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

Black Mage · 2009-12-28 09:20:38

First of all AWSOME GUIDE, ive been waiting for opensource 3d for a while now, since catalyst was removed from extra i didnt use it, and had to resort in playing games
exclusivly on windows

so i followed phoronix, and then when stuff updated followed arch, and well luckily i have no problems with getting my hands dirty so at some point when my patience broke,
i looked for help in the forums and found this guide

I am running 2.6.32-1 kernel on arch64 from testing, i have a radeon hd3650 card.
all the packages mentioned in this guide where installed from git, i updated them yesterday so they are all ver 20091227
also for trying out using bin32 wine, and 32bit games like enemy-territory i updated a bit the (check out my comment in each package):
lib32-libdrm-git, lib32-libgl-git and lib32-mesa-git.
and to actually be able to have something i added the lib32-ati-dri-git package.

My configuration is early start kms, to be able to use it i used the hack that was described here somewhere in the topic with the radeon-initrd package,
and putting "radeon" in HOOKS after "base" and "udev" instead of putting it in MODULES in /etc/mkinitcpio.conf.

KMS worked flawlesly since i upgraded my kernel to the testing one, even when 3d broke in some configurations.. (1680x1050 command prompt is so awsome ^_^)

My 3d broke twice, at first when i tried the git repo from this guide, their 64 bit repo is less updated and seems that at least one of the 20091222 package is broke
also a couple of days ago when i saw that mesa entered testing, i wanted to try a configuration with as much stable (non git) packages as possible, but i think
testing mesa package has broken 3d or something.

anyway right now i have working 3d, with some success rate on the next games:

Sauerbraten = it used to crash before my last update of the git packges, now it works but has a messy display (you dont see walls, floors and objects, only character
modules), but still the game itlself runs smoothly (i didnt check fps tough, i never do )

Smokin' Guns = works flawlessly

Teeworlds = its a 2d game, but it uses opengl and will run very slowly without 3d rendering, works flawlesly

Xmoto = as with Teeworlds a 2d game that uses opengl, works flawlesly

Frets on Fire = i remember it working with catalyst but showing no character output, one of the many reasons i hated catalyst, so its nice to see it working perfectly

Yo Frankie = has flikering, not all game objects are rendered, but like sauer it runs smoothly.

Vega Strike = crahes :\ i get to see some stuff perfectly rendered on screen before the crash.

As i mentioned i also install 32bit stuff to try out wine and binary blobs, so if you want to can add an explanation in the guide.
bassicly until tommorow you will be able to use the packages i linked to, by using the PKGBUILD in my comment, tommorow the server will probably update the git package versions.
but its very simple to adjust the packages, just change "pkgver" to the one in the 32bit package repo:
http://nightly.user-helfen-usern.de/index.php?dir=i686
delete the md5sums content, do makepkg -g, edit the md5sum with the hash yo get and do makepkg -s.
to make ati-dri-git work i had to restart, because i was to lazy to find out what script i needed to run.

anyway my test results with some games:

WoW = direct3d doesnt work, opengl shows a messy display, its very messy, but i remember the inteface so i manged to login, and play, so if you count out the fact that you can
barely understant what is what, with all the flickering and messy rendering, it actually worked flawlessly cant wait till the driver will fix this little issues

Teeworlds = my first success, i wanted a game with the simplies opengl bindings, so i downloaded the windows version, works perfectly

Enemy Territory = another GREAT SUCCESS, works great

ill test more games and report

Black Mage · 2009-12-28 09:22:17

oh btw, does anyone know how i can compile the 23rc kernel, without messing with kernel configurations, just using stock arch kerneles configurations with the kernel26rc package.

Nezmer · 2009-12-28 09:51:36

gee wrote:

gee wrote:
Yeah in the past years I have built many kernels to get reiser4 till i got bored of it.
I got radeonhd working with audio now, so I guess I will wait for 2.6.33 official as I am getting a bit tired on working on the driver for this computer.
As a side question, is radeon faster than radeonhd today or anything else apart from getting KMS?

Thank you for your help!
Well I have just got a 2.6.33 kernel for other reasons so I thought of trying this, but I could not get hdmi audio.
I could not find a howto either so maybe I am missing some steps...

hmm, HD audio support was added before rc1 It seems. So, you should be able to make it work.

Just to make sure:

grep r600_hdmi.o /path/to/kernel/source/drivers/gpu/drm/radeon/Makefile

The result should look like this:

r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o

If you're really interested, you can ask "dri-devel@lists.sourceforge.net". You should include dmesg output and any other needed logs of course.

handy · 2009-12-28 10:36:57

Black Mage wrote:

oh btw, does anyone know how i can compile the 23rc kernel, without messing with kernel configurations, just using stock arch kerneles configurations with the kernel26rc package.

If you can't find the particular kernel version in AUR... then?

By the way, thanks for your great post... I thoroughly enjoyed it, & learned from it too... Bravo...

