[Bug rtl-optimization/123994] [16 Regression] RISC-V rv64gcb: ICE at -O2 during RTL pass: late_combine

[rsandifo at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123994

Richard Sandiford <rsandifo at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |NEW
           Assignee|rsandifo at gcc dot gnu.org        |unassigned at gcc dot gnu.org

--- Comment #13 from Richard Sandiford <rsandifo at gcc dot gnu.org> ---
(In reply to Jeffrey A. Law from comment #12)
> I'd tend to lean towards the simpler fix for -16 and the more complete for
> -17 given the possibility of triggering new combinations.  While I could
> argue that any regressions caused by such a change were just latent and
> waiting to bite us, safer seems better given we're in the latter half of Feb.
Yeah, sounds good to me.