https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98032
--- Comment #13 from Richard Biener <rguenth at gcc dot gnu.org> --- (In reply to Richard Biener from comment #7) > The 10 branch has > > while (BINFO_PRIMARY_P (binfo)) > /* BINFO_VIRTUALS in a primary base isn't accurate, find the derived > class that actually owns the vtable. */ > binfo = BINFO_INHERITANCE_CHAIN (binfo); > > in lookup_vfn_in_binfo and that would apply here. Was added with > g:554eb7d2e1ef5660d6a8e1c12ee1d751a70bbf31 for PR95719. The testcase from that rev doesn't ICE on the 8/9 branches nor does it pass its dump scanning. I'll backport it crippled (dump-scanning removed) and have added the testcase from this PR.