Last edited by handy (2009-12-28 10:38:25)

Black Mage · 2009-12-28 15:18:13

ok i done some more testing.

i bassicly tried every direct3d app i could think off that was quickly avaible, from simple 3d apps to serious games, from old directx7 to directx9
most segfaulted before even starting, those that started, halted their loading right before the screen that should have 3d.
my conclusion is that right now the OS drivers and wine's direct3d implention dont play along.

anyway i tested

DeusEx = opengl renderer works, and i even found a fix for every issue i had with deus ex on linux (wine related) so now im happy with one of my
favorite games perfectly working on my open source drivers

Savage 1 = this one showd a complete white screen at first, but its a common issue not only with the radeon driver (altough it is a driver issue)
anyway the fix is to "echo "setsave vid_compressTextures 0" >> game/startup.cfg" each time the game starts (in the bin script of the game),
but this just gave me a blnk screen with writings on it and a glittering cursor, well at least i could now find the "quit" button

Savage 2 = didnt work throwed an error "Savage2 - Fatal Error: GLSL 1.20 not available." well its only "mostly" finished in the wiki

Freeorion = didnt compile from source, so i resorted to an i686 package, it segfaulted before starting, not sure if it has anything to do with
the driver.

handy you are welcome, what i meant was that when you compile a kernel you need to configure it first, bassicly choose wich drivers should
be build in, build as modules or not build at all, its a tedious process and you need to know your hardware and the naming of the approtiate drivers.
most important as far as i remember from my gentoo time, was to build in your motherboard chipset.

anyway i dont want to mess again with that, i just want to use the aur kernel26rc pacakage with the general setting for kernel compiling that the
distro uses.

so my question is, how exacly do i do that?

Last edited by Black Mage (2009-12-28 15:50:29)

gee · 2009-12-28 19:58:50

Nezmer wrote:

gee wrote:
gee wrote:
Yeah in the past years I have built many kernels to get reiser4 till i got bored of it.
I got radeonhd working with audio now, so I guess I will wait for 2.6.33 official as I am getting a bit tired on working on the driver for this computer.
As a side question, is radeon faster than radeonhd today or anything else apart from getting KMS?

Thank you for your help!
Well I have just got a 2.6.33 kernel for other reasons so I thought of trying this, but I could not get hdmi audio.
I could not find a howto either so maybe I am missing some steps...
hmm, HD audio support was added before rc1 It seems. So, you should be able to make it work.
Just to make sure:
grep r600_hdmi.o /path/to/kernel/source/drivers/gpu/drm/radeon/Makefile
The result should look like this:
r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o
If you're really interested, you can ask "dri-devel@lists.sourceforge.net". You should include dmesg output and any other needed logs of course.

I got an rc2 so I should have what is needed, but as with radeonhd maybe there is an option to activate it that I am missing. I will look at the dri mailing list, thank you!

gentux · 2009-12-28 21:04:43

Hi all,
I have a problem with a radeon hd4570 graphic card.
When I want to build xf86-video-ati-git from the git server like it is said in the first page, I get an error:

==> Starting configure...                                                     
autoreconf: Entering directory `.'                                            
autoreconf: configure.ac: not using Gettext                                   
autoreconf: running: aclocal                                                  
/usr/share/aclocal/progsreiserfs.m4:13: warning: underquoted definition of AC_CHECK_LIBREISERFS                                                             
/usr/share/aclocal/progsreiserfs.m4:13:   run info '(automake)Extending aclocal'                                                                            
/usr/share/aclocal/progsreiserfs.m4:13:   or see http://sources.redhat.com/automake/automake.html#Extending-aclocal                                         
autoreconf: configure.ac: tracing                                             
autoreconf: running: libtoolize --copy                                        
libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, `.'.                
libtoolize: copying file `./ltmain.sh'                                        
libtoolize: Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and   
libtoolize: rerunning libtoolize, to keep the correct libtool macros in-tree. 
libtoolize: Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.        
/usr/share/aclocal/progsreiserfs.m4:13: warning: underquoted definition of AC_CHECK_LIBREISERFS                                                             
/usr/share/aclocal/progsreiserfs.m4:13:   run info '(automake)Extending aclocal'                                                                            
/usr/share/aclocal/progsreiserfs.m4:13:   or see http://sources.redhat.com/automake/automake.html#Extending-aclocal                                         
autoreconf: running: /usr/bin/autoconf                                        
autoreconf: running: /usr/bin/autoheader                                      
autoreconf: running: automake --add-missing --copy --no-force                 
configure.ac:36: installing `./config.guess'                                  
configure.ac:36: installing `./config.sub'                                    
configure.ac:40: installing `./install-sh'                                    
configure.ac:40: installing `./missing'                                       
src/Makefile.am: installing `./depcomp'                                       
autoreconf: Leaving directory `.'                                             
configure: WARNING: unrecognized options: --enable-shave                      
checking for gcc... gcc                                                       
checking whether the C compiler works... yes                                  
checking for C compiler default output file name... a.out                     
checking for suffix of executables...                                         
checking whether we are cross compiling... no                                 
checking for suffix of object files... o                                      
checking whether we are using the GNU C compiler... yes                       
checking whether gcc accepts -g... yes                                        
checking for gcc option to accept ISO C89... none needed                      
checking how to run the C preprocessor... gcc -E                              
checking for grep that handles long lines and -e... /bin/grep                 
checking for egrep... /bin/grep -E                                            
checking for ANSI C header files... yes                                       
checking for sys/types.h... yes                                               
checking for sys/stat.h... yes                                                
checking for stdlib.h... yes                                                  
checking for string.h... yes                                                  
checking for memory.h... yes                                                  
checking for strings.h... yes                                                 
checking for inttypes.h... yes                                                
checking for stdint.h... yes                                                  
checking for unistd.h... yes                                                  
checking for gcc option to accept ISO C99... -std=gnu99                       
checking for pkg-config... /usr/bin/pkg-config                                
checking pkg-config is at least version 0.9.0... yes                          
checking build system type... x86_64-unknown-linux-gnu                        
checking host system type... x86_64-unknown-linux-gnu                         
checking for a BSD-compatible install... /bin/install -c                      
checking whether build environment is sane... yes                             
checking for a thread-safe mkdir -p... /bin/mkdir -p                          
checking for gawk... gawk                                                     
checking whether make sets $(MAKE)... yes                                     
checking for style of include used by make... GNU                             
checking dependency style of gcc -std=gnu99... gcc3                           
checking for special C compiler options needed for large files... no          
checking for _FILE_OFFSET_BITS value needed for large files... no             
checking whether to enable maintainer-specific portions of Makefiles... yes   
checking for a sed that does not truncate output... /bin/sed                  
checking for fgrep... /bin/grep -F                                            
checking for ld used by gcc -std=gnu99... /usr/bin/ld                         
checking if the linker (/usr/bin/ld) is GNU ld... yes                         
checking for BSD- or MS-compatible name lister (nm)... /usr/bin/nm -B         
checking the name lister (/usr/bin/nm -B) interface... BSD nm                 
checking whether ln -s works... yes                                           
checking the maximum length of command line arguments... 1572864              
checking whether the shell understands some XSI constructs... yes             
checking whether the shell understands "+="... yes                            
checking for /usr/bin/ld option to reload object files... -r                  
checking for objdump... objdump                                               
checking how to recognize dependent libraries... pass_all                     
checking for ar... ar                                                         
checking for strip... strip                                                   
checking for ranlib... ranlib                                                 
checking command to parse /usr/bin/nm -B output from gcc -std=gnu99 object... ok                                                                            
checking for dlfcn.h... yes                                                   
checking for objdir... .libs                                                  
checking if gcc -std=gnu99 supports -fno-rtti -fno-exceptions... no           
checking for gcc -std=gnu99 option to produce PIC... -fPIC -DPIC              
checking if gcc -std=gnu99 PIC flag -fPIC -DPIC works... yes                  
checking if gcc -std=gnu99 static flag -static works... yes                   
checking if gcc -std=gnu99 supports -c -o file.o... yes                       
checking if gcc -std=gnu99 supports -c -o file.o... (cached) yes              
checking whether the gcc -std=gnu99 linker (/usr/bin/ld -m elf_x86_64) supports shared libraries... yes                                                     
checking whether -lc should be explicitly linked in... no                     
checking dynamic linker characteristics... GNU/Linux ld.so                    
checking how to hardcode library paths into programs... immediate             
checking whether stripping libraries is possible... yes                       
checking if libtool supports shared libraries... yes                          
checking whether to build shared libraries... yes                             
checking whether to build static libraries... no                              
checking for gcc... (cached) gcc                                              
checking whether we are using the GNU C compiler... (cached) yes              
checking whether gcc accepts -g... (cached) yes                               
checking for gcc option to accept ISO C89... (cached) none needed             
checking if XINERAMA is defined... yes                                        
checking if RANDR is defined... yes                                           
checking if RENDER is defined... yes                                          
checking if XV is defined... yes                                              
checking if DPMSExtension is defined... yes                                   
checking for XORG... yes                                                      
checking for XEXT... yes                                                      
checking for ANSI C header files... (cached) yes                              
checking for DRI... yes                                                       
checking for dri.h... yes                                                     
checking for sarea.h... yes                                                   
checking for damage.h... yes                                                  
checking whether to include DRI support... yes                                
checking for xf86drmMode.h... yes                                             
checking for LIBDRM_RADEON... yes                                             
checking for xf86Modes.h... yes                                               
checking whether to enable EXA support... yes                                 
checking exa.h usability... yes                                               
checking exa.h presence... yes                                                
checking for exa.h... yes                                                     
checking whether EXA version is at least 2.0.0... yes                         
checking whether xf86XVFillKeyHelperDrawable is declared... yes               
checking whether xf86ModeBandwidth is declared... yes                         
checking whether xf86_crtc_clip_video_helper is declared... yes               
checking whether xf86RotateFreeShadow is declared... yes                      
checking whether pci_device_enable is declared... yes                         
checking whether XSERVER_LIBPCIACCESS is declared... yes                      
configure: X server has new mode code                                         
checking for PCIACCESS... yes                                                 
checking byteswap.h usability... yes                                          
checking byteswap.h presence... yes                                           
checking for byteswap.h... yes                                                
checking sys/endian.h usability... no                                         
checking sys/endian.h presence... no                                          
checking for sys/endian.h... no                                               
checking for /usr/share/sgml/X11/defs.ent... no                               
checking for linuxdoc... no                                                   
checking for ps2pdf... /usr/bin/ps2pdf                                        
checking Whether to build documentation... no                                 
checking Whether to build pdf documentation... yes                            
configure: The atimisc sub-driver has been split out to xf86-video-mach64:    
    git://anongit.freedesktop.org/git/xorg/driver/xf86-video-mach64           
Please install that driver as well for mach64-based cards.                    

