[PATCH 11/41] i386: Emulate MMX mmx_eq/mmx_gt<mode>3 with SSE

[PATCH 11/41] i386: Emulate MMX mmx_eq/mmx_gt<mode>3 with SSE

[H.J. Lu]

Emulate MMX mmx_eq/mmx_gt<mode>3 with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_eq<mode>3): Also allow
	TARGET_MMX_WITH_SSE.
	(*mmx_eq<mode>3): Also allow TARGET_MMX_WITH_SSE.  Add SSE
	support.
	(mmx_gt<mode>3): Likewise.
---
 gcc/config/i386/mmx.md | 43 +++++++++++++++++++++++++-----------------
 1 file changed, 26 insertions(+), 17 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index b3df46dd563..aeebb4f5741 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1017,30 +1017,39 @@
 (define_expand "mmx_eq<mode>3"
   [(set (match_operand:MMXMODEI 0 "register_operand")
         (eq:MMXMODEI
-	  (match_operand:MMXMODEI 1 "nonimmediate_operand")
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand")))]
-  "TARGET_MMX"
+	  (match_operand:MMXMODEI 1 "register_mmxmem_operand")
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (EQ, <MODE>mode, operands);")
 
 (define_insn "*mmx_eq<mode>3"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
         (eq:MMXMODEI
-	  (match_operand:MMXMODEI 1 "nonimmediate_operand" "%0")
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX && ix86_binary_operator_ok (EQ, <MODE>mode, operands)"
-  "pcmpeq<mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxcmp")
-   (set_attr "mode" "DI")])
+	  (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (EQ, <MODE>mode, operands)"
+  "@
+   pcmpeq<mmxvecsize>\t{%2, %0|%0, %2}
+   pcmpeq<mmxvecsize>\t{%2, %0|%0, %2}
+   vpcmpeq<mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcmp,ssecmp,ssecmp")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_gt<mode>3"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
         (gt:MMXMODEI
-	  (match_operand:MMXMODEI 1 "register_operand" "0")
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX"
-  "pcmpgt<mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxcmp")
-   (set_attr "mode" "DI")])
+	  (match_operand:MMXMODEI 1 "register_operand" "0,0,Yv")
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   pcmpgt<mmxvecsize>\t{%2, %0|%0, %2}
+   pcmpgt<mmxvecsize>\t{%2, %0|%0, %2}
+   vpcmpgt<mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcmp,ssecmp,ssecmp")
+   (set_attr "mode" "DI,TI,TI")])
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
-- 
2.20.1

[PATCH 15/41] i386: Emulate MMX sse_cvtpi2ps with SSE

[H.J. Lu]

Emulate MMX sse_cvtpi2ps with SSE2 cvtdq2ps, preserving upper 64 bits of
destination XMM register.  Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/sse.md (sse_cvtpi2ps): Changed to
	define_insn_and_split.  Also allow TARGET_MMX_WITH_SSE.  Add
	SSE emulation.
---
 gcc/config/i386/sse.md | 64 ++++++++++++++++++++++++++++++++++++------
 1 file changed, 56 insertions(+), 8 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index dd3a3d9ba67..3135ce4eace 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -4569,16 +4569,64 @@
 ;;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
-(define_insn "sse_cvtpi2ps"
-  [(set (match_operand:V4SF 0 "register_operand" "=x")
+(define_insn_and_split "sse_cvtpi2ps"
+  [(set (match_operand:V4SF 0 "register_operand" "=x,x,Yv")
 	(vec_merge:V4SF
 	  (vec_duplicate:V4SF
-	    (float:V2SF (match_operand:V2SI 2 "nonimmediate_operand" "ym")))
-	  (match_operand:V4SF 1 "register_operand" "0")
-	  (const_int 3)))]
-  "TARGET_SSE"
-  "cvtpi2ps\t{%2, %0|%0, %2}"
-  [(set_attr "type" "ssecvt")
+	    (float:V2SF (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv")))
+	  (match_operand:V4SF 1 "register_operand" "0,0,Yv")
+	  (const_int 3)))
+   (clobber (match_scratch:V4SF 3 "=X,x,Yv"))]
+  "TARGET_SSE || TARGET_MMX_WITH_SSE"
+  "@
+   cvtpi2ps\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+{
+  rtx op2 = lowpart_subreg (V4SImode, operands[2],
+			    GET_MODE (operands[2]));
+  /* Generate SSE2 cvtdq2ps.  */
+  rtx insn = gen_floatv4siv4sf2 (operands[3], op2);
+  emit_insn (insn);
+
+  /* Merge operands[3] with operands[0].  */
+  rtx mask, op1;
+  if (TARGET_AVX)
+    {
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4, GEN_INT (0), GEN_INT (1),
+					  GEN_INT (6), GEN_INT (7)));
+      op1 = gen_rtx_VEC_CONCAT (V8SFmode, operands[3], operands[1]);
+      op2 = gen_rtx_VEC_SELECT (V4SFmode, op1, mask);
+      insn = gen_rtx_SET (operands[0], op2);
+    }
+  else
+    {
+      /* NB: SSE can only concatenate OP0 and OP3 to OP0.  */
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4, GEN_INT (2), GEN_INT (3),
+					  GEN_INT (4), GEN_INT (5)));
+      op1 = gen_rtx_VEC_CONCAT (V8SFmode, operands[0], operands[3]);
+      op2 = gen_rtx_VEC_SELECT (V4SFmode, op1, mask);
+      insn = gen_rtx_SET (operands[0], op2);
+      emit_insn (insn);
+
+      /* Swap bits 0:63 with bits 64:127.  */
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4, GEN_INT (2), GEN_INT (3),
+					  GEN_INT (0), GEN_INT (1)));
+      rtx dest = lowpart_subreg (V4SImode, operands[0],
+				 GET_MODE (operands[0]));
+      op1 = gen_rtx_VEC_SELECT (V4SImode, dest, mask);
+      insn = gen_rtx_SET (dest, op1);
+    }
+  emit_insn (insn);
+  DONE;
+}
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "ssecvt")
    (set_attr "mode" "V4SF")])
 
 (define_insn "sse_cvtps2pi"
-- 
2.20.1

[PATCH 04/41] i386: Emulate MMX plusminus/sat_plusminus with SSE

[H.J. Lu]

Emulate MMX plusminus/sat_plusminus with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/mmx.md (MMXMODEI8): Require TARGET_SSE2 for V1DI.
	(plusminus:mmx_<plusminus_insn><mode>3): Check
	TARGET_MMX_WITH_SSE.
	(sat_plusminus:mmx_<plusminus_insn><mode>3): Likewise.
	(<plusminus_insn><mode>3): New.
	(*mmx_<plusminus_insn><mode>3): Add SSE emulation.
	(*mmx_<plusminus_insn><mode>3): Likewise.
---
 gcc/config/i386/mmx.md | 59 +++++++++++++++++++++++++++---------------
 1 file changed, 38 insertions(+), 21 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 0aa793395fb..587e31b299e 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -45,7 +45,7 @@
 
 ;; 8 byte integral modes handled by MMX (and by extension, SSE)
 (define_mode_iterator MMXMODEI [V8QI V4HI V2SI])
-(define_mode_iterator MMXMODEI8 [V8QI V4HI V2SI V1DI])
+(define_mode_iterator MMXMODEI8 [V8QI V4HI V2SI (V1DI "TARGET_SSE2")])
 
 ;; All 8-byte vector modes handled by MMX
 (define_mode_iterator MMXMODE [V8QI V4HI V2SI V1DI V2SF])
@@ -663,39 +663,56 @@
 (define_expand "mmx_<plusminus_insn><mode>3"
   [(set (match_operand:MMXMODEI8 0 "register_operand")
 	(plusminus:MMXMODEI8
-	  (match_operand:MMXMODEI8 1 "nonimmediate_operand")
-	  (match_operand:MMXMODEI8 2 "nonimmediate_operand")))]
-  "TARGET_MMX || (TARGET_SSE2 && <MODE>mode == V1DImode)"
+	  (match_operand:MMXMODEI8 1 "register_mmxmem_operand")
+	  (match_operand:MMXMODEI8 2 "register_mmxmem_operand")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
+
+(define_expand "<plusminus_insn><mode>3"
+  [(set (match_operand:MMXMODEI 0 "register_operand")
+	(plusminus:MMXMODEI
+	  (match_operand:MMXMODEI 1 "register_operand")
+	  (match_operand:MMXMODEI 2 "register_operand")))]
+  "TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
 
 (define_insn "*mmx_<plusminus_insn><mode>3"
-  [(set (match_operand:MMXMODEI8 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI8 0 "register_operand" "=y,x,Yv")
         (plusminus:MMXMODEI8
-	  (match_operand:MMXMODEI8 1 "nonimmediate_operand" "<comm>0")
-	  (match_operand:MMXMODEI8 2 "nonimmediate_operand" "ym")))]
-  "(TARGET_MMX || (TARGET_SSE2 && <MODE>mode == V1DImode))
+	  (match_operand:MMXMODEI8 1 "register_mmxmem_operand" "<comm>0,0,Yv")
+	  (match_operand:MMXMODEI8 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
-  "p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+  "@
+   p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}
+   p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}
+   vp<plusminus_mnemonic><mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sseadd,sseadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_<plusminus_insn><mode>3"
   [(set (match_operand:MMXMODE12 0 "register_operand")
 	(sat_plusminus:MMXMODE12
-	  (match_operand:MMXMODE12 1 "nonimmediate_operand")
-	  (match_operand:MMXMODE12 2 "nonimmediate_operand")))]
-  "TARGET_MMX"
+	  (match_operand:MMXMODE12 1 "register_mmxmem_operand")
+	  (match_operand:MMXMODE12 2 "register_mmxmem_operand")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
 
 (define_insn "*mmx_<plusminus_insn><mode>3"
-  [(set (match_operand:MMXMODE12 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODE12 0 "register_operand" "=y,x,Yv")
         (sat_plusminus:MMXMODE12
-	  (match_operand:MMXMODE12 1 "nonimmediate_operand" "<comm>0")
-	  (match_operand:MMXMODE12 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
-  "p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+	  (match_operand:MMXMODE12 1 "register_mmxmem_operand" "<comm>0,0,Yv")
+	  (match_operand:MMXMODE12 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
+  "@
+   p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}
+   p<plusminus_mnemonic><mmxvecsize>\t{%2, %0|%0, %2}
+   vp<plusminus_mnemonic><mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sseadd,sseadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_mulv4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 03/41] i386: Emulate MMX punpcklXX/punpckhXX with SSE punpcklXX

[H.J. Lu]

Emulate MMX punpcklXX/punpckhXX with SSE punpcklXX.  For MMX punpckhXX,
move bits 64:127 to bits 0:63 in SSE register.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/i386-protos.h (ix86_split_mmx_punpck): New
	prototype.
	* config/i386/i386.c (ix86_split_mmx_punpck): New function.
	* config/i386/mmx.m (mmx_punpckhbw): Changed to
	define_insn_and_split to support SSE emulation.
	(mmx_punpcklbw): Likewise.
	(mmx_punpckhwd): Likewise.
	(mmx_punpcklwd): Likewise.
	(mmx_punpckhdq): Likewise.
	(mmx_punpckldq): Likewise.
---
 gcc/config/i386/i386-protos.h |   1 +
 gcc/config/i386/i386.c        |  77 +++++++++++++++++++
 gcc/config/i386/mmx.md        | 138 ++++++++++++++++++++++------------
 3 files changed, 168 insertions(+), 48 deletions(-)

diff --git a/gcc/config/i386/i386-protos.h b/gcc/config/i386/i386-protos.h
index a53b48438ec..37581837a32 100644
--- a/gcc/config/i386/i386-protos.h
+++ b/gcc/config/i386/i386-protos.h
@@ -204,6 +204,7 @@ extern rtx ix86_split_stack_guard (void);
 
 extern void ix86_move_vector_high_sse_to_mmx (rtx);
 extern void ix86_split_mmx_pack (rtx[], enum rtx_code);
+extern void ix86_split_mmx_punpck (rtx[], bool);
 
 #ifdef TREE_CODE
 extern void init_cumulative_args (CUMULATIVE_ARGS *, tree, rtx, tree, int);
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index d31b69d9a82..a76c17beece 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -20275,6 +20275,83 @@ ix86_split_mmx_pack (rtx operands[], enum rtx_code code)
   ix86_move_vector_high_sse_to_mmx (op0);
 }
 
+/* Split MMX punpcklXX/punpckhXX with SSE punpcklXX.  */
+
+void
+ix86_split_mmx_punpck (rtx operands[], bool high_p)
+{
+  rtx op0 = operands[0];
+  rtx op1 = operands[1];
+  rtx op2 = operands[2];
+  machine_mode mode = GET_MODE (op0);
+  rtx mask;
+  /* The corresponding SSE mode.  */
+  machine_mode sse_mode, double_sse_mode;
+
+  switch (mode)
+    {
+    case E_V8QImode:
+      sse_mode = V16QImode;
+      double_sse_mode = V32QImode;
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (16,
+					  GEN_INT (0), GEN_INT (16),
+					  GEN_INT (1), GEN_INT (17),
+					  GEN_INT (2), GEN_INT (18),
+					  GEN_INT (3), GEN_INT (19),
+					  GEN_INT (4), GEN_INT (20),
+					  GEN_INT (5), GEN_INT (21),
+					  GEN_INT (6), GEN_INT (22),
+					  GEN_INT (7), GEN_INT (23)));
+      break;
+
+    case E_V4HImode:
+      sse_mode = V8HImode;
+      double_sse_mode = V16HImode;
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (8,
+					  GEN_INT (0), GEN_INT (8),
+					  GEN_INT (1), GEN_INT (9),
+					  GEN_INT (2), GEN_INT (10),
+					  GEN_INT (3), GEN_INT (11)));
+      break;
+
+    case E_V2SImode:
+      sse_mode = V4SImode;
+      double_sse_mode = V8SImode;
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4,
+					  GEN_INT (0), GEN_INT (4),
+					  GEN_INT (1), GEN_INT (5)));
+      break;
+
+    default:
+      gcc_unreachable ();
+    }
+
+  /* Generate SSE punpcklXX.  */
+  rtx dest = lowpart_subreg (sse_mode, op0, GET_MODE (op0));
+  op1 = lowpart_subreg (sse_mode, op1, GET_MODE (op1));
+  op2 = lowpart_subreg (sse_mode, op2, GET_MODE (op2));
+
+  op1 = gen_rtx_VEC_CONCAT (double_sse_mode, op1, op2);
+  op2 = gen_rtx_VEC_SELECT (sse_mode, op1, mask);
+  rtx insn = gen_rtx_SET (dest, op2);
+  emit_insn (insn);
+
+  if (high_p)
+    {
+      /* Move bits 64:127 to bits 0:63.  */
+      mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4, GEN_INT (2), GEN_INT (3),
+					  GEN_INT (0), GEN_INT (0)));
+      dest = lowpart_subreg (V4SImode, dest, GET_MODE (dest));
+      op1 = gen_rtx_VEC_SELECT (V4SImode, dest, mask);
+      insn = gen_rtx_SET (dest, op1);
+      emit_insn (insn);
+    }
+}
+
 /* Helper function of ix86_fixup_binary_operands to canonicalize
    operand order.  Returns true if the operands should be swapped.  */
 
diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 63a390923b6..0aa793395fb 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1064,87 +1064,129 @@
    (set_attr "type" "mmxshft,sselog,sselog")
    (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpckhbw"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpckhbw"
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V8QI
 	  (vec_concat:V16QI
-	    (match_operand:V8QI 1 "register_operand" "0")
-	    (match_operand:V8QI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V8QI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv"))
           (parallel [(const_int 4) (const_int 12)
                      (const_int 5) (const_int 13)
                      (const_int 6) (const_int 14)
                      (const_int 7) (const_int 15)])))]
-  "TARGET_MMX"
-  "punpckhbw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckhbw\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, true); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpcklbw"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpcklbw"
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V8QI
 	  (vec_concat:V16QI
-	    (match_operand:V8QI 1 "register_operand" "0")
-	    (match_operand:V8QI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V8QI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv"))
           (parallel [(const_int 0) (const_int 8)
                      (const_int 1) (const_int 9)
                      (const_int 2) (const_int 10)
                      (const_int 3) (const_int 11)])))]
-  "TARGET_MMX"
-  "punpcklbw\t{%2, %0|%0, %k2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpcklbw\t{%2, %0|%0, %k2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, false); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpckhwd"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpckhwd"
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V4HI
 	  (vec_concat:V8HI
-	    (match_operand:V4HI 1 "register_operand" "0")
-	    (match_operand:V4HI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V4HI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv"))
           (parallel [(const_int 2) (const_int 6)
                      (const_int 3) (const_int 7)])))]
-  "TARGET_MMX"
-  "punpckhwd\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckhwd\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, true); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpcklwd"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpcklwd"
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V4HI
 	  (vec_concat:V8HI
-	    (match_operand:V4HI 1 "register_operand" "0")
-	    (match_operand:V4HI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V4HI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv"))
           (parallel [(const_int 0) (const_int 4)
                      (const_int 1) (const_int 5)])))]
-  "TARGET_MMX"
-  "punpcklwd\t{%2, %0|%0, %k2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpcklwd\t{%2, %0|%0, %k2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, false); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpckhdq"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpckhdq"
+  [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V2SI
 	  (vec_concat:V4SI
-	    (match_operand:V2SI 1 "register_operand" "0")
-	    (match_operand:V2SI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V2SI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv"))
 	  (parallel [(const_int 1)
 		     (const_int 3)])))]
-  "TARGET_MMX"
-  "punpckhdq\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckhdq\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, true); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_punpckldq"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_punpckldq"
+  [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv")
 	(vec_select:V2SI
 	  (vec_concat:V4SI
-	    (match_operand:V2SI 1 "register_operand" "0")
-	    (match_operand:V2SI 2 "nonimmediate_operand" "ym"))
+	    (match_operand:V2SI 1 "register_operand" "0,0,Yv")
+	    (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv"))
 	  (parallel [(const_int 0)
 		     (const_int 2)])))]
-  "TARGET_MMX"
-  "punpckldq\t{%2, %0|%0, %k2}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckldq\t{%2, %0|%0, %k2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_punpck (operands, false); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_pinsrw"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 06/41] i386: Emulate MMX smulv4hi3_highpart with SSE

[H.J. Lu]

Emulate MMX mulv4hi3 with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_smulv4hi3_highpart): Also allow
	TARGET_MMX_WITH_SSE.
	(*mmx_smulv4hi3_highpart): Also allow TARGET_MMX_WITH_SSE. Add
	SSE support.
---
 gcc/config/i386/mmx.md | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index fd0189eae60..01c80602b5b 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -748,27 +748,32 @@
 	  (lshiftrt:V4SI
 	    (mult:V4SI
 	      (sign_extend:V4SI
-		(match_operand:V4HI 1 "nonimmediate_operand"))
+		(match_operand:V4HI 1 "register_mmxmem_operand"))
 	      (sign_extend:V4SI
-		(match_operand:V4HI 2 "nonimmediate_operand")))
+		(match_operand:V4HI 2 "register_mmxmem_operand")))
 	    (const_int 16))))]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);")
 
 (define_insn "*mmx_smulv4hi3_highpart"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(truncate:V4HI
 	  (lshiftrt:V4SI
 	    (mult:V4SI
 	      (sign_extend:V4SI
-		(match_operand:V4HI 1 "nonimmediate_operand" "%0"))
+		(match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv"))
 	      (sign_extend:V4SI
-		(match_operand:V4HI 2 "nonimmediate_operand" "ym")))
+		(match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))
 	    (const_int 16))))]
-  "TARGET_MMX && ix86_binary_operator_ok (MULT, V4HImode, operands)"
-  "pmulhw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxmul")
-   (set_attr "mode" "DI")])
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (MULT, V4HImode, operands)"
+  "@
+   pmulhw\t{%2, %0|%0, %2}
+   pmulhw\t{%2, %0|%0, %2}
+   vpmulhw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxmul,ssemul,ssemul")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_umulv4hi3_highpart"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 09/41] i386: Emulate MMX <any_logic><mode>3 with SSE

[H.J. Lu]

Emulate MMX <any_logic><mode>3 with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/mmx.md (any_logic:mmx_<code><mode>3): Also allow
	TARGET_MMX_WITH_SSE.
	(any_logic:<code><mode>3): New.
	(any_logic:*mmx_<code><mode>3): Also allow TARGET_MMX_WITH_SSE.
	Add SSE support.
---
 gcc/config/i386/mmx.md | 33 +++++++++++++++++++++++----------
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 6af05a1881e..33f6c2aa774 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1061,20 +1061,33 @@
 (define_expand "mmx_<code><mode>3"
   [(set (match_operand:MMXMODEI 0 "register_operand")
 	(any_logic:MMXMODEI
-	  (match_operand:MMXMODEI 1 "nonimmediate_operand")
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand")))]
-  "TARGET_MMX"
+	  (match_operand:MMXMODEI 1 "register_mmxmem_operand")
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
+
+(define_expand "<code><mode>3"
+  [(set (match_operand:MMXMODEI 0 "register_operand")
+	(any_logic:MMXMODEI
+	  (match_operand:MMXMODEI 1 "register_operand")
+	  (match_operand:MMXMODEI 2 "register_operand")))]
+  "TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (<CODE>, <MODE>mode, operands);")
 
 (define_insn "*mmx_<code><mode>3"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
         (any_logic:MMXMODEI
-	  (match_operand:MMXMODEI 1 "nonimmediate_operand" "%0")
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
-  "p<logic>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+	  (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
+  "@
+   p<logic>\t{%2, %0|%0, %2}
+   p<logic>\t{%2, %0|%0, %2}
+   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
-- 
2.20.1

[PATCH 08/41] i386: Emulate MMX ashr<mode>3/<shift_insn><mode>3 with SSE

[H.J. Lu]

Emulate MMX ashr<mode>3/<shift_insn><mode>3 with SSE.  Only SSE register
source operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_ashr<mode>3): Also allow
	TARGET_MMX_WITH_SSE.  Add SSE emulation.
	(mmx_<shift_insn><mode>3): Likewise.
	(ashr<mode>3): New.
	(<shift_insn><mode>3): Likewise.
---
 gcc/config/i386/mmx.md | 50 ++++++++++++++++++++++++++++++------------
 1 file changed, 36 insertions(+), 14 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index fe746a487d1..6af05a1881e 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -959,32 +959,54 @@
    (set_attr "mode" "DI")])
 
 (define_insn "mmx_ashr<mode>3"
-  [(set (match_operand:MMXMODE24 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODE24 0 "register_operand" "=y,x,Yv")
         (ashiftrt:MMXMODE24
-	  (match_operand:MMXMODE24 1 "register_operand" "0")
-	  (match_operand:DI 2 "nonmemory_operand" "yN")))]
-  "TARGET_MMX"
-  "psra<mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
+	  (match_operand:MMXMODE24 1 "register_operand" "0,0,Yv")
+	  (match_operand:DI 2 "nonmemory_operand" "yN,xN,YvN")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   psra<mmxvecsize>\t{%2, %0|%0, %2}
+   psra<mmxvecsize>\t{%2, %0|%0, %2}
+   vpsra<mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sseishft,sseishft")
    (set (attr "length_immediate")
      (if_then_else (match_operand 2 "const_int_operand")
        (const_string "1")
        (const_string "0")))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
+
+(define_expand "ashr<mode>3"
+  [(set (match_operand:MMXMODE24 0 "register_operand")
+        (ashiftrt:MMXMODE24
+	  (match_operand:MMXMODE24 1 "register_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_MMX_WITH_SSE")
 
 (define_insn "mmx_<shift_insn><mode>3"
-  [(set (match_operand:MMXMODE248 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODE248 0 "register_operand" "=y,x,Yv")
         (any_lshift:MMXMODE248
-	  (match_operand:MMXMODE248 1 "register_operand" "0")
-	  (match_operand:DI 2 "nonmemory_operand" "yN")))]
-  "TARGET_MMX"
-  "p<vshift><mmxvecsize>\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
+	  (match_operand:MMXMODE248 1 "register_operand" "0,0,Yv")
+	  (match_operand:DI 2 "nonmemory_operand" "yN,xN,YvN")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   p<vshift><mmxvecsize>\t{%2, %0|%0, %2}
+   p<vshift><mmxvecsize>\t{%2, %0|%0, %2}
+   vp<vshift><mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sseishft,sseishft")
    (set (attr "length_immediate")
      (if_then_else (match_operand 2 "const_int_operand")
        (const_string "1")
        (const_string "0")))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
+
+(define_expand "<shift_insn><mode>3"
+  [(set (match_operand:MMXMODE248 0 "register_operand")
+        (any_lshift:MMXMODE248
+	  (match_operand:MMXMODE248 1 "register_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_MMX_WITH_SSE")
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
-- 
2.20.1

[PATCH 01/41] i386: Allow MMX register modes in SSE registers

[H.J. Lu]

In 64-bit mode, SSE2 can be used to emulate MMX instructions without
3DNOW.  We can use SSE2 to support MMX register modes.

	PR target/89021
	* config/i386/i386-c.c (ix86_target_macros_internal): Define
	__MMX_WITH_SSE__ for TARGET_MMX_WITH_SSE.
	* config/i386/i386.c (ix86_set_reg_reg_cost): Add support for
	TARGET_MMX_WITH_SSE with VALID_MMX_REG_MODE.
	(ix86_vector_mode_supported_p): Likewise.
	* config/i386/i386.h (TARGET_MMX_WITH_SSE): New.
---
 gcc/config/i386/i386-c.c | 2 ++
 gcc/config/i386/i386.c   | 5 +++--
 gcc/config/i386/i386.h   | 2 ++
 3 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/gcc/config/i386/i386-c.c b/gcc/config/i386/i386-c.c
index 5e7e46fcebe..213e1b56c6b 100644
--- a/gcc/config/i386/i386-c.c
+++ b/gcc/config/i386/i386-c.c
@@ -548,6 +548,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
     def_or_undef (parse_in, "__CLDEMOTE__");
   if (isa_flag2 & OPTION_MASK_ISA_PTWRITE)
     def_or_undef (parse_in, "__PTWRITE__");
+  if (TARGET_MMX_WITH_SSE)
+    def_or_undef (parse_in, "__MMX_WITH_SSE__");
   if (TARGET_IAMCU)
     {
       def_or_undef (parse_in, "__iamcu");
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 3e5f52175d2..7d7dd80930e 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -40490,7 +40490,8 @@ ix86_set_reg_reg_cost (machine_mode mode)
 	  || (TARGET_AVX && VALID_AVX256_REG_MODE (mode))
 	  || (TARGET_SSE2 && VALID_SSE2_REG_MODE (mode))
 	  || (TARGET_SSE && VALID_SSE_REG_MODE (mode))
-	  || (TARGET_MMX && VALID_MMX_REG_MODE (mode)))
+	  || ((TARGET_MMX || TARGET_MMX_WITH_SSE)
+	      && VALID_MMX_REG_MODE (mode)))
 	units = GET_MODE_SIZE (mode);
     }
 
@@ -44316,7 +44317,7 @@ ix86_vector_mode_supported_p (machine_mode mode)
     return true;
   if (TARGET_AVX512F && VALID_AVX512F_REG_MODE (mode))
     return true;
-  if (TARGET_MMX && VALID_MMX_REG_MODE (mode))
+  if ((TARGET_MMX ||TARGET_MMX_WITH_SSE) && VALID_MMX_REG_MODE (mode))
     return true;
   if (TARGET_3DNOW && VALID_MMX_REG_MODE_3DNOW (mode))
     return true;
diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h
index 4fd8bc40a34..91b233022c2 100644
--- a/gcc/config/i386/i386.h
+++ b/gcc/config/i386/i386.h
@@ -201,6 +201,8 @@ see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
 #define TARGET_16BIT	TARGET_CODE16
 #define TARGET_16BIT_P(x)	TARGET_CODE16_P(x)
 
+#define TARGET_MMX_WITH_SSE	(TARGET_64BIT && TARGET_SSE2)
+
 #include "config/vxworks-dummy.h"
 
 #include "config/i386/i386-opts.h"
-- 
2.20.1

[PATCH 12/41] i386: Emulate MMX vec_dupv2si with SSE

[H.J. Lu]

Emulate MMX vec_dupv2si with SSE.  Add the "Yw" constraint to allow
broadcast from integer register for AVX512BW with TARGET_AVX512VL.
Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/constraints.md (Yw): New constraint.
	* config/i386/mmx.md (*vec_dupv2si): Changed to
	define_insn_and_split and also allow TARGET_MMX_WITH_SSE to
	support SSE emulation.
---
 gcc/config/i386/constraints.md |  6 ++++++
 gcc/config/i386/mmx.md         | 24 +++++++++++++++++-------
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/constraints.md b/gcc/config/i386/constraints.md
index 16075b4acf3..c546b20d9dc 100644
--- a/gcc/config/i386/constraints.md
+++ b/gcc/config/i386/constraints.md
@@ -110,6 +110,8 @@
 ;;  v	any EVEX encodable SSE register for AVX512VL target,
 ;;	otherwise any SSE register
 ;;  h	EVEX encodable SSE register with number factor of four
+;;  w	any EVEX encodable SSE register for AVX512BW with TARGET_AVX512VL
+;;	target.
 
 (define_register_constraint "Yz" "TARGET_SSE ? SSE_FIRST_REG : NO_REGS"
  "First SSE register (@code{%xmm0}).")
@@ -146,6 +148,10 @@
  "TARGET_AVX512VL ? ALL_SSE_REGS : TARGET_SSE ? SSE_REGS : NO_REGS"
  "@internal For AVX512VL, any EVEX encodable SSE register (@code{%xmm0-%xmm31}), otherwise any SSE register.")
 
+(define_register_constraint "Yw"
+ "TARGET_AVX512BW && TARGET_AVX512VL ? ALL_SSE_REGS : NO_REGS"
+ "@internal Any EVEX encodable SSE register (@code{%xmm0-%xmm31}) for AVX512BW with TARGET_AVX512VL target.")
+
 ;; We use the B prefix to denote any number of internal operands:
 ;;  f  FLAGS_REG
 ;;  g  GOT memory operand.
diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index aeebb4f5741..b441f36dfc6 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1381,14 +1381,24 @@
    (set_attr "length_immediate" "1")
    (set_attr "mode" "DI")])
 
-(define_insn "*vec_dupv2si"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+(define_insn_and_split "*vec_dupv2si"
+  [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv,Yw")
 	(vec_duplicate:V2SI
-	  (match_operand:SI 1 "register_operand" "0")))]
-  "TARGET_MMX"
-  "punpckldq\t%0, %0"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+	  (match_operand:SI 1 "register_operand" "0,0,Yv,r")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckldq\t%0, %0
+   #
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(set (match_dup 0)
+	(vec_duplicate:V4SI (match_dup 1)))]
+  "operands[0] = lowpart_subreg (V4SImode, operands[0],
+				 GET_MODE (operands[0]));"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx,x64_avx")
+   (set_attr "type" "mmxcvt,ssemov,ssemov,ssemov")
+   (set_attr "mode" "DI,TI,TI,TI")])
 
 (define_insn "*mmx_concatv2si"
   [(set (match_operand:V2SI 0 "register_operand"     "=y,y")
-- 
2.20.1

[PATCH 17/41] i386: Emulate MMX mmx_pinsrw with SSE

[H.J. Lu]

Emulate MMX mmx_pinsrw with SSE.  Only SSE register destination operand
is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_pinsrw): Also check TARGET_MMX and
	TARGET_MMX_WITH_SSE.
	(*mmx_pinsrw): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 33 +++++++++++++++++++++++----------
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 28725f48282..dea2be1d8e2 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1282,32 +1282,45 @@
             (match_operand:SI 2 "nonimmediate_operand"))
 	  (match_operand:V4HI 1 "register_operand")
           (match_operand:SI 3 "const_0_to_3_operand")))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
 {
   operands[2] = gen_lowpart (HImode, operands[2]);
   operands[3] = GEN_INT (1 << INTVAL (operands[3]));
 })
 
 (define_insn "*mmx_pinsrw"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
         (vec_merge:V4HI
           (vec_duplicate:V4HI
-            (match_operand:HI 2 "nonimmediate_operand" "rm"))
-	  (match_operand:V4HI 1 "register_operand" "0")
+            (match_operand:HI 2 "nonimmediate_operand" "rm,rm,rm"))
+	  (match_operand:V4HI 1 "register_operand" "0,0,Yv")
           (match_operand:SI 3 "const_int_operand")))]
-  "(TARGET_SSE || TARGET_3DNOW_A)
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ((unsigned) exact_log2 (INTVAL (operands[3]))
        < GET_MODE_NUNITS (V4HImode))"
 {
   operands[3] = GEN_INT (exact_log2 (INTVAL (operands[3])));
-  if (MEM_P (operands[2]))
-    return "pinsrw\t{%3, %2, %0|%0, %2, %3}";
+  if (TARGET_MMX_WITH_SSE && TARGET_AVX)
+    {
+      if (MEM_P (operands[2]))
+	return "vpinsrw\t{%3, %2, %1, %0|%0, %1, %2, %3}";
+      else
+	return "vpinsrw\t{%3, %k2, %1, %0|%0, %1, %k2, %3}";
+    }
   else
-    return "pinsrw\t{%3, %k2, %0|%0, %k2, %3}";
+    {
+      if (MEM_P (operands[2]))
+	return "pinsrw\t{%3, %2, %0|%0, %2, %3}";
+      else
+	return "pinsrw\t{%3, %k2, %0|%0, %k2, %3}";
+    }
 }
-  [(set_attr "type" "mmxcvt")
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,sselog,sselog")
    (set_attr "length_immediate" "1")
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_pextrw"
   [(set (match_operand:SI 0 "register_operand" "=r,r")
-- 
2.20.1

[PATCH 14/41] i386: Emulate MMX sse_cvtps2pi/sse_cvttps2pi with SSE

[H.J. Lu]

Emulate MMX sse_cvtps2pi/sse_cvttps2pi with SSE.

	PR target/89021
	* config/i386/sse.md (sse_cvtps2pi): Add SSE emulation.
	(sse_cvttps2pi): Likewise.
---
 gcc/config/i386/sse.md | 30 ++++++++++++++++++------------
 1 file changed, 18 insertions(+), 12 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 30bf7e23122..dd3a3d9ba67 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -4582,26 +4582,32 @@
    (set_attr "mode" "V4SF")])
 
 (define_insn "sse_cvtps2pi"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+  [(set (match_operand:V2SI 0 "register_operand" "=y,Yv")
 	(vec_select:V2SI
-	  (unspec:V4SI [(match_operand:V4SF 1 "nonimmediate_operand" "xm")]
+	  (unspec:V4SI [(match_operand:V4SF 1 "register_mmxmem_operand" "xm,YvBm")]
 		       UNSPEC_FIX_NOTRUNC)
 	  (parallel [(const_int 0) (const_int 1)])))]
-  "TARGET_SSE"
-  "cvtps2pi\t{%1, %0|%0, %q1}"
-  [(set_attr "type" "ssecvt")
-   (set_attr "unit" "mmx")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSE"
+  "@
+   cvtps2pi\t{%1, %0|%0, %q1}
+   %vcvtps2dq\t{%1, %0|%0, %1}"
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "ssecvt")
+   (set_attr "unit" "mmx,*")
    (set_attr "mode" "DI")])
 
 (define_insn "sse_cvttps2pi"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+  [(set (match_operand:V2SI 0 "register_operand" "=y,Yv")
 	(vec_select:V2SI
-	  (fix:V4SI (match_operand:V4SF 1 "nonimmediate_operand" "xm"))
+	  (fix:V4SI (match_operand:V4SF 1 "register_mmxmem_operand" "xm,YvBm"))
 	  (parallel [(const_int 0) (const_int 1)])))]
-  "TARGET_SSE"
-  "cvttps2pi\t{%1, %0|%0, %q1}"
-  [(set_attr "type" "ssecvt")
-   (set_attr "unit" "mmx")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSE"
+  "@
+   cvttps2pi\t{%1, %0|%0, %q1}
+   %vcvttps2dq\t{%1, %0|%0, %1}"
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "ssecvt")
+   (set_attr "unit" "mmx,*")
    (set_attr "prefix_rep" "0")
    (set_attr "mode" "SF")])
 
-- 
2.20.1

[PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On x86-64, since __m64 is returned and passed in XMM registers, we can
emulate MMX intrinsics with SSE instructions. To support it, we added

 #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)

;; Define instruction set of MMX instructions
(define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
  (const_string "base"))

         (eq_attr "mmx_isa" "native")
           (symbol_ref "!TARGET_MMX_WITH_SSE")
         (eq_attr "mmx_isa" "x64")
           (symbol_ref "TARGET_MMX_WITH_SSE")
         (eq_attr "mmx_isa" "x64_avx")
           (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
         (eq_attr "mmx_isa" "x64_noavx")
           (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")

We added SSE emulation to MMX patterns and disabled MMX alternatives with
TARGET_MMX_WITH_SSE.

Most of MMX instructions have equivalent SSE versions and results of some
SSE versions need to be reshuffled to the right order for MMX.  Thee are
couple tricky cases:

1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
mask operand and handle unmapped bits 64:127 at memory address by
adjusting source and mask operands together with memory address.

2. MMX movntq is emulated with SSE2 DImode movnti, which is available
in 64-bit mode.

3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
SSE emulation must clear the bit 4 in the shuffle control mask.

4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
the upper 64 bits of destination XMM register.

Tests are also added to check each SSE emulation of MMX intrinsics.

There are no regressions on i686 and x86-64.  For x86-64, GCC is also
tested with

--with-arch=native --with-cpu=native

on AVX2 and AVX512F machines.

