qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Shameer Kolothum <skolothumtho@nvidia.com>
To: <qemu-arm@nongnu.org>, <qemu-devel@nongnu.org>
Cc: <eric.auger@redhat.com>, <peter.maydell@linaro.org>,
	<jgg@nvidia.com>, <nicolinc@nvidia.com>, <ddutile@redhat.com>,
	<berrange@redhat.com>, <nathanc@nvidia.com>, <mochs@nvidia.com>,
	<smostafa@google.com>, <wangzhou1@hisilicon.com>,
	<jiangkunkun@huawei.com>, <jonathan.cameron@huawei.com>,
	<zhangfei.gao@linaro.org>, <zhenzhong.duan@intel.com>,
	<yi.l.liu@intel.com>, <kjaju@nvidia.com>
Subject: [PATCH v6 17/33] hw/arm/smmuv3: Add support for providing a direct MSI doorbell GPA
Date: Thu, 20 Nov 2025 13:21:57 +0000	[thread overview]
Message-ID: <20251120132213.56581-18-skolothumtho@nvidia.com> (raw)
In-Reply-To: <20251120132213.56581-1-skolothumtho@nvidia.com>

Accelerated SMMUv3 instances rely on the physical SMMUv3 for nested
translation (Guest Stage-1, Host Stage-2). In this mode the guest’s
Stage-1 tables are programmed directly into hardware, and QEMU should
not attempt to walk them for translation since doing so is not reliably
safe. For vfio-pci endpoints behind such a vSMMU, the only translation
QEMU is responsible for is the MSI doorbell used during KVM MSI setup.

Add a device property to carry the MSI doorbell GPA from the virt
machine, and expose it through a new get_msi_direct_gpa PCIIOMMUOp.
kvm_arch_fixup_msi_route() can then use this GPA directly instead of
attempting a software walk of guest translation tables.

This enables correct MSI routing with accelerated SMMUv3 while avoiding
unsafe accesses to page tables.

For meaningful use of vfio-pci devices with accelerated SMMUv3, both KVM
and a kernel irqchip are required. Enforce this requirement when accel=on
is selected.

Signed-off-by: Shameer Kolothum <skolothumtho@nvidia.com>
---
 hw/arm/smmuv3-accel.c   | 10 ++++++++++
 hw/arm/smmuv3.c         |  2 ++
 hw/arm/virt.c           | 22 ++++++++++++++++++++++
 include/hw/arm/smmuv3.h |  1 +
 4 files changed, 35 insertions(+)

diff --git a/hw/arm/smmuv3-accel.c b/hw/arm/smmuv3-accel.c
index 65b577f49a..8f7c0cda05 100644
--- a/hw/arm/smmuv3-accel.c
+++ b/hw/arm/smmuv3-accel.c
@@ -392,6 +392,15 @@ static void smmuv3_accel_unset_iommu_device(PCIBus *bus, void *opaque,
     }
 }
 
+static uint64_t smmuv3_accel_get_msi_gpa(PCIBus *bus, void *opaque, int devfn)
+{
+    SMMUState *bs = opaque;
+    SMMUv3State *s = ARM_SMMUV3(bs);
+
+    g_assert(s->msi_gpa);
+    return s->msi_gpa;
+}
+
 /*
  * Only allow PCIe bridges, pxb-pcie roots, and GPEX roots so vfio-pci
  * endpoints can sit downstream. Accelerated SMMUv3 requires a vfio-pci
@@ -496,6 +505,7 @@ static const PCIIOMMUOps smmuv3_accel_ops = {
     .get_viommu_flags = smmuv3_accel_get_viommu_flags,
     .set_iommu_device = smmuv3_accel_set_iommu_device,
     .unset_iommu_device = smmuv3_accel_unset_iommu_device,
+    .get_msi_direct_gpa = smmuv3_accel_get_msi_gpa,
 };
 
 /* Based on SMUUv3 GPBA.ABORT configuration, attach a corresponding HWPT */
diff --git a/hw/arm/smmuv3.c b/hw/arm/smmuv3.c
index 42c60b1ec8..f02e3ee46c 100644
--- a/hw/arm/smmuv3.c
+++ b/hw/arm/smmuv3.c
@@ -1998,6 +1998,8 @@ static const Property smmuv3_properties[] = {
      * Defaults to stage 1
      */
     DEFINE_PROP_STRING("stage", SMMUv3State, stage),
+    /* GPA of MSI doorbell, for SMMUv3 accel use. */
+    DEFINE_PROP_UINT64("msi-gpa", SMMUv3State, msi_gpa, 0),
 };
 
 static void smmuv3_instance_init(Object *obj)
diff --git a/hw/arm/virt.c b/hw/arm/virt.c
index 25fb2bab56..ea3231543a 100644
--- a/hw/arm/virt.c
+++ b/hw/arm/virt.c
@@ -3052,6 +3052,14 @@ static void virt_machine_device_pre_plug_cb(HotplugHandler *hotplug_dev,
             /* The new SMMUv3 device is specific to the PCI bus */
             object_property_set_bool(OBJECT(dev), "smmu_per_bus", true, NULL);
         }
+        if (object_property_find(OBJECT(dev), "accel") &&
+            object_property_get_bool(OBJECT(dev), "accel", &error_abort)) {
+            if (!kvm_enabled() || !kvm_irqchip_in_kernel()) {
+                error_setg(errp, "SMMUv3 accel=on requires KVM with "
+                           "kernel-irqchip=on support");
+                    return;
+            }
+        }
     }
 }
 
