i386: psrlq is not used for PERM<a,{0},1,2,3,4> [PR113871]

Introduce vec_shl_<mode> and vec_shr_<mode> expanders to improve

	'*a = __builtin_shufflevector(*a, (vect64){0}, 1, 2, 3, 4);'

and
	'*a = __builtin_shufflevector((vect64){0}, *a, 3, 4, 5, 6);'

shuffles.  The generated code improves from:

	movzwl  6(%rdi), %eax
	movzwl  4(%rdi), %edx
	salq    $16, %rax
	orq     %rdx, %rax
	movzwl  2(%rdi), %edx
	salq    $16, %rax
	orq     %rdx, %rax
	movq    %rax, (%rdi)

to:
	movq    (%rdi), %xmm0
	psrlq   $16, %xmm0
	movq    %xmm0, (%rdi)

and to:
	movq    (%rdi), %xmm0
	psllq   $16, %xmm0
	movq    %xmm0, (%rdi)

in the second case.

The patch handles 32-bit vectors as well and improves generated code from:

	movd    (%rdi), %xmm0
	pxor    %xmm1, %xmm1
	punpcklwd       %xmm1, %xmm0
	pshuflw $230, %xmm0, %xmm0
	movd    %xmm0, (%rdi)

to:
	movd    (%rdi), %xmm0
	psrld   $16, %xmm0
	movd    %xmm0, (%rdi)

and to:
	movd    (%rdi), %xmm0
	pslld   $16, %xmm0
	movd    %xmm0, (%rdi)

	PR target/113871

gcc/ChangeLog:

	* config/i386/mmx.md (V248FI): New mode iterator.
	(V24FI_32): DItto.
	(vec_shl_<V248FI:mode>): New expander.
	(vec_shl_<V24FI_32:mode>): Ditto.
	(vec_shr_<V248FI:mode>): Ditto.
	(vec_shr_<V24FI_32:mode>): Ditto.
	* config/i386/sse.md (vec_shl_<V_128:mode>): Simplify expander.
	(vec_shr_<V248FI:mode>): Ditto.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/pr113871-1a.c: New test.
	* gcc.target/i386/pr113871-1b.c: New test.
	* gcc.target/i386/pr113871-2a.c: New test.
	* gcc.target/i386/pr113871-2b.c: New test.
	* gcc.target/i386/pr113871-3a.c: New test.
	* gcc.target/i386/pr113871-3b.c: New test.
	* gcc.target/i386/pr113871-4a.c: New test.

9 files changed, 222 insertions, 14 deletions

diff --git a/gcc/config/i386/mmx.md b/gcc/config/i386/mmx.md
index 6215b12..075309c 100644
--- a/gcc/config/i386/mmx.md
+++ b/gcc/config/i386/mmx.md
@@ -84,6 +84,11 @@
 (define_mode_iterator V2FI [V2SF V2SI])
 
 (define_mode_iterator V24FI [V2SF V2SI V4HF V4HI])
+
+(define_mode_iterator V248FI [V2SF V2SI V4HF V4HI V8QI])
+
+(define_mode_iterator V24FI_32 [V2HF V2HI V4QI])
+
 ;; Mapping from integer vector mode to mnemonic suffix
 (define_mode_attr mmxvecsize
   [(V8QI "b") (V4QI "b") (V2QI "b")
@@ -3729,6 +3734,70 @@
   DONE;
 })
 
