On Tue, Jul 26, 2011 at 12:23 AM, Richard Guenther <richard.guent...@gmail.com> wrote: > Hmm. I suppose we'd need to see at the initial expand stage that the > move is going to be handled specially. For other strict-align targets > we end up with store/load-bit-field for unaligned accesses, so I suppose > SPU doesn't want to go down that path (via insv/extv)?
The problem is that almost all load/stores on spu are unaligned. So there will be less optimized done on them if we go down that path (it was tried at least twice and it produced worse code). Thanks, Andrew Pinski
