https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80377
--- Comment #6 from Jeffrey Walton <noloader at gmail dot com> --- (In reply to David Malcolm from comment #5) > Likewise, I tried but failed to reproduce it. > > A hunch: given the "-march=native", is there a chance that this bug is > dependent on the precise CPU flags on the host? Thanks David. Well, that's a good question. I was testing on a Skylake, which is 6th gen. It has AVX and AVX2, but it does not have AVX-512. Is there a way I can test it? Another possibility is Fedora 25 has an older version of the autocomplete library that's producing it.
