Bug#972514: fixed in nvidia-graphics-drivers 450.80.02-1

[David Headland]

> I have 455.23.04-1 from experimental installed since my last post,
> everything is stable, and clinfo and glxinfo report fully working OpenCL
> and GLX setups. 

[...]

> Is there any reason you made a new nvidia 450 release instead of
> propagating the already working nvidia 455 experimental builds?

It's interesting that you mention that, as switching to experimental was
the first thing I tried when I experienced this bug. Whilst it initially
seemed to solve the problem for me, I did run into another issue whereby
it would only appear to correctly initialise the display one per boot.
After this, if the HDMI link went down through DPMS blanking or
switching to another input into the monitor, nothing I could do would
cause my monitor to see a valid signal from the card. I tried all sorts
of things, like switching to a VC and back (VC switching didn't seem to
work either), unplugging the cable, power cycling the monitor, nothing
worked except rebooting the machine. The same thing happens if I switch
to another monitor completely.

It might be that 455 works for some but not all cards, perhaps? For
reference, this is what lspci shows for my card:

01:00.0 VGA compatible controller: NVIDIA Corporation GP108 [GeForce GT
1030] (rev a1)

All the best,
-Dave.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-nvidia-devel/attachments/20201029/2f654336/attachment.sig>