This is the mail archive of the
cygwin@cygwin.com
mailing list for the Cygwin project.
Re: Printf routine incorrectly prints NaN for -Inf
- From: Christopher Faylor <cgf at redhat dot com>
- To: cygwin at cygwin dot com
- Date: Mon, 11 Nov 2002 22:32:23 -0500
- Subject: Re: Printf routine incorrectly prints NaN for -Inf
- References: <000301c289f9$428771f0$a82ace84@research.pc.ece.mcgill.ca>
- Reply-to: cygwin at cygwin dot com
On Mon, Nov 11, 2002 at 10:11:57PM -0500, Peter Kabal wrote:
>gcc version 3.2 20020818 (prerelease)
>
>printf ("%g", log(0));
>
>This prints "NaN" instead of "-Inf" as it should (and did in the past).
>A check of the bit pattern returned shows that that log(0) returns a bit
>pattern corresponding to "-Inf", but printf prints "NaN" instead.
>
>I am not sure which C library is being used (a bug report to gcc folks
>got the response that newlib is the culprit). However, the compiler
>option -print-libgcc-file-name tells me it is looking for libgcc.a. So I
>am confused as to who to report the bug to.
Perhaps unsurprisingly, the gcc folks are correct. It's a newlib bug.
The newlib mailing list is newlib at sources dot redhat dot com .
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/