From: Richard Biener <rguenther@suse.de>
Date: Thu, 11 Jul 2024 07:56:56 +0000 (+0200)
Subject: tree-optimization/115868 - ICE with .MASK_CALL in simdclone
X-Git-Tag: basepoints/gcc-16~7539
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=abf3964711f05b6858d9775c3595ec2b45483e14;p=thirdparty%2Fgcc.git

tree-optimization/115868 - ICE with .MASK_CALL in simdclone

The following adjusts mask recording which didn't take into account
that we can merge call arguments from two vectors like

  _50 = {vect_d_1.253_41, vect_d_1.254_43};
  _51 = VIEW_CONVERT_EXPR<unsigned char>(mask__19.257_49);
  _52 = (unsigned int) _51;
  _53 = _Z3bazd.simdclone.7 (_50, _52);
  _54 = BIT_FIELD_REF <_53, 256, 0>;
  _55 = BIT_FIELD_REF <_53, 256, 256>;

The testcase g++.dg/vect/pr68762-2.cc exercises this on x86_64 with
partial vector usage enabled and AVX512 support.

	PR tree-optimization/115868
	* tree-vect-stmts.cc (vectorizable_simd_clone_call): Correctly
	compute the number of mask copies required for vect_record_loop_mask.
---

diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
index 2e4d500d1f2..8530a98e6d6 100644
--- a/gcc/tree-vect-stmts.cc
+++ b/gcc/tree-vect-stmts.cc
@@ -4349,9 +4349,14 @@ vectorizable_simd_clone_call (vec_info *vinfo, stmt_vec_info stmt_info,
 	    case SIMD_CLONE_ARG_TYPE_MASK:
 	      if (loop_vinfo
 		  && LOOP_VINFO_CAN_USE_PARTIAL_VECTORS_P (loop_vinfo))
-		vect_record_loop_mask (loop_vinfo,
-				       &LOOP_VINFO_MASKS (loop_vinfo),
-				       ncopies, vectype, op);
+		{
+		  unsigned nmasks
+		    = exact_div (ncopies * bestn->simdclone->simdlen,
+				 TYPE_VECTOR_SUBPARTS (vectype)).to_constant ();
+		  vect_record_loop_mask (loop_vinfo,
+					 &LOOP_VINFO_MASKS (loop_vinfo),
+					 nmasks, vectype, op);
+		}
 
 	      break;
 	    }