Return to BSD News archive
Newsgroups: comp.os.386bsd.development Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!caen!usenet.cis.ufl.edu!usenet.eel.ufl.edu!news.mathworks.com!europa.eng.gtefsd.com!howland.reston.ans.net!pipex!uunet!news.cygnus.com!jtc From: jtc@cygnus.com (J.T. Conklin) Subject: Re: slow math library In-Reply-To: graichen@omega.physik.fu-berlin.de's message of Mon, 21 Nov 1994 12:49:13 GMT Message-ID: <JTC.94Nov21103825@rtl.cygnus.com> Sender: news@cygnus.com Nntp-Posting-Host: rtl.cygnus.com Organization: Cygnus Support References: <EYMUB1UR@math.fu-berlin.de> Date: Mon, 21 Nov 1994 18:38:25 GMT Lines: 27 >>>>> "Thomas" == Thomas Graichen <graichen@omega.physik.fu-berlin.de> writes: Thomas> in the last day's i tried freebsd (1.1) ... but i've found one Thomas> problem for me - the math-library (floating-point-operations) Thomas> seem to be very slow (factor 2-4 slower like under linux on Thomas> the same machine) - i've also tried the msun lib (from 1.1.5) Thomas> - but i can't see the big improvement - but floating point Thomas> speed is important for me because i do mostly numerical stuff Thomas> on that machine - can anybody tell me if there is a faster Thomas> math-lib for freebsd - or if not - can somebody who knows Thomas> something about this try to improve the mathlib I maintain the SunPro fdlibm derived math library for NetBSD and DJGPP. I understand that the FreeBSD folks have integrated a copy I gave to Jordan into their OS as "msun". I'm surprised that you have found it to be so much slower than the Linux libm. Before someone can start addressing the differences, it would be useful to know what (math) operations your programs use. One thing that you might try is to recompile the library with -D_IEEE_LIBM and without -D_MULTI_LIBM and -D_POSIX_MODE. I don't really think it can make that much of a difference, since it only effects error handling. --jtc