obtained gcc-4.6.4.tar.bz2 from upstream website;upstream

verified gcc-4.6.4.tar.bz2.sig;
imported gcc-4.6.4 source tree from verified upstream tarball.

downloading a git-generated archive based on the 'upstream' tag
should provide you with a source tree that is binary identical
to the one extracted from the above tarball.

if you have obtained the source via the command 'git clone',
however, do note that line-endings of files in your working
directory might differ from line-endings of the respective
files in the upstream repository.

1 files changed, 83 insertions, 0 deletions

diff --git a/gcc/testsuite/gcc.c-torture/execute/pr39501.c b/gcc/testsuite/gcc.c-torture/execute/pr39501.c
new file mode 100644
index 000000000..3749d6e78
--- /dev/null
+++ b/gcc/testsuite/gcc.c-torture/execute/pr39501.c
@@ -0,0 +1,83 @@
+/* { dg-options "-ffast-math" } */
+#define min1(a,b) ((a) < (b) ? (a) : (b))
+#define max1(a,b) ((a) > (b) ? (a) : (b))
+
+#define min2(a,b) ((a) <= (b) ? (a) : (b))
+#define max2(a,b) ((a) >= (b) ? (a) : (b))
+
+#define F(type,n)						\
+  type __attribute__((noinline)) type##_##n(type a, type b)	\
+  {								\
+    return n(a, b);						\
+  }
+
+F(float,min1)
+F(float,min2)
+F(float,max1)
+F(float,max2)
+
+F(double,min1)
+F(double,min2)
+F(double,max1)
+F(double,max2)
+
+int main()
+{
+  if (float_min1(0.f, -1.f) != -1.f) abort();
+  if (float_min1(-1.f, 0.f) != -1.f) abort();
+  if (float_min1(0.f, 1.f)  != 0.f)  abort();
+  if (float_min1(1.f, 0.f)  != 0.f)  abort();
+  if (float_min1(-1.f, 1.f) != -1.f) abort();
+  if (float_min1(1.f, -1.f) != -1.f) abort();
+  
+  if (float_max1(0.f, -1.f) != 0.f)  abort();
+  if (float_max1(-1.f, 0.f) != 0.f)  abort();
+  if (float_max1(0.f, 1.f)  != 1.f)  abort();
+  if (float_max1(1.f, 0.f)  != 1.f)  abort();
+  if (float_max1(-1.f, 1.f) != 1.f)  abort();
+  if (float_max1(1.f, -1.f) != 1.f)  abort();
+  
+  if (float_min2(0.f, -1.f) != -1.f) abort();
+  if (float_min2(-1.f, 0.f) != -1.f) abort();
+  if (float_min2(0.f, 1.f)  != 0.f)  abort();
+  if (float_min2(1.f, 0.f)  != 0.f)  abort();
+  if (float_min2(-1.f, 1.f) != -1.f) abort();
+  if (float_min2(1.f, -1.f) != -1.f) abort();
+  
+  if (float_max2(0.f, -1.f) != 0.f)  abort();
+  if (float_max2(-1.f, 0.f) != 0.f)  abort();
+  if (float_max2(0.f, 1.f)  != 1.f)  abort();
+  if (float_max2(1.f, 0.f)  != 1.f)  abort();
+  if (float_max2(-1.f, 1.f) != 1.f)  abort();
+  if (float_max2(1.f, -1.f) != 1.f)  abort();
+  
+  if (double_min1(0., -1.) != -1.) abort();
+  if (double_min1(-1., 0.) != -1.) abort();
+  if (double_min1(0., 1.)  != 0.)  abort();
+  if (double_min1(1., 0.)  != 0.)  abort();
+  if (double_min1(-1., 1.) != -1.) abort();
+  if (double_min1(1., -1.) != -1.) abort();
+  
+  if (double_max1(0., -1.) != 0.)  abort();
+  if (double_max1(-1., 0.) != 0.)  abort();
+  if (double_max1(0., 1.)  != 1.)  abort();
+  if (double_max1(1., 0.)  != 1.)  abort();
+  if (double_max1(-1., 1.) != 1.)  abort();
+  if (double_max1(1., -1.) != 1.)  abort();
+  
+  if (double_min2(0., -1.) != -1.) abort();
+  if (double_min2(-1., 0.) != -1.) abort();
+  if (double_min2(0., 1.)  != 0.)  abort();
+  if (double_min2(1., 0.)  != 0.)  abort();
+  if (double_min2(-1., 1.) != -1.) abort();
+  if (double_min2(1., -1.) != -1.) abort();
+  
+  if (double_max2(0., -1.) != 0.)  abort();
+  if (double_max2(-1., 0.) != 0.)  abort();
+  if (double_max2(0., 1.)  != 1.)  abort();
+  if (double_max2(1., 0.)  != 1.)  abort();
+  if (double_max2(-1., 1.) != 1.)  abort();
+  if (double_max2(1., -1.) != 1.)  abort();
+  
+  exit(0);
+}