Re: fork/exec performance?
Re: fork/exec performance?
- Subject: Re: fork/exec performance?
- From: Ragnar Sundblad <email@hidden>
- Date: Thu, 03 Oct 2002 00:31:01 +0200
On Mon, 23 Sep 2002 18:06:53 -0400 Jim Magee <email@hidden> wrote:
vfork() yields 100X (yes, 100 times, not percent more) the performance
of fork() in most situations on Mac OS X. So, if you can live with
vfork() requirements, please do so. Getting common tools to use it
where appropriate appears to be the trick in your case. I don't think
Linux has vfork(), so people tend to stick with just fork() in portable
code.
Is there any chance of improving fork+exec performance in the OS
since, as you say, many programs currently use fork?
(Just for fun I did a quick and dirty change to bash to make it
use vfork+_exit, which included moving the fork up one stack
frame and some other small changes, to see if I could
improve automake, (gnu-)libtool and friends' performance
(they currently run really slow on mac os x, it is sad when the
compile time is just a fraction of the time spent to build
things.) The program seems to work for simple interactive use,
but crashes immediately on shellscripts, I haven't checked what
the problem is yet. Anyway, as far as I can tell, with my very
limited knowledge on both vfork world limitations and the bash
code, it could possibly be pretty hairy to (correctly) change
for example bash to use vfork instead of fork, it would be nice
if there was a smaller performance hit.
If anyone more knowledgeable would take on the bash-vfork task
instead I for one would be most grateful. Maybe if I filed a
radar... :-)
/ragge
_______________________________________________
darwin-kernel mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/darwin-kernel
Do not post admin requests to the list. They will be ignored.