Support reduc_{plus,xor,and,ior}_scal_m for vector integer mode.

BB vectorizer relies on the backend support of
.REDUC_{PLUS,IOR,XOR,AND} to vectorize reduction.

gcc/ChangeLog:

	PR target/112325
	* config/i386/sse.md (reduc_<code>_scal_<mode>): New expander.
	(REDUC_ANY_LOGIC_MODE): New iterator.
	(REDUC_PLUS_MODE): Extend to VxHI/SI/DImode.
	(REDUC_SSE_PLUS_MODE): Ditto.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/pr112325-1.c: New test.
	* gcc.target/i386/pr112325-2.c: New test.

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index d250a6cb802f8b2554e38a0d843edc961d954f9b..f94a77d0b6d40e6d1ef84ad9396e6af0e46a8732 100644 (file)
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -3417,7 +3417,9 @@
 
 (define_mode_iterator REDUC_SSE_PLUS_MODE
  [(V2DF "TARGET_SSE") (V4SF "TARGET_SSE")
-  (V8HF "TARGET_AVX512FP16 && TARGET_AVX512VL")])
+  (V8HF "TARGET_AVX512FP16 && TARGET_AVX512VL")
+  (V8HI "TARGET_SSE2") (V4SI "TARGET_SSE2")
+  (V2DI "TARGET_SSE2")])
 
 (define_expand "reduc_plus_scal_<mode>"
  [(plus:REDUC_SSE_PLUS_MODE
@@ -3458,8 +3460,12 @@
   (V8DF "TARGET_AVX512F && TARGET_EVEX512")
   (V16SF "TARGET_AVX512F && TARGET_EVEX512")
   (V32HF "TARGET_AVX512FP16 && TARGET_AVX512VL && TARGET_EVEX512")
-  (V32QI "TARGET_AVX")
-  (V64QI "TARGET_AVX512F && TARGET_EVEX512")])
+  (V32QI "TARGET_AVX") (V16HI "TARGET_AVX")
+  (V8SI "TARGET_AVX")  (V4DI "TARGET_AVX")
+  (V64QI "TARGET_AVX512F && TARGET_EVEX512")
+  (V32HI "TARGET_AVX512F && TARGET_EVEX512")
+  (V16SI "TARGET_AVX512F && TARGET_EVEX512")
+  (V8DI "TARGET_AVX512F && TARGET_EVEX512")])
 
 (define_expand "reduc_plus_scal_<mode>"
  [(plus:REDUC_PLUS_MODE
@@ -3597,6 +3603,42 @@
    (set_attr "prefix" "evex")
    (set_attr "mode" "<MODE>")])
 
+(define_expand "reduc_<code>_scal_<mode>"
+ [(any_logic:VI_128
+    (match_operand:<ssescalarmode> 0 "register_operand")
+    (match_operand:VI_128 1 "register_operand"))]
+ "TARGET_SSE2"
+{
+  rtx tmp = gen_reg_rtx (<MODE>mode);
+  ix86_expand_reduc (gen_<code><mode>3, tmp, operands[1]);
+  emit_insn (gen_vec_extract<mode><ssescalarmodelower> (operands[0],
+                                                      tmp, const0_rtx));
+  DONE;
+})
+
+(define_mode_iterator REDUC_ANY_LOGIC_MODE
+ [(V32QI "TARGET_AVX") (V16HI "TARGET_AVX")
+  (V8SI "TARGET_AVX")  (V4DI "TARGET_AVX")
+  (V64QI "TARGET_AVX512F && TARGET_EVEX512")
+  (V32HI "TARGET_AVX512F && TARGET_EVEX512")
+  (V16SI "TARGET_AVX512F && TARGET_EVEX512")
+  (V8DI "TARGET_AVX512F && TARGET_EVEX512")])
+
+(define_expand "reduc_<code>_scal_<mode>"
+ [(any_logic:REDUC_ANY_LOGIC_MODE
+   (match_operand:<ssescalarmode> 0 "register_operand")
+   (match_operand:REDUC_ANY_LOGIC_MODE 1 "register_operand"))]
+ ""
+{
+  rtx tmp = gen_reg_rtx (<ssehalfvecmode>mode);
+  emit_insn (gen_vec_extract_hi_<mode> (tmp, operands[1]));
+  rtx tmp2 = gen_reg_rtx (<ssehalfvecmode>mode);
+  rtx tmp3 = gen_lowpart (<ssehalfvecmode>mode, operands[1]);
+  emit_insn (gen_<code><ssehalfvecmodelower>3 (tmp2, tmp, tmp3));
+  emit_insn (gen_reduc_<code>_scal_<ssehalfvecmodelower> (operands[0], tmp2));
+  DONE;
+})
+
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
 ;; Parallel floating point comparisons

