Bug 336467 – x11-drivers/ati-drivers-10.[89]: doesn't compile with 2.6.36-rc3 or 2.6.35-r8

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Vasco Gervasi 2010-09-08 17:52:52 UTC]

Hi all,
I am using open driver now: mesa-9999, xf86-video-ati-9999, xorg-server-1.8.2 gallium software render, vanilla-sources-2.6.36_rc3.
But I want to try ati-drivers to full use my ati radeon HD5770; so I added fglrx to: VIDEO_CARDS="fglrx radeon"
But when it emerges ati-drivers, it gives me this error:


>>> Emerging (1 of 6) x11-drivers/ati-drivers-10.8
 * ati-driver-installer-10-8-x86.x86_64.run RMD160 SHA1 SHA256 size ;-) ...  [ ok ]
 * checking ebuild checksums ;-) ...                                     [ ok ]
 * checking auxfile checksums ;-) ...                                    [ ok ]
 * checking miscfile checksums ;-) ...                                   [ ok ]
 * CPV:  x11-drivers/ati-drivers-10.8
 * REPO: gentoo
 * Maintainer: [e-mail] [e-mail]
 * USE:  amd64 elibc_glibc kernel_linux modules userland_GNU
 * Determining the location of the kernel source code
 * Found kernel source directory:
 *     /usr/src/linux
 * Found kernel object directory:
 *     /lib/modules/2.6.36-rc3/build
 * Found sources for kernel version:
 *     2.6.36-rc3
 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * 
>>> Unpacking source...
>>> Source unpacked in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work
>>> Preparing source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Applying ati-drivers-fix_compilation-bug-297322.patch ...                                                                                                                                            [ ok ]
 * Applying ati-powermode-opt-path-2.patch ...                                                                                                                                                          [ ok ]
 * Converting 2.6.x/Makefile to use M= instead of SUBDIRS= ...                                                                                                                                          [ ok ]
>>> Unpacking ./../common/usr/src/ati/fglrx_sample_source.tgz to /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/extra
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Preparing fglrx module
make -j3 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build 
make -C /usr/src/linux M=/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x modules
make[1]: Entering directory `/usr/src/linux-2.6.36-rc3'
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_acpi.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_agp.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:410:5: error: unknown field ‘ioctl’ specified in initializer
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:410:5: warning: initialization from incompatible pointer type
In file included from /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:451:0:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h: In function ‘FGLDRM__vma_info’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h:497:2: warning: format ‘%08lx’ expects type ‘long unsigned int’, but argument 5 has type ‘phys_addr_t’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function ‘KCL_MapPageToPfn’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:1638:16: warning: unused variable ‘bus_addr’
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.o
make[2]: *** [/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o] Error 1
make[2]: *** Waiting for unfinished jobs....
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: its scope is only this definition or declaration, which is probably not what you want
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:60:49: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:67:5: warning: initialization from incompatible pointer type
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:74:5: warning: initialization from incompatible pointer type
make[1]: *** [_module_/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.36-rc3'
make: *** [kmod_build] Error 2
emake failed
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3698:  Called linux-mod_src_compile
 *   environment, line 2862:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

>>> Failed to emerge x11-drivers/ati-drivers-10.8, Log file:

>>>  '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'

 * Messages for package x11-drivers/ati-drivers-10.8:

 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3698:  Called linux-mod_src_compile
 *   environment, line 2862:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

Thanks

Reproducible: Always

Steps to Reproduce:
1. emerge -a ati-drivers
2.
3.



Expected Results:  
Fglrx should work

[Comment 1 Vasco Gervasi 2010-09-08 17:55:38 UTC]

Here seems to be a patch:

http://www.phoronix.com/forums/showthread.php?t=25485

[Comment 2 Ludovic Magerand 2010-09-12 01:02:39 UTC]

This bug is confirmed here, I got the same result with a similar configuration.

I quickly created an ebuild in my local overlay using the second version of the patch cited above, and it's working fine.

I think I badly added the patch in the ebuild as I'm not an ebuild expert, but the result is attached together with the patch files to be added in the 'files' directory.

If a dev can make a revision of the ebuild, I would test it before inclusion in the tree if needed.

