On Tue, 2008-03-18 at 02:01 +0100, Markus Amsler wrote:
> Michel Dänzer wrote:
> > On Mon, 2008-03-17 at 12:10 +0100, Jerome Glisse wrote:
> >
> >> Markus i think you should apply for an account you have provided
> >> several good patch for r300, there is the procedure on fdo.
> >>
> >
> > I'm afraid this may be a little premature yet:
> >
> I think so too, I'll keep sending patches for now (or are bugs with
> proposed fixes better?)
> > Commit f0c8863dbb71b2c3605bb8eb9526c7d8d13793cf ('r300: Simplify
> > r300VAPInputRoute0, check for valid input.') makes any 3D app lock up
> > the RV350 in my PowerBook immediately.
> >
> Looks like an endianess issue.
> > Commit c48212fe5861291efd134b35e67568c79b15f960 ('r300: Simplify
> > r300VAPInputRoute1.') causes non-opaque windows to be corrupted with
> > compiz.
> >
> I'll have a look at it, when I get compiz running.
Actually, that looks likely endianness related as well - AFAICT in both
cases you changed the loops from operating on a 32 bit word at a time to
16 bits.
I don't have time right now to look into this in detail though.
> That leaves the question, what's the best way to test patches? I ran
> quite some src/progs, wow in opengl/d3d, openarena and even ppracer
> without any issues.
Sounds like a nice selection. If these turn out to be endianness bugs
indeed, there isn't really a way to test that on a little endian
machine. Probably the best you can do is get a feeling for endianness
related issues and ask for testing on big endian when you suspect a
change might affect it.
--
Earthling Michel Dänzer | http://tungstengraphics.com
Libre software enthusiast | Debian, X and DRI developer
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Mesa3d-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
