From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:60355) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gJjQa-0007ob-65 for qemu-devel@nongnu.org; Mon, 05 Nov 2018 13:16:46 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gJjQS-0002vS-KY for qemu-devel@nongnu.org; Mon, 05 Nov 2018 13:16:42 -0500 Received: from mx1.redhat.com ([209.132.183.28]:43874) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gJjQG-0002dO-0E for qemu-devel@nongnu.org; Mon, 05 Nov 2018 13:16:27 -0500 Date: Mon, 5 Nov 2018 13:16:08 -0500 From: "Michael S. Tsirkin" Message-ID: <20181105181353.39804-10-mst@redhat.com> References: <20181105181353.39804-1-mst@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20181105181353.39804-1-mst@redhat.com> Subject: [Qemu-devel] [PULL 09/33] intel_iommu: handle invalid ce for shadow sync List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Peter Maydell , Peter Xu , Pei Zhang , Eric Auger , Maxime Coquelin , Marcel Apfelbaum , Paolo Bonzini , Richard Henderson , Eduardo Habkost From: Peter Xu We should handle VTD_FR_CONTEXT_ENTRY_P properly when synchronizing shadow page tables. Having invalid context entry there is perfectly valid when we move a device out of an existing domain. When that happens, instead of posting an error we invalidate the whole region. Without this patch, QEMU will crash if we do these steps: (1) start QEMU with VT-d IOMMU and two 10G NICs (ixgbe) (2) bind the NICs with vfio-pci in the guest (3) start testpmd with the NICs applied (4) stop testpmd (5) rebind the NIC back to ixgbe kernel driver The patch should fix it. Reported-by: Pei Zhang Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1627272 Signed-off-by: Peter Xu Reviewed-by: Eric Auger Reviewed-by: Maxime Coquelin Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- hw/i386/intel_iommu.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c index 25e54671f4..f24ebfca1c 100644 --- a/hw/i386/intel_iommu.c +++ b/hw/i386/intel_iommu.c @@ -38,6 +38,7 @@ #include "trace.h" static void vtd_address_space_refresh_all(IntelIOMMUState *s); +static void vtd_address_space_unmap(VTDAddressSpace *as, IOMMUNotifier *n); static void vtd_define_quad(IntelIOMMUState *s, hwaddr addr, uint64_t val, uint64_t wmask, uint64_t w1cmask) @@ -1066,11 +1067,27 @@ static int vtd_sync_shadow_page_table(VTDAddressSpace *vtd_as) { int ret; VTDContextEntry ce; + IOMMUNotifier *n; ret = vtd_dev_to_context_entry(vtd_as->iommu_state, pci_bus_num(vtd_as->bus), vtd_as->devfn, &ce); if (ret) { + if (ret == -VTD_FR_CONTEXT_ENTRY_P) { + /* + * It's a valid scenario to have a context entry that is + * not present. For example, when a device is removed + * from an existing domain then the context entry will be + * zeroed by the guest before it was put into another + * domain. When this happens, instead of synchronizing + * the shadow pages we should invalidate all existing + * mappings and notify the backends. + */ + IOMMU_NOTIFIER_FOREACH(n, &vtd_as->iommu) { + vtd_address_space_unmap(vtd_as, n); + } + ret = 0; + } return ret; } -- MST