H.J. Lu (40):
  i386: Allow MMX register modes in SSE registers
  i386: Emulate MMX packsswb/packssdw/packuswb with SSE2
  i386: Emulate MMX punpcklXX/punpckhXX with SSE punpcklXX
  i386: Emulate MMX plusminus/sat_plusminus with SSE
  i386: Emulate MMX mulv4hi3 with SSE
  i386: Emulate MMX smulv4hi3_highpart with SSE
  i386: Emulate MMX mmx_pmaddwd with SSE
  i386: Emulate MMX ashr<mode>3/<shift_insn><mode>3 with SSE
  i386: Emulate MMX <any_logic><mode>3 with SSE
  i386: Emulate MMX mmx_andnot<mode>3 with SSE
  i386: Emulate MMX mmx_eq/mmx_gt<mode>3 with SSE
  i386: Emulate MMX vec_dupv2si with SSE
  i386: Emulate MMX pshufw with SSE
  i386: Emulate MMX sse_cvtps2pi/sse_cvttps2pi with SSE
  i386: Emulate MMX sse_cvtpi2ps with SSE
  i386: Emulate MMX mmx_pextrw with SSE
  i386: Emulate MMX mmx_pinsrw with SSE
  i386: Emulate MMX V4HI smaxmin/V8QI umaxmin with SSE
  i386: Emulate MMX mmx_pmovmskb with SSE
  i386: Emulate MMX mmx_umulv4hi3_highpart with SSE
  i386: Emulate MMX maskmovq with SSE2 maskmovdqu
  i386: Emulate MMX mmx_uavgv8qi3 with SSE
  i386: Emulate MMX mmx_uavgv4hi3 with SSE
  i386: Emulate MMX mmx_psadbw with SSE
  i386: Emulate MMX movntq with SSE2 movntidi
  i386: Emulate MMX umulv1siv1di3 with SSE2
  i386: Make _mm_empty () as NOP for TARGET_MMX_WITH_SSE
  i386: Emulate MMX ssse3_ph<plusminus_mnemonic>wv4hi3 with SSE
  i386: Emulate MMX ssse3_ph<plusminus_mnemonic>dv2si3 with SSE
  i386: Emulate MMX ssse3_pmaddubsw with SSE
  i386: Emulate MMX ssse3_pmulhrswv4hi3 with SSE
  i386: Emulate MMX pshufb with SSE version
  i386: Emulate MMX ssse3_psign<mode>3 with SSE
  i386: Emulate MMX ssse3_palignrdi with SSE
  i386: Emulate MMX abs<mode>2 with SSE
  i386: Allow MMXMODE moves with TARGET_MMX_WITH_SSE
  i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE
  i386: Allow MMX intrinsic emulation with SSE
  i386: Enable TM MMX intrinsics with SSE2
  i386: Add tests for MMX intrinsic emulations with SSE

Uros Bizjak (1):
  Prevent allocation of MMX registers with TARGET_MMX_WITH_SSE

 gcc/config/i386/constraints.md                |   6 +
 gcc/config/i386/i386-builtin.def              | 126 +--
 gcc/config/i386/i386-c.c                      |   2 +
 gcc/config/i386/i386-protos.h                 |   4 +
 gcc/config/i386/i386.c                        | 189 +++-
 gcc/config/i386/i386.h                        |   2 +
 gcc/config/i386/i386.md                       |  17 +
 gcc/config/i386/mmintrin.h                    |  12 +-
 gcc/config/i386/mmx.md                        | 984 ++++++++++++------
 gcc/config/i386/predicates.md                 |   7 +
 gcc/config/i386/sse.md                        | 359 +++++--
 gcc/config/i386/xmmintrin.h                   |  61 ++
 gcc/testsuite/gcc.target/i386/mmx-vals.h      |  77 ++
 gcc/testsuite/gcc.target/i386/pr82483-1.c     |   2 +-
 gcc/testsuite/gcc.target/i386/pr82483-2.c     |   2 +-
 gcc/testsuite/gcc.target/i386/sse2-mmx-10.c   |  43 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-11.c   |  39 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-12.c   |  42 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-13.c   |  40 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-14.c   |  31 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-15.c   |  36 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-16.c   |  40 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-17.c   |  51 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c  |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c  |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-2.c    |  12 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-20.c   |  12 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-21.c   |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-22.c   |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-3.c    |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-4.c    |   4 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-5.c    |  11 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-6.c    |  11 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-7.c    |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-8.c    |   4 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-9.c    |  79 ++
 .../gcc.target/i386/sse2-mmx-cvtpi2ps.c       |  43 +
 .../gcc.target/i386/sse2-mmx-cvtps2pi.c       |  36 +
 .../gcc.target/i386/sse2-mmx-cvttps2pi.c      |  36 +
 .../gcc.target/i386/sse2-mmx-maskmovq.c       |  99 ++
 .../gcc.target/i386/sse2-mmx-packssdw.c       |  52 +
 .../gcc.target/i386/sse2-mmx-packsswb.c       |  52 +
 .../gcc.target/i386/sse2-mmx-packuswb.c       |  52 +
 .../gcc.target/i386/sse2-mmx-paddb.c          |  48 +
 .../gcc.target/i386/sse2-mmx-paddd.c          |  48 +
 .../gcc.target/i386/sse2-mmx-paddq.c          |  43 +
 .../gcc.target/i386/sse2-mmx-paddsb.c         |  48 +
 .../gcc.target/i386/sse2-mmx-paddsw.c         |  48 +
 .../gcc.target/i386/sse2-mmx-paddusb.c        |  48 +
 .../gcc.target/i386/sse2-mmx-paddusw.c        |  48 +
 .../gcc.target/i386/sse2-mmx-paddw.c          |  48 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c |  44 +
 .../gcc.target/i386/sse2-mmx-pandn.c          |  44 +
 .../gcc.target/i386/sse2-mmx-pavgb.c          |  52 +
 .../gcc.target/i386/sse2-mmx-pavgw.c          |  52 +
 .../gcc.target/i386/sse2-mmx-pcmpeqb.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pcmpeqd.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pcmpeqw.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pcmpgtb.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pcmpgtd.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pcmpgtw.c        |  48 +
 .../gcc.target/i386/sse2-mmx-pextrw.c         |  59 ++
 .../gcc.target/i386/sse2-mmx-pinsrw.c         |  61 ++
 .../gcc.target/i386/sse2-mmx-pmaddwd.c        |  47 +
 .../gcc.target/i386/sse2-mmx-pmaxsw.c         |  48 +
 .../gcc.target/i386/sse2-mmx-pmaxub.c         |  48 +
 .../gcc.target/i386/sse2-mmx-pminsw.c         |  48 +
 .../gcc.target/i386/sse2-mmx-pminub.c         |  48 +
 .../gcc.target/i386/sse2-mmx-pmovmskb.c       |  46 +
 .../gcc.target/i386/sse2-mmx-pmulhuw.c        |  51 +
 .../gcc.target/i386/sse2-mmx-pmulhw.c         |  53 +
 .../gcc.target/i386/sse2-mmx-pmullw.c         |  52 +
 .../gcc.target/i386/sse2-mmx-pmuludq.c        |  47 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-por.c  |  44 +
 .../gcc.target/i386/sse2-mmx-psadbw.c         |  58 ++
 .../gcc.target/i386/sse2-mmx-pshufw.c         | 248 +++++
 .../gcc.target/i386/sse2-mmx-pslld.c          |  52 +
 .../gcc.target/i386/sse2-mmx-pslldi.c         | 153 +++
 .../gcc.target/i386/sse2-mmx-psllq.c          |  47 +
 .../gcc.target/i386/sse2-mmx-psllqi.c         | 245 +++++
 .../gcc.target/i386/sse2-mmx-psllw.c          |  52 +
 .../gcc.target/i386/sse2-mmx-psllwi.c         | 105 ++
 .../gcc.target/i386/sse2-mmx-psrad.c          |  52 +
 .../gcc.target/i386/sse2-mmx-psradi.c         | 153 +++
 .../gcc.target/i386/sse2-mmx-psraw.c          |  52 +
 .../gcc.target/i386/sse2-mmx-psrawi.c         | 105 ++
 .../gcc.target/i386/sse2-mmx-psrld.c          |  52 +
 .../gcc.target/i386/sse2-mmx-psrldi.c         | 153 +++
 .../gcc.target/i386/sse2-mmx-psrlq.c          |  47 +
 .../gcc.target/i386/sse2-mmx-psrlqi.c         | 245 +++++
 .../gcc.target/i386/sse2-mmx-psrlw.c          |  52 +
 .../gcc.target/i386/sse2-mmx-psrlwi.c         | 105 ++
 .../gcc.target/i386/sse2-mmx-psubb.c          |  48 +
 .../gcc.target/i386/sse2-mmx-psubd.c          |  48 +
 .../gcc.target/i386/sse2-mmx-psubq.c          |  43 +
 .../gcc.target/i386/sse2-mmx-psubusb.c        |  48 +
 .../gcc.target/i386/sse2-mmx-psubusw.c        |  48 +
 .../gcc.target/i386/sse2-mmx-psubw.c          |  48 +
 .../gcc.target/i386/sse2-mmx-punpckhbw.c      |  53 +
 .../gcc.target/i386/sse2-mmx-punpckhdq.c      |  47 +
 .../gcc.target/i386/sse2-mmx-punpckhwd.c      |  49 +
 .../gcc.target/i386/sse2-mmx-punpcklbw.c      |  53 +
 .../gcc.target/i386/sse2-mmx-punpckldq.c      |  47 +
 .../gcc.target/i386/sse2-mmx-punpcklwd.c      |  49 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c |  44 +
 gcc/testsuite/gcc.target/i386/sse2-mmx.c      |   1 -
 112 files changed, 6418 insertions(+), 493 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/mmx-vals.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-10.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-11.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-12.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-13.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-14.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-15.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-16.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-17.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-20.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-21.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-22.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-3.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-4.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-5.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-6.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-7.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-8.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-9.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-por.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c

-- 
2.20.1

[PATCH 16/41] i386: Emulate MMX mmx_pextrw with SSE

[H.J. Lu]

Emulate MMX mmx_pextrw with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_pextrw): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 09e78ac5f74..28725f48282 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1310,16 +1310,20 @@
    (set_attr "mode" "DI")])
 
 (define_insn "mmx_pextrw"
-  [(set (match_operand:SI 0 "register_operand" "=r")
+  [(set (match_operand:SI 0 "register_operand" "=r,r")
         (zero_extend:SI
 	  (vec_select:HI
-	    (match_operand:V4HI 1 "register_operand" "y")
-	    (parallel [(match_operand:SI 2 "const_0_to_3_operand" "n")]))))]
-  "TARGET_SSE || TARGET_3DNOW_A"
-  "pextrw\t{%2, %1, %0|%0, %1, %2}"
-  [(set_attr "type" "mmxcvt")
+	    (match_operand:V4HI 1 "register_operand" "y,Yv")
+	    (parallel [(match_operand:SI 2 "const_0_to_3_operand" "n,n")]))))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "@
+   pextrw\t{%2, %1, %0|%0, %1, %2}
+   %vpextrw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "mmxcvt,sselog1")
    (set_attr "length_immediate" "1")
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI")])
 
 (define_expand "mmx_pshufw"
   [(match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 10/41] i386: Emulate MMX mmx_andnot<mode>3 with SSE

[H.J. Lu]

Emulate MMX mmx_andnot<mode>3 with SSE.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_andnot<mode>3): Also allow
	TARGET_MMX_WITH_SSE.  Add SSE support.
---
 gcc/config/i386/mmx.md | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 33f6c2aa774..b3df46dd563 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1049,14 +1049,18 @@
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
 (define_insn "mmx_andnot<mode>3"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
 	(and:MMXMODEI
-	  (not:MMXMODEI (match_operand:MMXMODEI 1 "register_operand" "0"))
-	  (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX"
-  "pandn\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+	  (not:MMXMODEI (match_operand:MMXMODEI 1 "register_operand" "0,0,Yv"))
+	  (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   pandn\t{%2, %0|%0, %2}
+   pandn\t{%2, %0|%0, %2}
+   vpandn\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_<code><mode>3"
   [(set (match_operand:MMXMODEI 0 "register_operand")
-- 
2.20.1

[PATCH 28/41] i386: Emulate MMX ssse3_ph<plusminus_mnemonic>wv4hi3 with SSE

[H.J. Lu]

Emulate MMX ssse3_ph<plusminus_mnemonic>wv4hi3 with SSE by moving bits
64:95 to bits 32:63 in SSE register.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_ph<plusminus_mnemonic>wv4hi3):
	Changed to define_insn_and_split to support SSE emulation.
---
 gcc/config/i386/sse.md | 34 ++++++++++++++++++++++++++--------
 1 file changed, 26 insertions(+), 8 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 3135ce4eace..5f29f2c3595 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15243,13 +15243,13 @@
    (set_attr "prefix" "orig,vex")
    (set_attr "mode" "TI")])
 
-(define_insn "ssse3_ph<plusminus_mnemonic>wv4hi3"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+(define_insn_and_split "ssse3_ph<plusminus_mnemonic>wv4hi3"
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(vec_concat:V4HI
 	  (vec_concat:V2HI
 	    (ssse3_plusminus:HI
 	      (vec_select:HI
-		(match_operand:V4HI 1 "register_operand" "0")
+		(match_operand:V4HI 1 "register_operand" "0,0,Yv")
 		(parallel [(const_int 0)]))
 	      (vec_select:HI (match_dup 1) (parallel [(const_int 1)])))
 	    (ssse3_plusminus:HI
@@ -15258,19 +15258,37 @@
 	  (vec_concat:V2HI
 	    (ssse3_plusminus:HI
 	      (vec_select:HI
-		(match_operand:V4HI 2 "nonimmediate_operand" "ym")
+		(match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")
 		(parallel [(const_int 0)]))
 	      (vec_select:HI (match_dup 2) (parallel [(const_int 1)])))
 	    (ssse3_plusminus:HI
 	      (vec_select:HI (match_dup 2) (parallel [(const_int 2)]))
 	      (vec_select:HI (match_dup 2) (parallel [(const_int 3)]))))))]
-  "TARGET_SSSE3"
-  "ph<plusminus_mnemonic>w\t{%2, %0|%0, %2}"
-  [(set_attr "type" "sseiadd")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   ph<plusminus_mnemonic>w\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+{
+  /* Generate SSE version of the operation.  */
+  rtx op0 = lowpart_subreg (V8HImode, operands[0],
+			    GET_MODE (operands[0]));
+  rtx op1 = lowpart_subreg (V8HImode, operands[1],
+			    GET_MODE (operands[1]));
+  rtx op2 = lowpart_subreg (V8HImode, operands[2],
+			    GET_MODE (operands[2]));
+  emit_insn (gen_ssse3_ph<plusminus_mnemonic>wv8hi3 (op0, op1, op2));
+  ix86_move_vector_high_sse_to_mmx (op0);
+  DONE;
+}
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sseiadd")
    (set_attr "atom_unit" "complex")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "avx2_ph<plusminus_mnemonic>dv8si3"
   [(set (match_operand:V8SI 0 "register_operand" "=x")
-- 
2.20.1

[PATCH 07/41] i386: Emulate MMX mmx_pmaddwd with SSE

[H.J. Lu]

Emulate MMX pmaddwd with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_pmaddwd): Also allow TARGET_MMX_WITH_SSE.
	(*mmx_pmaddwd): Also allow TARGET_MMX_WITH_SSE.  Add SSE support.
---
 gcc/config/i386/mmx.md | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 01c80602b5b..fe746a487d1 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -810,11 +810,11 @@
 	  (mult:V2SI
 	    (sign_extend:V2SI
 	      (vec_select:V2HI
-		(match_operand:V4HI 1 "nonimmediate_operand")
+		(match_operand:V4HI 1 "register_mmxmem_operand")
 		(parallel [(const_int 0) (const_int 2)])))
 	    (sign_extend:V2SI
 	      (vec_select:V2HI
-		(match_operand:V4HI 2 "nonimmediate_operand")
+		(match_operand:V4HI 2 "register_mmxmem_operand")
 		(parallel [(const_int 0) (const_int 2)]))))
 	  (mult:V2SI
 	    (sign_extend:V2SI
@@ -823,20 +823,20 @@
 	    (sign_extend:V2SI
 	      (vec_select:V2HI (match_dup 2)
 		(parallel [(const_int 1) (const_int 3)]))))))]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);")
 
 (define_insn "*mmx_pmaddwd"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+  [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv")
         (plus:V2SI
 	  (mult:V2SI
 	    (sign_extend:V2SI
 	      (vec_select:V2HI
-		(match_operand:V4HI 1 "nonimmediate_operand" "%0")
+		(match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv")
 		(parallel [(const_int 0) (const_int 2)])))
 	    (sign_extend:V2SI
 	      (vec_select:V2HI
-		(match_operand:V4HI 2 "nonimmediate_operand" "ym")
+		(match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")
 		(parallel [(const_int 0) (const_int 2)]))))
 	  (mult:V2SI
 	    (sign_extend:V2SI
@@ -845,10 +845,15 @@
 	    (sign_extend:V2SI
 	      (vec_select:V2HI (match_dup 2)
 		(parallel [(const_int 1) (const_int 3)]))))))]
-  "TARGET_MMX && ix86_binary_operator_ok (MULT, V4HImode, operands)"
-  "pmaddwd\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxmul")
-   (set_attr "mode" "DI")])
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (MULT, V4HImode, operands)"
+  "@
+   pmaddwd\t{%2, %0|%0, %2}
+   pmaddwd\t{%2, %0|%0, %2}
+   vpmaddwd\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxmul,sseiadd,sseiadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_pmulhrwv4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 05/41] i386: Emulate MMX mulv4hi3 with SSE

[H.J. Lu]

Emulate MMX mulv4hi3 with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_mulv4hi3): Also allow
	TARGET_MMX_WITH_SSE.
	(mulv4hi3): New.
	(*mmx_mulv4hi3): Also allow TARGET_MMX_WITH_SSE.  Add SSE
	support.
---
 gcc/config/i386/mmx.md | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 587e31b299e..fd0189eae60 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -716,19 +716,31 @@
 
 (define_expand "mmx_mulv4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
-        (mult:V4HI (match_operand:V4HI 1 "nonimmediate_operand")
-		   (match_operand:V4HI 2 "nonimmediate_operand")))]
-  "TARGET_MMX"
+        (mult:V4HI (match_operand:V4HI 1 "register_mmxmem_operand")
+		   (match_operand:V4HI 2 "register_mmxmem_operand")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);")
+
+(define_expand "mulv4hi3"
+  [(set (match_operand:V4HI 0 "register_operand")
+        (mult:V4HI (match_operand:V4HI 1 "register_operand")
+		   (match_operand:V4HI 2 "register_operand")))]
+  "TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);")
 
 (define_insn "*mmx_mulv4hi3"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
-        (mult:V4HI (match_operand:V4HI 1 "nonimmediate_operand" "%0")
-		   (match_operand:V4HI 2 "nonimmediate_operand" "ym")))]
-  "TARGET_MMX && ix86_binary_operator_ok (MULT, V4HImode, operands)"
-  "pmullw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxmul")
-   (set_attr "mode" "DI")])
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
+        (mult:V4HI (match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv")
+		   (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && ix86_binary_operator_ok (MULT, V4HImode, operands)"
+  "@
+   pmullw\t{%2, %0|%0, %2}
+   pmullw\t{%2, %0|%0, %2}
+   vpmullw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxmul,ssemul,ssemul")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_smulv4hi3_highpart"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 31/41] i386: Emulate MMX ssse3_pmulhrswv4hi3 with SSE

[H.J. Lu]

Emulate MMX ssse3_pmulhrswv4hi3 with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_pmulhrswv4hi3): Require TARGET_MMX
	or TARGET_MMX_WITH_SSE.
	(*ssse3_pmulhrswv4hi3): Add SSE emulation.
---
 gcc/config/i386/sse.md | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index e8d9bec9766..b08a577d1e4 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15670,38 +15670,44 @@
 	      (lshiftrt:V4SI
 		(mult:V4SI
 		  (sign_extend:V4SI
-		    (match_operand:V4HI 1 "nonimmediate_operand"))
+		    (match_operand:V4HI 1 "register_mmxmem_operand"))
 		  (sign_extend:V4SI
-		    (match_operand:V4HI 2 "nonimmediate_operand")))
+		    (match_operand:V4HI 2 "register_mmxmem_operand")))
 		(const_int 14))
 	      (match_dup 3))
 	    (const_int 1))))]
-  "TARGET_SSSE3"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
 {
   operands[3] = CONST1_RTX(V4HImode);
   ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);
 })
 
 (define_insn "*ssse3_pmulhrswv4hi3"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(truncate:V4HI
 	  (lshiftrt:V4SI
 	    (plus:V4SI
 	      (lshiftrt:V4SI
 		(mult:V4SI
 		  (sign_extend:V4SI
-		    (match_operand:V4HI 1 "nonimmediate_operand" "%0"))
+		    (match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv"))
 		  (sign_extend:V4SI
-		    (match_operand:V4HI 2 "nonimmediate_operand" "ym")))
+		    (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))
 		(const_int 14))
 	      (match_operand:V4HI 3 "const1_operand"))
 	    (const_int 1))))]
-  "TARGET_SSSE3 && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
-  "pmulhrsw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "sseimul")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && TARGET_SSSE3
+   && !(MEM_P (operands[1]) && MEM_P (operands[2]))"
+  "@
+   pmulhrsw\t{%2, %0|%0, %2}
+   pmulhrsw\t{%2, %0|%0, %2}
+   vpmulhrsw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sseimul")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "<ssse3_avx2>_pshufb<mode>3<mask_name>"
   [(set (match_operand:VI1_AVX512 0 "register_operand" "=x,x,v")
-- 
2.20.1

[PATCH 27/41] i386: Make _mm_empty () as NOP for TARGET_MMX_WITH_SSE

[H.J. Lu]

With SSE emulation of MMX intrinsics, we should make _mm_empty () as NOP
for TARGET_MMX_WITH_SSE.

	PR target/89021
	* config/i386/mmx.md (mmx_<emms>): Renamed to ...
	(*mmx_<emms>): This.
	(mmx_<emms>): New expander.
---
 gcc/config/i386/mmx.md | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index f27513f7f2c..8b5fe3948e2 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1849,7 +1849,35 @@
   [(UNSPECV_EMMS "emms")
    (UNSPECV_FEMMS "femms")])
 
-(define_insn "mmx_<emms>"
+(define_expand "mmx_<emms>"
+  [(parallel
+    [(unspec_volatile [(const_int 0)] EMMS)
+      (clobber (reg:XF ST0_REG))
+      (clobber (reg:XF ST1_REG))
+      (clobber (reg:XF ST2_REG))
+      (clobber (reg:XF ST3_REG))
+      (clobber (reg:XF ST4_REG))
+      (clobber (reg:XF ST5_REG))
+      (clobber (reg:XF ST6_REG))
+      (clobber (reg:XF ST7_REG))
+      (clobber (reg:DI MM0_REG))
+      (clobber (reg:DI MM1_REG))
+      (clobber (reg:DI MM2_REG))
+      (clobber (reg:DI MM3_REG))
+      (clobber (reg:DI MM4_REG))
+      (clobber (reg:DI MM5_REG))
+      (clobber (reg:DI MM6_REG))
+      (clobber (reg:DI MM7_REG))])]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+{
+   if (TARGET_MMX_WITH_SSE)
+     {
+       emit_insn (gen_nop ());
+       DONE;
+     }
+})
+
+(define_insn "*mmx_<emms>"
   [(unspec_volatile [(const_int 0)] EMMS)
    (clobber (reg:XF ST0_REG))
    (clobber (reg:XF ST1_REG))
-- 
2.20.1

[PATCH 02/41] i386: Emulate MMX packsswb/packssdw/packuswb with SSE2

[H.J. Lu]

Emulate MMX packsswb/packssdw/packuswb with SSE packsswb/packssdw/packuswb
plus moving bits 64:95 to bits 32:63 in SSE register.  Only SSE register
source operand is allowed.

2019-02-08  H.J. Lu  <hongjiu.lu@intel.com>
	    Uros Bizjak  <ubizjak@gmail.com>

	PR target/89021
	* config/i386/i386-protos.h (ix86_move_vector_high_sse_to_mmx):
	New prototype.
	(ix86_split_mmx_pack): Likewise.
	* config/i386/i386.c (ix86_move_vector_high_sse_to_mmx): New
	function.
	(ix86_split_mmx_pack): Likewise.
	* config/i386/i386.md (mmx_isa): New.
	(enabled): Also check mmx_isa.
	* config/i386/mmx.md (any_s_truncate): New code iterator.
	(s_trunsuffix): New code attr.
	(mmx_packsswb): Removed.
	(mmx_packssdw): Likewise.
	(mmx_packuswb): Likewise.
	(mmx_pack<s_trunsuffix>swb): New define_insn_and_split to emulate
	MMX packsswb/packuswb with SSE2.
	(mmx_packssdw): Likewise.
	* config/i386/predicates.md (register_mmxmem_operand): New.
---
 gcc/config/i386/i386-protos.h |  3 ++
 gcc/config/i386/i386.c        | 54 ++++++++++++++++++++++++++++
 gcc/config/i386/i386.md       | 13 +++++++
 gcc/config/i386/mmx.md        | 67 +++++++++++++++++++----------------
 gcc/config/i386/predicates.md |  7 ++++
 5 files changed, 114 insertions(+), 30 deletions(-)

diff --git a/gcc/config/i386/i386-protos.h b/gcc/config/i386/i386-protos.h
index 27f5cc13abf..a53b48438ec 100644
--- a/gcc/config/i386/i386-protos.h
+++ b/gcc/config/i386/i386-protos.h
@@ -202,6 +202,9 @@ extern void ix86_expand_vecop_qihi (enum rtx_code, rtx, rtx, rtx);
 
 extern rtx ix86_split_stack_guard (void);
 
+extern void ix86_move_vector_high_sse_to_mmx (rtx);
+extern void ix86_split_mmx_pack (rtx[], enum rtx_code);
+
 #ifdef TREE_CODE
 extern void init_cumulative_args (CUMULATIVE_ARGS *, tree, rtx, tree, int);
 #endif	/* TREE_CODE  */
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 7d7dd80930e..d31b69d9a82 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -20221,6 +20221,60 @@ ix86_expand_vector_move_misalign (machine_mode mode, rtx operands[])
     gcc_unreachable ();
 }
 
+/* Move bits 64:95 to bits 32:63.  */
+
+void
+ix86_move_vector_high_sse_to_mmx (rtx op)
+{
+  rtx mask = gen_rtx_PARALLEL (VOIDmode,
+			       gen_rtvec (4, GEN_INT (0), GEN_INT (2),
+					  GEN_INT (0), GEN_INT (0)));
+  rtx dest = lowpart_subreg (V4SImode, op, GET_MODE (op));
+  op = gen_rtx_VEC_SELECT (V4SImode, dest, mask);
+  rtx insn = gen_rtx_SET (dest, op);
+  emit_insn (insn);
+}
+
+/* Split MMX pack with signed/unsigned saturation with SSE/SSE2.  */
+
+void
+ix86_split_mmx_pack (rtx operands[], enum rtx_code code)
+{
+  rtx op0 = operands[0];
+  rtx op1 = operands[1];
+  rtx op2 = operands[2];
+
+  machine_mode dmode = GET_MODE (op0);
+  machine_mode smode = GET_MODE (op1);
+  machine_mode inner_dmode = GET_MODE_INNER (dmode);
+  machine_mode inner_smode = GET_MODE_INNER (smode);
+
+  /* Get the corresponding SSE mode for destination.  */
+  int nunits = 16 / GET_MODE_SIZE (inner_dmode);
+  machine_mode sse_dmode = mode_for_vector (GET_MODE_INNER (dmode),
+					    nunits).require ();
+  machine_mode sse_half_dmode = mode_for_vector (GET_MODE_INNER (dmode),
+						 nunits / 2).require ();
+
+  /* Get the corresponding SSE mode for source.  */
+  nunits = 16 / GET_MODE_SIZE (inner_smode);
+  machine_mode sse_smode = mode_for_vector (GET_MODE_INNER (smode),
+					    nunits).require ();
+
+  /* Generate SSE pack with signed/unsigned saturation.  */
+  rtx dest = lowpart_subreg (sse_dmode, op0, GET_MODE (op0));
+  op1 = lowpart_subreg (sse_smode, op1, GET_MODE (op1));
+  op2 = lowpart_subreg (sse_smode, op2, GET_MODE (op2));
+
+  op1 = gen_rtx_fmt_e (code, sse_half_dmode, op1);
+  op2 = gen_rtx_fmt_e (code, sse_half_dmode, op2);
+  rtx insn = gen_rtx_SET (dest, gen_rtx_VEC_CONCAT (sse_dmode,
+						    op1, op2));
+  emit_insn (insn);
+
+  ix86_move_vector_high_sse_to_mmx (op0);
+}
+
 /* Helper function of ix86_fixup_binary_operands to canonicalize
    operand order.  Returns true if the operands should be swapped.  */
 
diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 240384917df..04ec0eeaa57 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -792,6 +792,10 @@
 		    avx512vl,noavx512vl,x64_avx512dq,x64_avx512bw"
   (const_string "base"))
 
+;; Define instruction set of MMX instructions
+(define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
+  (const_string "base"))
+
 (define_attr "enabled" ""
   (cond [(eq_attr "isa" "x64") (symbol_ref "TARGET_64BIT")
 	 (eq_attr "isa" "x64_sse2")
@@ -830,6 +834,15 @@
 	 (eq_attr "isa" "noavx512dq") (symbol_ref "!TARGET_AVX512DQ")
 	 (eq_attr "isa" "avx512vl") (symbol_ref "TARGET_AVX512VL")
 	 (eq_attr "isa" "noavx512vl") (symbol_ref "!TARGET_AVX512VL")
+
+	 (eq_attr "mmx_isa" "native")
+	   (symbol_ref "!TARGET_MMX_WITH_SSE")
+	 (eq_attr "mmx_isa" "x64")
+	   (symbol_ref "TARGET_MMX_WITH_SSE")
+	 (eq_attr "mmx_isa" "x64_avx")
+	   (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
+	 (eq_attr "mmx_isa" "x64_noavx")
+	   (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
 	]
 	(const_int 1)))
 
diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 57060b9d233..63a390923b6 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1021,41 +1021,48 @@
 ;;
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 
-(define_insn "mmx_packsswb"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
+;; Used in signed and unsigned truncations with saturation.
+(define_code_iterator any_s_truncate [ss_truncate us_truncate])
+;; Instruction suffix for truncations with saturation.
+(define_code_attr s_trunsuffix [(ss_truncate "s") (us_truncate "u")])
+
+(define_insn_and_split "mmx_pack<s_trunsuffix>swb"
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
 	(vec_concat:V8QI
-	  (ss_truncate:V4QI
-	    (match_operand:V4HI 1 "register_operand" "0"))
-	  (ss_truncate:V4QI
-	    (match_operand:V4HI 2 "nonimmediate_operand" "ym"))))]
-  "TARGET_MMX"
-  "packsswb\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
-   (set_attr "mode" "DI")])
+	  (any_s_truncate:V4QI
+	    (match_operand:V4HI 1 "register_operand" "0,0,Yv"))
+	  (any_s_truncate:V4QI
+	    (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv"))))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   pack<s_trunsuffix>swb\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_pack (operands, <any_s_truncate:CODE>); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
-(define_insn "mmx_packssdw"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+(define_insn_and_split "mmx_packssdw"
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(vec_concat:V4HI
 	  (ss_truncate:V2HI
-	    (match_operand:V2SI 1 "register_operand" "0"))
+	    (match_operand:V2SI 1 "register_operand" "0,0,Yv"))
 	  (ss_truncate:V2HI
-	    (match_operand:V2SI 2 "nonimmediate_operand" "ym"))))]
-  "TARGET_MMX"
-  "packssdw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
-   (set_attr "mode" "DI")])
-
-(define_insn "mmx_packuswb"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
-	(vec_concat:V8QI
-	  (us_truncate:V4QI
-	    (match_operand:V4HI 1 "register_operand" "0"))
-	  (us_truncate:V4QI
-	    (match_operand:V4HI 2 "nonimmediate_operand" "ym"))))]
-  "TARGET_MMX"
-  "packuswb\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
-   (set_attr "mode" "DI")])
+	    (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv"))))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   packssdw\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+  "ix86_split_mmx_pack (operands, SS_TRUNCATE); DONE;"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sselog,sselog")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_punpckhbw"
   [(set (match_operand:V8QI 0 "register_operand" "=y")
diff --git a/gcc/config/i386/predicates.md b/gcc/config/i386/predicates.md
index 99226e86436..f3c2f72de54 100644
--- a/gcc/config/i386/predicates.md
+++ b/gcc/config/i386/predicates.md
@@ -49,6 +49,13 @@
   (and (match_code "reg")
        (match_test "MMX_REGNO_P (REGNO (op))")))
 
+;; Match register operands, but include memory operands for
+;; !TARGET_MMX_WITH_SSE.
+(define_predicate "register_mmxmem_operand"
+  (ior (match_operand 0 "register_operand")
+       (and (not (match_test "TARGET_MMX_WITH_SSE"))
+	    (match_operand 0 "memory_operand"))))
+
 ;; True if the operand is an SSE register.
 (define_predicate "sse_reg_operand"
   (and (match_code "reg")
-- 
2.20.1

[PATCH 40/41] i386: Enable TM MMX intrinsics with SSE2

[H.J. Lu]

This pach enables TM MMX intrinsics with SSE2 when MMX is disabled.

	PR target/89021
	* config/i386/i386.c (bdesc_tm): Enable MMX intrinsics with
	SSE2.
---
 gcc/config/i386/i386.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 073a2534d1f..319a98f824a 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -31065,13 +31065,13 @@ static const struct builtin_description bdesc_##kind[] =		    \
    we're lazy.  Add casts to make them fit.  */
 static const struct builtin_description bdesc_tm[] =
 {
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_WM64", (enum ix86_builtins) BUILT_IN_TM_STORE_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_WaRM64", (enum ix86_builtins) BUILT_IN_TM_STORE_WAR_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_WaWM64", (enum ix86_builtins) BUILT_IN_TM_STORE_WAW_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_RM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_RaRM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RAR_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_RaWM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RAW_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_RfWM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RFW_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_WM64", (enum ix86_builtins) BUILT_IN_TM_STORE_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_WaRM64", (enum ix86_builtins) BUILT_IN_TM_STORE_WAR_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_WaWM64", (enum ix86_builtins) BUILT_IN_TM_STORE_WAW_M64, UNKNOWN, VOID_FTYPE_PV2SI_V2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_RM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_RaRM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RAR_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_RaWM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RAW_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_RfWM64", (enum ix86_builtins) BUILT_IN_TM_LOAD_RFW_M64, UNKNOWN, V2SI_FTYPE_PCV2SI },
 
   { OPTION_MASK_ISA_SSE, 0, CODE_FOR_nothing, "__builtin__ITM_WM128", (enum ix86_builtins) BUILT_IN_TM_STORE_M128, UNKNOWN, VOID_FTYPE_PV4SF_V4SF },
   { OPTION_MASK_ISA_SSE, 0, CODE_FOR_nothing, "__builtin__ITM_WaRM128", (enum ix86_builtins) BUILT_IN_TM_STORE_WAR_M128, UNKNOWN, VOID_FTYPE_PV4SF_V4SF },
@@ -31089,7 +31089,7 @@ static const struct builtin_description bdesc_tm[] =
   { OPTION_MASK_ISA_AVX, 0, CODE_FOR_nothing, "__builtin__ITM_RaWM256", (enum ix86_builtins) BUILT_IN_TM_LOAD_RAW_M256, UNKNOWN, V8SF_FTYPE_PCV8SF },
   { OPTION_MASK_ISA_AVX, 0, CODE_FOR_nothing, "__builtin__ITM_RfWM256", (enum ix86_builtins) BUILT_IN_TM_LOAD_RFW_M256, UNKNOWN, V8SF_FTYPE_PCV8SF },
 
-  { OPTION_MASK_ISA_MMX, 0, CODE_FOR_nothing, "__builtin__ITM_LM64", (enum ix86_builtins) BUILT_IN_TM_LOG_M64, UNKNOWN, VOID_FTYPE_PCVOID },
+  { OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_nothing, "__builtin__ITM_LM64", (enum ix86_builtins) BUILT_IN_TM_LOG_M64, UNKNOWN, VOID_FTYPE_PCVOID },
   { OPTION_MASK_ISA_SSE, 0, CODE_FOR_nothing, "__builtin__ITM_LM128", (enum ix86_builtins) BUILT_IN_TM_LOG_M128, UNKNOWN, VOID_FTYPE_PCVOID },
   { OPTION_MASK_ISA_AVX, 0, CODE_FOR_nothing, "__builtin__ITM_LM256", (enum ix86_builtins) BUILT_IN_TM_LOG_M256, UNKNOWN, VOID_FTYPE_PCVOID },
 };
-- 
2.20.1

[PATCH 13/41] i386: Emulate MMX pshufw with SSE

[H.J. Lu]

Emulate MMX pshufw with SSE.  Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_pshufw): Also check TARGET_MMX and
	TARGET_MMX_WITH_SSE.
	(mmx_pshufw_1): Add SSE emulation.
	(*vec_dupv4hi): Changed to define_insn_and_split and also allow
	TARGET_MMX_WITH_SSE to support SSE emulation.
---
 gcc/config/i386/mmx.md | 81 +++++++++++++++++++++++++++++++++---------
 1 file changed, 65 insertions(+), 16 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index b441f36dfc6..09e78ac5f74 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1323,9 +1323,10 @@
 
 (define_expand "mmx_pshufw"
   [(match_operand:V4HI 0 "register_operand")
-   (match_operand:V4HI 1 "nonimmediate_operand")
+   (match_operand:V4HI 1 "register_mmxmem_operand")
    (match_operand:SI 2 "const_int_operand")]
-  "TARGET_SSE || TARGET_3DNOW_A"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
 {
   int mask = INTVAL (operands[2]);
   emit_insn (gen_mmx_pshufw_1 (operands[0], operands[1],
@@ -1337,14 +1338,15 @@
 })
 
 (define_insn "mmx_pshufw_1"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,Yv")
         (vec_select:V4HI
-          (match_operand:V4HI 1 "nonimmediate_operand" "ym")
+          (match_operand:V4HI 1 "register_mmxmem_operand" "ym,Yv")
           (parallel [(match_operand 2 "const_0_to_3_operand")
                      (match_operand 3 "const_0_to_3_operand")
                      (match_operand 4 "const_0_to_3_operand")
                      (match_operand 5 "const_0_to_3_operand")])))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
 {
   int mask = 0;
   mask |= INTVAL (operands[2]) << 0;
@@ -1353,11 +1355,20 @@
   mask |= INTVAL (operands[5]) << 6;
   operands[2] = GEN_INT (mask);
 
-  return "pshufw\t{%2, %1, %0|%0, %1, %2}";
+  switch (which_alternative)
+    {
+    case 0:
+      return "pshufw\t{%2, %1, %0|%0, %1, %2}";
+    case 1:
+      return "%vpshuflw\t{%2, %1, %0|%0, %1, %2}";
+    default:
+      gcc_unreachable ();
+    }
 }
-  [(set_attr "type" "mmxcvt")
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "mmxcvt,sselog")
    (set_attr "length_immediate" "1")
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI")])
 
 (define_insn "mmx_pswapdv2si2"
   [(set (match_operand:V2SI 0 "register_operand" "=y")
@@ -1370,16 +1381,54 @@
    (set_attr "prefix_extra" "1")
    (set_attr "mode" "DI")])
 
-(define_insn "*vec_dupv4hi"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+(define_insn_and_split "*vec_dupv4hi"
+  [(set (match_operand:V4HI 0 "register_operand" "=y,Yv,Yw")
 	(vec_duplicate:V4HI
 	  (truncate:HI
-	    (match_operand:SI 1 "register_operand" "0"))))]
-  "TARGET_SSE || TARGET_3DNOW_A"
-  "pshufw\t{$0, %0, %0|%0, %0, 0}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "length_immediate" "1")
-   (set_attr "mode" "DI")])
+	    (match_operand:SI 1 "register_operand" "0,Yv,r"))))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "@
+   pshufw\t{$0, %0, %0|%0, %0, 0}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+{
+  rtx op;
+  operands[0] = lowpart_subreg (V8HImode, operands[0],
+				GET_MODE (operands[0]));
+  if (TARGET_AVX2)
+    {
+      operands[1] = lowpart_subreg (HImode, operands[1],
+				    GET_MODE (operands[1]));
+      op = gen_rtx_VEC_DUPLICATE (V8HImode, operands[1]);
+    }
+  else
+    {
+      operands[1] = lowpart_subreg (V8HImode, operands[1],
+				    GET_MODE (operands[1]));
+      rtx mask = gen_rtx_PARALLEL (VOIDmode,
+				   gen_rtvec (8,
+					      GEN_INT (0),
+					      GEN_INT (0),
+					      GEN_INT (0),
+					      GEN_INT (0),
+					      GEN_INT (4),
+					      GEN_INT (5),
+					      GEN_INT (6),
+					      GEN_INT (7)));
+
+      op = gen_rtx_VEC_SELECT (V8HImode, operands[1], mask);
+    }
+  rtx insn = gen_rtx_SET (operands[0], op);
+  emit_insn (insn);
+  DONE;
+}
+  [(set_attr "mmx_isa" "native,x64,x64_avx")
+   (set_attr "type" "mmxcvt,sselog1,ssemov")
+   (set_attr "length_immediate" "1,1,0")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn_and_split "*vec_dupv2si"
   [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv,Yw")
-- 
2.20.1

[PATCH 33/41] i386: Emulate MMX ssse3_psign<mode>3 with SSE

[H.J. Lu]

Emulate MMX ssse3_psign<mode>3 with SSE.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_psign<mode>3): Add SSE emulation.
---
 gcc/config/i386/sse.md | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 79b35d95424..1d90af0a4b0 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15786,17 +15786,21 @@
    (set_attr "mode" "<sseinsnmode>")])
 
 (define_insn "ssse3_psign<mode>3"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
 	(unspec:MMXMODEI
-	  [(match_operand:MMXMODEI 1 "register_operand" "0")
-	   (match_operand:MMXMODEI 2 "nonimmediate_operand" "ym")]
+	  [(match_operand:MMXMODEI 1 "register_operand" "0,0,Yv")
+	   (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")]
 	  UNSPEC_PSIGN))]
-  "TARGET_SSSE3"
-  "psign<mmxvecsize>\t{%2, %0|%0, %2}";
-  [(set_attr "type" "sselog1")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   psign<mmxvecsize>\t{%2, %0|%0, %2}
+   psign<mmxvecsize>\t{%2, %0|%0, %2}
+   vpsign<mmxvecsize>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sselog1")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "<ssse3_avx2>_palignr<mode>_mask"
   [(set (match_operand:VI1_AVX512 0 "register_operand" "=v")
-- 
2.20.1

[PATCH 25/41] i386: Emulate MMX movntq with SSE2 movntidi

[H.J. Lu]

Emulate MMX movntq with SSE2 movntidi.  Only register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (sse_movntq): Add SSE2 emulation.
---
 gcc/config/i386/mmx.md | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 098e41e19c3..b06f0af984a 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -214,12 +214,16 @@
 })
 
 (define_insn "sse_movntq"
-  [(set (match_operand:DI 0 "memory_operand" "=m")
-	(unspec:DI [(match_operand:DI 1 "register_operand" "y")]
+  [(set (match_operand:DI 0 "memory_operand" "=m,m")
+	(unspec:DI [(match_operand:DI 1 "register_operand" "y,r")]
 		   UNSPEC_MOVNTQ))]
-  "TARGET_SSE || TARGET_3DNOW_A"
-  "movntq\t{%1, %0|%0, %1}"
-  [(set_attr "type" "mmxmov")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "@
+   movntq\t{%1, %0|%0, %1}
+   movnti\t{%1, %0|%0, %1}"
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "mmxmov,ssemov")
    (set_attr "mode" "DI")])
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-- 
2.20.1

[PATCH 41/41] i386: Add tests for MMX intrinsic emulations with SSE

[H.J. Lu]

Test MMX intrinsics with -msse2 in 32-bit mode and -msse2 -mno-mmx in
64-bit mode.

	PR target/89021
	* gcc.target/i386/mmx-vals.h: New file.
	* gcc.target/i386/sse2-mmx-2.c: Likewise.
	* gcc.target/i386/sse2-mmx-3.c: Likewise.
	* gcc.target/i386/sse2-mmx-4.c: Likewise.
	* gcc.target/i386/sse2-mmx-5.c: Likewise.
	* gcc.target/i386/sse2-mmx-6.c: Likewise.
	* gcc.target/i386/sse2-mmx-7.c: Likewise.
	* gcc.target/i386/sse2-mmx-8.c: Likewise.
	* gcc.target/i386/sse2-mmx-9.c: Likewise.
	* gcc.target/i386/sse2-mmx-10.c: Likewise.
	* gcc.target/i386/sse2-mmx-11.c: Likewise.
	* gcc.target/i386/sse2-mmx-12.c: Likewise.
	* gcc.target/i386/sse2-mmx-13.c: Likewise.
	* gcc.target/i386/sse2-mmx-14.c: Likewise.
	* gcc.target/i386/sse2-mmx-15.c: Likewise.
	* gcc.target/i386/sse2-mmx-16.c: Likewise.
	* gcc.target/i386/sse2-mmx-17.c: Likewise.
	* gcc.target/i386/sse2-mmx-18a.c: Likewise.
	* gcc.target/i386/sse2-mmx-18b.c: Likewise.
	* gcc.target/i386/sse2-mmx-18c.c: Likewise.
	* gcc.target/i386/sse2-mmx-19a.c: Likewise.
	* gcc.target/i386/sse2-mmx-18b.c: Likewise.
	* gcc.target/i386/sse2-mmx-19c.c: Likewise.
	* gcc.target/i386/sse2-mmx-19d.c: Likewise.
	* gcc.target/i386/sse2-mmx-19e.c: Likewise.
	* gcc.target/i386/sse2-mmx-20.c: Likewise.
	* gcc.target/i386/sse2-mmx-21.c: Likewise.
	* gcc.target/i386/sse2-mmx-22.c: Likewise.
	* gcc.target/i386/sse2-mmx-cvtpi2ps.c: Likewise.
	* gcc.target/i386/sse2-mmx-cvtps2pi.c: Likewise.
	* gcc.target/i386/sse2-mmx-cvttps2pi.c: Likewise.
	* gcc.target/i386/sse2-mmx-maskmovq.c: Likewise.
	* gcc.target/i386/sse2-mmx-packssdw.c: Likewise.
	* gcc.target/i386/sse2-mmx-packsswb.c: Likewise.
	* gcc.target/i386/sse2-mmx-packuswb.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddb.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddd.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddq.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddsb.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddsw.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddusb.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddusw.c: Likewise.
	* gcc.target/i386/sse2-mmx-paddw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pand.c: Likewise.
	* gcc.target/i386/sse2-mmx-pandn.c: Likewise.
	* gcc.target/i386/sse2-mmx-pavgb.c: Likewise.
	* gcc.target/i386/sse2-mmx-pavgw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpeqb.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpeqd.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpeqw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpgtb.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpgtd.c: Likewise.
	* gcc.target/i386/sse2-mmx-pcmpgtw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pextrw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pinsrw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmaddwd.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmaxsw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmaxub.c: Likewise.
	* gcc.target/i386/sse2-mmx-pminsw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pminub.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmovmskb.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmulhuw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmulhw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmullw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pmuludq.c: Likewise.
	* gcc.target/i386/sse2-mmx-por.c: Likewise.
	* gcc.target/i386/sse2-mmx-psadbw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pshufw.c: Likewise.
	* gcc.target/i386/sse2-mmx-pslld.c: Likewise.
	* gcc.target/i386/sse2-mmx-pslldi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psllq.c: Likewise.
	* gcc.target/i386/sse2-mmx-psllqi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psllw.c: Likewise.
	* gcc.target/i386/sse2-mmx-psllwi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrad.c: Likewise.
	* gcc.target/i386/sse2-mmx-psradi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psraw.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrawi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrld.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrldi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrlq.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrlqi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrlw.c: Likewise.
	* gcc.target/i386/sse2-mmx-psrlwi.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubb.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubd.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubq.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubusb.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubusw.c: Likewise.
	* gcc.target/i386/sse2-mmx-psubw.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpckhbw.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpckhdq.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpckhwd.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpcklbw.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpckldq.c: Likewise.
	* gcc.target/i386/sse2-mmx-punpcklwd.c: Likewise.
	* gcc.target/i386/sse2-mmx-pxor.c: Likewise.
---
 gcc/testsuite/gcc.target/i386/mmx-vals.h      |  77 ++++++
 gcc/testsuite/gcc.target/i386/sse2-mmx-10.c   |  43 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-11.c   |  39 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-12.c   |  42 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-13.c   |  40 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-14.c   |  31 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-15.c   |  36 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-16.c   |  40 +++
 gcc/testsuite/gcc.target/i386/sse2-mmx-17.c   |  51 ++++
 gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c  |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c  |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c  |   7 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-2.c    |  12 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-20.c   |  12 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-21.c   |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-22.c   |  14 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-3.c    |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-4.c    |   4 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-5.c    |  11 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-6.c    |  11 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-7.c    |  13 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-8.c    |   4 +
 gcc/testsuite/gcc.target/i386/sse2-mmx-9.c    |  79 ++++++
 .../gcc.target/i386/sse2-mmx-cvtpi2ps.c       |  43 +++
 .../gcc.target/i386/sse2-mmx-cvtps2pi.c       |  36 +++
 .../gcc.target/i386/sse2-mmx-cvttps2pi.c      |  36 +++
 .../gcc.target/i386/sse2-mmx-maskmovq.c       |  99 +++++++
 .../gcc.target/i386/sse2-mmx-packssdw.c       |  52 ++++
 .../gcc.target/i386/sse2-mmx-packsswb.c       |  52 ++++
 .../gcc.target/i386/sse2-mmx-packuswb.c       |  52 ++++
 .../gcc.target/i386/sse2-mmx-paddb.c          |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddd.c          |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddq.c          |  43 +++
 .../gcc.target/i386/sse2-mmx-paddsb.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddsw.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddusb.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddusw.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-paddw.c          |  48 ++++
 gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c |  44 ++++
 .../gcc.target/i386/sse2-mmx-pandn.c          |  44 ++++
 .../gcc.target/i386/sse2-mmx-pavgb.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-pavgw.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-pcmpeqb.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pcmpeqd.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pcmpeqw.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pcmpgtb.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pcmpgtd.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pcmpgtw.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-pextrw.c         |  59 +++++
 .../gcc.target/i386/sse2-mmx-pinsrw.c         |  61 +++++
 .../gcc.target/i386/sse2-mmx-pmaddwd.c        |  47 ++++
 .../gcc.target/i386/sse2-mmx-pmaxsw.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-pmaxub.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-pminsw.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-pminub.c         |  48 ++++
 .../gcc.target/i386/sse2-mmx-pmovmskb.c       |  46 ++++
 .../gcc.target/i386/sse2-mmx-pmulhuw.c        |  51 ++++
 .../gcc.target/i386/sse2-mmx-pmulhw.c         |  53 ++++
 .../gcc.target/i386/sse2-mmx-pmullw.c         |  52 ++++
 .../gcc.target/i386/sse2-mmx-pmuludq.c        |  47 ++++
 gcc/testsuite/gcc.target/i386/sse2-mmx-por.c  |  44 ++++
 .../gcc.target/i386/sse2-mmx-psadbw.c         |  58 ++++
 .../gcc.target/i386/sse2-mmx-pshufw.c         | 248 ++++++++++++++++++
 .../gcc.target/i386/sse2-mmx-pslld.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-pslldi.c         | 153 +++++++++++
 .../gcc.target/i386/sse2-mmx-psllq.c          |  47 ++++
 .../gcc.target/i386/sse2-mmx-psllqi.c         | 245 +++++++++++++++++
 .../gcc.target/i386/sse2-mmx-psllw.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-psllwi.c         | 105 ++++++++
 .../gcc.target/i386/sse2-mmx-psrad.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-psradi.c         | 153 +++++++++++
 .../gcc.target/i386/sse2-mmx-psraw.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-psrawi.c         | 105 ++++++++
 .../gcc.target/i386/sse2-mmx-psrld.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-psrldi.c         | 153 +++++++++++
 .../gcc.target/i386/sse2-mmx-psrlq.c          |  47 ++++
 .../gcc.target/i386/sse2-mmx-psrlqi.c         | 245 +++++++++++++++++
 .../gcc.target/i386/sse2-mmx-psrlw.c          |  52 ++++
 .../gcc.target/i386/sse2-mmx-psrlwi.c         | 105 ++++++++
 .../gcc.target/i386/sse2-mmx-psubb.c          |  48 ++++
 .../gcc.target/i386/sse2-mmx-psubd.c          |  48 ++++
 .../gcc.target/i386/sse2-mmx-psubq.c          |  43 +++
 .../gcc.target/i386/sse2-mmx-psubusb.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-psubusw.c        |  48 ++++
 .../gcc.target/i386/sse2-mmx-psubw.c          |  48 ++++
 .../gcc.target/i386/sse2-mmx-punpckhbw.c      |  53 ++++
 .../gcc.target/i386/sse2-mmx-punpckhdq.c      |  47 ++++
 .../gcc.target/i386/sse2-mmx-punpckhwd.c      |  49 ++++
 .../gcc.target/i386/sse2-mmx-punpcklbw.c      |  53 ++++
 .../gcc.target/i386/sse2-mmx-punpckldq.c      |  47 ++++
 .../gcc.target/i386/sse2-mmx-punpcklwd.c      |  49 ++++
 gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c |  44 ++++
 gcc/testsuite/gcc.target/i386/sse2-mmx.c      |   1 -
 98 files changed, 5137 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/mmx-vals.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-10.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-11.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-12.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-13.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-14.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-15.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-16.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-17.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-20.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-21.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-22.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-3.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-4.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-5.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-6.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-7.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-8.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-9.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-por.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c