@@ -3088,6 +3096,20 @@ static void virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
             }
 
             create_smmuv3_dev_dtb(vms, dev, bus);
+            if (object_property_find(OBJECT(dev), "accel") &&
+                object_property_get_bool(OBJECT(dev), "accel", &error_abort)) {
+                hwaddr db_start;
+
+                if (vms->msi_controller == VIRT_MSI_CTRL_ITS) {
+                    /* GITS_TRANSLATER page + offset */
+                    db_start = base_memmap[VIRT_GIC_ITS].base + 0x10000 + 0x40;
+                } else {
+                    /* MSI_SETSPI_NS page + offset */
+                    db_start = base_memmap[VIRT_GIC_V2M].base + 0x40;
+                }
+                object_property_set_uint(OBJECT(dev), "msi-gpa", db_start,
+                                         &error_abort);
+            }
         }
     }
 
diff --git a/include/hw/arm/smmuv3.h b/include/hw/arm/smmuv3.h
index e54ece2d38..5616a8a2be 100644
--- a/include/hw/arm/smmuv3.h
+++ b/include/hw/arm/smmuv3.h
@@ -67,6 +67,7 @@ struct SMMUv3State {
     /* SMMU has HW accelerator support for nested S1 + s2 */
     bool accel;
     struct SMMUv3AccelState *s_accel;
+    uint64_t msi_gpa;
 };
 
 typedef enum {
-- 
2.43.0



  parent reply	other threads:[~2025-11-20 13:25 UTC|newest]

Thread overview: 89+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-20 13:21 [PATCH v6 00/33] hw/arm/virt: Add support for user-creatable accelerated SMMUv3 Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 01/33] backends/iommufd: Introduce iommufd_backend_alloc_viommu Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 02/33] backends/iommufd: Introduce iommufd_backend_alloc_vdev Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 03/33] hw/arm/smmu-common: Factor out common helper functions and export Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 04/33] hw/arm/smmu-common: Make iommu ops part of SMMUState Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 05/33] hw/arm/smmuv3-accel: Introduce smmuv3 accel device Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 06/33] hw/arm/smmuv3-accel: Initialize shared system address space Shameer Kolothum
2025-12-08 17:05   ` Eric Auger
2025-11-20 13:21 ` [PATCH v6 07/33] hw/pci/pci: Move pci_init_bus_master() after adding device to bus Shameer Kolothum
2025-11-20 20:44   ` Nicolin Chen
2025-11-20 13:21 ` [PATCH v6 08/33] hw/pci/pci: Add optional supports_address_space() callback Shameer Kolothum
2025-11-20 20:51   ` Nicolin Chen
2025-11-21 10:38     ` Shameer Kolothum
2025-11-21 17:28       ` Nicolin Chen
2025-11-21 17:32         ` Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 09/33] hw/pci-bridge/pci_expander_bridge: Move TYPE_PXB_PCIE_DEV to header Shameer Kolothum
2025-11-20 20:52   ` Nicolin Chen
2025-11-20 13:21 ` [PATCH v6 10/33] hw/arm/smmuv3-accel: Restrict accelerated SMMUv3 to vfio-pci endpoints with iommufd Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 11/33] hw/arm/smmuv3: Implement get_viommu_cap() callback Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 12/33] hw/arm/smmuv3-accel: Add set/unset_iommu_device callback Shameer Kolothum
2025-12-09  7:57   ` Eric Auger
2025-11-20 13:21 ` [PATCH v6 13/33] hw/arm/smmuv3: propagate smmuv3_cmdq_consume() errors to caller Shameer Kolothum
2025-11-20 20:59   ` Nicolin Chen
2025-12-04 16:28   ` Eric Auger
2025-11-20 13:21 ` [PATCH v6 14/33] hw/arm/smmuv3-accel: Add nested vSTE install/uninstall support Shameer Kolothum
2025-12-09  8:14   ` Eric Auger
2025-11-20 13:21 ` [PATCH v6 15/33] hw/arm/smmuv3-accel: Install SMMUv3 GBPA based hwpt Shameer Kolothum
2025-11-20 21:03   ` Nicolin Chen
2025-11-20 13:21 ` [PATCH v6 16/33] hw/pci/pci: Introduce a callback to retrieve the MSI doorbell GPA directly Shameer Kolothum
2025-11-20 21:05   ` Nicolin Chen
2025-12-04 16:38   ` Eric Auger
2025-12-04 18:57     ` Shameer Kolothum
2025-12-08 17:03       ` Eric Auger
2025-11-20 13:21 ` Shameer Kolothum [this message]
2025-11-20 21:21   ` [PATCH v6 17/33] hw/arm/smmuv3: Add support for providing a direct MSI doorbell GPA Nicolin Chen
2025-11-21  9:57     ` Shameer Kolothum
2025-11-21 17:56       ` Nicolin Chen
2025-11-24  8:05         ` Shameer Kolothum
2025-11-24 18:34           ` Nicolin Chen
2025-11-24 19:01             ` Shameer Kolothum
2025-11-24 20:08               ` Nicolin Chen
2025-11-20 13:21 ` [PATCH v6 18/33] hw/arm/smmuv3-accel: Add support to issue invalidation cmd to host Shameer Kolothum
2025-11-20 13:21 ` [PATCH v6 19/33] hw/arm/smmuv3: Initialize ID registers early during realize() Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 20/33] hw/arm/smmuv3-accel: Get host SMMUv3 hw info and validate Shameer Kolothum
2025-11-20 21:27   ` Nicolin Chen
2025-11-20 21:30     ` Nicolin Chen
2025-11-20 13:22 ` [PATCH v6 21/33] hw/pci-host/gpex: Allow to generate preserve boot config DSM #5 Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 22/33] hw/arm/virt: Set PCI preserve_config for accel SMMUv3 Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 23/33] tests/qtest/bios-tables-test: Prepare for IORT revison upgrade Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 24/33] hw/arm/virt-acpi-build: Add IORT RMR regions to handle MSI nested binding Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 25/33] tests/qtest/bios-tables-test: Update IORT blobs after revision upgrade Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 26/33] hw/arm/smmuv3: Add accel property for SMMUv3 device Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 27/33] hw/arm/smmuv3-accel: Add a property to specify RIL support Shameer Kolothum
2025-11-20 21:34   ` Nicolin Chen via
2025-11-21 10:04     ` Shameer Kolothum
2025-11-20 13:22 ` [PATCH v6 28/33] hw/arm/smmuv3-accel: Add support for ATS Shameer Kolothum
2025-11-20 21:40   ` Nicolin Chen
2025-11-24 12:00   ` Zhangfei Gao
2025-11-24 12:48     ` Shameer Kolothum
2025-12-08 17:36   ` Eric Auger
2025-11-20 13:22 ` [PATCH v6 29/33] hw/arm/smmuv3-accel: Add property to specify OAS bits Shameer Kolothum
2025-11-20 21:47   ` Nicolin Chen
2025-12-08 17:17     ` Eric Auger
2025-11-20 13:22 ` [PATCH v6 30/33] backends/iommufd: Retrieve PASID width from iommufd_backend_get_device_info() Shameer Kolothum
2025-11-20 21:50   ` Nicolin Chen
2025-11-20 13:22 ` [PATCH v6 31/33] Extend get_cap() callback to support PASID Shameer Kolothum
2025-11-20 21:56   ` Nicolin Chen
2025-12-08 17:20   ` Eric Auger
2025-11-20 13:22 ` [PATCH v6 32/33] vfio: Synthesize vPASID capability to VM Shameer Kolothum
2025-11-20 21:59   ` Nicolin Chen
2025-12-09  9:51   ` Eric Auger
2025-12-09 11:17     ` Yi Liu
2025-11-20 13:22 ` [PATCH v6 33/33] hw/arm/smmuv3-accel: Add support for PASID enable Shameer Kolothum
2025-11-20 22:09   ` Nicolin Chen
2025-11-21 10:22     ` Shameer Kolothum
2025-11-21 17:50       ` Nicolin Chen
2025-11-21 18:36         ` Nicolin Chen
2025-11-21 18:44         ` Jason Gunthorpe
2025-11-24  8:17         ` Shameer Kolothum
2025-11-20 17:06 ` [PATCH v6 00/33] hw/arm/virt: Add support for user-creatable accelerated SMMUv3 Nicolin Chen
2025-11-24 12:09   ` Zhangfei Gao
2025-12-08 10:08 ` Duan, Zhenzhong
2025-12-08 11:15   ` Shameer Kolothum
2025-12-09  2:30     ` Duan, Zhenzhong
2025-12-09  3:33       ` Yi Liu
2025-12-09 10:31 ` Cédric Le Goater
2025-12-10 15:07   ` Shameer Kolothum
2025-12-10 16:07     ` Cédric Le Goater
2025-12-10 16:18       ` Eric Auger

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=20251120132213.56581-18-skolothumtho@nvidia.com \
    --to=skolothumtho@nvidia.com \
    --cc=berrange@redhat.com \
    --cc=ddutile@redhat.com \
    --cc=eric.auger@redhat.com \
    --cc=jgg@nvidia.com \
    --cc=jiangkunkun@huawei.com \
    --cc=jonathan.cameron@huawei.com \
    --cc=kjaju@nvidia.com \
    --cc=mochs@nvidia.com \
    --cc=nathanc@nvidia.com \
    --cc=nicolinc@nvidia.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=smostafa@google.com \
    --cc=wangzhou1@hisilicon.com \
    --cc=yi.l.liu@intel.com \
    --cc=zhangfei.gao@linaro.org \
    --cc=zhenzhong.duan@intel.com \
    /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).