configure: The r128 sub-driver has been split out to xf86-video-r128:
    git://anongit.freedesktop.org/git/xorg/driver/xf86-video-r128    
Please install that driver as well for rage128-based cards.          

configure: creating ./config.status
config.status: creating Makefile   
config.status: creating src/Makefile
config.status: creating man/Makefile
config.status: creating config.h    
config.status: executing depfiles commands
config.status: executing libtool commands 
configure: WARNING: unrecognized options: --enable-shave

        prefix:              /usr
        exec_prefix:         ${prefix}
        libdir:              ${exec_prefix}/lib
        includedir:          ${prefix}/include 

        Kernel modesetting:  yes

        CFLAGS:              -march=x86-64 -mtune=generic -O2 -pipe
        CXXFLAGS:            -march=x86-64 -mtune=generic -O2 -pipe
        Macros:                                                    

        Run 'make' to build xf86-video-ati

==> Starting make...
make  all-recursive 
make[1]: entrant dans le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build »                                               
Making all in src                                                             
make[2]: entrant dans le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build/src »
  CC     ati.lo
  CC     atimodule.lo
  CCLD   ati_drv.la
  CC     radeon_accel.lo
Dans le fichier inclus à partir de radeon.h:76,
          à partir de radeon_accel.c:79:
/usr/include/GL/glxint.h:36:19: erreur: GL/gl.h : Aucun fichier ou dossier de ce type
In file included from radeon.h:76,
                 from radeon_accel.c:79:
/usr/include/GL/glxint.h:103: erreur: expected specifier-qualifier-list before 'GLboolean'
make[2]: *** [radeon_accel.lo] Erreur 1
make[2]: quittant le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build/src »
make[1]: *** [all-recursive] Erreur 1
make[1]: quittant le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build »
make: *** [all] Erreur 2

Of course, I have built the others packages before (libdrm-git glproto-git dri2proto-git libgl-git mesa-git ati-dri-git).
I have no idea how to solve this problem.
Thanks for your help !

Nezmer · 2009-12-28 21:04:47

gee wrote:

I got an rc2 so I should have what is needed, but as with radeonhd maybe there is an option to activate it that I am missing. I will look at the dri mailing list, thank you!

Looking at the code, It should be enabled by default.
To be sure, Add "radeon.audio=1" to the kernel boot parameters.

gee · 2009-12-28 22:11:57

Awesome I will try that tonight thank you.
Since you know so much about all that, do you actually work on any of these projects?

gee · 2009-12-28 22:13:36

gentux wrote:

Hi all,
I have a problem with a radeon hd4570 graphic card.
When I want to build xf86-video-ati-git from the git server like it is said in the first page, I get an error:

