On Thursday, October 02, 2014 11:53:27 AM Matt Turner wrote: > On Wed, Oct 1, 2014 at 12:18 PM, Jason Ekstrand <ja...@jlekstrand.net> wrote: > > Previously, we had a MAX_SAMPLER_MESSAGE_SIZE which we used instead. > > However, some FB write messages can validly be longer than this so we need > > something different. Since MAX_SAMPLER_MESSAGE_SIZE is validly useful on > > its own, we leave it alone and add a new MAX_GRF_SIZE that's big enough for > > FB writes. > > I might call it MAX_FB_WRITE_SIZE or something. MAX_GRF_SIZE is a > little generic, but in the register allocation code it seems like the > thing you want to convey is "this is the largest group of registers > we'll allocate" and MAX_FB_WRITE_SIZE doesn't obviously do that. > > MAX_VGRF_SIZE? That seems better for a future where we want to > indirectly address a large block of registers. But we're not going to > want to register coalesce arrays. Maybe need two macros? > > Ken?
Yes, please. MAX_GRF_SIZE is pretty generic. MAX_MESSAGE_LENGTH, MAX_FB_WRITE_SIZE, or MAX_VGRF_SIZE would all be reasonable in my book.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
