On Du, 15 mai 11, 11:56:18, Russell L. Carter wrote:
> I left out the step of replacing Driver="nouveau" with Driver="nvidia"
> in /etc/X11/xorg.conf, sorry.

Does this mean that everything is ok now?

Regards,
Andrei
--
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic

I had this happen with recent xorg update from Sid.




Log said that (proprietary) Nvidia driver was no longer compatable with the new Xorg version. So I ran its installer (version 260) with update which downloaded and installed version 270 (but did not keep that installer!) and this works fine.




Since 270 is the current version on Debian, how would I go from Nvidia's installer to Debian's version, preferrably using the DKMS?




The proprietary version works first time, every time, but still (after all this time) leaves the text mode of the terminal, i.e. from control/alt/F1, messed up.

On 05/15/2011 12:17 PM, Andrei Popescu wrote:
> On Du, 15 mai 11, 11:56:18, Russell L. Carter wrote:
>> I left out the step of replacing Driver="nouveau" with Driver="nvidia"
>> in /etc/X11/xorg.conf, sorry.
>
> Does this mean that everything is ok now?

Ah, sorry, no, I left out the step of switching from
nouveau to nvidia in xorg.conf from my description, I did
do it correctly when I switched. (I was hoping not to
derail the focus, but I guess I did anyway).

So current status is the current testing nvidia-kernel-dkms
breaks at the depmod step.

Thanks,
Russell

>
> Regards,
> Andrei


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 4DD03278.3000409@pinyon.org">http://lists.debian.org/4DD03278.3000409@pinyon.org

>15/05/2011 20:52, Russell L. Carter wrote:
>
> First, thanks much to the people on the other side of the globe who
> see a new day before me. I have dug deeper and can now get more specific
> about the nvidia blank screen problem. (For the record, I have a
> fixed hardware config that I've tracked debian-testing on for two
> years, and also tracking nvidia drivers. I used to just download
> them from nvidia.com but for the past few months I've been using
> dkms, which has worked beautifully.)
>
[...]
> (The recent memcpy/memmove issue is orthogonal to kernel modules,
> right?)
[...]

Not sure about that, I take it you are running the current testing/Sid
libc6 (2.13-4). The nvidia-kernel-dkms module sure compiles here on
Wheezy/Sid amd64, and xorg doesn't crash when loading nvidia module. I
don't run the Debian kernel though (custom 2.6.38.6).


> Setting up nvidia-kernel-dkms (270.41.06-1) ...
> Loading new nvidia-270.41.06 DKMS files...
> First Installation: checking all kernels...
> Building only for 2.6.38-2-amd64
> Building initial module for 2.6.38-2-amd64
> Done.
>
> nvidia.ko:
> Running module version sanity check.
> - Original module
> - No original module exists within this kernel
> - Installation
> - Installing to /lib/modules/2.6.38-2-amd64/updates/dkms/
>
> depmod...
> Message from syslogd@feyerabend at May 15 11:35:38 ...
> kernel:[ 177.226221] Oops: 0000 [#1] SMP
>
> Message from syslogd@feyerabend at May 15 11:35:38 ...
> kernel:[ 177.226223] last sysfs file: /sys/bus/acpi/drivers/NVIDIA ACPI Video
> Driver/uevent
>
> Message from syslogd@feyerabend at May 15 11:35:38 ...
> kernel:[ 177.226406] Stack:
>
> Message from syslogd@feyerabend at May 15 11:35:38 ...
> kernel:[ 177.226416] Call Trace:
>
> Message from syslogd@feyerabend at May 15 11:35:38 ...
> kernel:[ 177.226841] Code: 00 ba 00 00 00 00 be 3c 00 00 00 41 ff 55 20 48 89
> c3 b9 01 00 00 00 ba 00 00 00 00 be 15 00 00 00 4c 89 ef 41 ff 55 20 49 89 c6
> <48> 8b 05 f7 6a c6 00 48 89 45 10 8b 05 f5 6a c6 00 89 45 18 0f
>
[snip crash trace]

> May 15 11:35:38 feyerabend kernel: [ 177.226982] ---[ end trace
> 58be261eea03ecf3 ]---
> root@feyerabend:~#
>
> [end of message]
>

Do you have the logs from the upgrade right before the crash, what
packages got upgraded ?
In between your different trials you cleaned up thoroughly ? I am
thinking Nvidia .run here.

I see new nvidia packages entered Sid, one is "nvidia-installer-cleanup"
which may find some residual components of a previous install, it's
worth a shot if you are still stuck with this problem.


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 4DD18FB6.4050408@googlemail.com">http://lists.debian.org/4DD18FB6.4050408@googlemail.com

On 05/16/2011 01:57 PM, tv.debian@googlemail.com wrote:
>> 15/05/2011 20:52, Russell L. Carter wrote:
>>

[...]

>>
>
> Do you have the logs from the upgrade right before the crash, what
> packages got upgraded ?
> In between your different trials you cleaned up thoroughly ? I am
> thinking Nvidia .run here.

Not now... I reinstalled from scratch and after spending a few hours
learning the ins and outs of the udev bug (fix by rm -f /run, hmm) I have
a brand spanking new installation running nvidia-kernel-dkms successfully.

I didn't know about "Nvidia .run", I will certainly look closer at this
possibility if I have problems in the future.

> I see new nvidia packages entered Sid, one is "nvidia-installer-cleanup"
> which may find some residual components of a previous install, it's
> worth a shot if you are still stuck with this problem.

Thanks for the info, I appreciate the help.

Regards,
Russell


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 4DD19F3B.8060803@pinyon.org">http://lists.debian.org/4DD19F3B.8060803@pinyon.org

On Lu, 16 mai 11, 15:03:39, Russell L. Carter wrote:
>
> Not now... I reinstalled from scratch and after spending a few hours
> learning the ins and outs of the udev bug (fix by rm -f /run, hmm) I have
> a brand spanking new installation running nvidia-kernel-dkms successfully.

- current udev in sid falls back to /dev/.udev if /run is not available
- /run *is* necessary now and correctly supported by current initscripts
in sid

Conclusion: this hack is not necessary anymore.

Regards,
Andrei
--
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic