[Bug lto/91375] [8/9/10 Regression] ICE on valid code in subbinfo_with_vtable_at_offset at ipa-devirt.c:2760 since r256685

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91375

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---

> 
> so I guess that flag_devirtualize is to blame since it may differ between
> CUs and type merging may chose the instance from a !flag_devirtualize one.
> A fix could be as simple as removing the !flag_devirtualize case.

Yes, the removal of '!flag_devirtualize' looks logical to me.