diff --git a/gcc/testsuite/gcc.target/i386/mmx-vals.h b/gcc/testsuite/gcc.target/i386/mmx-vals.h
new file mode 100644
index 00000000000..62d0c1cb514
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/mmx-vals.h
@@ -0,0 +1,77 @@
+/* Routine to check correctness of the results */
+
+__attribute__((unused))
+static int
+saturate_b (int i)
+{
+  if (i > 127)
+    i = 127;
+  else if (i < -128)
+    i = -128;
+  return i;
+}
+
+__attribute__((unused))
+static int
+saturate_w (int i)
+{
+  if (i > 32767)
+    i = 32767;
+  else if (i < -32768)
+    i = -32768;
+  return i;
+}
+
+__attribute__((unused))
+static int
+saturate_ub (int i)
+{
+  if (i > 255)
+    i = 255;
+  else if (i < 0)
+    i = 0;
+  return i;
+}
+
+__attribute__((unused))
+static int
+saturate_uw (int i)
+{
+  if (i > 65535)
+    i = 65535;
+  else if (i < 0)
+    i = 0;
+  return i;
+}
+
+static long long MMXops[] =
+{
+  0x3467512347612976LL, 0x000000000000000eLL,
+  0x3467512347612976LL, 0x0000000000000014LL,
+  0x3467512347612976LL, 0x000000000000003cLL,
+  0x0000000000000000LL, 0xFFFFFFFFFFFFFFFFLL,
+  0xFFFFFFFFFFFFFFFFLL, 0x0000000000000000LL,
+  0x0000000000000001LL, 0x1000000000000000LL,
+  0x1000000000000000LL, 0x0000000000000001LL,
+  0xFF00FF00FF00FF00LL, 0x00FF00FF00FF00FFLL,
+  0xFFFFFFFFFFFFFFFFLL, 0x0101010101010101LL,
+  0x0101010101010101LL, 0xFFFFFFFFFFFFFFFFLL,
+  0x0123456789ABCDEFLL, 0x0123456789ABCDEFLL,
+  0x3467512347612976LL, 0x1839876340879234LL,
+  0x0000000000000000LL, 0x0000000000000000LL,
+  0xFFFFFFFFFFFFFFFFLL, 0xFFFFFFFFFFFFFFFFLL,
+  0x7F7F7F7F7F7F7F7FLL, 0x7F7F7F7F7F7F7F7FLL,
+  0x7F7F7F7F7F7F7F7FLL, 0x0101010101010101LL,
+  0x7F7F7F7F7F7F7F7FLL, 0x4782082349761237LL,
+  0x0000000000000000LL, 0x7F7F7F7F7F7F7F7FLL,
+  0x8080808080808080LL, 0x8080808080808080LL,
+  0x0101010101010101LL, 0x8080808080808080LL,
+  0x8080808080808080LL, 0x0000000000000000LL,
+  0x2372347120982458LL, 0x8080808080808080LL,
+  0xFFFFFFFFFFFFFFFFLL, 0x8080808080808080LL,
+  0x7F7F7F7F7F7F7F7FLL, 0xFFFFFFFFFFFFFFFFLL,
+  0x8080808080808080LL, 0x7F7F7F7F7F7F7F7FLL,
+  0xFFFFFFFFFFFFFFFFLL, 0x7F7F7F7F7F7F7F7FLL
+};
+
+#define MMX_num_ops (sizeof (MMXops) / sizeof (MMXops[0]))
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-10.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-10.c
new file mode 100644
index 00000000000..4da7ed3a3e9
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-10.c
@@ -0,0 +1,43 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_from_int  (long long *ll1, long long *r)
+{
+  int i1 = *(int *) ll1;
+  *(__m64 *) r = _m_from_int (i1);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *ll1, long long *r)
+{
+  int *res = (int *) r;
+  res[0] = *(int *) ll1;
+  res[1] = 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      test_from_int (&MMXops[i], &r);
+      compute_correct_result (&MMXops[i], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-11.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-11.c
new file mode 100644
index 00000000000..6737ec5f2d4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-11.c
@@ -0,0 +1,39 @@
+/* { dg-do run { target { ! ia32 } } } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2 -mno-mmx" } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_from_long_long  (long long *ll1, long long *r)
+{
+  *(__m64 *) r = _mm_cvtsi64_m64 (*ll1);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *ll1, long long *r)
+{
+  *r = *ll1;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      test_from_long_long (&MMXops[i], &r);
+      compute_correct_result (&MMXops[i], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-12.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-12.c
new file mode 100644
index 00000000000..b626daa0aab
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-12.c
@@ -0,0 +1,42 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_to_int  (long long *ll1, long long *r)
+{
+  __m64 m = *(__m64 *) ll1; 
+  *(int *) r = _m_to_int (m);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *ll1, long long *r)
+{
+  int *i1 = (int *) ll1;
+  *(int *) r = *i1;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r = 0, ck = 0;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      test_to_int (&MMXops[i], &r);
+      compute_correct_result (&MMXops[i], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-13.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-13.c
new file mode 100644
index 00000000000..fd1eed66daa
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-13.c
@@ -0,0 +1,40 @@
+/* { dg-do run { target { ! ia32 } } } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2 -mno-mmx" } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_to_long_long  (long long *ll1, long long *r)
+{
+  __m64 m = *(__m64 *) ll1; 
+  *r = _mm_cvtm64_si64 (m);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *ll1, long long *r)
+{
+  *r = *ll1;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      test_to_long_long (&MMXops[i], &r);
+      compute_correct_result (&MMXops[i], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-14.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-14.c
new file mode 100644
index 00000000000..bc21ba711da
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-14.c
@@ -0,0 +1,31 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_setzero (long long *r)
+{
+  *(__m64 *) r = _mm_setzero_si64 ();
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *r)
+{
+  *r = 0x0LL;
+}
+
+static void
+sse2_test (void)
+{
+  long long r, ck;
+
+  /* Run the MMX tests */
+  test_setzero (&r);
+  compute_correct_result (&ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-15.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-15.c
new file mode 100644
index 00000000000..ab8451b591d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-15.c
@@ -0,0 +1,36 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_set (int x, int y, long long *r)
+{
+  *(__m64 *) r = _mm_set_pi32 (x, y);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (int x, int y, long long *res_p)
+{
+  int *res = (int *) res_p;
+  res[0] = y;
+  res[1] = x;
+}
+
+static void
+sse2_test (void)
+{
+  int x, y;
+  long long r, ck;
+
+  /* Run the MMX tests */
+  x = 0x0badbeef;
+  y = 0x0badfeed;
+  test_set (x, y, &r);
+  compute_correct_result (x, y, &ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-16.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-16.c
new file mode 100644
index 00000000000..b348b9f1c64
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-16.c
@@ -0,0 +1,40 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_set (int i0, int i1, int i2, int i3, long long *r)
+{
+  *(__m64 *) r = _mm_set_pi16 (i0, i1, i2, i3);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (int i0, int i1, int i2, int i3, long long *res_p)
+{
+  short *res = (short *) res_p;
+  res[0] = i3;
+  res[1] = i2;
+  res[2] = i1;
+  res[3] = i0;
+}
+
+static void
+sse2_test (void)
+{
+  short i0, i1, i2, i3;
+  long long r, ck;
+
+  /* Run the MMX tests */
+  i0 = 0x0bad;
+  i1 = 0xbeef;
+  i2 = 0x0bad;
+  i3 = 0xfeed;
+  test_set (i0, i1, i2, i3, &r);
+  compute_correct_result (i0, i1, i2, i3, &ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-17.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-17.c
new file mode 100644
index 00000000000..756cbeb33da
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-17.c
@@ -0,0 +1,51 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_set (char i0, char i1, char i2, char i3,
+	  char i4, char i5, char i6, char i7, long long *r)
+{
+  *(__m64 *) r = _mm_set_pi8 (i0, i1, i2, i3, i4, i5, i6, i7);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (char i0, char i1, char i2, char i3,
+			char i4, char i5, char i6, char i7,
+			long long *res_p)
+{
+  char *res = (char *) res_p;
+  res[0] = i7;
+  res[1] = i6;
+  res[2] = i5;
+  res[3] = i4;
+  res[4] = i3;
+  res[5] = i2;
+  res[6] = i1;
+  res[7] = i0;
+}
+
+static void
+sse2_test (void)
+{
+  char i0, i1, i2, i3, i4, i5, i6, i7;
+  long long r, ck;
+
+  /* Run the MMX tests */
+  i0 = 0x12;
+  i1 = 0x34;
+  i2 = 0x56;
+  i3 = 0x78;
+  i4 = 0x90;
+  i5 = 0xab;
+  i6 = 0xcd;
+  i7 = 0xef;
+  test_set (i0, i1, i2, i3, i4, i5, i6, i7, &r);
+  compute_correct_result (i0, i1, i2, i3, i4, i5, i6, i7, &ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c
new file mode 100644
index 00000000000..3505a5c0cf4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c
@@ -0,0 +1,14 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx -mno-avx512vl" } */
+/* { dg-final { scan-assembler-times "pshufd" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include <mmintrin.h>
+
+__m64
+foo (int i)
+{
+  __v2si x = { i, i };
+  return (__m64) x;
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c
new file mode 100644
index 00000000000..9b267b17346
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx512bw -mavx512vl" } */
+/* { dg-final { scan-assembler-times "pbroadcastd" 1 } } */
+/* { dg-final { scan-assembler-not "movd" } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-18a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c
new file mode 100644
index 00000000000..394f05b6b49
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx512bw -mno-avx512vl" } */
+/* { dg-final { scan-assembler-times "pshufd" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-18a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c
new file mode 100644
index 00000000000..9715ace241f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c
@@ -0,0 +1,14 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx -mno-avx -mtune=intel" } */
+/* { dg-final { scan-assembler-times "pshuflw" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include <mmintrin.h>
+
+__m64
+foo (short i)
+{
+  __v4hi x = { i, i, i, i };
+  return (__m64) x;
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c
new file mode 100644
index 00000000000..a6d42313336
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx512bw -mavx512vl" } */
+/* { dg-final { scan-assembler-times "pbroadcastw" 1 } } */
+/* { dg-final { scan-assembler-not "movd" } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-19a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c
new file mode 100644
index 00000000000..b02dc8c2ffd
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx -mno-avx2 -mtune=intel" } */
+/* { dg-final { scan-assembler-times "pshuflw" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-19a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c
new file mode 100644
index 00000000000..54691883c9c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx512bw -mno-avx512vl -mtune=intel" } */
+/* { dg-final { scan-assembler-times "pbroadcastw" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-19a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c
new file mode 100644
index 00000000000..8be973cc4fc
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -mno-mmx -mavx2 -mno-avx512f -mtune=intel" } */
+/* { dg-final { scan-assembler-times "pbroadcastw" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "movl" } } */
+
+#include "sse2-mmx-19a.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-2.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-2.c
new file mode 100644
index 00000000000..e4cee2da83e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-2.c
@@ -0,0 +1,12 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler-not "%xmm" } } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <mmintrin.h>
+
+float
+foo (__m64 x)
+{
+  return ((__v2sf) x)[0];
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-20.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-20.c
new file mode 100644
index 00000000000..173fa154d40
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-20.c
@@ -0,0 +1,12 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <mmintrin.h>
+
+int
+foo (__m64 x)
+{
+  return ((__v2si) x)[0];
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-21.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-21.c
new file mode 100644
index 00000000000..8f5341e2de6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-21.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler-times "pshufd" 1 } } */
+/* { dg-final { scan-assembler-times "movd" 1 } } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <mmintrin.h>
+
+int
+foo (__m64 x)
+{
+  return ((__v2si) x)[1];
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-22.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-22.c
new file mode 100644
index 00000000000..94fa5874365
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-22.c
@@ -0,0 +1,14 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include <mmintrin.h>
+
+void
+foo (void)
+{
+  _mm_empty ();
+}
+
+/* { dg-final { scan-assembler-times "emms" 1 { target ia32 } } } */
+/* { dg-final { scan-assembler-not "emms" { target { ! ia32 } } } } */
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-3.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-3.c
new file mode 100644
index 00000000000..77f518b6c5f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-3.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler "cvtdq2ps" } } */
+/* { dg-final { scan-assembler-not "cvtpi2ps" } } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <xmmintrin.h>
+
+__m128
+foo (__m128 i1, __m64 i2)
+{
+  return _mm_cvtpi32_ps (i1, i2);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-4.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-4.c
new file mode 100644
index 00000000000..d923724fc1c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-4.c
@@ -0,0 +1,4 @@
+/* { dg-do run { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+
+#include "mmx-4.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-5.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-5.c
new file mode 100644
index 00000000000..1953dc89bb1
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-5.c
@@ -0,0 +1,11 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <xmmintrin.h>
+
+int
+foo (__m64 i)
+{
+  return _m_pextrw (i, 2);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-6.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-6.c
new file mode 100644
index 00000000000..f73444f493b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-6.c
@@ -0,0 +1,11 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <xmmintrin.h>
+
+__m64
+foo (__m64 i, int w)
+{
+  return _m_pinsrw (i, w, 2);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-7.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-7.c
new file mode 100644
index 00000000000..6ea491d2715
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-7.c
@@ -0,0 +1,13 @@
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-final { scan-assembler "movnti" } } */
+/* { dg-final { scan-assembler-not "movntq" } } */
+/* { dg-final { scan-assembler-not "%mm" } } */
+
+#include <xmmintrin.h>
+
+void
+foo (__m64 *p, __m64 i)
+{
+  _mm_stream_pi (p, i);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-8.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-8.c
new file mode 100644
index 00000000000..342c2fa4f25
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-8.c
@@ -0,0 +1,4 @@
+/* { dg-do run { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+
+#include "mmx-8.c"
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-9.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-9.c
new file mode 100644
index 00000000000..f0bf7256c0e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-9.c
@@ -0,0 +1,79 @@
+/* { dg-do run { target { ! ia32 } } } */
+/* { dg-options "-O2 -msse2 -mno-mmx" } */
+/* { dg-require-effective-target sse2 } */
+
+#include "sse2-check.h"
+
+#include <string.h>
+
+#define FLOAT_X	2.3456
+#define FLOAT_Y	-4.5987
+
+static float expected_x = FLOAT_X;
+static float expected_y = FLOAT_Y;
+static __v2sf expected1 = { FLOAT_X, FLOAT_Y };
+static __v2sf expected2 = { FLOAT_X, 0 };
+static __v2sf expected3 = { FLOAT_X, FLOAT_X };
+
+float
+__attribute__((noinline, noclone))
+foo1 (__m64 x)
+{
+  return ((__v2sf) x)[0];
+}
+
+float
+__attribute__((noinline, noclone))
+foo2 (__m64 x)
+{
+  return ((__v2sf) x)[1];
+}
+
+__m64
+__attribute__((noinline, noclone))
+foo3 (float x)
+{
+  return __extension__ (__m64) (__v2sf) { x, 0 };
+}
+
+__m64
+__attribute__((noinline, noclone))
+foo4 (float x)
+{
+  return __extension__ (__m64) (__v2sf) { x, x };
+}
+
+__m64
+__attribute__((noinline, noclone))
+foo5 (float x, float y)
+{
+  return __extension__ (__m64) (__v2sf) { x, y };
+}
+
+void
+__attribute__((noinline))
+sse2_test (void)
+{
+  __m64 res;
+  float x;
+
+  x = foo1 ((__m64) expected1);
+  if (x != expected_x)
+    abort ();
+
+  x = foo2 ((__m64) expected1);
+  if (x != expected_y)
+    abort ();
+
+  res = foo3 (FLOAT_X); 
+  if (memcmp (&res, &expected2, sizeof (res)))
+    abort ();
+
+  res = foo4 (FLOAT_X); 
+  if (memcmp (&res, &expected3, sizeof (res)))
+    abort ();
+
+  res = foo5 (FLOAT_X, FLOAT_Y); 
+  if (memcmp (&res, &expected1, sizeof (res)))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c
new file mode 100644
index 00000000000..4ab38d6f015
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c
@@ -0,0 +1,43 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include <string.h>
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_cvtpi32_ps (__m128 *i1, __m64 *i2, __m128 *r)
+{
+  *(__m128 *) r = _mm_cvtpi32_ps (*i1, *i2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (__m128 *dst_p, __m64 *src_p, __m128 *res_p)
+{
+  int *src = (int *) src_p;
+  float *res = (float *) res_p;
+  *res_p = *dst_p;
+  int i;
+  __m128 r;
+  for (i = 0; i < 2; i++)
+    {
+      r = _mm_cvt_si2ss (*dst_p, src[i]);
+      res[i] = ((__v4sf) r)[0];
+    }
+}
+
+static void
+sse2_test (void)
+{
+  __m128 r, ck;
+  __v4sf x = { 1.99f, -3.9f, -4.9f, 3.8f };
+  __v2si y = { 30, -39 };
+
+  /* Run the MMX tests */
+  test_cvtpi32_ps ((__m128 *) &x, (__m64 *) &y, &r);
+  compute_correct_result ((__m128 *) &x, (__m64 *) &y, &ck);
+  if (memcmp (&ck, &r, sizeof (r)))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c
new file mode 100644
index 00000000000..6084c9ff3d5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c
@@ -0,0 +1,36 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_cvtps_pi32 (__m128 *src_p, long long *r)
+{
+  *(__m64 *) r = _mm_cvtps_pi32 (*src_p);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (__m128 *src_p, long long *res_p)
+{
+  __v4sf *src = (__v4sf *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = _mm_cvt_ss2si (_mm_set_ss ((*src)[i]));
+}
+
+static void
+sse2_test (void)
+{
+  long long r, ck;
+  __v4sf x = { 1.99f, -3.9f, -4.9f, 3.8f };
+
+  /* Run the MMX tests */
+  test_cvtps_pi32 ((__m128 *) &x, &r);
+  compute_correct_result ((__m128 *) &x, &ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c
new file mode 100644
index 00000000000..6f12b2f064f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c
@@ -0,0 +1,36 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_cvttps_pi32 (__m128 *src_p, long long *r)
+{
+  *(__m64 *) r = _mm_cvttps_pi32 (*src_p);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (__m128 *src_p, long long *res_p)
+{
+  __v4sf *src = (__v4sf *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = _mm_cvtt_ss2si (_mm_set_ss ((*src)[i]));
+}
+
+static void
+sse2_test (void)
+{
+  long long r, ck;
+  __v4sf x = { 1.99f, -3.9f, -4.9f, 3.8f };
+
+  /* Run the MMX tests */
+  test_cvttps_pi32 ((__m128 *) &x, &r);
+  compute_correct_result ((__m128 *) &x, &ck);
+  if (ck != r)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c
new file mode 100644
index 00000000000..29fab1914f6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c
@@ -0,0 +1,99 @@
+/* { dg-do run { target { *-*-linux* } } } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include <unistd.h>
+#include <sys/mman.h>
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_maskmovq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+   _m_maskmovq (t1, t2, (char *) r);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    if ((src[i] & 0x80) != 0)
+      res[i] = dst[i];
+}
+
+static void
+do_maskmovq_test (long long *r)
+{
+  int i;
+  long long ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      r[0] = -1LL;
+      ck = -1LL;
+      test_maskmovq (&MMXops[i], &MMXops[i], r);
+      compute_correct_result (&MMXops[i], &MMXops[i], &ck);
+      if (*r != ck)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
+
+static void
+sse2_test (void)
+{
+  char *buf;
+  long long *r;
+  size_t page_size = sysconf(_SC_PAGESIZE);
+
+  buf = mmap (0, 3 * page_size, PROT_READ | PROT_WRITE,
+	      MAP_PRIVATE | MAP_ANON, -1, 0);
+  if (buf == MAP_FAILED)
+    {
+      perror ("mmap");
+      abort ();
+    }
+
+  if (mprotect (buf, page_size, PROT_NONE))
+    {
+      perror ("mprotect");
+      abort ();
+    }
+
+  if (mprotect (buf + 2 * page_size, page_size, PROT_NONE))
+    {
+      perror ("mprotect");
+      abort ();
+    }
+
+  r = (long long *) (buf + page_size);
+  do_maskmovq_test (r);
+
+  r = (long long *) (buf + page_size + 3);
+  do_maskmovq_test (r);
+
+  r = (long long *) (buf + page_size + 11);
+  do_maskmovq_test (r);
+
+  r = (long long *) (buf + 2 * page_size - 16);
+  do_maskmovq_test (r);
+
+  r = (long long *) (buf + 2 * page_size - 16 + 3);
+  do_maskmovq_test (r);
+
+  r = (long long *) (buf + 2 * page_size - 16 + 8);
+  do_maskmovq_test (r);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c
new file mode 100644
index 00000000000..ef5fded4568
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_packssdw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_packssdw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  short *res = (short *) res_p;
+  int i;
+
+  for (i = 0; i < 2; i++)
+    {
+      res[i] = saturate_w (dst[i]);
+      res[i + 2] = saturate_w (src[i]);
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_packssdw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c
new file mode 100644
index 00000000000..bcc34040b2b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_packsswb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_packsswb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  char *res = (char *) res_p;
+  int i;
+
+  for (i = 0; i < 4; i++)
+    {
+      res[i] = saturate_b (dst[i]);
+      res[i + 4] = saturate_b (src[i]);
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_packsswb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c
new file mode 100644
index 00000000000..ac2da068d64
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_packuswb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_packuswb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+
+  for (i = 0; i < 4; i++)
+    {
+      res[i] = saturate_ub (dst[i]);
+      res[i + 4] = saturate_ub (src[i]);
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_packuswb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c
new file mode 100644
index 00000000000..b12534908ec
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] + src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c
new file mode 100644
index 00000000000..d71b21713a6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = dst[i] + src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c
new file mode 100644
index 00000000000..b015393cf03
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c
@@ -0,0 +1,43 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _mm_add_si64 (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  res_p[0] = dst_p[0] + src_p[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c
new file mode 100644
index 00000000000..f8236beef32
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddsb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddsb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = saturate_b (dst[i] + src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddsb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c
new file mode 100644
index 00000000000..cc52fe65e66
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddsw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddsw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = saturate_w (dst[i] + src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddsw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c
new file mode 100644
index 00000000000..c802bcf4d10
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddusb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddusb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = saturate_ub (dst[i] + src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddusb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c
new file mode 100644
index 00000000000..c0f3c8601fb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddusw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddusw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned short *dst = (unsigned short *) dst_p;
+  unsigned short *src = (unsigned short *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = saturate_uw (dst[i] + src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddusw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c
new file mode 100644
index 00000000000..6921ae8da3c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_paddw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_paddw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] + src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_paddw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c
new file mode 100644
index 00000000000..0047a7e4d93
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c
@@ -0,0 +1,44 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pand  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pand (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  res[0] = dst[0] & src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pand (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c
new file mode 100644
index 00000000000..a0cce55a4ea
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c
@@ -0,0 +1,44 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pandn  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pandn (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  res[0] = ~dst[0] & src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pandn (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c
new file mode 100644
index 00000000000..fbb586e1d21
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pavgb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pavgb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+  unsigned int tmp;
+  for (i = 0; i < 8; i++)
+    {
+      tmp = dst[i] + src[i] + 1;
+      res[i] = tmp >> 1;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pavgb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c
new file mode 100644
index 00000000000..81cdc213541
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pavgw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pavgw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned short *dst = (unsigned short *) dst_p;
+  unsigned short *src = (unsigned short *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  unsigned int tmp;
+  for (i = 0; i < 4; i++)
+    {
+      tmp = dst[i] + src[i] + 1;
+      res[i] = tmp >> 1;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pavgw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c
new file mode 100644
index 00000000000..d355ba0a757
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpeqb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpeqb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] == src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpeqb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c
new file mode 100644
index 00000000000..bd896249212
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpeqd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpeqd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = dst[i] == src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpeqd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c
new file mode 100644
index 00000000000..ae15437f1cb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpeqw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpeqw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] == src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpeqw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c
new file mode 100644
index 00000000000..f175c698803
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpgtb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpgtb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] > src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpgtb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c
new file mode 100644
index 00000000000..451afb54c3d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpgtd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpgtd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = dst[i] > src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpgtd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c
new file mode 100644
index 00000000000..f3ebe8eedb9
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pcmpgtw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pcmpgtw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] > src[i] ? -1 : 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pcmpgtw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c
new file mode 100644
index 00000000000..bb48740a7ca
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c
@@ -0,0 +1,59 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include <string.h>
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pextrw (__m64 *i, unsigned int imm, int *r)
+{
+  switch (imm)
+    {
+    case 0:
+      *r = _m_pextrw (*i, 0);
+      break;
+    case 1:
+      *r = _m_pextrw (*i, 1);
+      break;
+    case 2:
+      *r = _m_pextrw (*i, 2);
+      break;
+    case 3:
+      *r = _m_pextrw (*i, 3);
+      break;
+    default:
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (__m64 *src_p, unsigned int imm, int *res_p)
+{
+  short *src = (short *) src_p;
+  if (imm < 4)
+    *res_p = src[imm];
+}
+
+static void
+sse2_test (void)
+{
+  int r, ck;
+  int i;
+  int failed = 0;
+  __v4hi y = { 3320, -3339, 48, 4392 };
+
+  /* Run the MMX tests */
+  for (i = 0; i < 4; i++)
+    {
+      test_pextrw ((__m64 *) &y, i, &r);
+      compute_correct_result ((__m64 *) &y, i, &ck);
+      if (r != ck)
+	failed++;
+    }
+
+  if (failed)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c
new file mode 100644
index 00000000000..c25ddd96daa
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c
@@ -0,0 +1,61 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include <string.h>
+#include "sse2-check.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pinsrw  (__m64 *i, int val, unsigned int imm, int *r)
+{
+  switch (imm)
+    {
+    case 0:
+       *(__m64 *) r = _m_pinsrw  (*i, val, 0);
+      break;
+    case 1:
+       *(__m64 *) r = _m_pinsrw  (*i, val, 1);
+      break;
+    case 2:
+       *(__m64 *) r = _m_pinsrw  (*i, val, 2);
+      break;
+    case 3:
+       *(__m64 *) r = _m_pinsrw  (*i, val, 3);
+      break;
+    default:
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (__m64 *src_p, int val, unsigned int imm,
+			int *res_p)
+{
+  short *res = (short *) res_p;
+  *(__m64 *) res_p = *src_p;
+  if (imm < 4)
+    res[imm] = val;
+}
+
+static void
+sse2_test (void)
+{
+  int r, ck;
+  int i;
+  int failed = 0;
+  __v4hi y = { 3320, -3339, 48, 4392 };
+
+  /* Run the MMX tests */
+  for (i = 0; i < 4; i++)
+    {
+      test_pinsrw  ((__m64 *) &y, 0x1234, i, &r);
+      compute_correct_result ((__m64 *) &y, 0x1234, i, &ck);
+      if (r != ck)
+	failed++;
+    }
+
+  if (failed)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c
new file mode 100644
index 00000000000..1eace62088a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmaddwd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmaddwd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  int *res = (int *) res_p;
+  res[0] = dst[0] * src[0] + dst[1] * src[1];
+  res[1] = dst[2] * src[2] + dst[3] * src[3];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmaddwd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c
new file mode 100644
index 00000000000..ec3b68f09d5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmaxsw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmaxsw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] > src[i] ? dst[i] : src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmaxsw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c
new file mode 100644
index 00000000000..085d6a0eaf5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmaxub  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmaxub (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] > src[i] ? dst[i] : src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmaxub (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c
new file mode 100644
index 00000000000..f64d9b2e2b7
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pminsw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pminsw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] < src[i] ? dst[i] : src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pminsw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c
new file mode 100644
index 00000000000..372e5ef8764
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pminub  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pminub (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] < src[i] ? dst[i] : src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pminub (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c
new file mode 100644
index 00000000000..46b18d20558
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c
@@ -0,0 +1,46 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmovmskb  (long long *ll1, int *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  *r = _m_pmovmskb (t1);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, int *res_p)
+{
+  char *src = (char *) src_p;
+  int i;
+  int res = 0;
+  for (i = 0; i < 8; i++)
+    res |= ((src[i] & 0x80) >> 7) << i;
+  *res_p = res;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  int r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      test_pmovmskb (&MMXops[i], &r);
+      compute_correct_result (&MMXops[i], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c
new file mode 100644
index 00000000000..0f8fcbe02b8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c
@@ -0,0 +1,51 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmulhuw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmulhuw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned short *dst = (unsigned short *) dst_p;
+  unsigned short *src = (unsigned short *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    {
+      unsigned int t = dst[i] * src[i];
+      res[i] = t >> 16;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmulhuw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c
new file mode 100644
index 00000000000..b89c58f3b6d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c
@@ -0,0 +1,53 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmulhw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmulhw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int tmp;
+  int i;
+  for (i = 0; i < 4; i++)
+    {
+      tmp = dst[i] * src[i];
+      tmp >>= 16;
+      res[i] = tmp;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmulhw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c
new file mode 100644
index 00000000000..13b89622560
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmullw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pmullw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int tmp;
+  int i;
+  for (i = 0; i < 4; i++)
+    {
+      tmp = dst[i] * src[i];
+      res[i] = tmp;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmullw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c
new file mode 100644
index 00000000000..819e0e13c03
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pmuludq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _mm_mul_su32 (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned int *dst = (unsigned int*) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  unsigned long long *res = (unsigned long long *) res_p;
+  res[0] = dst[0];
+  res[0] *= src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pmuludq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-por.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-por.c
new file mode 100644
index 00000000000..9dc6072fffa
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-por.c
@@ -0,0 +1,44 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_por  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_por (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  res[0] = dst[0] | src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_por (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c
new file mode 100644
index 00000000000..223d3b280bd
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c
@@ -0,0 +1,58 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psadbw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psadbw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  int tmp;
+  unsigned int sum = 0;
+  for (i = 0; i < 8; i++)
+    {
+      tmp = dst[i] - src[i];
+      if (tmp < 0)
+	tmp = -tmp;
+      sum += tmp;
+    }
+  res[0] = sum;
+  for (i = 1; i < 4; i++)
+    res[i] = 0;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psadbw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c
new file mode 100644
index 00000000000..b7236586216
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c
@@ -0,0 +1,248 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pshufw  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_pshufw (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_pshufw (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_pshufw (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_pshufw (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_pshufw (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_pshufw (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_pshufw (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_pshufw (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_pshufw (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_pshufw (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_pshufw (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_pshufw (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_pshufw (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_pshufw (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_pshufw (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_pshufw (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_pshufw (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_pshufw (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_pshufw (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_pshufw (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_pshufw (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_pshufw (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_pshufw (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_pshufw (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_pshufw (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_pshufw (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_pshufw (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_pshufw (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_pshufw (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_pshufw (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_pshufw (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_pshufw (t1, 31);
+      break;
+    case 32:
+      *(__m64 *) r = _m_pshufw (t1, 32);
+      break;
+    case 33:
+      *(__m64 *) r = _m_pshufw (t1, 33);
+      break;
+    case 34:
+      *(__m64 *) r = _m_pshufw (t1, 34);
+      break;
+    case 35:
+      *(__m64 *) r = _m_pshufw (t1, 35);
+      break;
+    case 36:
+      *(__m64 *) r = _m_pshufw (t1, 36);
+      break;
+    case 37:
+      *(__m64 *) r = _m_pshufw (t1, 37);
+      break;
+    case 38:
+      *(__m64 *) r = _m_pshufw (t1, 38);
+      break;
+    case 39:
+      *(__m64 *) r = _m_pshufw (t1, 39);
+      break;
+    case 40:
+      *(__m64 *) r = _m_pshufw (t1, 40);
+      break;
+    case 41:
+      *(__m64 *) r = _m_pshufw (t1, 41);
+      break;
+    case 42:
+      *(__m64 *) r = _m_pshufw (t1, 42);
+      break;
+    case 43:
+      *(__m64 *) r = _m_pshufw (t1, 43);
+      break;
+    case 44:
+      *(__m64 *) r = _m_pshufw (t1, 44);
+      break;
+    case 45:
+      *(__m64 *) r = _m_pshufw (t1, 45);
+      break;
+    case 46:
+      *(__m64 *) r = _m_pshufw (t1, 46);
+      break;
+    case 47:
+      *(__m64 *) r = _m_pshufw (t1, 47);
+      break;
+    case 48:
+      *(__m64 *) r = _m_pshufw (t1, 48);
+      break;
+    case 49:
+      *(__m64 *) r = _m_pshufw (t1, 49);
+      break;
+    case 50:
+      *(__m64 *) r = _m_pshufw (t1, 50);
+      break;
+    case 51:
+      *(__m64 *) r = _m_pshufw (t1, 51);
+      break;
+    case 52:
+      *(__m64 *) r = _m_pshufw (t1, 52);
+      break;
+    case 53:
+      *(__m64 *) r = _m_pshufw (t1, 53);
+      break;
+    case 54:
+      *(__m64 *) r = _m_pshufw (t1, 54);
+      break;
+    case 55:
+      *(__m64 *) r = _m_pshufw (t1, 55);
+      break;
+    case 56:
+      *(__m64 *) r = _m_pshufw (t1, 56);
+      break;
+    case 57:
+      *(__m64 *) r = _m_pshufw (t1, 57);
+      break;
+    case 58:
+      *(__m64 *) r = _m_pshufw (t1, 58);
+      break;
+    case 59:
+      *(__m64 *) r = _m_pshufw (t1, 59);
+      break;
+    case 60:
+      *(__m64 *) r = _m_pshufw (t1, 60);
+      break;
+    case 61:
+      *(__m64 *) r = _m_pshufw (t1, 61);
+      break;
+    case 62:
+      *(__m64 *) r = _m_pshufw (t1, 62);
+      break;
+    case 63:
+      *(__m64 *) r = _m_pshufw (t1, 63);
+      break;
+    default:
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  unsigned long long src = *(unsigned long long *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  unsigned int shift;
+  for (i = 0; i < 4; i++)
+    {
+      shift = ((imm >> (2 * i)) & 0x3) * 16;
+      res[i] = (src >> shift) & 0xffff;
+    }
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      if (i > 63)
+	break;
+      test_pshufw (&MMXops[i], i, &r);
+      compute_correct_result (&MMXops[i], i, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c
new file mode 100644
index 00000000000..86983c0b848
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pslld  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pslld (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned int *dst = (unsigned int *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  unsigned int *res = (unsigned int *) res_p;
+  int i;
+  if (src[1] || src[0] > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = dst[i] << src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pslld (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c
new file mode 100644
index 00000000000..07f0ae9d83c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c
@@ -0,0 +1,153 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllwi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_pslldi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_pslldi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_pslldi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_pslldi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_pslldi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_pslldi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_pslldi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_pslldi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_pslldi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_pslldi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_pslldi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_pslldi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_pslldi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_pslldi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_pslldi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_pslldi (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_pslldi (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_pslldi (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_pslldi (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_pslldi (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_pslldi (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_pslldi (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_pslldi (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_pslldi (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_pslldi (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_pslldi (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_pslldi (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_pslldi (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_pslldi (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_pslldi (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_pslldi (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_pslldi (t1, 31);
+      break;
+    default:
+      *(__m64 *) r = _m_pslldi (t1, 32);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  unsigned int *src = (unsigned int *) src_p;
+  unsigned int *res = (unsigned int *) res_p;
+  int i;
+  if (imm > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = src[i] << imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psllwi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c
new file mode 100644
index 00000000000..37d2e0b22cc
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psllq (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  if (src[0] > 63)
+    res[0] = 0;
+  else
+    res[0] = dst[0] << src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psllq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c
new file mode 100644
index 00000000000..655b369e04f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c
@@ -0,0 +1,245 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllwi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psllqi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psllqi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psllqi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psllqi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psllqi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psllqi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psllqi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psllqi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psllqi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psllqi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psllqi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psllqi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psllqi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psllqi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psllqi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psllqi (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_psllqi (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_psllqi (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_psllqi (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_psllqi (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_psllqi (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_psllqi (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_psllqi (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_psllqi (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_psllqi (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_psllqi (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_psllqi (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_psllqi (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_psllqi (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_psllqi (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_psllqi (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_psllqi (t1, 31);
+      break;
+    case 32:
+      *(__m64 *) r = _m_psllqi (t1, 32);
+      break;
+    case 33:
+      *(__m64 *) r = _m_psllqi (t1, 33);
+      break;
+    case 34:
+      *(__m64 *) r = _m_psllqi (t1, 34);
+      break;
+    case 35:
+      *(__m64 *) r = _m_psllqi (t1, 35);
+      break;
+    case 36:
+      *(__m64 *) r = _m_psllqi (t1, 36);
+      break;
+    case 37:
+      *(__m64 *) r = _m_psllqi (t1, 37);
+      break;
+    case 38:
+      *(__m64 *) r = _m_psllqi (t1, 38);
+      break;
+    case 39:
+      *(__m64 *) r = _m_psllqi (t1, 39);
+      break;
+    case 40:
+      *(__m64 *) r = _m_psllqi (t1, 40);
+      break;
+    case 41:
+      *(__m64 *) r = _m_psllqi (t1, 41);
+      break;
+    case 42:
+      *(__m64 *) r = _m_psllqi (t1, 42);
+      break;
+    case 43:
+      *(__m64 *) r = _m_psllqi (t1, 43);
+      break;
+    case 44:
+      *(__m64 *) r = _m_psllqi (t1, 44);
+      break;
+    case 45:
+      *(__m64 *) r = _m_psllqi (t1, 45);
+      break;
+    case 46:
+      *(__m64 *) r = _m_psllqi (t1, 46);
+      break;
+    case 47:
+      *(__m64 *) r = _m_psllqi (t1, 47);
+      break;
+    case 48:
+      *(__m64 *) r = _m_psllqi (t1, 48);
+      break;
+    case 49:
+      *(__m64 *) r = _m_psllqi (t1, 49);
+      break;
+    case 50:
+      *(__m64 *) r = _m_psllqi (t1, 50);
+      break;
+    case 51:
+      *(__m64 *) r = _m_psllqi (t1, 51);
+      break;
+    case 52:
+      *(__m64 *) r = _m_psllqi (t1, 52);
+      break;
+    case 53:
+      *(__m64 *) r = _m_psllqi (t1, 53);
+      break;
+    case 54:
+      *(__m64 *) r = _m_psllqi (t1, 54);
+      break;
+    case 55:
+      *(__m64 *) r = _m_psllqi (t1, 55);
+      break;
+    case 56:
+      *(__m64 *) r = _m_psllqi (t1, 56);
+      break;
+    case 57:
+      *(__m64 *) r = _m_psllqi (t1, 57);
+      break;
+    case 58:
+      *(__m64 *) r = _m_psllqi (t1, 58);
+      break;
+    case 59:
+      *(__m64 *) r = _m_psllqi (t1, 59);
+      break;
+    case 60:
+      *(__m64 *) r = _m_psllqi (t1, 60);
+      break;
+    case 61:
+      *(__m64 *) r = _m_psllqi (t1, 61);
+      break;
+    case 62:
+      *(__m64 *) r = _m_psllqi (t1, 62);
+      break;
+    case 63:
+      *(__m64 *) r = _m_psllqi (t1, 63);
+      break;
+    default:
+      *(__m64 *) r = _m_psllqi (t1, 64);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *src, unsigned int imm,
+			unsigned long long *res)
+{
+  int i;
+  if (imm > 63)
+    res[0] = 0;
+  else
+    res[0] = src[0] << imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psllwi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c
new file mode 100644
index 00000000000..7c27d144185
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psllw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned short *dst = (unsigned short *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  if (src[1] || src[0] > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = dst[i] << src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psllw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c
new file mode 100644
index 00000000000..458463b2073
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c
@@ -0,0 +1,105 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllwi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psllwi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psllwi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psllwi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psllwi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psllwi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psllwi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psllwi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psllwi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psllwi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psllwi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psllwi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psllwi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psllwi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psllwi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psllwi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psllwi (t1, 15);
+      break;
+    default:
+      *(__m64 *) r = _m_psllwi (t1, 16);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  unsigned short *src = (unsigned short *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  if (imm > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = src[i] << imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psllwi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c
new file mode 100644
index 00000000000..494bcb4ccbd
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrad  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psrad (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  if (src[1] || src[0] > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = dst[i] < 0 ? -1 : 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = dst[i] >> src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psrad (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c
new file mode 100644
index 00000000000..4ae2ac848bc
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c
@@ -0,0 +1,153 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psradi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psradi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psradi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psradi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psradi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psradi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psradi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psradi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psradi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psradi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psradi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psradi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psradi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psradi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psradi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psradi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psradi (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_psradi (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_psradi (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_psradi (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_psradi (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_psradi (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_psradi (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_psradi (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_psradi (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_psradi (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_psradi (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_psradi (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_psradi (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_psradi (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_psradi (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_psradi (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_psradi (t1, 31);
+      break;
+    default:
+      *(__m64 *) r = _m_psradi (t1, 32);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  if (imm > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = src[i] < 0 ? -1 : 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = src[i] >> imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psradi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c
new file mode 100644
index 00000000000..dd097f2e16b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psraw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psraw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  if (src[1] || src[0] > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = dst[i] < 0 ? -1 : 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = dst[i] >> src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psraw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c
new file mode 100644
index 00000000000..30a2d8bd08b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c
@@ -0,0 +1,105 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrawi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psrawi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psrawi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psrawi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psrawi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psrawi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psrawi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psrawi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psrawi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psrawi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psrawi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psrawi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psrawi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psrawi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psrawi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psrawi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psrawi (t1, 15);
+      break;
+    default:
+      *(__m64 *) r = _m_psrawi (t1, 16);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  if (imm > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = src[i] < 0 ? -1 : 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = src[i] >> imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psrawi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c
new file mode 100644
index 00000000000..3858be180a8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrld  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psrld (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  if (src[1] || src[0] > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = dst[i] >> src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psrld (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c
new file mode 100644
index 00000000000..9315b6f0137
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c
@@ -0,0 +1,153 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrldi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psrldi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psrldi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psrldi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psrldi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psrldi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psrldi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psrldi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psrldi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psrldi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psrldi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psrldi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psrldi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psrldi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psrldi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psrldi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psrldi (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_psrldi (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_psrldi (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_psrldi (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_psrldi (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_psrldi (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_psrldi (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_psrldi (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_psrldi (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_psrldi (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_psrldi (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_psrldi (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_psrldi (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_psrldi (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_psrldi (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_psrldi (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_psrldi (t1, 31);
+      break;
+    default:
+      *(__m64 *) r = _m_psrldi (t1, 32);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  if (imm > 31)
+    for (i = 0; i < 2; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 2; i++)
+      res[i] = src[i] >> imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psrldi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c
new file mode 100644
index 00000000000..064fb4aab7e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrlq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psrlq (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  if (src[0] > 63)
+    res[0] = 0;
+  else
+    res[0] = dst[0] >> src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psrlq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c
new file mode 100644
index 00000000000..35de178efea
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c
@@ -0,0 +1,245 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psllwi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psrlqi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psrlqi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psrlqi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psrlqi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psrlqi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psrlqi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psrlqi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psrlqi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psrlqi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psrlqi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psrlqi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psrlqi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psrlqi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psrlqi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psrlqi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psrlqi (t1, 15);
+      break;
+    case 16:
+      *(__m64 *) r = _m_psrlqi (t1, 16);
+      break;
+    case 17:
+      *(__m64 *) r = _m_psrlqi (t1, 17);
+      break;
+    case 18:
+      *(__m64 *) r = _m_psrlqi (t1, 18);
+      break;
+    case 19:
+      *(__m64 *) r = _m_psrlqi (t1, 19);
+      break;
+    case 20:
+      *(__m64 *) r = _m_psrlqi (t1, 20);
+      break;
+    case 21:
+      *(__m64 *) r = _m_psrlqi (t1, 21);
+      break;
+    case 22:
+      *(__m64 *) r = _m_psrlqi (t1, 22);
+      break;
+    case 23:
+      *(__m64 *) r = _m_psrlqi (t1, 23);
+      break;
+    case 24:
+      *(__m64 *) r = _m_psrlqi (t1, 24);
+      break;
+    case 25:
+      *(__m64 *) r = _m_psrlqi (t1, 25);
+      break;
+    case 26:
+      *(__m64 *) r = _m_psrlqi (t1, 26);
+      break;
+    case 27:
+      *(__m64 *) r = _m_psrlqi (t1, 27);
+      break;
+    case 28:
+      *(__m64 *) r = _m_psrlqi (t1, 28);
+      break;
+    case 29:
+      *(__m64 *) r = _m_psrlqi (t1, 29);
+      break;
+    case 30:
+      *(__m64 *) r = _m_psrlqi (t1, 30);
+      break;
+    case 31:
+      *(__m64 *) r = _m_psrlqi (t1, 31);
+      break;
+    case 32:
+      *(__m64 *) r = _m_psrlqi (t1, 32);
+      break;
+    case 33:
+      *(__m64 *) r = _m_psrlqi (t1, 33);
+      break;
+    case 34:
+      *(__m64 *) r = _m_psrlqi (t1, 34);
+      break;
+    case 35:
+      *(__m64 *) r = _m_psrlqi (t1, 35);
+      break;
+    case 36:
+      *(__m64 *) r = _m_psrlqi (t1, 36);
+      break;
+    case 37:
+      *(__m64 *) r = _m_psrlqi (t1, 37);
+      break;
+    case 38:
+      *(__m64 *) r = _m_psrlqi (t1, 38);
+      break;
+    case 39:
+      *(__m64 *) r = _m_psrlqi (t1, 39);
+      break;
+    case 40:
+      *(__m64 *) r = _m_psrlqi (t1, 40);
+      break;
+    case 41:
+      *(__m64 *) r = _m_psrlqi (t1, 41);
+      break;
+    case 42:
+      *(__m64 *) r = _m_psrlqi (t1, 42);
+      break;
+    case 43:
+      *(__m64 *) r = _m_psrlqi (t1, 43);
+      break;
+    case 44:
+      *(__m64 *) r = _m_psrlqi (t1, 44);
+      break;
+    case 45:
+      *(__m64 *) r = _m_psrlqi (t1, 45);
+      break;
+    case 46:
+      *(__m64 *) r = _m_psrlqi (t1, 46);
+      break;
+    case 47:
+      *(__m64 *) r = _m_psrlqi (t1, 47);
+      break;
+    case 48:
+      *(__m64 *) r = _m_psrlqi (t1, 48);
+      break;
+    case 49:
+      *(__m64 *) r = _m_psrlqi (t1, 49);
+      break;
+    case 50:
+      *(__m64 *) r = _m_psrlqi (t1, 50);
+      break;
+    case 51:
+      *(__m64 *) r = _m_psrlqi (t1, 51);
+      break;
+    case 52:
+      *(__m64 *) r = _m_psrlqi (t1, 52);
+      break;
+    case 53:
+      *(__m64 *) r = _m_psrlqi (t1, 53);
+      break;
+    case 54:
+      *(__m64 *) r = _m_psrlqi (t1, 54);
+      break;
+    case 55:
+      *(__m64 *) r = _m_psrlqi (t1, 55);
+      break;
+    case 56:
+      *(__m64 *) r = _m_psrlqi (t1, 56);
+      break;
+    case 57:
+      *(__m64 *) r = _m_psrlqi (t1, 57);
+      break;
+    case 58:
+      *(__m64 *) r = _m_psrlqi (t1, 58);
+      break;
+    case 59:
+      *(__m64 *) r = _m_psrlqi (t1, 59);
+      break;
+    case 60:
+      *(__m64 *) r = _m_psrlqi (t1, 60);
+      break;
+    case 61:
+      *(__m64 *) r = _m_psrlqi (t1, 61);
+      break;
+    case 62:
+      *(__m64 *) r = _m_psrlqi (t1, 62);
+      break;
+    case 63:
+      *(__m64 *) r = _m_psrlqi (t1, 63);
+      break;
+    default:
+      *(__m64 *) r = _m_psrlqi (t1, 64);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *src, unsigned int imm,
+			unsigned long long *res)
+{
+  int i;
+  if (imm > 63)
+    res[0] = 0;
+  else
+    res[0] = src[0] >> imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psllwi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c
new file mode 100644
index 00000000000..d2c1680ba5d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c
@@ -0,0 +1,52 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrlw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psrlw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  unsigned int *src = (unsigned int *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  if (src[1] || src[0] > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = dst[i] >> src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psrlw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c
new file mode 100644
index 00000000000..0dbffa56299
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c
@@ -0,0 +1,105 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psrlwi  (long long *ll1, unsigned int imm, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  switch (imm)
+    {
+    case 0:
+      *(__m64 *) r = _m_psrlwi (t1, 0);
+      break;
+    case 1:
+      *(__m64 *) r = _m_psrlwi (t1, 1);
+      break;
+    case 2:
+      *(__m64 *) r = _m_psrlwi (t1, 2);
+      break;
+    case 3:
+      *(__m64 *) r = _m_psrlwi (t1, 3);
+      break;
+    case 4:
+      *(__m64 *) r = _m_psrlwi (t1, 4);
+      break;
+    case 5:
+      *(__m64 *) r = _m_psrlwi (t1, 5);
+      break;
+    case 6:
+      *(__m64 *) r = _m_psrlwi (t1, 6);
+      break;
+    case 7:
+      *(__m64 *) r = _m_psrlwi (t1, 7);
+      break;
+    case 8:
+      *(__m64 *) r = _m_psrlwi (t1, 8);
+      break;
+    case 9:
+      *(__m64 *) r = _m_psrlwi (t1, 9);
+      break;
+    case 10:
+      *(__m64 *) r = _m_psrlwi (t1, 10);
+      break;
+    case 11:
+      *(__m64 *) r = _m_psrlwi (t1, 11);
+      break;
+    case 12:
+      *(__m64 *) r = _m_psrlwi (t1, 12);
+      break;
+    case 13:
+      *(__m64 *) r = _m_psrlwi (t1, 13);
+      break;
+    case 14:
+      *(__m64 *) r = _m_psrlwi (t1, 14);
+      break;
+    case 15:
+      *(__m64 *) r = _m_psrlwi (t1, 15);
+      break;
+    default:
+      *(__m64 *) r = _m_psrlwi (t1, 16);
+      break;
+    }
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *src_p, unsigned int imm,
+			long long *res_p)
+{
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  if (imm > 15)
+    for (i = 0; i < 4; i++)
+      res[i] = 0;
+  else
+    for (i = 0; i < 4; i++)
+      res[i] = src[i] >> imm;
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  unsigned int count;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i++)
+    {
+      count = MMXops[i];
+      test_psrlwi (&MMXops[i], count, &r);
+      compute_correct_result (&MMXops[i], count, &ck);
+      if (ck != r)
+	  fail++;
+      }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c
new file mode 100644
index 00000000000..de0076a1654
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psubb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = dst[i] - src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c
new file mode 100644
index 00000000000..344f632b6d6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psubd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  int i;
+  for (i = 0; i < 2; i++)
+    res[i] = dst[i] - src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c
new file mode 100644
index 00000000000..613f302bf6a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c
@@ -0,0 +1,43 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _mm_sub_si64 (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  res_p[0] = dst_p[0] - src_p[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c
new file mode 100644
index 00000000000..ad6112a5d33
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubusb  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psubusb (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned char *dst = (unsigned char *) dst_p;
+  unsigned char *src = (unsigned char *) src_p;
+  unsigned char *res = (unsigned char *) res_p;
+  int i;
+  for (i = 0; i < 8; i++)
+    res[i] = saturate_ub (dst[i] - src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubusb (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c
new file mode 100644
index 00000000000..8a8a9c0fb4d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubusw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psubusw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  unsigned short *dst = (unsigned short *) dst_p;
+  unsigned short *src = (unsigned short *) src_p;
+  unsigned short *res = (unsigned short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = saturate_uw (dst[i] - src[i]);
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubusw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c
new file mode 100644
index 00000000000..4d53c8c419a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_psubw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_psubw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  int i;
+  for (i = 0; i < 4; i++)
+    res[i] = dst[i] - src[i];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_psubw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c
new file mode 100644
index 00000000000..07281f2c3a3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c
@@ -0,0 +1,53 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpckhbw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpckhbw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  res[0] = dst[4];
+  res[1] = src[4];
+  res[2] = dst[5];
+  res[3] = src[5];
+  res[4] = dst[6];
+  res[5] = src[6];
+  res[6] = dst[7];
+  res[7] = src[7];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpckhbw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c
new file mode 100644
index 00000000000..6c67af92b19
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpckhdq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpckhdq (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  res[0] = dst[1];
+  res[1] = src[1];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpckhdq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c
new file mode 100644
index 00000000000..b6c348323ea
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c
@@ -0,0 +1,49 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpckhwd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpckhwd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  res[0] = dst[2];
+  res[1] = src[2];
+  res[2] = dst[3];
+  res[3] = src[3];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpckhwd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c
new file mode 100644
index 00000000000..dcf6d3f25bb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c
@@ -0,0 +1,53 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpcklbw  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpcklbw (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  char *dst = (char *) dst_p;
+  char *src = (char *) src_p;
+  char *res = (char *) res_p;
+  res[0] = dst[0];
+  res[1] = src[0];
+  res[2] = dst[1];
+  res[3] = src[1];
+  res[4] = dst[2];
+  res[5] = src[2];
+  res[6] = dst[3];
+  res[7] = src[3];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpcklbw (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c
new file mode 100644
index 00000000000..463c6d2c989
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpckldq  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpckldq (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  int *dst = (int *) dst_p;
+  int *src = (int *) src_p;
+  int *res = (int *) res_p;
+  res[0] = dst[0];
+  res[1] = src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpckldq (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c
new file mode 100644
index 00000000000..acd62f5bd74
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c
@@ -0,0 +1,49 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_punpcklwd  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_punpcklwd (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (long long *dst_p, long long *src_p,
+			long long *res_p)
+{
+  short *dst = (short *) dst_p;
+  short *src = (short *) src_p;
+  short *res = (short *) res_p;
+  res[0] = dst[0];
+  res[1] = src[0];
+  res[2] = dst[1];
+  res[3] = src[1];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_punpcklwd (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c b/gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c
new file mode 100644
index 00000000000..be632989c35
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c
@@ -0,0 +1,44 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -fno-strict-aliasing -msse2" } */
+/* { dg-additional-options "-mno-mmx" { target { ! ia32 } } } */
+
+#include "sse2-check.h"
+#include "mmx-vals.h"
+
+__attribute__((noinline, noclone))
+static void
+test_pxor  (long long *ll1, long long *ll2, long long *r)
+{
+  __m64 t1 = *(__m64 *) ll1;
+  __m64 t2 = *(__m64 *) ll2;
+  *(__m64 *) r = _m_pxor (t1, t2);
+}
+
+/* Routine to manually compute the results */
+static void
+compute_correct_result (unsigned long long *dst,
+			unsigned long long *src,
+			unsigned long long *res)
+{
+  res[0] = dst[0] ^ src[0];
+}
+
+static void
+sse2_test (void)
+{
+  int i;
+  long long r, ck;
+  int fail = 0;
+
+  /* Run the MMX tests */
+  for (i = 0; i < MMX_num_ops; i += 2)
+    {
+      test_pxor (&MMXops[i], &MMXops[i + 1], &r);
+      compute_correct_result (&MMXops[i], &MMXops[i + 1], &ck);
+      if (ck != r)
+	fail++;
+    }
+
+  if (fail != 0)
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse2-mmx.c b/gcc/testsuite/gcc.target/i386/sse2-mmx.c
index fb226a8e8f3..338cb9da289 100644
--- a/gcc/testsuite/gcc.target/i386/sse2-mmx.c
+++ b/gcc/testsuite/gcc.target/i386/sse2-mmx.c
@@ -4,7 +4,6 @@
 
 #include "sse2-check.h"
 
-#include <mmintrin.h>
 
 #define N 4
 
-- 
2.20.1

[PATCH 20/41] i386: Emulate MMX mmx_umulv4hi3_highpart with SSE

[H.J. Lu]

Emulate MMX mmx_umulv4hi3_highpart with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_umulv4hi3_highpart): Also check
	TARGET_MMX and TARGET_MMX_WITH_SSE.
	(*mmx_umulv4hi3_highpart): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 5ae04de205d..5a342256cbc 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -781,28 +781,34 @@
 	  (lshiftrt:V4SI
 	    (mult:V4SI
 	      (zero_extend:V4SI
-		(match_operand:V4HI 1 "nonimmediate_operand"))
+		(match_operand:V4HI 1 "register_mmxmem_operand"))
 	      (zero_extend:V4SI
-		(match_operand:V4HI 2 "nonimmediate_operand")))
+		(match_operand:V4HI 2 "register_mmxmem_operand")))
 	    (const_int 16))))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
   "ix86_fixup_binary_operands_no_copy (MULT, V4HImode, operands);")
 
 (define_insn "*mmx_umulv4hi3_highpart"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(truncate:V4HI
 	  (lshiftrt:V4SI
 	    (mult:V4SI
 	      (zero_extend:V4SI
-		(match_operand:V4HI 1 "nonimmediate_operand" "%0"))
+		(match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv"))
 	      (zero_extend:V4SI
-		(match_operand:V4HI 2 "nonimmediate_operand" "ym")))
+		(match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))
 	  (const_int 16))))]
-  "(TARGET_SSE || TARGET_3DNOW_A)
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ix86_binary_operator_ok (MULT, V4HImode, operands)"
-  "pmulhuw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxmul")
-   (set_attr "mode" "DI")])
+  "@
+   pmulhuw\t{%2, %0|%0, %2}
+   pmulhuw\t{%2, %0|%0, %2}
+   vpmulhuw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxmul,ssemul,ssemul")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_pmaddwd"
   [(set (match_operand:V2SI 0 "register_operand")
-- 
2.20.1

[PATCH 19/41] i386: Emulate MMX mmx_pmovmskb with SSE

[H.J. Lu]

Emulate MMX mmx_pmovmskb with SSE by zero-extending result of SSE pmovmskb
from QImode to SImode.  Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_pmovmskb): Changed to
	define_insn_and_split to support SSE emulation.
---
 gcc/config/i386/mmx.md | 30 +++++++++++++++++++++++-------
 1 file changed, 23 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index edfb8623701..5ae04de205d 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1763,14 +1763,30 @@
   [(set_attr "type" "mmxshft")
    (set_attr "mode" "DI")])
 
-(define_insn "mmx_pmovmskb"
-  [(set (match_operand:SI 0 "register_operand" "=r")
-	(unspec:SI [(match_operand:V8QI 1 "register_operand" "y")]
+(define_insn_and_split "mmx_pmovmskb"
+  [(set (match_operand:SI 0 "register_operand" "=r,r")
+	(unspec:SI [(match_operand:V8QI 1 "register_operand" "y,x")]
 		   UNSPEC_MOVMSK))]
-  "TARGET_SSE || TARGET_3DNOW_A"
-  "pmovmskb\t{%1, %0|%0, %1}"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "@
+   pmovmskb\t{%1, %0|%0, %1}
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(set (match_dup 0)
+        (unspec:SI [(match_dup 1)] UNSPEC_MOVMSK))
+   (set (match_dup 0)
+	(zero_extend:SI (match_dup 2)))]
+{
+  /* Generate SSE pmovmskb and zero-extend from QImode to SImode.  */
+  operands[1] = lowpart_subreg (V16QImode, operands[1],
+				GET_MODE (operands[1]));
+  operands[2] = lowpart_subreg (QImode, operands[0],
+				GET_MODE (operands[0]));
+}
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "mmxcvt,ssemov")
+   (set_attr "mode" "DI,TI")])
 
 (define_expand "mmx_maskmovq"
   [(set (match_operand:V8QI 0 "memory_operand")
-- 
2.20.1

[PATCH 37/41] i386: Allow MMXMODE moves with TARGET_MMX_WITH_SSE

[H.J. Lu]

	PR target/89021
	* config/i386/mmx.md (MMXMODE:mov<mode>): Also allow
	TARGET_MMX_WITH_SSE.
	(MMXMODE:*mov<mode>_internal): Likewise.
	(MMXMODE:movmisalign<mode>): Likewise.
---
 gcc/config/i386/mmx.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 8b5fe3948e2..c612d6b9e5c 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -70,7 +70,7 @@
 (define_expand "mov<mode>"
   [(set (match_operand:MMXMODE 0 "nonimmediate_operand")
 	(match_operand:MMXMODE 1 "nonimmediate_operand"))]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_move (<MODE>mode, operands);
   DONE;
@@ -81,7 +81,7 @@
     "=r ,o ,r,r ,m ,?!y,!y,?!y,m  ,r  ,?!y,v,v,v,m,r,v,!y,*x")
 	(match_operand:MMXMODE 1 "nonimm_or_0_operand"
     "rCo,rC,C,rm,rC,C  ,!y,m  ,?!y,?!y,r  ,C,v,m,v,v,r,*x,!y"))]
-  "TARGET_MMX
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
    && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
 {
   switch (get_attr_type (insn))
@@ -207,7 +207,7 @@
 (define_expand "movmisalign<mode>"
   [(set (match_operand:MMXMODE 0 "nonimmediate_operand")
 	(match_operand:MMXMODE 1 "nonimmediate_operand"))]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_move (<MODE>mode, operands);
   DONE;
-- 
2.20.1

[PATCH 24/41] i386: Emulate MMX mmx_psadbw with SSE

[H.J. Lu]

Emulate MMX mmx_psadbw with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_psadbw): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index d647dc28baa..098e41e19c3 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1771,14 +1771,19 @@
    (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_psadbw"
-  [(set (match_operand:V1DI 0 "register_operand" "=y")
-        (unspec:V1DI [(match_operand:V8QI 1 "register_operand" "0")
-		      (match_operand:V8QI 2 "nonimmediate_operand" "ym")]
+  [(set (match_operand:V1DI 0 "register_operand" "=y,x,Yv")
+        (unspec:V1DI [(match_operand:V8QI 1 "register_operand" "0,0,Yv")
+		      (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv")]
 		     UNSPEC_PSADBW))]
-  "TARGET_SSE || TARGET_3DNOW_A"
-  "psadbw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
-   (set_attr "mode" "DI")])
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "@
+   psadbw\t{%2, %0|%0, %2}
+   psadbw\t{%2, %0|%0, %2}
+   vpsadbw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sseiadd,sseiadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn_and_split "mmx_pmovmskb"
   [(set (match_operand:SI 0 "register_operand" "=r,r")
-- 
2.20.1

[PATCH 32/41] i386: Emulate MMX pshufb with SSE version

[H.J. Lu]

Emulate MMX version of pshufb with SSE version by masking out the bit 3
of the shuffle control byte.  Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_pshufbv8qi3): Changed to
	define_insn_and_split.  Also allow TARGET_MMX_WITH_SSE.  Add
	SSE emulation.
---
 gcc/config/i386/sse.md | 46 +++++++++++++++++++++++++++++++++---------
 1 file changed, 37 insertions(+), 9 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index b08a577d1e4..79b35d95424 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15728,17 +15728,45 @@
    (set_attr "btver2_decode" "vector")
    (set_attr "mode" "<sseinsnmode>")])
 
-(define_insn "ssse3_pshufbv8qi3"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
-	(unspec:V8QI [(match_operand:V8QI 1 "register_operand" "0")
-		      (match_operand:V8QI 2 "nonimmediate_operand" "ym")]
-		     UNSPEC_PSHUFB))]
-  "TARGET_SSSE3"
-  "pshufb\t{%2, %0|%0, %2}";
-  [(set_attr "type" "sselog1")
+(define_insn_and_split "ssse3_pshufbv8qi3"
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
+	(unspec:V8QI [(match_operand:V8QI 1 "register_operand" "0,0,Yv")
+		      (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv")]
+		     UNSPEC_PSHUFB))
+   (clobber (match_scratch:V4SI 3 "=X,x,Yv"))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   pshufb\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(set (match_dup 3) (match_dup 5))
+   (set (match_dup 3)
+	(and:V4SI (match_dup 3) (match_dup 2)))
+   (set (match_dup 0)
+	(unspec:V16QI [(match_dup 1) (match_dup 4)] UNSPEC_PSHUFB))]
+{
+  /* Emulate MMX version of pshufb with SSE version by masking out the
+     bit 3 of the shuffle control byte.  */
+  operands[0] = lowpart_subreg (V16QImode, operands[0],
+				GET_MODE (operands[0]));
+  operands[1] = lowpart_subreg (V16QImode, operands[1],
+				GET_MODE (operands[1]));
+  operands[2] = lowpart_subreg (V4SImode, operands[2],
+				GET_MODE (operands[2]));
+  operands[4] = lowpart_subreg (V16QImode, operands[3],
+				GET_MODE (operands[3]));
+  rtvec par = gen_rtvec (4, GEN_INT (0xf7f7f7f7),
+			 GEN_INT (0xf7f7f7f7),
+			 GEN_INT (0xf7f7f7f7),
+			 GEN_INT (0xf7f7f7f7));
+  rtx vec_const = gen_rtx_CONST_VECTOR (V4SImode, par);
+  operands[5] = force_const_mem (V4SImode, vec_const);
+}
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "<ssse3_avx2>_psign<mode>3"
   [(set (match_operand:VI124_AVX2 0 "register_operand" "=x,x")
-- 
2.20.1

[PATCH 21/41] i386: Emulate MMX maskmovq with SSE2 maskmovdqu

[H.J. Lu]

Emulate MMX maskmovq with SSE2 maskmovdqu for TARGET_MMX_WITH_SSE by
zero-extending source and mask operands to 128 bits.  Handle unmapped
bits 64:127 at memory address by adjusting source and mask operands
together with memory address.

	PR target/89021
	* config/i386/xmmintrin.h: Emulate MMX maskmovq with SSE2
	maskmovdqu for __MMX_WITH_SSE__.
---
 gcc/config/i386/xmmintrin.h | 61 +++++++++++++++++++++++++++++++++++++
 1 file changed, 61 insertions(+)

diff --git a/gcc/config/i386/xmmintrin.h b/gcc/config/i386/xmmintrin.h
index 58284378514..a915f6c87d7 100644
--- a/gcc/config/i386/xmmintrin.h
+++ b/gcc/config/i386/xmmintrin.h
@@ -1165,7 +1165,68 @@ _m_pshufw (__m64 __A, int const __N)
 extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm_maskmove_si64 (__m64 __A, __m64 __N, char *__P)
 {
+#ifdef __MMX_WITH_SSE__
+  /* Emulate MMX maskmovq with SSE2 maskmovdqu and handle unmapped bits
+     64:127 at address __P.  */
+  typedef long long __v2di __attribute__ ((__vector_size__ (16)));
+  typedef char __v16qi __attribute__ ((__vector_size__ (16)));
+  /* Zero-extend __A and __N to 128 bits.  */
+  __v2di __A128 = __extension__ (__v2di) { ((__v1di) __A)[0], 0 };
+  __v2di __N128 = __extension__ (__v2di) { ((__v1di) __N)[0], 0 };
+
+  /* Check the alignment of __P.  */
+  __SIZE_TYPE__ offset = ((__SIZE_TYPE__) __P) & 0xf;
+  if (offset)
+    {
+      /* If the misalignment of __P > 8, subtract __P by 8 bytes.
+	 Otherwise, subtract __P by the misalignment.  */
+      if (offset > 8)
+	offset = 8;
+      __P = (char *) (((__SIZE_TYPE__) __P) - offset);
+
+      /* Shift __A128 and __N128 to the left by the adjustment.  */
+      switch (offset)
+	{
+	case 1:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 8);
+	  break;
+	case 2:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 2 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 2 * 8);
+	  break;
+	case 3:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 3 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 3 * 8);
+	  break;
+	case 4:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 4 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 4 * 8);
+	  break;
+	case 5:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 5 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 5 * 8);
+	  break;
+	case 6:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 6 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 6 * 8);
+	  break;
+	case 7:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 7 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 7 * 8);
+	  break;
+	case 8:
+	  __A128 = __builtin_ia32_pslldqi128 (__A128, 8 * 8);
+	  __N128 = __builtin_ia32_pslldqi128 (__N128, 8 * 8);
+	  break;
+	default:
+	  break;
+	}
+    }
+  __builtin_ia32_maskmovdqu ((__v16qi)__A128, (__v16qi)__N128, __P);
+#else
   __builtin_ia32_maskmovq ((__v8qi)__A, (__v8qi)__N, __P);
+#endif
 }
 
 extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-- 
2.20.1

[PATCH 35/41] i386: Emulate MMX abs<mode>2 with SSE

[H.J. Lu]

Emulate MMX abs<mode>2 with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/sse.md (abs<mode>2): Add SSE emulation.
---
 gcc/config/i386/sse.md | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index b69a467291c..97ec3795b82 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15973,16 +15973,19 @@
 })
 
 (define_insn "abs<mode>2"
-  [(set (match_operand:MMXMODEI 0 "register_operand" "=y")
+  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,Yv")
 	(abs:MMXMODEI
-	  (match_operand:MMXMODEI 1 "nonimmediate_operand" "ym")))]
-  "TARGET_SSSE3"
-  "pabs<mmxvecsize>\t{%1, %0|%0, %1}";
-  [(set_attr "type" "sselog1")
+	  (match_operand:MMXMODEI 1 "register_mmxmem_operand" "ym,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   pabs<mmxvecsize>\t{%1, %0|%0, %1}
+   %vpabs<mmxvecsize>\t{%1, %0|%0, %1}"
+  [(set_attr "mmx_isa" "native,x64")
+   (set_attr "type" "sselog1")
    (set_attr "prefix_rep" "0")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI")])
 
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
-- 
2.20.1

[PATCH 30/41] i386: Emulate MMX ssse3_pmaddubsw with SSE

[H.J. Lu]

Emulate MMX ssse3_pmaddubsw with SSE.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_pmaddubsw): Add SSE emulation.
---
 gcc/config/i386/sse.md | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 551a1cb1eb2..e8d9bec9766 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15555,17 +15555,17 @@
    (set_attr "mode" "TI")])
 
 (define_insn "ssse3_pmaddubsw"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(ss_plus:V4HI
 	  (mult:V4HI
 	    (zero_extend:V4HI
 	      (vec_select:V4QI
-		(match_operand:V8QI 1 "register_operand" "0")
+		(match_operand:V8QI 1 "register_operand" "0,0,Yv")
 		(parallel [(const_int 0) (const_int 2)
 			   (const_int 4) (const_int 6)])))
 	    (sign_extend:V4HI
 	      (vec_select:V4QI
-		(match_operand:V8QI 2 "nonimmediate_operand" "ym")
+		(match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv")
 		(parallel [(const_int 0) (const_int 2)
 			   (const_int 4) (const_int 6)]))))
 	  (mult:V4HI
@@ -15577,13 +15577,17 @@
 	      (vec_select:V4QI (match_dup 2)
 		(parallel [(const_int 1) (const_int 3)
 			   (const_int 5) (const_int 7)]))))))]
-  "TARGET_SSSE3"
-  "pmaddubsw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "sseiadd")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   pmaddubsw\t{%2, %0|%0, %2}
+   pmaddubsw\t{%2, %0|%0, %2}
+   vpmaddubsw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sseiadd")
    (set_attr "atom_unit" "simul")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_mode_iterator PMULHRSW
   [V8HI (V16HI "TARGET_AVX2")])
-- 
2.20.1

[PATCH 26/41] i386: Emulate MMX umulv1siv1di3 with SSE2

[H.J. Lu]

Emulate MMX umulv1siv1di3 with SSE2.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/mmx.md (sse2_umulv1siv1di3): Add SSE emulation
	support.
	(*sse2_umulv1siv1di3): Add SSE2 emulation.
---
 gcc/config/i386/mmx.md | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index b06f0af984a..f27513f7f2c 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -905,30 +905,36 @@
         (mult:V1DI
 	  (zero_extend:V1DI
 	    (vec_select:V1SI
-	      (match_operand:V2SI 1 "nonimmediate_operand")
+	      (match_operand:V2SI 1 "register_mmxmem_operand")
 	      (parallel [(const_int 0)])))
 	  (zero_extend:V1DI
 	    (vec_select:V1SI
-	      (match_operand:V2SI 2 "nonimmediate_operand")
+	      (match_operand:V2SI 2 "register_mmxmem_operand")
 	      (parallel [(const_int 0)])))))]
-  "TARGET_SSE2"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSE2"
   "ix86_fixup_binary_operands_no_copy (MULT, V2SImode, operands);")
 
 (define_insn "*sse2_umulv1siv1di3"
-  [(set (match_operand:V1DI 0 "register_operand" "=y")
+  [(set (match_operand:V1DI 0 "register_operand" "=y,x,Yv")
         (mult:V1DI
 	  (zero_extend:V1DI
 	    (vec_select:V1SI
-	      (match_operand:V2SI 1 "nonimmediate_operand" "%0")
+	      (match_operand:V2SI 1 "register_mmxmem_operand" "%0,0,Yv")
 	      (parallel [(const_int 0)])))
 	  (zero_extend:V1DI
 	    (vec_select:V1SI
-	      (match_operand:V2SI 2 "nonimmediate_operand" "ym")
+	      (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv")
 	      (parallel [(const_int 0)])))))]
-  "TARGET_SSE2 && ix86_binary_operator_ok (MULT, V2SImode, operands)"
-  "pmuludq\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxmul")
-   (set_attr "mode" "DI")])
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && TARGET_SSE2
+   && ix86_binary_operator_ok (MULT, V2SImode, operands)"
+  "@
+   pmuludq\t{%2, %0|%0, %2}
+   pmuludq\t{%2, %0|%0, %2}
+   vpmuludq\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxmul,ssemul,ssemul")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_<code>v4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

[PATCH 29/41] i386: Emulate MMX ssse3_ph<plusminus_mnemonic>dv2si3 with SSE

[H.J. Lu]

Emulate MMX ssse3_ph<plusminus_mnemonic>dv2si3 with SSE by moving bits
64:95 to bits 32:63 in SSE register.  Only SSE register source operand
is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_ph<plusminus_mnemonic>dv2si3):
	Changed to define_insn_and_split to support SSE emulation.
---
 gcc/config/i386/sse.md | 34 ++++++++++++++++++++++++++--------
 1 file changed, 26 insertions(+), 8 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 5f29f2c3595..551a1cb1eb2 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15367,26 +15367,44 @@
    (set_attr "prefix" "orig,vex")
    (set_attr "mode" "TI")])
 
-(define_insn "ssse3_ph<plusminus_mnemonic>dv2si3"
-  [(set (match_operand:V2SI 0 "register_operand" "=y")
+(define_insn_and_split "ssse3_ph<plusminus_mnemonic>dv2si3"
+  [(set (match_operand:V2SI 0 "register_operand" "=y,x,Yv")
 	(vec_concat:V2SI
 	  (plusminus:SI
 	    (vec_select:SI
-	      (match_operand:V2SI 1 "register_operand" "0")
+	      (match_operand:V2SI 1 "register_operand" "0,0,Yv")
 	      (parallel [(const_int 0)]))
 	    (vec_select:SI (match_dup 1) (parallel [(const_int 1)])))
 	  (plusminus:SI
 	    (vec_select:SI
-	      (match_operand:V2SI 2 "nonimmediate_operand" "ym")
+	      (match_operand:V2SI 2 "register_mmxmem_operand" "ym,x,Yv")
 	      (parallel [(const_int 0)]))
 	    (vec_select:SI (match_dup 2) (parallel [(const_int 1)])))))]
-  "TARGET_SSSE3"
-  "ph<plusminus_mnemonic>d\t{%2, %0|%0, %2}"
-  [(set_attr "type" "sseiadd")
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
+  "@
+   ph<plusminus_mnemonic>d\t{%2, %0|%0, %2}
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(const_int 0)]
+{
+  /* Generate SSE version of the operation.  */
+  rtx op0 = lowpart_subreg (V4SImode, operands[0],
+			    GET_MODE (operands[0]));
+  rtx op1 = lowpart_subreg (V4SImode, operands[1],
+			    GET_MODE (operands[1]));
+  rtx op2 = lowpart_subreg (V4SImode, operands[2],
+			    GET_MODE (operands[2]));
+  emit_insn (gen_ssse3_ph<plusminus_mnemonic>dv4si3 (op0, op1, op2));
+  ix86_move_vector_high_sse_to_mmx (op0);
+  DONE;
+}
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sseiadd")
    (set_attr "atom_unit" "complex")
    (set_attr "prefix_extra" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "avx2_pmaddubsw256"
   [(set (match_operand:V16HI 0 "register_operand" "=x,v")
-- 
2.20.1

[PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[H.J. Lu]

	PR target/89021
	* config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
	mmx_ok to true if TARGET_MMX_WITH_SSE is true.
	(ix86_expand_vector_init_one_nonzero): Likewise.
	(ix86_expand_vector_init_one_var): Likewise.
	(ix86_expand_vector_init_general): Likewise.
	(ix86_expand_vector_init): Likewise.
	(ix86_expand_vector_set): Likewise.
	(ix86_expand_vector_extract): Likewise.
	* config/i386/mmx.md (*vec_dupv2sf): Changed to
	define_insn_and_split to support SSE emulation.
	(*vec_extractv2sf_0): Likewise.
	(*vec_extractv2sf_1): Likewise.
	(*vec_extractv2si_0): Likewise.
	(*vec_extractv2si_1): Likewise.
	(*vec_extractv2si_zext_mem): Likewise.
	(vec_setv2sf): Also allow TARGET_MMX_WITH_SSE.
	(vec_extractv2sf_1 splitter): Likewise.
	(vec_extractv2sfsf): Likewise.
	(vec_setv2si): Likewise.
	(vec_extractv2si_1 splitter): Likewise.
	(vec_extractv2sisi): Likewise.
	(vec_setv4hi): Likewise.
	(vec_extractv4hihi): Likewise.
	(vec_setv8qi): Likewise.
	(vec_extractv8qiqi): Likewise.
---
 gcc/config/i386/i386.c |  8 +++++
 gcc/config/i386/mmx.md | 66 +++++++++++++++++++++++++++---------------
 2 files changed, 50 insertions(+), 24 deletions(-)

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index a76c17beece..25e0dc43a9e 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -42620,6 +42620,7 @@ ix86_expand_vector_init_duplicate (bool mmx_ok, machine_mode mode,
 {
   bool ok;
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2SImode:
@@ -42779,6 +42780,7 @@ ix86_expand_vector_init_one_nonzero (bool mmx_ok, machine_mode mode,
   bool use_vector_set = false;
   rtx (*gen_vec_set_0) (rtx, rtx, rtx) = NULL;
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2DImode:
@@ -42972,6 +42974,7 @@ ix86_expand_vector_init_one_var (bool mmx_ok, machine_mode mode,
   XVECEXP (const_vec, 0, one_var) = CONST0_RTX (GET_MODE_INNER (mode));
   const_vec = gen_rtx_CONST_VECTOR (mode, XVEC (const_vec, 0));
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2DFmode:
@@ -43357,6 +43360,7 @@ ix86_expand_vector_init_general (bool mmx_ok, machine_mode mode,
   machine_mode quarter_mode = VOIDmode;
   int n, i;
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2SFmode:
@@ -43556,6 +43560,8 @@ ix86_expand_vector_init (bool mmx_ok, rtx target, rtx vals)
   int i;
   rtx x;
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
+
   /* Handle first initialization from vector elts.  */
   if (n_elts != XVECLEN (vals, 0))
     {
@@ -43655,6 +43661,7 @@ ix86_expand_vector_set (bool mmx_ok, rtx target, rtx val, int elt)
   machine_mode mmode = VOIDmode;
   rtx (*gen_blendm) (rtx, rtx, rtx, rtx);
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2SFmode:
@@ -44010,6 +44017,7 @@ ix86_expand_vector_extract (bool mmx_ok, rtx target, rtx vec, int elt)
   bool use_vec_extr = false;
   rtx tmp;
 
+  mmx_ok |= TARGET_MMX_WITH_SSE;
   switch (mode)
     {
     case E_V2SImode:
diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index c612d6b9e5c..f98952fd8a0 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -555,14 +555,23 @@
    (set_attr "prefix_extra" "1")
    (set_attr "mode" "V2SF")])
 
-(define_insn "*vec_dupv2sf"
-  [(set (match_operand:V2SF 0 "register_operand" "=y")
+(define_insn_and_split "*vec_dupv2sf"
+  [(set (match_operand:V2SF 0 "register_operand" "=y,x,Yv")
 	(vec_duplicate:V2SF
-	  (match_operand:SF 1 "register_operand" "0")))]
-  "TARGET_MMX"
-  "punpckldq\t%0, %0"
-  [(set_attr "type" "mmxcvt")
-   (set_attr "mode" "DI")])
+	  (match_operand:SF 1 "register_operand" "0,0,Yv")))]
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
+  "@
+   punpckldq\t%0, %0
+   #
+   #"
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(set (match_dup 0)
+	(vec_duplicate:V4SF (match_dup 1)))]
+  "operands[0] = lowpart_subreg (V4SFmode, operands[0],
+				 GET_MODE (operands[0]));"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxcvt,ssemov,ssemov")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "*mmx_concatv2sf"
   [(set (match_operand:V2SF 0 "register_operand"     "=y,y")
@@ -580,7 +589,7 @@
   [(match_operand:V2SF 0 "register_operand")
    (match_operand:SF 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_set (false, operands[0], operands[1],
 			  INTVAL (operands[2]));
@@ -594,11 +603,13 @@
 	(vec_select:SF
 	  (match_operand:V2SF 1 "nonimmediate_operand" " xm,x,ym,y,m,m")
 	  (parallel [(const_int 0)])))]
-  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
   "#"
   "&& reload_completed"
   [(set (match_dup 0) (match_dup 1))]
-  "operands[1] = gen_lowpart (SFmode, operands[1]);")
+  "operands[1] = gen_lowpart (SFmode, operands[1]);"
+  [(set_attr "mmx_isa" "*,*,native,native,*,*")])
 
 ;; Avoid combining registers from different units in a single alternative,
 ;; see comment above inline_secondary_memory_needed function in i386.c
@@ -607,7 +618,8 @@
 	(vec_select:SF
 	  (match_operand:V2SF 1 "nonimmediate_operand" " 0,x,x,o,o,o,o")
 	  (parallel [(const_int 1)])))]
-  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
   "@
    punpckhdq\t%0, %0
    %vmovshdup\t{%1, %0|%0, %1}
@@ -617,6 +629,7 @@
    #
    #"
   [(set_attr "isa" "*,sse3,noavx,*,*,*,*")
+   (set_attr "mmx_isa" "native,*,*,native,*,*,*")
    (set_attr "type" "mmxcvt,sse,sseshuf1,mmxmov,ssemov,fmov,imov")
    (set (attr "length_immediate")
      (if_then_else (eq_attr "alternative" "2")
@@ -634,7 +647,7 @@
 	(vec_select:SF
 	  (match_operand:V2SF 1 "memory_operand")
 	  (parallel [(const_int 1)])))]
-  "TARGET_MMX && reload_completed"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && reload_completed"
   [(set (match_dup 0) (match_dup 1))]
   "operands[1] = adjust_address (operands[1], SFmode, 4);")
 
@@ -642,7 +655,7 @@
   [(match_operand:SF 0 "register_operand")
    (match_operand:V2SF 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_extract (false, operands[0], operands[1],
 			      INTVAL (operands[2]));
@@ -1526,7 +1539,7 @@
   [(match_operand:V2SI 0 "register_operand")
    (match_operand:SI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_set (false, operands[0], operands[1],
 			  INTVAL (operands[2]));
@@ -1540,11 +1553,13 @@
 	(vec_select:SI
 	  (match_operand:V2SI 1 "nonimmediate_operand" "xm,x,ym,y,m")
 	  (parallel [(const_int 0)])))]
-  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
   "#"
   "&& reload_completed"
   [(set (match_dup 0) (match_dup 1))]
-  "operands[1] = gen_lowpart (SImode, operands[1]);")
+  "operands[1] = gen_lowpart (SImode, operands[1]);"
+  [(set_attr "mmx_isa" "*,*,native,native,*")])
 
 ;; Avoid combining registers from different units in a single alternative,
 ;; see comment above inline_secondary_memory_needed function in i386.c
@@ -1553,7 +1568,8 @@
 	(vec_select:SI
 	  (match_operand:V2SI 1 "nonimmediate_operand" " 0,x,x,o,o,o")
 	  (parallel [(const_int 1)])))]
-  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
   "@
    punpckhdq\t%0, %0
    %vpshufd\t{$0xe5, %1, %0|%0, %1, 0xe5}
@@ -1562,6 +1578,7 @@
    #
    #"
   [(set_attr "isa" "*,sse2,noavx,*,*,*")
+   (set_attr "mmx_isa" "native,*,*,native,*,*")
    (set_attr "type" "mmxcvt,sseshuf1,sseshuf1,mmxmov,ssemov,imov")
    (set (attr "length_immediate")
      (if_then_else (eq_attr "alternative" "1,2")
@@ -1575,7 +1592,7 @@
 	(vec_select:SI
 	  (match_operand:V2SI 1 "memory_operand")
 	  (parallel [(const_int 1)])))]
-  "TARGET_MMX && reload_completed"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && reload_completed"
   [(set (match_dup 0) (match_dup 1))]
   "operands[1] = adjust_address (operands[1], SImode, 4);")
 
@@ -1592,13 +1609,14 @@
 {
   operands[1] = adjust_address (operands[1], SImode, INTVAL (operands[2]) * 4);
 }
-  [(set_attr "isa" "*,sse2,*")])
+  [(set_attr "isa" "*,sse2,*")
+   (set_attr "mmx_isa" "native,*,*")])
 
 (define_expand "vec_extractv2sisi"
   [(match_operand:SI 0 "register_operand")
    (match_operand:V2SI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_extract (false, operands[0], operands[1],
 			      INTVAL (operands[2]));
@@ -1618,7 +1636,7 @@
   [(match_operand:V4HI 0 "register_operand")
    (match_operand:HI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_set (false, operands[0], operands[1],
 			  INTVAL (operands[2]));
@@ -1629,7 +1647,7 @@
   [(match_operand:HI 0 "register_operand")
    (match_operand:V4HI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_extract (false, operands[0], operands[1],
 			      INTVAL (operands[2]));
@@ -1649,7 +1667,7 @@
   [(match_operand:V8QI 0 "register_operand")
    (match_operand:QI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_set (false, operands[0], operands[1],
 			  INTVAL (operands[2]));
@@ -1660,7 +1678,7 @@
   [(match_operand:QI 0 "register_operand")
    (match_operand:V8QI 1 "register_operand")
    (match_operand 2 "const_int_operand")]
-  "TARGET_MMX"
+  "TARGET_MMX || TARGET_MMX_WITH_SSE"
 {
   ix86_expand_vector_extract (false, operands[0], operands[1],
 			      INTVAL (operands[2]));
-- 
2.20.1

[PATCH 18/41] i386: Emulate MMX V4HI smaxmin/V8QI umaxmin with SSE

[H.J. Lu]

Emulate MMX V4HI smaxmin/V8QI umaxmin with SSE.  Only SSE register source
operand is allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_<code>v4hi3): Also check TARGET_MMX
	and TARGET_MMX_WITH_SSE.
	(mmx_<code>v8qi3): Likewise.
	(smaxmin:<code>v4hi3): New.
	(umaxmin:<code>v8qi3): Likewise.
	(smaxmin:*mmx_<code>v4hi3): Add SSE emulation.
	(umaxmin:*mmx_<code>v8qi3): Likewise.
---
 gcc/config/i386/mmx.md | 68 +++++++++++++++++++++++++++++-------------
 1 file changed, 48 insertions(+), 20 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index dea2be1d8e2..edfb8623701 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -923,40 +923,68 @@
 (define_expand "mmx_<code>v4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
         (smaxmin:V4HI
-	  (match_operand:V4HI 1 "nonimmediate_operand")
-	  (match_operand:V4HI 2 "nonimmediate_operand")))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+	  (match_operand:V4HI 1 "register_mmxmem_operand")
+	  (match_operand:V4HI 2 "register_mmxmem_operand")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "ix86_fixup_binary_operands_no_copy (<CODE>, V4HImode, operands);")
+
+(define_expand "<code>v4hi3"
+  [(set (match_operand:V4HI 0 "register_operand")
+        (smaxmin:V4HI
+	  (match_operand:V4HI 1 "register_operand")
+	  (match_operand:V4HI 2 "register_operand")))]
+  "TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (<CODE>, V4HImode, operands);")
 
 (define_insn "*mmx_<code>v4hi3"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
         (smaxmin:V4HI
-	  (match_operand:V4HI 1 "nonimmediate_operand" "%0")
-	  (match_operand:V4HI 2 "nonimmediate_operand" "ym")))]
-  "(TARGET_SSE || TARGET_3DNOW_A)
+	  (match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv")
+	  (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ix86_binary_operator_ok (<CODE>, V4HImode, operands)"
-  "p<maxmin_int>w\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+  "@
+   p<maxmin_int>w\t{%2, %0|%0, %2}
+   p<maxmin_int>w\t{%2, %0|%0, %2}
+   vp<maxmin_int>w\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sseiadd,sseiadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_<code>v8qi3"
   [(set (match_operand:V8QI 0 "register_operand")
         (umaxmin:V8QI
-	  (match_operand:V8QI 1 "nonimmediate_operand")
-	  (match_operand:V8QI 2 "nonimmediate_operand")))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+	  (match_operand:V8QI 1 "register_mmxmem_operand")
+	  (match_operand:V8QI 2 "register_mmxmem_operand")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
+  "ix86_fixup_binary_operands_no_copy (<CODE>, V8QImode, operands);")
+
+(define_expand "<code>v8qi3"
+  [(set (match_operand:V8QI 0 "register_operand")
+        (umaxmin:V8QI
+	  (match_operand:V8QI 1 "register_operand")
+	  (match_operand:V8QI 2 "register_operand")))]
+  "TARGET_MMX_WITH_SSE"
   "ix86_fixup_binary_operands_no_copy (<CODE>, V8QImode, operands);")
 
 (define_insn "*mmx_<code>v8qi3"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
         (umaxmin:V8QI
-	  (match_operand:V8QI 1 "nonimmediate_operand" "%0")
-	  (match_operand:V8QI 2 "nonimmediate_operand" "ym")))]
-  "(TARGET_SSE || TARGET_3DNOW_A)
+	  (match_operand:V8QI 1 "register_mmxmem_operand" "%0,0,Yv")
+	  (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv")))]
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ix86_binary_operator_ok (<CODE>, V8QImode, operands)"
-  "p<maxmin_int>b\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxadd")
-   (set_attr "mode" "DI")])
+  "@
+   p<maxmin_int>b\t{%2, %0|%0, %2}
+   p<maxmin_int>b\t{%2, %0|%0, %2}
+   vp<maxmin_int>b\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxadd,sseiadd,sseiadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_ashr<mode>3"
   [(set (match_operand:MMXMODE24 0 "register_operand" "=y,x,Yv")
-- 
2.20.1

[PATCH 34/41] i386: Emulate MMX ssse3_palignrdi with SSE

[H.J. Lu]

Emulate MMX version of palignrq with SSE version by concatenating 2
64-bit MMX operands into a single 128-bit SSE operand, followed by
SSE psrldq.  Only SSE register source operand is allowed.

	PR target/89021
	* config/i386/sse.md (ssse3_palignrdi): Changed to
	define_insn_and_split to support SSE emulation.
---
 gcc/config/i386/sse.md | 58 ++++++++++++++++++++++++++++++++++--------
 1 file changed, 48 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 1d90af0a4b0..b69a467291c 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -15855,23 +15855,61 @@
    (set_attr "prefix" "orig,vex,evex")
    (set_attr "mode" "<sseinsnmode>")])
 
-(define_insn "ssse3_palignrdi"
-  [(set (match_operand:DI 0 "register_operand" "=y")
-	(unspec:DI [(match_operand:DI 1 "register_operand" "0")
-		    (match_operand:DI 2 "nonimmediate_operand" "ym")
-		    (match_operand:SI 3 "const_0_to_255_mul_8_operand" "n")]
+(define_insn_and_split "ssse3_palignrdi"
+  [(set (match_operand:DI 0 "register_operand" "=y,x,Yv")
+	(unspec:DI [(match_operand:DI 1 "register_operand" "0,0,Yv")
+		    (match_operand:DI 2 "register_mmxmem_operand" "ym,x,Yv")
+		    (match_operand:SI 3 "const_0_to_255_mul_8_operand" "n,n,n")]
 		   UNSPEC_PALIGNR))]
-  "TARGET_SSSE3"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && TARGET_SSSE3"
 {
-  operands[3] = GEN_INT (INTVAL (operands[3]) / 8);
-  return "palignr\t{%3, %2, %0|%0, %2, %3}";
+  switch (which_alternative)
+    {
+    case 0:
+      operands[3] = GEN_INT (INTVAL (operands[3]) / 8);
+      return "palignr\t{%3, %2, %0|%0, %2, %3}";
+    case 1:
+    case 2:
+      return "#";
+    default:
+      gcc_unreachable ();
+    }
 }
-  [(set_attr "type" "sseishft")
+  "TARGET_MMX_WITH_SSE && reload_completed"
+  [(set (match_dup 0)
+	(lshiftrt:V1TI (match_dup 0) (match_dup 3)))]
+{
+  /* Emulate MMX palignrdi with SSE psrldq.  */
+  rtx op0 = lowpart_subreg (V2DImode, operands[0],
+			    GET_MODE (operands[0]));
+  rtx insn;
+  if (TARGET_AVX)
+    insn = gen_vec_concatv2di (op0, operands[2], operands[1]);
+  else
+    {
+      /* NB: SSE can only concatenate OP0 and OP1 to OP0.  */
+      insn = gen_vec_concatv2di (op0, operands[1], operands[2]);
+      emit_insn (insn);
+      /* Swap bits 0:63 with bits 64:127.  */
+      rtx mask = gen_rtx_PARALLEL (VOIDmode,
+				   gen_rtvec (4, GEN_INT (2),
+					      GEN_INT (3),
+					      GEN_INT (0),
+					      GEN_INT (1)));
+      rtx op1 = lowpart_subreg (V4SImode, op0, GET_MODE (op0));
+      rtx op2 = gen_rtx_VEC_SELECT (V4SImode, op1, mask);
+      insn = gen_rtx_SET (op1, op2);
+    }
+  emit_insn (insn);
+  operands[0] = lowpart_subreg (V1TImode, op0, GET_MODE (op0));
+}
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "sseishft")
    (set_attr "atom_unit" "sishuf")
    (set_attr "prefix_extra" "1")
    (set_attr "length_immediate" "1")
    (set (attr "prefix_rex") (symbol_ref "x86_extended_reg_mentioned_p (insn)"))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 ;; Mode iterator to handle singularity w/ absence of V2DI and V4DI
 ;; modes for abs instruction on pre AVX-512 targets.
-- 
2.20.1

[PATCH 36/41] Prevent allocation of MMX registers with TARGET_MMX_WITH_SSE

[H.J. Lu]

From: Uros Bizjak <ubizjak@gmail.com>

2019-02-14  Uroš Bizjak  <ubizjak@gmail.com>

	PR target/89021
	* config/i386/i386.md (*zero_extendsidi2): Add mmx_isa attribute.
	* config/i386/sse.md (*vec_concatv2sf_sse4_1): Ditto.
	(*vec_concatv2sf_sse): Ditto.
	(*vec_concatv2si_sse4_1): Ditto.
	(*vec_concatv2si): Ditto.
	(*vec_concatv4si_0): Ditto.
	(*vec_concatv2di_0): Ditto.
---
 gcc/config/i386/i386.md |  4 ++++
 gcc/config/i386/sse.md  | 16 ++++++++++++++--
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
index 04ec0eeaa57..4cbbd4cf685 100644
--- a/gcc/config/i386/i386.md
+++ b/gcc/config/i386/i386.md
@@ -3683,6 +3683,10 @@
 	      (const_string "avx512bw")
 	   ]
 	   (const_string "*")))
+   (set (attr "mmx_isa")
+     (if_then_else (eq_attr "alternative" "5,6")
+		   (const_string "native")
+		   (const_string "*")))
    (set (attr "type")
      (cond [(eq_attr "alternative" "0,1,2,4")
 	      (const_string "multi")
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 97ec3795b82..4b415d255e0 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -7209,6 +7209,10 @@
 	      (const_string "mmxmov")
 	   ]
 	   (const_string "sselog")))
+   (set (attr "mmx_isa")
+     (if_then_else (eq_attr "alternative" "7,8")
+		   (const_string "native")
+		   (const_string "*")))
    (set (attr "prefix_data16")
      (if_then_else (eq_attr "alternative" "3,4")
 		   (const_string "1")
@@ -7244,7 +7248,8 @@
    movss\t{%1, %0|%0, %1}
    punpckldq\t{%2, %0|%0, %2}
    movd\t{%1, %0|%0, %1}"
-  [(set_attr "type" "sselog,ssemov,mmxcvt,mmxmov")
+  [(set_attr "mmx_isa" "*,*,native,native")
+   (set_attr "type" "sselog,ssemov,mmxcvt,mmxmov")
    (set_attr "mode" "V4SF,SF,DI,DI")])
 
 (define_insn "*vec_concatv4sf"
@@ -14520,6 +14525,10 @@
    punpckldq\t{%2, %0|%0, %2}
    movd\t{%1, %0|%0, %1}"
   [(set_attr "isa" "noavx,noavx,avx,avx512dq,noavx,noavx,avx,*,*,*")
+   (set (attr "mmx_isa")
+     (if_then_else (eq_attr "alternative" "8,9")
+		   (const_string "native")
+		   (const_string "*")))
    (set (attr "type")
      (cond [(eq_attr "alternative" "7")
 	      (const_string "ssemov")
@@ -14557,6 +14566,7 @@
    punpckldq\t{%2, %0|%0, %2}
    movd\t{%1, %0|%0, %1}"
   [(set_attr "isa" "sse2,sse2,*,*,*,*")
+   (set_attr "mmx_isa" "*,*,*,*,native,native")
    (set_attr "type" "sselog,ssemov,sselog,ssemov,mmxcvt,mmxmov")
    (set_attr "mode" "TI,TI,V4SF,SF,DI,DI")])
 
@@ -14586,7 +14596,8 @@
   "@
    %vmovq\t{%1, %0|%0, %1}
    movq2dq\t{%1, %0|%0, %1}"
-  [(set_attr "type" "ssemov")
+  [(set_attr "mmx_isa" "*,native")
+   (set_attr "type" "ssemov")
    (set_attr "prefix" "maybe_vex,orig")
    (set_attr "mode" "TI")])
 
@@ -14661,6 +14672,7 @@
    %vmovq\t{%1, %0|%0, %1}
    movq2dq\t{%1, %0|%0, %1}"
   [(set_attr "isa" "x64,*,*")
+   (set_attr "mmx_isa" "*,*,native")
    (set_attr "type" "ssemov")
    (set_attr "prefix_rex" "1,*,*")
    (set_attr "prefix" "maybe_vex,maybe_vex,orig")
-- 
2.20.1

[PATCH 23/41] i386: Emulate MMX mmx_uavgv4hi3 with SSE

[H.J. Lu]

Emulate MMX mmx_uavgv4hi3 with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_uavgv4hi3): Also check TARGET_MMX and
	TARGET_MMX_WITH_SSE.
	(*mmx_uavgv4hi3): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 8866354dea9..d647dc28baa 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1736,33 +1736,39 @@
 	    (plus:V4SI
 	      (plus:V4SI
 		(zero_extend:V4SI
-		  (match_operand:V4HI 1 "nonimmediate_operand"))
+		  (match_operand:V4HI 1 "register_mmxmem_operand"))
 		(zero_extend:V4SI
-		  (match_operand:V4HI 2 "nonimmediate_operand")))
+		  (match_operand:V4HI 2 "register_mmxmem_operand")))
 	      (const_vector:V4SI [(const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)]))
 	    (const_int 1))))]
-  "TARGET_SSE || TARGET_3DNOW_A"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
   "ix86_fixup_binary_operands_no_copy (PLUS, V4HImode, operands);")
 
 (define_insn "*mmx_uavgv4hi3"
-  [(set (match_operand:V4HI 0 "register_operand" "=y")
+  [(set (match_operand:V4HI 0 "register_operand" "=y,x,Yv")
 	(truncate:V4HI
 	  (lshiftrt:V4SI
 	    (plus:V4SI
 	      (plus:V4SI
 		(zero_extend:V4SI
-		  (match_operand:V4HI 1 "nonimmediate_operand" "%0"))
+		  (match_operand:V4HI 1 "register_mmxmem_operand" "%0,0,Yv"))
 		(zero_extend:V4SI
-		  (match_operand:V4HI 2 "nonimmediate_operand" "ym")))
+		  (match_operand:V4HI 2 "register_mmxmem_operand" "ym,x,Yv")))
 	      (const_vector:V4SI [(const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)]))
 	    (const_int 1))))]
-  "(TARGET_SSE || TARGET_3DNOW_A)
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ix86_binary_operator_ok (PLUS, V4HImode, operands)"
-  "pavgw\t{%2, %0|%0, %2}"
-  [(set_attr "type" "mmxshft")
-   (set_attr "mode" "DI")])
+  "@
+   pavgw\t{%2, %0|%0, %2}
+   pavgw\t{%2, %0|%0, %2}
+   vpavgw\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sseiadd,sseiadd")
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_insn "mmx_psadbw"
   [(set (match_operand:V1DI 0 "register_operand" "=y")
-- 
2.20.1

[PATCH 39/41] i386: Allow MMX intrinsic emulation with SSE

[H.J. Lu]

Allow MMX intrinsic emulation with SSE/SSE2/SSSE3.  Don't enable MMX ISA
by default with TARGET_MMX_WITH_SSE.

