Nouveau Archive on lore.kernel.org
 help / color / Atom feed
* [Nouveau] Request for information and/or assistance
@ 2021-03-04  2:20 o1bigtenor
  2021-03-08  9:04 ` Karol Herbst
  0 siblings, 1 reply; 3+ messages in thread
From: o1bigtenor @ 2021-03-04  2:20 UTC (permalink / raw)
  To: nouveau

Greetings

Running debian testing, 2 nvidia gpus (GP107 and GF110) and 5 monitors
(1 - 3840x2160 and 2 1920x1080 on the GP107 and 2 - 1920x1080 on the
GF110) using one screen (7680x3000).

I would still like to modify some things but can't find any docs on
how to do that and my attempts have only made things worse and I can
function, even though some things are right awkward, on the system.
What I've run into is that my graphics system either refuses to come
out of hibernation or decides to come back on but in a locked state. I
then log into this machine from another on the network and do a reboot
for a 'graceful' restart (rather than use the kill button - - - the
hard shut off).

This event, graphics shutdown, is happening in direct relationship to
how much I have happening on the desktops of which I have lots.

In searching it seems that nouveau support for the GP107 is not yet
complete - - - - or that's what it seems like.

Is there someone out there who has some idea what the problem is?
(There are quite a number of different individuals mentioning this
issue in one of the bug reports.)

I would be happen to revise the nouveau pages on xrandr updating the
information from 1.3 to 1.5 and possibly some of the other pages as
well. I have no programming skills but I do have reasonable writing
skills - - - - would rather trade than to continue taking that 1/2
hour it takes to set my system up after each reboot - - - - taking an
hour or even 1.5 hours a day doing this is a huge drain on my
resources and would rather change this.

Please advise!

Regards

Aj
_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [Nouveau] Request for information and/or assistance
  2021-03-04  2:20 [Nouveau] Request for information and/or assistance o1bigtenor
@ 2021-03-08  9:04 ` Karol Herbst
  2021-03-08 13:13   ` o1bigtenor
  0 siblings, 1 reply; 3+ messages in thread
From: Karol Herbst @ 2021-03-08  9:04 UTC (permalink / raw)
  To: o1bigtenor; +Cc: nouveau

On Thu, Mar 4, 2021 at 3:21 AM o1bigtenor <o1bigtenor@gmail.com> wrote:
>
> Greetings
>
> Running debian testing, 2 nvidia gpus (GP107 and GF110) and 5 monitors
> (1 - 3840x2160 and 2 1920x1080 on the GP107 and 2 - 1920x1080 on the
> GF110) using one screen (7680x3000).
>
> I would still like to modify some things but can't find any docs on
> how to do that and my attempts have only made things worse and I can
> function, even though some things are right awkward, on the system.
> What I've run into is that my graphics system either refuses to come
> out of hibernation or decides to come back on but in a locked state. I
> then log into this machine from another on the network and do a reboot
> for a 'graceful' restart (rather than use the kill button - - - the
> hard shut off).
>
> This event, graphics shutdown, is happening in direct relationship to
> how much I have happening on the desktops of which I have lots.
>
> In searching it seems that nouveau support for the GP107 is not yet
> complete - - - - or that's what it seems like.
>
> Is there someone out there who has some idea what the problem is?
> (There are quite a number of different individuals mentioning this
> issue in one of the bug reports.)
>
> I would be happen to revise the nouveau pages on xrandr updating the
> information from 1.3 to 1.5 and possibly some of the other pages as
> well. I have no programming skills but I do have reasonable writing
> skills - - - - would rather trade than to continue taking that 1/2
> hour it takes to set my system up after each reboot - - - - taking an
> hour or even 1.5 hours a day doing this is a huge drain on my
> resources and would rather change this.
>
> Please advise!
>

I think at this point we can only guess what's going on there and
having system logs would help us figure out what's wrong. Mind sharing
system logs from the time the crash or whatever that is happens?
Output of dmesg via ssh could be enough.

