Compiz Running With Mesa On R600/700 GPUs

[Louise]

I'm the pointy-haired boss. Things work better when I don't code

I didn't thought managers and bosses read Dilbert

Almost every strip about the pointy haired boss is an exact match on the boss, where I work! It is so scary

[bridgman]

I keep a copy of Dogbert's Guide to Management on my desk just in case.

[Louise]

I keep a copy of Dogbert's Guide to Management on my desk just in case.

So it IS true. All managers are evil.

Evil doesn't come by its self. It have to be managed. -Dogbert

Edit: does -> doesn't

[kernelOfTruth]

I'm using the modified ebuild for x11-base/x11-drm

anyone getting the same error message ?

Switched to a new branch 'branch-r6xx-r7xx-3d'
>>> Unpacked to /var/tmp/portage/x11-base/x11-drm-99999999/work/drm
>>> Source unpacked in /var/tmp/portage/x11-base/x11-drm-99999999/work
>>> Preparing source in /var/tmp/portage/x11-base/x11-drm-99999999/work/drm ...
* Converting drm/linux-core/Makefile to use M= instead of SUBDIRS= ... [ ok ]
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/x11-base/x11-drm-99999999/work/drm ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/x11-base/x11-drm-99999999/work/drm ...
* Building DRM in /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core...
ln: creating symbolic link `Module.symvers': File exists
* Preparing drm module
make -j5 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= DRM_MODULES=radeon.ko LINUXDIR=/usr/src/linux M=/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core modules
sh ../scripts/create_linux_pci_lists.sh < ../shared-core/drm_pciids.txt
make -C /usr/src/linux M=`/bin/pwd` DRMSRCDIR=`/bin/pwd` modules
make[1]: Entering directory `/usr/src/linux-2.6.30-zen3'
/usr/src/linux-2.6.30-zen3/arch/x86/Makefile:82: stack protector enabled but no compiler support
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_auth.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_bufs.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_dma.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_drawable.o
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_bufs.c: In function ‘drm_rmmap_locked’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_bufs.c:402: warning: enumeration value ‘_DRM_GEM’ not handled in switch
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.c: In function ‘drm_context_switch’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.c:234: warning: pointer targets in passing argument 2 of ‘test_and_set_bit’ differ in signedness
/usr/src/linux-2.6.30-zen3/arch/x86/include/asm/bitops.h:195: note: expected ‘volatile long unsigned int *’ but argument is of type ‘volatile long int *’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.c:242: warning: pointer targets in passing argument 2 of ‘clear_bit’ differ in signedness
/usr/src/linux-2.6.30-zen3/arch/x86/include/asm/bitops.h:98: note: expected ‘volatile long unsigned int *’ but argument is of type ‘volatile long int *’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.c: In function ‘drm_context_switch_complete’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_context.c:272: warning: pointer targets in passing argument 2 of ‘clear_bit’ differ in signedness
/usr/src/linux-2.6.30-zen3/arch/x86/include/asm/bitops.h:98: note: expected ‘volatile long unsigned int *’ but argument is of type ‘volatile long int *’
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_drv.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_fops.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_ioctl.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_irq.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_lock.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_memory.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_stub.o
CC [M] /var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.o
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_memory.c: In function ‘agp_remap’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_memory.c:287: error: ‘struct agp_memory’ has no member named ‘memory’
make[2]: *** [/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_memory.o] Error 1
make[2]: *** Waiting for unfinished jobs....
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.c: In function ‘drm_gem_one_name_info’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.c:602: warning: format ‘%d’ expects type ‘int’, but argument 3 has type ‘size_t’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.c:610: warning: format ‘%9d’ expects type ‘int’, but argument 4 has type ‘size_t’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.c: In function ‘drm__vma_info’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_proc.c:691: warning: format ‘%08lx’ expects type ‘long unsigned int’, but argument 5 has type ‘phys_addr_t’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_stub.c: In function ‘__check_debug’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_stub.c:48: warning: pointer targets in return differ in signedness
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.c: In function ‘drm_do_vm_fault’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.c:154: error: ‘struct agp_memory’ has no member named ‘memory’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.c:161: error: ‘struct agp_memory’ has no member named ‘memory’
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.c: In function ‘drm_vm_shm_close’:
/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.c:261: warning: enumeration value ‘_DRM_GEM’ not handled in switch
make[2]: *** [/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core/drm_vm.o] Error 1
make[1]: *** [_module_/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.30-zen3'
make: *** [modules] Error 2
*
* ERROR: x11-base/x11-drm-99999999 failed.
* Call stack:
* ebuild.sh, line 49: Called src_compile
* environment, line 3685: Called src_compile_linux
* environment, line 3714: Called linux-mod_src_compile
* environment, line 2891: Called die
* The specific snippet of code:
* eval "emake HOSTCC=\"$(tc-getBUILD_CC)\" CROSS_COMPILE=${CHOST}- LDFLAGS=\"$(get_abi_LDFLAGS)\" ${BUILD_FIXES} ${BUILD_PARAMS} ${BUILD_TARGETS} " || die "Unable to emake HOSTCC="$(tc-getBUILD_CC)" CROSS_COMPILE=${CHOST}- LDFLAGS="$(get_abi_LDFLAGS)" ${BUILD_FIXES} ${BUILD_PARAMS} ${BUILD_TARGETS}";
* The die message:
* Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= DRM_MODULES='radeon.ko' LINUXDIR='/usr/src/linux' M='/var/tmp/portage/x11-base/x11-drm-99999999/work/drm/linux-core' modules
*
* If you need support, post the topmost build error, and the call stack if relevant.
* A complete build log is located at '/var/log/portage/x11-base:x11-drm-99999999:20090806-213423.log'.
* The ebuild environment file is located at '/var/tmp/portage/x11-base/x11-drm-99999999/temp/environment'.
* This ebuild used the following eclasses from overlays:
* /usr/gentoo/overlays/x11-overlay/eclass/git.eclass
* This ebuild is from an overlay named 'x11-overlay': '/usr/gentoo/overlays/x11-overlay/'

