daily CVS update output

2024-03-14 Thread NetBSD source update


Updating src tree:
P src/distrib/evbarm/instkernel/sshramdisk/Makefile
P src/distrib/sets/lists/tests/mi
P src/distrib/utils/embedded/conf/armv7.conf
P src/doc/3RDPARTY
P src/etc/mtree/NetBSD.dist.tests
P src/lib/libusbhid/usb_hid_usages
P src/share/man/man4/ihidev.4
P src/sys/arch/hppa/conf/GENERIC
P src/sys/arch/sparc/include/limits.h
P src/sys/dev/ic/dwc_gmac.c
P src/tests/usr.bin/Makefile
U src/tests/usr.bin/stat/Makefile
U src/tests/usr.bin/stat/t_stat.sh
P src/usr.bin/stat/stat.c

Updating xsrc tree:


Killing core files:




Updating file list:
-rw-rw-r--  1 srcmastr  netbsd  45774907 Mar 15 03:04 ls-lRA.gz


Graphics card issues with new system build

2024-03-14 Thread Paul Goyette

OK, I'm getting frustrated as hell.  :-)

I've got my new system built, and managed to get the dmesg uploaded to
NYCBug ...

It's got a AMD Ryzen 9 7950X3D CPU, which has internal video configured
as acpivga0 - I disable this (using userconf) when I boot because I need
to use the older-but-supported GeForce 730 graphics card.

Boot comes up normally and starts X (via xdm).  However, I'm getting the
following message logged on the console.  As far as I can tell, they're
logged when switching from the xdm console session to a non-X session;
but it is difficult to be certain of the timing.

nouveau0: error: DRM: core notifier timeout

I also sometimes get

nouveau0: failed to idle channel n [user]

when stopping xdm.

During normal operation, things are OK, other than "jerky" cursor
tracking.  Most often, the cursor "sticks" in one spot for a few
seconds, and then continues to track.  "One spot" is very often, but
NOT ALWAYS, at a point where the mouse icon needs to change (ie, it
enters or leaves a window or button, etc.)

After a while, the system just hangs (suspicion: it's hitting one of
the core notifier timeouts, and hangs forever rather than time-out).
I can still access the machine via network log-in but cannot get
anything done "locally".  At this point I can't even switch to the
non-X console sessions.  The hang-time varies, but comes much sooner
when there is heavy system activity - like build.sh - going on.  It
even comes, however, with no activity at all - just staring at the
screen for a few hours triggers the hang.

I've got a bunch of newer video cards, but our current video drivers
don't support them, and end up using genfb with a "default" mode-line
that makes everything look wrong.  And I've already cannabilized the
old machine (which had a now non-functional GeForce 1050 Ti - maybe
it was a 1080 but I can't find any marking) so I can't go back.

At this point, I am stuck and don't know how to proceed.  I'm willing
to buy one more video card (but only one!) if I have a high likelihood
of it working.  But if there's something else going on which is likely
to be fatal no matter what video card I choose I'd rather not spend
more cash.  If you do recommend another card, please be very explicit
in identifying the make and model and any other relevant attributes
(how much graphics memory, for example).

All reasonable suggestions will be appreciated.  If there are specific
debug commands to try, please let me know.  (But also know that I don't
know much about the video subsystem - this is way beyond my usual level
of knowledge.)


+-+--+--+
| Paul Goyette (.sig) | PGP Key fingerprint: | E-mail addresses:|
| (Retired)   | 1B11 1849 721C 56C8 F63A | p...@whooppee.com|
| Software Developer  | 6E2E 05FD 15CE 9F2D 5102 | pgoye...@netbsd.org  |
| & Network Engineer  |  | pgoyett...@gmail.com |
+-+--+--+


Re: dwiic errors

2024-03-14 Thread Paul Goyette

On Thu, 14 Mar 2024, Manuel Bouyer wrote:


It could also be some sensors I guess. Any chance to see what attaches
at dwiic0 ? Maybe entering ddb before the console gets spammed ?


The error messages start immediately after attaching the dwiic.  I
have waited for several minutes and the timeout messages continue
with nothing else.


FWIW I have a laptop with the touchpad as ihidev@dwiic and it works
fine with RC6


Yeah, this system seems to behave somewhat oddly.

+-+--+--+
| Paul Goyette (.sig) | PGP Key fingerprint: | E-mail addresses:|
| (Retired)   | 1B11 1849 721C 56C8 F63A | p...@whooppee.com|
| Software Developer  | 6E2E 05FD 15CE 9F2D 5102 | pgoye...@netbsd.org  |
| & Network Engineer  |  | pgoyett...@gmail.com |
+-+--+--+


Re: dwiic errors

2024-03-14 Thread Manuel Bouyer
On Thu, Mar 14, 2024 at 08:42:47AM -0700, Paul Goyette wrote:
> On Thu, 14 Mar 2024, Michael van Elst wrote:
> 
> > p...@whooppee.com (Paul Goyette) writes:
> > 
> > > as soon as you proceed past this point (including normal non-single-
> > > user boot), the dwiic starts spewing time-out messages.  These
> > > messages come every 0.5 second or so, and there's usually a hundred
> > > or more messages before they stop;  in some cases the messages have
> > > continued to stream by for several minutes (at which point I pressed
> > > the reset button).  The value for %d is always 0 or 1.
> > 
> > Probably result of
> > 
> > GENERIC:ihidev* at iic?
> > 
> > that is probing for a modern laptop touchpad.
> > 
> > Can you disable ihidev instead of dwiic and see what happens then ?
> 
> No change.  It attaches dwiic0 and then starts with the messages.

It could also be some sensors I guess. Any chance to see what attaches
at dwiic0 ? Maybe entering ddb before the console gets spammed ?

FWIW I have a laptop with the touchpad as ihidev@dwiic and it works
fine with RC6

-- 
Manuel Bouyer 
 NetBSD: 26 ans d'experience feront toujours la difference
--


Re: dwiic errors

2024-03-14 Thread Paul Goyette

On Thu, 14 Mar 2024, Michael van Elst wrote:


p...@whooppee.com (Paul Goyette) writes:


as soon as you proceed past this point (including normal non-single-
user boot), the dwiic starts spewing time-out messages.  These
messages come every 0.5 second or so, and there's usually a hundred
or more messages before they stop;  in some cases the messages have
continued to stream by for several minutes (at which point I pressed
the reset button).  The value for %d is always 0 or 1.


Probably result of

GENERIC:ihidev* at iic?

that is probing for a modern laptop touchpad.

Can you disable ihidev instead of dwiic and see what happens then ?


No change.  It attaches dwiic0 and then starts with the messages.


+-+--+--+
| Paul Goyette (.sig) | PGP Key fingerprint: | E-mail addresses:|
| (Retired)   | 1B11 1849 721C 56C8 F63A | p...@whooppee.com|
| Software Developer  | 6E2E 05FD 15CE 9F2D 5102 | pgoye...@netbsd.org  |
| & Network Engineer  |  | pgoyett...@gmail.com |
+-+--+--+


Re: dwiic errors

2024-03-14 Thread Michael van Elst
p...@whooppee.com (Paul Goyette) writes:

>as soon as you proceed past this point (including normal non-single-
>user boot), the dwiic starts spewing time-out messages.  These
>messages come every 0.5 second or so, and there's usually a hundred
>or more messages before they stop;  in some cases the messages have
>continued to stream by for several minutes (at which point I pressed
>the reset button).  The value for %d is always 0 or 1.

Probably result of

GENERIC:ihidev* at iic?

that is probing for a modern laptop touchpad.

Can you disable ihidev instead of dwiic and see what happens then ?