasn't found it should fallback to "classic" and still run? (bit like the TTM buffer manager did)
Offline
Kernel devs close the merge window after rc1. There's no possible way to merge a complete experimental branch between 2.6.27-rc3 and 2.6.27-rc4 as f ar as I know.
The GEM stuff should fallback to classic mode when the kernel module doesn't support it, there's code included for this logic. Somehow there's a problem with the interfacing to the 2D driver that makes problems here.
Offline
On phoronix forums, someone posted that you need to get the git-head of xorg-server to get mesa 7.1rc4 to work . I've not had time to try this yet though, having reverted my mesa back to 7.0.3.
Offline
New mesa release! 7.1, is that going to move into the arch repos?
Offline
well.... I did manage to get all this sorted out and working shortly after my last post using mesa 7.1rc3....However, Like an IDIOT I went and upgraded to the release, that's in testing now, version. As well as the xserver in testing repo. I still have direct rendering, however, I noticed that I am now having a problem with firefox&flash (seperately and together). It doesn't matter if I change my DE or WM. (have been using gnome/compiz) I tried with just openbox, openbox/gnome, and xfce *with no change(actually was worse in some areas)* (I thought maybe a lighter DE/WM would help)... I tried even built firefox from the spookyet PKG, and reinstalled flash to no avail. Does anyone have an idea why firefox would be affected by this?
Offline
Oops its not software rendering, its just vsync!!!
Does anyone know how to disable this?
Last edited by Raedwulf (2008-09-02 09:28:11)
Offline
Sorted it - driconf lets you configure it!
Offline