> Regards
>
> Aj
> _______________________________________________
> Nouveau mailing list
> Nouveau@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/nouveau
>

_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [Nouveau] Request for information and/or assistance
  2021-03-08  9:04 ` Karol Herbst
@ 2021-03-08 13:13   ` o1bigtenor
  0 siblings, 0 replies; 3+ messages in thread
From: o1bigtenor @ 2021-03-08 13:13 UTC (permalink / raw)
  To: Karol Herbst; +Cc: nouveau

On Mon, Mar 8, 2021 at 3:05 AM Karol Herbst <kherbst@redhat.com> wrote:
>
> On Thu, Mar 4, 2021 at 3:21 AM o1bigtenor <o1bigtenor@gmail.com> wrote:
> >
> > Greetings
> >
> > Running debian testing, 2 nvidia gpus (GP107 and GF110) and 5 monitors
> > (1 - 3840x2160 and 2 1920x1080 on the GP107 and 2 - 1920x1080 on the
> > GF110) using one screen (7680x3000).
> >Direct firmware load for nouveau/nvc8_fuc084 failed with error -2
> > I would still like to modify some things but can't find any docs on
> > how to do that and my attempts have only made things worse and I can
> > function, even though some things are right awkward, on the system.
> > What I've run into is that my graphics system either refuses to come
> > out of hibernation or decides to come back on but in a locked state. I
> > then log into this machine from another on the network and do a reboot
> > for a 'graceful' restart (rather than use the kill button - - - the
> > hard shut off).
> >
> > This event, graphics shutdown, is happening in direct relationship to
> > how much I have happening on the desktops of which I have lots.
> >
> > In searching it seems that nouveau support for the GP107 is not yet
> > complete - - - - or that's what it seems like.
> >
> > Is there someone out there who has some idea what the problem is?
> > (There are quite a number of different individuals mentioning this
> > issue in one of the bug reports.)
> >
> > I would be happy to revise the nouveau pages on xrandr updating the
> > information from 1.3 to 1.5 and possibly some of the other pages as
> > well. I have no programming skills but I do have reasonable writing
> > skills - - - - would rather trade than to continue taking that 1/2
> > hour it takes to set my system up after each reboot - - - - taking an
> > hour or even 1.5 hours a day doing this is a huge drain on my
> > resources and would rather change this.
> >
> > Please advise!
> >
>
> I think at this point we can only guess what's going on there and
> having system logs would help us figure out what's wrong. Mind sharing
> system logs from the time the crash or whatever that is happens?
> Output of dmesg via ssh could be enough.
>

As requested:

debianbase kernel: [  436.180582] nouveau 0000:02:00.0: Direct
firmware load for nouveau/nvc8_fuc084 failed with error -2
Mar  5 20:24:40 debianbase kernel: [  436.180597] nouveau
0000:02:00.0: Direct firmware load for nouveau/nvc8_fuc084d failed
with error -2
Mar  6 00:00:02 debianbase kernel: [13358.378698] audit: type=1400
audit(1615010402.464:22): apparmor="DENIED" operation="capable"
profile="/usr/sbin/cupsd" pid=13168 comm="cupsd" capability=12
capname="net_admin"
Mar  6 00:00:02 debianbase kernel: [13358.415978] audit: type=1400
audit(1615010402.500:23): apparmor="DENIED" operation="capable"
profile="/usr/sbin/cups-browsed" pid=13170 comm="cups-browsed"
capability=23  capname="sys_nice"
Mar  6 07:40:56 debianbase kernel: [41012.952323] perf: interrupt took
too long (2511 > 2500), lowering kernel.perf_event_max_sample_rate to
79500
Mar  6 12:04:47 debianbase kernel: [56843.957285] perf: interrupt took
too long (3152 > 3138), lowering kernel.perf_event_max_sample_rate to
63250
Mar  6 15:56:25 debianbase kernel: [70741.868103] perf: interrupt took
too long (3952 > 3940), lowering kernel.perf_event_max_sample_rate to
50500
Mar  7 00:00:01 debianbase kernel: [99758.601985] audit: type=1400
audit(1615096801.491:24): apparmor="DENIED" operation="capable"
profile="/usr/sbin/cupsd" pid=56136 comm="cupsd" capability=12
capname="net_admin"
Mar  7 00:00:01 debianbase kernel: [99758.625644] audit: type=1400
audit(1615096801.515:25): apparmor="DENIED" operation="capable"
profile="/usr/sbin/cups-browsed" pid=56139 comm="cups-browsed"
capability=23  capname="sys_nice"

Is this sufficient?
_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, back to index

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-04  2:20 [Nouveau] Request for information and/or assistance o1bigtenor
2021-03-08  9:04 ` Karol Herbst
2021-03-08 13:13   ` o1bigtenor

Nouveau Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/nouveau/0 nouveau/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 nouveau nouveau/ https://lore.kernel.org/nouveau \
		nouveau@lists.freedesktop.org
	public-inbox-index nouveau

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.freedesktop.lists.nouveau


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git