Sender: tutu AT duboi DOT com
Message-ID: <37282C1C.F6758A83@duboi.com>
Date: Thu, 29 Apr 1999 11:53:32 +0200
From: Olivier Tubach <tutu AT duboi DOT com>
X-Mailer: Mozilla 4.51 [en] (X11; I; IRIX 6.3 IP32)
X-Accept-Language: fr-FR, en
MIME-Version: 1.0
To: pgcc AT delorie DOT com
Subject: Re: pgcc... Do I really got it ?
References: <19990428201537 DOT A27287 AT win DOT tue DOT nl> <Pine DOT LNX DOT 4 DOT 10 DOT 9904281936460 DOT 16107-100000 AT billabong DOT demon DOT co DOT uk> <19990428223259 DOT B28723 AT win DOT tue DOT nl> <19990429014944 DOT I18899 AT cerebro DOT laendle>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: pgcc AT delorie DOT com

> Benchmark it against a known non-optimized version ;-> (sorry, I couldn't
> resist).

I would have not resisted either !

> Judging from the original mail, I think the binary was indeed pentium
> optimized. Possible sources of errors include:
>
> - not _really_ using pgcc to compile the binary.
> - pgcc didn't optimize at all
> - wrong testing methodology
>
> However, its very very rare that a pgcc-optimized cpu-intensive program
> shows _exactly_ the same execution time as a gcc-optimized.

I've redone the tests, with the time command,
when vanilla gcc -O3  runs in 54 secs,
pgcc -O6 ..[killer options]... runs in 53.5 secs.
So you were right !

Olivier Tubach