This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/59471] [4.9 Regression] ICE using vector extensions (non-top-level BIT_FIELD_REF, IMAGPART_EXPR or REALPART_EXPR)
- From: "jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 16 Dec 2013 15:08:17 +0000
- Subject: [Bug middle-end/59471] [4.9 Regression] ICE using vector extensions (non-top-level BIT_FIELD_REF, IMAGPART_EXPR or REALPART_EXPR)
- Auto-submitted: auto-generated
- References: <bug-59471-4 at http dot gcc dot gnu dot org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59471
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |jakub at gcc dot gnu.org,
| |jamborm at gcc dot gnu.org
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
This started failing when the assertion has been added:
http://gcc.gnu.org/ml/gcc-patches/2013-05/msg01265.html
There is no code to ensure that it won't happen though.
So, either the verification patch should be reverted and SRA adjusted to cope
with this, or something in gimple-expr.[ch] ? will need to be adjusted to
reject it (but e.g. making is_gimple_addressable more expensive might be a
problem).