Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 6 Nov 2000 14:38:07 +0200 (IST)
From:      Roman Shterenzon <roman@harmonic.co.il>
To:        Randell Jesup <rjesup@wgate.com>
Cc:        freebsd-stable@FreeBSD.ORG
Subject:   Re: "Malloc type lacks magic" show-stopper solved
Message-ID:  <Pine.LNX.4.10.10011061432050.24920-100000@shark.harmonic.co.il>
In-Reply-To: <ybubsvyfhze.fsf@jesup.eng.tvol.net.jesup.eng.tvol.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 2 Nov 2000, Randell Jesup wrote:

> Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at> writes:
> >David O'Brien wrote:
> >> Forgot to mention.  I'm considering removing the ability to do -O2 and
> >> -O3 from the system C compiler again.... people are just proving over
> >> and over how much they want to shoot their foot off.
> >
> >Please don't. We are already seeing enough problems due to differences
> >between FSF GCC and FreeBSD's patched version. Please let's reduce those
> >differences, not introduce anything further.
> 
>         Removing -O2 and -O3 from userland would be a Very Bad Thing, and
> would cause us considerable consternation.  Proper use of -O2, -O3, and
> various -f's and -m's can get us 10's of % on some heavily compute-bound
> code.  This translates to considerable differences in $$$ in hardware.

Do you have any specific examples? 

I, myself tried to compile some applications using -O vs. -O2 and even
having -march=pentium , -march=k6 (I've K6-2) and tried to benchmark the
resulting binaries.
The results were quite similar, and the differences could have been
blaimed on the 2nd level cache.
I haven't found any "considerable differences", so I stick with -O .

This wasn't true for 2.7.2.3 I think, since -O2 there produced faster
binary than -O (I tested it on 3.4-RELEASE once).

I don't remember -O3 producing ever faster code for me anyway. It just
produced larger binary, probably filling delayed branch slots or whatever
it is in Pentium :) (Don't comment on this one, I'm just kidding).

--Roman Shterenzon, UNIX System Administrator and Consultant
[ Xpert UNIX Systems Ltd., Herzlia, Israel. Tel: +972-9-9522361 ]



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.10.10011061432050.24920-100000>