Date: Thu, 28 Apr 2005 06:16:39 -0400 From: David Schultz <das@FreeBSD.ORG> To: Stefan Farfeleder <stefanf@FreeBSD.ORG> Cc: cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/lib/msun/src s_ceill.c s_floorl.c s_truncl.c Message-ID: <20050428101639.GA15653@VARK.MIT.EDU> In-Reply-To: <20050428090356.GB67414@wombat.fafoe.narf.at> References: <200504220830.j3M8UXSC030662@repoman.freebsd.org> <20050422084616.GB87979@wombat.fafoe.narf.at> <20050423132715.GA59057@VARK.MIT.EDU> <20050428090356.GB67414@wombat.fafoe.narf.at>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Apr 28, 2005, Stefan Farfeleder wrote: > On Sat, Apr 23, 2005 at 09:27:15AM -0400, David Schultz wrote: > > On Fri, Apr 22, 2005, Stefan Farfeleder wrote: > > > > > Please ignore my last message, I'll just use double addition to generate > > > the exception. > > > > Couldn't that result in a spurious underflow when logn double x gets > > converted to double precision? Consider x = 0x1p-2000L. > > Indeed, that does happen. Should I back out the last change and leave > it broken on sparc64 until someone fixes the quad emulation? Yeah, I think so. There's no sense in having a hack to do things right in one case when the real problem is that things are broken for all the other cases, including addition.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050428101639.GA15653>