For pr82483-1.c and pr82483-2.c, "-mssse3 -mno-mmx" compiles in 64-bit
mode since MMX intrinsics can be emulated wit SSE.

gcc/

	PR target/89021
	* config/i386/i386-builtin.def: Enable MMX intrinsics with
	SSE/SSE2/SSSE3.
	* config/i386/i386.c (ix86_init_mmx_sse_builtins): Likewise.
	(ix86_expand_builtin): Allow SSE/SSE2/SSSE3 to emulate MMX
	intrinsics with TARGET_MMX_WITH_SSE.
	* config/i386/mmintrin.h: Only require SSE2 if __MMX_WITH_SSE__
	is defined.

gcc/testsuite/

	PR target/89021
	* gcc.target/i386/pr82483-1.c: Error only on ia32.
	* gcc.target/i386/pr82483-2.c: Likewise.
---
 gcc/config/i386/i386-builtin.def          | 126 +++++++++++-----------
 gcc/config/i386/i386.c                    |  29 ++++-
 gcc/config/i386/mmintrin.h                |  12 ++-
 gcc/testsuite/gcc.target/i386/pr82483-1.c |   2 +-
 gcc/testsuite/gcc.target/i386/pr82483-2.c |   2 +-
 5 files changed, 101 insertions(+), 70 deletions(-)

diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index 88005f4687f..10a9d631f29 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -100,7 +100,7 @@ BDESC (0, 0, CODE_FOR_fnstsw, "__builtin_ia32_fnstsw", IX86_BUILTIN_FNSTSW, UNKN
 BDESC (0, 0, CODE_FOR_fnclex, "__builtin_ia32_fnclex", IX86_BUILTIN_FNCLEX, UNKNOWN, (int) VOID_FTYPE_VOID)
 
 /* MMX */
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_emms, "__builtin_ia32_emms", IX86_BUILTIN_EMMS, UNKNOWN, (int) VOID_FTYPE_VOID)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_emms, "__builtin_ia32_emms", IX86_BUILTIN_EMMS, UNKNOWN, (int) VOID_FTYPE_VOID)
 
 /* 3DNow! */
 BDESC (OPTION_MASK_ISA_3DNOW, 0, CODE_FOR_mmx_femms, "__builtin_ia32_femms", IX86_BUILTIN_FEMMS, UNKNOWN, (int) VOID_FTYPE_VOID)
@@ -442,68 +442,68 @@ BDESC (0, 0, CODE_FOR_rotrqi3, "__builtin_ia32_rorqi", IX86_BUILTIN_RORQI, UNKNO
 BDESC (0, 0, CODE_FOR_rotrhi3, "__builtin_ia32_rorhi", IX86_BUILTIN_RORHI, UNKNOWN, (int) UINT16_FTYPE_UINT16_INT)
 
 /* MMX */
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_addv8qi3, "__builtin_ia32_paddb", IX86_BUILTIN_PADDB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_addv4hi3, "__builtin_ia32_paddw", IX86_BUILTIN_PADDW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_addv2si3, "__builtin_ia32_paddd", IX86_BUILTIN_PADDD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_subv8qi3, "__builtin_ia32_psubb", IX86_BUILTIN_PSUBB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_subv4hi3, "__builtin_ia32_psubw", IX86_BUILTIN_PSUBW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_subv2si3, "__builtin_ia32_psubd", IX86_BUILTIN_PSUBD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ssaddv8qi3, "__builtin_ia32_paddsb", IX86_BUILTIN_PADDSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ssaddv4hi3, "__builtin_ia32_paddsw", IX86_BUILTIN_PADDSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_sssubv8qi3, "__builtin_ia32_psubsb", IX86_BUILTIN_PSUBSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_sssubv4hi3, "__builtin_ia32_psubsw", IX86_BUILTIN_PSUBSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_usaddv8qi3, "__builtin_ia32_paddusb", IX86_BUILTIN_PADDUSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_usaddv4hi3, "__builtin_ia32_paddusw", IX86_BUILTIN_PADDUSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ussubv8qi3, "__builtin_ia32_psubusb", IX86_BUILTIN_PSUBUSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ussubv4hi3, "__builtin_ia32_psubusw", IX86_BUILTIN_PSUBUSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_mulv4hi3, "__builtin_ia32_pmullw", IX86_BUILTIN_PMULLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_smulv4hi3_highpart, "__builtin_ia32_pmulhw", IX86_BUILTIN_PMULHW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_andv2si3, "__builtin_ia32_pand", IX86_BUILTIN_PAND, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_andnotv2si3, "__builtin_ia32_pandn", IX86_BUILTIN_PANDN, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_iorv2si3, "__builtin_ia32_por", IX86_BUILTIN_POR, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_xorv2si3, "__builtin_ia32_pxor", IX86_BUILTIN_PXOR, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_eqv8qi3, "__builtin_ia32_pcmpeqb", IX86_BUILTIN_PCMPEQB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_eqv4hi3, "__builtin_ia32_pcmpeqw", IX86_BUILTIN_PCMPEQW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_eqv2si3, "__builtin_ia32_pcmpeqd", IX86_BUILTIN_PCMPEQD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_gtv8qi3, "__builtin_ia32_pcmpgtb", IX86_BUILTIN_PCMPGTB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_gtv4hi3, "__builtin_ia32_pcmpgtw", IX86_BUILTIN_PCMPGTW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_gtv2si3, "__builtin_ia32_pcmpgtd", IX86_BUILTIN_PCMPGTD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpckhbw, "__builtin_ia32_punpckhbw", IX86_BUILTIN_PUNPCKHBW, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpckhwd, "__builtin_ia32_punpckhwd", IX86_BUILTIN_PUNPCKHWD, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpckhdq, "__builtin_ia32_punpckhdq", IX86_BUILTIN_PUNPCKHDQ, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpcklbw, "__builtin_ia32_punpcklbw", IX86_BUILTIN_PUNPCKLBW, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpcklwd, "__builtin_ia32_punpcklwd", IX86_BUILTIN_PUNPCKLWD, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_punpckldq, "__builtin_ia32_punpckldq", IX86_BUILTIN_PUNPCKLDQ, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_packsswb, "__builtin_ia32_packsswb", IX86_BUILTIN_PACKSSWB, UNKNOWN, (int) V8QI_FTYPE_V4HI_V4HI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_packssdw, "__builtin_ia32_packssdw", IX86_BUILTIN_PACKSSDW, UNKNOWN, (int) V4HI_FTYPE_V2SI_V2SI)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_packuswb, "__builtin_ia32_packuswb", IX86_BUILTIN_PACKUSWB, UNKNOWN, (int) V8QI_FTYPE_V4HI_V4HI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_pmaddwd, "__builtin_ia32_pmaddwd", IX86_BUILTIN_PMADDWD, UNKNOWN, (int) V2SI_FTYPE_V4HI_V4HI)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv4hi3, "__builtin_ia32_psllwi", IX86_BUILTIN_PSLLWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv2si3, "__builtin_ia32_pslldi", IX86_BUILTIN_PSLLDI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv1di3, "__builtin_ia32_psllqi", IX86_BUILTIN_PSLLQI, UNKNOWN, (int) V1DI_FTYPE_V1DI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv4hi3, "__builtin_ia32_psllw", IX86_BUILTIN_PSLLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv2si3, "__builtin_ia32_pslld", IX86_BUILTIN_PSLLD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashlv1di3, "__builtin_ia32_psllq", IX86_BUILTIN_PSLLQ, UNKNOWN, (int) V1DI_FTYPE_V1DI_V1DI_COUNT)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv4hi3, "__builtin_ia32_psrlwi", IX86_BUILTIN_PSRLWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv2si3, "__builtin_ia32_psrldi", IX86_BUILTIN_PSRLDI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv1di3, "__builtin_ia32_psrlqi", IX86_BUILTIN_PSRLQI, UNKNOWN, (int) V1DI_FTYPE_V1DI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv4hi3, "__builtin_ia32_psrlw", IX86_BUILTIN_PSRLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv2si3, "__builtin_ia32_psrld", IX86_BUILTIN_PSRLD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_lshrv1di3, "__builtin_ia32_psrlq", IX86_BUILTIN_PSRLQ, UNKNOWN, (int) V1DI_FTYPE_V1DI_V1DI_COUNT)
-
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashrv4hi3, "__builtin_ia32_psrawi", IX86_BUILTIN_PSRAWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashrv2si3, "__builtin_ia32_psradi", IX86_BUILTIN_PSRADI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashrv4hi3, "__builtin_ia32_psraw", IX86_BUILTIN_PSRAW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
-BDESC (OPTION_MASK_ISA_MMX, 0, CODE_FOR_mmx_ashrv2si3, "__builtin_ia32_psrad", IX86_BUILTIN_PSRAD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_addv8qi3, "__builtin_ia32_paddb", IX86_BUILTIN_PADDB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_addv4hi3, "__builtin_ia32_paddw", IX86_BUILTIN_PADDW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_addv2si3, "__builtin_ia32_paddd", IX86_BUILTIN_PADDD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_subv8qi3, "__builtin_ia32_psubb", IX86_BUILTIN_PSUBB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_subv4hi3, "__builtin_ia32_psubw", IX86_BUILTIN_PSUBW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_subv2si3, "__builtin_ia32_psubd", IX86_BUILTIN_PSUBD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ssaddv8qi3, "__builtin_ia32_paddsb", IX86_BUILTIN_PADDSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ssaddv4hi3, "__builtin_ia32_paddsw", IX86_BUILTIN_PADDSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_sssubv8qi3, "__builtin_ia32_psubsb", IX86_BUILTIN_PSUBSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_sssubv4hi3, "__builtin_ia32_psubsw", IX86_BUILTIN_PSUBSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_usaddv8qi3, "__builtin_ia32_paddusb", IX86_BUILTIN_PADDUSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_usaddv4hi3, "__builtin_ia32_paddusw", IX86_BUILTIN_PADDUSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ussubv8qi3, "__builtin_ia32_psubusb", IX86_BUILTIN_PSUBUSB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ussubv4hi3, "__builtin_ia32_psubusw", IX86_BUILTIN_PSUBUSW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_mulv4hi3, "__builtin_ia32_pmullw", IX86_BUILTIN_PMULLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_smulv4hi3_highpart, "__builtin_ia32_pmulhw", IX86_BUILTIN_PMULHW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_andv2si3, "__builtin_ia32_pand", IX86_BUILTIN_PAND, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_andnotv2si3, "__builtin_ia32_pandn", IX86_BUILTIN_PANDN, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_iorv2si3, "__builtin_ia32_por", IX86_BUILTIN_POR, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_xorv2si3, "__builtin_ia32_pxor", IX86_BUILTIN_PXOR, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_eqv8qi3, "__builtin_ia32_pcmpeqb", IX86_BUILTIN_PCMPEQB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_eqv4hi3, "__builtin_ia32_pcmpeqw", IX86_BUILTIN_PCMPEQW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_eqv2si3, "__builtin_ia32_pcmpeqd", IX86_BUILTIN_PCMPEQD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_gtv8qi3, "__builtin_ia32_pcmpgtb", IX86_BUILTIN_PCMPGTB, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_gtv4hi3, "__builtin_ia32_pcmpgtw", IX86_BUILTIN_PCMPGTW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_gtv2si3, "__builtin_ia32_pcmpgtd", IX86_BUILTIN_PCMPGTD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpckhbw, "__builtin_ia32_punpckhbw", IX86_BUILTIN_PUNPCKHBW, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpckhwd, "__builtin_ia32_punpckhwd", IX86_BUILTIN_PUNPCKHWD, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpckhdq, "__builtin_ia32_punpckhdq", IX86_BUILTIN_PUNPCKHDQ, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpcklbw, "__builtin_ia32_punpcklbw", IX86_BUILTIN_PUNPCKLBW, UNKNOWN, (int) V8QI_FTYPE_V8QI_V8QI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpcklwd, "__builtin_ia32_punpcklwd", IX86_BUILTIN_PUNPCKLWD, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_punpckldq, "__builtin_ia32_punpckldq", IX86_BUILTIN_PUNPCKLDQ, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_packsswb, "__builtin_ia32_packsswb", IX86_BUILTIN_PACKSSWB, UNKNOWN, (int) V8QI_FTYPE_V4HI_V4HI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_packssdw, "__builtin_ia32_packssdw", IX86_BUILTIN_PACKSSDW, UNKNOWN, (int) V4HI_FTYPE_V2SI_V2SI)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_packuswb, "__builtin_ia32_packuswb", IX86_BUILTIN_PACKUSWB, UNKNOWN, (int) V8QI_FTYPE_V4HI_V4HI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_pmaddwd, "__builtin_ia32_pmaddwd", IX86_BUILTIN_PMADDWD, UNKNOWN, (int) V2SI_FTYPE_V4HI_V4HI)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv4hi3, "__builtin_ia32_psllwi", IX86_BUILTIN_PSLLWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv2si3, "__builtin_ia32_pslldi", IX86_BUILTIN_PSLLDI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv1di3, "__builtin_ia32_psllqi", IX86_BUILTIN_PSLLQI, UNKNOWN, (int) V1DI_FTYPE_V1DI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv4hi3, "__builtin_ia32_psllw", IX86_BUILTIN_PSLLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv2si3, "__builtin_ia32_pslld", IX86_BUILTIN_PSLLD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashlv1di3, "__builtin_ia32_psllq", IX86_BUILTIN_PSLLQ, UNKNOWN, (int) V1DI_FTYPE_V1DI_V1DI_COUNT)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv4hi3, "__builtin_ia32_psrlwi", IX86_BUILTIN_PSRLWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv2si3, "__builtin_ia32_psrldi", IX86_BUILTIN_PSRLDI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv1di3, "__builtin_ia32_psrlqi", IX86_BUILTIN_PSRLQI, UNKNOWN, (int) V1DI_FTYPE_V1DI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv4hi3, "__builtin_ia32_psrlw", IX86_BUILTIN_PSRLW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv2si3, "__builtin_ia32_psrld", IX86_BUILTIN_PSRLD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_lshrv1di3, "__builtin_ia32_psrlq", IX86_BUILTIN_PSRLQ, UNKNOWN, (int) V1DI_FTYPE_V1DI_V1DI_COUNT)
+
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashrv4hi3, "__builtin_ia32_psrawi", IX86_BUILTIN_PSRAWI, UNKNOWN, (int) V4HI_FTYPE_V4HI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashrv2si3, "__builtin_ia32_psradi", IX86_BUILTIN_PSRADI, UNKNOWN, (int) V2SI_FTYPE_V2SI_SI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashrv4hi3, "__builtin_ia32_psraw", IX86_BUILTIN_PSRAW, UNKNOWN, (int) V4HI_FTYPE_V4HI_V4HI_COUNT)
+BDESC (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0, CODE_FOR_mmx_ashrv2si3, "__builtin_ia32_psrad", IX86_BUILTIN_PSRAD, UNKNOWN, (int) V2SI_FTYPE_V2SI_V2SI_COUNT)
 
 /* 3DNow! */
 BDESC (OPTION_MASK_ISA_3DNOW, 0, CODE_FOR_mmx_pf2id, "__builtin_ia32_pf2id", IX86_BUILTIN_PF2ID, UNKNOWN, (int) V2SI_FTYPE_V2SF)
diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 25e0dc43a9e..073a2534d1f 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -31764,14 +31764,17 @@ ix86_init_mmx_sse_builtins (void)
 	       VOID_FTYPE_UNSIGNED, IX86_BUILTIN_XABORT);
 
   /* MMX access to the vec_init patterns.  */
-  def_builtin_const (OPTION_MASK_ISA_MMX, 0, "__builtin_ia32_vec_init_v2si",
+  def_builtin_const (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0,
+		     "__builtin_ia32_vec_init_v2si",
 		     V2SI_FTYPE_INT_INT, IX86_BUILTIN_VEC_INIT_V2SI);
 
-  def_builtin_const (OPTION_MASK_ISA_MMX, 0, "__builtin_ia32_vec_init_v4hi",
+  def_builtin_const (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0,
+		     "__builtin_ia32_vec_init_v4hi",
 		     V4HI_FTYPE_HI_HI_HI_HI,
 		     IX86_BUILTIN_VEC_INIT_V4HI);
 
-  def_builtin_const (OPTION_MASK_ISA_MMX, 0, "__builtin_ia32_vec_init_v8qi",
+  def_builtin_const (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0,
+		     "__builtin_ia32_vec_init_v8qi",
 		     V8QI_FTYPE_QI_QI_QI_QI_QI_QI_QI_QI,
 		     IX86_BUILTIN_VEC_INIT_V8QI);
 
@@ -31793,7 +31796,8 @@ ix86_init_mmx_sse_builtins (void)
 		     "__builtin_ia32_vec_ext_v4hi",
 		     HI_FTYPE_V4HI_INT, IX86_BUILTIN_VEC_EXT_V4HI);
 
-  def_builtin_const (OPTION_MASK_ISA_MMX, 0, "__builtin_ia32_vec_ext_v2si",
+  def_builtin_const (OPTION_MASK_ISA_MMX | OPTION_MASK_ISA_SSE2, 0,
+		     "__builtin_ia32_vec_ext_v2si",
 		     SI_FTYPE_V2SI_INT, IX86_BUILTIN_VEC_EXT_V2SI);
 
   def_builtin_const (OPTION_MASK_ISA_SSE2, 0, "__builtin_ia32_vec_ext_v16qi",
@@ -36926,6 +36930,23 @@ ix86_expand_builtin (tree exp, rtx target, rtx subtarget,
        == (OPTION_MASK_ISA_FMA | OPTION_MASK_ISA_FMA4))
       && (isa & (OPTION_MASK_ISA_FMA | OPTION_MASK_ISA_FMA4)) != 0)
     isa |= (OPTION_MASK_ISA_FMA | OPTION_MASK_ISA_FMA4);
+  /* Use SSE/SSE2/SSSE3 to emulate MMX intrinsics in 64-bit mode when
+     MMX is disabled.  */
+  if (TARGET_MMX_WITH_SSE)
+    {
+      if (((bisa & (OPTION_MASK_ISA_SSE | OPTION_MASK_ISA_MMX))
+	   == (OPTION_MASK_ISA_SSE | OPTION_MASK_ISA_MMX))
+	  && (isa & (OPTION_MASK_ISA_SSE | OPTION_MASK_ISA_MMX)) != 0)
+	isa |= (OPTION_MASK_ISA_SSE | OPTION_MASK_ISA_MMX);
+      if (((bisa & (OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_MMX))
+	   == (OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_MMX))
+	  && (isa & (OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_MMX)) != 0)
+	isa |= (OPTION_MASK_ISA_SSE2 | OPTION_MASK_ISA_MMX);
+      if (((bisa & (OPTION_MASK_ISA_SSSE3 | OPTION_MASK_ISA_MMX))
+	   == (OPTION_MASK_ISA_SSSE3 | OPTION_MASK_ISA_MMX))
+	  && (isa & (OPTION_MASK_ISA_SSSE3 | OPTION_MASK_ISA_MMX)) != 0)
+	isa |= (OPTION_MASK_ISA_SSSE3 | OPTION_MASK_ISA_MMX);
+    }
   if ((bisa & isa) != bisa || (bisa2 & isa2) != bisa2)
     {
       char *opts = ix86_target_string (bisa, bisa2, 0, 0, NULL, NULL,
diff --git a/gcc/config/i386/mmintrin.h b/gcc/config/i386/mmintrin.h
index 238b3df3121..c4b2e0c7b25 100644
--- a/gcc/config/i386/mmintrin.h
+++ b/gcc/config/i386/mmintrin.h
@@ -29,7 +29,9 @@
 
 #if defined __x86_64__ && !defined __SSE__ || !defined __MMX__
 #pragma GCC push_options
-#ifdef __x86_64__
+#ifdef __MMX_WITH_SSE__
+#pragma GCC target("sse2")
+#elif defined __x86_64__
 #pragma GCC target("sse,mmx")
 #else
 #pragma GCC target("mmx")
@@ -315,7 +317,11 @@ _m_paddd (__m64 __m1, __m64 __m2)
 /* Add the 64-bit values in M1 to the 64-bit values in M2.  */
 #ifndef __SSE2__
 #pragma GCC push_options
+#ifdef __MMX_WITH_SSE__
+#pragma GCC target("sse2")
+#else
 #pragma GCC target("sse2,mmx")
+#endif
 #define __DISABLE_SSE2__
 #endif /* __SSE2__ */
 
@@ -427,7 +433,11 @@ _m_psubd (__m64 __m1, __m64 __m2)
 /* Add the 64-bit values in M1 to the 64-bit values in M2.  */
 #ifndef __SSE2__
 #pragma GCC push_options
+#ifdef __MMX_WITH_SSE__
+#pragma GCC target("sse2")
+#else
 #pragma GCC target("sse2,mmx")
+#endif
 #define __DISABLE_SSE2__
 #endif /* __SSE2__ */
 
diff --git a/gcc/testsuite/gcc.target/i386/pr82483-1.c b/gcc/testsuite/gcc.target/i386/pr82483-1.c
index 59a59dc8dfe..b2028d8dc5e 100644
--- a/gcc/testsuite/gcc.target/i386/pr82483-1.c
+++ b/gcc/testsuite/gcc.target/i386/pr82483-1.c
@@ -1,7 +1,7 @@
 /* PR target/82483 */
 /* { dg-do compile } */
 /* { dg-options "-mssse3 -mno-mmx -Wno-psabi" } */
-/* { dg-error "needs isa option" "" { target *-*-* } 0 } */
+/* { dg-error "needs isa option" "" { target ia32 } 0 } */
 
 #include <x86intrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/pr82483-2.c b/gcc/testsuite/gcc.target/i386/pr82483-2.c
index 305ddbd6c64..c92de405cb3 100644
--- a/gcc/testsuite/gcc.target/i386/pr82483-2.c
+++ b/gcc/testsuite/gcc.target/i386/pr82483-2.c
@@ -1,7 +1,7 @@
 /* PR target/82483 */
 /* { dg-do compile } */
 /* { dg-options "-mssse3 -mno-mmx -Wno-psabi" } */
-/* { dg-error "needs isa option" "" { target *-*-* } 0 } */
+/* { dg-error "needs isa option" "" { target ia32 } 0 } */
 
 #include <x86intrin.h>
 
-- 
2.20.1

[PATCH 22/41] i386: Emulate MMX mmx_uavgv8qi3 with SSE

[H.J. Lu]

Emulate MMX mmx_uavgv8qi3 with SSE.  Only SSE register source operand is
allowed.

	PR target/89021
	* config/i386/mmx.md (mmx_uavgv8qi3): Also check TARGET_MMX
	and TARGET_MMX_WITH_SSE.
	(*mmx_uavgv8qi3): Add SSE emulation.
---
 gcc/config/i386/mmx.md | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 5a342256cbc..8866354dea9 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -1679,50 +1679,55 @@
 	    (plus:V8HI
 	      (plus:V8HI
 		(zero_extend:V8HI
-		  (match_operand:V8QI 1 "nonimmediate_operand"))
+		  (match_operand:V8QI 1 "register_mmxmem_operand"))
 		(zero_extend:V8HI
-		  (match_operand:V8QI 2 "nonimmediate_operand")))
+		  (match_operand:V8QI 2 "register_mmxmem_operand")))
 	      (const_vector:V8HI [(const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)]))
 	    (const_int 1))))]
-  "TARGET_SSE || TARGET_3DNOW"
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)"
   "ix86_fixup_binary_operands_no_copy (PLUS, V8QImode, operands);")
 
 (define_insn "*mmx_uavgv8qi3"
-  [(set (match_operand:V8QI 0 "register_operand" "=y")
+  [(set (match_operand:V8QI 0 "register_operand" "=y,x,Yv")
 	(truncate:V8QI
 	  (lshiftrt:V8HI
 	    (plus:V8HI
 	      (plus:V8HI
 		(zero_extend:V8HI
-		  (match_operand:V8QI 1 "nonimmediate_operand" "%0"))
+		  (match_operand:V8QI 1 "register_mmxmem_operand" "%0,0,Yv"))
 		(zero_extend:V8HI
-		  (match_operand:V8QI 2 "nonimmediate_operand" "ym")))
+		  (match_operand:V8QI 2 "register_mmxmem_operand" "ym,x,Yv")))
 	      (const_vector:V8HI [(const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)
 				  (const_int 1) (const_int 1)]))
 	    (const_int 1))))]
-  "(TARGET_SSE || TARGET_3DNOW)
+  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
+   && (TARGET_SSE || TARGET_3DNOW_A)
    && ix86_binary_operator_ok (PLUS, V8QImode, operands)"
 {
   /* These two instructions have the same operation, but their encoding
      is different.  Prefer the one that is de facto standard.  */
-  if (TARGET_SSE || TARGET_3DNOW_A)
+  if (TARGET_MMX_WITH_SSE && TARGET_AVX)
+    return "vpavgb\t{%2, %1, %0|%0, %1, %2}";
+  else if (TARGET_SSE || TARGET_3DNOW_A)
     return "pavgb\t{%2, %0|%0, %2}";
   else
     return "pavgusb\t{%2, %0|%0, %2}";
 }
-  [(set_attr "type" "mmxshft")
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
+   (set_attr "type" "mmxshft,sseiadd,sseiadd")
    (set (attr "prefix_extra")
      (if_then_else
        (not (ior (match_test "TARGET_SSE")
 		 (match_test "TARGET_3DNOW_A")))
        (const_string "1")
        (const_string "*")))
-   (set_attr "mode" "DI")])
+   (set_attr "mode" "DI,TI,TI")])
 
 (define_expand "mmx_uavgv4hi3"
   [(set (match_operand:V4HI 0 "register_operand")
-- 
2.20.1

Re: [PATCH 27/41] i386: Make _mm_empty () as NOP for TARGET_MMX_WITH_SSE

[Uros Bizjak]

On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> With SSE emulation of MMX intrinsics, we should make _mm_empty () as NOP
> for TARGET_MMX_WITH_SSE.

NAK. NOP can be emitter only for !TARGET_MMX.

Uros.

>
> 	PR target/89021
> 	* config/i386/mmx.md (mmx_<emms>): Renamed to ...
> 	(*mmx_<emms>): This.
> 	(mmx_<emms>): New expander.
> ---
>  gcc/config/i386/mmx.md | 30 +++++++++++++++++++++++++++++-
>  1 file changed, 29 insertions(+), 1 deletion(-)
>
> diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
> index f27513f7f2c..8b5fe3948e2 100644
> --- a/gcc/config/i386/mmx.md
> +++ b/gcc/config/i386/mmx.md
> @@ -1849,7 +1849,35 @@
>    [(UNSPECV_EMMS "emms")
>     (UNSPECV_FEMMS "femms")])
>
> -(define_insn "mmx_<emms>"
> +(define_expand "mmx_<emms>"
> +  [(parallel
> +    [(unspec_volatile [(const_int 0)] EMMS)
> +      (clobber (reg:XF ST0_REG))
> +      (clobber (reg:XF ST1_REG))
> +      (clobber (reg:XF ST2_REG))
> +      (clobber (reg:XF ST3_REG))
> +      (clobber (reg:XF ST4_REG))
> +      (clobber (reg:XF ST5_REG))
> +      (clobber (reg:XF ST6_REG))
> +      (clobber (reg:XF ST7_REG))
> +      (clobber (reg:DI MM0_REG))
> +      (clobber (reg:DI MM1_REG))
> +      (clobber (reg:DI MM2_REG))
> +      (clobber (reg:DI MM3_REG))
> +      (clobber (reg:DI MM4_REG))
> +      (clobber (reg:DI MM5_REG))
> +      (clobber (reg:DI MM6_REG))
> +      (clobber (reg:DI MM7_REG))])]
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
> +{
> +   if (TARGET_MMX_WITH_SSE)
> +     {
> +       emit_insn (gen_nop ());
> +       DONE;
> +     }
> +})
> +
> +(define_insn "*mmx_<emms>"
>    [(unspec_volatile [(const_int 0)] EMMS)
>     (clobber (reg:XF ST0_REG))
>     (clobber (reg:XF ST1_REG))
> --
> 2.20.1
>
>

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> On x86-64, since __m64 is returned and passed in XMM registers, we can
> emulate MMX intrinsics with SSE instructions. To support it, we added
>
>  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
>
> ;; Define instruction set of MMX instructions
> (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
>   (const_string "base"))
>
>          (eq_attr "mmx_isa" "native")
>            (symbol_ref "!TARGET_MMX_WITH_SSE")
>          (eq_attr "mmx_isa" "x64")
>            (symbol_ref "TARGET_MMX_WITH_SSE")
>          (eq_attr "mmx_isa" "x64_avx")
>            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
>          (eq_attr "mmx_isa" "x64_noavx")
>            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
>
> We added SSE emulation to MMX patterns and disabled MMX alternatives with
> TARGET_MMX_WITH_SSE.
>
> Most of MMX instructions have equivalent SSE versions and results of some
> SSE versions need to be reshuffled to the right order for MMX.  Thee are
> couple tricky cases:
>
> 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> mask operand and handle unmapped bits 64:127 at memory address by
> adjusting source and mask operands together with memory address.
>
> 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> in 64-bit mode.
>
> 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> SSE emulation must clear the bit 4 in the shuffle control mask.
>
> 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> the upper 64 bits of destination XMM register.
>
> Tests are also added to check each SSE emulation of MMX intrinsics.
>
> There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> tested with
>
> --with-arch=native --with-cpu=native
>
> on AVX2 and AVX512F machines.

An idea that would take patch a step further also on 32 bit targets:

*Assuming* that operations on XMM registers are as fast (or perhaps
faster) than operations on MMX registers, we can change mmx_isa
attribute in e.g.

+  "@
+   p<logic>\t{%2, %0|%0, %2}
+   p<logic>\t{%2, %0|%0, %2}
+   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")

to:

[(set_attr "isa" "*,noavx,avx")
 (set_attr "mmx_isa" "native,*,*")]

So, for x86_64 everything stays the same, but for x86_32 we now allow
intrinsics to use xmm registers in addition to mmx registers. We can't
disable MMX for x64_32 anyway due to ISA constraints (and some tricky
cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
& similar, which are more efficient with MMX regs), but RA has much
more freedom to allocate the most effective register set even for
32bit targets.

WDYT?

Uros.

