https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88836
--- Comment #4 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org>
---
My preferred fix for this is now to finish off the secondary
(and ill-fated :-)) combine pass I posted at:
https://gcc.gnu.org/pipermail/gcc-patches/2019-November/534728.html
I've been doing this on a spare-time basis since then and hope
to have something to post in time for the end of stage 1.
That approach is more general in that it also fixes some missed
optimisations in the handling of the first-faulting load ACLE intrinsics.
