Henri Verbeet : msvcrt: Don't depend on the system's implementation of acos () & asin().

[Alexandre Julliard]

Module: wine
Branch: master
Commit: 71aa14af993dd065b1882675fb9f01af44ed232a
URL:    http://source.winehq.org/git/wine.git/?a=commit;h=71aa14af993dd065b1882675fb...

Author: Henri Verbeet <hverbeet(a)gmail.com>
Date:   Fri Aug 29 02:11:53 2008 +0200

msvcrt: Don't depend on the system's implementation of acos() & asin().

---

 dlls/msvcrt/math.c |    9 +++++++--
 1 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/dlls/msvcrt/math.c b/dlls/msvcrt/math.c
index b449af5..25ac69f 100644
--- a/dlls/msvcrt/math.c
+++ b/dlls/msvcrt/math.c
@@ -57,7 +57,12 @@ static MSVCRT_matherr_func MSVCRT_default_matherr_func = NULL;
 double CDECL MSVCRT_acos( double x )
 {
   if (x < -1.0 || x > 1.0 || !finite(x)) *MSVCRT__errno() = MSVCRT_EDOM;
-  return acos(x);
+  /* glibc implements acos() as the FPU equivalent of atan2(sqrt(1 - x ^ 2), x).
+   * asin() uses a similar construction. This is bad because as x gets nearer to
+   * 1 the error in the expression "1 - x^2" can get relatively large due to
+   * cancellation. The sqrt() makes things worse. A safer way to calculate
+   * acos() is to use atan2(sqrt((1 - x) * (1 + x)), x). */
+  return atan2(sqrt((1 - x) * (1 + x)), x);
 }
 
 /*********************************************************************
@@ -66,7 +71,7 @@ double CDECL MSVCRT_acos( double x )
 double CDECL MSVCRT_asin( double x )
 {
   if (x < -1.0 || x > 1.0 || !finite(x)) *MSVCRT__errno() = MSVCRT_EDOM;
-  return asin(x);
+  return atan2(x, sqrt((1 - x) * (1 + x)));
 }
 
 /*********************************************************************