Re: [Dri-devel] Re: Radeon / Mesa problem?

[Keith W. <ke...@va...>]

Petr Sebor wrote:
> 
> Hello Keith,
> 
> I haven't had enough time to play with the pipeline to test if DrawElements with lower than
> 3k vertices will survive, but this is the yesterday's backtrace I promised. I hope I will be
> able to do more tests today...
> 
> Thanks!
> Petr
> 
> 0x4661c2fb in clipTEX0_RGBA0_FOG (VB=0x85c7b20, dst=3000, t=0.552259386, in=32, out=1) at interp_tmp.h:73
> 73                      VB->Color[0]->data[out] );
> 
> #0  0x4661c2fb in clipTEX0_RGBA0_FOG (VB=0x85c6fc0, dst=3000, t=0.515004754, in=32, out=1) at interp_tmp.h:73
> #1  0x4661e7c2 in viewclip_polygon_4 (VB=0x85c6fc0, n=3, vlist=0xbfffc184, mask=2 '\002') at general_clip.h:43
> #2  0x466ee21d in gl_render_clipped_triangle (ctx=0x82ddf98, n=3, vlist=0xbfffc184, pv=1) at vbrender.c:210
> #3  0x466edb1b in indexed_render_tris (VB=0x85c6fc0, state=0x46739760, elt=0x86834b8, start=0, count=5766) at vbindirect.c:217
> #4  0x466edcd1 in gl_render_elts (VB=0x85c6fc0) at vbindirect.c:319
> #5  0x46689e2c in gl_run_pipeline (VB=0x85c6fc0) at pipeline.c:525
> #6  0x46633ce0 in gl_cva_force_precalc (ctx=0x82ddf98) at cva.c:826
> #7  0x466e54d1 in _mesa_DrawElements (mode=4, count=5766, type=5123, indices=0x831ecd0) at varray.c:1282
> 

Hmm.  Looks like it is using the locked paths, but according to what you said
it really shouldn't be.  

A question:  Do the arrays have 5766 vertices, or are you rendering 5766
elements?  There's a difference, of course.

Keith