------- Comment #4 from jakub at gcc dot gnu dot org 2007-11-05 10:10 -------
There are two issues IMHO, one is that in this case the vectorizer should
create VECTOR_CST rather than CONSTRUCTOR, because it is really known to be
constant. That cures this ICE, though you are right that CONSTRUCTORs inside
of
VIEW_CONVERT_EXPR shouldn't be diagnosed as invalid, both CONSTRUCTORs with
empty initializer and all VECTOR_TYPE CONSTRUCTORs are left in GIMPLE by the
gimplifier.
--
jakub at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |jakub at gcc dot gnu dot org
|dot org |
Status|UNCONFIRMED |ASSIGNED
Ever Confirmed|0 |1
Last reconfirmed|0000-00-00 00:00:00 |2007-11-05 10:10:49
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33993