[Comment 3 Ludovic Magerand 2010-09-12 01:04:29 UTC]

Created attachment 246923 [details]
The ebuild

[Comment 4 Ludovic Magerand 2010-09-12 01:04:49 UTC]

Created attachment 246924 [details]
The patch

[Comment 5 Vasco Gervasi 2010-09-12 07:56:52 UTC]

Using these patch and ebuild, it doesn't work:

* ati-driver-installer-10-8-x86.x86_64.run RMD160 SHA1 SHA256 size ;-) ...  [ ok ]
 * checking ebuild checksums ;-) ...                                     [ ok ]
 * checking auxfile checksums ;-) ...                                    [ ok ]
 * checking miscfile checksums ;-) ...                                   [ ok ]
 * CPV:  x11-drivers/ati-drivers-10.8
 * REPO: gentoo
 * Maintainer: [e-mail] [e-mail]
 * USE:  amd64 elibc_glibc kernel_linux modules userland_GNU
 * Determining the location of the kernel source code
 * Found kernel source directory:
 *     /usr/src/linux
 * Found kernel object directory:
 *     /lib/modules/2.6.36-rc3/build
 * Found sources for kernel version:
 *     2.6.36-rc3
 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * 
>>> Unpacking source...
>>> Source unpacked in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work
>>> Preparing source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Applying ati-drivers-2.6.36.patch ...                                                                                                                                                                [ ok ]
 * Applying ati-drivers-fix_compilation-bug-297322.patch ...                                                                                                                                            [ ok ]
 * Applying ati-powermode-opt-path-2.patch ...                                                                                                                                                          [ ok ]
 * Converting 2.6.x/Makefile to use M= instead of SUBDIRS= ...                                                                                                                                          [ ok ]
>>> Unpacking ./../common/usr/src/ati/fglrx_sample_source.tgz to /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/extra
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Preparing fglrx module
make -j3 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build 
make -C /usr/src/linux M=/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x modules
make[1]: Entering directory `/usr/src/linux-2.6.36-rc3'
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_acpi.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_agp.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:415:5: warning: initialization from incompatible pointer type
In file included from /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:459:0:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h: In function ‘FGLDRM__vma_info’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h:497:2: warning: format ‘%08lx’ expects type ‘long unsigned int’, but argument 5 has type ‘phys_addr_t’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function ‘KCL_MapPageToPfn’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:1646:16: warning: unused variable ‘bus_addr’
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: its scope is only this definition or declaration, which is probably not what you want
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:60:49: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:67:5: warning: initialization from incompatible pointer type
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:74:5: warning: initialization from incompatible pointer type
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_io.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_RegisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:5: error: implicit declaration of function ‘register_ioctl32_conversion’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:46: error: ‘ioctl_trans_handler_t’ undeclared (first use in this function)
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:46: note: each undeclared identifier is reported only once for each function it appears in
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:68: error: expected ‘)’ before ‘handler’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_UnregisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:186:5: error: implicit declaration of function ‘unregister_ioctl32_conversion’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_RegisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:176:1: warning: control reaches end of non-void function
make[2]: *** [/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [_module_/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.36-rc3'
make: *** [kmod_build] Error 2
emake failed
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3703:  Called linux-mod_src_compile
 *   environment, line 2867:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

>>> Failed to emerge x11-drivers/ati-drivers-10.8, Log file:

>>>  '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'

 * Messages for package x11-drivers/ati-drivers-10.8:

 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3703:  Called linux-mod_src_compile
 *   environment, line 2867:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

[Comment 6 Ludovic Magerand 2010-09-12 09:42:53 UTC]

I guess your system is 64bits ?
It looks like the error has something to do with 32bits code being compiled as 64bits, or the other way around, and as you don't use the 'multilib' use flag, this can't work.
Can you try it with use flag 'multilib' enable, just to check if this is really something related to that ?

[Comment 7 Vasco Gervasi 2010-09-12 10:09:29 UTC]

Yes I am using ~amd64 with no-multilib, how can I enable multilib use flag for ati-drivers? I have tried with USE="multilib" emerge -a ati-drivers but:
[ebuild  N    ] x11-drivers/ati-drivers-10.8  USE="modules -debug (-multilib) -qt4" 0 kB

so multilib use isn't enable

[Comment 8 Vasco Gervasi 2010-09-12 10:12:29 UTC]

Momentaly I have change profile

[Comment 9 Vasco Gervasi 2010-09-12 10:14:52 UTC]

Enabling multilib use to ati-drivers same result:

