From: "Cédric Le Goater" <clg@redhat.com>
To: Shameer Kolothum <skolothumtho@nvidia.com>,
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: Re: [PATCH v6 26/33] hw/arm/smmuv3: Add accel property for SMMUv3 device
Date: Thu, 11 Dec 2025 16:11:34 +0100 [thread overview]
Message-ID: <9a54ba3b-1a14-4240-80a7-715e20b99dac@redhat.com> (raw)
In-Reply-To: <20251120132213.56581-27-skolothumtho@nvidia.com>
On 11/20/25 14:22, Shameer Kolothum wrote:
> Introduce an “accel” property to enable accelerator mode.
>
> Live migration is currently unsupported when accelerator mode is enabled,
> so a migration blocker is added.
>
> Because this mode relies on IORT RMR for MSI support, accelerator mode is
> not supported for device tree boot.
>
> Also, in the accelerated SMMUv3 case, the host SMMUv3 is configured in nested
> mode (S1 + S2), and the guest owns the Stage-1 page table. Therefore, we
> expose only Stage-1 to the guest to ensure it uses the correct page table
> format.
>
> Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>
> Reviewed-by: Eric Auger <eric.auger@redhat.com>
> Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
> Signed-off-by: Shameer Kolothum <skolothumtho@nvidia.com>
> ---
> hw/arm/smmuv3.c | 26 ++++++++++++++++++++++++++
> hw/arm/virt-acpi-build.c | 4 +---
> hw/arm/virt.c | 15 +++++++++++----
> include/hw/arm/smmuv3.h | 1 +
> 4 files changed, 39 insertions(+), 7 deletions(-)
>
> diff --git a/hw/arm/smmuv3.c b/hw/arm/smmuv3.c
> index dba5abc8d3..8352dd5757 100644
> --- a/hw/arm/smmuv3.c
> +++ b/hw/arm/smmuv3.c
> @@ -20,6 +20,7 @@
> #include "qemu/bitops.h"
> #include "hw/irq.h"
> #include "hw/sysbus.h"
> +#include "migration/blocker.h"
> #include "migration/vmstate.h"
> #include "hw/qdev-properties.h"
> #include "hw/qdev-core.h"
> @@ -1915,6 +1916,17 @@ static void smmu_reset_exit(Object *obj, ResetType type)
> smmuv3_accel_reset(s);
> }
>
> +static bool smmu_validate_property(SMMUv3State *s, Error **errp)
> +{
> +#ifndef CONFIG_ARM_SMMUV3_ACCEL
> + if (s->accel) {
> + error_setg(errp, "accel=on support not compiled in");
> + return false;
> + }
> +#endif
> + return true;
> +}
> +
> static void smmu_realize(DeviceState *d, Error **errp)
> {
> SMMUState *sys = ARM_SMMU(d);
> @@ -1923,8 +1935,17 @@ static void smmu_realize(DeviceState *d, Error **errp)
> SysBusDevice *dev = SYS_BUS_DEVICE(d);
> Error *local_err = NULL;
>
> + if (!smmu_validate_property(s, errp)) {
> + return;
> + }
You should really consider adding a new QOM model for the "accel" mode.
> if (s->accel) {
> smmuv3_accel_init(s);
> + error_setg(&s->migration_blocker, "Migration not supported with SMMUv3 "
> + "accelerator mode enabled");
> + if (migrate_add_blocker(&s->migration_blocker, errp) < 0) {
> + return;
> + }
I would add the migration blocker in a separate patch.
> }
>
> c->parent_realize(d, &local_err);
> @@ -2023,6 +2044,7 @@ static const Property smmuv3_properties[] = {
> * Defaults to stage 1
> */
> DEFINE_PROP_STRING("stage", SMMUv3State, stage),
> + DEFINE_PROP_BOOL("accel", SMMUv3State, accel, false),
> /* GPA of MSI doorbell, for SMMUv3 accel use. */
> DEFINE_PROP_UINT64("msi-gpa", SMMUv3State, msi_gpa, 0),
> };
> @@ -2046,6 +2068,10 @@ static void smmuv3_class_init(ObjectClass *klass, const void *data)
> device_class_set_props(dc, smmuv3_properties);
> dc->hotpluggable = false;
> dc->user_creatable = true;
> +
> + object_class_property_set_description(klass, "accel",
> + "Enable SMMUv3 accelerator support. Allows host SMMUv3 to be "
> + "configured in nested mode for vfio-pci dev assignment");
> }
>
> static int smmuv3_notify_flag_changed(IOMMUMemoryRegion *iommu,
> diff --git a/hw/arm/virt-acpi-build.c b/hw/arm/virt-acpi-build.c
> index 7a7b2e62c1..fd78c39317 100644
> --- a/hw/arm/virt-acpi-build.c
> +++ b/hw/arm/virt-acpi-build.c
> @@ -400,9 +400,7 @@ static int iort_smmuv3_devices(Object *obj, void *opaque)
> }
>
> bus = PCI_BUS(object_property_get_link(obj, "primary-bus", &error_abort));
> - if (object_property_find(obj, "accel")) {
> - sdev.accel = object_property_get_bool(obj, "accel", &error_abort);
> - }
> + sdev.accel = object_property_get_bool(obj, "accel", &error_abort);
> pbus = PLATFORM_BUS_DEVICE(vms->platform_bus_dev);
> sbdev = SYS_BUS_DEVICE(obj);
> sdev.base = platform_bus_get_mmio_addr(pbus, sbdev, 0);
> diff --git a/hw/arm/virt.c b/hw/arm/virt.c
> index 8503879c3d..51b15aef37 100644
> --- a/hw/arm/virt.c
> +++ b/hw/arm/virt.c
> @@ -3052,13 +3052,21 @@ 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 (object_property_get_bool(OBJECT(dev), "accel", &error_abort)) {
> + char *stage;
> +
> if (!kvm_enabled() || !kvm_irqchip_in_kernel()) {
> error_setg(errp, "SMMUv3 accel=on requires KVM with "
> "kernel-irqchip=on support");
> return;
> }
> + stage = object_property_get_str(OBJECT(dev), "stage", &error_fatal);
> + /* If no stage specified, SMMUv3 will default to stage 1 */
> + if (*stage && strcmp("1", stage)) {
It would be less "ugly" to check SMMUStage enum values instead
of using string compare :/
> + error_setg(errp, "Only stage1 is supported for SMMUV3 with "
> + "accel=on");
Can't this condition be checked in the realize handler instead ?
C.
> + return;
> + }
> }
> }
> }
> @@ -3096,8 +3104,7 @@ 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)) {
> + if (object_property_get_bool(OBJECT(dev), "accel", &error_abort)) {
> hwaddr db_start;
>
> if (vms->msi_controller == VIRT_MSI_CTRL_ITS) {
> diff --git a/include/hw/arm/smmuv3.h b/include/hw/arm/smmuv3.h
> index 5616a8a2be..9c39acd5ca 100644
> --- a/include/hw/arm/smmuv3.h
> +++ b/include/hw/arm/smmuv3.h
> @@ -68,6 +68,7 @@ struct SMMUv3State {
> bool accel;
> struct SMMUv3AccelState *s_accel;
> uint64_t msi_gpa;
> + Error *migration_blocker;
> };
>
> typedef enum {
next prev parent reply other threads:[~2025-12-11 15:12 UTC|newest]
Thread overview: 99+ 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
2025-12-11 11:03 ` Cédric Le Goater
2025-11-20 13:21 ` [PATCH v6 05/33] hw/arm/smmuv3-accel: Introduce smmuv3 accel device Shameer Kolothum
2025-12-11 12:54 ` Cédric Le Goater
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
2025-12-11 14:40 ` Cédric Le Goater
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
2025-12-11 13:41 ` Cédric Le Goater
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 ` [PATCH v6 17/33] hw/arm/smmuv3: Add support for providing a direct MSI doorbell GPA Shameer Kolothum
2025-11-20 21:21 ` 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
2025-12-11 14:03 ` Cédric Le Goater
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
2025-12-11 15:11 ` Cédric Le Goater [this message]
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
2025-12-11 15:14 ` Cédric Le Goater
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
2025-12-11 15:23 ` Cédric Le Goater
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
2025-12-11 15:24 ` Cédric Le Goater
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
2025-12-11 15:26 ` Cédric Le Goater
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=9a54ba3b-1a14-4240-80a7-715e20b99dac@redhat.com \
--to=clg@redhat.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=skolothumtho@nvidia.com \
--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).