https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122976
Martin Jambor <jamborm at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|unassigned at gcc dot gnu.org |jamborm at gcc dot
gnu.org
Status|NEW |ASSIGNED
--- Comment #6 from Martin Jambor <jamborm at gcc dot gnu.org> ---
With checking compiler, the testcase runs into normal intraprocedural SRA
verification failure until r16-3201-gee67004474d521 (Andrew Pinski:
forwprop: Copy prop aggregates into args).
The failing assert warns of mismatching offset, which looks fairly bad.
>From the very quick first glance at the dumps, the accesses stem from
"propagation across assignments" into something that looks like it was
created by IPA-SRA and then inlined. Perhaps another issue with
build_user_friendly_ref_for_offset. In any case, I will investigate what
was going on.