>
> H.J. Lu (40):
>   i386: Allow MMX register modes in SSE registers
>   i386: Emulate MMX packsswb/packssdw/packuswb with SSE2
>   i386: Emulate MMX punpcklXX/punpckhXX with SSE punpcklXX
>   i386: Emulate MMX plusminus/sat_plusminus with SSE
>   i386: Emulate MMX mulv4hi3 with SSE
>   i386: Emulate MMX smulv4hi3_highpart with SSE
>   i386: Emulate MMX mmx_pmaddwd with SSE
>   i386: Emulate MMX ashr<mode>3/<shift_insn><mode>3 with SSE
>   i386: Emulate MMX <any_logic><mode>3 with SSE
>   i386: Emulate MMX mmx_andnot<mode>3 with SSE
>   i386: Emulate MMX mmx_eq/mmx_gt<mode>3 with SSE
>   i386: Emulate MMX vec_dupv2si with SSE
>   i386: Emulate MMX pshufw with SSE
>   i386: Emulate MMX sse_cvtps2pi/sse_cvttps2pi with SSE
>   i386: Emulate MMX sse_cvtpi2ps with SSE
>   i386: Emulate MMX mmx_pextrw with SSE
>   i386: Emulate MMX mmx_pinsrw with SSE
>   i386: Emulate MMX V4HI smaxmin/V8QI umaxmin with SSE
>   i386: Emulate MMX mmx_pmovmskb with SSE
>   i386: Emulate MMX mmx_umulv4hi3_highpart with SSE
>   i386: Emulate MMX maskmovq with SSE2 maskmovdqu
>   i386: Emulate MMX mmx_uavgv8qi3 with SSE
>   i386: Emulate MMX mmx_uavgv4hi3 with SSE
>   i386: Emulate MMX mmx_psadbw with SSE
>   i386: Emulate MMX movntq with SSE2 movntidi
>   i386: Emulate MMX umulv1siv1di3 with SSE2
>   i386: Make _mm_empty () as NOP for TARGET_MMX_WITH_SSE
>   i386: Emulate MMX ssse3_ph<plusminus_mnemonic>wv4hi3 with SSE
>   i386: Emulate MMX ssse3_ph<plusminus_mnemonic>dv2si3 with SSE
>   i386: Emulate MMX ssse3_pmaddubsw with SSE
>   i386: Emulate MMX ssse3_pmulhrswv4hi3 with SSE
>   i386: Emulate MMX pshufb with SSE version
>   i386: Emulate MMX ssse3_psign<mode>3 with SSE
>   i386: Emulate MMX ssse3_palignrdi with SSE
>   i386: Emulate MMX abs<mode>2 with SSE
>   i386: Allow MMXMODE moves with TARGET_MMX_WITH_SSE
>   i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE
>   i386: Allow MMX intrinsic emulation with SSE
>   i386: Enable TM MMX intrinsics with SSE2
>   i386: Add tests for MMX intrinsic emulations with SSE
>
> Uros Bizjak (1):
>   Prevent allocation of MMX registers with TARGET_MMX_WITH_SSE
>
>  gcc/config/i386/constraints.md                |   6 +
>  gcc/config/i386/i386-builtin.def              | 126 +--
>  gcc/config/i386/i386-c.c                      |   2 +
>  gcc/config/i386/i386-protos.h                 |   4 +
>  gcc/config/i386/i386.c                        | 189 +++-
>  gcc/config/i386/i386.h                        |   2 +
>  gcc/config/i386/i386.md                       |  17 +
>  gcc/config/i386/mmintrin.h                    |  12 +-
>  gcc/config/i386/mmx.md                        | 984 ++++++++++++------
>  gcc/config/i386/predicates.md                 |   7 +
>  gcc/config/i386/sse.md                        | 359 +++++--
>  gcc/config/i386/xmmintrin.h                   |  61 ++
>  gcc/testsuite/gcc.target/i386/mmx-vals.h      |  77 ++
>  gcc/testsuite/gcc.target/i386/pr82483-1.c     |   2 +-
>  gcc/testsuite/gcc.target/i386/pr82483-2.c     |   2 +-
>  gcc/testsuite/gcc.target/i386/sse2-mmx-10.c   |  43 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-11.c   |  39 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-12.c   |  42 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-13.c   |  40 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-14.c   |  31 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-15.c   |  36 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-16.c   |  40 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-17.c   |  51 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c  |  14 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c  |  14 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c  |   7 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-2.c    |  12 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-20.c   |  12 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-21.c   |  13 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-22.c   |  14 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-3.c    |  13 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-4.c    |   4 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-5.c    |  11 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-6.c    |  11 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-7.c    |  13 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-8.c    |   4 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-9.c    |  79 ++
>  .../gcc.target/i386/sse2-mmx-cvtpi2ps.c       |  43 +
>  .../gcc.target/i386/sse2-mmx-cvtps2pi.c       |  36 +
>  .../gcc.target/i386/sse2-mmx-cvttps2pi.c      |  36 +
>  .../gcc.target/i386/sse2-mmx-maskmovq.c       |  99 ++
>  .../gcc.target/i386/sse2-mmx-packssdw.c       |  52 +
>  .../gcc.target/i386/sse2-mmx-packsswb.c       |  52 +
>  .../gcc.target/i386/sse2-mmx-packuswb.c       |  52 +
>  .../gcc.target/i386/sse2-mmx-paddb.c          |  48 +
>  .../gcc.target/i386/sse2-mmx-paddd.c          |  48 +
>  .../gcc.target/i386/sse2-mmx-paddq.c          |  43 +
>  .../gcc.target/i386/sse2-mmx-paddsb.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-paddsw.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-paddusb.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-paddusw.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-paddw.c          |  48 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c |  44 +
>  .../gcc.target/i386/sse2-mmx-pandn.c          |  44 +
>  .../gcc.target/i386/sse2-mmx-pavgb.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-pavgw.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-pcmpeqb.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pcmpeqd.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pcmpeqw.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pcmpgtb.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pcmpgtd.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pcmpgtw.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-pextrw.c         |  59 ++
>  .../gcc.target/i386/sse2-mmx-pinsrw.c         |  61 ++
>  .../gcc.target/i386/sse2-mmx-pmaddwd.c        |  47 +
>  .../gcc.target/i386/sse2-mmx-pmaxsw.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-pmaxub.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-pminsw.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-pminub.c         |  48 +
>  .../gcc.target/i386/sse2-mmx-pmovmskb.c       |  46 +
>  .../gcc.target/i386/sse2-mmx-pmulhuw.c        |  51 +
>  .../gcc.target/i386/sse2-mmx-pmulhw.c         |  53 +
>  .../gcc.target/i386/sse2-mmx-pmullw.c         |  52 +
>  .../gcc.target/i386/sse2-mmx-pmuludq.c        |  47 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-por.c  |  44 +
>  .../gcc.target/i386/sse2-mmx-psadbw.c         |  58 ++
>  .../gcc.target/i386/sse2-mmx-pshufw.c         | 248 +++++
>  .../gcc.target/i386/sse2-mmx-pslld.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-pslldi.c         | 153 +++
>  .../gcc.target/i386/sse2-mmx-psllq.c          |  47 +
>  .../gcc.target/i386/sse2-mmx-psllqi.c         | 245 +++++
>  .../gcc.target/i386/sse2-mmx-psllw.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-psllwi.c         | 105 ++
>  .../gcc.target/i386/sse2-mmx-psrad.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-psradi.c         | 153 +++
>  .../gcc.target/i386/sse2-mmx-psraw.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-psrawi.c         | 105 ++
>  .../gcc.target/i386/sse2-mmx-psrld.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-psrldi.c         | 153 +++
>  .../gcc.target/i386/sse2-mmx-psrlq.c          |  47 +
>  .../gcc.target/i386/sse2-mmx-psrlqi.c         | 245 +++++
>  .../gcc.target/i386/sse2-mmx-psrlw.c          |  52 +
>  .../gcc.target/i386/sse2-mmx-psrlwi.c         | 105 ++
>  .../gcc.target/i386/sse2-mmx-psubb.c          |  48 +
>  .../gcc.target/i386/sse2-mmx-psubd.c          |  48 +
>  .../gcc.target/i386/sse2-mmx-psubq.c          |  43 +
>  .../gcc.target/i386/sse2-mmx-psubusb.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-psubusw.c        |  48 +
>  .../gcc.target/i386/sse2-mmx-psubw.c          |  48 +
>  .../gcc.target/i386/sse2-mmx-punpckhbw.c      |  53 +
>  .../gcc.target/i386/sse2-mmx-punpckhdq.c      |  47 +
>  .../gcc.target/i386/sse2-mmx-punpckhwd.c      |  49 +
>  .../gcc.target/i386/sse2-mmx-punpcklbw.c      |  53 +
>  .../gcc.target/i386/sse2-mmx-punpckldq.c      |  47 +
>  .../gcc.target/i386/sse2-mmx-punpcklwd.c      |  49 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c |  44 +
>  gcc/testsuite/gcc.target/i386/sse2-mmx.c      |   1 -
>  112 files changed, 6418 insertions(+), 493 deletions(-)
>  create mode 100644 gcc/testsuite/gcc.target/i386/mmx-vals.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-10.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-11.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-12.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-13.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-14.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-15.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-16.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-17.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18a.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18b.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-18c.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19a.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19b.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19c.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19d.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-19e.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-20.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-21.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-22.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-3.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-4.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-5.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-6.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-7.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-8.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-9.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtpi2ps.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvtps2pi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-cvttps2pi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-maskmovq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packssdw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packsswb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-packuswb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddsw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddusw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-paddw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pand.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pandn.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pavgw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpeqw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pcmpgtw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pextrw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pinsrw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaddwd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxsw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmaxub.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminsw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pminub.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmovmskb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhuw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmulhw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmullw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pmuludq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-por.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psadbw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pshufw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslld.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pslldi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllqi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psllwi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrad.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psradi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psraw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrawi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrld.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrldi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlqi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psrlwi.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusb.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubusw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-psubw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhbw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhdq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckhwd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklbw.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpckldq.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-punpcklwd.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sse2-mmx-pxor.c
>
> --
> 2.20.1
>
>

Re: [PATCH 27/41] i386: Make _mm_empty () as NOP for TARGET_MMX_WITH_SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 1:45 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> > With SSE emulation of MMX intrinsics, we should make _mm_empty () as NOP
> > for TARGET_MMX_WITH_SSE.
>
> NAK. NOP can be emitter only for !TARGET_MMX.
>

I changed it to

  "TARGET_MMX || TARGET_MMX_WITH_SSE"
{
   if (!TARGET_MMX)
     {
       emit_insn (gen_nop ());
       DONE;
     }
})


-- 
H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 2:33 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > emulate MMX intrinsics with SSE instructions. To support it, we added
> >
> >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> >
> > ;; Define instruction set of MMX instructions
> > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> >   (const_string "base"))
> >
> >          (eq_attr "mmx_isa" "native")
> >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> >          (eq_attr "mmx_isa" "x64")
> >            (symbol_ref "TARGET_MMX_WITH_SSE")
> >          (eq_attr "mmx_isa" "x64_avx")
> >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> >          (eq_attr "mmx_isa" "x64_noavx")
> >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> >
> > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > TARGET_MMX_WITH_SSE.
> >
> > Most of MMX instructions have equivalent SSE versions and results of some
> > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > couple tricky cases:
> >
> > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > mask operand and handle unmapped bits 64:127 at memory address by
> > adjusting source and mask operands together with memory address.
> >
> > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > in 64-bit mode.
> >
> > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > SSE emulation must clear the bit 4 in the shuffle control mask.
> >
> > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > the upper 64 bits of destination XMM register.
> >
> > Tests are also added to check each SSE emulation of MMX intrinsics.
> >
> > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > tested with
> >
> > --with-arch=native --with-cpu=native
> >
> > on AVX2 and AVX512F machines.
>
> An idea that would take patch a step further also on 32 bit targets:
>
> *Assuming* that operations on XMM registers are as fast (or perhaps
> faster) than operations on MMX registers, we can change mmx_isa
> attribute in e.g.
>
> +  "@
> +   p<logic>\t{%2, %0|%0, %2}
> +   p<logic>\t{%2, %0|%0, %2}
> +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
>
> to:
>
> [(set_attr "isa" "*,noavx,avx")
>  (set_attr "mmx_isa" "native,*,*")]
>
> So, for x86_64 everything stays the same, but for x86_32 we now allow
> intrinsics to use xmm registers in addition to mmx registers. We can't
> disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> & similar, which are more efficient with MMX regs), but RA has much
> more freedom to allocate the most effective register set even for
> 32bit targets.
>
> WDYT?
>

Since MMX registers are used to pass and return __m64 values,
we can't really get rid of MMX instructions in 32-bit mode.  If people
have to stay with 32-bit mode, they need MMX.  I don't think we should
extend TARGET_MMX_WITH_SSE to 32-bit mode.

-- 
H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 2:42 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 2:33 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > On 2/16/19, H.J. Lu <hjl.tools@gmail.com> wrote:
> > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > >
> > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > >
> > > ;; Define instruction set of MMX instructions
> > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > >   (const_string "base"))
> > >
> > >          (eq_attr "mmx_isa" "native")
> > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > >          (eq_attr "mmx_isa" "x64")
> > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > >          (eq_attr "mmx_isa" "x64_avx")
> > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > >          (eq_attr "mmx_isa" "x64_noavx")
> > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > >
> > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > TARGET_MMX_WITH_SSE.
> > >
> > > Most of MMX instructions have equivalent SSE versions and results of some
> > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > couple tricky cases:
> > >
> > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > mask operand and handle unmapped bits 64:127 at memory address by
> > > adjusting source and mask operands together with memory address.
> > >
> > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > in 64-bit mode.
> > >
> > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > >
> > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > the upper 64 bits of destination XMM register.
> > >
> > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > >
> > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > tested with
> > >
> > > --with-arch=native --with-cpu=native
> > >
> > > on AVX2 and AVX512F machines.
> >
> > An idea that would take patch a step further also on 32 bit targets:
> >
> > *Assuming* that operations on XMM registers are as fast (or perhaps
> > faster) than operations on MMX registers, we can change mmx_isa
> > attribute in e.g.
> >
> > +  "@
> > +   p<logic>\t{%2, %0|%0, %2}
> > +   p<logic>\t{%2, %0|%0, %2}
> > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> >
> > to:
> >
> > [(set_attr "isa" "*,noavx,avx")
> >  (set_attr "mmx_isa" "native,*,*")]
> >
> > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > intrinsics to use xmm registers in addition to mmx registers. We can't
> > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > & similar, which are more efficient with MMX regs), but RA has much
> > more freedom to allocate the most effective register set even for
> > 32bit targets.
> >
> > WDYT?
> >
>
> Since MMX registers are used to pass and return __m64 values,
> we can't really get rid of MMX instructions in 32-bit mode.  If people
> have to stay with 32-bit mode, they need MMX.  I don't think we should
> extend TARGET_MMX_WITH_SSE to 32-bit mode.

No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
should not *disable* SSE alternatives on 32bit targets.

Uros.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 4:53 PM Uros Bizjak <ubizjak@gmail.com> wrote:

> > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > >
> > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > >
> > > > ;; Define instruction set of MMX instructions
> > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > >   (const_string "base"))
> > > >
> > > >          (eq_attr "mmx_isa" "native")
> > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > >          (eq_attr "mmx_isa" "x64")
> > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > >          (eq_attr "mmx_isa" "x64_avx")
> > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > >
> > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > TARGET_MMX_WITH_SSE.
> > > >
> > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > couple tricky cases:
> > > >
> > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > adjusting source and mask operands together with memory address.
> > > >
> > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > in 64-bit mode.
> > > >
> > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > >
> > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > the upper 64 bits of destination XMM register.
> > > >
> > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > >
> > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > tested with
> > > >
> > > > --with-arch=native --with-cpu=native
> > > >
> > > > on AVX2 and AVX512F machines.
> > >
> > > An idea that would take patch a step further also on 32 bit targets:
> > >
> > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > faster) than operations on MMX registers, we can change mmx_isa
> > > attribute in e.g.
> > >
> > > +  "@
> > > +   p<logic>\t{%2, %0|%0, %2}
> > > +   p<logic>\t{%2, %0|%0, %2}
> > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > >
> > > to:
> > >
> > > [(set_attr "isa" "*,noavx,avx")
> > >  (set_attr "mmx_isa" "native,*,*")]
> > >
> > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > & similar, which are more efficient with MMX regs), but RA has much
> > > more freedom to allocate the most effective register set even for
> > > 32bit targets.
> > >
> > > WDYT?
> > >
> >
> > Since MMX registers are used to pass and return __m64 values,
> > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > extend TARGET_MMX_WITH_SSE to 32-bit mode.
>
> No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> should not *disable* SSE alternatives on 32bit targets.

The correct isa attribute definition would be:

[(set_attr "isa" "*,sse2_noavx,avx")
 (set_attr "mmx_isa" "native,*,*")]

Uros.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[Uros Bizjak]

On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
>         PR target/89021
>         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
>         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
>         (ix86_expand_vector_init_one_nonzero): Likewise.
>         (ix86_expand_vector_init_one_var): Likewise.
>         (ix86_expand_vector_init_general): Likewise.
>         (ix86_expand_vector_init): Likewise.
>         (ix86_expand_vector_set): Likewise.
>         (ix86_expand_vector_extract): Likewise.

Please move this initialization to .md files. We don't support V2SF, so

(define_expand "vec_setv2sf"
  [(match_operand:V2SF 0 "register_operand")
   (match_operand:SF 1 "register_operand")
   (match_operand 2 "const_int_operand")]
  "TARGET_MMX"
{
  ix86_expand_vector_set (false, operands[0], operands[1],
              INTVAL (operands[2]));
  DONE;
})

should remain false, while in:

(define_expand "vec_setv2si"
  [(match_operand:V2SI 0 "register_operand")
   (match_operand:SI 1 "register_operand")
   (match_operand 2 "const_int_operand")]
  "TARGET_MMX"
{
  ix86_expand_vector_set (false, operands[0], operands[1],
              INTVAL (operands[2]));
  DONE;
})

we should use TARGET_MMX_WITH_SSE instead of false in the call to
ix86_expand_vector_set. The same change for the other MMX modes: V4HI
and V8QI.

Same for other exported vector initializers: ix86_expand_vector_init
and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
not V2SF (non-MMX modes are agnostic to mmx_ok argument).

Uros.


>         * config/i386/mmx.md (*vec_dupv2sf): Changed to
>         define_insn_and_split to support SSE emulation.
>         (*vec_extractv2sf_0): Likewise.
>         (*vec_extractv2sf_1): Likewise.
>         (*vec_extractv2si_0): Likewise.
>         (*vec_extractv2si_1): Likewise.
>         (*vec_extractv2si_zext_mem): Likewise.
>         (vec_setv2sf): Also allow TARGET_MMX_WITH_SSE.
>         (vec_extractv2sf_1 splitter): Likewise.
>         (vec_extractv2sfsf): Likewise.
>         (vec_setv2si): Likewise.
>         (vec_extractv2si_1 splitter): Likewise.
>         (vec_extractv2sisi): Likewise.
>         (vec_setv4hi): Likewise.
>         (vec_extractv4hihi): Likewise.
>         (vec_setv8qi): Likewise.
>         (vec_extractv8qiqi): Likewise.
> ---
>  gcc/config/i386/i386.c |  8 +++++
>  gcc/config/i386/mmx.md | 66 +++++++++++++++++++++++++++---------------
>  2 files changed, 50 insertions(+), 24 deletions(-)
>
> diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
> index a76c17beece..25e0dc43a9e 100644
> --- a/gcc/config/i386/i386.c
> +++ b/gcc/config/i386/i386.c
> @@ -42620,6 +42620,7 @@ ix86_expand_vector_init_duplicate (bool mmx_ok, machine_mode mode,
>  {
>    bool ok;
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2SImode:
> @@ -42779,6 +42780,7 @@ ix86_expand_vector_init_one_nonzero (bool mmx_ok, machine_mode mode,
>    bool use_vector_set = false;
>    rtx (*gen_vec_set_0) (rtx, rtx, rtx) = NULL;
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2DImode:
> @@ -42972,6 +42974,7 @@ ix86_expand_vector_init_one_var (bool mmx_ok, machine_mode mode,
>    XVECEXP (const_vec, 0, one_var) = CONST0_RTX (GET_MODE_INNER (mode));
>    const_vec = gen_rtx_CONST_VECTOR (mode, XVEC (const_vec, 0));
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2DFmode:
> @@ -43357,6 +43360,7 @@ ix86_expand_vector_init_general (bool mmx_ok, machine_mode mode,
>    machine_mode quarter_mode = VOIDmode;
>    int n, i;
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2SFmode:
> @@ -43556,6 +43560,8 @@ ix86_expand_vector_init (bool mmx_ok, rtx target, rtx vals)
>    int i;
>    rtx x;
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
> +
>    /* Handle first initialization from vector elts.  */
>    if (n_elts != XVECLEN (vals, 0))
>      {
> @@ -43655,6 +43661,7 @@ ix86_expand_vector_set (bool mmx_ok, rtx target, rtx val, int elt)
>    machine_mode mmode = VOIDmode;
>    rtx (*gen_blendm) (rtx, rtx, rtx, rtx);
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2SFmode:
> @@ -44010,6 +44017,7 @@ ix86_expand_vector_extract (bool mmx_ok, rtx target, rtx vec, int elt)
>    bool use_vec_extr = false;
>    rtx tmp;
>
> +  mmx_ok |= TARGET_MMX_WITH_SSE;
>    switch (mode)
>      {
>      case E_V2SImode:
> diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
> index c612d6b9e5c..f98952fd8a0 100644
> --- a/gcc/config/i386/mmx.md
> +++ b/gcc/config/i386/mmx.md
> @@ -555,14 +555,23 @@
>     (set_attr "prefix_extra" "1")
>     (set_attr "mode" "V2SF")])
>
> -(define_insn "*vec_dupv2sf"
> -  [(set (match_operand:V2SF 0 "register_operand" "=y")
> +(define_insn_and_split "*vec_dupv2sf"
> +  [(set (match_operand:V2SF 0 "register_operand" "=y,x,Yv")
>         (vec_duplicate:V2SF
> -         (match_operand:SF 1 "register_operand" "0")))]
> -  "TARGET_MMX"
> -  "punpckldq\t%0, %0"
> -  [(set_attr "type" "mmxcvt")
> -   (set_attr "mode" "DI")])
> +         (match_operand:SF 1 "register_operand" "0,0,Yv")))]
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
> +  "@
> +   punpckldq\t%0, %0
> +   #
> +   #"
> +  "TARGET_MMX_WITH_SSE && reload_completed"
> +  [(set (match_dup 0)
> +       (vec_duplicate:V4SF (match_dup 1)))]
> +  "operands[0] = lowpart_subreg (V4SFmode, operands[0],
> +                                GET_MODE (operands[0]));"
> +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> +   (set_attr "type" "mmxcvt,ssemov,ssemov")
> +   (set_attr "mode" "DI,TI,TI")])
>
>  (define_insn "*mmx_concatv2sf"
>    [(set (match_operand:V2SF 0 "register_operand"     "=y,y")
> @@ -580,7 +589,7 @@
>    [(match_operand:V2SF 0 "register_operand")
>     (match_operand:SF 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_set (false, operands[0], operands[1],
>                           INTVAL (operands[2]));
> @@ -594,11 +603,13 @@
>         (vec_select:SF
>           (match_operand:V2SF 1 "nonimmediate_operand" " xm,x,ym,y,m,m")
>           (parallel [(const_int 0)])))]
> -  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> +   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
>    "#"
>    "&& reload_completed"
>    [(set (match_dup 0) (match_dup 1))]
> -  "operands[1] = gen_lowpart (SFmode, operands[1]);")
> +  "operands[1] = gen_lowpart (SFmode, operands[1]);"
> +  [(set_attr "mmx_isa" "*,*,native,native,*,*")])
>
>  ;; Avoid combining registers from different units in a single alternative,
>  ;; see comment above inline_secondary_memory_needed function in i386.c
> @@ -607,7 +618,8 @@
>         (vec_select:SF
>           (match_operand:V2SF 1 "nonimmediate_operand" " 0,x,x,o,o,o,o")
>           (parallel [(const_int 1)])))]
> -  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> +   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
>    "@
>     punpckhdq\t%0, %0
>     %vmovshdup\t{%1, %0|%0, %1}
> @@ -617,6 +629,7 @@
>     #
>     #"
>    [(set_attr "isa" "*,sse3,noavx,*,*,*,*")
> +   (set_attr "mmx_isa" "native,*,*,native,*,*,*")
>     (set_attr "type" "mmxcvt,sse,sseshuf1,mmxmov,ssemov,fmov,imov")
>     (set (attr "length_immediate")
>       (if_then_else (eq_attr "alternative" "2")
> @@ -634,7 +647,7 @@
>         (vec_select:SF
>           (match_operand:V2SF 1 "memory_operand")
>           (parallel [(const_int 1)])))]
> -  "TARGET_MMX && reload_completed"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && reload_completed"
>    [(set (match_dup 0) (match_dup 1))]
>    "operands[1] = adjust_address (operands[1], SFmode, 4);")
>
> @@ -642,7 +655,7 @@
>    [(match_operand:SF 0 "register_operand")
>     (match_operand:V2SF 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_extract (false, operands[0], operands[1],
>                               INTVAL (operands[2]));
> @@ -1526,7 +1539,7 @@
>    [(match_operand:V2SI 0 "register_operand")
>     (match_operand:SI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_set (false, operands[0], operands[1],
>                           INTVAL (operands[2]));
> @@ -1540,11 +1553,13 @@
>         (vec_select:SI
>           (match_operand:V2SI 1 "nonimmediate_operand" "xm,x,ym,y,m")
>           (parallel [(const_int 0)])))]
> -  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> +   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
>    "#"
>    "&& reload_completed"
>    [(set (match_dup 0) (match_dup 1))]
> -  "operands[1] = gen_lowpart (SImode, operands[1]);")
> +  "operands[1] = gen_lowpart (SImode, operands[1]);"
> +  [(set_attr "mmx_isa" "*,*,native,native,*")])
>
>  ;; Avoid combining registers from different units in a single alternative,
>  ;; see comment above inline_secondary_memory_needed function in i386.c
> @@ -1553,7 +1568,8 @@
>         (vec_select:SI
>           (match_operand:V2SI 1 "nonimmediate_operand" " 0,x,x,o,o,o")
>           (parallel [(const_int 1)])))]
> -  "TARGET_MMX && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> +   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
>    "@
>     punpckhdq\t%0, %0
>     %vpshufd\t{$0xe5, %1, %0|%0, %1, 0xe5}
> @@ -1562,6 +1578,7 @@
>     #
>     #"
>    [(set_attr "isa" "*,sse2,noavx,*,*,*")
> +   (set_attr "mmx_isa" "native,*,*,native,*,*")
>     (set_attr "type" "mmxcvt,sseshuf1,sseshuf1,mmxmov,ssemov,imov")
>     (set (attr "length_immediate")
>       (if_then_else (eq_attr "alternative" "1,2")
> @@ -1575,7 +1592,7 @@
>         (vec_select:SI
>           (match_operand:V2SI 1 "memory_operand")
>           (parallel [(const_int 1)])))]
> -  "TARGET_MMX && reload_completed"
> +  "(TARGET_MMX || TARGET_MMX_WITH_SSE) && reload_completed"
>    [(set (match_dup 0) (match_dup 1))]
>    "operands[1] = adjust_address (operands[1], SImode, 4);")
>
> @@ -1592,13 +1609,14 @@
>  {
>    operands[1] = adjust_address (operands[1], SImode, INTVAL (operands[2]) * 4);
>  }
> -  [(set_attr "isa" "*,sse2,*")])
> +  [(set_attr "isa" "*,sse2,*")
> +   (set_attr "mmx_isa" "native,*,*")])
>
>  (define_expand "vec_extractv2sisi"
>    [(match_operand:SI 0 "register_operand")
>     (match_operand:V2SI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_extract (false, operands[0], operands[1],
>                               INTVAL (operands[2]));
> @@ -1618,7 +1636,7 @@
>    [(match_operand:V4HI 0 "register_operand")
>     (match_operand:HI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_set (false, operands[0], operands[1],
>                           INTVAL (operands[2]));
> @@ -1629,7 +1647,7 @@
>    [(match_operand:HI 0 "register_operand")
>     (match_operand:V4HI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_extract (false, operands[0], operands[1],
>                               INTVAL (operands[2]));
> @@ -1649,7 +1667,7 @@
>    [(match_operand:V8QI 0 "register_operand")
>     (match_operand:QI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_set (false, operands[0], operands[1],
>                           INTVAL (operands[2]));
> @@ -1660,7 +1678,7 @@
>    [(match_operand:QI 0 "register_operand")
>     (match_operand:V8QI 1 "register_operand")
>     (match_operand 2 "const_int_operand")]
> -  "TARGET_MMX"
> +  "TARGET_MMX || TARGET_MMX_WITH_SSE"
>  {
>    ix86_expand_vector_extract (false, operands[0], operands[1],
>                               INTVAL (operands[2]));
> --
> 2.20.1
>

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 8:24 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >
> >         PR target/89021
> >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> >         (ix86_expand_vector_init_one_nonzero): Likewise.
> >         (ix86_expand_vector_init_one_var): Likewise.
> >         (ix86_expand_vector_init_general): Likewise.
> >         (ix86_expand_vector_init): Likewise.
> >         (ix86_expand_vector_set): Likewise.
> >         (ix86_expand_vector_extract): Likewise.
>
> Please move this initialization to .md files. We don't support V2SF, so
>
> (define_expand "vec_setv2sf"
>   [(match_operand:V2SF 0 "register_operand")
>    (match_operand:SF 1 "register_operand")
>    (match_operand 2 "const_int_operand")]
>   "TARGET_MMX"
> {
>   ix86_expand_vector_set (false, operands[0], operands[1],
>               INTVAL (operands[2]));
>   DONE;
> })
>
> should remain false, while in:
>
> (define_expand "vec_setv2si"
>   [(match_operand:V2SI 0 "register_operand")
>    (match_operand:SI 1 "register_operand")
>    (match_operand 2 "const_int_operand")]
>   "TARGET_MMX"
> {
>   ix86_expand_vector_set (false, operands[0], operands[1],
>               INTVAL (operands[2]));
>   DONE;
> })
>
> we should use TARGET_MMX_WITH_SSE instead of false in the call to
> ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> and V8QI.
>
> Same for other exported vector initializers: ix86_expand_vector_init
> and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> not V2SF (non-MMX modes are agnostic to mmx_ok argument).
>

There is limited support for V2SF with MMX.  My goal is to emulate
all MMX features with SSE so that -msse2 -mno-mmx works for all
MMX codes which don't use MMX inline asm.  Leaving out V2SF
emulation creates a hole in my implementation.  Do you have testcases
to show issues with V2SF emulation?

-- 
H.J.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 6:03 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 8:24 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > >
> > >         PR target/89021
> > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > >         (ix86_expand_vector_init_one_var): Likewise.
> > >         (ix86_expand_vector_init_general): Likewise.
> > >         (ix86_expand_vector_init): Likewise.
> > >         (ix86_expand_vector_set): Likewise.
> > >         (ix86_expand_vector_extract): Likewise.
> >
> > Please move this initialization to .md files. We don't support V2SF, so
> >
> > (define_expand "vec_setv2sf"
> >   [(match_operand:V2SF 0 "register_operand")
> >    (match_operand:SF 1 "register_operand")
> >    (match_operand 2 "const_int_operand")]
> >   "TARGET_MMX"
> > {
> >   ix86_expand_vector_set (false, operands[0], operands[1],
> >               INTVAL (operands[2]));
> >   DONE;
> > })
> >
> > should remain false, while in:
> >
> > (define_expand "vec_setv2si"
> >   [(match_operand:V2SI 0 "register_operand")
> >    (match_operand:SI 1 "register_operand")
> >    (match_operand 2 "const_int_operand")]
> >   "TARGET_MMX"
> > {
> >   ix86_expand_vector_set (false, operands[0], operands[1],
> >               INTVAL (operands[2]));
> >   DONE;
> > })
> >
> > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > and V8QI.
> >
> > Same for other exported vector initializers: ix86_expand_vector_init
> > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> >
>
> There is limited support for V2SF with MMX.  My goal is to emulate
> all MMX features with SSE so that -msse2 -mno-mmx works for all
> MMX codes which don't use MMX inline asm.  Leaving out V2SF
> emulation creates a hole in my implementation.  Do you have testcases
> to show issues with V2SF emulation?

No, but V2SF should be avoided from emulation.

Uros.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 7:57 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 4:53 PM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > >
> > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > >
> > > > > ;; Define instruction set of MMX instructions
> > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > >   (const_string "base"))
> > > > >
> > > > >          (eq_attr "mmx_isa" "native")
> > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > >          (eq_attr "mmx_isa" "x64")
> > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > >
> > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > TARGET_MMX_WITH_SSE.
> > > > >
> > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > couple tricky cases:
> > > > >
> > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > adjusting source and mask operands together with memory address.
> > > > >
> > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > in 64-bit mode.
> > > > >
> > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > >
> > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > the upper 64 bits of destination XMM register.
> > > > >
> > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > >
> > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > tested with
> > > > >
> > > > > --with-arch=native --with-cpu=native
> > > > >
> > > > > on AVX2 and AVX512F machines.
> > > >
> > > > An idea that would take patch a step further also on 32 bit targets:
> > > >
> > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > attribute in e.g.
> > > >
> > > > +  "@
> > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > >
> > > > to:
> > > >
> > > > [(set_attr "isa" "*,noavx,avx")
> > > >  (set_attr "mmx_isa" "native,*,*")]
> > > >
> > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > more freedom to allocate the most effective register set even for
> > > > 32bit targets.
> > > >
> > > > WDYT?
> > > >
> > >
> > > Since MMX registers are used to pass and return __m64 values,
> > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> >
> > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > should not *disable* SSE alternatives on 32bit targets.

I don't think my patch set disables any SSE alternatives in 32-bit
mode.   However,
it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
alternatives in

(define_insn "*mmx_<code><mode>3"
  [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
        (any_logic:MMXMODEI
          (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
          (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
  "(TARGET_MMX || TARGET_MMX_WITH_SSE)
   && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
  "@
   p<logic>\t{%2, %0|%0, %2}
   p<logic>\t{%2, %0|%0, %2}
   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
   (set_attr "type" "mmxadd,sselog,sselog")
   (set_attr "mode" "DI,TI,TI")])

register_mmxmem_operand must return true for SSE alternatives:

;; Match register operands, but include memory operands for
;; !TARGET_MMX_WITH_SSE.
(define_predicate "register_mmxmem_operand"
  (ior (match_operand 0 "register_operand")
       (and (not (match_test "TARGET_MMX_WITH_SSE"))
            (match_operand 0 "memory_operand"))))

How do you enable SSE alternatives in 32-bit mode without enabling
TARGET_MMX_WITH_SSE for 32-bit mode?

> The correct isa attribute definition would be:
>
> [(set_attr "isa" "*,sse2_noavx,avx")
>  (set_attr "mmx_isa" "native,*,*")]
>
> Uros.



-- 
H.J.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 9:08 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 6:03 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >
> > On Sun, Feb 17, 2019 at 8:24 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> > >
> > > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > > >
> > > >         PR target/89021
> > > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > > >         (ix86_expand_vector_init_one_var): Likewise.
> > > >         (ix86_expand_vector_init_general): Likewise.
> > > >         (ix86_expand_vector_init): Likewise.
> > > >         (ix86_expand_vector_set): Likewise.
> > > >         (ix86_expand_vector_extract): Likewise.
> > >
> > > Please move this initialization to .md files. We don't support V2SF, so
> > >
> > > (define_expand "vec_setv2sf"
> > >   [(match_operand:V2SF 0 "register_operand")
> > >    (match_operand:SF 1 "register_operand")
> > >    (match_operand 2 "const_int_operand")]
> > >   "TARGET_MMX"
> > > {
> > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > >               INTVAL (operands[2]));
> > >   DONE;
> > > })
> > >
> > > should remain false, while in:
> > >
> > > (define_expand "vec_setv2si"
> > >   [(match_operand:V2SI 0 "register_operand")
> > >    (match_operand:SI 1 "register_operand")
> > >    (match_operand 2 "const_int_operand")]
> > >   "TARGET_MMX"
> > > {
> > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > >               INTVAL (operands[2]));
> > >   DONE;
> > > })
> > >
> > > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > > and V8QI.
> > >
> > > Same for other exported vector initializers: ix86_expand_vector_init
> > > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> > >
> >
> > There is limited support for V2SF with MMX.  My goal is to emulate
> > all MMX features with SSE so that -msse2 -mno-mmx works for all
> > MMX codes which don't use MMX inline asm.  Leaving out V2SF
> > emulation creates a hole in my implementation.  Do you have testcases
> > to show issues with V2SF emulation?
>
> No, but V2SF should be avoided from emulation.

Then -msse2 -mno-mmx won't work on some MMX codes like:

#include <mmintrin.h>

float
foo (__m64 x)
{
  return ((__v2sf) x)[0];
}

Is there any reason why it shouldn't work for -msse2 -mno-mmx?

-- 
H.J.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 6:15 PM H.J. Lu <hjl.tools@gmail.com> wrote:

> > > > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > > > >
> > > > >         PR target/89021
> > > > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > > > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > > > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > > > >         (ix86_expand_vector_init_one_var): Likewise.
> > > > >         (ix86_expand_vector_init_general): Likewise.
> > > > >         (ix86_expand_vector_init): Likewise.
> > > > >         (ix86_expand_vector_set): Likewise.
> > > > >         (ix86_expand_vector_extract): Likewise.
> > > >
> > > > Please move this initialization to .md files. We don't support V2SF, so
> > > >
> > > > (define_expand "vec_setv2sf"
> > > >   [(match_operand:V2SF 0 "register_operand")
> > > >    (match_operand:SF 1 "register_operand")
> > > >    (match_operand 2 "const_int_operand")]
> > > >   "TARGET_MMX"
> > > > {
> > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > >               INTVAL (operands[2]));
> > > >   DONE;
> > > > })
> > > >
> > > > should remain false, while in:
> > > >
> > > > (define_expand "vec_setv2si"
> > > >   [(match_operand:V2SI 0 "register_operand")
> > > >    (match_operand:SI 1 "register_operand")
> > > >    (match_operand 2 "const_int_operand")]
> > > >   "TARGET_MMX"
> > > > {
> > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > >               INTVAL (operands[2]));
> > > >   DONE;
> > > > })
> > > >
> > > > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > > > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > > > and V8QI.
> > > >
> > > > Same for other exported vector initializers: ix86_expand_vector_init
> > > > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > > > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> > > >
> > >
> > > There is limited support for V2SF with MMX.  My goal is to emulate
> > > all MMX features with SSE so that -msse2 -mno-mmx works for all
> > > MMX codes which don't use MMX inline asm.  Leaving out V2SF
> > > emulation creates a hole in my implementation.  Do you have testcases
> > > to show issues with V2SF emulation?
> >
> > No, but V2SF should be avoided from emulation.
>
> Then -msse2 -mno-mmx won't work on some MMX codes like:
>
> #include <mmintrin.h>
>
> float
> foo (__m64 x)
> {
>   return ((__v2sf) x)[0];
> }
>
> Is there any reason why it shouldn't work for -msse2 -mno-mmx?

Does this require 3DNOW insn? If not, it is OK also for V2SF.

Uros.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 6:10 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 7:57 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > On Sun, Feb 17, 2019 at 4:53 PM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > >
> > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > >
> > > > > > ;; Define instruction set of MMX instructions
> > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > >   (const_string "base"))
> > > > > >
> > > > > >          (eq_attr "mmx_isa" "native")
> > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > >
> > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > TARGET_MMX_WITH_SSE.
> > > > > >
> > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > couple tricky cases:
> > > > > >
> > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > adjusting source and mask operands together with memory address.
> > > > > >
> > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > in 64-bit mode.
> > > > > >
> > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > >
> > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > the upper 64 bits of destination XMM register.
> > > > > >
> > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > >
> > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > tested with
> > > > > >
> > > > > > --with-arch=native --with-cpu=native
> > > > > >
> > > > > > on AVX2 and AVX512F machines.
> > > > >
> > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > >
> > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > attribute in e.g.
> > > > >
> > > > > +  "@
> > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > >
> > > > > to:
> > > > >
> > > > > [(set_attr "isa" "*,noavx,avx")
> > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > >
> > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > more freedom to allocate the most effective register set even for
> > > > > 32bit targets.
> > > > >
> > > > > WDYT?
> > > > >
> > > >
> > > > Since MMX registers are used to pass and return __m64 values,
> > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > >
> > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > should not *disable* SSE alternatives on 32bit targets.
>
> I don't think my patch set disables any SSE alternatives in 32-bit
> mode.   However,
> it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> alternatives in
>
> (define_insn "*mmx_<code><mode>3"
>   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
>         (any_logic:MMXMODEI
>           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
>           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
>   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
>    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
>   "@
>    p<logic>\t{%2, %0|%0, %2}
>    p<logic>\t{%2, %0|%0, %2}
>    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
>   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
>    (set_attr "type" "mmxadd,sselog,sselog")
>    (set_attr "mode" "DI,TI,TI")])
>
> register_mmxmem_operand must return true for SSE alternatives:

It returns true for register and memory operands for 32bit targets, because

#define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)

> ;; Match register operands, but include memory operands for
> ;; !TARGET_MMX_WITH_SSE.
> (define_predicate "register_mmxmem_operand"
>   (ior (match_operand 0 "register_operand")
>        (and (not (match_test "TARGET_MMX_WITH_SSE"))
>             (match_operand 0 "memory_operand"))))
>
> How do you enable SSE alternatives in 32-bit mode without enabling
> TARGET_MMX_WITH_SSE for 32-bit mode?

Check the new attribute definitions below:

> > The correct isa attribute definition would be:
> >
> > [(set_attr "isa" "*,sse2_noavx,avx")
> >  (set_attr "mmx_isa" "native,*,*")]

Uros.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 9:22 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 6:15 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> > > > > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > > > > >
> > > > > >         PR target/89021
> > > > > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > > > > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > > > > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > > > > >         (ix86_expand_vector_init_one_var): Likewise.
> > > > > >         (ix86_expand_vector_init_general): Likewise.
> > > > > >         (ix86_expand_vector_init): Likewise.
> > > > > >         (ix86_expand_vector_set): Likewise.
> > > > > >         (ix86_expand_vector_extract): Likewise.
> > > > >
> > > > > Please move this initialization to .md files. We don't support V2SF, so
> > > > >
> > > > > (define_expand "vec_setv2sf"
> > > > >   [(match_operand:V2SF 0 "register_operand")
> > > > >    (match_operand:SF 1 "register_operand")
> > > > >    (match_operand 2 "const_int_operand")]
> > > > >   "TARGET_MMX"
> > > > > {
> > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > >               INTVAL (operands[2]));
> > > > >   DONE;
> > > > > })
> > > > >
> > > > > should remain false, while in:
> > > > >
> > > > > (define_expand "vec_setv2si"
> > > > >   [(match_operand:V2SI 0 "register_operand")
> > > > >    (match_operand:SI 1 "register_operand")
> > > > >    (match_operand 2 "const_int_operand")]
> > > > >   "TARGET_MMX"
> > > > > {
> > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > >               INTVAL (operands[2]));
> > > > >   DONE;
> > > > > })
> > > > >
> > > > > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > > > > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > > > > and V8QI.
> > > > >
> > > > > Same for other exported vector initializers: ix86_expand_vector_init
> > > > > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > > > > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> > > > >
> > > >
> > > > There is limited support for V2SF with MMX.  My goal is to emulate
> > > > all MMX features with SSE so that -msse2 -mno-mmx works for all
> > > > MMX codes which don't use MMX inline asm.  Leaving out V2SF
> > > > emulation creates a hole in my implementation.  Do you have testcases
> > > > to show issues with V2SF emulation?
> > >
> > > No, but V2SF should be avoided from emulation.
> >
> > Then -msse2 -mno-mmx won't work on some MMX codes like:
> >
> > #include <mmintrin.h>
> >
> > float
> > foo (__m64 x)
> > {
> >   return ((__v2sf) x)[0];
> > }
> >
> > Is there any reason why it shouldn't work for -msse2 -mno-mmx?
>
> Does this require 3DNOW insn? If not, it is OK also for V2SF.

