[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC] Page table sharing
- To: Linus Torvalds <torvalds@transmeta.com>
- Subject: Re: [RFC] Page table sharing
- From: Rik van Riel <riel@conectiva.com.br>
- Date: Mon, 18 Feb 2002 22:53:45 -0300 (BRT)
- cc: Daniel Phillips <phillips@bonn-fries.net>,Hugh Dickins <hugh@veritas.com>, <dmccr@us.ibm.com>,Kernel Mailing List <linux-kernel@vger.kernel.org>,<linux-mm@kvack.org>, Robert Love <rml@tech9.net>,<mingo@redhat.co>, Andrew Morton <akpm@zip.com.au>,<manfred@colorfullife.com>, <wli@holomorphy.com>
- Fake-Sender: owner-linux-mm@kvack.org
- In-Reply-To: <Pine.LNX.4.33.0202181746090.24597-100000@home.transmeta.com>
- Original-Recipient: rfc822;linux-mm-archive@humbolt.geo.uu.nl
- Sender: Rik van Riel <riel@nl.linux.org>
On Mon, 18 Feb 2002, Linus Torvalds wrote:
> On Mon, 18 Feb 2002, Rik van Riel wrote:
> >
> > We'll need protection from the swapout code.
>
> Absolutely NOT.
>
> If the swapout code unshares or shares the PMD, that's a major bug.
>
> The swapout code doesn't need to know one way or the other, because the
> swapout code never actually touches the pmd itself, it just follows the
> pointers - it doesn't ever need to worry about the pmd counts at all.
The swapout code can remove a page from the page table
while another process is in the process of unsharing
the page table.
We really want to make sure that the copied-over page
table doesn't point to a page which just got swapped
out.
regards,
Rik
--
"Linux holds advantages over the single-vendor commercial OS"
-- Microsoft's "Competing with Linux" document
http://www.surriel.com/ http://distro.conectiva.com/
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/