>>> Emerging (4 of 5) x11-drivers/ati-drivers-10.8
 * ati-driver-installer-10-8-x86.x86_64.run RMD160 SHA1 SHA256 size ;-) ...  [ ok ]
 * checking ebuild checksums ;-) ...                                     [ ok ]
 * checking auxfile checksums ;-) ...                                    [ ok ]
 * checking miscfile checksums ;-) ...                                   [ ok ]
 * CPV:  x11-drivers/ati-drivers-10.8
 * REPO: gentoo
 * Maintainer: [e-mail] [e-mail]
 * USE:  amd64 elibc_glibc kernel_linux modules multilib userland_GNU
 * Determining the location of the kernel source code
 * Found kernel source directory:
 *     /usr/src/linux
 * Found kernel object directory:
 *     /lib/modules/2.6.36-rc3/build
 * Found sources for kernel version:
 *     2.6.36-rc3
 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * 
>>> Unpacking source...
>>> Source unpacked in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work
>>> Preparing source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Applying ati-drivers-2.6.36.patch ...                                 [ ok ]
 * Applying ati-drivers-fix_compilation-bug-297322.patch ...             [ ok ]
 * Applying ati-powermode-opt-path-2.patch ...                           [ ok ]
 * Converting 2.6.x/Makefile to use M= instead of SUBDIRS= ...           [ ok ]
>>> Unpacking ./../common/usr/src/ati/fglrx_sample_source.tgz to /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/extra
>>> Source prepared.
>>> Configuring source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
>>> Source configured.
>>> Compiling source in /var/tmp/portage/x11-drivers/ati-drivers-10.8/work ...
 * Preparing fglrx module
make -j3 HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS= GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build 
make -C /usr/src/linux M=/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x modules
make[1]: Entering directory `/usr/src/linux-2.6.36-rc3'
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_acpi.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_agp.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:415:5: warning: initialization from incompatible pointer type
In file included from /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:459:0:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h: In function ‘FGLDRM__vma_info’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/drm_proc.h:497:2: warning: format ‘%08lx’ expects type ‘long unsigned int’, but argument 5 has type ‘phys_addr_t’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c: In function ‘KCL_MapPageToPfn’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/firegl_public.c:1646:16: warning: unused variable ‘bus_addr’
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:51:44: warning: its scope is only this definition or declaration, which is probably not what you want
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:60:49: warning: ‘struct tty_struct’ declared inside parameter list
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:67:5: warning: initialization from incompatible pointer type
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_debug.c:74:5: warning: initialization from incompatible pointer type
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.o
  CC [M]  /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_io.o
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_RegisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:5: error: implicit declaration of function ‘register_ioctl32_conversion’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:46: error: ‘ioctl_trans_handler_t’ undeclared (first use in this function)
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:46: note: each undeclared identifier is reported only once for each function it appears in
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:174:68: error: expected ‘)’ before ‘handler’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_UnregisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:186:5: error: implicit declaration of function ‘unregister_ioctl32_conversion’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c: In function ‘KCL_IOCTL_RegisterConversion32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:176:1: warning: control reaches end of non-void function
make[2]: *** [/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [_module_/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.36-rc3'
make: *** [kmod_build] Error 2
emake failed
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3703:  Called linux-mod_src_compile
 *   environment, line 2867:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

>>> Failed to emerge x11-drivers/ati-drivers-10.8, Log file:

>>>  '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'

 * Messages for package x11-drivers/ati-drivers-10.8:

 * You have DRM support built in to the kernel
 * Direct rendering will not work.
 * 
 * Please note that this driver supports only graphic cards based on
 * r600 chipset and newer.
 * This represent the ATI Radeon HD series at this moment.
 * 
 * If your card is older then usage of x11-drivers/xf86-video-ati
 * as replacement is highly recommended. Rather than staying with
 * old versions of this driver.
 * For migration informations please reffer to:
 * http://www.gentoo.org/proj/en/desktop/x/x11/ati-migration-guide.xml
 * ERROR: x11-drivers/ati-drivers-10.8 failed:
 *   Unable to emake HOSTCC=x86_64-pc-linux-gnu-gcc CROSS_COMPILE=x86_64-pc-linux-gnu- LDFLAGS=  GCC_VER_MAJ=4 KVER=2.6.36-rc3 KDIR=/usr/src/linux kmod_build
 * 
 * Call stack:
 *     ebuild.sh, line   56:  Called src_compile
 *   environment, line 3703:  Called linux-mod_src_compile
 *   environment, line 2867:  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}";
 * 
 * If you need support, post the output of 'emerge --info =x11-drivers/ati-drivers-10.8',
 * the complete build log and the output of 'emerge -pqv =x11-drivers/ati-drivers-10.8'.
 * The complete build log is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/build.log'.
 * The ebuild environment file is located at '/var/tmp/portage/x11-drivers/ati-drivers-10.8/temp/environment'.
 * S: '/var/tmp/portage/x11-drivers/ati-drivers-10.8/work'

