lkml.org 
[lkml]   [1998]   [Sep]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: AVL and hash in memory management
Date
From
In message <19980920153215.A15492@caffeine.ix.net.nz>, Chris Wedgwood writes:
+-----
| On Sat, Sep 19, 1998 at 11:14:21PM -0400, sfrost@ns.snowman.net wrote:
| > That is defintely something else to look at, also I've seen alot of
| > shell scripts that uses '#!/bin/sh', but are really for bash and
| > won't run under the 'sh' found on other systems, would be better if
| > they did '#!/bin/bash', or that 'sh' not just be 'bash', but that's
| > all something for the distributors really...
|
| `ash' is available is much faster in many cases.
+--->8

But it won't run many of the scripts that come with Red Hat or Slackware ---
I've tried it. They both assume that /bin/sh is bash in all its g(l)ory.

(Then again, I once tried making /bin/sh ksh on an SCO box to see what would
happen and promptly discovered that many of SCO's scripts still used '^' to
build pipelines. Go figure. :-)

| Oh - they are for me to, but in total, I doubt I would spend more
| that 0.5% of my CPU time running them compared to a squid or
| somesuch.
+--->8

I spend a bit more than that on scripts that do a lot of forking ---
configure scripts, build scripts, various administration scripts that use sh
rather than assume Perl because they need to run on 9 different platforms
before our Perl has been installed, etc.

Here's a thought in that direction: how many things on Linux are Perl or
Python, etc., scripts not because they need Perl/Python/etc. features but
simply to avoid the overhead of fork()/exec() for everything they do? I
know I've done that quite often.

| For me, a 10% speed increase in network IO or disk IO compared to a
| 300% increase in running ./configure is more worth while.
+--->8

You might be surprised just how often fork() is used and how much wall time
is consumed by it, except possibly on install-and-forget database servers.
(I bet web servers get hammered by fork() latency, though.)

--
brandon s. allbery [os/2][linux][solaris][japh] allbery@kf8nh.apk.net
system administrator [WAY too many hats] allbery@ece.cmu.edu
electrical and computer engineering KF8NH
carnegie mellon university



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.269 / U:0.060 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site