[tree-optimization/111721] VECT: Support SLP for MASK_LEN_GATHER_LOAD with dummy mask

[Juzhe-Zhong]

This patch fixes following FAILs for RVV:
FAIL: gcc.dg/vect/vect-gather-1.c -flto -ffat-lto-objects  scan-tree-dump vect 
"Loop contains only SLP stmts"
FAIL: gcc.dg/vect/vect-gather-1.c scan-tree-dump vect "Loop contains only SLP 
stmts"

Bootstrap on X86 and regtest passed.

Tested on aarch64 passed.

Ok for trunk ?

        PR tree-optimization/111721

gcc/ChangeLog:

        * tree-vect-slp.cc (vect_get_and_check_slp_defs): Support SLP for dummy 
mask -1.
        * tree-vect-stmts.cc (vectorizable_load): Ditto.

---
 gcc/tree-vect-slp.cc   | 14 ++++++++++++--
 gcc/tree-vect-stmts.cc |  8 +++++++-
 2 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/gcc/tree-vect-slp.cc b/gcc/tree-vect-slp.cc
index 43d742e3c92..23ca0318e31 100644
--- a/gcc/tree-vect-slp.cc
+++ b/gcc/tree-vect-slp.cc
@@ -756,8 +756,7 @@ vect_get_and_check_slp_defs (vec_info *vinfo, unsigned char 
swap,
        {
          tree type = TREE_TYPE (oprnd);
          dt = dts[i];
-         if ((dt == vect_constant_def
-              || dt == vect_external_def)
+         if (dt == vect_external_def
              && !GET_MODE_SIZE (vinfo->vector_mode).is_constant ()
              && (TREE_CODE (type) == BOOLEAN_TYPE
                  || !can_duplicate_and_interleave_p (vinfo, stmts.length (),
@@ -769,6 +768,17 @@ vect_get_and_check_slp_defs (vec_info *vinfo, unsigned 
char swap,
                                 "for variable-length SLP %T\n", oprnd);
              return -1;
            }
+         if (dt == vect_constant_def
+             && !GET_MODE_SIZE (vinfo->vector_mode).is_constant ()
+             && !can_duplicate_and_interleave_p (vinfo, stmts.length (), type))
+           {
+             if (dump_enabled_p ())
+               dump_printf_loc (MSG_MISSED_OPTIMIZATION, vect_location,
+                                "Build SLP failed: invalid type of def "
+                                "for variable-length SLP %T\n",
+                                oprnd);
+             return -1;
+           }
 
          /* For the swapping logic below force vect_reduction_def
             for the reduction op in a SLP reduction group.  */
diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
index 6ce4868d3e1..6c47121e158 100644
--- a/gcc/tree-vect-stmts.cc
+++ b/gcc/tree-vect-stmts.cc
@@ -9859,10 +9859,16 @@ vectorizable_load (vec_info *vinfo,
       mask_index = internal_fn_mask_index (ifn);
       if (mask_index >= 0 && slp_node)
        mask_index = vect_slp_child_index_for_operand (call, mask_index);
+      slp_tree slp_op = NULL;
       if (mask_index >= 0
          && !vect_check_scalar_mask (vinfo, stmt_info, slp_node, mask_index,
-                                     &mask, NULL, &mask_dt, &mask_vectype))
+                                     &mask, &slp_op, &mask_dt, &mask_vectype))
        return false;
+      /* MASK_LEN_GATHER_LOAD dummy mask -1 should always match the
+        MASK_VECTYPE.  */
+      if (mask_index >= 0 && slp_node && mask_dt == vect_constant_def
+         && !vect_maybe_update_slp_op_vectype (slp_op, mask_vectype))
+       gcc_unreachable ();
     }
 
   tree vectype = STMT_VINFO_VECTYPE (stmt_info);
-- 
2.36.3