From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mukesh Rathor Subject: Re: [PATCH 01/24] PVH xen: Add readme docs/misc/pvh-readme.txt Date: Fri, 19 Jul 2013 14:33:39 -0700 Message-ID: <20130719143339.24fb0f9c@mantra.us.oracle.com> References: <1374114788-27652-1-git-send-email-mukesh.rathor@oracle.com> <1374114788-27652-2-git-send-email-mukesh.rathor@oracle.com> <1374142143.26728.52.camel@kazak.uk.xensource.com> <20130718112128.0dff80a9@mantra.us.oracle.com> <1374225360.13645.25.camel@kazak.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1374225360.13645.25.camel@kazak.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 Campbell Cc: Xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On Fri, 19 Jul 2013 10:16:00 +0100 Ian Campbell wrote: > On Thu, 2013-07-18 at 11:21 -0700, Mukesh Rathor wrote: > > > Is there a description somewhere in the series of what PVH means > > > in terms of the guest visible ABI? i.e. documentation of the > > > delta from the regular PV mode? I had a skim through and didn't > > > spot it. > > > > No. The ABIs are not as affected, thanks to pre-existing auto > > translate mode. > > Not even a little bit of variation from that? In the current series, it's a PV domU guest with auto translate, so not really. There are some changes to the implementation, like in case of VCPUOP_initialise or XEN_DOMCTL_setvcpucontext for PVH, we must set context in VMCS also. The upcoming dom0 patch will introduce a new ABI, (unless you already did it for ARM and PVH will just piggyback on it). BTW, is there such a doc for ARM I can look at for reference? > > But, I suppose after all the patches are checked in, I can write > > up something. > > I could live with that, but I have heard mutterings that some people > are finding it hard to review the patches without knowing the > interface the are supposed to be implementing, which is pretty fair I > think. Ah, I see. I will try to enhance the patch comment prolog in the next version. Hopefully, that will help. > Could you perhaps enumerate the exact set of XENFEAT flags which > must/must not be used/supported by a PVH guest in a document > somewhere? That would hopefully explain the vast majority of the > differences between trad-PV and PVH and be pretty succinct I think. > Anything which isn't explained away by a particular feature flag > might need additional explanation. Ok, done. I put that in the pvh-readme. thanks Mukesh