X-pop3-spooler: POP3MAIL 2.1.0 b 4 980420 -bs- Message-ID: <19980913210931.22376@cerebro.laendle> Date: Sun, 13 Sep 1998 21:09:31 +0200 From: Marc Lehmann To: jduan AT boxer DOT chem DOT indiana DOT edu Cc: beastium Subject: Re: pgcc 1.0.3a: unexplainable -m[arch] flag results Mail-Followup-To: jduan AT boxer DOT chem DOT indiana DOT edu, beastium References: <35FAF8BC DOT 6A54D0EA AT worldnet DOT att DOT net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <35FAF8BC.6A54D0EA@worldnet.att.net>; from Jack Duan on Sat, Sep 12, 1998 at 05:42:04AM -0500 X-Operating-System: Linux version 2.1.120 (root AT cerebro) (gcc version pgcc-2.91.57 19980901 (egcs-1.1 release)) Status: RO Content-Length: 1545 Lines: 35 On Sat, Sep 12, 1998 at 05:42:04AM -0500, Jack Duan wrote: > > flag: option + -fomit-frame-pointer > > compress uncompress > -------- ---------- > bzip2 v.9b (-O2) 144.390u 27.700u > -O2 -m486 142.750u > 27.690u < > -O6 149.030u 29.830u > -O6 -mpentium 150.710u 29.860u > -O6 -mpentiumpro 136.570u 28.490u > -O6 -m486 139.720u 29.450u > -O6 -mamdk6 > 124.360u < 28.540u > > ------------------------ END ------------------------------------ > > So, what makes AMDK6 optimization faster than Pentium/PPro running on my > PII? Well, I don't know ;) Probably a bug somewhere ;) I will include the amdk6 support from egcs soon, maybe that improves it further. I'll investigate this. > Also, noticing that '-O2 -m486' is even faster than '-O6 -mpentium'?!! Thats easy to explain: -mpentium is poison for a ppro/p-ii, even worse than -m386 or -m486. -----==- | ----==-- _ | ---==---(_)__ __ ____ __ Marc Lehmann +-- --==---/ / _ \/ // /\ \/ / pcg AT goof DOT com |e| -=====/_/_//_/\_,_/ /_/\_\ --+ The choice of a GNU generation | |