From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Campbell Subject: Re: [Minios-devel] [PATCH v4 0/] Begin to disentangle libxenctrl and provide some stable libraries Date: Thu, 3 Dec 2015 09:44:36 +0000 Message-ID: <1449135876.4424.115.camel@citrix.com> References: <1445440941.9563.163.camel@citrix.com> <1445442435.9563.184.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Stefano Stabellini Cc: Ian Jackson , Roger Pau Monne , Wei Liu , xen-devel List-Id: xen-devel@lists.xenproject.org On Thu, 2015-11-19 at 16:20 +0000, Stefano Stabellini wrote: (Backing up this thread a bit). > > It seems like much of this would be candidates for adding to > > libxendevicemodel, but the underlying unstable interfaces pose a problem > > there. I'm going to leave this for another day. > > Does the Xen<->libxc interface need to be stable for libxendevicemodel > to have a stable ABI? > I think that libxendevicemodel could be stable, no matter what hypercall > is eventually called underneath. Am I missing something? I remember this morning why this matters, it's not the user space library ABI etc, but rather that if we want libxendevicemodel to be able to lock its fd to a specific domain (which we eventually do) then the kernel driver needs to be able to introspect the privcmd hypercalls[0] so the underlying hypercall does really need to be stable (since we don't want to be constantly reving the kernel driver when something changes). Given a stable hypercall interface underneath given the wrapping library a stable ABI is a no brainer. Ian. [0] or to provide a dedicate ioctl or something, either way it is the kernel which needs to be providing the lockdown functionality.