Re: AMDGPU Driver patches/bugs

[Chavdar Ivanov <ci4ic4%gmail.com@localhost>]

FYI I tried amdgpu driver today on a fresh -current amd64 post the
above mentioned patches, using FirePro W5000; Xorg (system) started
fine and looked to be working well, but firefox 109.1 froze the
machine when I tried the benchmark from http://cad.onshape.com/check -
no crash, dark screen, no respond to ping. With the radeon driver this
hardware performs next to perfect for me - this test gives some 60
million triangles and lines per second. The radeon driver also returns
some 5000 points on glmark2; I didn't do a second attempt with the
amdgpu though - it probably would have froze again.

On Tue, 21 Feb 2023 at 22:23, Taylor R Campbell <riastradh%netbsd.org@localhost> wrote:
>
> > Date: Tue, 21 Feb 2023 13:20:13 -0800
> > From: Jeff Frasca <thatguy%jeff-frasca.name@localhost>
> >
> > I was going to try the radeon driver again, because I want to see if
> > my wayland compositor works better against it than the AMDGPU driver
> > (I'm getting some weird corruption problems with my compositor that
> > do not happen under Linux, but that's probably my code).
>
> We have seen other weird minor graphics corruption problems with X,
> even with xcompmgr or picom running.  I probably made another stupid
> bug, maybe in cacheability attributes or something, buried somewhere
> in the megabytes of diffs...
>
> > I'll send the FP stuff to tech-kern and CC you.  For the PRs, that's
> > the sendpr.cgi on netbsd.org, right?
>
> Sure, https://gnats.NetBSD.org -> report a new bug, which goes to some
> sendpr.cgi URL, or send-pr(1) if you like to do things from a local
> mailer (but it requires a working local mailer which can make outgoing
> SMTP connections, often blocked on residential networks).



-- 
----