Unichrome

Benno Schulenberg wrote:
> > My lsmod says:
> >
> >    via                    32096  2
> >    drm                    59672  3 via
> >
> > (agpgart is compiled into my kernel).
> 
> Out of curiosity, what about via-agp?  Or is that only needed when 
> using a separate AGP videocard?

I don't have it loaded.  I've wondered what it's for, as it isn't
auto-loaded when X starts, and unlike the other modules it can't be
unloaded after starting and killing X because it's refcount stays at 1.

Perhaps loading it will fix the AGP problems I'm having.  Worth a
quick try (although I don't have the hardware for a few days so I
can't try immediately).

> >    gcc version 3.4.3 20050110 (Gentoo Linux 3.4.3.20050110,
> > ssp-3.4.3.20050110-0, pie-8.7.7)
> 
> Yep, using gcc 3.4.3 it loads fine and doesn't oops, with or without 
> preemption.  So it's a problem with gcc 3.3.5 (tried that one 
> without preemption too, still oopses).
> 
> But when kernel and modules are compiled with gcc 3.4.3, X can't 
> find a dri card, there is not even a /dev/dri dir:
> 
>   drmOpenDevice: node name is /dev/dri/card0
>   drmOpenDevice: open result is -1, (No such device)
>   drmOpenDevice: open result is -1, (No such device)
>   drmOpenDevice: Open failed
> 
> And so on, for 14 cards.  Making the node manually doesn't help, it 
> just gets deleted when starting X.  What's wrong here?  Xorg I 
> can't compile with gcc 3.4.3, it fails with an Internal Compiler 
> Error.  And drm and Mesa CVS I pulled again just now, still no dice, 
> no dri access.  What to try next?

I compiled X just a couple of days ago with that compiler, but the X I
compiled is 6.8.1 + patches (actually Gentoo's xorg-x11-6.8.1.902 +
Unichrome R30).  No internal compiler errors.

Bear in mind both X and GCC are patched in Gentoo, and either may have
been patched to fix those problems.

I have no idea why /dev/dri/card0 isn't opening for you.  It seems
unlikely to be related to the GCC version.  Are you sure you have
loaded the DRM modules?  Have you tried both the linux-2.6 one
(via.ko) and linux-core (drm.ko + via.ko)?

-- Jamie

Jamie Lokier wrote:
> Benno Schulenberg wrote:
> > Out of curiosity, what about via-agp?  Or is that only needed
> > when using a separate AGP videocard?
>
> I don't have it loaded.  I've wondered what it's for, as it isn't
> auto-loaded when X starts, and unlike the other modules it can't
> be unloaded after starting and killing X because it's refcount
> stays at 1.

Exactly the same here.

> I compiled X just a couple of days ago with that compiler, but
> the X I compiled is 6.8.1 + patches (actually Gentoo's
> xorg-x11-6.8.1.902 + Unichrome R30).  No internal compiler
> errors.
>
> Bear in mind both X and GCC are patched in Gentoo, and either may
> have been patched to fix those problems.

I've used the same compiler you used:  gcc (GCC) 3.4.3 20050110 
(Gentoo Hardened Linux 3.4.3.20050110, ssp-3.4.3.20050110-0, 
pie-8.7.7), on Xorg 6.8.2... ICE.  This is on an Athlon XP.
Of course, this is a hardened gcc, but I tried the vanilla one too, 
same thing:

  i686-pc-linux-gnu-gcc -c -ansi -pedantic -Wno-return-type -w   
  -I../.. -I../../exports/include   -Dlinux -D__i386__ 
  -D_POSIX_C_SOURCE=199309L   -D_POSIX_SOURCE -D_XOPEN_SOURCE 
  -D_BSD_SOURCE -D_SVID_SOURCE-D_GNU_SOURCE  -DFUNCPROTO=15 
  -DNARROWPROTO -DXTHREADS  -D_REENTRANT -DXUSE_MTSAFE_API   
  -DMALLOC_0_RETURNS_NULL  -DHAS_SNPRINTF -DLIBX11 
  -DPOSTLOCALELIBDIR=\"lib\"       -march=athlon-xp -Os -pipe 
  -fomit-frame-pointer -fstack-protector -mno-sse2 -mno-sse3 
  -fno-strict-aliasing   cmsMath.c -o unshared/cmsMath.o
  cmsMath.c: In function `_XcmsCubeRoot':
  cmsMath.c:91: internal compiler error: in try_split, at 
  emit-rtl.c:3343

> I have no idea why /dev/dri/card0 isn't opening for you.  It
> seems unlikely to be related to the GCC version.  Are you sure
> you have loaded the DRM modules?

Yes, no doubt, they are loaded.

> Have you tried both the 
> linux-2.6 one (via.ko) and linux-core (drm.ko + via.ko)?

Until yesterday I've been using the linux-2.6 one without a problem, 
and am using it again now.  But Thomas said we should be using the 
linux-core couple instead, but those simply don't work for me.

Benno

Jaxon Lee wrote:
> The opensource 3D drivers seem upset that they can't find .drirc.
> 
> libGL error:
> Can't open configuration file /etc/drirc: No such file or directory.
> libGL error:
> Can't open configuration file /home/akshunj/.drirc: No such file or 
> directory.
> 
> Where do I get it?  What's supposed to be in it?  Anybody know?

The .drirc file is just a config file that allows you to tweak certain 
device-specific options on a global or per-application basis.  You don't 
need it, so it's a bit of an over-statement for libGL to call this an 
"error".  In fact, I'm not even sure that the Unichrome driver supports 
any configurable options.  Hm...

Am Montag, den 14.02.2005, 07:45 -0800 schrieb Ian Romanick:
> Jaxon Lee wrote:
> > The opensource 3D drivers seem upset that they can't find .drirc.
> >=20
> > libGL error:
> > Can't open configuration file /etc/drirc: No such file or directory.
> > libGL error:
> > Can't open configuration file /home/akshunj/.drirc: No such file or=20
> > directory.
> >=20
> > Where do I get it?  What's supposed to be in it?  Anybody know?
>=20
> The .drirc file is just a config file that allows you to tweak certain=20
> device-specific options on a global or per-application basis.  You don't=20
> need it, so it's a bit of an over-statement for libGL to call this an=20
> "error".

The configuration parser uses __driUtilMessage to print this message.
Don't know why this ends up being printed as an error message. Sometimes
this seems to be the last message from the driver before it bails out for
some other reason, so people incorrectly blame it on the missing
configuration file. Should I rephrase the message to clearly state that
this is *not an error* or just not print anything if the file can't be
read?

>   In fact, I'm not even sure that the Unichrome driver supports=20
> any configurable options.  Hm...

You wouldn't see this message if the driver didn't support
configuration.


--=20
| Felix K=FChling <fxkuehl@...>                     http://fxk.de.vu |
| PGP Fingerprint: 6A3C 9566 5B30 DDED 73C3  B152 151C 5CC1 D888 E595 |