==> Starting configure...                                                     
autoreconf: Entering directory `.'                                            
autoreconf: configure.ac: not using Gettext                                   
autoreconf: running: aclocal                                                  
/usr/share/aclocal/progsreiserfs.m4:13: warning: underquoted definition of AC_CHECK_LIBREISERFS                                                             
/usr/share/aclocal/progsreiserfs.m4:13:   run info '(automake)Extending aclocal'                                                                            
/usr/share/aclocal/progsreiserfs.m4:13:   or see http://sources.redhat.com/automake/automake.html#Extending-aclocal                                         
autoreconf: configure.ac: tracing                                             
autoreconf: running: libtoolize --copy                                        
libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, `.'.                
libtoolize: copying file `./ltmain.sh'                                        
libtoolize: Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and   
libtoolize: rerunning libtoolize, to keep the correct libtool macros in-tree. 
libtoolize: Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.        
/usr/share/aclocal/progsreiserfs.m4:13: warning: underquoted definition of AC_CHECK_LIBREISERFS                                                             
/usr/share/aclocal/progsreiserfs.m4:13:   run info '(automake)Extending aclocal'                                                                            
/usr/share/aclocal/progsreiserfs.m4:13:   or see http://sources.redhat.com/automake/automake.html#Extending-aclocal                                         
autoreconf: running: /usr/bin/autoconf                                        
autoreconf: running: /usr/bin/autoheader                                      
autoreconf: running: automake --add-missing --copy --no-force                 
configure.ac:36: installing `./config.guess'                                  
configure.ac:36: installing `./config.sub'                                    
configure.ac:40: installing `./install-sh'                                    
configure.ac:40: installing `./missing'                                       
src/Makefile.am: installing `./depcomp'                                       
autoreconf: Leaving directory `.'                                             
configure: WARNING: unrecognized options: --enable-shave                      
checking for gcc... gcc                                                       
checking whether the C compiler works... yes                                  
checking for C compiler default output file name... a.out                     
checking for suffix of executables...                                         
checking whether we are cross compiling... no                                 
checking for suffix of object files... o                                      
checking whether we are using the GNU C compiler... yes                       
checking whether gcc accepts -g... yes                                        
checking for gcc option to accept ISO C89... none needed                      
checking how to run the C preprocessor... gcc -E                              
checking for grep that handles long lines and -e... /bin/grep                 
checking for egrep... /bin/grep -E                                            
checking for ANSI C header files... yes                                       
checking for sys/types.h... yes                                               
checking for sys/stat.h... yes                                                
checking for stdlib.h... yes                                                  
checking for string.h... yes                                                  
checking for memory.h... yes                                                  
checking for strings.h... yes                                                 
checking for inttypes.h... yes                                                
checking for stdint.h... yes                                                  
checking for unistd.h... yes                                                  
checking for gcc option to accept ISO C99... -std=gnu99                       
checking for pkg-config... /usr/bin/pkg-config                                
checking pkg-config is at least version 0.9.0... yes                          
checking build system type... x86_64-unknown-linux-gnu                        
checking host system type... x86_64-unknown-linux-gnu                         
checking for a BSD-compatible install... /bin/install -c                      
checking whether build environment is sane... yes                             
checking for a thread-safe mkdir -p... /bin/mkdir -p                          
checking for gawk... gawk                                                     
checking whether make sets $(MAKE)... yes                                     
checking for style of include used by make... GNU                             
checking dependency style of gcc -std=gnu99... gcc3                           
checking for special C compiler options needed for large files... no          
checking for _FILE_OFFSET_BITS value needed for large files... no             
checking whether to enable maintainer-specific portions of Makefiles... yes   
checking for a sed that does not truncate output... /bin/sed                  
checking for fgrep... /bin/grep -F                                            
checking for ld used by gcc -std=gnu99... /usr/bin/ld                         
checking if the linker (/usr/bin/ld) is GNU ld... yes                         
checking for BSD- or MS-compatible name lister (nm)... /usr/bin/nm -B         
checking the name lister (/usr/bin/nm -B) interface... BSD nm                 
checking whether ln -s works... yes                                           
checking the maximum length of command line arguments... 1572864              
checking whether the shell understands some XSI constructs... yes             
checking whether the shell understands "+="... yes                            
checking for /usr/bin/ld option to reload object files... -r                  
checking for objdump... objdump                                               
checking how to recognize dependent libraries... pass_all                     
checking for ar... ar                                                         
checking for strip... strip                                                   
checking for ranlib... ranlib                                                 
checking command to parse /usr/bin/nm -B output from gcc -std=gnu99 object... ok                                                                            
checking for dlfcn.h... yes                                                   
checking for objdir... .libs                                                  
checking if gcc -std=gnu99 supports -fno-rtti -fno-exceptions... no           
checking for gcc -std=gnu99 option to produce PIC... -fPIC -DPIC              
checking if gcc -std=gnu99 PIC flag -fPIC -DPIC works... yes                  
checking if gcc -std=gnu99 static flag -static works... yes                   
checking if gcc -std=gnu99 supports -c -o file.o... yes                       
checking if gcc -std=gnu99 supports -c -o file.o... (cached) yes              
checking whether the gcc -std=gnu99 linker (/usr/bin/ld -m elf_x86_64) supports shared libraries... yes                                                     
checking whether -lc should be explicitly linked in... no                     
checking dynamic linker characteristics... GNU/Linux ld.so                    
checking how to hardcode library paths into programs... immediate             
checking whether stripping libraries is possible... yes                       
checking if libtool supports shared libraries... yes                          
checking whether to build shared libraries... yes                             
checking whether to build static libraries... no                              
checking for gcc... (cached) gcc                                              
checking whether we are using the GNU C compiler... (cached) yes              
checking whether gcc accepts -g... (cached) yes                               
checking for gcc option to accept ISO C89... (cached) none needed             
checking if XINERAMA is defined... yes                                        
checking if RANDR is defined... yes                                           
checking if RENDER is defined... yes                                          
checking if XV is defined... yes                                              
checking if DPMSExtension is defined... yes                                   
checking for XORG... yes                                                      
checking for XEXT... yes                                                      
checking for ANSI C header files... (cached) yes                              
checking for DRI... yes                                                       
checking for dri.h... yes                                                     
checking for sarea.h... yes                                                   
checking for damage.h... yes                                                  
checking whether to include DRI support... yes                                
checking for xf86drmMode.h... yes                                             
checking for LIBDRM_RADEON... yes                                             
checking for xf86Modes.h... yes                                               
checking whether to enable EXA support... yes                                 
checking exa.h usability... yes                                               
checking exa.h presence... yes                                                
checking for exa.h... yes                                                     
checking whether EXA version is at least 2.0.0... yes                         
checking whether xf86XVFillKeyHelperDrawable is declared... yes               
checking whether xf86ModeBandwidth is declared... yes                         
checking whether xf86_crtc_clip_video_helper is declared... yes               
checking whether xf86RotateFreeShadow is declared... yes                      
checking whether pci_device_enable is declared... yes                         
checking whether XSERVER_LIBPCIACCESS is declared... yes                      
configure: X server has new mode code                                         
checking for PCIACCESS... yes                                                 
checking byteswap.h usability... yes                                          
checking byteswap.h presence... yes                                           
checking for byteswap.h... yes                                                
checking sys/endian.h usability... no                                         
checking sys/endian.h presence... no                                          
checking for sys/endian.h... no                                               
checking for /usr/share/sgml/X11/defs.ent... no                               
checking for linuxdoc... no                                                   
checking for ps2pdf... /usr/bin/ps2pdf                                        
checking Whether to build documentation... no                                 
checking Whether to build pdf documentation... yes                            
configure: The atimisc sub-driver has been split out to xf86-video-mach64:    
    git://anongit.freedesktop.org/git/xorg/driver/xf86-video-mach64           
Please install that driver as well for mach64-based cards.                    

