Hi guys, I found out a patch at phoronix (http://www.phoronix.com/forums/showthread.php?t=12546) that enables installation of ati-drivers-8.532 on 2.6.27*. It is simple and works quite well at home (core2quad, Radeon4870, compiz w/ AIGLX). I'm not sure if such a bug entry for a not-yet released kernel is acceptable for you. Feel free to close if it's not. Reproducible: Always Steps to Reproduce: 1.Emerge =sys-kernel/git-sources-2.6.27* 2.Emerge =x11-ati-drivers/ati-drivers-8.532 3. Actual Results: Emerge fails. Expected Results: Emerge succeeds.
Created attachment 166904 [details, diff] Patch for 8.532 to compile against 2.6.27
please, how to modify ebuild to use this patch? thanks
I would edit the patch too: --- common/lib/modules/fglrx/build_mod/firegl_public.c.orig 2008-09-19 10:57:09.000000000 +0200 +++ common/lib/modules/fglrx/build_mod/firegl_public.c 2008-09-19 10:55:57.000000000 +0200 instead of: --- firegl_public.c.orig 2008-09-19 10:57:09.000000000 +0200 +++ firegl_public.c 2008-09-19 10:55:57.000000000 +0200 and the ebuild something like this: if kernel_is 2 6 27; then epatch "${FILESDIR}"/ati-drivers-2.6.27.patch fi (ebuild part untested)
*** Bug 241872 has been marked as a duplicate of this bug. ***
Looks like 8.10 is out and fixes the 2.6.27 compilation issues, see the driver release page (for amd64): http://ati.amd.com/support/drivers/linux64/linux64-radeon.html Given this, is the patch still necessary?
(In reply to comment #5) > Looks like 8.10 is out and fixes the 2.6.27 compilation issues, see the driver > release page (for amd64): > > http://ati.amd.com/support/drivers/linux64/linux64-radeon.html > > Given this, is the patch still necessary? Also, I tried bumping the driver to the newest version in my local overlay, and it installs & runs just fine.No patches needed except for what's already in FILESDIR. So except for xorg-server 1.5 support, it seems ATI has done a decent job with fixing 8.10 so that it works with .27 kernels.
8.542 is now in portage. Enjoy!
Nop, the patch is still necessary for 2.6.27, otherwise I get : /var/tmp/paludis/x11-drivers-ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:2352: error: too many arguments to function 'on_each_cpu' Notice that it is the 8.543 driver (that only canonical owns at this time) that is compatible with kernel-2.6.27/xorg-1.5. Please reopen the issue.
(In reply to comment #8) > Nop, the patch is still necessary for 2.6.27, otherwise I get : Same here, but with portage (warnings snipped): /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function 'KCL_flush_tlb_onepage': /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:2352: error: too many arguments to function 'on_each_cpu' /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function '__ke_flush_cache': /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:2791: error: too many arguments to function 'smp_call_function' /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function 'KCL_enable_pat': /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:4063: error: too many arguments to function 'smp_call_function' /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function 'KCL_disable_pat': /var/tmp/portage/x11-drivers/ati-drivers-8.542/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:4082: error: too many arguments to function 'smp_call_function'
Created attachment 168786 [details] ati-drivers-8.542.ebuild ati-drivers-8.542.ebuild
Created attachment 168788 [details, diff] Patch for 8.542 to compile against 2.6.27
I just updated the patch and submitted ebuild for 8.542. Thanx volker for your comments. The ebuild diff, compared to the official ebuild is : @@ -167,6 +169,10 @@ ln -s "${ARCH_DIR}"/lib/modules/fglrx/build_mod/libfglrx_ip.a.GCC$(gcc-major-version) \ || die "symlinking precompiled core failed" + if kernel_is 2 6 27; then + epatch "${FILESDIR}/ati-drivers-2.6.27.patch" + fi + convert_to_m 2.6.x/Makefile || die "convert_to_m failed"
Could someone with the right permission reopen this bug please ?
It seems I can do it myself :-) Reopened.
Just committed to CVS. Thanks everyone!
Now in portage tree. Closing the bug.
Created attachment 176407 [details, diff] Patch for the ebuild to support versions > 2.6.27 (i.e. 2.6.28)
Is there any reason this patch should not be applied to previous versions of ati-drivers? All versions of ati-drivers > 8.532 seem to have removed support for mobility radeon 9600 M10(specifically chip id 0x4e50), so all versions of ati-drivers newer than 8.542 will not work for me(compiles, but will not load). This patch is not applied to versions < 8.542 in portage, which makes installation very difficult for anyone with a radeon with chipid 0x4e50 using kernel 2.6.27.(had the same problem in 2.6.26 as well)
(In reply to comment #18) > "ati-drivers newer than 8.542 will not work for me" What I meant to say was "ati-drivers 8.542 or newer will not work for me"