From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jan Beulich" Subject: RE: iommu=0 leading to panic when system defaults to using x2apic Date: Tue, 14 Dec 2010 08:21:51 +0000 Message-ID: <4D07372F0200007800027BBF@vpn.id2.novell.com> References: <4D0215E0020000780002734A@vpn.id2.novell.com> <4D0224E602000078000273AD@vpn.id2.novell.com> <749B9D3DBF0F054390025D9EAFF47F22317AA2F4@shsmsx501.ccr.corp.intel.com> <4D0257D7020000780002747B@vpn.id2.novell.com> <987664A83D2D224EAE907B061CE93D530193BB9EB8@orsmsx505.amr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <987664A83D2D224EAE907B061CE93D530193BB9EB8@orsmsx505.amr.corp.intel.com> Content-Disposition: inline List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Allen M Kay , Yang Z Zhang Cc: Keir Fraser , "xen-devel@lists.xensource.com" , Weidong Han List-Id: xen-devel@lists.xenproject.org >>> On 10.12.10 at 19:26, "Kay, Allen M" wrote: > The architectural requirement is actually between interrupt remapping = and=20 > x2apic. Since interrupt remapping is part of the VT-d feature so = current=20 > software requires all VT-d features enabled in order for x2apic to be = enabled. >=20 > Strictly speaking DMA remapping is not required for x2apic. However, = queued=20 > invalidation is required since interrupt remapping requires queued=20 > invalidation. So x2apic dependency is as follows: >=20 > x2apic->interrupt remapping->queued invalidation >=20 > Due to historical reasons, the new VT-d features were built on top of = the old=20 > ones as they become available. Is there a requirement to separate this = out? =20 > If so, we will need to re-design iommu boot parameter which took a while = to=20 > get it right so most systems can now boot successfully. And why is it that Xen panics when the BIOS pre-enabled x2apic mode without also enabling interrupt re-mapping, while Linux (afaict from looking at 2.6.32 code and output from an affected machine) simply enables interrupt re-mapping in this case? Thanks, Jan