Patchwork [i386] : Introduce Y4 register constraint and merge SSE4_1 patterns

login
register
mail settings
Submitter Uros Bizjak
Date June 2, 2011, 5:46 p.m.
Message ID <BANLkTimqaeEuVOfk=Fefh_sTyrrk6nDy+g@mail.gmail.com>
Download mbox | patch
Permalink /patch/98448/
State New
Headers show

Comments

Uros Bizjak - June 2, 2011, 5:46 p.m.
Hello!

... and some unrelated cleanups involving simplifying a couple of
switch statements.

2011-06-02  Uros Bizjak  <ubizjak@gmail.com>

	* config/i386/i386.c (standard_sse_constant_p) <case 1>:
	Simplify switch statement.
	* config/i386/i386.md (*movdf_internal_rex64) <case 8,9,10>: Ditto.
	(*movdf_internal) <case 6,7,8>: Ditto.

	* config/i386/constraints.md (Y4): New constraint.
	* config/i386/sse.md (vec_set<mode>_0): Merge with
	*vec_set<mode>_0_sse4_1 and *vec_set<mode>_0_sse2.
	(*vec_extractv2di_1): Merge from *vec_extractv2di_1_sse2 and
	*vec_extractv2di_1_sse.
	(*vec_concatv2di_rex64): Merge from *vec_concatv2di_rex64_sse4_1
	and *vec_concatv2di_rex64_sse.

testsuite/ChangeLog:

2011-06-02  Uros Bizjak  <ubizjak@gmail.com>

	* gcc.target/i386/sse2-init-v2di-2: Update scan-assembler-times string.

Bootstrapped and regression tested on x86_64-pc-linux-gnu, committed
to mainline SVN.

Uros.

Patch

Index: testsuite/gcc.target/i386/sse2-init-v2di-2.c
===================================================================
--- testsuite/gcc.target/i386/sse2-init-v2di-2.c	(revision 174566)
+++ testsuite/gcc.target/i386/sse2-init-v2di-2.c	(working copy)
@@ -10,4 +10,4 @@  test (long long b)
   return _mm_cvtsi64_si128 (b); 
 }
 
-/* { dg-final { scan-assembler-times "\\*vec_concatv2di_rex64_sse4_1/4" 1 } } */
+/* { dg-final { scan-assembler-times "\\*vec_concatv2di_rex64/4" 1 } } */
Index: config/i386/i386.md
===================================================================
--- config/i386/i386.md	(revision 174566)
+++ config/i386/i386.md	(working copy)
@@ -2956,18 +2956,15 @@ 
     case 10:
       switch (get_attr_mode (insn))
 	{
-	case MODE_V4SF:
-	  return "%vmovaps\t{%1, %0|%0, %1}";
-	case MODE_V2DF:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vmovaps\t{%1, %0|%0, %1}";
-	  else
-	    return "%vmovapd\t{%1, %0|%0, %1}";
 	case MODE_TI:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vmovaps\t{%1, %0|%0, %1}";
-	  else
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
 	    return "%vmovdqa\t{%1, %0|%0, %1}";
+	case MODE_V2DF:
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
+	    return "%vmovapd\t{%1, %0|%0, %1}";
+	case MODE_V4SF:
+	  return "%vmovaps\t{%1, %0|%0, %1}";
+
 	case MODE_DI:
 	  return "%vmovq\t{%1, %0|%0, %1}";
 	case MODE_DF:
@@ -3102,18 +3099,15 @@ 
     case 8:
       switch (get_attr_mode (insn))
 	{
-	case MODE_V4SF:
-	  return "%vmovaps\t{%1, %0|%0, %1}";
-	case MODE_V2DF:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vmovaps\t{%1, %0|%0, %1}";
-	  else
-	    return "%vmovapd\t{%1, %0|%0, %1}";
 	case MODE_TI:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vmovaps\t{%1, %0|%0, %1}";
-	  else
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
 	    return "%vmovdqa\t{%1, %0|%0, %1}";
+	case MODE_V2DF:
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
+	    return "%vmovapd\t{%1, %0|%0, %1}";
+	case MODE_V4SF:
+	  return "%vmovaps\t{%1, %0|%0, %1}";
+
 	case MODE_DI:
 	  return "%vmovq\t{%1, %0|%0, %1}";
 	case MODE_DF:
Index: config/i386/constraints.md
===================================================================
--- config/i386/constraints.md	(revision 174566)
+++ config/i386/constraints.md	(working copy)
@@ -99,6 +99,9 @@ 
 (define_register_constraint "Y2" "TARGET_SSE2 ? SSE_REGS : NO_REGS"
  "@internal Any SSE register, when SSE2 is enabled.")
 
