>obscure fork() implementation and ignores native one. I hoped using native
would improve performance of the shell scripts where fork() is the main
bottleneck. To my surprise the native implementation isn’t faster than CygWin
What tools have you used to measure the native fork()?
–
Maxim Shatskih, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of Maxim S. Shatskih[SMTP:xxxxx@storagecraft.com]
Reply To: Windows System Software Devs Interest List
Sent: Friday, December 22, 2006 1:21 PM
To: Windows System Software Devs Interest List
Subject: Re:[ntdev] Subsystems concept in Windows>obscure fork() implementation and ignores native one. I hoped using native
>would improve performance of the shell scripts where fork() is the main
>bottleneck. To my surprise the native implementation isn’t faster than CygWinWhat tools have you used to measure the native fork()?
Honestly, I don’t remember. It is several years and my memory is worse because of constant overflow 
I don’t even remember if I really measured it or only read the explanation from CygWin people who tried to improve performance this way and found there was no real advantage. Can’t find it now, everything I found were your complains about it 
What I remember to measure was normal process creation and the surprise how it is slow even if the image section is already mapped to memory by different process (i.e. no disk access necessary). It was in tens ms which may not seem too much until one notices shell scripts and makefiles can easily cause hundreds forks and process creation overhead is then in seconds.
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]