Message-ID: <4AE82CE6.7090502@onevision.de>
Date: Wed, 28 Oct 2009 12:37:10 +0100
From: Roland Schwingel <roland.schwingel@onevision.de>
User-Agent: Thunderbird 2.0.0.23 (Windows/20090812)
MIME-Version: 1.0
To: cygwin@cygwin.com
Subject: Cygwin 1.5 vs 1.7: speed
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm
Sender: cygwin-owner@cygwin.com
Mail-Followup-To: cygwin@cygwin.com

Hi...

When evaluating 1.7 I do also conduct some speed tests with cygwin 1.5. 
I am using 1.5.25 und 1.7.0-62 on windows xp sp3 32bit here for my tests 
on a quad core machine with enough ram.

It seems that 1.7 is around 10-25% slower in my scenarios here. I did 
some compilation tests. Each test was performed 10 times on each cygwin 
version.
I am using a private cygwin 1.5 build of the mingw-w64 
gcc(4.4.2)/binutils(2.19) in cygwin to build on cygwin for mingw 32 and 
64bit.
The toolchain was (as already mentioned) build with cygwin 1.5 and is 
just copied over to my 1.7 installation. The toolchain built with 1.5 
appears to operate fine on 1.7.

When I compile I always compile to the local disk. When my sources are 
located on the local harddrive compilation is around 10% slower with 
cygwin 1.7 compared to 1.5 and around 25% slower if I place my sources 
on samba 3.4 share (the linux samba server was purely private for me 
during my tests). I am a little bit astonished about the speed gap when 
putting the files local vs. putting them on a network share.

Any clues? I just made an additional test compiling stock libxml2-2.6.32 
using the stock gcc3 coming with both cygwin versions and I am coming to 
roughly the same numbers. Is 1.7 truly slower than 1.5 or aren't all 
final optimizations done right now?

Roland


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