From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH/RFC 1/2] anon-inodes: Remove fd_install() from anon_inode_getfd() Date: Wed, 27 Feb 2008 21:20:05 +0200 Message-ID: <47C5B7E5.6030708@qumranet.com> References: <20080225191043.GA32342@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: Christoph Hellwig , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Davide Libenzi , kvm-devel@lists.sourceforge.net, akpm@linux-foundation.org To: Roland Dreier Return-path: Received: from bzq-179-150-194.static.bezeqint.net ([212.179.150.194]:52303 "EHLO il.qumranet.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756201AbYB0TWK (ORCPT ); Wed, 27 Feb 2008 14:22:10 -0500 In-Reply-To: Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Roland Dreier wrote: > The anonymous inodes interface anon_inode_getfd() calls fd_install() > for the newly created fd, which does not work for some use cases where > the caller must do futher initialization before exposing the file to > userspace. This is also probably not the safest interface, since the > caller must be sure that it is OK if userspace closes the fd before > anon_inode_getfd() even returns. > > Therefore, change the anonymous inodes interface so that the caller is > responsible for calling fd_install(), and change the name of the > function from anon_inode_getfd() to anon_inode_allocfd() so that any > code using the old interface breaks at compilation rather than failing > in a strange way. Fix up all the in-kernel users to use the new > interface. > > The kvm changes are Acked-by: Avi Kivity -- Any sufficiently difficult bug is indistinguishable from a feature.