kernel is 2.6.30-zen3

[agd5f]

I'm using the modified ebuild for x11-base/x11-drm

anyone getting the same error message ?

kernel is 2.6.30-zen3

Might be specific to the zen kernel.

[nanonyme]

That should be fixed in git. Also iirc that API change isn't supposed to happen in 2.6.30 but 2.6.31... (so yes, very likely specific to zen kernel; see if the bug persists on a stock 2.6.30)

[kernelOfTruth]

Might be specific to the zen kernel.

thanks agd5f and nanonyme

it most probably was a problem with zen3 - I read in fgo (forums.gentoo.org) that some parts of drm/graphics related stuff seem to be missing in that release,

now I'm running 2.6.31-rc5-zen0 and it compiled fine

compositing and/or GL is now working with compiz (indirect-rendering) and kwin (opengl backend)

with compiz the screen flashes during each refreshing of the screen it seems

with kwin (4.3) the performance seems to be somewhat better, after some time it however pauses/suspends compositing because it's too slow

great work guys !

the drivers are really getting into a feature-rich and productive state with big steps

thanks a lot

my next graphics card (most probably) also will be an AMD/ATI

[descendent87]

Using libdrm-git, mesa-git, glproto-git, dri2proto-git, libgl-git, ati-dri-r6xx-r7xx, xf86-video-ati-git (all from AUR on Arch) I now have desktop effects working on KDE (with a lot of artifacts) and glxgears is getting about 500fps on my HD3200

[MùPùF]

Using libdrm-git, mesa-git, glproto-git, dri2proto-git, libgl-git, ati-dri-r6xx-r7xx, xf86-video-ati-git (all from AUR on Arch) I now have desktop effects working on KDE (with a lot of artifacts) and glxgears is getting about 500fps on my HD3200

Same here, I'll upload a video tonight

PS: to make my system usable, I removed the drm package from the kernel. I just have minor screen corruption now and it works quite fast (gtkperf is under 4 seconds!) but, of course, no 3D support

[Perry3D]

Same here, I'll upload a video tonight

PS: to make my system usable, I removed the drm package from the kernel. I just have minor screen corruption now and it works quite fast (gtkperf is under 4 seconds!) but, of course, no 3D support

Do you also use Arch? I thought, that the drm-modules from the kernel are overwritten by the drm-modules-r6xx-r7xx-3d package. There should be no difference if you remove the original kernel modules.