iommufd: Enable noiommu mode for cdev
Subsystems (6)
v2 cover iommufd: Enable noiommu mode for cdev
VFIO's unsafe_noiommu_mode has long provided a way for userspace drivers to operate on platforms lacking a hardware IOMMU. Today, IOMMUFD also supports No-IOMMU mode for group-based devices under vfio_compat mode. However, IOMMUFD's native character device (cdev) does not yet support No-IOMMU mode, which is the purpose of this patch.
In summary, we have:
|-------------------------+------+---------------| | Device access mode | VFIO | IOMMUFD | |-------------------------+------+---------------| | group /dev/vfio/$GROUP | Yes | Yes | |-------------------------+------+---------------| | cdev /dev/vfio/devices/ | No | This patch | |-------------------------+------+---------------|
Beyond enabling cdev for IOMMUFD, this patch also addresses the following deficiencies in the current No-IOMMU mode suggested by Jason[1]:
- Devices operating under No-IOMMU mode are limited to device-level UAPI access, without container or IOAS-level capabilities. Consequently, user-space drivers lack structured mechanisms for page pinning and often resort to mlock(), which is less robust than pin_user_pages() used for devices backed by a physical IOMMU. For example, mlock() does not prevent page migration.
- There is no architectural mechanism for obtaining physical addresses for DMA. As a workaround, user-space drivers frequently rely on /proc/pagemap tricks or hardcoded values.
By allowing noiommu device access to IOMMUFD IOAS and HWPT objects, this patch brings No-IOMMU mode closer to full citizenship within the IOMMU subsystem. In addition to addressing the two deficiencies mentioned above, the expectation is that it will also enable No-IOMMU devices to seamlessly participate in live update sessions via KHO [2].
Furthermore, these devices will use the IOMMUFD-based ownership checking model for VFIO_DEVICE_PCI_HOT_RESET, eliminating the need for an iommufd_access object as required in a previous attempt [3].
ChangeLog:
V2:
- Fix build depenency by adding IOMMU_SUPPORT in [8/11]
- Add an optimization to scan beyond the first page for a contiguous physical address range and return its length instead of a single page.[4/11]
Since RFC[4]:
- Abandoned dummy iommu driver approach as patch 1-3 absorbed the changes into iommufd.
[1] https://lore.kernel.org/linux-iommu/[email protected]/
[2] https://lore.kernel.org/linux-pci/[email protected]/
[3] https://lore.kernel.org/kvm/[email protected]/
[4] https://lore.kernel.org/linux-iommu/[email protected]/
Thanks,
Jacob
Jacob Pan (8):
iommufd: Add an ioctl IOMMU_IOAS_GET_PA to query PA from IOVA
vfio: Allow null group for noiommu without containers
vfio: Introduce and set noiommu flag on vfio_device
vfio: Update noiommu device detection logic for cdev
vfio: Enable cdev noiommu mode under iommufd vfio:selftest: Handle VFIO noiommu cdev selftests/vfio: Add iommufd noiommu mode selftest for cdev
Doc: Update VFIO NOIOMMU mode
Jason Gunthorpe (3):
iommufd: Support a HWPT without an iommu driver for noiommu
iommufd: Move igroup allocation to a function
iommufd: Allow binding to a noiommu device
Documentation/driver-api/vfio.rst | 44 +- drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/device.c | 161 +++-- drivers/iommu/iommufd/hw_pagetable.c | 11 +- drivers/iommu/iommufd/hwpt_noiommu.c | 91 +++ drivers/iommu/iommufd/io_pagetable.c | 60 ++ drivers/iommu/iommufd/ioas.c | 28 + drivers/iommu/iommufd/iommufd_private.h | 5 + drivers/iommu/iommufd/main.c | 3 + drivers/vfio/Kconfig | 7 +- drivers/vfio/group.c | 35 +- drivers/vfio/iommufd.c | 7 - drivers/vfio/vfio.h | 34 +- drivers/vfio/vfio_main.c | 22 +- include/linux/vfio.h | 10 + include/uapi/linux/iommufd.h | 25 + tools/testing/selftests/vfio/Makefile | 1 + .../selftests/vfio/lib/vfio_pci_device.c | 25 +- .../vfio/vfio_iommufd_noiommu_test.c | 549 ++++++++++++++++++ 19 files changed, 1027 insertions(+), 92 deletions(-) create mode 100644 drivers/iommu/iommufd/hwpt_noiommu.c create mode 100644 tools/testing/selftests/vfio/vfio_iommufd_noiommu_test.c
Signature · 1 line
2.34.1
PATCH v2 1/11 iommufd: Support a HWPT without an iommu driver for noiommu
From: Jason Gunthorpe <[email protected]>
Create just a little part of a real iommu driver, enough to slot in under the dev_iommu_ops() and allow iommufd to call domain_alloc_paging_flags() and fail everything else.
This allows explicitly creating a HWPT under an IOAS.
--- drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/hw_pagetable.c | 11 ++- drivers/iommu/iommufd/hwpt_noiommu.c | 91 +++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 2 + 4 files changed, 103 insertions(+), 2 deletions(-) create mode 100644 drivers/iommu/iommufd/hwpt_noiommu.c
drivers/iommu/iommufd/Makefile +1 -0
@@ -10,6 +10,7 @@ iommufd-y := \ vfio_compat.o \ viommu.o +iommufd-$(CONFIG_VFIO_NOIOMMU) += hwpt_noiommu.o iommufd-$(CONFIG_IOMMUFD_TEST) += selftest.o obj-$(CONFIG_IOMMUFD) += iommufd.o
drivers/iommu/iommufd/hw_pagetable.c +9 -2
@@ -8,6 +8,13 @@#include "../iommu-priv.h" #include "iommufd_private.h" +static const struct iommu_ops *get_iommu_ops (struct iommufd_device *idev) +{ +if (IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && !idev->igroup->group) +return &iommufd_noiommu_ops; +return dev_iommu_ops (idev->dev); +} +static void __iommufd_hwpt_destroy (struct iommufd_hw_pagetable *hwpt) {if (hwpt->domain)
@@ -114,7 +121,7 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,IOMMU_HWPT_ALLOC_DIRTY_TRACKING |IOMMU_HWPT_FAULT_ID_VALID |IOMMU_HWPT_ALLOC_PASID ; -const struct iommu_ops *ops =dev_iommu_ops (idev->dev); +const struct iommu_ops *ops =get_iommu_ops (idev);struct iommufd_hwpt_paging *hwpt_paging;struct iommufd_hw_pagetable *hwpt;int rc;
@@ -229,7 +236,7 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, struct iommufd_device *idev, u32 flags, const struct iommu_user_data *user_data) { - const struct iommu_ops *ops = dev_iommu_ops(idev->dev); + const struct iommu_ops *ops = get_iommu_ops(idev);struct iommufd_hwpt_nested *hwpt_nested;struct iommufd_hw_pagetable *hwpt;int rc;
drivers/iommu/iommufd/hwpt_noiommu.c new file +91 -0
91 lines not on this page · raw message
drivers/iommu/iommufd/iommufd_private.h +2 -0
@@ -464,6 +464,8 @@ static inline void iommufd_hw_pagetable_put(struct iommufd_ctx *ictx,refcount_dec (&hwpt->obj.users); } +extern struct iommu_ops iommufd_noiommu_ops; +struct iommufd_attach ;struct iommufd_group {
Signature · 1 line
2.34.1
Re: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
Quoted 147 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
Create just a little part of a real iommu driver, enough to slot in under the
dev_iommu_ops()and allow iommufd to calldomain_alloc_paging_flags()and fail everything else.This allows explicitly creating a HWPT under an IOAS.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>---
drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/hw_pagetable.c | 11 ++- drivers/iommu/iommufd/hwpt_noiommu.c | 91 +++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 2 + 4 files changed, 103 insertions(+), 2 deletions(-)create mode 100644
drivers/iommu/iommufd/hwpt_noiommu.cdrivers/iommu/iommufd/Makefile +1
-0@@ -10,6 +10,7 @@ iommufd-y := \ vfio_compat.o \ viommu.o +iommufd-$(CONFIG_VFIO_NOIOMMU) += hwpt_noiommu.oiommufd-$(CONFIG_IOMMUFD_TEST) += selftest.o
obj-$(CONFIG_IOMMUFD) += iommufd.o
drivers/iommu/iommufd/hw_pagetable.c +9
-2@@ -8,6 +8,13 @@#include "../iommu-priv.h" #include "iommufd_private.h" +static const struct iommu_ops *get_iommu_ops (struct iommufd_device *idev) +{ +if (IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && !idev->igroup->group) +return &iommufd_noiommu_ops; +return dev_iommu_ops (idev->dev); +} +static void __iommufd_hwpt_destroy (struct iommufd_hw_pagetable *hwpt) {if (hwpt->domain)@@ -114,7 +121,7 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,IOMMU_HWPT_ALLOC_DIRTY_TRACKING |IOMMU_HWPT_FAULT_ID_VALID |IOMMU_HWPT_ALLOC_PASID ; -const struct iommu_ops *ops =dev_iommu_ops (idev->dev); +const struct iommu_ops *ops =get_iommu_ops (idev);struct iommufd_hwpt_paging *hwpt_paging;struct iommufd_hw_pagetable *hwpt;int rc;@@ -229,7 +236,7 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, struct iommufd_device *idev, u32 flags, const struct iommu_user_data *user_data) { - const struct iommu_ops *ops = dev_iommu_ops(idev->dev); + const struct iommu_ops *ops = get_iommu_ops(idev);struct iommufd_hwpt_nested *hwpt_nested;struct iommufd_hw_pagetable *hwpt;int rc;drivers/iommu/iommufd/hwpt_noiommu.c new file +75
-0@@ -0,0 +1,91 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* +* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES +*/ +#include <linux/iommu.h> +#include <linux/generic_pt/iommu.h> +#include "iommufd_private.h" + +static const struct iommu_domain_ops noiommu_amdv1_ops; + +struct noiommu_domain { +union { +struct iommu_domain domain; +struct pt_iommu_amdv1 amdv1; + }; +spinlock_t lock; +}; +PT_IOMMU_CHECK_DOMAIN(struct noiommu_domain, amdv1.iommu, domain); + +static void noiommu_change_top (struct pt_iommu *iommu_table, +phys_addr_t top_paddr,unsigned int top_level) +{ +} + +static spinlock_t *noiommu_get_top_lock (struct pt_iommu *iommupt) +{ +struct noiommu_domain *domain = + container_of(iommupt, struct noiommu_domain, amdv1.iommu); + +return &domain->lock; +} + +static const struct pt_iommu_driver_ops noiommu_driver_ops = { + .get_top_lock = noiommu_get_top_lock, + .change_top = noiommu_change_top, +}; + +static struct iommu_domain * +noiommu_alloc_paging_flags (struct device *dev,u32 flags, +const struct iommu_user_data *user_data) +{ +struct pt_iommu_amdv1_cfg cfg = {}; +struct noiommu_domain *dom; +int rc; + +if (flags || user_data) +return ERR_PTR (-EOPNOTSUPP ); + + cfg.common.hw_max_vasz_lg2 =64 ; + cfg.common.hw_max_oasz_lg2 =52 ; + cfg.starting_level =2 ; + cfg.common.features = + (BIT (PT_FEAT_DYNAMIC_TOP ) |BIT (PT_FEAT_AMDV1_ENCRYPT_TABLES ) | +BIT (PT_FEAT_AMDV1_FORCE_COHERENCE )); + + dom =kzalloc (sizeof (*dom),GFP_KERNEL ); +if (!dom) +return ERR_PTR (-ENOMEM ); + +spin_lock_init (&dom->lock); + dom->amdv1.iommu.nid =NUMA_NO_NODE ; + dom->amdv1.iommu.driver_ops = &noiommu_driver_ops; + dom->domain.ops = &noiommu_amdv1_ops; + +/* Use mock page table which is based on AMDV1 */ + rc =pt_iommu_amdv1_init (&dom->amdv1, &cfg,GFP_KERNEL ); +if (rc) { +kfree (dom); +return ERR_PTR (rc); + } + +return &dom->domain; +} + +static void noiommu_domain_free (struct iommu_domain *iommu_domain)
drivers/iommu/iommufd/hwpt_noiommu.c line 85
Quoted 10 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +10
-0+{ +struct noiommu_domain *domain = + container_of(iommu_domain, struct noiommu_domain, domain); + +pt_iommu_deinit (&domain->amdv1.iommu); +kfree (domain); +} + +static const struct iommu_domain_ops noiommu_amdv1_ops = { +IOMMU_PT_DOMAIN_OPS (amdv1),
I understand that this fits in really well into the iommufd/hwpt construction, but do we need page tables for this as all the iova-to-phys information should be available in the IOPT in IOAS? As the get_pa() function introduced in the later patch is only used for noiommu use-cases, it can use the IOPT to get the physical addresses?