[Comment 10 Ludovic Magerand 2010-09-12 16:47:36 UTC]

This new error is somewhat unrelated from the original bug and is a another bug in my point of view.
This new error has to do with 2 undeclared symbols which are normally defined in the kernel contrary to the original bug which is about a badly defined function inside the driver code. 
So this new bug comes probably from a missing option in the kernel configuration, and seems related to 32/64 bits code compatibility (I can compile the ebuild here, but have both the kernel and user space configured to work with 32 or 64 bits code).
But it's all my guess because this is all far away from my area of expertise.

[Comment 11 Vasco Gervasi 2010-09-12 16:57:44 UTC]

Thanks.

[Comment 12 Kiichiro 2010-09-18 13:47:12 UTC]

Created attachment 247851 [details]
The patch

[Comment 13 Enrico Tagliavini 2010-09-21 15:47:03 UTC]

Created attachment 248274 [details, diff]
first patch to make ati-drivers 10.9 working with kernel .36-rc*

[Comment 14 Enrico Tagliavini 2010-09-21 15:47:42 UTC]

Created attachment 248275 [details, diff]
second patch to make ati-drivers 10.9 working with kernel >=.36-rc4-git2

[Comment 15 Enrico Tagliavini 2010-09-21 15:50:27 UTC]

Created attachment 248277 [details, diff]
the ebuild that uses the previous patches

with this ebuild and the 2 patches i'm able to run ati-drivers-10.9 on kernel sys-kernel/git-sources-2.6.36_rc4-r5 (and should work with every rc, it compiles against sys-kernel/vanilla-sources-2.6.36_rc4)

This ebuild is just a mod of the one present in the main portage tree (just added)

[Comment 16 Chí-Thanh Christopher Nguyễn 2010-09-22 11:37:06 UTC]

*** Bug 338325 has been marked as a duplicate of this bug. ***

[Comment 17 Chí-Thanh Christopher Nguyễn 2010-09-22 11:37:14 UTC]

*** Bug 338336 has been marked as a duplicate of this bug. ***

[Comment 18 Enrico Tagliavini 2010-09-22 15:25:41 UTC]

Created attachment 248366 [details]
new ebuild patch to fix kernel .35 too

kernel .35.5 (and so also gentoo-sources .35-r8 and newer) needs the patch too. this patch applies to the 10.9 current tree ebuild. be aware of the possible security hole, it is possible to stay with an older kernel like .35-r7 or .34..... but they are broken too i guess.

[Comment 19 Enrico Tagliavini 2010-09-22 15:28:36 UTC]

I forgotten to say that this patch might not work with all kernels (eg tuxonice-sources and other) i have done this for vanilla, gentoo and git sources for now (hardened still doesn't need this. if you can't use this patch with something else, send me the build log and the KV_EXTRA value so i will might be able to fix it

[Comment 20 Enrico Tagliavini 2010-09-22 16:08:31 UTC]

I add for convenience more info about the exploit. https://bugzilla.redhat.com/show_bug.cgi?id=634457 as you can see is not good. i'm working on a patch to fix this, but i have to use GPL code, so i have to ask some license expert some help understanding if this work is legal or not

[Comment 21 Ming-Wei 2010-09-22 22:05:46 UTC]

(In reply to comment #20)
> I add for convenience more info about the exploit.
> https://bugzilla.redhat.com/show_bug.cgi?id=634457 as you can see is not good.
> i'm working on a patch to fix this, but i have to use GPL code, so i have to
> ask some license expert some help understanding if this work is legal or not


Here is the fix from Ubuntu
http://launchpadlibrarian.net/56199871/compat_alloc_cve-2010-3081.patch 

[Comment 22 Thomas Raschbacher 2010-09-23 05:37:26 UTC]

the unbuntu patch seems to build fine here ;) although i do not know enough about the ati-drivers ebuild to fix it with that patch ;) (the patch uses the unbuntu make.sh which sets an environment var which is detected from the makefile

[Comment 23 Thomas Raschbacher 2010-09-23 06:40:16 UTC]

