Path: utzoo!attcan!uunet!cbmvax!jesup
From: jesup@cbmvax.UUCP (Randell Jesup)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Binary loading times
Message-ID: <5734@cbmvax.UUCP>
Date: 18 Jan 89 23:24:39 GMT
References: <15043@cisunx.UUCP>
Reply-To: jesup@cbmvax.UUCP (Randell Jesup)
Organization: Commodore Technology, West Chester, PA
Lines: 25

In article <15043@cisunx.UUCP> ejkst@unix.cis.pittsburgh.edu (Eric J. Kennedy) writes:
>
>I noticed some behavior that I thought rather odd today.  I have
>AmigaSpice and Superbase Professional on my hard disk.  Spice takes up
>445K on the disk and SbPro takes up 354K.  When I execute these
>programs, however, spice loads in just over 3 seconds, while sbpro
>takes over 13 seconds to load!  When spice loads, the hard disk light
>stays on steadily, and I can hear the heads moving
>"tick...tick...tick...".  When sbpro loads, the drive does its usual
>flickering and seeking all over creation.

	Yup.  There are 114 (if memory holds) hunks in sbpro.  This slows
loads down to a crawl.  FFS (and LoadSeg) are much more efficient when they
have nice, big hunks to play with.  And you're right, if they are very large
there may be a problem loading them on a small-memory, fragmented system.
You can, however, specify to blink to merge hunks, but only up to some
upper size bound.  This gets you a small number of hunks, but not gigantic ones.

>Can anyone verify or oppose these guesses?  How can I tell if a file is
>fragmented or how many hunks are in an executable?

	Run dumpobj on it. (or was that objdump?)

-- 
Randell Jesup, Commodore Engineering {uunet|rutgers|allegra}!cbmvax!jesup