+(define_expand "vec_shl_<mode>"
+  [(set (match_operand:V248FI 0 "register_operand")
+	(ashift:V1DI
+	  (match_operand:V248FI 1 "nonimmediate_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_MMX_WITH_SSE"
+{
+  rtx op0 = gen_reg_rtx (V1DImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_mmx_ashlv1di3
+	      (op0, gen_lowpart (V1DImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
+})
+
+(define_expand "vec_shl_<mode>"
+  [(set (match_operand:V24FI_32 0 "register_operand")
+	(ashift:V1SI
+	  (match_operand:V24FI_32 1 "nonimmediate_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_SSE2"
+{
+  rtx op0 = gen_reg_rtx (V1SImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_mmx_ashlv1si3
+	      (op0, gen_lowpart (V1SImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
+})
+
+(define_expand "vec_shr_<mode>"
+  [(set (match_operand:V248FI 0 "register_operand")
+	(lshiftrt:V1DI
+	  (match_operand:V248FI 1 "nonimmediate_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_MMX_WITH_SSE"
+{
+  rtx op0 = gen_reg_rtx (V1DImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_mmx_lshrv1di3
+	      (op0, gen_lowpart (V1DImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
+})
+
+(define_expand "vec_shr_<mode>"
+  [(set (match_operand:V24FI_32 0 "register_operand")
+	(lshiftrt:V1SI
+	  (match_operand:V24FI_32 1 "nonimmediate_operand")
+	  (match_operand:DI 2 "nonmemory_operand")))]
+  "TARGET_SSE2"
+{
+  rtx op0 = gen_reg_rtx (V1SImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_mmx_lshrv1si3
+	      (op0, gen_lowpart (V1SImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
+})
+
 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
 ;;
 ;; Parallel integral comparisons
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index acd1090..1bc614a 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -16498,29 +16498,35 @@
   "operands[3] = XVECEXP (operands[2], 0, 0);")
 
 (define_expand "vec_shl_<mode>"
-  [(set (match_dup 3)
+  [(set (match_operand:V_128 0 "register_operand")
 	(ashift:V1TI
-	 (match_operand:V_128 1 "register_operand")
-	 (match_operand:SI 2 "const_0_to_255_mul_8_operand")))
-   (set (match_operand:V_128 0 "register_operand") (match_dup 4))]
+	 (match_operand:V_128 1 "nonimmediate_operand")
+	 (match_operand:SI 2 "const_0_to_255_mul_8_operand")))]
   "TARGET_SSE2"
 {
-  operands[1] = gen_lowpart (V1TImode, operands[1]);
-  operands[3] = gen_reg_rtx (V1TImode);
-  operands[4] = gen_lowpart (<MODE>mode, operands[3]);
+  rtx op0 = gen_reg_rtx (V1TImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_sse2_ashlv1ti3
+	      (op0, gen_lowpart (V1TImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
 })
 
 (define_expand "vec_shr_<mode>"
-  [(set (match_dup 3)
+  [(set (match_operand:V_128 0 "register_operand")
 	(lshiftrt:V1TI
-	 (match_operand:V_128 1 "register_operand")
-	 (match_operand:SI 2 "const_0_to_255_mul_8_operand")))
-   (set (match_operand:V_128 0 "register_operand") (match_dup 4))]
+	 (match_operand:V_128 1 "nonimmediate_operand")
+	 (match_operand:SI 2 "const_0_to_255_mul_8_operand")))]
   "TARGET_SSE2"
 {
-  operands[1] = gen_lowpart (V1TImode, operands[1]);
-  operands[3] = gen_reg_rtx (V1TImode);
-  operands[4] = gen_lowpart (<MODE>mode, operands[3]);
+  rtx op0 = gen_reg_rtx (V1TImode);
+  rtx op1 = force_reg (<MODE>mode, operands[1]);
+
+  emit_insn (gen_sse2_lshrv1ti3
+	      (op0, gen_lowpart (V1TImode, op1), operands[2]));
+  emit_move_insn (operands[0], gen_lowpart (<MODE>mode, op0));
+  DONE;
 })
 