+(define_register_constraint "Y4" "TARGET_SSE4_1 ? SSE_REGS : NO_REGS"
+ "@internal Any SSE register, when SSE4_1 is enabled.")
+
 (define_register_constraint "Yi"
  "TARGET_SSE2 && TARGET_INTER_UNIT_MOVES ? SSE_REGS : NO_REGS"
  "@internal Any SSE register, when SSE2 and inter-unit moves are enabled.")
Index: config/i386/sse.md
===================================================================
--- config/i386/sse.md	(revision 174566)
+++ config/i386/sse.md	(working copy)
@@ -3376,79 +3376,35 @@ 
 
 ;; Avoid combining registers from different units in a single alternative,
 ;; see comment above inline_secondary_memory_needed function in i386.c
-(define_insn "*vec_set<mode>_0_sse4_1"
+(define_insn "vec_set<mode>_0"
   [(set (match_operand:VI4F_128 0 "nonimmediate_operand"
-	  "=x,x,x ,x,x,x  ,x  ,m,m,m")
+	  "=Y4,Y2,Y2,x,x,x,Y4 ,x  ,m,m,m")
 	(vec_merge:VI4F_128
 	  (vec_duplicate:VI4F_128
 	    (match_operand:<ssescalarmode> 2 "general_operand"
-	  " x,m,*r,x,x,*rm,*rm,x,*r,fF"))
+	  " Y4,m ,*r,m,x,x,*rm,*rm,x,*r,fF"))
 	  (match_operand:VI4F_128 1 "vector_move_operand"
-	  " C,C,C ,0,x,0  ,x  ,0,0 ,0")
+	  " C ,C ,C ,C,0,x,0  ,x  ,0,0 ,0")
 	  (const_int 1)))]
-  "TARGET_SSE4_1"
+  "TARGET_SSE"
   "@
    %vinsertps\t{$0xe, %d2, %0|%0, %d2, 0xe}
    %vmov<ssescalarmodesuffix>\t{%2, %0|%0, %2}
    %vmovd\t{%2, %0|%0, %2}
    movss\t{%2, %0|%0, %2}
+   movss\t{%2, %0|%0, %2}
    vmovss\t{%2, %1, %0|%0, %1, %2}
    pinsrd\t{$0, %2, %0|%0, %2, 0}
    vpinsrd\t{$0, %2, %1, %0|%0, %1, %2, 0}
    #
    #
    #"
-  [(set_attr "isa" "base,base,base,noavx,avx,noavx,avx,base,base,base")
-   (set_attr "type" "sselog,ssemov,ssemov,ssemov,ssemov,sselog,sselog,*,*,*")
-   (set_attr "prefix_extra" "*,*,*,*,*,1,1,*,*,*")
-   (set_attr "length_immediate" "*,*,*,*,*,1,1,*,*,*")
-   (set_attr "prefix" "maybe_vex,maybe_vex,maybe_vex,orig,vex,orig,vex,*,*,*")
-   (set_attr "mode" "SF,<ssescalarmode>,SI,SF,SF,TI,TI,*,*,*")])
-
-;; Avoid combining registers from different units in a single alternative,
-;; see comment above inline_secondary_memory_needed function in i386.c
-(define_insn "*vec_set<mode>_0_sse2"
-  [(set (match_operand:VI4F_128 0 "nonimmediate_operand"
-	  "=x,x ,x,m,m ,m")
-	(vec_merge:VI4F_128
-	  (vec_duplicate:VI4F_128
-	    (match_operand:<ssescalarmode> 2 "general_operand"
-	  " m,*r,x,x,*r,fF"))
-	  (match_operand:VI4F_128 1 "vector_move_operand"
-	  " C, C,0,0,0 ,0")
-	  (const_int 1)))]
-  "TARGET_SSE2"
-  "@
-   mov<ssescalarmodesuffix>\t{%2, %0|%0, %2}
-   movd\t{%2, %0|%0, %2}
-   movss\t{%2, %0|%0, %2}
-   #
-   #
-   #"
-  [(set_attr "type" "ssemov")
-   (set_attr "mode" "<ssescalarmode>,SI,SF,*,*,*")])
-
-;; Avoid combining registers from different units in a single alternative,
-;; see comment above inline_secondary_memory_needed function in i386.c
-(define_insn "vec_set<mode>_0"
-  [(set (match_operand:VI4F_128 0 "nonimmediate_operand"
-	  "=x,x,m,m ,m")
-	(vec_merge:VI4F_128
-	  (vec_duplicate:VI4F_128
-	    (match_operand:<ssescalarmode> 2 "general_operand"
-	  " m,x,x,*r,fF"))
-	  (match_operand:VI4F_128 1 "vector_move_operand"
-	  " C,0,0,0 ,0")
-	  (const_int 1)))]
-  "TARGET_SSE"
-  "@
-   movss\t{%2, %0|%0, %2}
-   movss\t{%2, %0|%0, %2}
-   #
-   #
-   #"
-  [(set_attr "type" "ssemov")
-   (set_attr "mode" "SF,SF,*,*,*")])
+  [(set_attr "isa" "base,base,base,noavx,noavx,avx,noavx,avx,base,base,base")
+   (set_attr "type" "sselog,ssemov,ssemov,ssemov,ssemov,ssemov,sselog,sselog,*,*,*")
+   (set_attr "prefix_extra" "*,*,*,*,*,*,1,1,*,*,*")
+   (set_attr "length_immediate" "*,*,*,*,*,*,1,1,*,*,*")
+   (set_attr "prefix" "maybe_vex,maybe_vex,maybe_vex,orig,orig,vex,orig,vex,*,*,*")
+   (set_attr "mode" "SF,<ssescalarmode>,SI,SF,SF,SF,TI,TI,*,*,*")])
 
 ;; A subset is vec_setv4sf.
 (define_insn "*vec_setv4sf_sse4_1"
@@ -6214,7 +6170,7 @@ 
    (set_attr "prefix" "maybe_vex")
    (set_attr "mode" "TI")])
 
-;; It must come before *vec_extractv2di_1_sse since it is preferred.
+;; It must come before *vec_extractv2di_1_rex64 since it is preferred.
 (define_insn "*sse4_1_pextrq"
   [(set (match_operand:DI 0 "nonimmediate_operand" "=rm")
 	(vec_select:DI
@@ -6478,39 +6434,26 @@ 
    (set_attr "prefix" "maybe_vex,orig,vex,maybe_vex,orig")
    (set_attr "mode" "V2SF,TI,TI,TI,DI")])
 
-(define_insn "*vec_extractv2di_1_sse2"
-  [(set (match_operand:DI 0 "nonimmediate_operand"     "=m,x,x,x")
+(define_insn "*vec_extractv2di_1"
+  [(set (match_operand:DI 0 "nonimmediate_operand"     "=m,Y2,Y2,Y2,x,x")
 	(vec_select:DI
-	  (match_operand:V2DI 1 "nonimmediate_operand" " x,0,x,o")
+	  (match_operand:V2DI 1 "nonimmediate_operand" " x,0 ,Y2,o ,x,o")
 	  (parallel [(const_int 1)])))]
-  "!TARGET_64BIT
-   && TARGET_SSE2 && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
+  "!TARGET_64BIT && TARGET_SSE
+   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
   "@
    %vmovhps\t{%1, %0|%0, %1}
    psrldq\t{$8, %0|%0, 8}
    vpsrldq\t{$8, %1, %0|%0, %1, 8}
-   %vmovq\t{%H1, %0|%0, %H1}"
-  [(set_attr "isa" "base,noavx,avx,base")
-   (set_attr "type" "ssemov,sseishft1,sseishft1,ssemov")
-   (set_attr "length_immediate" "*,1,1,*")
-   (set_attr "memory" "*,none,none,*")
-   (set_attr "prefix" "maybe_vex,orig,vex,maybe_vex")
-   (set_attr "mode" "V2SF,TI,TI,TI")])
-
-;; Not sure this is ever used, but it doesn't hurt to have it. -aoliva
-(define_insn "*vec_extractv2di_1_sse"
-  [(set (match_operand:DI 0 "nonimmediate_operand"     "=m,x,x")
-	(vec_select:DI
-	  (match_operand:V2DI 1 "nonimmediate_operand" " x,x,o")
-	  (parallel [(const_int 1)])))]
-  "!TARGET_SSE2 && TARGET_SSE
-   && !(MEM_P (operands[0]) && MEM_P (operands[1]))"
-  "@
-   movhps\t{%1, %0|%0, %1}
+   %vmovq\t{%H1, %0|%0, %H1}
    movhlps\t{%1, %0|%0, %1}
    movlps\t{%H1, %0|%0, %H1}"
-  [(set_attr "type" "ssemov")
-   (set_attr "mode" "V2SF,V4SF,V2SF")])
+  [(set_attr "isa" "base,noavx,avx,base,noavx,noavx")
+   (set_attr "type" "ssemov,sseishft1,sseishft1,ssemov,ssemov,ssemov")
