On Sun, Sep 28, 2008 at 11:14 PM, Zack Rusin <[EMAIL PROTECTED]> wrote: > On Sunday 28 September 2008 15:04:42 Stephane Marchesin wrote: >> >> Oh, I'm thinking about doing driver-specific llvm backends here. > > Yea, that was always my plan. > >> I've realised that the difference between GPUs really didn't allow common >> intermediate representation. Furthermore, the llvm tablegen format is >> very powerful and I don't think it'll be much more work. >> >> So basically instead of doing >> TGSI -> LLVM -> TGSI -> driver > > The only scenario for this was D3D9 for which the driver has to live in the > Windows kernel, and getting LLVM running in a Windows kernel would be > excidingly difficult. So the use-case was client-side library does the > conversion to LLVM, runs the LLVM optimization passes, transforms back into > TGSI and feeds the optimized TGSI to the driver. >
Wait, you got some level of llvm stuff running already ? Or is that completely un-mergeable ? I just hope I'm not doing duplicated work here... >> I think we should be doing >> TGSI -> LLVM -> driver > > Yea, for all cases where you can get LLVM code-generator into the actual > driver that's what we want. Ok, great! It's good when things make sense and everyone agrees on the right aproach :) Stephane ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
