https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114353
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hubicka at gcc dot gnu.org
Last reconfirmed| |2024-03-15
Keywords| |ice-on-invalid-code
Ever confirmed|0 |1
Status|UNCONFIRMED |NEW
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
Yeah, LTO objects are not "portable" and since they are more-or-less
unstructured mismatches produce this kind of "errors" (and there's no way
to know for the compiler). Not exactly great engineering but this is how it is
"expected" to not work.
Much stricter "versioning" would be necessary to prevent this (maybe similar to
the checksum we use for PCH validation). I wonder what the C++ module support
does, it should face the very same issue.
