Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT sourceware DOT cygnus DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help: <mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com
Date: Mon, 29 Nov 1999 14:47:39 +0200
From: Paul Sokolovsky <paul-ml AT is DOT lg DOT ua>
X-Mailer: The Bat! (v1.32) S/N 34D3AC61
Reply-To: Paul Sokolovsky <paul-ml AT is DOT lg DOT ua>
X-Priority: 3 (Normal)
Message-ID: <10616.991129@is.lg.ua>
To: Chris Faylor <cygwin AT sourceware DOT cygnus DOT com>
Subject: Re[2]: cygwin on 95 slower than NT
In-reply-To: <19991126122322.A2084@cygnus.com>
References: <19991126122322 DOT A2084 AT cygnus DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Hello Chris,

Chris Faylor <cgf AT cygnus DOT com> wrote:

>>>>Iam using cygwin shell on NT and 95.  On windows 95 the shell scripts
>>>>are considerably slow.  Like it take 1-2 seconds for each command.  Is
>>>>this natural?.  Is there any parameter to be adjusted in the DOS Shell.
>>>>Iam a shell maniac I badly need a fast shell on Win95.  Could you
>>>>help?.

>>It's known issue of Cygwin (and other POSIX layers, e.g.  UWIN).  They
>>all by some reason (probably because they themselves were developed on
>>NT, without enough attention to other Win32 systems) count Win9x as
>>'degraded mode'.

CF> Oh yeah.  That was it.  If only we'd paid more attention to Windows 95,
CF> Cygwin would be much faster.  I knew that we should have used the
CF> "GoFasterOnWin9x (TRUE);' function.

    Joke, guys, joke. I can laugh you even more: I was so amused by
assurance that sane POSIX implementation cannot be done on Win95 that
take making proof of that as my thesis (i.e. I stated that I would
implement such thing and it will be as bad as already existing).
Consider my condition when I had to announce on the defend that I
failed achieving objectives of my thesis! For some unknown reason
stupid thing didn't want to work badly - it did screen output quite
fast, process files fast also and didn't corrupt them trying to cut
\r\n to \n or vice-versa. But don't hold breath, story has happy end:
I was granted my Master degree.

CF> If anyone thinks they can optimize things so that console I/O works
CF> better on Windows 95, I'll be thrilled to consider a patch.

    Back from humor, if you consider only "optimization patches",
probably nothing can be done - I believe that there's really nothing
unneeded in cygwin, as comprehensive POSIX implementation.

    But take an other perspective: how many programs require general
POSIX terminal interface? My estimate that no more than 20% At least
fileutils, textutils, shellutils, binutils - most commonly used
packages doesn't use it. Make lightweight write() path for them -
directly to WriteFile() and then see the difference.

CF> -chris


Best regards,
 Paul                            mailto:paul-ml AT is DOT lg DOT ua



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