https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116818
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
Last reconfirmed| |2024-09-23
Status|UNCONFIRMED |ASSIGNED
Assignee|unassigned at gcc dot gnu.org |rguenth at gcc dot
gnu.org
--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
I think this is SLP missing the corresponding
/* As a last resort, trying using a gather load or scatter store.
??? Although the code can handle all group sizes correctly,
it probably isn't a win to use separate strided accesses based
on nearby locations. Or, even if it's a win over scalar code,
it might not be a win over vectorizing at a lower VF, if that
allows us to use contiguous accesses. */
if (*memory_access_type == VMAT_ELEMENTWISE
&& single_element_p
&& loop_vinfo
&& vect_use_strided_gather_scatters_p (stmt_info, loop_vinfo,
masked_p, gs_info))
*memory_access_type = VMAT_GATHER_SCATTER;
we can try moving that down to also catch SLP now that we set VMAT_ELEMENTWISE
in some cases. That fixes the testcase but I have no idea what fallout
is elsewhere.
I'll propose that to the CI.