Quoted 22 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +6
-0+ .free = noiommu_domain_free, +}; + +struct iommu_ops iommufd_noiommu_ops = { + .domain_alloc_paging_flags = noiommu_alloc_paging_flags, +};drivers/iommu/iommufd/iommufd_private.h +2
-0@@ -464,6 +464,8 @@ static inline void iommufd_hw_pagetable_put(struct iommufd_ctx *ictx,refcount_dec (&hwpt->obj.users); } +extern struct iommu_ops iommufd_noiommu_ops; +struct iommufd_attach ;struct iommufd_group {Signature · 1 line
2.34.1
Jason Gunthorpe
maintainerRe: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
Quoted 6 lines · Samiullah Khawaja
drivers/iommu/iommufd/hwpt_noiommu.c line 85
Quoted 2 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +2
-0+static const struct iommu_domain_ops noiommu_amdv1_ops = { +IOMMU_PT_DOMAIN_OPS (amdv1),I understand that this fits in really well into the iommufd/hwpt construction, but do we need page tables for this as all the iova-to-phys information should be available in the IOPT in IOAS?
Yes we do! That is the whole point.
In iommufd once you pin the memory the phys is stored in only two possible ways:
1) Inside an xarray if an access is used
2) Inside at least one iommu_domain
That's it. So to fit noiommu into this scheme, and have it rely on the existing pinning, we either have to make it use an access or make it use an iommu_domain -> a real one that can store phys.
Maybe a comment is helpful, but using the domain like this to store the pinned phys has been the vfio design from day 1..
Quoted 2 lines
get_pa()function introduced in the later patch is only used for noiommu use-cases, it can use the IOPT to get the physical addresses?
No.
Jason
Re: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
Quoted 14 lines · Jason Gunthorpe
Quoted 6 lines · Samiullah Khawaja
drivers/iommu/iommufd/hwpt_noiommu.c line 85
Quoted 2 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +2
-0+static const struct iommu_domain_ops noiommu_amdv1_ops = { +IOMMU_PT_DOMAIN_OPS (amdv1),I understand that this fits in really well into the iommufd/hwpt construction, but do we need page tables for this as all the iova-to-phys information should be available in the IOPT in IOAS?
Yes we do! That is the whole point.
In iommufd once you pin the memory the phys is stored in only two possible ways:
Quoted 6 lines
1) Inside an xarray if an access is used
2) Inside at least oneiommu_domainThat's it. So to fit noiommu into this scheme, and have it rely on the existing pinning, we either have to make it use an access or make it use an
iommu_domain-> a real one that can store phys.
Thanks for the explanation.
I missed the part where once pinning is done, the pfns are only available in those two places.
Quoted 10 lines
Maybe a comment is helpful, but using the domain like this to store the pinned phys has been the vfio design from day 1..
Quoted 2 lines
get_pa()function introduced in the later patch is only used for noiommu use-cases, it can use the IOPT to get the physical addresses?No.
Jason
Re: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
In reply to Jacob PanQuoted 147 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
Create just a little part of a real iommu driver, enough to slot in under the
dev_iommu_ops()and allow iommufd to calldomain_alloc_paging_flags()and fail everything else.This allows explicitly creating a HWPT under an IOAS.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/hw_pagetable.c | 11 ++- drivers/iommu/iommufd/hwpt_noiommu.c | 91 +++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 2 + 4 files changed, 103 insertions(+), 2 deletions(-) create mode 100644 drivers/iommu/iommufd/hwpt_noiommu.cdrivers/iommu/iommufd/Makefile +1
-0@@ -10,6 +10,7 @@ iommufd-y := \ vfio_compat.o \ viommu.o +iommufd-$(CONFIG_VFIO_NOIOMMU) += hwpt_noiommu.o iommufd-$(CONFIG_IOMMUFD_TEST) += selftest.o obj-$(CONFIG_IOMMUFD) += iommufd.odrivers/iommu/iommufd/hw_pagetable.c +9
-2@@ -8,6 +8,13 @@#include "../iommu-priv.h" #include "iommufd_private.h" +static const struct iommu_ops *get_iommu_ops (struct iommufd_device *idev) +{ +if (IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && !idev->igroup->group) +return &iommufd_noiommu_ops; +return dev_iommu_ops (idev->dev); +} +static void __iommufd_hwpt_destroy (struct iommufd_hw_pagetable *hwpt) {if (hwpt->domain)@@ -114,7 +121,7 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas,IOMMU_HWPT_ALLOC_DIRTY_TRACKING |IOMMU_HWPT_FAULT_ID_VALID |IOMMU_HWPT_ALLOC_PASID ; -const struct iommu_ops *ops =dev_iommu_ops (idev->dev); +const struct iommu_ops *ops =get_iommu_ops (idev);struct iommufd_hwpt_paging *hwpt_paging;struct iommufd_hw_pagetable *hwpt;int rc;@@ -229,7 +236,7 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, struct iommufd_device *idev, u32 flags, const struct iommu_user_data *user_data) { - const struct iommu_ops *ops = dev_iommu_ops(idev->dev); + const struct iommu_ops *ops = get_iommu_ops(idev);struct iommufd_hwpt_nested *hwpt_nested;struct iommufd_hw_pagetable *hwpt;int rc;drivers/iommu/iommufd/hwpt_noiommu.c new file +75
-0@@ -0,0 +1,91 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* +* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES +*/ +#include <linux/iommu.h> +#include <linux/generic_pt/iommu.h> +#include "iommufd_private.h" + +static const struct iommu_domain_ops noiommu_amdv1_ops; + +struct noiommu_domain { +union { +struct iommu_domain domain; +struct pt_iommu_amdv1 amdv1; + }; +spinlock_t lock; +}; +PT_IOMMU_CHECK_DOMAIN(struct noiommu_domain, amdv1.iommu, domain); + +static void noiommu_change_top (struct pt_iommu *iommu_table, +phys_addr_t top_paddr,unsigned int top_level) +{ +} + +static spinlock_t *noiommu_get_top_lock (struct pt_iommu *iommupt) +{ +struct noiommu_domain *domain = + container_of(iommupt, struct noiommu_domain, amdv1.iommu); + +return &domain->lock; +} + +static const struct pt_iommu_driver_ops noiommu_driver_ops = { + .get_top_lock = noiommu_get_top_lock, + .change_top = noiommu_change_top, +}; + +static struct iommu_domain * +noiommu_alloc_paging_flags (struct device *dev,u32 flags, +const struct iommu_user_data *user_data) +{ +struct pt_iommu_amdv1_cfg cfg = {}; +struct noiommu_domain *dom; +int rc; + +if (flags || user_data) +return ERR_PTR (-EOPNOTSUPP ); + + cfg.common.hw_max_vasz_lg2 =64 ; + cfg.common.hw_max_oasz_lg2 =52 ; + cfg.starting_level =2 ; + cfg.common.features = + (BIT (PT_FEAT_DYNAMIC_TOP ) |BIT (PT_FEAT_AMDV1_ENCRYPT_TABLES ) | +BIT (PT_FEAT_AMDV1_FORCE_COHERENCE )); + + dom =kzalloc (sizeof (*dom),GFP_KERNEL ); +if (!dom) +return ERR_PTR (-ENOMEM ); + +spin_lock_init (&dom->lock); + dom->amdv1.iommu.nid =NUMA_NO_NODE ; + dom->amdv1.iommu.driver_ops = &noiommu_driver_ops; + dom->domain.ops = &noiommu_amdv1_ops; + +/* Use mock page table which is based on AMDV1 */ + rc =pt_iommu_amdv1_init (&dom->amdv1, &cfg,GFP_KERNEL ); +if (rc) { +kfree (dom); +return ERR_PTR (rc); + } + +return &dom->domain; +} + +static void noiommu_domain_free (struct iommu_domain *iommu_domain)
drivers/iommu/iommufd/hwpt_noiommu.c line 85
Quoted 10 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +10
-0+{ +struct noiommu_domain *domain = + container_of(iommu_domain, struct noiommu_domain, domain); + +pt_iommu_deinit (&domain->amdv1.iommu); +kfree (domain); +} + +static const struct iommu_domain_ops noiommu_amdv1_ops = { +IOMMU_PT_DOMAIN_OPS (amdv1),
I see the appeal of re-using an existing page table implementation to keep track of iovas which -as far as I understand- are used as tokens for DMA pinned pages later, but maybe at least add some paragraph about that, as it is not immediately clear and that's a different design from the legacy noiommu VFIO code.
