X-Spam-Check-By: sourceware.org
Message-ID: <013601c65209$d594cd40$b3db87d4@multiplay.co.uk>
From: "Steven Hartland" <killing AT multiplay DOT co DOT uk>
To: <cygwin AT cygwin DOT com>,
   =?iso-8859-1?Q?Ren=E9_Berber?= <r DOT berber AT computer DOT org>
References:  <048501c651b8$0862c1c0$a501a8c0 AT CAM DOT ARTIMI DOT COM> <001401c651d0$872c9090$b3db87d4 AT multiplay DOT co DOT uk> <89F9E813-1E27-4FEE-9359-6C73A8CD871A AT ptnix DOT com> <005a01c651ee$4dd35680$b3db87d4 AT multiplay DOT co DOT uk> <801E2E71-2C61-46B7-9CEE-77838077FD3E AT ptnix DOT com> <006f01c651f3$40250ab0$b3db87d4 AT multiplay DOT co DOT uk> <00b601c651f6$b8aeb5a0$b3db87d4 AT multiplay DOT co DOT uk> <e0a3io$qpt$1 AT sea DOT gmane DOT org>
Subject: Re: very poor cygwin scp performance in some situations
Date: Tue, 28 Mar 2006 02:49:15 +0100
MIME-Version: 1.0
Content-Type: text/plain; 	format=flowed; 	charset="iso-8859-1"; 	reply-type=original
Content-Transfer-Encoding: 8bit
X-Spam-Processed: multiplay.co.uk, Tue, 28 Mar 2006 02:49:22 +0100 	(not processed: message from valid local sender)
X-MDRemoteIP: 212.135.219.179
X-Return-Path: killing AT multiplay DOT co DOT uk
X-MDaemon-Deliver-To: cygwin AT cygwin DOT com
X-MDAV-Processed: multiplay.co.uk, Tue, 28 Mar 2006 02:49:23 +0100
X-IsSubscribed: yes
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

Interesting idea that I just tried setting high prio on the command line
version and it does benefit obtaining 6.0MB/s but not quite matching
the 6.2MB/s of the daemonised version.
Note: Using realtime prio drops the rate back to 4.3MB/s

Oh just tried using above normal and that hits 6.2MB/s so that looks
like its the reason but why just removing -D should produce the
same effects is beyond me.

Max rate seen is now straight /usr/sbin/ssh using above normal prio
notching up 6.4MB/s not that its much difference as +-1second
makes the difference. Interesting neither the less.

    Steve
----- Original Message ----- 
From: "René Berber"
> Also very interesting is when running sshd from the command line with
> /usr/sbin/sshd -D vs running it via cygrunsvr I get:
> * cmdline = 192MB   4.4MB/s   00:44
> * cygrunsvr = 192MB   6.4MB/s   00:30
> * cmdline without -D = 192MB   6.4MB/s   00:30
>
> I tried running:
> /usr/sbin/sshd -D 2>&1 > /dev/null but no difference from the standard
> so the difference might be the closed stdin which would support select
> being at least a contributing factor?

Changing performance options (in System Properties -> Advanced -> Performance
Options -> Advanced) to give services preference may have the same impact.

If it does, that means that the sshd process just needs a higher priority to
give better performance (sounds too obvious) at other processes expense.

Your results are interesting.



================================================
This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. 

In the event of misdirection, illegible or incomplete transmission please telephone (023) 8024 3137
or return the E.mail to postmaster AT multiplay DOT co DOT uk.


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/