[Bug math/5680] math functions (atanh, log, log1p) are inconsistent in treatment of errno

[Bug math/5680] math functions (atanh, log, log1p) are inconsistent in treatment of errno

[jsm28 at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=5680

Joseph Myers <jsm28 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

--- Comment #1 from Joseph Myers <jsm28 at gcc dot gnu.org> 2012-02-29 20:12:25 UTC ---
atanh was fixed by:

2009-04-26  Ulrich Drepper  <drepper@redhat.com>

        * sysdeps/ieee754/k_standard.c (__kernel_standard): Pole errors in
        atanh should set ERANGE.

All functions also raise divide-by-zero.  So the only bug left here is for
log1p, which should set errno to ERANGE.  That's one particular case of the
problems for log1p detailed in bug 6792.

*** This bug has been marked as a duplicate of bug 6792 ***

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/5680] math functions (atanh, log, log1p) are inconsistent in treatment of errno

[fweimer at redhat dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=5680

Florian Weimer <fweimer at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
              Flags|                            |security-

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug math/5680] New: math functions (atanh, log, log1p) are inconsistent in treatment of errno

[dickinsm at gmail dot com]

In version 2.5 of glibc, on SuSE Linux 10.2, it appears that the libm functions atanh, log and log1p (and 
possibly others) are inconsistent in their handling of singularities:

log(0.0) returns -Infinity and sets errno = ERANGE
log1p(-1.0) returns -Infinity and doesn't set errno
atanh(-1.0) returns -Infinity and sets errno = EDOM.

According to the C99 standard (appendix F), and IEEE-754r, all of these operations should behave in 
the same way---they should all return -Infinity and raise the "divide-by-zero" exception. This makes 
sense mathematically:  all three function calls correspond to singularities of log type.  Admittedly, the 
C99 standard has little to say about errno, but I don't understand why these three calls behave 
differently with respect to setting of errno.

-- 
           Summary: math functions (atanh, log, log1p) are inconsistent in
                    treatment of errno
           Product: glibc
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: math
        AssignedTo: aj at suse dot de
        ReportedBy: dickinsm at gmail dot com
                CC: glibc-bugs at sources dot redhat dot com
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://sourceware.org/bugzilla/show_bug.cgi?id=5680

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/5680] math functions (atanh, log, log1p) are inconsistent in treatment of errno

[mtk dot manpages at gmail dot com]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mtk dot manpages at gmail
                   |                            |dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=5680

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.