From mboxrd@z Thu Jan 1 00:00:00 1970 From: Malcolm Crossley Subject: Re: VT-d interrup remapping errata workaround Date: Mon, 8 Jul 2013 13:56:53 +0100 Message-ID: <51DAB715.5070007@citrix.com> References: <51DACCD702000078000E342C@nat28.tlf.novell.com> <51DAB47E.3070507@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <51DAB47E.3070507@citrix.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: xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On 08/07/13 13:45, Malcolm Crossley wrote: > On 08/07/13 13:29, Jan Beulich wrote: >> All, >> >> just having spotted the backport of Linux commit 03bbcb2e I notice >> a certain discrepancy with the Xen commit having the same purpose >> as well as with the actual specification updates: >> >> The Linux solution keys off of device IDs 3403 and 3406, as listed in >> the specification update, but this way fails to cover the X58 chipset, >> which has - under different numbers (62 and 69) - the same errata >> (the Linux commit message also only mentions erratum 53 for the >> 55x0 chipsets, albeit I believe 47 is as much of an issue there as it is >> for Xen). >> >> The Xen solution keys off of ID 342e, with no explanation in the >> commit description on where this association comes from. >> >> I therefore wonder whether >> - the Xen solution may have false positives >> - the Linux solution would still have false negatives even after >> adding ID 3405 >> >> Jan >> > Hi Jan, > > I have used the VT-d control PCI device which is common across > X58/5520/5500 > > See page 233 > http://www.intel.co.uk/content/dam/doc/datasheet/x58-express-chipset-datasheet.pdf Whoops got the page number wrong it's page 223 > and > > page 255 > http://www.intel.co.uk/content/dam/doc/datasheet/5520-5500-chipset-ioh-datasheet.pdf > > It seems the Linux workaround only applies to the 5520 and 5500 chipsets. > > The pci ID database suggests 0x342e will detect 7500 chipset (which > does not have the interrupt remapping errata) but that chipset never > reports revision 0x13 so the errata workaround will never be applied. > Also, if you check the devhide registers on page 318 of the 5520/5500 datasheet you can see the Device 0 (DMI port which the Linux uses for the PCI ID match) can be hidden from the PCI scan but device 20 function 0 can't be hidden (what Xen use's). You may find on a Dual IOH system that the DMI port for one of the IOH's is disabled. > Malcolm > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel