public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug tree-optimization/97043] New: latent wrong-code with SLP vectorization
@ 2020-09-14  9:01 rguenth at gcc dot gnu.org
  2020-09-14  9:16 ` [Bug tree-optimization/97043] " rguenth at gcc dot gnu.org
                   ` (7 more replies)
  0 siblings, 8 replies; 9+ messages in thread
From: rguenth at gcc dot gnu.org @ 2020-09-14  9:01 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97043

            Bug ID: 97043
           Summary: latent wrong-code with SLP vectorization
           Product: gcc
           Version: 10.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: rguenth at gcc dot gnu.org
  Target Milestone: ---

There's a latent wrong-code bug on the branches visible with the
gcc.dg/vect/pr81410.c testcase.  The issue is avoided on trunk
by means of delaying optimizing/validating of SLP permutations until
the vectorization factor is final whilst on branches we throw away
the permutation prematurely via

              if (!this_load_permuted
                  /* The load requires permutation when unrolling exposes
                     a gap either because the group is larger than the SLP
                     group-size or because there is a gap between the groups. 
*/
                  && (known_eq (unrolling_factor, 1U)
                      || (group_size == DR_GROUP_SIZE (first_stmt_info)
                          && DR_GROUP_GAP (first_stmt_info) == 0)))
                {
                  SLP_TREE_LOAD_PERMUTATION (load_node).release ();

because unrolling_factor is 1 but later is upped to 2 due to hybrid
SLP/non-SLP vectorization.  This causes us to run into the gap adjustment
code added by the PR81410 fix:

              /* With SLP permutation we load the gaps as well, without
                 we need to skip the gaps after we manage to fully load
                 all elements.  group_gap_adj is DR_GROUP_SIZE here.  */
              group_elt += nunits;
              if (maybe_ne (group_gap_adj, 0U)
                  && !slp_perm
                  && known_eq (group_elt, group_size - group_gap_adj))
                {
                  poly_wide_int bump_val
                    = (wi::to_wide (TYPE_SIZE_UNIT (elem_type))
                       * group_gap_adj);
                  tree bump = wide_int_to_tree (sizetype, bump_val);
                  dataref_ptr = bump_vector_ptr (dataref_ptr, ptr_incr, gsi,
                                                 stmt_info, bump);
                  group_elt = 0;
                }

which notes that we do not load the gaps.  Now, alignment analysis
correctly analyzes the load of x[] to be aligned due to the VF being 2:

      poly_uint64 vf = LOOP_VINFO_VECT_FACTOR (loop_vinfo);
      step_preserves_misalignment_p
        = multiple_p (DR_STEP_ALIGNMENT (dr_info->dr) * vf, vect_align_c);

but that assumes contiguous aligned loads.  The
VMAT_CONTIGUOUS-with-gap-without-permutation vectorization OTOH assumes that
each individual instance
of the group is aligned which it is not.  Trying to fix up there
would also require unaligned access support checking at analysis time
as well as possible cost adjustments.

^ permalink raw reply	[flat|nested] 9+ messages in thread
end of thread, other threads:[~2022-02-18  8:21 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-09-14  9:01 [Bug tree-optimization/97043] New: latent wrong-code with SLP vectorization rguenth at gcc dot gnu.org
2020-09-14  9:16 ` [Bug tree-optimization/97043] " rguenth at gcc dot gnu.org
2020-09-14  9:30 ` rguenth at gcc dot gnu.org
2020-09-14 13:22 ` cvs-commit at gcc dot gnu.org
2020-10-06 12:20 ` rguenth at gcc dot gnu.org
2021-02-03  8:26 ` rguenth at gcc dot gnu.org
2021-02-03  8:27 ` rguenth at gcc dot gnu.org
2021-02-08 10:43 ` rguenth at gcc dot gnu.org
2022-02-18  8:21 ` cvs-commit at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).