cap_t has been removed from include/linux/capability.h and ati-driver uses it "vi /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c" 1886 KCL_TYPE_Cap ATI_API_CALL KCL_PosixSecurityCapGetEffectiveVector() 1887 { 1888 return cap_t(current->cap_effective); 1889 } 1890 1891 /** \brief Set vector of effective security caps for the current process 1892 * \param cap OS dependent vector of effective security caps 1893 */ 1894 void ATI_API_CALL KCL_PosixSecurityCapSetEffectiveVector(KCL_TYPE_Cap cap) 1895 { 1896 cap_t(current->cap_effective) = cap; 1897 } Reproducible: Always Steps to Reproduce: 1. emerge ati-drivers Actual Results: * Messages for package x11-drivers/ati-drivers-8.471.3: * * ERROR: x11-drivers/ati-drivers-8.471.3 failed. * Call stack: * ebuild.sh, line 49: Called src_compile * environment, line 3204: Called linux-mod_src_compile * environment, line 2422: Called die * The specific snippet of code: * eval "emake HOSTCC=\"$(tc-getBUILD_CC)\" CC=\"$(get-KERNEL_CC)\" LDFLAGS=\"$(get_abi_LDFLAGS)\" ${BUILD_FIXES} ${BUILD_PARAMS} ${BUILD_TARGETS} " || die "Unable to emake HOSTCC="$(tc-getBUILD_CC)" CC="$(get-KERNEL_CC)" LDFLAGS="$(get_abi_LDFLAGS)" ${BUILD_FIXES} ${BUILD_PARAMS} ${BUILD_TARGETS}"; * The die message: * Unable to emake HOSTCC=i686-pc-linux-gnu-gcc CC=i686-pc-linux-gnu-gcc LDFLAGS= GCC_VER_MAJ=4 KVER=2.6.25-gentoo-r1 KDIR=/usr/src/linux kmod_build * * If you need support, post the topmost build error, and the call stack if relevant. * A complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-8.471.3/temp/build.log'. * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-8.471.3/temp/environment'. * * Preparing fglrx module make -C /usr/src/linux M=/var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x modules make[1]: Entering directory `/usr/src/linux-2.6.25-gentoo-r1' CC [M] /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function 'KCL_PosixSecurityCapGetEffectiveVector': /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:1888: error: implicit declaration of function 'cap_t' /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function 'KCL_PosixSecurityCapSetEffectiveVector': /var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:1896: error: invalid lvalue in assignment make[2]: *** [/var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o] Error 1 make[1]: *** [_module_/var/tmp/portage/x11-drivers/ati-drivers-8.471.3/work/common/lib/modules/fglrx/build_mod/2.6.x] Error 2 make[1]: Leaving directory `/usr/src/linux-2.6.25-gentoo-r1' make: *** [kmod_build] Error 2 * * ERROR: x11-drivers/ati-drivers-8.471.3 failed.
Hope you're ready to help fixing the drivers :)
the cap stuff is easy to fix, the tlb and memory management stuff got unexposed and require upstream to take care.
Luca, any chance you could add a kernel version check to the ebuild so that it deliberately bails out on 2.6.25? Or if it works on some configurations, perhaps just an ewarn would be useful. Thanks.
applying the two patches found on google for this problem, and then altering the kernel sources via the method in the Gentoo ATI Drivers HOWTO page will make this compile correctly. I do not know if this can be accommodated via portage as it would require patches for both the kernel and the drivers.
Created attachment 150533 [details, diff] working patch
The attached file is the patch to fix the original problem the compilation fails with, and then the process listed at http://gentoo-wiki.com/Ati#Build_ati-drivers_on_rt-kernels_failed will fix the subsequent problem.
Woah, that page suggests to remove GPL-only annotations! That's bloody illegal. Greg KH would probably kill you with his own hands for that.
i don't know about the legality of it, as i'm strictly an end-user. all i know is it works.
Oh yeah, I forgot that the GPL and copyright law doesn't apply to end users.
For the record, I would just like to point out that a dependency conflict now exists between x11-drivers/ati-drivers and sys-kernel/gentoo-sources or sys-kernel/vanilla-sources - both kernel ebuilds have now got 2.5.25 releases tagged as stable (at least on x86 and amd64). Given that the problem seems to have already been resolved in more recent upstream packages, it seems a good idea to resolve this conflict by selecting one of such more recent packages and marking it as stable.
I agree. It would be nice having my X server running without any hack... (In reply to comment #10) > For the record, I would just like to point out that a dependency conflict now > exists between x11-drivers/ati-drivers and sys-kernel/gentoo-sources or > sys-kernel/vanilla-sources - both kernel ebuilds have now got 2.5.25 releases > tagged as stable (at least on x86 and amd64). Given that the problem seems to > have already been resolved in more recent upstream packages, it seems a good > idea to resolve this conflict by selecting one of such more recent packages and > marking it as stable. >
