Xref:	news2.mv.net comp.os.msdos.djgpp:3806
From:	Thomas Demmer <demmer AT LSTM DOT Ruhr-UNI-Bochum DOT De>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re2: More libm Bugs
Date:	Mon, 13 May 1996 19:30:33 +0100
Organization:	Lehrstuhl fuer Stroemungsmechanik
Lines:	42
Message-ID:	<31977FC9.167E@LSTM.Ruhr-UNI-Bochum.De>
References:	<31977B9C DOT 41C6 AT LSTM DOT Ruhr-UNI-Bochum DOT De>
NNTP-Posting-Host:	bvb.lstm.ruhr-uni-bochum.de
Mime-Version:	1.0
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Thomas Demmer wrote:
> 
> William Menninger told me about another libm bug,
> that I could trace down to nearly the same error as
> the ldexp() bug. The reason is, again, the wrong use of
> fscale, this time in e_exp.S
> As there are several more functions using fscale, it is likely
> that more mathmatical functions are broken with libm.
> I will check the assembler sources and (hopefully) put tomorrow a
> fixed(?) version on
> 
> ftp://ftp.lstm.ruhr-uni-bochum.de/pub/djgpp/libm.zip
> 
> BTW, Eli told me that there will be a fixed version of
> libc soon. Maybe libm should also get fixed by then...
> 

OK, I just put the file at the above URL. Maybe someone with
sources for thoroghly testing math functions for errors and
could either send them to me, or download the zip file and test
the library him/herself. Be warned: I have _NOT_ completely 
checked all changed functions, so there might be other errors
by now (sigh).



-- 
Ciao
Tom



*************************************************************
* Thomas Demmer                                             *
* Lehrstuhl fuer Stroemungsmechanik                         *
* Ruhr-Uni-Bochum                                           *
* Universitaetsstr. 150                                     *
* D-44780  Bochum                                           *
* Tel: +49 234 700 6434                                     *
* Fax: +49 234 709 4162                                     *
* http://www.lstm.ruhr-uni-bochum.de/~demmer                *
*************************************************************

- Raw text -