[Bug rtl-optimization/123121] [13/14/15/16 regression] wrong code at -O3 with "-fno-ipa-cp -fschedule-insns" on x86_64-linux-gnu

[vmakarov at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123121

--- Comment #6 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
I've reproduced the bug.  It seems wrong LRA rematerialization.

There are several ways to fix it, e.g. through modification of
availability/liveness calculation of rematerialization candidates or through
constraining possible rematerialization candidates.

I am investigating what approach is better.  But I think I'll fix it on this
week.