From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH] IOMMU: don't disable bus mastering on faults for devices used by Xen or Dom0 Date: Tue, 6 Nov 2012 15:24:22 +0100 Message-ID: <1352211862.2155.13.camel@Solace> References: <5097FD2902000078000A66BF@nat28.tlf.novell.com> <5098E19D02000078000A697A@nat28.tlf.novell.com> <20121106094456.GA45690@ocelot.phlegethon.org> <1352203585.505.5.camel@Solace> <509915CD02000078000A6ABF@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============6881749480603285661==" Return-path: In-Reply-To: <509915CD02000078000A6ABF@nat28.tlf.novell.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: Jan Beulich Cc: Keir Fraser , wei.huang2@amd.com, Tim Deegan , xen-devel , weiwang.dd@gmail.com, xiantao.zhang@intel.com List-Id: xen-devel@lists.xenproject.org --===============6881749480603285661== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-GrSN0qQMMqtJRlrXr+kS" --=-GrSN0qQMMqtJRlrXr+kS Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, 2012-11-06 at 12:51 +0000, Jan Beulich wrote: > > Right. But moving the fault handling code to softirq should have alread= y > > helped solving/mitigating that, hasn't it? >=20 > It helps keeping Xen alive, but doesn't for any specific domain > (including Dom0). >=20 Ok. Yes, now I remember... That was the main purpose of that work. Thanks and sorry. :-P > > When implementing and testing that, I wasn't able to reproduce any > > livelock situation (although I can't exclude that to be at least partly > > due to my inexperience, especially at the time, with I/O > > virtualization)... Jan, have you (after killing the 'disable > > bus-mastering part' of course)? >=20 > No, I haven't - you'd have to have a device that doesn't stop I/O > after a finite amount was done (or program one that way, e.g. by > handing it a cyclic list of SG descriptors or alike). Wasn't it at your > (Citrix) end that the problem was actually observed/reported? >=20 Yes but, as Tim said, I was never able to get in touch with the original faulting behavior/piece of hardware. Trying to simulate that as you're suggesting is basically what I did, but, at that time, I was using a passed-to-a-guest device, and wasn't bypassing (for anyone) the ~BUS_MASTERING thing... That's why I asked whether you observed a different behaviour. Anyway, I see that we're talking about different issues and understand (now) what you're trying to solve and why that is needed, so, again, sorry for dragging the discussion a bit out of scope. :-) Thanks and Regards, Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://retis.sssup.it/people/faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-GrSN0qQMMqtJRlrXr+kS Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEABECAAYFAlCZHZYACgkQk4XaBE3IOsQBUgCfUYT23GPWRULGUGoRsxMbuMNW OGIAn2GSo5gEpmbvUAyGHa/ffMKC+2jB =vMqn -----END PGP SIGNATURE----- --=-GrSN0qQMMqtJRlrXr+kS-- --===============6881749480603285661== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============6881749480603285661==--