 (define_expand "ashlv1ti3"
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-1a.c b/gcc/testsuite/gcc.target/i386/pr113871-1a.c
new file mode 100644
index 0000000..f720927
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-1a.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2" } */
+
+typedef char vect64 __attribute__((vector_size(8)));
+
+void f (vect64 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect64){0}, 1, 2, 3, 4, 5, 6, 7, 8);
+}
+
+/* { dg-final { scan-assembler "psrlq" } } */
+
+void g(vect64 *a)
+{
+  *a = __builtin_shufflevector((vect64){0}, *a, 7, 8, 9, 10, 11, 12, 13, 14);
+}
+
+/* { dg-final { scan-assembler "psllq" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-1b.c b/gcc/testsuite/gcc.target/i386/pr113871-1b.c
new file mode 100644
index 0000000..705cf5c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-1b.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -msse2" } */
+
+typedef char vect32 __attribute__((vector_size(4)));
+
+void f (vect32 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect32){0}, 1, 2, 3, 4);
+}
+
+/* { dg-final { scan-assembler "psrld" } } */
+
+void g(vect32 *a)
+{
+  *a = __builtin_shufflevector((vect32){0}, *a, 3, 4, 5, 6);
+}
+
+/* { dg-final { scan-assembler "pslld" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-2a.c b/gcc/testsuite/gcc.target/i386/pr113871-2a.c
new file mode 100644
index 0000000..5430f69
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-2a.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2" } */
+
+typedef short vect64 __attribute__((vector_size(8)));
+
+void f (vect64 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect64){0}, 1, 2, 3, 4);
+}
+
+/* { dg-final { scan-assembler "psrlq" } } */
+
+void g(vect64 *a)
+{
+  *a = __builtin_shufflevector((vect64){0}, *a, 3, 4, 5, 6);
+}
+
+/* { dg-final { scan-assembler "psllq" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-2b.c b/gcc/testsuite/gcc.target/i386/pr113871-2b.c
new file mode 100644
index 0000000..06e2a44
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-2b.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -msse2" } */
+
+typedef short vect32 __attribute__((vector_size(4)));
+
+void f (vect32 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect32){0}, 1, 2);
+}
+
+/* { dg-final { scan-assembler "psrld" } } */
+
+void g(vect32 *a)
+{
+  *a = __builtin_shufflevector((vect32){0}, *a, 1, 2);
+}
+
+/* { dg-final { scan-assembler "pslld" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-3a.c b/gcc/testsuite/gcc.target/i386/pr113871-3a.c
new file mode 100644
index 0000000..825d48e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-3a.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2" } */
+
+typedef _Float16 vect64 __attribute__((vector_size(8)));
+
+void f (vect64 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect64){0}, 1, 2, 3, 4);
+}
+
+/* { dg-final { scan-assembler "psrlq" } } */
+
+void g(vect64 *a)
+{
+  *a = __builtin_shufflevector((vect64){0}, *a, 3, 4, 5, 6);
+}
+
+/* { dg-final { scan-assembler "psllq" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-3b.c b/gcc/testsuite/gcc.target/i386/pr113871-3b.c
new file mode 100644
index 0000000..f8e0299
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-3b.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -msse2" } */
+
+typedef _Float16 vect32 __attribute__((vector_size(4)));
+
+void f (vect32 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect32){0}, 1, 2);
+}
+
+/* { dg-final { scan-assembler "psrld" } } */
+
+void g(vect32 *a)
+{
+  *a = __builtin_shufflevector((vect32){0}, *a, 1, 2);
+}
+
+/* { dg-final { scan-assembler "pslld" } } */
diff --git a/gcc/testsuite/gcc.target/i386/pr113871-4a.c b/gcc/testsuite/gcc.target/i386/pr113871-4a.c
new file mode 100644
index 0000000..3887b1f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr113871-4a.c
@@ -0,0 +1,19 @@
+/* PR target/113871 */
+/* { dg-do compile { target { ! ia32 } } } */
+/* { dg-options "-O2" } */
+
+typedef int vect64 __attribute__((vector_size(8)));
+
+void f (vect64 *a)
+{
+  *a = __builtin_shufflevector(*a, (vect64){0}, 1, 2);
+}
+
+/* { dg-final { scan-assembler "psrlq" } } */
+
+void g(vect64 *a)
+{
+  *a = __builtin_shufflevector((vect64){0}, *a, 1, 2);
+}
+
+/* { dg-final { scan-assembler "psllq" } } */