From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH 01/11] fs: add O_BENEATH_ONLY flag to openat(2) Date: Wed, 9 Jul 2014 01:48:59 -0700 Message-ID: <20140709084859.GA4618@infradead.org> References: <1404124096-21445-1-git-send-email-drysdale@google.com> <1404124096-21445-2-git-send-email-drysdale@google.com> <20140708120331.GA30459@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: David Drysdale Cc: Christoph Hellwig , LSM List , "linux-kernel@vger.kernel.org" , Greg Kroah-Hartman , Alexander Viro , Meredydd Luff , Kees Cook , James Morris , Linux API List-Id: linux-api@vger.kernel.org On Tue, Jul 08, 2014 at 05:54:24PM +0100, David Drysdale wrote: > > How is this implemented in FreeBSD? I can't find any references to > > O_BENEATH_ONLY except for your patchset. > > FreeBSD have the relative-only behaviour for openat() relative to a > Capsicum capability dfd [1], and for a process in capability-mode [2], > but they don't have the O_BENEATH_ONLY as a separately-accessible > openat() flag. However, it seemed like a more widely useful idea so > separating it out was suggested. In that case we should make sure to use the same name and semantics for it. As far as I'm concerned I'd prefer a less clumsy name like O_BENEATH.