NVidia 319.12 breaks desktop

[clubex]

Nvidia 319.12 fail to build nvidia module against 3.8-6.towo-siduction-amd64.

/var/lib/dkms/nvidia/319.12/build/make.log

Code Select Expand


DKMS make.log for nvidia-319.12 for kernel 3.8-6.towo-siduction-amd64 (x86_64)
Wed Apr 10 12:45:47 BST 2013
make: Entering directory `/var/lib/dkms/nvidia/319.12/build'
make -C /lib/modules/3.8-6.towo-siduction-amd64/build  M=`/bin/pwd` modules
make[1]: Entering directory `/usr/src/linux-headers-3.8-6.towo-siduction-amd64'
 CC [M]  /var/lib/dkms/nvidia/319.12/build/nv.o
In file included from include/linux/kernel.h:10:0,
                from include/linux/sched.h:15,
                from include/linux/utsname.h:5,
                from /var/lib/dkms/nvidia/319.12/build/nv-linux.h:46,
                from /var/lib/dkms/nvidia/319.12/build/nv.c:13:
include/linux/bitops.h: In function 'hweight_long':
include/linux/bitops.h:66:41: warning: signed and unsigned type in conditional expression [-Wsign-compare]
In file included from /usr/src/linux-headers-3.8-6.towo-siduction-amd64/arch/x86/include/asm/uaccess.h:552:0,
                from include/linux/poll.h:11,
                from /var/lib/dkms/nvidia/319.12/build/nv-linux.h:117,
                from /var/lib/dkms/nvidia/319.12/build/nv.c:13:
/usr/src/linux-headers-3.8-6.towo-siduction-amd64/arch/x86/include/asm/uaccess_64.h: In function 'copy_from_user':
/usr/src/linux-headers-3.8-6.towo-siduction-amd64/arch/x86/include/asm/uaccess_64.h:62:6: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
/var/lib/dkms/nvidia/319.12/build/nv.c: At top level:
/var/lib/dkms/nvidia/319.12/build/nv.c:37:30: fatal error: conftest/patches.h: No such file or directory
compilation terminated.
make[2]: *** [/var/lib/dkms/nvidia/319.12/build/nv.o] Error 1
make[1]: *** [_module_/var/lib/dkms/nvidia/319.12/build] Error 2
make[1]: Leaving directory `/usr/src/linux-headers-3.8-6.towo-siduction-amd64'
make: *** [modules] Error 2
make: Leaving directory `/var/lib/dkms/nvidia/319.12/build'


Again I'm not worried about my machine; this is solely a warning to others.

[dibl]

Confirmed -- same problem, same make.log report.  I was able to uninstall and reinstall 304.88.

[towo]

Should be fixed in the new uploaded version.

[clubex]

Looks like a siduction build bug so hopefully it'll soon be fixed.

[towo]

Code Select Expand


nvidia-kernel-dkms (319.12-0.siduction.2) wird eingerichtet ...
Loading new nvidia-319.12 DKMS files...
Building for 3.8-3.towo-siduction-amd64 and 3.8-6.towo-siduction-amd64
Building initial module for 3.8-3.towo-siduction-amd64
Done.

nvidia:
Running module version sanity check.
- Original module
  - No original module exists within this kernel
- Installation
  - Installing to /lib/modules/3.8-3.towo-siduction-amd64/updates/dkms/

depmod....

DKMS: install completed.
Building initial module for 3.8-6.towo-siduction-amd64
Done.

nvidia:
Running module version sanity check.
- Original module
  - No original module exists within this kernel
- Installation
  - Installing to /lib/modules/3.8-6.towo-siduction-amd64/updates/dkms/

depmod....

DKMS: install completed.

So it seems fixed now.

[dibl]

Yep, 319.12 installed successfully here.

vielen Dank, towo!

[clubex]

Excellent: Many thanks towo

[piper]

