> > Actually, I'm almost 100% certain that there are lots of other strict > aliasing violations in the Mesa code. That's why we've always disabled it. > > More generally, IMO it's unrealistic to rely on strict aliasing for > optimization, because very few people really understand it (I'm not one > of them).
I personally think we should get past the, aliasing is hard, lets go shopping, I get a 30K code size reduction on r600_dri.so 6780217 367820 1991392 9139429 8b74e5 lib/gallium/r600_dri.so 6746389 367820 1991392 9105601 8af0c1 lib/gallium/r600_dri.so That isn't a small amount, and I'd think at least for Intel targetting atoms with small caches it might matter. However it might also not be that size once we fixed all the aliasing. Granted there might be other things that would get us that sort of reduction but I'm not sure what they are. Dave. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