Made a patch for the ebuild (and changed the ubuntu patch) to build, load and run ;)

[Comment 24 Thomas Raschbacher 2010-09-23 06:43:57 UTC]

Created attachment 248420 [details, diff]
ati-drivers-10.9.ebuild.patch

[Comment 25 Thomas Raschbacher 2010-09-23 06:44:18 UTC]

Created attachment 248421 [details, diff]
ati-drivers-compat_alloc_cvs-2010-3081.patch

[Comment 26 Thomas Raschbacher 2010-09-23 06:48:48 UTC]

Thought about putting it in my overlay when i realised that i don'T have the required ssh keys on my laptop but prolly better if it goes in live tree asap anyway. if there are some concerns about live tree let me know i'll stick it in my dev overlay later so it is easier for ppl to install it

[Comment 27 Helmut Jarausch 2010-09-23 07:42:27 UTC]

(In reply to comment #26)
> Thought about putting it in my overlay when i realised that i don'T have the
> required ssh keys on my laptop but prolly better if it goes in live tree asap
> anyway. if there are some concerns about live tree let me know i'll stick it in
> my dev overlay later so it is easier for ppl to install it
> 

Thanks! Using your two patches works just fine here (2.6.35-gentoo-r8 + ati-drivers-10.9)
Helmut.

[Comment 28 Markus Peloquin 2010-09-23 08:02:40 UTC]

I just used the patches from Thomas Raschbacher and everything went swimmingly.  Thanks!

[Comment 29 Ming-Wei 2010-09-23 09:06:18 UTC]

(In reply to comment #26)
> Thought about putting it in my overlay when i realised that i don'T have the
> required ssh keys on my laptop but prolly better if it goes in live tree asap
> anyway. if there are some concerns about live tree let me know i'll stick it in
> my dev overlay later so it is easier for ppl to install it


Thanks, I was going to look at the patch and make an ebuild today, but you beat me to it :D 

Compiles file, going to reboot. If I don't report back then it means it also runs fine :D

[Comment 30 Thomas Raschbacher 2010-09-23 09:28:50 UTC]

(In reply to comment #29)
> (In reply to comment #26)
> > Thought about putting it in my overlay when i realised that i don'T have the
> > required ssh keys on my laptop but prolly better if it goes in live tree asap
> > anyway. if there are some concerns about live tree let me know i'll stick it in
> > my dev overlay later so it is easier for ppl to install it
> 
> 
> Thanks, I was going to look at the patch and make an ebuild today, but you beat
> me to it :D 
> 
> Compiles file, going to reboot. If I don't report back then it means it also
> runs fine :D
> 
should ;) btw u can save yourself a reboot just quit X (don'T forget to run /etc/init.d/xdm stop as well) and then unload the old module.
so in total it'd be the following (log out of your X-session first) on a root-terminal:

/etc/init.d/xdm stop
rmmod fglrx
modprobe fglrx
/etc/init.d/xdm start

[Comment 31 Ming-Wei 2010-09-25 11:50:26 UTC]

> should ;) btw u can save yourself a reboot just quit X (don'T forget to run
> /etc/init.d/xdm stop as well) and then unload the old module.
> so in total it'd be the following (log out of your X-session first) on a
> root-terminal:
> 
> /etc/init.d/xdm stop
> rmmod fglrx
> modprobe fglrx
> /etc/init.d/xdm start

thanks Thomas, everything works fine. Well I had to reboot, I was using 2.6.35.4 and now 2.6.35.5 :D

[Comment 32 Thomas Raschbacher 2010-09-28 08:14:06 UTC]

@luca: do you plan to put a fix into the official tree soon? if not i'll stick it into my overlay since i'm too lazy to just copy stuff to all my machines with ati cards ;)

[Comment 33 Thomas Raschbacher 2010-09-28 09:59:15 UTC]

@lu_zero:
err just noticed that you did commit 10.9-r1 already with the patch. i assume you just forgot to close this then ;)

[Comment 34 Tomáš Chvátal (RETIRED) 2010-10-02 15:40:24 UTC]

*** Bug 339365 has been marked as a duplicate of this bug. ***

[Comment 35 Enrico Tagliavini 2010-10-10 00:24:58 UTC]

*** Bug 340259 has been marked as a duplicate of this bug. ***

[Comment 36 Enrico Tagliavini 2010-10-13 17:36:53 UTC]

Ok 15 days without activity i think this can be closed :)