Installed fine here, boot-up errors and dmesg errors

Code Select Expand

piper@x1:~$ dmesg
[  338.359178]  [<ffffffff810f720b>] ? do_vfs_ioctl+0x3b8/0x3fa
[  338.359187]  [<ffffffff810f7907>] ? poll_select_copy_remaining+0xe0/0xf5
[  338.359193]  [<ffffffff810f729a>] ? sys_ioctl+0x4d/0x7b
[  338.359199]  [<ffffffff8134a2a9>] ? system_call_fastpath+0x16/0x1b
[  338.359220] BUG: using smp_processor_id() in preemptible [00000000] code: kwin/5068
[  338.359338] caller is _nv012291rm+0x9/0xe [nvidia]
[  338.359347] Pid: 5068, comm: kwin Tainted: P           O 3.8-6.towo-siduction-amd64 #1
[  338.359349] Call Trace:
[  338.359362]  [<ffffffff8119e2e9>] ? debug_smp_processor_id+0xc9/0xe0
[  338.359485]  [<ffffffffa0a0269b>] ? _nv012291rm+0x9/0xe [nvidia]
[  338.359599]  [<ffffffffa055a7bf>] ? _nv015429rm+0x39/0x23e [nvidia]
[  338.359707]  [<ffffffffa053f40c>] ? _nv000696rm+0x1c/0x59 [nvidia]
[  338.359831]  [<ffffffffa09ff43d>] ? _nv000790rm+0x98d/0x9ee [nvidia]
[  338.359947]  [<ffffffffa0a0840b>] ? rm_ioctl+0x76/0x100 [nvidia]
[  338.360083]  [<ffffffffa0a2587f>] ? nv_kern_ioctl+0x334/0x39a [nvidia]
[  338.360111]  [<ffffffff8127adef>] ? sys_recvfrom+0xf5/0x123
[  338.360222]  [<ffffffffa0a2591b>] ? nv_kern_unlocked_ioctl+0x19/0x1d [nvidia]
[  338.360234]  [<ffffffff810f6a26>] ? vfs_ioctl+0x1e/0x31
[  338.360239]  [<ffffffff810f720b>] ? do_vfs_ioctl+0x3b8/0x3fa
[  338.360249]  [<ffffffff810f7907>] ? poll_select_copy_remaining+0xe0/0xf5
[  338.360258]  [<ffffffff810f729a>] ? sys_ioctl+0x4d/0x7b

Code Select Expand

piper@x1:~$ dmesg | grep nvidia
[  479.472849] caller is _nv012291rm+0x9/0xe [nvidia]
[  479.472982]  [<ffffffffa0a0269b>] ? _nv012291rm+0x9/0xe [nvidia]
[  479.473141]  [<ffffffffa055a7bf>] ? _nv015429rm+0x39/0x23e [nvidia]
[  479.473290]  [<ffffffffa053f40c>] ? _nv000696rm+0x1c/0x59 [nvidia]
[  479.473441]  [<ffffffffa09ff43d>] ? _nv000790rm+0x98d/0x9ee [nvidia]
[  479.473566]  [<ffffffffa0a0840b>] ? rm_ioctl+0x76/0x100 [nvidia]
[  479.473680]  [<ffffffffa0a2587f>] ? nv_kern_ioctl+0x334/0x39a [nvidia]
[  479.473797]  [<ffffffffa0a2591b>] ? nv_kern_unlocked_ioctl+0x19/0x1d [nvidia]

Graphics:  Card: NVIDIA GF106 [GeForce GTS 450] X.Org: 1.12.4 driver: nvidia Resolution: 1920x1200@60.0hz
          GLX Renderer: GeForce GTS 450/PCIe/SSE2 GLX Version: 4.3.0 NVIDIA 319.12

[michaa7]

slightly OT and BTW:

is NVidia 319.12 still incompatible with GeForce6xxx and am I stuck with 304.64-4? I have it on hold since some weeks.

Is there a way to automaticly install the newest GeForce6xxx Nvidia driver ( does nvidia-detect the job? To me it seems it only distinguishes between current and both legacy drivers. And I don't like to mess around with nvidia packages without having enough information of what should and could happen)

[towo]

slightly OT and BTW:

is NVidia 319.12 still incompatible with GeForce6xxx and am I stuck with 304.64-4? I have it on hold since some weeks.

Is there a way to automaticly install the newest GeForce6xxx Nvidia driver ( does nvidia-detect the job? To me it seems it only distinguishes between current and both legacy drivers. And I don't like to mess around with nvidia packages without having enough information of what should and could happen)

For Geforce 6xxx and 7xxx only the 304.series will support those cards.
For that, you should install version 304.88-1, it fixes a security bug.

Btw, the driver 319.12 seems to have massive problems with preemt-enabled kernels, i will ask in #nvidia.
For now, i have switched back my system to 313.3ß.

[piper]

Btw, the driver 319.12 seems to have massive problems with preemt-enabled kernels, i will ask in #nvidia.
For now, i have switched back my system to 313.3ß.

I think I will do the same for now, thanks towo

[dibl]

Btw, the driver 319.12 seems to have massive problems with preemt-enabled kernels, i will ask in #nvidia.

What would be the sign of trouble on this? I was away from my system and now it has 2.5 hours uptime and no sign of any problem.

[towo]

On my system after a while my desktop has no response anymore, i had to kill my x-session.

[dibl]

OK thanks. I'm going to be brave and keep using this desktop, but not upgrade 3 others that have Nvidia GPUs, until I hear this is resolved.  If my desktop breaks I can always go back a version on the driver.

[Yeti]

Hello,

how to install 319.12-0.siduction.2 ?

Regards
yeti