home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.next.misc
- Path: sparky!uunet!munnari.oz.au!metro!mama!andy
- From: andy@research.canon.oz.au (Andy Newman)
- Subject: Re: ksh
- Message-ID: <BznAJD.6uC@research.canon.oz.au>
- Sender: news@research.canon.oz.au
- Organization: Canon Information Systems Research Australia
- References: <FISCHER.92Dec12191250@solsort.iesd.auc.dk> <Bz9voD.AF@research.canon.oz.au> <1gomlsINNmot@darkstar.UCSC.EDU>
- Date: Tue, 22 Dec 1992 05:14:00 GMT
- Lines: 26
-
- isbell@cats.ucsc.edu (Art Isbell) writes:
- >
- >Have you possibly neglected to strip bash?
- >
- >259[~]% ls -l /usr/local/bin/bash /bin/sh
- >-rwxr-xr-x 2 root 122880 Jul 21 06:16 /bin/sh*
- >-rwxr-xr-x 1 root 221184 Oct 5 11:28 /usr/local/bin/bash*
- >260[~]% bash -version
- >bash -version
- >GNU bash, version 1.12.1
-
- I'm talking about incore memory use. An ls is not a good indicator of
- code/data size anyway, there's lots of junk in Mach-O executable
- files. Try size instead...
-
- $ size /bin/sh /usr/local/bin/bash
- __TEXT __DATA __OBJC others dec hex
- 114688 24576 0 8192 147456 24000 /bin/sh
- 245760 16384 0 8192 270336 42000 /usr/local/bin/bash
- $
-
- (This method has it own problem as it is measuring static code size and not
- the working set of the acutal executable.)
-
- --
- Andy Newman (andy@research.canon.oz.au)
-