home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: netcom.com!milod
- From: milod@netcom.com (John DiCamillo)
- Subject: Re: The Good, the Bad, the Ugly, and the Wicked ...
- Message-ID: <milodDpzqCu.GE8@netcom.com>
- Organization: NETCOM On-line Communication Services (408 261-4700 guest)
- References: <31570B8E.5A12@vmark.com> <4je5rq$7qg@mimas.brunel.ac.uk> <4jes0t$gth@decaxp.HARVARD.EDU> <31630E30.5A02@oma.com> <4kbq3q$1i8@gaia.ns.utk.edu> <JSA.96Apr9131057@organon.com> <RMARTIN.96Apr10133335@rcm.oma.com> <JSA.96Apr11153135@organon.com> <milodDprBA9.FyH@netcom.com> <JSA.96Apr16205124@organon.com>
- Date: Wed, 17 Apr 1996 05:10:06 GMT
- Sender: milod@netcom15.netcom.com
-
- jsa@organon.com (Jon S Anthony) writes:
- >In article <milodDprBA9.FyH@netcom.com> milod@netcom.com (John DiCamillo) writes:
-
- >> >1. Yes, we (Matt and I) both missed your point (assuming it was the
- >> > one you are now claiming) - probably because this is not what you
- >> > said.
- >>
- >> Huh? What Robert wrote was perfectly clear both times. For reasons
- >> unknown (time pressure? the desire to argue on the net?) both you
- >> and Matt mis-interpreted Robert's post as if it had read 'any old
- >> malloc can be *used* in a deterministic manner'
-
- >Probably, because that is what he _said_. Here, (since you chop off
- >the context) take a look again:
-
- >[From R Martin...]
- >> : Malloc and new can *always* be made to be deterministic. That is
- >> : one of the major attractions to manual memory management in
- >> : real time systems. You can use malloc/free (new/delete) pairs which are
- >> : 100% predictable.
- >>
-
- >He seems to say that Malloc and new (the _standard_ manual memory
- >management stuff in typical libraries) can be made deterministic by
- >always calling new and free in pairs. You can read between the lines
- >and say "oh, of course he really didn't mean that", but I didn't. If
- >he had been clearer and stated that manual memory management based on
- >user defined allocation and deallocation operations can be made
- >deterministic, that would have been different. This is exactly what
- >he says in the later post.
-
- But you did "read between the lines" in assuming that Robert was
- referring to a "typical" library. Since this thread has been about
- memory management in *embedded* systems, the 'hosted' implementation
- definitions of the standard C library apply. Therefore, it is only
- reasonable to interpret "malloc and new" as being those provided by
- an embedded compiler vendor, who may choose to provide deterministic
- allocation.
- Oh never mind -- talk about beating a dead horse :-)
- Sorry for taking up your time!
-
- --
- ciao,
- milo
- ================================================================
- John DiCamillo Fiery the Angels Fell
- milod@netcom.com Deep thunder rode around their shores
-