+   (set_attr "length_immediate" "*,1,1,*,*,*")
+   (set_attr "memory" "*,none,none,*,*,*")
+   (set_attr "prefix" "maybe_vex,orig,vex,maybe_vex,orig,orig")
+   (set_attr "mode" "V2SF,TI,TI,TI,V4SF,V2SF")])
 
 (define_insn "*vec_dupv4si_avx"
   [(set (match_operand:V4SI 0 "register_operand"     "=x,x")
@@ -6570,8 +6513,8 @@ 
 	  (match_operand:SI 2 "vector_move_operand"  "rm,rm,x,x, C,*ym, C")))]
   "TARGET_SSE4_1"
   "@
-   pinsrd\t{$0x1, %2, %0|%0, %2, 0x1}
-   vpinsrd\t{$0x1, %2, %1, %0|%0, %1, %2, 0x1}
+   pinsrd\t{$1, %2, %0|%0, %2, 1}
+   vpinsrd\t{$1, %2, %1, %0|%0, %1, %2, 1}
    punpckldq\t{%2, %0|%0, %2}
    vpunpckldq\t{%2, %1, %0|%0, %1, %2}
    %vmovd\t{%1, %0|%0, %1}
@@ -6633,18 +6576,18 @@ 
    (set_attr "mode" "TI,TI,V4SF,V2SF,V2SF")])
 
 ;; movd instead of movq is required to handle broken assemblers.
-(define_insn "*vec_concatv2di_rex64_sse4_1"
+(define_insn "*vec_concatv2di_rex64"
   [(set (match_operand:V2DI 0 "register_operand"
-	  "=x, x, x,Yi,!x,x,x,x,x")
+	  "=Y4,x ,x ,Yi,!x,x,x,x,x")
 	(vec_concat:V2DI
 	  (match_operand:DI 1 "nonimmediate_operand"
-	  " 0, x,xm,r ,*y,0,x,0,x")
+	  " 0 ,x ,xm,r ,*y,0,x,0,x")
 	  (match_operand:DI 2 "vector_move_operand"
-	  "rm,rm, C,C ,C ,x,x,m,m")))]
-  "TARGET_64BIT && TARGET_SSE4_1"
+	  " rm,rm,C ,C ,C ,x,x,m,m")))]
+  "TARGET_64BIT"
   "@
-   pinsrq\t{$0x1, %2, %0|%0, %2, 0x1}
-   vpinsrq\t{$0x1, %2, %1, %0|%0, %1, %2, 0x1}
+   pinsrq\t{$1, %2, %0|%0, %2, 1}
+   vpinsrq\t{$1, %2, %1, %0|%0, %1, %2, 1}
    %vmovq\t{%1, %0|%0, %1}
    %vmovd\t{%1, %0|%0, %1}
    movq2dq\t{%1, %0|%0, %1}
@@ -6653,7 +6596,11 @@ 
    movhps\t{%2, %0|%0, %2}
    vmovhps\t{%2, %1, %0|%0, %1, %2}"
   [(set_attr "isa" "noavx,avx,base,base,base,noavx,avx,noavx,avx")
-   (set_attr "type" "sselog,sselog,ssemov,ssemov,ssemov,sselog,sselog,ssemov,ssemov")
+   (set (attr "type")
+     (if_then_else
+       (eq_attr "alternative" "0,1,5,6")
+       (const_string "sselog")
+       (const_string "ssemov")))
    (set (attr "prefix_rex")
      (if_then_else
        (and (eq_attr "alternative" "0,3")
@@ -6665,24 +6612,6 @@ 
    (set_attr "prefix" "orig,vex,maybe_vex,maybe_vex,orig,orig,vex,orig,vex")
    (set_attr "mode" "TI,TI,TI,TI,TI,TI,TI,V2SF,V2SF")])
 
