Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264628AbUFTAUh (ORCPT ); Sat, 19 Jun 2004 20:20:37 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264777AbUFTAUh (ORCPT ); Sat, 19 Jun 2004 20:20:37 -0400 Received: from bay-bridge.veritas.com ([143.127.3.10]:28454 "EHLO MTVMIME03.enterprise.veritas.com") by vger.kernel.org with ESMTP id S264628AbUFTAUg (ORCPT ); Sat, 19 Jun 2004 20:20:36 -0400 Date: Sun, 20 Jun 2004 01:20:25 +0100 (BST) From: Hugh Dickins X-X-Sender: hugh@localhost.localdomain To: Stas Sergeev cc: Andrew Morton , Christoph Rohland , Subject: Re: [patch][rfc] expandable anonymous shared mappings In-Reply-To: <40D4023E.8010500@aknet.ru> Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1035 Lines: 25 On Sat, 19 Jun 2004, Stas Sergeev wrote: > Hugh Dickins wrote: > > shared anonymous only becomes interesting when you fork. > I disagree with this. The way I am using it may look horrible, > but yes, I do use it without the fork(). Then I think you have no reason to use MAP_SHARED: use MAP_PRIVATE and you should get the behaviour you require, without kernel change. Shared anonymous is peculiar: although mapping is anonymous (nothing shared with unrelated mms), modifications are shared between parent and children. It's half-way between anonymous and file-backed. We agree that it might be nice to let the object used to support that be extended if mremap extends the mapping. But it might instead just be needless feature creep. Sorry, your case does not persuade me yet. Hugh - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/