[Bug ada/11391] Invalid return values from math libraries

["bdavis9659 at comcast dot net" <gcc-bugzilla at gcc dot gnu dot org>]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11391



------- Additional Comments From bdavis9659 at comcast dot net  2003-09-14 01:10 -------
maybe i am missing something, but it looks correct with
gcc 3.4

see below. included is a c program to calculate the sine.

(i686,gnu,linux)
-----------------------------------------------------------
 
[bdavis@localhost ~]$ gnatmake -v test_322
 
GNATMAKE 3.4 20030913 (experimental) Copyright 1995-2002 Free Software
Foundation, Inc.
  "test_322.ali" being checked ...
  -> "test_322.adb" time stamp mismatch
gcc -c test_322.adb
End of compilation
gnatbind -x test_322.ali
gnatlink test_322.ali
[bdavis@localhost ~]$ ./test_322
 
GNAT 3.2.x (Not ACT GNAT 5.0) problem 1
   Incorrect math library operations
Sin( 5.23000000000000E-01) =>  4.99481355518642E-01
[bdavis@localhost ~]$ cat sine_test.c
#include <stdio.h>
#include <math.h>
 
int main(void)
{
   printf("   %f \n", sin(0.523));
   return 0;
 
}
[bdavis@localhost ~]$ cc -o sine_test sine_test.c -lm
[bdavis@localhost ~]$ ./sine_test
   0.499481