Thanks,
Mostafa
Quoted 22 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +6
-0+ .free = noiommu_domain_free, +}; + +struct iommu_ops iommufd_noiommu_ops = { + .domain_alloc_paging_flags = noiommu_alloc_paging_flags, +};drivers/iommu/iommufd/iommufd_private.h +2
-0@@ -464,6 +464,8 @@ static inline void iommufd_hw_pagetable_put(struct iommufd_ctx *ictx,refcount_dec (&hwpt->obj.users); } +extern struct iommu_ops iommufd_noiommu_ops; +struct iommufd_attach ;struct iommufd_group {Signature · 1 line
2.34.1
Re: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
Hi Mostafa,
Quoted 158 lines · On Sun, 22 Mar 2026 09:24:37 +0000 Mostafa Saleh <[email protected]> wrote:
Quoted 157 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
Create just a little part of a real iommu driver, enough to slot in under the
dev_iommu_ops()and allow iommufd to calldomain_alloc_paging_flags()and fail everything else.This allows explicitly creating a HWPT under an IOAS.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/hw_pagetable.c | 11 ++- drivers/iommu/iommufd/hwpt_noiommu.c | 91 +++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 2 + 4 files changed, 103 insertions(+), 2 deletions(-) create mode 100644 drivers/iommu/iommufd/hwpt_noiommu.c(quoted diff) +0
-0b/drivers/iommu/iommufd/Makefile index
71d692c9a8f4..2b1a020b14a6100644 --- a/drivers/iommu/iommufd/Makefile
+++ b/drivers/iommu/iommufd/Makefiledrivers/iommu/iommufd/Makefile +1
-0@@ -10,6 +10,7 @@ iommufd-y := \ vfio_compat.o \ viommu.o +iommufd-$(CONFIG_VFIO_NOIOMMU) += hwpt_noiommu.o iommufd-$(CONFIG_IOMMUFD_TEST) += selftest.o obj-$(CONFIG_IOMMUFD) += iommufd.o(quoted diff) +0
-0
b/drivers/iommu/iommufd/hw_pagetable.cindexfe789c2dc0c9..37316d77277d100644 ---a/drivers/iommu/iommufd/hw_pagetable.c+++b/drivers/iommu/iommufd/hw_pagetable.c@@ -8,6 +8,13 @@
#include "../iommu-priv.h"
#include "iommufd_private.h"+static const struct
iommu_ops*get_iommu_ops(structiommufd_device*idev) +{(quoted diff) +1
-0+ if (IS_ENABLED(CONFIG_VFIO_NOIOMMU) &&!idev->igroup->group)
drivers/iommu/iommufd/hw_pagetable.c +4
-0+return &iommufd_noiommu_ops; +return dev_iommu_ops (idev->dev); +} + static void __iommufd_hwpt_destroy(struct iommufd_hw_pagetable*hwpt) {
drivers/iommu/iommufd/hw_pagetable.c +2
-2if (hwpt->domain)@@ -114,7 +121,7 @@ iommufd_hwpt_paging_alloc(struct iommufd_ctx *ictx, struct iommufd_ioas *ioas, IOMMU_HWPT_ALLOC_DIRTY_TRACKING |IOMMU_HWPT_FAULT_ID_VALID |IOMMU_HWPT_ALLOC_PASID ; -const struct iommu_ops *ops =dev_iommu_ops (idev->dev); +const struct iommu_ops *ops =get_iommu_ops (idev);struct iommufd_hwpt_paging *hwpt_paging;struct iommufd_hw_pagetable *hwpt;int rc;@@ -229,7 +236,7 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, struct iommufd_device *idev, u32 flags, const struct iommu_user_data *user_data) { -const struct iommu_ops *ops =dev_iommu_ops (idev->dev); +const struct iommu_ops *ops =get_iommu_ops (idev);struct iommufd_hwpt_nested *hwpt_nested;struct iommufd_hw_pagetable *hwpt;int rc;(quoted diff) +0
-0
b/drivers/iommu/iommufd/hwpt_noiommu.cnew file mode 100644 index 000000000000..0aa99f581ca3drivers/iommu/iommufd/hwpt_noiommu.c new file +21
-0@@ -0,0 +1,91 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* +* Copyright (c) 2021-2022, NVIDIA CORPORATION & AFFILIATES +*/ +#include <linux/iommu.h> +#include <linux/generic_pt/iommu.h> +#include "iommufd_private.h" + +static const struct iommu_domain_ops noiommu_amdv1_ops; + +struct noiommu_domain { +union { +struct iommu_domain domain; +struct pt_iommu_amdv1 amdv1; + }; +spinlock_t lock; +}; +PT_IOMMU_CHECK_DOMAIN(struct noiommu_domain, amdv1.iommu, domain); + +static void noiommu_change_top (struct pt_iommu *iommu_table, + phys_addr_t top_paddr, unsigned int
top_level) +{drivers/iommu/iommufd/hwpt_noiommu.c new file +6
-0+} + +static spinlock_t *noiommu_get_top_lock (struct pt_iommu *iommupt) +{ +struct noiommu_domain *domain = + container_of(iommupt, struct noiommu_domain,amdv1.iommu); +
drivers/iommu/iommufd/hwpt_noiommu.c new file +24
-0+return &domain->lock; +} + +static const struct pt_iommu_driver_ops noiommu_driver_ops = { + .get_top_lock = noiommu_get_top_lock, + .change_top = noiommu_change_top, +}; + +static struct iommu_domain * +noiommu_alloc_paging_flags (struct device *dev,u32 flags, +const struct iommu_user_data *user_data) +{ +struct pt_iommu_amdv1_cfg cfg = {}; +struct noiommu_domain *dom; +int rc; + +if (flags || user_data) +return ERR_PTR (-EOPNOTSUPP ); + + cfg.common.hw_max_vasz_lg2 =64 ; + cfg.common.hw_max_oasz_lg2 =52 ; + cfg.starting_level =2 ; + cfg.common.features = + (BIT(PT_FEAT_DYNAMIC_TOP) |
BIT(PT_FEAT_AMDV1_ENCRYPT_TABLES)|drivers/iommu/iommufd/hwpt_noiommu.c new file +25
-0+ BIT(PT_FEAT_AMDV1_FORCE_COHERENCE)); + + dom =kzalloc (sizeof (*dom),GFP_KERNEL ); +if (!dom) +return ERR_PTR (-ENOMEM ); + +spin_lock_init (&dom->lock); + dom->amdv1.iommu.nid =NUMA_NO_NODE ; + dom->amdv1.iommu.driver_ops = &noiommu_driver_ops; + dom->domain.ops = &noiommu_amdv1_ops; + +/* Use mock page table which is based on AMDV1 */ + rc =pt_iommu_amdv1_init (&dom->amdv1, &cfg,GFP_KERNEL ); +if (rc) { +kfree (dom); +return ERR_PTR (rc); + } + +return &dom->domain; +} + +static void noiommu_domain_free (struct iommu_domain *iommu_domain) +{ +struct noiommu_domain *domain = + container_of(iommu_domain, struct noiommu_domain,domain); +
drivers/iommu/iommufd/hwpt_noiommu.c new file +4
-0+pt_iommu_deinit (&domain->amdv1.iommu); +kfree (domain); +} +
Quoted 9 lines
drivers/iommu/iommufd/hwpt_noiommu.c line 85
Quoted 2 lines
drivers/iommu/iommufd/hwpt_noiommu.c new file +2
-0+static const struct iommu_domain_ops noiommu_amdv1_ops = { +IOMMU_PT_DOMAIN_OPS (amdv1),I see the appeal of re-using an existing page table implementation to keep track of iovas which -as far as I understand- are used as tokens for DMA pinned pages later, but maybe at least add some paragraph about that, as it is not immediately clear and that's a different design from the legacy noiommu VFIO code.
Indeed it is a little confusing where we use the same VFIO noiommu knobs but with extended set of features. The legacy VFIO noiommu mode does not support container/IOAS level APIs thus no need for domain ops.
I also tried to explain the new design in the doc patch[11/11] with summaries of API limitations between legacy VFIO noiommu mode and this new mode under iommufd.
+-------------------+---------------------+---------------------+ | Feature | VFIO group | VFIO device cdev | +===================+=====================+=====================+ | VFIO device UAPI | Yes | Yes | +-------------------+---------------------+---------------------+ | VFIO container | No | No | +-------------------+---------------------+---------------------+ | IOMMUFD IOAS | No | Yes* | +-------------------+---------------------+---------------------+
How about adding the following comments:
@@ -81,6 +81,17 @@ static void noiommu_domain_free(struct iommu_domain *iommu_domain)
kfree(domain);
}
+/*
+ * AMDV1 is used as a dummy page table for no-IOMMU mode, similar to the
+ * iommufd selftest mock page table.
+ * Unlike legacy VFIO no-IOMMU mode, where no container level APIs are
+ * supported, this allows IOAS and hwpt objects to exist without hardware
+ * IOMMU support. IOVAs are used only for IOVA-to-PA lookups not for
+ * hardware translation in DMA.
+ *
+ * This is only used with iommufd and cdev-based interfaces and does not
+ * apply to legacy VFIO group-container based noiommu mode.
+ */
static const struct iommu_domain_ops noiommu_amdv1_ops = {
IOMMU_PT_DOMAIN_OPS(amdv1),Jason Gunthorpe
maintainerRe: [PATCH V2 01/11] iommufd: Support a HWPT without an iommu driver for noiommu
Quoted 4 lines · Jacob Pan
+/*
+ * AMDV1 is used as a dummy page table for no-IOMMU mode, similar to the
+ * iommufd selftest mock page table.
+ * Unlike legacy VFIO no-IOMMU mode, where no container level APIs are
Quoted 8 lines
+ * supported, this allows IOAS and hwpt objects to exist without hardware
+ * IOMMU support. IOVAs are used only for IOVA-to-PA lookups not for
+ * hardware translation in DMA.
+ *
+ * This is only used with iommufd and cdev-based interfaces and does not
+ * apply to legacy VFIO group-container based noiommu mode.
+ */
static const structiommu_domain_opsnoiommu_amdv1_ops= {
That seems clear
Jason
PATCH v2 2/11 iommufd: Move igroup allocation to a function
In reply to Jacob PanFrom: Jason Gunthorpe <[email protected]>
So it can be reused in the next patch which allows binding to noiommu device.
