Date: Sat, 15 Feb 2003 13:10:50 -0600 From: David Booth <wbooth@austin.rr.com> To: java@FreeBSD.ORG Subject: Re: Math.pow bug for jdk1.3.1-p8 ? Message-ID: <200302151310.51070.wbooth@austin.rr.com> In-Reply-To: <20030215174554.GA97285@moo.holy.cow> References: <3E4C9DDD.4040204@gddsn.org.cn> <20030216.005020.894433699.shudo@localhost> <20030215174554.GA97285@moo.holy.cow>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday 15 February 2003 11:45 am, parv wrote: > in message <20030216.005020.894433699.shudo@localhost>, wrote > shudo@computer.org thusly... > > > From: Greg Lewis <glewis@eyesbeyond.com> > > > > > On Sat, Feb 15, 2003 at 05:54:21PM +0900, shudo@computer.org > > > wrote: > > ... > > > > > It is certain that there is a problem around fdlibm, FreeBSD's > > > > gcc 2.95.4 or how the fdlibm compiled (compiler flags?). > > > > > > Sounds like its quite possibly due to problems with > > > optimisation. Obviously, java_g uses no optimisation when > > > compiling, > > ... > > > Next thing we can do to track the cause down is to compile > > e_pow.c, w_sqrt.c, e_sqrt.c, s_fabs.c, s_scalbn.c and s_copysign.c > > with different compiler options (-O0, -O2 and so on) and see the > > result of the ieee754_pow() function. > > I have only -O option set. I would love to test all the variations, > but my computer is too slow for me to wait the two-three iterations > to test. > A few of more data points: 1.3.1p7 built on 5.0 release gcc3.2.1 -O2 TestPow works (i=512); 1.3.1p8 built on 5.0 release gcc3.2.1 -O2 TestPow fails (i=17); 1.3.1p7 built on 4.7 stable gcc3.2.2 -O2 TestPow works (i=512); 1.3.1p8 built on 4.7 stable gcc3.2.1 -O2 TestPow fails (i=17); 1.3.1p8 built on 4.7 stable gcc3.2.2 -O2 TestPow fials (i=17); > > - Parv To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-java" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200302151310.51070.wbooth>