From: eric.auger@redhat.com (Eric Auger)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v14 14/16] vfio/type1: Check doorbell safety
Date: Wed, 12 Oct 2016 13:22:22 +0000 [thread overview]
Message-ID: <1476278544-3397-15-git-send-email-eric.auger@redhat.com> (raw)
In-Reply-To: <1476278544-3397-1-git-send-email-eric.auger@redhat.com>
On x86 IRQ remapping is abstracted by the IOMMU. On ARM this is abstracted
by the msi controller.
Since we currently have no way to detect whether the MSI controller is
upstream or downstream to the IOMMU we rely on the MSI doorbell information
registered by the interrupt controllers. In case at least one doorbell
does not implement proper isolation, we state the assignment is unsafe
with regard to interrupts. This is a coarse assessment but should allow to
wait for a better system description.
At this point ARM sMMU still advertises IOMMU_CAP_INTR_REMAP. This is
removed in next patch.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
---
v13 -> v15:
- check vfio_msi_resv before checking whether msi doorbell is safe
v9 -> v10:
- coarse safety assessment based on MSI doorbell info
v3 -> v4:
- rename vfio_msi_parent_irq_remapping_capable into vfio_safe_irq_domain
and irq_remapping into safe_irq_domains
v2 -> v3:
- protect vfio_msi_parent_irq_remapping_capable with
CONFIG_GENERIC_MSI_IRQ_DOMAIN
---
drivers/vfio/vfio_iommu_type1.c | 30 +++++++++++++++++++++++++++++-
1 file changed, 29 insertions(+), 1 deletion(-)
diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index e0c97ef..c18ba9d 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -442,6 +442,29 @@ static void vfio_unmap_unpin(struct vfio_iommu *iommu, struct vfio_dma *dma)
}
/**
+ * vfio_msi_resv - Return whether any VFIO iommu domain requires
+ * MSI mapping
+ *
+ * @iommu: vfio iommu handle
+ *
+ * Return: true of MSI mapping is needed, false otherwise
+ */
+static bool vfio_msi_resv(struct vfio_iommu *iommu)
+{
+ struct iommu_domain_msi_resv msi_resv;
+ struct vfio_domain *d;
+ int ret;
+
+ list_for_each_entry(d, &iommu->domain_list, next) {
+ ret = iommu_domain_get_attr(d->domain, DOMAIN_ATTR_MSI_RESV,
+ &msi_resv);
+ if (!ret)
+ return true;
+ }
+ return false;
+}
+
+/**
* vfio_set_msi_aperture - Sets the msi aperture on all domains
* requesting MSI mapping
*
@@ -945,8 +968,13 @@ static int vfio_iommu_type1_attach_group(void *iommu_data,
INIT_LIST_HEAD(&domain->group_list);
list_add(&group->next, &domain->group_list);
+ /*
+ * to advertise safe interrupts either the IOMMU or the MSI controllers
+ * must support IRQ remapping (aka. interrupt translation)
+ */
if (!allow_unsafe_interrupts &&
- !iommu_capable(bus, IOMMU_CAP_INTR_REMAP)) {
+ (!iommu_capable(bus, IOMMU_CAP_INTR_REMAP) &&
+ !(vfio_msi_resv(iommu) && iommu_msi_doorbell_safe()))) {
pr_warn("%s: No interrupt remapping support. Use the module param \"allow_unsafe_interrupts\" to enable VFIO IOMMU support on this platform\n",
__func__);
ret = -EPERM;
--
1.9.1
next prev parent reply other threads:[~2016-10-12 13:22 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-10-12 13:22 [PATCH v14 00/16] KVM PCIe/MSI passthrough on ARM/ARM64 Eric Auger
2016-10-12 13:22 ` [PATCH v14 01/16] iommu/iova: fix __alloc_and_insert_iova_range Eric Auger
2016-10-12 13:22 ` [PATCH v14 02/16] iommu: Introduce DOMAIN_ATTR_MSI_RESV Eric Auger
2016-10-12 13:22 ` [PATCH v14 03/16] iommu/dma: Allow MSI-only cookies Eric Auger
2016-10-12 13:22 ` [PATCH v14 04/16] iommu/dma: MSI doorbell alloc/free Eric Auger
2016-10-14 11:25 ` Punit Agrawal
2016-10-17 14:25 ` Auger Eric
2016-10-12 13:22 ` [PATCH v14 05/16] iommu/dma: Introduce iommu_calc_msi_resv Eric Auger
2016-10-12 13:22 ` [PATCH v14 06/16] iommu/arm-smmu: Implement domain_get_attr for DOMAIN_ATTR_MSI_RESV Eric Auger
2016-10-12 13:22 ` [PATCH v14 07/16] irqchip/gic-v2m: Register the MSI doorbell Eric Auger
2016-10-12 13:22 ` [PATCH v14 08/16] irqchip/gicv3-its: " Eric Auger
2016-10-12 13:22 ` [PATCH v14 09/16] vfio: Introduce a vfio_dma type field Eric Auger
2016-10-12 13:22 ` [PATCH v14 10/16] vfio/type1: vfio_find_dma accepting a type argument Eric Auger
2016-10-12 13:22 ` [PATCH v14 11/16] vfio/type1: Implement recursive vfio_find_dma_from_node Eric Auger
2016-10-12 13:22 ` [PATCH v14 12/16] vfio/type1: Handle unmap/unpin and replay for VFIO_IOVA_RESERVED slots Eric Auger
2016-10-12 13:22 ` [PATCH v14 13/16] vfio: Allow reserved msi iova registration Eric Auger
2016-10-12 13:22 ` Eric Auger [this message]
2016-11-03 13:45 ` [PATCH v14 14/16] vfio/type1: Check doorbell safety Diana Madalina Craciun
2016-11-03 14:14 ` Auger Eric
2016-10-12 13:22 ` [PATCH v14 15/16] iommu/arm-smmu: Do not advertise IOMMU_CAP_INTR_REMAP Eric Auger
2016-10-12 13:22 ` [PATCH v14 16/16] vfio/type1: Introduce MSI_RESV capability Eric Auger
2016-10-14 11:24 ` [PATCH v14 00/16] KVM PCIe/MSI passthrough on ARM/ARM64 Punit Agrawal
2016-10-17 14:19 ` Auger Eric
2016-10-17 15:08 ` Punit Agrawal
2016-10-20 17:32 ` Will Deacon
2016-10-21 9:26 ` Auger Eric
2016-10-24 19:39 ` Robin Murphy
2016-11-02 16:15 ` Auger Eric
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1476278544-3397-15-git-send-email-eric.auger@redhat.com \
--to=eric.auger@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).