Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
Sender:	cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To:	mailing list cygwin AT sourceware DOT cygnus DOT com
X-Authentication-Warning:	modi.xraylith.wisc.edu: khan owned process doing -bs
Date:	Fri, 23 Apr 1999 09:01:28 -0500 (CDT)
From:	Mumit Khan <khan AT xraylith DOT wisc DOT edu>
To:	Aldo Mazzilli <aldo DOT mazzilli AT inria DOT fr>
cc:	gnu-win32 AT cygnus DOT com
Subject:	Re: egcs vs gcc
In-Reply-To:	<37206EA0.7312@inria.fr>
Message-ID:	<Pine.SUN.3.93.990423085734.23125J-100000@modi.xraylith.wisc.edu>
MIME-Version:	1.0

On Fri, 23 Apr 1999, Aldo Mazzilli wrote:
> 
> I have to build 2 cross-compilers to Win32 :
> - i386-pc-cygwin32
> - i386-pc-mingw32.
> 
> And, I have to choose the compiler between EGCS and GCC.
> I just want your advice concerning the best choose I can do between EGCS
> and GCC.
> 
> My choose must depend on :
> - reexamination degree of the compiler
> - security degree of the compiler

Don't know what either of the above mean, sorry.

> - efficiency of produced code
> - size of produced code

Cygwin code is always going to be a bit less efficient that Mingw code
simply because of the emulation layer. Whether it's acceptable or not
depends on what your application does! For floating point intensive code,
Cygwin's math library provided by newlib is not that great, and you'll
get faster results (and fewer bugs) with MS runtime.

> Is there any benchmarks concerning this study.
> if not, which benchmark C suite can I use to build Benchmarks Charts.

Try your own application. That's usually always the best benchmark.

My suggestion: don't start out by choosing one over the other; try both 
and then decide.

What does this have to do with crossgcc?? I'm removing it from the list
of recipients.

Regards,
Mumit



--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com

- Raw text -