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, 173 insertions, 0 deletions

diff --git a/gcc/testsuite/c-c++-common/dfp/convert-int-fold.c b/gcc/testsuite/c-c++-common/dfp/convert-int-fold.c
new file mode 100644
index 000000000..038559d41
--- /dev/null
+++ b/gcc/testsuite/c-c++-common/dfp/convert-int-fold.c
@@ -0,0 +1,173 @@
+/* { dg-options "-O2" } */
+
+/* N1150 5.1 Conversion between decimal floating integer.
+   C99 6.3.1.4(1a) New.
+   These should all be folded at compile time.  */
+
+#include "dfp-dbg.h"
+
+#ifdef __cplusplus
+#define BOOL bool
+#else
+#define BOOL _Bool
+#endif
+
+extern void link_error (void);
+
+int
+main ()
+{
+  _Decimal32 d32;
+  _Decimal64 d64;
+  _Decimal128 d128;
+  unsigned int ui;
+  unsigned long ul;
+  unsigned long long ull;
+  int si;
+  long sl;
+  long long sll;
+  BOOL b;
+
+  /* C99 Section 6.7.2 Type specifiers.  Type _Bool is 
+     mentioned in this section.  Conversions between 
+     BOOL and DFP types.  */
+
+  /* Decimal float to unsigned integer.  */
+  d32 = 456.789df;
+  d64 = 23.456789dd;
+  d128 = 1234.5678dl;
+
+  ui = d32;
+  if (ui != 456U)
+    link_error ();
+  ul = d32;
+  if (ul != 456UL)
+    link_error ();
+  ull = d32;
+  if (ull != 456ULL)
+    link_error ();
+
+  ui = d64;
+  if (ui != 23U)
+    link_error ();
+  ul = d64;
+  if (ul != 23UL)
+    link_error ();
+  ull = d64;
+  if (ull != 23ULL)
+    link_error ();
+
+  ui = d128;
+  if (ui != 1234U)
+    link_error ();
+  ul = d128;
+  if (ul != 1234UL)
+    link_error ();
+  ull = d128;
+  if (ull != 1234ULL)
+    link_error ();
+
+  /* Decimal float to signed integer.  */
+
+  /* Decimal float to BOOL.  */
+  d32 = 1.23df;
+  d64 = -3.4dd;
+  d128 = 0.00003dl;
+
+  b = d32;
+  if (!b)
+    link_error ();
+  b = d64;
+  if (!b)
+    link_error ();
+  b = d128;
+  if (!b)
+    link_error ();
+
+  /* Unsigned integer to decimal float.  */
+  ui = 987U;
+  ul = 345678UL;
+  ull = 1234567ULL;
+
+  d32 = ui;
+  if (d32 != 987.0df)
+    link_error ();
+  d32 = ul;
+  if (d32 != 345678.0df)
+    link_error ();
+  d32 = ull;
+  if (d32 != 1234567.df)
+    link_error ();
+
+  d64 = ui;
+  if (d64 != 987.0dd)
+    link_error ();
+  d64 = ul;
+  if (d64 != 345678.0dd)
+    link_error ();
+  d64 = ull;
+  if (d64 != 1234567.dd)
+    link_error ();
+
+  d128 = ui;
+  if (d128 != 987.0dl)
+    link_error ();
+  d128 = ul;
+  if (d128 != 345678.0dl)
+    link_error ();
+  d128 = ull;
+  if (d128 != 1234567.dl)
+    link_error ();
+
+  /* Signed integer to decimal float.  */
+  si = -987;
+  sl = -345678;
+  sll = -1234567;
+
+  d32 = si;
+  if (d32 != -987.0df)
+    link_error ();
+  d32 = sl;
+  if (d32 != -345678.0df)
+    link_error ();
+  d32 = sll;
+  if (d32 != -1234567.df)
+    link_error ();
+
+  d64 = si;
+  if (d64 != -987.0dd)
+    link_error ();
+  d64 = sl;
+  if (d64 != -345678.0dd)
+    link_error ();
+  d64 = sll;
+  if (d64 != -1234567.dd)
+    link_error ();
+
+  d128 = si;
+  if (d128 != -987.0dl)
+    link_error ();
+  d128 = sl;
+  if (d128 != -345678.0dl)
+    link_error ();
+  d128 = sll;
+  if (d128 != -1234567.dl)
+    link_error ();
+
+  /* BOOL to decimal float.  */
+  d32 = 0.0DF;
+  d64 = 0.0DD;
+  d128 = 0.0DL;
+  
+  b = d32;
+  if (b)
+    link_error ();
+  b = d64;
+  if (b)
+    link_error ();
+  b = d128;
+  if (b)
+    link_error ();
+
+  return 0;
+}