On Mon, Apr 11, 2016 at 12:15:10PM +0200, Olivier Hainque wrote: > > > On Mar 28, 2016, at 19:41 , Richard Henderson <r...@redhat.com> wrote: > > > > But I expect for stage4, the best solution is to strengthen the stack_tie > > pattern to block all memory. Early scheduling of the stack frame > > deallocation (a simple logic insn) can't really be that important to > > performance. > > Something like the attached patch, at least for next stage1 until the > more general issue is sorted out ?
It's a bit heavy; as a workaround, we may want this clobber in the stack deallocation in the epilogue, but not in all other places stack_tie is used. And for stage 1 we do not really want a workaround, we want to fix the actual problem ;-) Segher