diff --git a/gcc/testsuite/gcc.target/i386/pr112325-1.c b/gcc/testsuite/gcc.target/i386/pr112325-1.c
new file mode 100644 (file)
index 0000000..56e20c1
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr112325-1.c
@@ -0,0 +1,116 @@
+/* { dg-do compile } */
+/* { dg-options "-mavx512vl -mavx512bw -O2 -mtune=generic -mprefer-vector-width=512 -fdump-tree-slp2" } */
+/* { dg-final { scan-tree-dump-times ".REDUC_PLUS" 3 "slp2" } } */
+/* { dg-final { scan-tree-dump-times ".REDUC_IOR" 4 "slp2" } } */
+
+int
+__attribute__((noipa))
+plus_v4si (int* a)
+{
+  int sum = 0;
+  sum += a[0];
+  sum += a[1];
+  sum += a[2];
+  sum += a[3];
+  return sum;
+}
+
+short
+__attribute__((noipa))
+plus_v8hi (short* a)
+{
+  short sum = 0;
+  sum += a[0];
+  sum += a[1];
+  sum += a[2];
+  sum += a[3];
+  sum += a[4];
+  sum += a[5];
+  sum += a[6];
+  sum += a[7];
+  return sum;
+}
+
+long long
+__attribute__((noipa))
+plus_v8di (long long* a)
+{
+  long long sum = 0;
+  sum += a[0];
+  sum += a[1];
+  sum += a[2];
+  sum += a[3];
+  sum += a[4];
+  sum += a[5];
+  sum += a[6];
+  sum += a[7];
+  return sum;
+}
+
+int
+__attribute__((noipa))
+ior_v4si (int* a)
+{
+  int sum = 0;
+  sum |= a[0];
+  sum |= a[1];
+  sum |= a[2];
+  sum |= a[3];
+  return sum;
+}
+
+short
+__attribute__((noipa))
+ior_v8hi (short* a)
+{
+  short sum = 0;
+  sum |= a[0];
+  sum |= a[1];
+  sum |= a[2];
+  sum |= a[3];
+  sum |= a[4];
+  sum |= a[5];
+  sum |= a[6];
+  sum |= a[7];
+  return sum;
+}
+
+long long
+__attribute__((noipa))
+ior_v8di (long long* a)
+{
+  long long sum = 0;
+  sum |= a[0];
+  sum |= a[1];
+  sum |= a[2];
+  sum |= a[3];
+  sum |= a[4];
+  sum |= a[5];
+  sum |= a[6];
+  sum |= a[7];
+  return sum;
+}
+
+char
+__attribute__((noipa))
+ior_v16qi (char* a)
+{
+  char sum = 0;
+  sum |= a[0];
+  sum |= a[1];
+  sum |= a[2];
+  sum |= a[3];
+  sum |= a[4];
+  sum |= a[5];
+  sum |= a[6];
+  sum |= a[7];
+  sum |= a[8];
+  sum |= a[9];
+  sum |= a[10];
+  sum |= a[11];
+  sum |= a[12];
+  sum |= a[13];
+  sum |= a[14];
+  sum |= a[15];
+  return sum;
+}

diff --git a/gcc/testsuite/gcc.target/i386/pr112325-2.c b/gcc/testsuite/gcc.target/i386/pr112325-2.c
new file mode 100644 (file)
index 0000000..650006b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr112325-2.c
@@ -0,0 +1,38 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msse2" } */
+/* { dg-require-effective-target sse2 } */
+
+#include "sse2-check.h"
+#include "pr112325-1.c"
+
+static void
+sse2_test (void)
+{
+  int d[4] = { 3, 11, 22, 89};
+  short w[8] = { 3, 11, 22, 89, 4, 9, 13, 7};
+  char b[16] = { 3, 11, 22, 89, 4, 9, 13, 7, 2, 6, 5, 111, 163, 88, 11, 235};
+  long long q[8] = { 3, 11, 22, 89, 4, 9, 13, 7};
+
+  /* if (plus_v4si (d) != 125) */
+  /*   __builtin_abort (); */
+
+  /* if (plus_v8hi (w) != 158) */
+  /*   __builtin_abort (); */
+
+  /* if (plus_v8di (q) != 158) */
+  /*   __builtin_abort (); */
+
+  /* if (ior_v4si (d) != 95) */
+  /*   __builtin_abort (); */
+
+  /* if (ior_v8hi (w) != 95) */
+  /*   __builtin_abort (); */
+
+  /* if (ior_v16qi (b) != (char)255) */
+  /*   __builtin_abort (); */
+
+  if (ior_v8di (q) != 95)
+    __builtin_abort ();
+
+    return;
+}