[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


From: Ralf Wildenhues
Subject: Re: _AS_DETECT_BETTER_SHELL speedup
Date: Sat, 29 Jan 2005 14:58:48 +0100
User-agent: Mutt/1.4.1i

* Alexandre Duret-Lutz wrote on Sat, Jan 29, 2005 at 02:07:53PM CET:
> It seems to me that replacing
>    (foo) 2>/dev/null
> by
>    (exec foo) 2>/dev/null
> would also replace two forks by one, and it looks a lot less
> risky.  
> However my understanding is that modern shells already detect
> these cases and perform the optimization by themselves.  I know
> for sure that Zsh has the machinery for this (along with some
> other hairy tricks to limit forks at other places), and, from
> what I can observe from the command line, so does Bash.  

Yes, after testing around a little, that is also my observation.
I should have done that before (grumble).  The only noticeable
difference I found was when foo was a shell builtin (old shells
and new shells alike).

Sorry, I did not mean to generate so much noise,

reply via email to

[Prev in Thread] Current Thread [Next in Thread]