From:	Gunnar DOT Isaksson AT saab DOT se (Gunnar Isaksson)
Subject:	Bugs in tan(x) and atan(x)
20 Jan 1997 02:28:07 -0800 :
Approved:	cygnus DOT gnu-win32 AT cygnus DOT com
Distribution:	cygnus
Message-ID:	<32E32961.60BC.cygnus.gnu-win32@saab.se>
Mime-Version:	1.0
X-Mailer:	Mozilla 3.01Gold (X11; I; HP-UX A.09.05 9000/735)
Original-To:	gnu-win32 AT cygnus DOT com
Original-Sender:	owner-gnu-win32 AT cygnus DOT com

Dear Sirs,

A few months ago I reported some bugs in
the mathlibrary concerning the tan(x) and atan(x)
functions and a workaround by using tanf(x) and
atanf(x).

I am sorry to report that this workaround does
not work in the 17.1 beta release of the Cygnus
GCC implementation.

Now both tan(x) and atanf(x) contains exact the
same error. atan(x) and atanf(x) has also the same
error. The errors can't be seen for small x-values
and is noticed only for large x-values.

I hope this information helps you in catching
bugs in GCC. My impression is that your version
of GCC is very stable and works great.

Best regards,

     Gunnar Isaksson 
-- 
=============================================    
    Gunnar Isaksson   Hamradio: SM5IUF
    Internet: Gunnar DOT Isaksson AT saab DOT se
_____________________________________________
-
For help on using this list, send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".

- Raw text -