On Fri, Oct 25, 2019 at 10:40 AM Craig Blackmore <craig.blackm...@embecosm.com> wrote: > The sched2 pass undoes some of the addresses generated by the RISC-V > shorten_memrefs code size optimization (patch 1/2) and consequently increases > code size. This patch prevents sched-deps.c from changing an address if it is > expected to increase address cost.
This should be rewritten as a target hook, and then we can define the hook to do what we want for RISC-V. It isn't OK to make this change for other targets without testing them. If the default hook does nothing, then other targets won't be affected. Jim
