Date: Wed, 25 Feb 2004 13:50:35 -0500 From: Kenneth Culver <culverk@sweetdreamsracing.biz> To: freebsd-amd64@FreeBSD.org, David O'Brien <obrien@FreeBSD.org> Cc: "'freebsd-amd64@FreeBSD.org'" <freebsd-amd64@FreeBSD.org> Subject: Re: Performance comparison, ULE vs 4BSD and AMD64 vs i386 Message-ID: <20040225135035.v66800cwkgw08wwc@www.sweetdreamsracing.biz> In-Reply-To: <20040225183234.GG7567@dragon.nuxi.com> References: <1077658664.92943.15.camel@.rochester.rr.com> <20040225110754.hcogcccokg84k44k@www.sweetdreamsracing.biz> <20040225183234.GG7567@dragon.nuxi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Quoting David O'Brien <obrien@FreeBSD.org>: > On Wed, Feb 25, 2004 at 11:07:54AM -0500, Kenneth Culver wrote: >> The buildworld problem could just be because it takes longer for the >> compiler to >> generate amd64 code. In fact, I'm almost willing to bet that's the >> case since >> amd64 has 2x the GPR's that x86 does. It's likely that it's harder >> to optimize >> for it. Does anyone who knows compilers care to comment? > > s/harder/slower/ > > It is defineately easier to optimize for amd64 because if its increased # > of registers. But I'm not sure even "slower" is a valid claim -- on the > i386 the compiler has to do a lot of time figuring out the best way to do > the spill code (when, where). > > -- > -- David (obrien@FreeBSD.org) Ok well maybe not with amd64, but I thought that when you add registers it can be harder/slower/whatever to optimize for since there are many more ways to do so. Like on PowerPC or Alpha or whatever there are a LOT more GPR's than there are on even Athlon64... I guess only having 2x the GPR's doesn't make a whole lot of difference... but it still wouldn't surprise me if the gcc people didn't have all the kinks worked out yet. Ken
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040225135035.v66800cwkgw08wwc>