https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65000
Richard Henderson <rth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
Assignee|unassigned at gcc dot gnu.org |rth at gcc dot gnu.org
--- Comment #10 from Richard Henderson <rth at gcc dot gnu.org> ---
Created attachment 34715
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=34715&action=edit
proposed patch
Yes, marking the regions of __builtin_eh_copy_values sounds reasonable
(and yes, it does fix the ICE here).
I think it would be valuable to know what's different about ubsan that
we haven't seen this same ICE with plain C++. I may spend a little bit
of time trying to reverse-engineer such a test case while this patch
goes through normal regression testing.