configure: The r128 sub-driver has been split out to xf86-video-r128:
    git://anongit.freedesktop.org/git/xorg/driver/xf86-video-r128    
Please install that driver as well for rage128-based cards.          

configure: creating ./config.status
config.status: creating Makefile   
config.status: creating src/Makefile
config.status: creating man/Makefile
config.status: creating config.h    
config.status: executing depfiles commands
config.status: executing libtool commands 
configure: WARNING: unrecognized options: --enable-shave

        prefix:              /usr
        exec_prefix:         ${prefix}
        libdir:              ${exec_prefix}/lib
        includedir:          ${prefix}/include 

        Kernel modesetting:  yes

        CFLAGS:              -march=x86-64 -mtune=generic -O2 -pipe
        CXXFLAGS:            -march=x86-64 -mtune=generic -O2 -pipe
        Macros:                                                    

        Run 'make' to build xf86-video-ati

==> Starting make...
make  all-recursive 
make[1]: entrant dans le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build »                                               
Making all in src                                                             
make[2]: entrant dans le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build/src »
  CC     ati.lo
  CC     atimodule.lo
  CCLD   ati_drv.la
  CC     radeon_accel.lo
Dans le fichier inclus à partir de radeon.h:76,
          à partir de radeon_accel.c:79:
/usr/include/GL/glxint.h:36:19: erreur: GL/gl.h : Aucun fichier ou dossier de ce type
In file included from radeon.h:76,
                 from radeon_accel.c:79:
/usr/include/GL/glxint.h:103: erreur: expected specifier-qualifier-list before 'GLboolean'
make[2]: *** [radeon_accel.lo] Erreur 1
make[2]: quittant le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build/src »
make[1]: *** [all-recursive] Erreur 1
make[1]: quittant le répertoire « /home/loic/Downloads/R700/xf86-video-ati-git/src/xf86-video-ati-build »
make: *** [all] Erreur 2

Of course, I have built the others packages before (libdrm-git glproto-git dri2proto-git libgl-git mesa-git ati-dri-git).
I have no idea how to solve this problem.
Thanks for your help !

I had the same problem and manually downloaded gl.h from the mesa repo.
Not sure why it is not there anymore.

Although I am not sure if that fully helped me build it, now I use the nightly repo so I don't care anymore.

descendent87 · 2009-12-28 22:36:16

Is everything needed now in the repo's or do you still need to build git packages to get 3D acceleration?

gee · 2009-12-28 23:02:52

The nightly repo gives me everything but radeonhd I believe.

Nezmer · 2009-12-28 23:35:27

gee wrote:

Awesome I will try that tonight thank you.
Since you know so much about all that, do you actually work on any of these projects?

I'm not a programmer. I just looked at those lines in "drivers/gpu/drm/radeon/radeon_drv.c" :

int radeon_no_wb;
int radeon_modeset = -1;
int radeon_dynclks = -1;
int radeon_r4xx_atom = 0;
int radeon_agpmode = 0; 
int radeon_vram_limit = 0; 
int radeon_gart_size = 512; /* default gart size */
int radeon_benchmarking = 0; 
int radeon_testing = 0; 
int radeon_connector_table = 0; 
int radeon_tv = 1; 
int radeon_new_pll = 1; 
int radeon_audio = 1;

Those are module parameters you can generally use when you modprobe or you can pass them to your boot loader. So If you for example prevent the radeon module from auto-loading and modprobe it manually, you can do this:

modprobe -v radeon modeset=1 dynclks=1 audio=1

Additionally, you can see the current parameter values through "/sys/module/radeon/parameters/".

I don't know why I just wrote all this!

bmentink · 2009-12-29 00:53:07

Perry3D wrote:

In kernel26 from testing KMS is now disabled by default. You have to set it manually.
I didn't tested it. Maybe the modeset kernel parameter suffice.
/edit: I use:
- kernel26-git, kernel26-firmware-git + radeon_ucode
- glproto-git, dri2proto-git
- libdrm-git
- mesa-split
- xf86-video-ati-git
And it works very nice. Compiz is fast as hell without tearing. Also videos in xbmc (gl output).

Ok, I have most packages installed. I have installed kernel26-firmware-git and am trying to compile radeon_ucode, however I am getting
the following error:

[

bmentink@Laptop2 radeon_ucode]$ makepkg
==> Making package: radeon_ucode 1-2 any (Tue Dec 29 13:48:20 NZDT 2009)
==> Checking Runtime Dependencies...
==> Checking Buildtime Dependencies...
==> Retrieving Sources...
  -> Found R700_rlc.bin in build dir
  -> Found R600_rlc.bin in build dir
  -> Found LICENSE.radeon in build dir
==> Validating source files with md5sums...
    R700_rlc.bin ... Passed
    R600_rlc.bin ... Passed
    LICENSE.radeon ... Passed
==> Extracting Sources...
==> Removing existing pkg/ directory...
==> Entering fakeroot environment...
==> Starting build()...
/lib/firmware/radeon directory does not exist. Please install
    kernel26-firmware or any package that provides kernel firmware
    files.
==> ERROR: Build Failed.
    Aborting..

