From mboxrd@z Thu Jan 1 00:00:00 1970 From: George Dunlap Subject: Re: [PATCH v2 1/2] libxl: Introduce functions to add and remove host USB devices to an HVM guest Date: Wed, 20 Mar 2013 17:04:41 +0000 Message-ID: <5149EC29.6070401@eu.citrix.com> References: <1363797697-27179-1-git-send-email-george.dunlap@eu.citrix.com> <20809.59721.561353.891856@mariner.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20809.59721.561353.891856@mariner.uk.xensource.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Ian Jackson Cc: Roger Pau Monne , Ian Campbell , "xen-devel@lists.xen.org" List-Id: xen-devel@lists.xenproject.org On 20/03/13 16:52, Ian Jackson wrote: > George Dunlap writes ("[PATCH v2 1/2] libxl: Introduce functions to add and remove host USB devices to an HVM guest"): >> Devices must be removed by "id", an identifying string, which must be >> specified when the device is created. The caller can either pass one >> in to request; if none is passed in, then libxl will choose one and >> pass it back to the caller. In the second case, it will return a >> string containing the id used in the "id" field of the device struct. > IMO we shouldn't expose these ids like this. Rather, we should do > what we do with block, network, etc. and have functions for listing > and removing devices by their virtual names ("device tree path" or > whatever as the guest sees it). > > If QMP is deficient then this may mean keeping a separate list > somewhere. Right -- I was hoping to have such an implementation as a "nice-to-have", which I would try to get in as a best-effort, rather than a requirement. -George