From:	mdruiter AT cs DOT vu DOT nl (Ruiter de M)
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: DJGPP vs. mingwin32 I/O performances
Date:	20 May 1998 12:49:32 GMT
Organization:	Fac. Wiskunde & Informatica, VU, Amsterdam
Lines:	34
Message-ID:	<6jujgs$bhe$1@star.cs.vu.nl>
References:	<3561B91D DOT 702A AT epfl DOT ch>
NNTP-Posting-Host:	galjas.cs.vu.nl
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Lionel Portmann (Lionel DOT Portmann AT epfl DOT ch) wrote:
> I wish to use gcc on a PC in order to do some I/O intensive work
> (parsing big ASCII files). The program will only produce displays in a
> console.

I bet GNU tools already exist for your specific task.

> I tried both ports of gcc v2.8.1 for PC, DJGPP and MINGWIN32.
> I compiled the very same source code with both environnement,
> and found that the mingwin32 executable was a least 1.5 times slower 
> than djgpp one. I read in a FAQ that djgpp has to switch between
> protected mode and real mode every time it has to do I/O access,
> so I expected mingwin32 to be faster than djgpp. 

I think DJGPP doesn't suffer very much from the switch, because most
I/O is buffered. That means big blocks are read/written at once,
causing just one switch per block. I don't know mingwin32, but I guess
it buffers I/O too. Maybe less efficiently? :-)

> I have to add that at compilation time, mingwin32 gave me more warnings
> than djgpp, which might indicate that the settings are not equivalent.

Have you tried the `-Wall' switch to gcc? Or maybe even `-W', which
warns about everything :) . Try `info gcc invoking warn' on the
command line.

> On the other hand, -O3 switch was ON in both versions. 

`-On' is used by most (all?) C-compilers. It is a very common switch.
-- 
Groeten, Michel.        http://www.cs.vu.nl/~mdruiter
  ____________
  \  /====\  /          "You know, Beavis, you need things that suck,
   \/      \/           to have things that are cool", Butt-Head.

- Raw text -