sse.md (VI4SD_AVX2): Removed.

	* config/i386/sse.md (VI4SD_AVX2): Removed.
	(VI48_AVX2, VI128_128, VI48_128, VI48_256): New mode iterators.
	(vashl<mode>3): Use VI12_128 iterator instead of VI124_128.
	Add another expander using VI48_128 iterator for
	TARGET_AVX2 || TARGET_XOP and another using VI48_256 iterator
	for TARGET_AVX2.
	(vlshr<mode>3): Likewise.  Change register_operand predicate to
	nonimmediate_operand on last operand in the VI12_128 expander.
	(vashr<mode>3): Use VI128_128 iterator instead of VI124_128.
	(vashrv4si3, vashrv8si3): New expanders.
	(avx2_ashrvv8si, avx2_ashrvv4si, avx2_<lshift>vv8si,
	avx2_<lshift>vv2di): Removed.
	(avx2_ashrv<mode>): New insn with VI4_AVX2 iterator.
	(avx2_<lshift>v<mode>): Macroize using VI48_AVX2
	iterator.  Simplify pattern.

	* gcc.dg/vshift-1.c: New test.
	* gcc.dg/vshift-2.c: New test.
	* gcc.target/i386/xop-vshift-1.c: New test.
	* gcc.target/i386/xop-vshift-2.c: New test.
	* gcc.target/i386/avx2-vshift-1.c: New test.

From-SVN: r180609

gcc/ChangeLog

+2011-10-28  Jakub Jelinek  <jakub@redhat.com>
+
+	* config/i386/sse.md (VI4SD_AVX2): Removed.
+	(VI48_AVX2, VI128_128, VI48_128, VI48_256): New mode iterators.
+	(vashl<mode>3): Use VI12_128 iterator instead of VI124_128.
+	Add another expander using VI48_128 iterator for
+	TARGET_AVX2 || TARGET_XOP and another using VI48_256 iterator
+	for TARGET_AVX2.
+	(vlshr<mode>3): Likewise.  Change register_operand predicate to
+	nonimmediate_operand on last operand in the VI12_128 expander.
+	(vashr<mode>3): Use VI128_128 iterator instead of VI124_128.
+	(vashrv4si3, vashrv8si3): New expanders.
+	(avx2_ashrvv8si, avx2_ashrvv4si, avx2_<lshift>vv8si,
+	avx2_<lshift>vv2di): Removed.
+	(avx2_ashrv<mode>): New insn with VI4_AVX2 iterator.
+	(avx2_<lshift>v<mode>): Macroize using VI48_AVX2
+	iterator.  Simplify pattern.
+
 2010-10-28  Richard Guenther  <rguenther@suse.de>
 
 	PR driver/50876

gcc/testsuite/ChangeLog

+2011-10-28  Jakub Jelinek  <jakub@redhat.com>
+
+	* gcc.dg/vshift-1.c: New test.
+	* gcc.dg/vshift-2.c: New test.
+	* gcc.target/i386/xop-vshift-1.c: New test.
+	* gcc.target/i386/xop-vshift-2.c: New test.
+	* gcc.target/i386/avx2-vshift-1.c: New test.
+
 2011-10-28  Chung-Lin Tang  <cltang@codesourcery.com>
 
 	PR rtl-optimization/49720

gcc/testsuite/gcc.dg/vshift-1.c

+/* { dg-do run } */
+/* { dg-options "-O3" } */
+
+#include <stdlib.h>
+
+#define N 64
+
+#ifndef TYPE1
+#define TYPE1 int
+#define TYPE2 long long
+#endif
+
+signed TYPE1 a[N], b[N], g[N];
+unsigned TYPE1 c[N], h[N];
+signed TYPE2 d[N], e[N], j[N];
+unsigned TYPE2 f[N], k[N];
+
+__attribute__((noinline)) void
+f1 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    g[i] = a[i] << b[i];
+}
+
+__attribute__((noinline)) void
+f2 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    g[i] = a[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f3 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    h[i] = c[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f4 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] << e[i];
+}
+
+__attribute__((noinline)) void
+f5 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] >> e[i];
+}
+
+__attribute__((noinline)) void
+f6 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    k[i] = f[i] >> e[i];
+}
+
+__attribute__((noinline)) void
+f7 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] << b[i];
+}
+
+__attribute__((noinline)) void
+f8 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f9 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    k[i] = f[i] >> b[i];
+}
+
+int
+main ()
+{
+  int i;
+  for (i = 0; i < N; i++)
+    {
+      asm ("");
+      c[i] = (random () << 1) | (random () & 1);
+      b[i] = (i * 85) & (sizeof (TYPE1) * __CHAR_BIT__ - 1);
+      a[i] = c[i];
+      d[i] = (random () << 1) | (random () & 1);
+      d[i] |= (unsigned long long) c[i] << 32;
+      e[i] = (i * 85) & (sizeof (TYPE2) * __CHAR_BIT__ - 1);
+      f[i] = d[i];
+    }
+  f1 ();
+  f3 ();
+  f4 ();
+  f6 ();
+  for (i = 0; i < N; i++)
+    if (g[i] != (signed TYPE1) (a[i] << b[i])
+	|| h[i] != (unsigned TYPE1) (c[i] >> b[i])
+	|| j[i] != (signed TYPE2) (d[i] << e[i])
+	|| k[i] != (unsigned TYPE2) (f[i] >> e[i]))
+      abort ();
+  f2 ();
+  f5 ();
+  f9 ();
+  for (i = 0; i < N; i++)
+    if (g[i] != (signed TYPE1) (a[i] >> b[i])
+	|| j[i] != (signed TYPE2) (d[i] >> e[i])
+	|| k[i] != (unsigned TYPE2) (f[i] >> b[i]))
+      abort ();
+  f7 ();
+  for (i = 0; i < N; i++)
+    if (j[i] != (signed TYPE2) (d[i] << b[i]))
+      abort ();
+  f8 ();
+  for (i = 0; i < N; i++)
+    if (j[i] != (signed TYPE2) (d[i] >> b[i]))
+      abort ();
+  return 0;
+}