--- drivers/iommu/iommufd/device.c | 48 +++++++++++++++++++++------------- 1 file changed, 30 insertions(+), 18 deletions(-)
drivers/iommu/iommufd/device.c +30 -18
@@ -30,8 +30,9 @@ static void iommufd_group_release(struct kref *kref)WARN_ON (!xa_empty (&igroup->pasid_attach)); -xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), igroup, -NULL ,GFP_KERNEL ); +if (igroup->group) +xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), + igroup,NULL ,GFP_KERNEL );iommu_group_put (igroup->group);mutex_destroy (&igroup->lock);kfree (igroup);
@@ -56,6 +57,30 @@ static bool iommufd_group_try_get(struct iommufd_group *igroup,return kref_get_unless_zero (&igroup->ref); } +static struct iommufd_group *iommufd_alloc_group (struct iommufd_ctx *ictx, +struct iommu_group *group) +{ +struct iommufd_group *new_igroup; + + new_igroup =kzalloc (sizeof (*new_igroup),GFP_KERNEL ); +if (!new_igroup) +return ERR_PTR (-ENOMEM ); + +kref_init (&new_igroup->ref); +mutex_init (&new_igroup->lock); +xa_init (&new_igroup->pasid_attach); + new_igroup->sw_msi_start =PHYS_ADDR_MAX ; +/* group reference moves into new_igroup */ + new_igroup->group = group; + +/* +* The ictx is not additionally refcounted here becase all objects using +* an igroup must put it before their destroy completes. +*/ + new_igroup->ictx = ictx; +return new_igroup; +} +/* * iommufd needs to store some more data for each iommu_group, we keep a * parallel xarray indexed by iommu_group id to hold this instead of putting it
@@ -87,25 +112,12 @@ static struct iommufd_group *iommufd_get_group(struct iommufd_ctx *ictx, }xa_unlock (&ictx->groups); - new_igroup =kzalloc_obj (*new_igroup); -if (!new_igroup) { + new_igroup =iommufd_alloc_group (ictx, group); +if (IS_ERR (new_igroup)) {iommu_group_put (group); -return ERR_PTR (-ENOMEM ); +return new_igroup; } -kref_init (&new_igroup->ref); -mutex_init (&new_igroup->lock); -xa_init (&new_igroup->pasid_attach); - new_igroup->sw_msi_start =PHYS_ADDR_MAX ; -/* group reference moves into new_igroup */ - new_igroup->group = group; - -/* -* The ictx is not additionally refcounted here becase all objects using -* an igroup must put it before their destroy completes. -*/ - new_igroup->ictx = ictx; -/* * We dropped the lock so igroup is invalid. NULL is a safe and likely * value to assume for the xa_cmpxchg algorithm.
Signature · 1 line
2.34.1
Re: [PATCH V2 02/11] iommufd: Move igroup allocation to a function
Quoted 90 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
So it can be reused in the next patch which allows binding to noiommu device.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>---
drivers/iommu/iommufd/device.c| 48 +++++++++++++++++++++-------------1 file changed, 30 insertions(+), 18 deletions(-)drivers/iommu/iommufd/device.c +30
-18@@ -30,8 +30,9 @@ static void iommufd_group_release(struct kref *kref)WARN_ON (!xa_empty (&igroup->pasid_attach)); -xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), igroup, -NULL ,GFP_KERNEL ); +if (igroup->group) +xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), + igroup,NULL ,GFP_KERNEL );iommu_group_put (igroup->group);mutex_destroy (&igroup->lock);kfree (igroup);@@ -56,6 +57,30 @@ static bool iommufd_group_try_get(struct iommufd_group *igroup,return kref_get_unless_zero (&igroup->ref); } +static struct iommufd_group *iommufd_alloc_group (struct iommufd_ctx *ictx, +struct iommu_group *group) +{ +struct iommufd_group *new_igroup; + + new_igroup =kzalloc (sizeof (*new_igroup),GFP_KERNEL ); +if (!new_igroup) +return ERR_PTR (-ENOMEM ); + +kref_init (&new_igroup->ref); +mutex_init (&new_igroup->lock); +xa_init (&new_igroup->pasid_attach); + new_igroup->sw_msi_start =PHYS_ADDR_MAX ; +/* group reference moves into new_igroup */ + new_igroup->group = group; + +/* +* The ictx is not additionally refcounted here becase all objects using +* an igroup must put it before their destroy completes. +*/ + new_igroup->ictx = ictx; +return new_igroup; +} +/* * iommufd needs to store some more data for each iommu_group, we keep a * parallel xarray indexed by iommu_group id to hold this instead of putting it @@ -87,25 +112,12 @@ static struct iommufd_group *iommufd_get_group(struct iommufd_ctx *ictx, }xa_unlock (&ictx->groups); - new_igroup =kzalloc_obj (*new_igroup); -if (!new_igroup) { + new_igroup =iommufd_alloc_group (ictx, group); +if (IS_ERR (new_igroup)) {iommu_group_put (group); -return ERR_PTR (-ENOMEM ); +return new_igroup; } -kref_init (&new_igroup->ref); -mutex_init (&new_igroup->lock); -xa_init (&new_igroup->pasid_attach); - new_igroup->sw_msi_start =PHYS_ADDR_MAX ; -/* group reference moves into new_igroup */ - new_igroup->group = group; - -/* -* The ictx is not additionally refcounted here becase all objects using -* an igroup must put it before their destroy completes. -*/ - new_igroup->ictx = ictx; -/* * We dropped the lock so igroup is invalid. NULL is a safe and likely * value to assume for the xa_cmpxchg algorithm. Signature · 1 line
2.34.1
Thanks,
Sami
Re: [PATCH V2 02/11] iommufd: Move igroup allocation to a function
In reply to Jacob PanQuoted 16 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
So it can be reused in the next patch which allows binding to noiommu device.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/device.c | 48 +++++++++++++++++++++------------- 1 file changed, 30 insertions(+), 18 deletions(-)drivers/iommu/iommufd/device.c +0
-0
drivers/iommu/iommufd/device.c static void iommufd_group_release(struct kref *kref) · line 35
Quoted 8 lines
drivers/iommu/iommufd/device.c +3
-2@@ -30,8 +30,9 @@ static void iommufd_group_release(struct kref *kref)WARN_ON (!xa_empty (&igroup->pasid_attach)); -xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), igroup, -NULL ,GFP_KERNEL ); +if (igroup->group) +xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), + igroup,NULL ,GFP_KERNEL );
Is that a separate fix or perhaps belongs to the next patch making it possible to have NULL groups.
