https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66719
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed| |2015-07-02
Assignee|unassigned at gcc dot gnu.org |rguenth at gcc dot
gnu.org
Ever confirmed|0 |1
--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
/vol/gcc/src/hg/trunk/local/gcc/testsuite/gcc.dg/vect/bb-slp-32.c:10:11: note:
=== vect_slp_analyze_data_ref_dependences ===
/vol/gcc/src/hg/trunk/local/gcc/testsuite/gcc.dg/vect/bb-slp-32.c:10:11: note:
not vectorized: unsupported unaligned load.*p_2(D)
so it doesn't consider vectorizing this, the XFAIL was removed with r222849:
-/* { dg-final { scan-tree-dump "vectorization is not profitable" "slp2" {
xfail { vect_no_align && { ! vect_hw_misalign } } } } } */
+/* { dg-final { scan-tree-dump "vectorization is not profitable" "slp2" } } */
as fix for PR62283 where you reported the testcase now succeeding for
solaris-sparc (building the vector from scalars).
But then I have removed the alignment check in the SLP build machinery which
now no longer makes this path trigger. So we again need the XFAIL.
