vect: Recog mul_highpart pattern [PR100696]

This patch is to extend the existing pattern mulhs handlings
to cover normal multiply highpart pattern recognization, it
introduces one new internal function IFN_MULH for 1:1 map to
[su]mul_highpart optab.  Since it covers MULT_HIGHPART_EXPR
with optab support, i386 part change is to ensure it follows
the consistent costing path.

Bootstrapped & regtested on powerpc64le-linux-gnu P9,
x86_64-redhat-linux and aarch64-linux-gnu.

gcc/ChangeLog:

	PR tree-optimization/100696
	* internal-fn.c (first_commutative_argument): Add info for IFN_MULH.
	* internal-fn.def (IFN_MULH): New internal function.
	* tree-vect-patterns.c (vect_recog_mulhs_pattern): Add support to
	recog normal multiply highpart as IFN_MULH.
	* config/i386/i386.c (ix86_add_stmt_cost): Adjust for combined
	function CFN_MULH.

gcc/testsuite/ChangeLog:

	PR tree-optimization/100696
	* gcc.target/i386/pr100637-3w.c: Adjust for mul_highpart recog.

gcc/config/i386/i386.c

@@ -22568,6 +22568,9 @@ ix86_add_stmt_cost (class vec_info *vinfo, void *data, int count,
 				   mode == SFmode ? ix86_cost->fmass
 				   : ix86_cost->fmasd);
 	break;
+      case CFN_MULH:
+	stmt_cost = ix86_multiplication_cost (ix86_cost, mode);
+	break;
       default:
 	break;
       }

gcc/internal-fn.c

@@ -3703,6 +3703,7 @@ first_commutative_argument (internal_fn fn)
     case IFN_FNMS:
     case IFN_AVG_FLOOR:
     case IFN_AVG_CEIL:
+    case IFN_MULH:
     case IFN_MULHS:
     case IFN_MULHRS:
     case IFN_FMIN:

gcc/internal-fn.def

@@ -169,6 +169,8 @@ DEF_INTERNAL_SIGNED_OPTAB_FN (AVG_FLOOR, ECF_CONST | ECF_NOTHROW, first,
 DEF_INTERNAL_SIGNED_OPTAB_FN (AVG_CEIL, ECF_CONST | ECF_NOTHROW, first,
 			      savg_ceil, uavg_ceil, binary)
 
+DEF_INTERNAL_SIGNED_OPTAB_FN (MULH, ECF_CONST | ECF_NOTHROW, first,
+			      smul_highpart, umul_highpart, binary)
 DEF_INTERNAL_SIGNED_OPTAB_FN (MULHS, ECF_CONST | ECF_NOTHROW, first,
 			      smulhs, umulhs, binary)
 DEF_INTERNAL_SIGNED_OPTAB_FN (MULHRS, ECF_CONST | ECF_NOTHROW, first,

gcc/testsuite/gcc.target/i386/pr100637-3w.c

@@ -1,6 +1,6 @@
 /* PR target/100637 */
 /* { dg-do compile } */
-/* { dg-options "-O2 -ftree-vectorize -msse4" } */
+/* { dg-options "-O2 -ftree-vectorize -msse4 -fno-vect-cost-model" } */
 
 short r[2], a[2], b[2];
 unsigned short ur[2], ua[2], ub[2];
@@ -13,7 +13,7 @@ void mulh (void)
     r[i] = ((int) a[i] * b[i]) >> 16;
 }
 
-/* { dg-final { scan-assembler "pmulhw" { xfail *-*-* } } } */
+/* { dg-final { scan-assembler "pmulhw" } } */
 
 void mulhu (void)
 {
@@ -23,7 +23,7 @@ void mulhu (void)
     ur[i] = ((unsigned int) ua[i] * ub[i]) >> 16;
 }
 
-/* { dg-final { scan-assembler "pmulhuw" { xfail *-*-* } } } */
+/* { dg-final { scan-assembler "pmulhuw" } } */
 
 void mulhrs (void)
 {

gcc/tree-vect-patterns.c

@@ -1934,8 +1934,15 @@ vect_recog_over_widening_pattern (vec_info *vinfo,
 
    1) Multiply high with scaling
      TYPE res = ((TYPE) a * (TYPE) b) >> c;
+     Here, c is bitsize (TYPE) / 2 - 1.
+
    2) ... or also with rounding
      TYPE res = (((TYPE) a * (TYPE) b) >> d + 1) >> 1;
+     Here, d is bitsize (TYPE) / 2 - 2.
+
+   3) Normal multiply high
+     TYPE res = ((TYPE) a * (TYPE) b) >> e;
+     Here, e is bitsize (TYPE) / 2.
 
    where only the bottom half of res is used.  */
 
@@ -1980,7 +1987,6 @@ vect_recog_mulhs_pattern (vec_info *vinfo,
   stmt_vec_info mulh_stmt_info;
   tree scale_term;
   internal_fn ifn;
-  unsigned int expect_offset;
 
   /* Check for the presence of the rounding term.  */
   if (gimple_assign_rhs_code (rshift_input_stmt) == PLUS_EXPR)
@@ -2029,25 +2035,37 @@ vect_recog_mulhs_pattern (vec_info *vinfo,
 
       /* Get the scaling term.  */
       scale_term = gimple_assign_rhs2 (plus_input_stmt);
+      /* Check that the scaling factor is correct.  */
+      if (TREE_CODE (scale_term) != INTEGER_CST)
+	return NULL;
+
+      /* Check pattern 2).  */
+      if (wi::to_widest (scale_term) + target_precision + 2
+	  != TYPE_PRECISION (lhs_type))
+	return NULL;
 
-      expect_offset = target_precision + 2;
       ifn = IFN_MULHRS;
     }
   else
     {
       mulh_stmt_info = rshift_input_stmt_info;
       scale_term = gimple_assign_rhs2 (last_stmt);
+      /* Check that the scaling factor is correct.  */
+      if (TREE_CODE (scale_term) != INTEGER_CST)
+	return NULL;
 
-      expect_offset = target_precision + 1;
-      ifn = IFN_MULHS;
+      /* Check for pattern 1).  */
+      if (wi::to_widest (scale_term) + target_precision + 1
+	  == TYPE_PRECISION (lhs_type))
+	ifn = IFN_MULHS;
+      /* Check for pattern 3).  */
+      else if (wi::to_widest (scale_term) + target_precision
+	       == TYPE_PRECISION (lhs_type))
+	ifn = IFN_MULH;
+      else
+	return NULL;
     }
 
-  /* Check that the scaling factor is correct.  */
-  if (TREE_CODE (scale_term) != INTEGER_CST
-      || wi::to_widest (scale_term) + expect_offset
-	   != TYPE_PRECISION (lhs_type))
-    return NULL;
-
   /* Check whether the scaling input term can be seen as two widened
      inputs multiplied together.  */
   vect_unpromoted_value unprom_mult[2];