Path: csiph.com!x330-a1.tempe.blueboxinc.net!feeder1.hal-mli.net!border3.nntp.dca.giganews.com!Xl.tags.giganews.com!border1.nntp.dca.giganews.com!nntp.giganews.com!local2.nntp.dca.giganews.com!nntp.earthlink.com!news.earthlink.com.POSTED!not-for-mail
NNTP-Posting-Date: Thu, 12 May 2011 07:01:33 -0500
Message-ID: <4DCBCC1F.41A5@mindspring.com>
Date: Thu, 12 May 2011 08:01:35 -0400
From: pete <pfiland@mindspring.com>
Reply-To: pfiland@mindspring.com
Organization: PF
X-Mailer: Mozilla 3.04Gold (WinNT; I)
MIME-Version: 1.0
Newsgroups: comp.lang.c
Subject: Re: Would someone tell me if I'm on the wrong track?
References: <70cf2865-5c4a-47a5-a8e8-ac369c2b91b3@k22g2000yqh.googlegroups.com> <4dca2cf0$0$18239$4fafbaef@reader2.news.tin.it> <35d5a07f-b868-4f6e-83fd-e333baba79fc@gu8g2000vbb.googlegroups.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 52
X-Usenet-Provider: http://www.giganews.com
NNTP-Posting-Host: 4.154.223.89
X-Trace: sv3-5oOBJTOk0MIiL1IJBEAaa1MRLKK86o/rRZ6S1YM0BrCxA0qmeKtIP/tqmF4zUZ65HNQ1rEzcHB4z0M8!1U7YUm56CPlBGr+cdI2zG6RljAjrWQuGDqrDFDRagE+InUTIp2EAc5S6Or96bU85ijoMgcWFII/C!NCDV0g05Cw==
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
X-Postfilter: 1.3.40
X-Original-Bytes: 2921
Xref: x330-a1.tempe.blueboxinc.net comp.lang.c:3798

luser- -droog wrote:
> 
> On May 11, 1:31 am, "io_x" <a...@b.c.invalid> wrote:
> > "luser- -droog" <mijo...@yahoo.com> ha scritto nel messaggionews:70cf2865-5c4a-47a5-a8e8-ac369c2b91b3@k22g2000yqh.googlegroups.com...
> >
> >
> >
> > > I'm trying to build a new memory allocation scheme for
> > > my postscript interpreter to satisfy these requirements:
> > > % pointer + offset + size + short tag <= 32bits
> > > % 'garbage-collection'-ready (structure the heap for easy iteration)
> > > % all memory is in one region for easy restore from disk
> >
> > > So I've written a prototype where the "pointer" is an index
> > > into an extendable address table stored in the heap itself.
> > > And it seems to work! But I don't trust it, you know?
> >
> > > I'd appreciate any comments the experts here could offer.
> >
> [...]
> > > /* grow memory by sz */
> > > void growmem(mfile *mem, unsigned sz) {
> > >    unsigned char *tmp;
> > >    if (sz < pgsz) sz = pgsz;
> > >    else sz = (sz/pgsz + 1) * pgsz;
> >
> > what about if this mult above overflow?
> >
> > >    sz += mem->max;
> >
> > what about if this sz += overflow?
> > i not understand well the remain (too much complex
> > for my little think)
> > but i think, it is better to see if sz overflow
> > and in input to write "if((int) sz <0)  return  -1;"
> > so that each function can fail.
> > never exist one function can not fail
> 
> unsigned sz;
> (int) sz < 0
> 
> Does this just check the high bit (assuming 2's complement)?

I don't know, 
but I don't see the relevance of ((int) sz < 0).

If you want to check for wrap around after (sz += mem->max), then 

    if (sz - mem->max > sz) ThenItWrapped;

-- 
pete