Date: Wed, 25 Feb 2004 10:58:16 -0800 From: "David O'Brien" <obrien@freebsd.org> To: Brooks Davis <brooks@one-eyed-alien.net> Cc: freebsd-amd64@freebsd.org Subject: Re: Performance comparison, ULE vs 4BSD and AMD64 vs i386 Message-ID: <20040225185816.GI7567@dragon.nuxi.com> In-Reply-To: <20040225184512.GA8620@Odin.AC.HMC.Edu> References: <1077658664.92943.15.camel@.rochester.rr.com> <20040225110754.hcogcccokg84k44k@www.sweetdreamsracing.biz> <20040225183234.GG7567@dragon.nuxi.com> <20040225184512.GA8620@Odin.AC.HMC.Edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Feb 25, 2004 at 10:45:14AM -0800, Brooks Davis wrote: > installed base of i386 machines. Even if generating amd64 code is > easier then generating i386 code, it's probably still a bit early to > expect the compiler to do it quickly. Oh, sorry, minor mis-read. No one for any platform spends any time trying to make GCC compile faster -- notice the compile speeds of GCC 2.7, 2.95, 3.0, 3.1, 3.2, 3.3[*]... each version *compiles* slower than the previous version. It is the quality and level of optimized code *produced* that any one spends major time on. No real effort has been spent making the i386 code generator run fast. [*] finally for GCC 3.4, compile time speed is being looked at to try to fix some of the really badly implimented algorthums used in GCC for its running (not generated code). But this is an architecturally neutral effort. -- -- David (obrien@FreeBSD.org)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040225185816.GI7567>