* config/i386/sse.md (round<mode>2_sfix): New expander.

[pf3gnuchains/gcc-fork.git] / gcc / config / i386 / sse.md

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 688b5be..147646b 100644 (file)
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -7501,6 +7501,16 @@
    (set_attr "prefix" "maybe_vex,orig,vex,maybe_vex,orig,orig")
    (set_attr "mode" "V2SF,TI,TI,TI,V4SF,V2SF")])
 
+(define_expand "vec_dupv4si"
+  [(set (match_operand:V4SI 0 "register_operand" "")
+       (vec_duplicate:V4SI
+         (match_operand:SI 1 "nonimmediate_operand" "")))]
+  "TARGET_SSE"
+{
+  if (!TARGET_AVX)
+    operands[1] = force_reg (V4SImode, operands[1]);
+})
+
 (define_insn "*vec_dupv4si"
   [(set (match_operand:V4SI 0 "register_operand"     "=x,x,x")
        (vec_duplicate:V4SI
@@ -7517,6 +7527,16 @@
    (set_attr "prefix" "maybe_vex,vex,orig")
    (set_attr "mode" "TI,V4SF,V4SF")])
 
+(define_expand "vec_dupv2di"
+  [(set (match_operand:V2DI 0 "register_operand" "")
+       (vec_duplicate:V2DI
+         (match_operand:DI 1 "nonimmediate_operand" "")))]
+  "TARGET_SSE"
+{
+  if (!TARGET_AVX)
+    operands[1] = force_reg (V2DImode, operands[1]);
+})
+
 (define_insn "*vec_dupv2di"
   [(set (match_operand:V2DI 0 "register_operand"     "=x,x,x,x")
        (vec_duplicate:V2DI
@@ -9882,6 +9902,45 @@
    (set_attr "prefix" "maybe_vex")
    (set_attr "mode" "<MODE>")])
 
+(define_expand "<sse4_1>_round<ssemodesuffix>_sfix<avxsizesuffix>"
+  [(match_operand:<sseintvecmode> 0 "register_operand" "")
+   (match_operand:VF1 1 "nonimmediate_operand" "")
+   (match_operand:SI 2 "const_0_to_15_operand" "")]
+  "TARGET_ROUND"
+{
+  rtx tmp = gen_reg_rtx (<MODE>mode);
+
+  emit_insn
+    (gen_<sse4_1>_round<ssemodesuffix><avxsizesuffix> (tmp, operands[1],
+                                                      operands[2]));
+  emit_insn
+    (gen_fix_trunc<mode><sseintvecmodelower>2 (operands[0], tmp));
+  DONE;
+})
+
+(define_expand "<sse4_1>_round<ssemodesuffix>_vec_pack_sfix<avxsizesuffix>"
+  [(match_operand:<ssepackfltmode> 0 "register_operand" "")
+   (match_operand:VF2 1 "nonimmediate_operand" "")
+   (match_operand:VF2 2 "nonimmediate_operand" "")
+   (match_operand:SI 3 "const_0_to_15_operand" "")]
+  "TARGET_ROUND"
+{
+  rtx tmp0, tmp1;
+
+  tmp0 = gen_reg_rtx (<MODE>mode);
+  tmp1 = gen_reg_rtx (<MODE>mode);
+
+  emit_insn
+    (gen_<sse4_1>_round<ssemodesuffix><avxsizesuffix> (tmp0, operands[1],
+                                                      operands[3]));
+  emit_insn
+    (gen_<sse4_1>_round<ssemodesuffix><avxsizesuffix> (tmp1, operands[2],
+                                                      operands[3]));
+  emit_insn
+    (gen_vec_pack_sfix_trunc_<mode> (operands[0], tmp0, tmp1));
+  DONE;
+})
+
 (define_insn "sse4_1_round<ssescalarmodesuffix>"
   [(set (match_operand:VF_128 0 "register_operand" "=x,x")
        (vec_merge:VF_128
@@ -9937,6 +9996,39 @@
   operands[5] = GEN_INT (ROUND_TRUNC);
 })
 
+(define_expand "round<mode>2_sfix"
+  [(match_operand:<sseintvecmode> 0 "register_operand" "")
+   (match_operand:VF1 1 "nonimmediate_operand" "")]
+  "TARGET_ROUND && !flag_trapping_math"
+{
+  rtx tmp = gen_reg_rtx (<MODE>mode);
+
+  emit_insn (gen_round<mode>2 (tmp, operands[1]));
+
+  emit_insn
+    (gen_fix_trunc<mode><sseintvecmodelower>2 (operands[0], tmp));
+  DONE;
+})
+
+(define_expand "round<mode>2_vec_pack_sfix"
+  [(match_operand:<ssepackfltmode> 0 "register_operand" "")
+   (match_operand:VF2 1 "nonimmediate_operand" "")
+   (match_operand:VF2 2 "nonimmediate_operand" "")]
+  "TARGET_ROUND && !flag_trapping_math"
+{
+  rtx tmp0, tmp1;
+
+  tmp0 = gen_reg_rtx (<MODE>mode);
+  tmp1 = gen_reg_rtx (<MODE>mode);
+
+  emit_insn (gen_round<mode>2 (tmp0, operands[1]));
+  emit_insn (gen_round<mode>2 (tmp1, operands[2]));
+
+  emit_insn
+    (gen_vec_pack_sfix_trunc_<mode> (operands[0], tmp0, tmp1));
+  DONE;
+})
+
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
 ;; Intel SSE4.2 string/text processing instructions
@@ -12065,6 +12157,10 @@
    && avx_vperm2f128_parallel (operands[3], <MODE>mode)"
 {
   int mask = avx_vperm2f128_parallel (operands[3], <MODE>mode) - 1;
+  if (mask == 0x12)
+    return "vinsert<i128>\t{$0, %x2, %1, %0|%0, %1, %x2, 0}";
+  if (mask == 0x20)
+    return "vinsert<i128>\t{$1, %x2, %1, %0|%0, %1, %x2, 1}";
   operands[3] = GEN_INT (mask);
   return "vperm2<i128>\t{%3, %2, %1, %0|%0, %1, %2, %3}";
 }