Message 73

Re: Scan problem

X-Added: With Flames (bblib $Revision: 1.4 $)
Return-path: <deepay@cs.cmu.edu>
X-Andrew-Authenticated-as: 0;andrew.cmu.edu;Network-Mail
Received: from po0.andrew.cmu.edu via trymail for bb+academic.cs.15-721@andrew.cmu.edu
          ID </afs/andrew.cmu.edu/usr0/bb/Mailbox/kvkUDAS00UdY48UU5r>;
          Mon,  8 Oct 2001 15:51:40 -0400 (EDT)
Received: from mx1.andrew.cmu.edu (MX1.ANDREW.CMU.EDU [128.2.10.111])
        by po0.andrew.cmu.edu (8.9.3/8.9.3) with ESMTP id PAA02174
        for <bb+academic.cs.15-721@ams.andrew.cmu.edu>; Mon, 8 Oct 2001 15:51:39 -0400 (EDT)
Received: from snapdragon.cald.cs.cmu.edu (SNAPDRAGON.CALD.CS.CMU.EDU [128.2.191.164])
        by mx1.andrew.cmu.edu (8.12.0.Beta16/8.12.0.Beta16) with SMTP id f98JpceD015067;
        Mon, 8 Oct 2001 15:51:38 -0400
Received: from snapdragon.cald.cs.cmu.edu by snapdragon.cald.cs.cmu.edu
          id aa11806; 8 Oct 2001 15:51 EDT
Date: Mon, 8 Oct 2001 15:51:01 -0400 (EDT)
From: Deepayan Chakrabarti <deepay@cs.cmu.edu>
X-Sender: deepay@snapdragon.cald.cs.cmu.edu
To: Neal Burns <nburns@andrew.cmu.edu>
cc: Spiros Papadimitriou <spapadim+@cs.cmu.edu>,
        post+academic.cs.15-721@andrew.cmu.edu
Subject: Re: Scan problem
In-Reply-To: <Pine.LNX.3.96L.1011008152514.11546A-100000@surreal>
Message-ID: <Pine.LNX.3.95L.1011008154531.11784A-100000@snapdragon.cald.cs.cmu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

I unpinned the _firstDirPageId page, and also the first data-record page.
I confirmed that this removes all the pinned buffers. But apparently the
test files have lots of cases where it checks that after openScan(), at
least one page should be pinned. I am not doing this, and I can think of
no workaround...

Come on, Spiros! Be a nice guy, and earn the friendly appreciation of all
of us! Remove that test in line 181 in heap_driver.C Its really a pain...

 -deepay



 On Mon, 8 Oct 2001, Neal Burns wrote:

> This is a real problem. I tried unpinning the first page in the file from
> my Scan2 class's constructor, and this didn't help. Did you have any luck,
> Deepayan?
> 
> Neal
> 
> On Mon, 8 Oct 2001, Deepayan Chakrabarti wrote:
> 
> > now it compiles, but there's another problem. MyScan::MyScan has to call
> > Scan::Scan, which ends up pinning 2 pages. MyScan doesnt unpin these 2
> > pages because of which we have 2 pinned pages at the end of the Scan. So
> > heap_driver.C complains..
> >   I am assuming the 2 pinned pages are: the first directory page, and the
> > page containing the first actual record. So in MyScan::MyScan, I'll try
> > unpinning these two... I hope that works.
> > 
> >  -deepay
> > 
> > 
> >  On Sun, 7 Oct 2001, Spiros Papadimitriou wrote:
> > 
> > > Two minor updates to the code were made early this afternoon, which you
> > > may already have noticed:
> > > 
> > > - The Scan methods have been turned virtual.  On your part, you should
> > >   update the declarations in your scan implementation (ie. add "virtual" to
> > >   the destructor and the other two public methods).
> > > 
> > > - The pin test (right after openScan() returns) has been removed.  If your
> > >   code passed before, it will still pass (no test was changed or added), so
> > >   it does not affect you.
> > >  
> > > Spiros
> > > 
> > > 
> > > 
> > > 
> > 
> > 
> 
> 
>