Thanks,
Mostafa
Quoted 66 lines
drivers/iommu/iommufd/device.c +27
-16iommu_group_put (igroup->group);mutex_destroy (&igroup->lock);kfree (igroup);@@ -56,6 +57,30 @@ static bool iommufd_group_try_get(struct iommufd_group *igroup,return kref_get_unless_zero (&igroup->ref); } +static struct iommufd_group *iommufd_alloc_group (struct iommufd_ctx *ictx, +struct iommu_group *group) +{ +struct iommufd_group *new_igroup; + + new_igroup =kzalloc (sizeof (*new_igroup),GFP_KERNEL ); +if (!new_igroup) +return ERR_PTR (-ENOMEM ); + +kref_init (&new_igroup->ref); +mutex_init (&new_igroup->lock); +xa_init (&new_igroup->pasid_attach); + new_igroup->sw_msi_start =PHYS_ADDR_MAX ; +/* group reference moves into new_igroup */ + new_igroup->group = group; + +/* +* The ictx is not additionally refcounted here becase all objects using +* an igroup must put it before their destroy completes. +*/ + new_igroup->ictx = ictx; +return new_igroup; +} +/* * iommufd needs to store some more data for each iommu_group, we keep a * parallel xarray indexed by iommu_group id to hold this instead of putting it @@ -87,25 +112,12 @@ static struct iommufd_group *iommufd_get_group(struct iommufd_ctx *ictx, }xa_unlock (&ictx->groups); - new_igroup =kzalloc_obj (*new_igroup); -if (!new_igroup) { + new_igroup =iommufd_alloc_group (ictx, group); +if (IS_ERR (new_igroup)) {iommu_group_put (group); -return ERR_PTR (-ENOMEM ); +return new_igroup; } -kref_init (&new_igroup->ref); -mutex_init (&new_igroup->lock); -xa_init (&new_igroup->pasid_attach); - new_igroup->sw_msi_start =PHYS_ADDR_MAX ; -/* group reference moves into new_igroup */ - new_igroup->group = group; - -/* -* The ictx is not additionally refcounted here becase all objects using -* an igroup must put it before their destroy completes. -*/ - new_igroup->ictx = ictx; -/* * We dropped the lock so igroup is invalid. NULL is a safe and likely * value to assume for the xa_cmpxchg algorithm. Signature · 1 line
2.34.1
Re: [PATCH V2 02/11] iommufd: Move igroup allocation to a function
Hi Mostafa,
Quoted 21 lines · On Sun, 22 Mar 2026 09:41:17 +0000 Mostafa Saleh <[email protected]> wrote:
Quoted 20 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
So it can be reused in the next patch which allows binding to noiommu device.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/device.c | 48+++++++++++++++++++++------------- 1 file changed, 30 insertions(+), 18 deletions(-)
(quoted diff) +0
-0
b/drivers/iommu/iommufd/device.cindex344d620cdecc..54d73016468f100644 ---a/drivers/iommu/iommufd/device.c
+++b/drivers/iommu/iommufd/device.cdrivers/iommu/iommufd/device.c +0
-0@@ -30,8 +30,9 @@ static void iommufd_group_release(struct kref *kref)WARN_ON (!xa_empty (&igroup->pasid_attach));
Quoted 10 lines
Review static void iommufd_group_release(struct kref · line 37
Quoted 7 lines
(quoted diff) +3
-2- xa_cmpxchg(&igroup->ictx->groups, iommu_group_id(igroup->group), igroup, - NULL, GFP_KERNEL); + if (igroup->group) + xa_cmpxchg(&igroup->ictx->groups, iommu_group_id(igroup->group), + igroup, NULL, GFP_KERNEL);Is that a separate fix or perhaps belongs to the next patch making it possible to have NULL groups.
Yes, I agree. NULL groups are not possible here. Should be merged with the next patch.
Re: [PATCH V2 02/11] iommufd: Move igroup allocation to a function
In reply to Jacob PanQuoted 90 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
So it can be reused in the next patch which allows binding to noiommu device.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>---
drivers/iommu/iommufd/device.c| 48 +++++++++++++++++++++-------------1 file changed, 30 insertions(+), 18 deletions(-)drivers/iommu/iommufd/device.c +30
-18@@ -30,8 +30,9 @@ static void iommufd_group_release(struct kref *kref)WARN_ON (!xa_empty (&igroup->pasid_attach)); -xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), igroup, -NULL ,GFP_KERNEL ); +if (igroup->group) +xa_cmpxchg (&igroup->ictx->groups,iommu_group_id (igroup->group), + igroup,NULL ,GFP_KERNEL );iommu_group_put (igroup->group);mutex_destroy (&igroup->lock);kfree (igroup);@@ -56,6 +57,30 @@ static bool iommufd_group_try_get(struct iommufd_group *igroup,return kref_get_unless_zero (&igroup->ref); } +static struct iommufd_group *iommufd_alloc_group (struct iommufd_ctx *ictx, +struct iommu_group *group) +{ +struct iommufd_group *new_igroup; + + new_igroup =kzalloc (sizeof (*new_igroup),GFP_KERNEL ); +if (!new_igroup) +return ERR_PTR (-ENOMEM ); + +kref_init (&new_igroup->ref); +mutex_init (&new_igroup->lock); +xa_init (&new_igroup->pasid_attach); + new_igroup->sw_msi_start =PHYS_ADDR_MAX ; +/* group reference moves into new_igroup */ + new_igroup->group = group; + +/* +* The ictx is not additionally refcounted here becase all objects using +* an igroup must put it before their destroy completes. +*/ + new_igroup->ictx = ictx; +return new_igroup; +} +/* * iommufd needs to store some more data for each iommu_group, we keep a * parallel xarray indexed by iommu_group id to hold this instead of putting it @@ -87,25 +112,12 @@ static struct iommufd_group *iommufd_get_group(struct iommufd_ctx *ictx, }xa_unlock (&ictx->groups); - new_igroup =kzalloc_obj (*new_igroup); -if (!new_igroup) { + new_igroup =iommufd_alloc_group (ictx, group); +if (IS_ERR (new_igroup)) {iommu_group_put (group); -return ERR_PTR (-ENOMEM ); +return new_igroup; } -kref_init (&new_igroup->ref); -mutex_init (&new_igroup->lock); -xa_init (&new_igroup->pasid_attach); - new_igroup->sw_msi_start =PHYS_ADDR_MAX ; -/* group reference moves into new_igroup */ - new_igroup->group = group; - -/* -* The ictx is not additionally refcounted here becase all objects using -* an igroup must put it before their destroy completes. -*/ - new_igroup->ictx = ictx; -/* * We dropped the lock so igroup is invalid. NULL is a safe and likely * value to assume for the xa_cmpxchg algorithm. Signature · 1 line
2.34.1
PATCH v2 3/11 iommufd: Allow binding to a noiommu device
In reply to Jacob PanFrom: Jason Gunthorpe <[email protected]>
Allow iommufd to bind devices without an IOMMU (noiommu mode) by creating a dummy IOMMU group for such devices and skipping hwpt operations.
This enables noiommu devices to operate through the same iommufd API as IOMMU- capable devices.
