On Mon, 2009-08-31 at 07:37 -0700, Brian Paul wrote:
> Eric Anholt wrote:
> > On Sat, 2009-08-29 at 11:24 -0700, Jose Fonseca wrote:
> >> Eric,
> >>
> >> This change broke softpipe on master (eg. terrain), and probably all 
> >> gallium based drivers. I'm pretty sure the vbo module was working 
> >> correctly without it, so there might be a different in the interpretation 
> >> of the MapBufferRange semantics between the drivers, but I couldn't find 
> >> it just by llooking at intel's and mesa state tracker implementation of 
> >> MapBufferRange.
> >>
> >> Was there any particular bug this change fixed?
> > 
> > Yes, garbage was drawn all over the screen when enabling VBOs in the VBO
> > module.  obj->Pointer should pretty clearly be the pointer to the mapped
> > range, as if you're creating a temporary space (system memory or
> > temporary buffer object) as part of MapRange, where else would it point
> > -- to an unmapped address?  It also then matches the
> > ARB_map_buffer_range semantics.  

Fair enough.

> Looks like a quick fix in
> > st_bufferobj_map_range.
> 
> I'll check in a fix for this shortly...

Thanks.

I should also change pipe_screen::buffer_flush_mapped_range semantics to
do the same. Even if Gallium allows simultaneous mappings of the same
buffer, it is better to have the same semantics everywhere.

Weird, I never received Eric's email, just Brian's reply. Was it just
me?

Jose


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Mesa3d-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev

Reply via email to