Date: Wed, 14 Oct 1998 11:58:28 +0300 (IDT) From: Eli Zaretskii X-Sender: eliz AT is To: Kbwms AT aol DOT com cc: rudd AT cyberoptics DOT com, djgpp-workers AT delorie DOT com Subject: Re: libc math function upgrade work In-Reply-To: <376d8718.36237de5@aol.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: djgpp-workers AT delorie DOT com On Tue, 13 Oct 1998 Kbwms AT aol DOT com wrote: > Rudd's analysis was correct: > > >> I would guess that these times are dominated by something other than > >> function computations, since they seem so similar. > > My claim that some of his functions are twice as fast as those in libm.a > has yet to be verified by credible evidence. Building the tests with -pg and looking at the cumulative time of the relevant math functions is the way to go, I think. In my view, it is important to know how much speedup does one gain by using this version as opposed to both libm and the current libc versions. (I would expect to see a significant difference, at least relative to libm.) Otherwise, the merits of getting current versions replaced are not clear enough. IIRC, one of Eric's original goals for doing all this work was to gain speed, so this aspect seems to be important to Eric as well.