Re: libstdc++/4586: math headers not up to current standards not sure if still a prob in 3.0+

Re: libstdc++/4586: math headers not up to current standards not sure if still a prob in 3.0+

[bkoz]

Synopsis: math headers not up to current standards not sure if still a prob in 3.0+

Responsible-Changed-From-To: unassigned->bkoz
Responsible-Changed-By: bkoz
Responsible-Changed-When: Thu Oct 25 22:53:56 2001
Responsible-Changed-Why:
    Mine.
State-Changed-From-To: open->closed
State-Changed-By: bkoz
State-Changed-When: Thu Oct 25 22:53:56 2001
State-Changed-Why:
    Not a bug in libstdc++-v3 math headers. 

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=4586&database=gcc

libstdc++/4586: math headers not up to current standards not sure if still a prob in 3.0+

[emt3734]

>Number:         4586
>Category:       libstdc++
>Synopsis:       math headers not up to current standards not sure if still a prob in 3.0+
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Oct 16 20:56:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     emt3734@rit.edu
>Release:        gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-85)
>Organization:
>Environment:
Redhat 7.1
>Description:
according the C++ standards, the C++ math functions should have several versions for example:

float cos(float x);
double cos(double x);
long double cos(long double x);

and obviously the compiler should choose the appropriate version according to the data type given, in my headers, there is only the double version (as per older standards).

I am not sure if this is corrected in gcc 3.0+, it is not a big deal, just noticed it when working on a calculator program.

Evan Teran
>How-To-Repeat:

>Fix:
well my headers define high precision versions of these functions (ex. cosl instead of cos).  so in the cmath header you could simply do 

inline long double cos (long double x) { return cosl (x); }

and so on for each function, and likewise for the float version, but that one is less of a big deal, as a double provides more precision than a float.
>Release-Note:
>Audit-Trail:
>Unformatted: