On Tue, Mar 3, 2026 at 1:12 PM Richard Biener <[email protected]> wrote: > > On Thu, Feb 26, 2026 at 1:21 AM H.J. Lu <[email protected]> wrote: > > > > ix86_access_stack_p can be quite expensive. Cache the result and call it > > only if there are symbolic constant loads. This reduces the compile time > > of PR target/124165 test from 202 seconds to 55 seconds. > > I have my reservations about the overall architecture of > ix86_find_max_used_stack_alignment > but I verified the patch delivers the expected speedup and I agree > that this seems to be > a way forward without re-architecting ix86_find_max_used_stack_alignment, > which, > I think would likely not be appropriate at this point.
Thanks for your review! > > Richard. > > > gcc/ > > > > PR target/124165 > > * config/i386/i386-protos.h (symbolic_reference_mentioned_p): > > Change the argument type from rtx to const_rtx. > > * config/i386/i386.cc (symbolic_reference_mentioned_p): Likewise. > > (ix86_access_stack_p): Add 2 auto_bitmap[] arguments. Cache > > the register BB domination result. > > (ix86_symbolic_const_load_p_1): New. > > (ix86_symbolic_const_load_p): Likewise. > > (ix86_find_max_used_stack_alignment): If there is no symbolic > > constant load into the register, don't call ix86_access_stack_p. LGTM. Thanks, Uros.