gcc/testsuite/gcc.dg/vshift-2.c

+/* { dg-do run } */
+/* { dg-options "-O3" } */
+
+#define TYPE1 char
+#define TYPE2 short
+
+#include "vshift-1.c"

gcc/testsuite/gcc.target/i386/avx2-vshift-1.c

+/* { dg-do run } */
+/* { dg-options "-O3 -mavx2" } */
+/* { dg-require-effective-target avx2 } */
+
+#ifndef CHECK_H
+#define CHECK_H "avx2-check.h"
+#endif
+
+#ifndef TEST
+#define TEST avx2_test
+#endif
+
+#include "xop-vshift-1.c"

gcc/testsuite/gcc.target/i386/xop-vshift-1.c

+/* { dg-do run } */
+/* { dg-options "-O3 -mxop" } */
+/* { dg-require-effective-target xop } */
+
+#ifndef CHECK_H
+#define CHECK_H "xop-check.h"
+#endif
+
+#ifndef TEST
+#define TEST xop_test
+#endif
+
+#include CHECK_H
+
+#define N 64
+
+#ifndef TYPE1
+#define TYPE1 int
+#define TYPE2 long long
+#endif
+
+signed TYPE1 a[N], b[N], g[N];
+unsigned TYPE1 c[N], h[N];
+signed TYPE2 d[N], e[N], j[N];
+unsigned TYPE2 f[N], k[N];
+
+__attribute__((noinline)) void
+f1 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    g[i] = a[i] << b[i];
+}
+
+__attribute__((noinline)) void
+f2 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    g[i] = a[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f3 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    h[i] = c[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f4 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] << e[i];
+}
+
+__attribute__((noinline)) void
+f5 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] >> e[i];
+}
+
+__attribute__((noinline)) void
+f6 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    k[i] = f[i] >> e[i];
+}
+
+__attribute__((noinline)) void
+f7 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] << b[i];
+}
+
+__attribute__((noinline)) void
+f8 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    j[i] = d[i] >> b[i];
+}
+
+__attribute__((noinline)) void
+f9 (void)
+{
+  int i;
+  for (i = 0; i < N; i++)
+    k[i] = f[i] >> b[i];
+}
+
+static void
+TEST ()
+{
+  int i;
+  for (i = 0; i < N; i++)
+    {
+      asm ("");
+      c[i] = (random () << 1) | (random () & 1);
+      b[i] = (i * 85) & (sizeof (TYPE1) * __CHAR_BIT__ - 1);
+      a[i] = c[i];
+      d[i] = (random () << 1) | (random () & 1);
+      d[i] |= (unsigned long long) c[i] << 32;
+      e[i] = (i * 85) & (sizeof (TYPE2) * __CHAR_BIT__ - 1);
+      f[i] = d[i];
+    }
+  f1 ();
+  f3 ();
+  f4 ();
+  f6 ();
+  for (i = 0; i < N; i++)
+    if (g[i] != (signed TYPE1) (a[i] << b[i])
+	|| h[i] != (unsigned TYPE1) (c[i] >> b[i])
+	|| j[i] != (signed TYPE2) (d[i] << e[i])
+	|| k[i] != (unsigned TYPE2) (f[i] >> e[i]))
+      abort ();
+  f2 ();
+  f5 ();
+  f9 ();
+  for (i = 0; i < N; i++)
+    if (g[i] != (signed TYPE1) (a[i] >> b[i])
+	|| j[i] != (signed TYPE2) (d[i] >> e[i])
+	|| k[i] != (unsigned TYPE2) (f[i] >> b[i]))
+      abort ();
+  f7 ();
+  for (i = 0; i < N; i++)
+    if (j[i] != (signed TYPE2) (d[i] << b[i]))
+      abort ();
+  f8 ();
+  for (i = 0; i < N; i++)
+    if (j[i] != (signed TYPE2) (d[i] >> b[i]))
+      abort ();
+}

gcc/testsuite/gcc.target/i386/xop-vshift-2.c

+/* { dg-do run } */
+/* { dg-options "-O3 -mxop" } */
+/* { dg-require-effective-target xop } */
+
+#define TYPE1 char
+#define TYPE2 short
+
+#include "xop-vshift-1.c"