From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH 2/2][RFC] KVM: Emulate MSI-X table and PBA in kernel Date: Thu, 30 Dec 2010 13:07:29 +0200 Message-ID: <20101230110728.GC6441@redhat.com> References: <1293007495-32325-1-git-send-email-sheng@linux.intel.com> <201012291655.19535.sheng@linux.intel.com> <20101229092824.GA2876@redhat.com> <201012301532.42341.sheng@linux.intel.com> <20101230074748.GB7889@redhat.com> <4D1C5124.2090409@redhat.com> <20101230103256.GB6441@redhat.com> <4D1C60DE.4020800@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Sheng Yang , Marcelo Tosatti , kvm@vger.kernel.org, Alex Williamson To: Avi Kivity Return-path: Received: from mx1.redhat.com ([209.132.183.28]:30436 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753514Ab0L3LIa (ORCPT ); Thu, 30 Dec 2010 06:08:30 -0500 Content-Disposition: inline In-Reply-To: <4D1C60DE.4020800@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On Thu, Dec 30, 2010 at 12:37:18PM +0200, Avi Kivity wrote: > On 12/30/2010 12:32 PM, Michael S. Tsirkin wrote: > >On Thu, Dec 30, 2010 at 11:30:12AM +0200, Avi Kivity wrote: > >> On 12/30/2010 09:47 AM, Michael S. Tsirkin wrote: > >> >I am not really suggesting this. What I say is PBA is unimplemen= ted > >> >let us not commit to an interface yet. > >> > >> What happens to a guest that tries to use PBA? > >> It's a mandatory part of MSI-X, no? > > > >Yes. Unfortunately the pending bit is in fact a communication channe= l > >used for function specific purposes when mask bit is set, > >and 0 when unset. The spec even seems to *require* this use: > > > >I refer to this: > > > > For MSI and MSI-X, while a vector is masked, the function is prohib= ited > > from sending the associated message, and the function must set the > > associated Pending bit whenever the function would otherwise send t= he > > message. When software unmasks a vector whose associated Pending bi= t is > > set, the function must schedule sending the associated message, and > > clear the Pending bit as soon as the message has been sent. Note th= at > > clearing the MSI-X Function Mask bit may result in many messages ne= eding > > to be sent. > > > > > > If a masked vector has its Pending bit set, and the associated > > underlying interrupt events are somehow satisfied (usually by softw= are > > though the exact manner is function-specific), the function must cl= ear > > the Pending bit, to avoid sending a spurious interrupt message late= r > > when software unmasks the vector. However, if a subsequent interrup= t > > event occurs while the vector is still masked, the function must ag= ain > > set the Pending bit. > > > > > > Software is permitted to mask one or more vectors indefinitely, and > > service their associated interrupt events strictly based on polling > > their Pending bits. A function must set and clear its Pending bits = as > > necessary to support this =E2=80=9Cpure polling=E2=80=9D mode of op= eration. > > > >For assigned devices, supporting this would require > >that the mask bits on the device are set if the mask bit in > >guest is set (otherwise pending bits are disabled). >=20 > Can't this be done by setting the real mask bit when the guest reads > the virtual pending bit, then reading the real pending bit? =46unction specific is function-specific, but most likely not, by that time the pending bit in the device might be clear: 'clear the Pending bit as soon as the message has been sent' > >Existing code does not support PBA in assigned devices, so at least = it's > >not a regression there, and the virtio spec says nothing about this = so > >we should be fine. >=20 > Why isn't it subject to the pci spec? >=20 > If an interrupt condition exits, the bit should be set. I wish. But this is not what the spec says above. It says if vector is unmasked, bit must be cleared. > --=20 > error compiling committee.c: too many arguments to function