nvptx.c (nvptx_generate_vector_shuffle): Deal with complex types.

	gcc/
	* config/nvptx/nvptx.c (nvptx_generate_vector_shuffle): Deal with
	complex types.

	libgomp/
	* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c: New.
	* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c: New.

From-SVN: r230325

3 files changed, 109 insertions, 0 deletions

diff --git a/libgomp/ChangeLog b/libgomp/ChangeLog
index 4e0cddb..9ca963a 100644
--- a/libgomp/ChangeLog
+++ b/libgomp/ChangeLog
@@ -1,3 +1,8 @@
+2015-11-13  Nathan Sidwell  <nathan@codesourcery.com>
+
+	* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c: New.
+	* testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c: New.
+
 2015-11-12  James Norris  <jnorris@codesourcery.com>
 	    Joseph Myers  <joseph@codesourcery.com>
 
diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c
new file mode 100644
index 0000000..314e511
--- /dev/null
+++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-dbl.c
@@ -0,0 +1,52 @@
+
+#include <complex.h>
+
+/* Double float has 53 bits of fraction. */
+#define FRAC (1.0 / (1LL << 48))
+
+int close_enough (double _Complex a, double _Complex b)
+{
+  double _Complex diff = a - b;
+  double mag2_a = __real__(a) * __real__ (a) + __imag__ (a) * __imag__ (a);
+  double mag2_diff = (__real__(diff) * __real__ (diff)
+		     + __imag__ (diff) * __imag__ (diff));
+
+  return mag2_diff / mag2_a < (FRAC * FRAC);
+}
+
+int main (void)
+{
+#define N 100
+  double _Complex ary[N], sum, prod, tsum, tprod;
+  int ix;
+
+  sum = tsum = 0;
+  prod = tprod = 1;
+  
+  for (ix = 0; ix < N;  ix++)
+    {
+      double frac = ix * (1.0 / 1024) + 1.0;
+      
+      ary[ix] = frac + frac * 2.0i - 1.0i;
+      sum += ary[ix];
+      prod *= ary[ix];
+    }
+
+#pragma acc parallel vector_length(32) copyin(ary) copy (tsum, tprod)
+  {
+#pragma acc loop vector reduction(+:tsum) reduction (*:tprod)
+    for (ix = 0; ix < N; ix++)
+      {
+	tsum += ary[ix];
+	tprod *= ary[ix];
+      }
+  }
+
+  if (!close_enough (sum, tsum))
+    return 1;
+
+  if (!close_enough (prod, tprod))
+    return 1;
+
+  return 0;
+}
diff --git a/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c b/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c
new file mode 100644
index 0000000..b3bde65
--- /dev/null
+++ b/libgomp/testsuite/libgomp.oacc-c-c++-common/reduction-cplx-flt.c
@@ -0,0 +1,52 @@
+
+#include <complex.h>
+
+/* Single float has 23 bits of fraction. */
+#define FRAC (1.0f / (1 << 20))
+
+int close_enough (float _Complex a, float _Complex b)
+{
+  float _Complex diff = a - b;
+  float mag2_a = __real__(a) * __real__ (a) + __imag__ (a) * __imag__ (a);
+  float mag2_diff = (__real__(diff) * __real__ (diff)
+		     + __imag__ (diff) * __imag__ (diff));
+
+  return mag2_diff / mag2_a < (FRAC * FRAC);
+}
+
+int main (void)
+{
+#define N 100
+  float _Complex ary[N], sum, prod, tsum, tprod;
+  int ix;
+
+  sum = tsum = 0;
+  prod = tprod = 1;
+  
+  for (ix = 0; ix < N;  ix++)
+    {
+      float frac = ix * (1.0f / 1024) + 1.0f;
+      
+      ary[ix] = frac + frac * 2.0i - 1.0i;
+      sum += ary[ix];
+      prod *= ary[ix];
+    }
+
+#pragma acc parallel vector_length(32) copyin(ary) copy (tsum, tprod)
+  {
+#pragma acc loop vector reduction(+:tsum) reduction (*:tprod)
+    for (ix = 0; ix < N; ix++)
+      {
+	tsum += ary[ix];
+	tprod *= ary[ix];
+      }
+  }
+
+  if (!close_enough (sum, tsum))
+    return 1;
+
+  if (!close_enough (prod, tprod))
+    return 1;
+
+  return 0;
+}