On 971346407 seconds since the Beginning of the UNIX epoch
Johan Danielsson wrote:
>
>"R. C. Dowdeswell" <elric@mabelode.imrryr.org> writes:
>
>> Try the attached patch. 
>
>No cigar yet.

Okay, I have a quick test.  I think that the ptr that is screwed
is dc->dc_ramdac_intr.  So, in this patch I don't reset it to
NULL after successfully completing the interrupt.  Hopefully this
will show the problem.  If not, then tga_intr may be being passed
a NULL -- but I would imagine that in that case you'd crash well
before you have a chance to run XalphaNetBSD.

>> Have you tried running the Xserver with the -noaccel option to see if this
>> is being caused by the accelerations?  This would help track down the bug
>> a bit.
>
>That makes it go away.

Oh, one thing, so this is an intermittant failure?  Or is this a different
card that you are running X on with a bit more success?

Thanks,

 == Roland Dowdeswell                      http://www.Imrryr.ORG/~elric/  ==
 == The Unofficial NetBSD Web Pages        http://www.Imrryr.ORG/NetBSD/  ==
 == The NetBSD Project                            http://www.NetBSD.ORG/  ==

Index: tga.c
===================================================================
RCS file: /cvsroot/syssrc/sys/dev/pci/tga.c,v
retrieving revision 1.26
diff -u -u -r1.26 tga.c
--- tga.c	2000/06/26 04:56:25	1.26
+++ tga.c	2000/10/14 09:58:24
@@ -596,7 +599,9 @@
 		}
 	}
 	dc->dc_ramdac_intr(dcrc);
+#if 0
 	dc->dc_ramdac_intr = NULL;
+#endif
 	TGAWREG(dc, TGA_REG_SISR, 0x00000001);
 	TGAREGWB(dc, TGA_REG_SISR, 1);
 	return (1);