Re: [patch 2/3] MAP_NOZERO - implement sys_brk2()

From: Davide Libenzi <davidel@xmailserver.org>
To: Rik van Riel <riel@redhat.com>
Cc: Ulrich Drepper <drepper@gmail.com>,
	Hugh Dickins <hugh@veritas.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [patch 2/3] MAP_NOZERO - implement sys_brk2()
Date: Wed, 27 Jun 2007 12:22:52 -0700 (PDT)	[thread overview]
Message-ID: <Pine.LNX.4.64.0706271218280.5219@alien.or.mcafeemobile.com> (raw)
In-Reply-To: <4682B3B0.2040500@redhat.com>

On Wed, 27 Jun 2007, Rik van Riel wrote:

> Ulrich Drepper wrote:
> > On 6/27/07, Hugh Dickins <hugh@veritas.com> wrote:
> > > Not so: if an mmap can be done by extending either adjacent vma (prot
> > > and flags and file and offset all match up), that's what's done and no
> > > separate vma is created.  (And adjacent vmas get merged when mprotect
> > > removes the difference in protection.)
> > 
> > mmap return values are randomized.  If they would be mergable
> > something would be wrong.
> 
> You can easily pass a hint address to mmap(), causing it to
> extend a malloc arena instead of creating a new VMA.

You mean something like the F_DUPFD, where you pass an fd and that serves 
as from-there-up hint?

- Davide