From owner-freebsd-current  Wed Sep 30 12:55:59 1998
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id MAA09020
          for freebsd-current-outgoing; Wed, 30 Sep 1998 12:55:59 -0700 (PDT)
          (envelope-from owner-freebsd-current@FreeBSD.ORG)
Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA09011
          for <current@FreeBSD.ORG>; Wed, 30 Sep 1998 12:55:56 -0700 (PDT)
          (envelope-from julian@whistle.com)
Received: (from daemon@localhost)
	by alpo.whistle.com (8.8.5/8.8.5) id MAA12663;
	Wed, 30 Sep 1998 12:47:06 -0700 (PDT)
Received: from current1.whistle.com(207.76.205.22)
 via SMTP by alpo.whistle.com, id smtpdr12652; Wed Sep 30 19:46:59 1998
Date: Wed, 30 Sep 1998 12:46:52 -0700 (PDT)
From: Julian Elischer <julian@whistle.com>
To: David Greenman <dg@root.com>
cc: Terry Lambert <tlambert@primenet.com>, current@FreeBSD.ORG
Subject: Re: VM mmap file extension bug still exists? 
In-Reply-To: <199809300920.CAA04277@implode.root.com>
Message-ID: <Pine.BSF.3.95.980930123944.13393B-100000@current1.whistle.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

well, yes, I considered doing that.
but couldn't convince myself that there was no case where the user is
mapping only PART of a file, and that 'size' might represent the
mapped region size (in pages) where "size_in_bytes" is the filesize.

In many cases these are the same, but I wasn't confident enough in my
understanding of this code to state publically that one value could always
be derived from the other. 
In fact I managed to0 convince myself that it was likely that this was not
always the case, and took the safe route.

julian

On Wed, 30 Sep 1998, David Greenman wrote:

> >> It's definitely a workaround.  The real fix is to add a parementer to
> >> the mapping function.
> >
> >Ok here's the patch done the 'correct way'
> >I have looked at the problem with terry and he has convinced me that
> >there IS a problem there.
> 
>    Well, as I said in previous email, I'm not convinced that adding another
> argument to the pager_alloc functions is "correct", and I think your patches
> pretty well demonstrate that. It appears to me that in every case, the
> "size_in_bytes" is just the non-rounded/indexed version of the original
> "size" parameter, so why don't we simply change the definition of "size" to
> be an off_t number of bytes and then do the OFF_TO_IDX translation in the
> pager(s) as necessary, instead of at the caller?
> 
> -DG
> 
> David Greenman
> Co-founder/Principal Architect, The FreeBSD Project
> 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message