V2SF vector manipulation, like init, dup, extract, insert, doesn't
require 3DNOW.  They are tested with -msse2 -mo-mmx in my
tests for MMX intrinsic emulations with SSE.

-- 
H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 9:27 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 6:10 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >
> > On Sun, Feb 17, 2019 at 7:57 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> > >
> > > On Sun, Feb 17, 2019 at 4:53 PM Uros Bizjak <ubizjak@gmail.com> wrote:
> > >
> > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > >
> > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > >
> > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > >   (const_string "base"))
> > > > > > >
> > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > >
> > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > >
> > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > couple tricky cases:
> > > > > > >
> > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > adjusting source and mask operands together with memory address.
> > > > > > >
> > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > in 64-bit mode.
> > > > > > >
> > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > >
> > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > the upper 64 bits of destination XMM register.
> > > > > > >
> > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > >
> > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > tested with
> > > > > > >
> > > > > > > --with-arch=native --with-cpu=native
> > > > > > >
> > > > > > > on AVX2 and AVX512F machines.
> > > > > >
> > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > >
> > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > attribute in e.g.
> > > > > >
> > > > > > +  "@
> > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > >
> > > > > > to:
> > > > > >
> > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > >
> > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > more freedom to allocate the most effective register set even for
> > > > > > 32bit targets.
> > > > > >
> > > > > > WDYT?
> > > > > >
> > > > >
> > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > >
> > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > should not *disable* SSE alternatives on 32bit targets.
> >
> > I don't think my patch set disables any SSE alternatives in 32-bit
> > mode.   However,
> > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > alternatives in
> >
> > (define_insn "*mmx_<code><mode>3"
> >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> >         (any_logic:MMXMODEI
> >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> >   "@
> >    p<logic>\t{%2, %0|%0, %2}
> >    p<logic>\t{%2, %0|%0, %2}
> >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> >    (set_attr "type" "mmxadd,sselog,sselog")
> >    (set_attr "mode" "DI,TI,TI")])
> >
> > register_mmxmem_operand must return true for SSE alternatives:
>
> It returns true for register and memory operands for 32bit targets, because
>
> #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)

Will

(match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]

work well with RA?  I got some wrong code before register_mmxmem_operand
was added to match "ym,x,Yv".

> > ;; Match register operands, but include memory operands for
> > ;; !TARGET_MMX_WITH_SSE.
> > (define_predicate "register_mmxmem_operand"
> >   (ior (match_operand 0 "register_operand")
> >        (and (not (match_test "TARGET_MMX_WITH_SSE"))
> >             (match_operand 0 "memory_operand"))))
> >
> > How do you enable SSE alternatives in 32-bit mode without enabling
> > TARGET_MMX_WITH_SSE for 32-bit mode?
>
> Check the new attribute definitions below:
> > > The correct isa attribute definition would be:
> > >
> > > [(set_attr "isa" "*,sse2_noavx,avx")
> > >  (set_attr "mmx_isa" "native,*,*")]
>
> Uros.





-- 
H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 6:37 PM H.J. Lu <hjl.tools@gmail.com> wrote:

> > > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > > >
> > > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > > >
> > > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > > >   (const_string "base"))
> > > > > > > >
> > > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > > >
> > > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > > >
> > > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > > couple tricky cases:
> > > > > > > >
> > > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > > adjusting source and mask operands together with memory address.
> > > > > > > >
> > > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > > in 64-bit mode.
> > > > > > > >
> > > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > > >
> > > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > > the upper 64 bits of destination XMM register.
> > > > > > > >
> > > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > > >
> > > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > > tested with
> > > > > > > >
> > > > > > > > --with-arch=native --with-cpu=native
> > > > > > > >
> > > > > > > > on AVX2 and AVX512F machines.
> > > > > > >
> > > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > > >
> > > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > > attribute in e.g.
> > > > > > >
> > > > > > > +  "@
> > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > >
> > > > > > > to:
> > > > > > >
> > > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > > >
> > > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > > more freedom to allocate the most effective register set even for
> > > > > > > 32bit targets.
> > > > > > >
> > > > > > > WDYT?
> > > > > > >
> > > > > >
> > > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > > >
> > > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > > should not *disable* SSE alternatives on 32bit targets.
> > >
> > > I don't think my patch set disables any SSE alternatives in 32-bit
> > > mode.   However,
> > > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > > alternatives in
> > >
> > > (define_insn "*mmx_<code><mode>3"
> > >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> > >         (any_logic:MMXMODEI
> > >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> > >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> > >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> > >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> > >   "@
> > >    p<logic>\t{%2, %0|%0, %2}
> > >    p<logic>\t{%2, %0|%0, %2}
> > >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > >    (set_attr "type" "mmxadd,sselog,sselog")
> > >    (set_attr "mode" "DI,TI,TI")])
> > >
> > > register_mmxmem_operand must return true for SSE alternatives:
> >
> > It returns true for register and memory operands for 32bit targets, because
> >
> > #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
>
> Will
>
> (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]
>
> work well with RA?  I got some wrong code before register_mmxmem_operand
> was added to match "ym,x,Yv".

I see no reason why it shouldn't.

Uros.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[Uros Bizjak]

On Sun, Feb 17, 2019 at 6:28 PM H.J. Lu <hjl.tools@gmail.com> wrote:

> > > > > > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > > > > > >
> > > > > > >         PR target/89021
> > > > > > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > > > > > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > > > > > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > > > > > >         (ix86_expand_vector_init_one_var): Likewise.
> > > > > > >         (ix86_expand_vector_init_general): Likewise.
> > > > > > >         (ix86_expand_vector_init): Likewise.
> > > > > > >         (ix86_expand_vector_set): Likewise.
> > > > > > >         (ix86_expand_vector_extract): Likewise.
> > > > > >
> > > > > > Please move this initialization to .md files. We don't support V2SF, so
> > > > > >
> > > > > > (define_expand "vec_setv2sf"
> > > > > >   [(match_operand:V2SF 0 "register_operand")
> > > > > >    (match_operand:SF 1 "register_operand")
> > > > > >    (match_operand 2 "const_int_operand")]
> > > > > >   "TARGET_MMX"
> > > > > > {
> > > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > > >               INTVAL (operands[2]));
> > > > > >   DONE;
> > > > > > })
> > > > > >
> > > > > > should remain false, while in:
> > > > > >
> > > > > > (define_expand "vec_setv2si"
> > > > > >   [(match_operand:V2SI 0 "register_operand")
> > > > > >    (match_operand:SI 1 "register_operand")
> > > > > >    (match_operand 2 "const_int_operand")]
> > > > > >   "TARGET_MMX"
> > > > > > {
> > > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > > >               INTVAL (operands[2]));
> > > > > >   DONE;
> > > > > > })
> > > > > >
> > > > > > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > > > > > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > > > > > and V8QI.
> > > > > >
> > > > > > Same for other exported vector initializers: ix86_expand_vector_init
> > > > > > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > > > > > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> > > > > >
> > > > >
> > > > > There is limited support for V2SF with MMX.  My goal is to emulate
> > > > > all MMX features with SSE so that -msse2 -mno-mmx works for all
> > > > > MMX codes which don't use MMX inline asm.  Leaving out V2SF
> > > > > emulation creates a hole in my implementation.  Do you have testcases
> > > > > to show issues with V2SF emulation?
> > > >
> > > > No, but V2SF should be avoided from emulation.
> > >
> > > Then -msse2 -mno-mmx won't work on some MMX codes like:
> > >
> > > #include <mmintrin.h>
> > >
> > > float
> > > foo (__m64 x)
> > > {
> > >   return ((__v2sf) x)[0];
> > > }
> > >
> > > Is there any reason why it shouldn't work for -msse2 -mno-mmx?
> >
> > Does this require 3DNOW insn? If not, it is OK also for V2SF.
>
> V2SF vector manipulation, like init, dup, extract, insert, doesn't
> require 3DNOW.  They are tested with -msse2 -mo-mmx in my
> tests for MMX intrinsic emulations with SSE.

That's OK then. Please move initialization to .md files, as suggested before.

Uros.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 10:49 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 6:37 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> > > > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > > > >
> > > > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > > > >
> > > > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > > > >   (const_string "base"))
> > > > > > > > >
> > > > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > > > >
> > > > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > > > >
> > > > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > > > couple tricky cases:
> > > > > > > > >
> > > > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > > > adjusting source and mask operands together with memory address.
> > > > > > > > >
> > > > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > > > in 64-bit mode.
> > > > > > > > >
> > > > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > > > >
> > > > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > > > the upper 64 bits of destination XMM register.
> > > > > > > > >
> > > > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > > > >
> > > > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > > > tested with
> > > > > > > > >
> > > > > > > > > --with-arch=native --with-cpu=native
> > > > > > > > >
> > > > > > > > > on AVX2 and AVX512F machines.
> > > > > > > >
> > > > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > > > >
> > > > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > > > attribute in e.g.
> > > > > > > >
> > > > > > > > +  "@
> > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > > >
> > > > > > > > to:
> > > > > > > >
> > > > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > > > >
> > > > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > > > more freedom to allocate the most effective register set even for
> > > > > > > > 32bit targets.
> > > > > > > >
> > > > > > > > WDYT?
> > > > > > > >
> > > > > > >
> > > > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > > > >
> > > > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > > > should not *disable* SSE alternatives on 32bit targets.
> > > >
> > > > I don't think my patch set disables any SSE alternatives in 32-bit
> > > > mode.   However,
> > > > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > > > alternatives in
> > > >
> > > > (define_insn "*mmx_<code><mode>3"
> > > >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> > > >         (any_logic:MMXMODEI
> > > >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> > > >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> > > >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> > > >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> > > >   "@
> > > >    p<logic>\t{%2, %0|%0, %2}
> > > >    p<logic>\t{%2, %0|%0, %2}
> > > >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > >    (set_attr "type" "mmxadd,sselog,sselog")
> > > >    (set_attr "mode" "DI,TI,TI")])
> > > >
> > > > register_mmxmem_operand must return true for SSE alternatives:
> > >
> > > It returns true for register and memory operands for 32bit targets, because
> > >
> > > #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> >
> > Will
> >
> > (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]
> >
> > work well with RA?  I got some wrong code before register_mmxmem_operand
> > was added to match "ym,x,Yv".
>
> I see no reason why it shouldn't.

This will be equivalent to replace register_operand in

[(match_operand:VI1_AVX512VLBW 1 "register_operand" "v")

with nonimmediate_operand.  If it should work, I can do it in i386.md and
sse.md to check it out.

-- 
H.J.

Re: [PATCH 38/41] i386: Allow MMX vector expanders with TARGET_MMX_WITH_SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 10:50 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 6:28 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> > > > > > > On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> > > > > > > >
> > > > > > > >         PR target/89021
> > > > > > > >         * config/i386/i386.c (ix86_expand_vector_init_duplicate): Set
> > > > > > > >         mmx_ok to true if TARGET_MMX_WITH_SSE is true.
> > > > > > > >         (ix86_expand_vector_init_one_nonzero): Likewise.
> > > > > > > >         (ix86_expand_vector_init_one_var): Likewise.
> > > > > > > >         (ix86_expand_vector_init_general): Likewise.
> > > > > > > >         (ix86_expand_vector_init): Likewise.
> > > > > > > >         (ix86_expand_vector_set): Likewise.
> > > > > > > >         (ix86_expand_vector_extract): Likewise.
> > > > > > >
> > > > > > > Please move this initialization to .md files. We don't support V2SF, so
> > > > > > >
> > > > > > > (define_expand "vec_setv2sf"
> > > > > > >   [(match_operand:V2SF 0 "register_operand")
> > > > > > >    (match_operand:SF 1 "register_operand")
> > > > > > >    (match_operand 2 "const_int_operand")]
> > > > > > >   "TARGET_MMX"
> > > > > > > {
> > > > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > > > >               INTVAL (operands[2]));
> > > > > > >   DONE;
> > > > > > > })
> > > > > > >
> > > > > > > should remain false, while in:
> > > > > > >
> > > > > > > (define_expand "vec_setv2si"
> > > > > > >   [(match_operand:V2SI 0 "register_operand")
> > > > > > >    (match_operand:SI 1 "register_operand")
> > > > > > >    (match_operand 2 "const_int_operand")]
> > > > > > >   "TARGET_MMX"
> > > > > > > {
> > > > > > >   ix86_expand_vector_set (false, operands[0], operands[1],
> > > > > > >               INTVAL (operands[2]));
> > > > > > >   DONE;
> > > > > > > })
> > > > > > >
> > > > > > > we should use TARGET_MMX_WITH_SSE instead of false in the call to
> > > > > > > ix86_expand_vector_set. The same change for the other MMX modes: V4HI
> > > > > > > and V8QI.
> > > > > > >
> > > > > > > Same for other exported vector initializers: ix86_expand_vector_init
> > > > > > > and ix86_expand_vector_extract. We support V2SI, V4HI and V8QI, but
> > > > > > > not V2SF (non-MMX modes are agnostic to mmx_ok argument).
> > > > > > >
> > > > > >
> > > > > > There is limited support for V2SF with MMX.  My goal is to emulate
> > > > > > all MMX features with SSE so that -msse2 -mno-mmx works for all
> > > > > > MMX codes which don't use MMX inline asm.  Leaving out V2SF
> > > > > > emulation creates a hole in my implementation.  Do you have testcases
> > > > > > to show issues with V2SF emulation?
> > > > >
> > > > > No, but V2SF should be avoided from emulation.
> > > >
> > > > Then -msse2 -mno-mmx won't work on some MMX codes like:
> > > >
> > > > #include <mmintrin.h>
> > > >
> > > > float
> > > > foo (__m64 x)
> > > > {
> > > >   return ((__v2sf) x)[0];
> > > > }
> > > >
> > > > Is there any reason why it shouldn't work for -msse2 -mno-mmx?
> > >
> > > Does this require 3DNOW insn? If not, it is OK also for V2SF.
> >
> > V2SF vector manipulation, like init, dup, extract, insert, doesn't
> > require 3DNOW.  They are tested with -msse2 -mo-mmx in my
> > tests for MMX intrinsic emulations with SSE.
>
> That's OK then. Please move initialization to .md files, as suggested before.
>

I will give it a try.

-- 
H.J.

Re: [PATCH 36/41] Prevent allocation of MMX registers with TARGET_MMX_WITH_SSE

[Uros Bizjak]

On Sat, Feb 16, 2019 at 11:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> From: Uros Bizjak <ubizjak@gmail.com>
>
> 2019-02-14  Uroš Bizjak  <ubizjak@gmail.com>
>
>         PR target/89021
>         * config/i386/i386.md (*zero_extendsidi2): Add mmx_isa attribute.
>         * config/i386/sse.md (*vec_concatv2sf_sse4_1): Ditto.
>         (*vec_concatv2sf_sse): Ditto.
>         (*vec_concatv2si_sse4_1): Ditto.
>         (*vec_concatv2si): Ditto.
>         (*vec_concatv4si_0): Ditto.
>         (*vec_concatv2di_0): Ditto.

It looks like I forgot to add sse2_cvtpi2pd, sse2_cvtpd2pi and sse2_cvttpd2pi.

Uros.

> ---
>  gcc/config/i386/i386.md |  4 ++++
>  gcc/config/i386/sse.md  | 16 ++++++++++++++--
>  2 files changed, 18 insertions(+), 2 deletions(-)
>
> diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md
> index 04ec0eeaa57..4cbbd4cf685 100644
> --- a/gcc/config/i386/i386.md
> +++ b/gcc/config/i386/i386.md
> @@ -3683,6 +3683,10 @@
>               (const_string "avx512bw")
>            ]
>            (const_string "*")))
> +   (set (attr "mmx_isa")
> +     (if_then_else (eq_attr "alternative" "5,6")
> +                  (const_string "native")
> +                  (const_string "*")))
>     (set (attr "type")
>       (cond [(eq_attr "alternative" "0,1,2,4")
>               (const_string "multi")
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index 97ec3795b82..4b415d255e0 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -7209,6 +7209,10 @@
>               (const_string "mmxmov")
>            ]
>            (const_string "sselog")))
> +   (set (attr "mmx_isa")
> +     (if_then_else (eq_attr "alternative" "7,8")
> +                  (const_string "native")
> +                  (const_string "*")))
>     (set (attr "prefix_data16")
>       (if_then_else (eq_attr "alternative" "3,4")
>                    (const_string "1")
> @@ -7244,7 +7248,8 @@
>     movss\t{%1, %0|%0, %1}
>     punpckldq\t{%2, %0|%0, %2}
>     movd\t{%1, %0|%0, %1}"
> -  [(set_attr "type" "sselog,ssemov,mmxcvt,mmxmov")
> +  [(set_attr "mmx_isa" "*,*,native,native")
> +   (set_attr "type" "sselog,ssemov,mmxcvt,mmxmov")
>     (set_attr "mode" "V4SF,SF,DI,DI")])
>
>  (define_insn "*vec_concatv4sf"
> @@ -14520,6 +14525,10 @@
>     punpckldq\t{%2, %0|%0, %2}
>     movd\t{%1, %0|%0, %1}"
>    [(set_attr "isa" "noavx,noavx,avx,avx512dq,noavx,noavx,avx,*,*,*")
> +   (set (attr "mmx_isa")
> +     (if_then_else (eq_attr "alternative" "8,9")
> +                  (const_string "native")
> +                  (const_string "*")))
>     (set (attr "type")
>       (cond [(eq_attr "alternative" "7")
>               (const_string "ssemov")
> @@ -14557,6 +14566,7 @@
>     punpckldq\t{%2, %0|%0, %2}
>     movd\t{%1, %0|%0, %1}"
>    [(set_attr "isa" "sse2,sse2,*,*,*,*")
> +   (set_attr "mmx_isa" "*,*,*,*,native,native")
>     (set_attr "type" "sselog,ssemov,sselog,ssemov,mmxcvt,mmxmov")
>     (set_attr "mode" "TI,TI,V4SF,SF,DI,DI")])
>
> @@ -14586,7 +14596,8 @@
>    "@
>     %vmovq\t{%1, %0|%0, %1}
>     movq2dq\t{%1, %0|%0, %1}"
> -  [(set_attr "type" "ssemov")
> +  [(set_attr "mmx_isa" "*,native")
> +   (set_attr "type" "ssemov")
>     (set_attr "prefix" "maybe_vex,orig")
>     (set_attr "mode" "TI")])
>
> @@ -14661,6 +14672,7 @@
>     %vmovq\t{%1, %0|%0, %1}
>     movq2dq\t{%1, %0|%0, %1}"
>    [(set_attr "isa" "x64,*,*")
> +   (set_attr "mmx_isa" "*,*,native")
>     (set_attr "type" "ssemov")
>     (set_attr "prefix_rex" "1,*,*")
>     (set_attr "prefix" "maybe_vex,maybe_vex,orig")
> --
> 2.20.1
>

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Sun, Feb 17, 2019 at 12:46 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Sun, Feb 17, 2019 at 10:49 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > On Sun, Feb 17, 2019 at 6:37 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >
> > > > > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > > > > >
> > > > > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > > > > >
> > > > > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > > > > >   (const_string "base"))
> > > > > > > > > >
> > > > > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > > > > >
> > > > > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > > > > >
> > > > > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > > > > couple tricky cases:
> > > > > > > > > >
> > > > > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > > > > adjusting source and mask operands together with memory address.
> > > > > > > > > >
> > > > > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > > > > in 64-bit mode.
> > > > > > > > > >
> > > > > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > > > > >
> > > > > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > > > > the upper 64 bits of destination XMM register.
> > > > > > > > > >
> > > > > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > > > > >
> > > > > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > > > > tested with
> > > > > > > > > >
> > > > > > > > > > --with-arch=native --with-cpu=native
> > > > > > > > > >
> > > > > > > > > > on AVX2 and AVX512F machines.
> > > > > > > > >
> > > > > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > > > > >
> > > > > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > > > > attribute in e.g.
> > > > > > > > >
> > > > > > > > > +  "@
> > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > > > >
> > > > > > > > > to:
> > > > > > > > >
> > > > > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > > > > >
> > > > > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > > > > more freedom to allocate the most effective register set even for
> > > > > > > > > 32bit targets.
> > > > > > > > >
> > > > > > > > > WDYT?
> > > > > > > > >
> > > > > > > >
> > > > > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > > > > >
> > > > > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > > > > should not *disable* SSE alternatives on 32bit targets.
> > > > >
> > > > > I don't think my patch set disables any SSE alternatives in 32-bit
> > > > > mode.   However,
> > > > > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > > > > alternatives in
> > > > >
> > > > > (define_insn "*mmx_<code><mode>3"
> > > > >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> > > > >         (any_logic:MMXMODEI
> > > > >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> > > > >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> > > > >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> > > > >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> > > > >   "@
> > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > >    (set_attr "type" "mmxadd,sselog,sselog")
> > > > >    (set_attr "mode" "DI,TI,TI")])
> > > > >
> > > > > register_mmxmem_operand must return true for SSE alternatives:
> > > >
> > > > It returns true for register and memory operands for 32bit targets, because
> > > >
> > > > #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > >
> > > Will
> > >
> > > (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]
> > >
> > > work well with RA?  I got some wrong code before register_mmxmem_operand
> > > was added to match "ym,x,Yv".
> >
> > I see no reason why it shouldn't.
>
> This will be equivalent to replace register_operand in
>
> [(match_operand:VI1_AVX512VLBW 1 "register_operand" "v")
>
> with nonimmediate_operand.  If it should work, I can do it in i386.md and
> sse.md to check it out.
>

I tried:

sed -i -e "s/\"register_operand\"[
\t]\+\(\"[^=^\+^f]\+\"[^=]\+$\)/\"nonimmediate_operand\" \1/" i386.md

and got

(gdb) call debug_rtx (insn)
(insn 65 19 67 2 (parallel [
            (set (reg/f:SI 97)
                (plus:SI (mem/u/c:SI (plus:SI (reg:SI 82)
                            (const:SI (unspec:SI [
                                        (symbol_ref:SI
("gomp_tls_data") [flags 0x62] <var_decl 0x7fffea6c5e10
gomp_tls_data>)
                                    ] UNSPEC_GOTNTPOFF))) [17  S4 A8])
                    (mem/u/c:SI (const_int 0 [0]) [0  S4 A8 AS2])))
            (clobber (reg:CC 17 flags))
        ]) "/export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c":139:7
-1
     (expr_list:REG_DEAD (reg:SI 82)
        (expr_list:REG_UNUSED (reg:CC 17 flags)
            (expr_list:REG_EQUIV (symbol_ref:SI ("gomp_tls_data")
[flags 0x62] <var_decl 0x7fffea6c5e10 gomp_tls_data>)
                (nil)))))
(gdb) c
Continuing.
during RTL pass: ira
/export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c: In
function ‘gomp_test_nest_lock_25’:
/export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c:149:1:
internal compiler error: in elimination_costs_in_insn, at
reload1.c:3640
  149 | }
      | ^
0x108b258 elimination_costs_in_insn
/export/gnu/import/git/gitlab/x86-gcc/gcc/reload1.c:3637
0x108596f calculate_elim_costs_all_insns()
/export/gnu/import/git/gitlab/x86-gcc/gcc/reload1.c:1609
0xe61a7a ira_costs()
/export/gnu/import/git/gitlab/x86-gcc/gcc/ira-costs.c:2298
0xe56613 ira_build()
/export/gnu/import/git/gitlab/x86-gcc/gcc/ira-build.c:3432
0xe4b31d ira
/export/gnu/import/git/gitlab/x86-gcc/gcc/ira.c:5346
0xe4bba0 execute
/export/gnu/import/git/gitlab/x86-gcc/gcc/ira.c:5657
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.


-- 
H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[Uros Bizjak]

On Mon, Feb 18, 2019 at 3:22 PM H.J. Lu <hjl.tools@gmail.com> wrote:

> > > > > > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > > > > > >
> > > > > > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > > > > > >
> > > > > > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > > > > > >   (const_string "base"))
> > > > > > > > > > >
> > > > > > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > > > > > >
> > > > > > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > > > > > >
> > > > > > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > > > > > couple tricky cases:
> > > > > > > > > > >
> > > > > > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > > > > > adjusting source and mask operands together with memory address.
> > > > > > > > > > >
> > > > > > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > > > > > in 64-bit mode.
> > > > > > > > > > >
> > > > > > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > > > > > >
> > > > > > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > > > > > the upper 64 bits of destination XMM register.
> > > > > > > > > > >
> > > > > > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > > > > > >
> > > > > > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > > > > > tested with
> > > > > > > > > > >
> > > > > > > > > > > --with-arch=native --with-cpu=native
> > > > > > > > > > >
> > > > > > > > > > > on AVX2 and AVX512F machines.
> > > > > > > > > >
> > > > > > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > > > > > >
> > > > > > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > > > > > attribute in e.g.
> > > > > > > > > >
> > > > > > > > > > +  "@
> > > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > > > > >
> > > > > > > > > > to:
> > > > > > > > > >
> > > > > > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > > > > > >
> > > > > > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > > > > > more freedom to allocate the most effective register set even for
> > > > > > > > > > 32bit targets.
> > > > > > > > > >
> > > > > > > > > > WDYT?
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > > > > > >
> > > > > > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > > > > > should not *disable* SSE alternatives on 32bit targets.
> > > > > >
> > > > > > I don't think my patch set disables any SSE alternatives in 32-bit
> > > > > > mode.   However,
> > > > > > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > > > > > alternatives in
> > > > > >
> > > > > > (define_insn "*mmx_<code><mode>3"
> > > > > >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> > > > > >         (any_logic:MMXMODEI
> > > > > >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> > > > > >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> > > > > >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> > > > > >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> > > > > >   "@
> > > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > > >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > >    (set_attr "type" "mmxadd,sselog,sselog")
> > > > > >    (set_attr "mode" "DI,TI,TI")])
> > > > > >
> > > > > > register_mmxmem_operand must return true for SSE alternatives:
> > > > >
> > > > > It returns true for register and memory operands for 32bit targets, because
> > > > >
> > > > > #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > >
> > > > Will
> > > >
> > > > (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]
> > > >
> > > > work well with RA?  I got some wrong code before register_mmxmem_operand
> > > > was added to match "ym,x,Yv".
> > >
> > > I see no reason why it shouldn't.
> >
> > This will be equivalent to replace register_operand in
> >
> > [(match_operand:VI1_AVX512VLBW 1 "register_operand" "v")
> >
> > with nonimmediate_operand.  If it should work, I can do it in i386.md and
> > sse.md to check it out.
> >
>
> I tried:
>
> sed -i -e "s/\"register_operand\"[
> \t]\+\(\"[^=^\+^f]\+\"[^=]\+$\)/\"nonimmediate_operand\" \1/" i386.md

I don't know what is the point in changing these operands, but

(match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv")

should work without problems.

Uros.

> and got
>
> (gdb) call debug_rtx (insn)
> (insn 65 19 67 2 (parallel [
>             (set (reg/f:SI 97)
>                 (plus:SI (mem/u/c:SI (plus:SI (reg:SI 82)
>                             (const:SI (unspec:SI [
>                                         (symbol_ref:SI
> ("gomp_tls_data") [flags 0x62] <var_decl 0x7fffea6c5e10
> gomp_tls_data>)
>                                     ] UNSPEC_GOTNTPOFF))) [17  S4 A8])
>                     (mem/u/c:SI (const_int 0 [0]) [0  S4 A8 AS2])))
>             (clobber (reg:CC 17 flags))
>         ]) "/export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c":139:7
> -1
>      (expr_list:REG_DEAD (reg:SI 82)
>         (expr_list:REG_UNUSED (reg:CC 17 flags)
>             (expr_list:REG_EQUIV (symbol_ref:SI ("gomp_tls_data")
> [flags 0x62] <var_decl 0x7fffea6c5e10 gomp_tls_data>)
>                 (nil)))))
> (gdb) c
> Continuing.
> during RTL pass: ira
> /export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c: In
> function ‘gomp_test_nest_lock_25’:
> /export/gnu/import/git/gitlab/x86-gcc/libgomp/config/linux/lock.c:149:1:
> internal compiler error: in elimination_costs_in_insn, at
> reload1.c:3640
>   149 | }
>       | ^
> 0x108b258 elimination_costs_in_insn
> /export/gnu/import/git/gitlab/x86-gcc/gcc/reload1.c:3637
> 0x108596f calculate_elim_costs_all_insns()
> /export/gnu/import/git/gitlab/x86-gcc/gcc/reload1.c:1609
> 0xe61a7a ira_costs()
> /export/gnu/import/git/gitlab/x86-gcc/gcc/ira-costs.c:2298
> 0xe56613 ira_build()
> /export/gnu/import/git/gitlab/x86-gcc/gcc/ira-build.c:3432
> 0xe4b31d ira
> /export/gnu/import/git/gitlab/x86-gcc/gcc/ira.c:5346
> 0xe4bba0 execute
> /export/gnu/import/git/gitlab/x86-gcc/gcc/ira.c:5657
> Please submit a full bug report,
> with preprocessed source if appropriate.
> Please include the complete backtrace with any bug report.
>
>
> --
> H.J.

Re: [PATCH 00/41] V8: Emulate MMX intrinsics with SSE

[H.J. Lu]

On Mon, Feb 18, 2019 at 6:37 AM Uros Bizjak <ubizjak@gmail.com> wrote:
>
> On Mon, Feb 18, 2019 at 3:22 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> > > > > > > > > > > > On x86-64, since __m64 is returned and passed in XMM registers, we can
> > > > > > > > > > > > emulate MMX intrinsics with SSE instructions. To support it, we added
> > > > > > > > > > > >
> > > > > > > > > > > >  #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > > > > > > > > >
> > > > > > > > > > > > ;; Define instruction set of MMX instructions
> > > > > > > > > > > > (define_attr "mmx_isa" "base,native,x64,x64_noavx,x64_avx"
> > > > > > > > > > > >   (const_string "base"))
> > > > > > > > > > > >
> > > > > > > > > > > >          (eq_attr "mmx_isa" "native")
> > > > > > > > > > > >            (symbol_ref "!TARGET_MMX_WITH_SSE")
> > > > > > > > > > > >          (eq_attr "mmx_isa" "x64")
> > > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE")
> > > > > > > > > > > >          (eq_attr "mmx_isa" "x64_avx")
> > > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && TARGET_AVX")
> > > > > > > > > > > >          (eq_attr "mmx_isa" "x64_noavx")
> > > > > > > > > > > >            (symbol_ref "TARGET_MMX_WITH_SSE && !TARGET_AVX")
> > > > > > > > > > > >
> > > > > > > > > > > > We added SSE emulation to MMX patterns and disabled MMX alternatives with
> > > > > > > > > > > > TARGET_MMX_WITH_SSE.
> > > > > > > > > > > >
> > > > > > > > > > > > Most of MMX instructions have equivalent SSE versions and results of some
> > > > > > > > > > > > SSE versions need to be reshuffled to the right order for MMX.  Thee are
> > > > > > > > > > > > couple tricky cases:
> > > > > > > > > > > >
> > > > > > > > > > > > 1. MMX maskmovq and SSE2 maskmovdqu aren't equivalent.  We emulate MMX
> > > > > > > > > > > > maskmovq with SSE2 maskmovdqu by zeroing out the upper 64 bits of the
> > > > > > > > > > > > mask operand and handle unmapped bits 64:127 at memory address by
> > > > > > > > > > > > adjusting source and mask operands together with memory address.
> > > > > > > > > > > >
> > > > > > > > > > > > 2. MMX movntq is emulated with SSE2 DImode movnti, which is available
> > > > > > > > > > > > in 64-bit mode.
> > > > > > > > > > > >
> > > > > > > > > > > > 3. MMX pshufb takes a 3-bit index while SSE pshufb takes a 4-bit index.
> > > > > > > > > > > > SSE emulation must clear the bit 4 in the shuffle control mask.
> > > > > > > > > > > >
> > > > > > > > > > > > 4. To emulate MMX cvtpi2p with SSE2 cvtdq2ps, we must properly preserve
> > > > > > > > > > > > the upper 64 bits of destination XMM register.
> > > > > > > > > > > >
> > > > > > > > > > > > Tests are also added to check each SSE emulation of MMX intrinsics.
> > > > > > > > > > > >
> > > > > > > > > > > > There are no regressions on i686 and x86-64.  For x86-64, GCC is also
> > > > > > > > > > > > tested with
> > > > > > > > > > > >
> > > > > > > > > > > > --with-arch=native --with-cpu=native
> > > > > > > > > > > >
> > > > > > > > > > > > on AVX2 and AVX512F machines.
> > > > > > > > > > >
> > > > > > > > > > > An idea that would take patch a step further also on 32 bit targets:
> > > > > > > > > > >
> > > > > > > > > > > *Assuming* that operations on XMM registers are as fast (or perhaps
> > > > > > > > > > > faster) than operations on MMX registers, we can change mmx_isa
> > > > > > > > > > > attribute in e.g.
> > > > > > > > > > >
> > > > > > > > > > > +  "@
> > > > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > > > +   p<logic>\t{%2, %0|%0, %2}
> > > > > > > > > > > +   vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > > > > > > +  [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > > > > > >
> > > > > > > > > > > to:
> > > > > > > > > > >
> > > > > > > > > > > [(set_attr "isa" "*,noavx,avx")
> > > > > > > > > > >  (set_attr "mmx_isa" "native,*,*")]
> > > > > > > > > > >
> > > > > > > > > > > So, for x86_64 everything stays the same, but for x86_32 we now allow
> > > > > > > > > > > intrinsics to use xmm registers in addition to mmx registers. We can't
> > > > > > > > > > > disable MMX for x64_32 anyway due to ISA constraints (and some tricky
> > > > > > > > > > > cases, e.g. monvti that works only for 64bit targets and e.g. maskmovq
> > > > > > > > > > > & similar, which are more efficient with MMX regs), but RA has much
> > > > > > > > > > > more freedom to allocate the most effective register set even for
> > > > > > > > > > > 32bit targets.
> > > > > > > > > > >
> > > > > > > > > > > WDYT?
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Since MMX registers are used to pass and return __m64 values,
> > > > > > > > > > we can't really get rid of MMX instructions in 32-bit mode.  If people
> > > > > > > > > > have to stay with 32-bit mode, they need MMX.  I don't think we should
> > > > > > > > > > extend TARGET_MMX_WITH_SSE to 32-bit mode.
> > > > > > > > >
> > > > > > > > > No, TARGET_MMX_WITH_SSE is still enabled only for 64bit targets. We
> > > > > > > > > should not *disable* SSE alternatives on 32bit targets.
> > > > > > >
> > > > > > > I don't think my patch set disables any SSE alternatives in 32-bit
> > > > > > > mode.   However,
> > > > > > > it DOES NOT enable any SSE alternatives in 32-bit mode.  To really enable SSE
> > > > > > > alternatives in
> > > > > > >
> > > > > > > (define_insn "*mmx_<code><mode>3"
> > > > > > >   [(set (match_operand:MMXMODEI 0 "register_operand" "=y,x,Yv")
> > > > > > >         (any_logic:MMXMODEI
> > > > > > >           (match_operand:MMXMODEI 1 "register_mmxmem_operand" "%0,0,Yv")
> > > > > > >           (match_operand:MMXMODEI 2 "register_mmxmem_operand" "ym,x,Yv")))]
> > > > > > >   "(TARGET_MMX || TARGET_MMX_WITH_SSE)
> > > > > > >    && ix86_binary_operator_ok (<CODE>, <MODE>mode, operands)"
> > > > > > >   "@
> > > > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > > > >    p<logic>\t{%2, %0|%0, %2}
> > > > > > >    vp<logic>\t{%2, %1, %0|%0, %1, %2}"
> > > > > > >   [(set_attr "mmx_isa" "native,x64_noavx,x64_avx")
> > > > > > >    (set_attr "type" "mmxadd,sselog,sselog")
> > > > > > >    (set_attr "mode" "DI,TI,TI")])
> > > > > > >
> > > > > > > register_mmxmem_operand must return true for SSE alternatives:
> > > > > >
> > > > > > It returns true for register and memory operands for 32bit targets, because
> > > > > >
> > > > > > #define TARGET_MMX_WITH_SSE (TARGET_64BIT && TARGET_SSE2)
> > > > >
> > > > > Will
> > > > >
> > > > > (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv"))))]
> > > > >
> > > > > work well with RA?  I got some wrong code before register_mmxmem_operand
> > > > > was added to match "ym,x,Yv".
> > > >
> > > > I see no reason why it shouldn't.
> > >
> > > This will be equivalent to replace register_operand in
> > >
> > > [(match_operand:VI1_AVX512VLBW 1 "register_operand" "v")
> > >
> > > with nonimmediate_operand.  If it should work, I can do it in i386.md and
> > > sse.md to check it out.
> > >
> >
> > I tried:
> >
> > sed -i -e "s/\"register_operand\"[
> > \t]\+\(\"[^=^\+^f]\+\"[^=]\+$\)/\"nonimmediate_operand\" \1/" i386.md
>
> I don't know what is the point in changing these operands, but

The point is we can't replace register_operand with nonimmediate_operand
in all places.

> (match_operand:V4HI 2 "nonimmediate_operand" "ym,x,Yv")
>
> should work without problems.
>

32-bit MMX has very low priority.  I will try it in the second phase.

-- 
H.J.