Date: Wed, 29 Oct 2003 00:06:57 -0800 From: Marcel Moolenaar <marcel@xcllnt.net> To: "David O'Brien" <obrien@FreeBSD.org> Cc: ia64@FreeBSD.org Subject: Re: 20031027 snapshot on ftp2.freebsd.org Message-ID: <20031029080657.GA49037@dhcp01.pn.xcllnt.net> In-Reply-To: <20031029075209.GE56958@dragon.nuxi.com> References: <20031028212041.GA46637@dhcp01.pn.xcllnt.net> <20031029073130.GB56958@dragon.nuxi.com> <20031029073808.GA48975@dhcp01.pn.xcllnt.net> <20031029075209.GE56958@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Oct 28, 2003 at 11:52:09PM -0800, David O'Brien wrote: > On Tue, Oct 28, 2003 at 11:38:08PM -0800, Marcel Moolenaar wrote: > > On Tue, Oct 28, 2003 at 11:31:30PM -0800, David O'Brien wrote: > > > On Tue, Oct 28, 2003 at 01:20:41PM -0800, Marcel Moolenaar wrote: > > > > The snapshot took 7.5 hours to build on an idle HP rx2600 with an > > > > UP kernel without debugging options. > > > > > > Is that using NOPORTS & NODOCS or a complete snapshot build including > > > building ISO's? > > > > A complete snapshot including ISO images. > > I wish 'make release' was an established benchark Intel used. Only cross-releases would be meaningful in that case. The ia64 is designed with the assumption that the compiler does most of the ILP related work. The compiler simply has more to do on ia64, so compilations are almost certainly slower than for i386 or amd64. worldstone comparisons between different architectures are mostly meaningless for that reason. Not to mention that a -O compile for ia64 simply doesn't produce good enough code to make the processor really sweat. > I just did the same on my 1P AMD Opteron box -- 4 hrs. Sweet! -- Marcel Moolenaar USPA: A-39004 marcel@xcllnt.net
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20031029080657.GA49037>