.

Bit strange as I installed the firmware package.

Nezmer · 2009-12-29 01:05:25

bmentink wrote:

Perry3D wrote:
In kernel26 from testing KMS is now disabled by default. You have to set it manually.
I didn't tested it. Maybe the modeset kernel parameter suffice.
/edit: I use:
- kernel26-git, kernel26-firmware-git + radeon_ucode
- glproto-git, dri2proto-git
- libdrm-git
- mesa-split
- xf86-video-ati-git
And it works very nice. Compiz is fast as hell without tearing. Also videos in xbmc (gl output).
Ok, I have most packages installed. I have installed kernel26-firmware-git and am trying to compile radeon_ucode, however I am getting
the following error:

[
bmentink@Laptop2 radeon_ucode]$ makepkg
==> Making package: radeon_ucode 1-2 any (Tue Dec 29 13:48:20 NZDT 2009)
==> Checking Runtime Dependencies...
==> Checking Buildtime Dependencies...
==> Retrieving Sources...
  -> Found R700_rlc.bin in build dir
  -> Found R600_rlc.bin in build dir
  -> Found LICENSE.radeon in build dir
==> Validating source files with md5sums...
    R700_rlc.bin ... Passed
    R600_rlc.bin ... Passed
    LICENSE.radeon ... Passed
==> Extracting Sources...
==> Removing existing pkg/ directory...
==> Entering fakeroot environment...
==> Starting build()...
/lib/firmware/radeon directory does not exist. Please install
    kernel26-firmware or any package that provides kernel firmware
    files.
==> ERROR: Build Failed.
    Aborting..
.
Bit strange as I installed the firmware package.

That's a check I added to the PKGBUILD to make sure other in-tree firmware files exist. Can you post the output of

find /lib/firmware/radeon

and

pacman -Ql kernel26-firmware-git

Perry3D · 2009-12-29 01:11:17

At the moment i have a problem compiling the kernel26-firmware-git package:

==> Connecting to GIT server....
Your configuration specifies to merge with the ref 'master'
from the remote, but no such ref was fetched.

Are you using the x64 architecture? Then i can send you the last working package.

gee · 2009-12-29 01:20:27

Nezmer wrote:

I'm not a programmer. I just looked at those lines in "drivers/gpu/drm/radeon/radeon_drv.c" :
int radeon_no_wb;
int radeon_modeset = -1;
int radeon_dynclks = -1;
int radeon_r4xx_atom = 0;
int radeon_agpmode = 0; 
int radeon_vram_limit = 0; 
int radeon_gart_size = 512; /* default gart size */
int radeon_benchmarking = 0; 
int radeon_testing = 0; 
int radeon_connector_table = 0; 
int radeon_tv = 1; 
int radeon_new_pll = 1; 
int radeon_audio = 1;
Those are module parameters you can generally use when you modprobe or you can pass them to your boot loader. So If you for example prevent the radeon module from auto-loading and modprobe it manually, you can do this:
modprobe -v radeon modeset=1 dynclks=1 audio=1
Additionally, you can see the current parameter values through "/sys/module/radeon/parameters/".
I don't know why I just wrote all this!

Cool thank you

Nezmer · 2009-12-29 01:21:31

Edited to add more explanation.

After looking at the git tree kernel26-firmware-git package uses. It contains a collection of out-of-tree firmware files like wireless firmware. Its name is ambiguous IMHO and I think installing it is a bad idea. Additionally, Its provides line is completely wrong.

@bmentink
You need in-tree firmware files. Files you can find in "kernelSourceTree/firmware/". Look at testing/kernel26-firmware for reference. If you have those firmware files properly installed even without a separate firmware package, you can remove the dependency from the radeon_ucode PKGBUILD.

Last edited by Nezmer (2009-12-29 01:31:48)

bmentink · 2009-12-29 01:45:34

Nezmer wrote:

Edited to add more explanation.
After looking at the git tree kernel26-firmware-git package uses. It contains a collection of out-of-tree firmware files like wireless firmware. Its name is ambiguous IMHO and I think installing it is a bad idea. Additionally, Its provides line is completely wrong.
@bmentink
You need in-tree firmware files. Files you can find in "kernelSourceTree/firmware/". Look at testing/kernel26-firmware for reference. If you have those firmware files properly installed even without a separate firmware package, you can remove the dependency from the radeon_ucode PKGBUILD.

Thanks, I have taken your advice and removed the kernel26-firmware-git package and have installed the package from testing. The radeon_ucode package now installs fine.

Before I go ahead and install the kernel26-git(it compiled fine), is their any advice you can give me regarding back ups ..... I am a bit nervous.:/

I have made an extra copy of my kernel and put another entry in grub/menu.1st .... just so I have a back-door in case the kernel borks ..
By the way, when compiling the kernel, I just accepted the default configuration ... I presume this is ok?

Cheers,

gee · 2009-12-29 04:51:21

....

Last edited by gee (2009-12-29 06:36:45)

bmentink · 2009-12-29 04:59:58

bmentink wrote:

Nezmer wrote:
Edited to add more explanation.
After looking at the git tree kernel26-firmware-git package uses. It contains a collection of out-of-tree firmware files like wireless firmware. Its name is ambiguous IMHO and I think installing it is a bad idea. Additionally, Its provides line is completely wrong.
@bmentink
You need in-tree firmware files. Files you can find in "kernelSourceTree/firmware/". Look at testing/kernel26-firmware for reference. If you have those firmware files properly installed even without a separate firmware package, you can remove the dependency from the radeon_ucode PKGBUILD.
Thanks, I have taken your advice and removed the kernel26-firmware-git package and have installed the package from testing. The radeon_ucode package now installs fine.
Before I go ahead and install the kernel26-git(it compiled fine), is their any advice you can give me regarding back ups ..... I am a bit nervous.:/
I have made an extra copy of my kernel and put another entry in grub/menu.1st .... just so I have a back-door in case the kernel borks ..
By the way, when compiling the kernel, I just accepted the default configuration ... I presume this is ok?
Cheers,

An update: I installed the "testing" version of the kernel and together with all the-git versions of the radeon drivers I now have 3D acceleration .... yay.
I get 2000fps with glxgears instead of 300fps and all videos play great including HD.

However, I am now missing a library (libGLU.so) when trying to use some applications.
I presume this library is in libgl? I have installed libgl-git ....
(I do have a libGL.so in /usr/lib but not ....GLU.so)

Can anyone help on this ..

Thanks,

Anyone know what is going on here ...

Nezmer · 2009-12-29 10:40:38

/usr/lib/libGLU.so is owned by mesa-git.

If you have this file and still encountering problems, please tell us.

Cdh · 2009-12-29 10:44:00

chris@chrispc> pkgfile libGLU.so
extra/mesa
testing/mesa
pkgfile libGLU.so  7,40s user 1,17s system 73% cpu 11,696 total
chris@chrispc> locate libGLU.so
/usr/google-earth/libGLU.so.1
/usr/lib/libGLU.so
/usr/lib/libGLU.so.1
/usr/lib/libGLU.so.1.3.070800
chris@chrispc> pacman -Qo /usr/lib/libGLU.so
/usr/lib/libGLU.so ist in mesa-git 20091228-1 enthalten
chris@chrispc>

Looks fine for me. Maybe reinstall mesa-git?

bmentink · 2009-12-29 19:43:44

Nezmer wrote:

/usr/lib/libGLU.so is owned by mesa-git.
If you have this file and still encountering problems, please tell us.

mesa-git was the problem, I un-installed mesa-git and installed mesa from the normal place, and all was good again.
It seems the package is missing from mesa-git, or is installed in the wrong place.

I am not sure why I need mesa-git as glxgears is working at full speed (2000fps) and HD movies play great, also tested on some 3D games with no issues.

So to recap, I have installed:

libdrm-git
libgl-git
xf86-video-ati-git
ati-dri-git

mesa (normal)
2.6.32-2 Kernel from testing

Cheers,

PS: And it seems KMS is working too, as I have the module installed in /etc/mkinitcpio.conf and it has loaded fine.
Is there any way to test this for sure?

Last edited by bmentink (2009-12-29 19:46:18)

Nezmer · 2009-12-29 19:57:18

bmentink wrote:

PS: And it seems KMS is working too, as I have the module installed in /etc/mkinitcpio.conf and it has loaded fine.
Is there any way to test this for sure?

cat /sys/module/radeon/parameters/modeset

BTW, What I said earlier about kernel26-firmware-git is completely inaccurate. I was looking at the wrong git tree. You probably had issues because the git repo "git://git.kernel.org/pub/scm/linux/kernel/git/dwmw2/linux-firmware-from-kernel.git" looks empty now. It looks empty from here at least.

Arch Linux

#501 2009-12-28 09:20:38

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

#502 2009-12-28 09:22:17

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

#503 2009-12-28 09:51:36

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

#504 2009-12-28 10:36:57

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

#505 2009-12-28 15:18:13

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

#506 2009-12-28 19:58:50

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

#507 2009-12-28 21:04:43

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

#508 2009-12-28 21:04:47

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

#509 2009-12-28 22:11:57

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

#510 2009-12-28 22:13:36

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

#511 2009-12-28 22:36:16

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

#512 2009-12-28 23:02:52

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

#513 2009-12-28 23:35:27

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

#514 2009-12-29 00:53:07

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

#515 2009-12-29 01:05:25

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

#516 2009-12-29 01:11:17

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

#517 2009-12-29 01:20:27

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

#518 2009-12-29 01:21:31

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

#519 2009-12-29 01:45:34

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

#520 2009-12-29 04:51:21

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

#521 2009-12-29 04:59:58

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

#522 2009-12-29 10:40:38

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

#523 2009-12-29 10:44:00

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

#524 2009-12-29 19:43:44

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

#525 2009-12-29 19:57:18

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

Board footer