request: patch for catalyst (10.7) to get 2.6.36-rc* working

**kernelOfTruth** · 08-20-2010, 02:59 AM

Hi guys,

anyone has already a patch to get the fglrx / catalyst drivers to work with 2.6.36-rc* kernels ?

**nanonyme** · 08-20-2010, 06:53 AM

Come on now, why do you even want to run Catalyst on such experimental kernels?

**kernelOfTruth** · 08-20-2010, 08:19 AM

Originally Posted by nanonyme

Come on now, why do you even want to run Catalyst on such experimental kernels?

it has been since the beginning for me that way - those bleeding edge kernels were the only possibility to run my hardware at that time (6 years ago) - and I'm kind of keeping that tradition

besides that the most severe problems don't affect me / have already been fixed

**Spasswolf** · 08-21-2010, 08:14 AM

There you are, I'm currently running the opencl 1.1 testing only driver (ati-driver-installer-8.753.1-x86.x86_64.run) with kernel 2.6.36-rc1, I didn't do much testing yet but torcs and some opencl examples from the sdk seen to work fine:

Code:

diff -aur work.orig/common/lib/modules/fglrx/build_mod/firegl_public.c work/common/lib/modules/fglrx/build_mod/firegl_public.c
--- work.orig/common/lib/modules/fglrx/build_mod/firegl_public.c	2010-08-20 21:51:48.000000000 +0200
+++ work/common/lib/modules/fglrx/build_mod/firegl_public.c	2010-08-20 22:02:32.000000000 +0200
@@ -320,7 +320,7 @@
     return firegl_release((KCL_IO_FILE_Handle)filp);
 }
 
-int ip_firegl_ioctl(struct inode* inode, struct file* filp, unsigned int cmd, unsigned long arg)
+int ip_firegl_ioctl(struct file* filp, unsigned int cmd, unsigned long arg)
 {
     return firegl_ioctl((KCL_IO_FILE_Handle)filp, cmd, arg);
 }
@@ -407,7 +407,7 @@
 #endif
     open:    ip_firegl_open,
     release: ip_firegl_release,
-    ioctl:   ip_firegl_ioctl,
+    unlocked_ioctl:   ip_firegl_ioctl,
     mmap:    ip_firegl_mmap,
 
     write:   ip_firegl_write,

This is a version of the patch with proper kernelversion detection, but the KERNEL_VERSION macro doesn't work with -rc kernels:

Code:

diff -aur work.orig/common/lib/modules/fglrx/build_mod/firegl_public.c work/common/lib/modules/fglrx/build_mod/firegl_public.c
--- work.orig/common/lib/modules/fglrx/build_mod/firegl_public.c	2010-08-20 21:51:48.000000000 +0200
+++ work/common/lib/modules/fglrx/build_mod/firegl_public.c	2010-08-20 23:46:37.000000000 +0200
@@ -320,7 +320,11 @@
     return firegl_release((KCL_IO_FILE_Handle)filp);
 }
 
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
+int ip_firegl_ioctl(struct file* filp, unsigned int cmd, unsigned long arg)
+#else
 int ip_firegl_ioctl(struct inode* inode, struct file* filp, unsigned int cmd, unsigned long arg)
+#endif
 {
     return firegl_ioctl((KCL_IO_FILE_Handle)filp, cmd, arg);
 }
@@ -407,7 +411,11 @@
 #endif
     open:    ip_firegl_open,
     release: ip_firegl_release,
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,36)
+    unlocked_ioctl:   ip_firegl_ioctl,
+#else
     ioctl:   ip_firegl_ioctl,
+#endif
     mmap:    ip_firegl_mmap,
 
     write:   ip_firegl_write,

**Spasswolf** · 08-21-2010, 10:29 AM

The second patch (and the KERNEL_VERSION macro) works with 2.6.36-rc1, too.

**kernelOfTruth** · 08-21-2010, 02:05 PM

Originally Posted by Spasswolf

The second patch (and the KERNEL_VERSION macro) works with 2.6.36-rc1, too.

confirmed !

it's running here now (compiz-fusion, so compositing and glxgears are working fine)

thanks a lot Spasswolf !

**kernelOfTruth** · 09-15-2010, 08:15 AM

does this still work for you with 2.6.36-rc4+ ?

**kernelOfTruth** · 09-15-2010, 08:16 AM

error message for me

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-rc4_plus'
/usr/src/linux-2.6.36-rc4_plus/arch/x86/Makefile:81: stack protector enabled but no compiler support
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
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_debug.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
CC [M] /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_pci.o
CC [M] /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_wait.o
CC [M] /var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_str.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
/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_AllocUserSpace32’:
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:196:5: error: implicit declaration of function ‘compat_alloc_user_space’
/var/tmp/portage/x11-drivers/ati-drivers-10.8/work/common/lib/modules/fglrx/build_mod/2.6.x/kcl_ioctl.c:196:5: warning: return makes pointer from integer without a cast
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....
/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’
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-rc4_plus'
make: *** [kmod_build] Error 2
emake failed
[31;01m*[0m ERROR: x11-drivers/ati-drivers-10.8 failed:
[31;01m*[0m 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-rc4_plus+ KDIR=/usr/src/linux kmod_build
[31;01m*[0m
[31;01m*[0m Call stack:
[31;01m*[0m ebuild.sh, line 56: Called src_compile
[31;01m*[0m environment, line 3721: Called linux-mod_src_compile
[31;01m*[0m environment, line 2885: Called die
[31;01m*[0m The specific snippet of code:
[31;01m*[0m 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}";

**kernelOfTruth** · 09-15-2010, 09:48 AM

looks like a very very old error to me:

[all variants] Ati,: fglrx module and vanilla kernel 2.6.24.x and 2.6.25.x on x86_64

Code:

--- kcl_ioctl.c.old	2010-09-15 16:38:39.399690030 +0200
+++ kcl_ioctl.c	2010-09-15 16:39:55.632689146 +0200
@@ -193,7 +193,7 @@
  */
 void* ATI_API_CALL KCL_IOCTL_AllocUserSpace32(long size)
 {
-    return compat_alloc_user_space(size);
+    return KCL_IOCTL_AllocUserSpace32(size);
 }
 
 #endif // __x86_64__

**Spasswolf** · 09-15-2010, 02:36 PM

My guess is you're missing CONFIG_COMPAT=y in your .config, which is selected automatically when enabling IA32_EMULATION (at least on an x86_64 system).

In 2.6.36-rc-4-git2 this will break again, because compat_alloc_user_space is renamed to arch_compat_alloc_user_space:
http://www.kernel.org/diff/diffview....-git2.bz2;z=15

Thread: request: patch for catalyst (10.7) to get 2.6.36-rc* working

Thread Tools

Search Thread

Display

request: patch for catalyst (10.7) to get 2.6.36-rc* working

Posting Permissions