-;; movd instead of movq is required to handle broken assemblers.
-(define_insn "*vec_concatv2di_rex64_sse"
-  [(set (match_operand:V2DI 0 "register_operand"     "=Y2,Yi,!Y2,Y2,x,x")
-	(vec_concat:V2DI
-	  (match_operand:DI 1 "nonimmediate_operand" "Y2m,r ,*y ,0 ,0,0")
-	  (match_operand:DI 2 "vector_move_operand"  " C ,C ,C  ,Y2,x,m")))]
-  "TARGET_64BIT && TARGET_SSE"
-  "@
-   movq\t{%1, %0|%0, %1}
-   movd\t{%1, %0|%0, %1}
-   movq2dq\t{%1, %0|%0, %1}
-   punpcklqdq\t{%2, %0|%0, %2}
-   movlhps\t{%2, %0|%0, %2}
-   movhps\t{%2, %0|%0, %2}"
-  [(set_attr "type" "ssemov,ssemov,ssemov,sselog,ssemov,ssemov")
-   (set_attr "prefix_rex" "*,1,*,*,*,*")
-   (set_attr "mode" "TI,TI,TI,TI,V4SF,V2SF")])
-
 (define_insn "vec_concatv2di"
   [(set (match_operand:V2DI 0 "register_operand"     "=Y2,?Y2,Y2,x,x,x,x")
 	(vec_concat:V2DI
Index: config/i386/i386.c
===================================================================
--- config/i386/i386.c	(revision 174566)
+++ config/i386/i386.c	(working copy)
@@ -8601,33 +8601,28 @@  standard_sse_constant_opcode (rtx insn, 
     case 1:
       switch (get_attr_mode (insn))
 	{
-	case MODE_V4SF:
-	  return "%vxorps\t%0, %d0";
-	case MODE_V2DF:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vxorps\t%0, %d0";
-	  else
-	    return "%vxorpd\t%0, %d0";
 	case MODE_TI:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "%vxorps\t%0, %d0";
-	  else
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
 	    return "%vpxor\t%0, %d0";
-	case MODE_V8SF:
-	  return "vxorps\t%x0, %x0, %x0";
-	case MODE_V4DF:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "vxorps\t%x0, %x0, %x0";
-	  else
-	    return "vxorpd\t%x0, %x0, %x0";
+	case MODE_V2DF:
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
+	    return "%vxorpd\t%0, %d0";
+	case MODE_V4SF:
+	  return "%vxorps\t%0, %d0";
+
 	case MODE_OI:
-	  if (TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
-	    return "vxorps\t%x0, %x0, %x0";
-	  else
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
 	    return "vpxor\t%x0, %x0, %x0";
+	case MODE_V4DF:
+	  if (!TARGET_SSE_PACKED_SINGLE_INSN_OPTIMAL)
+	    return "vxorpd\t%x0, %x0, %x0";
+	case MODE_V8SF:
+	  return "vxorps\t%x0, %x0, %x0";
+
 	default:
 	  break;
 	}
+
     case 2:
       return "%vpcmpeqd\t%0, %d0";
     default: