https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121821
--- Comment #7 from Sam James <sjames at gcc dot gnu.org> --- As discussed on IRC, we suspect this is fixed by PR122867, but I can't verify it because with lto+pgo, I get PR122456, and with just pgo, I get an ada miscompilation (but hopefully it's something else already known, not going to worry for now anyway until we have other issue sorted). But I did get past the point where I *think* it failed to assemble before.