--- drivers/iommu/iommufd/device.c | 113 ++++++++++++++++++++++----------- 1 file changed, 76 insertions(+), 37 deletions(-)
drivers/iommu/iommufd/device.c +76 -37
@@ -23,6 +23,11 @@ struct iommufd_attach {struct xarray device_array; }; +static bool is_vfio_noiommu (struct iommufd_device *idev) +{ +return !device_iommu_mapped (idev->dev) || !idev->dev->iommu; +} +static void iommufd_group_release (struct kref *kref) { struct iommufd_group *igroup =
@@ -205,32 +210,17 @@ void iommufd_device_destroy(struct iommufd_object *obj)struct iommufd_device *idev = container_of(obj, struct iommufd_device, obj); -iommu_device_release_dma_owner (idev->dev); +if (!is_vfio_noiommu (idev)) +iommu_device_release_dma_owner (idev->dev);iommufd_put_group (idev->igroup);if (!iommufd_selftest_is_mock_dev (idev->dev))iommufd_ctx_put (idev->ictx); } -/** -* iommufd_device_bind - Bind a physical device to an iommu fd -* @ictx: iommufd file descriptor -* @dev: Pointer to a physical device struct -* @id: Output ID number to return to userspace for this device -* -* A successful bind establishes an ownership over the device and returns -* struct iommufd_device pointer, otherwise returns error pointer. -* -* A driver using this API must set driver_managed_dma and must not touch -* the device until this routine succeeds and establishes ownership. -* -* Binding a PCI device places the entire RID under iommufd control. -* -* The caller must undo this with iommufd_device_unbind() -*/ -struct iommufd_device *iommufd_device_bind (struct iommufd_ctx *ictx, -struct device *dev,u32 *id) +static int iommufd_bind_iommu (struct iommufd_device *idev) { -struct iommufd_device *idev; +struct iommufd_ctx *ictx = idev->ictx; +struct device *dev = idev->dev;struct iommufd_group *igroup;int rc;
@@ -239,11 +229,11 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx,* to restore cache coherency. */ if (!device_iommu_capable (dev,IOMMU_CAP_CACHE_COHERENCY )) -return ERR_PTR (-EINVAL ); +return -EINVAL ; - igroup =iommufd_get_group (ictx, dev); + igroup =iommufd_get_group (idev->ictx, dev);if (IS_ERR (igroup)) -return ERR_CAST (igroup); +return PTR_ERR (igroup);/* * For historical compat with VFIO the insecure interrupt path is
@@ -269,21 +259,66 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx,if (rc) goto out_group_put; +/* igroup refcount moves into iommufd_device */ + idev->igroup = igroup; +return 0 ; + +out_group_put: +iommufd_put_group (igroup); +return rc; +} + +/** +* iommufd_device_bind - Bind a physical device to an iommu fd +* @ictx: iommufd file descriptor +* @dev: Pointer to a physical device struct +* @id: Output ID number to return to userspace for this device +* +* A successful bind establishes an ownership over the device and returns +* struct iommufd_device pointer, otherwise returns error pointer. +* +* A driver using this API must set driver_managed_dma and must not touch +* the device until this routine succeeds and establishes ownership. +* +* Binding a PCI device places the entire RID under iommufd control. +* +* The caller must undo this with iommufd_device_unbind() +*/ +struct iommufd_device *iommufd_device_bind (struct iommufd_ctx *ictx, +struct device *dev,u32 *id) +{ +struct iommufd_device *idev; +int rc; + idev =iommufd_object_alloc (ictx, idev,IOMMUFD_OBJ_DEVICE ); -if (IS_ERR (idev)) { - rc =PTR_ERR (idev); - goto out_release_owner; - } +if (IS_ERR (idev)) +return idev; idev->ictx = ictx; -if (!iommufd_selftest_is_mock_dev (dev)) -iommufd_ctx_get (ictx); idev->dev = dev; idev->enforce_cache_coherency =device_iommu_capable (dev,IOMMU_CAP_ENFORCE_CACHE_COHERENCY ); + +if (!is_vfio_noiommu (idev)) { + rc =iommufd_bind_iommu (idev); +if (rc) +return ERR_PTR (rc); + }else { +struct iommufd_group *igroup; + +/* +* Create a dummy igroup, lots of stuff expects ths igroup to be +* present, but a NULL igroup->group is OK +*/ + igroup =iommufd_alloc_group (ictx,NULL ); +if (IS_ERR (igroup)) +return ERR_CAST (igroup); + idev->igroup = igroup; + } + +if (!iommufd_selftest_is_mock_dev (dev)) +iommufd_ctx_get (ictx);/* The calling driver is a user until iommufd_device_unbind() */ refcount_inc (&idev->obj.users); -/* igroup refcount moves into iommufd_device */ - idev->igroup = igroup;/* * If the caller fails after this success it must call
@@ -295,11 +330,6 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, *id = idev->obj.id;return idev; -out_release_owner: -iommu_device_release_dma_owner (dev); -out_group_put: -iommufd_put_group (igroup); -return ERR_PTR (rc); }EXPORT_SYMBOL_NS_GPL (iommufd_device_bind,"IOMMUFD" );
@@ -513,6 +543,9 @@ static int iommufd_hwpt_attach_device(struct iommufd_hw_pagetable *hwpt,struct iommufd_attach_handle *handle;int rc; +if (is_vfio_noiommu (idev)) +return 0 ; +if (!iommufd_hwpt_compatible_device (hwpt, idev))return -EINVAL ;
@@ -560,6 +593,9 @@ static void iommufd_hwpt_detach_device(struct iommufd_hw_pagetable *hwpt, {struct iommufd_attach_handle *handle; +if (is_vfio_noiommu (idev)) +return ; + handle =iommufd_device_get_attach_handle (idev, pasid);if (pasid ==IOMMU_NO_PASID )iommu_detach_group_handle (hwpt->domain, idev->igroup->group);
@@ -578,6 +614,9 @@ static int iommufd_hwpt_replace_device(struct iommufd_device *idev,struct iommufd_attach_handle *handle, *old_handle;int rc; +if (is_vfio_noiommu (idev)) +return 0 ; +if (!iommufd_hwpt_compatible_device (hwpt, idev))return -EINVAL ;
@@ -653,7 +692,7 @@ int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt, goto err_release_devid; } -if (attach_resv) { +if (attach_resv && !is_vfio_noiommu (idev)) { rc =iommufd_device_attach_reserved_iova (idev, hwpt_paging);if (rc) goto err_release_devid;
Signature · 1 line
2.34.1
Re: [PATCH V2 03/11] iommufd: Allow binding to a noiommu device
Quoted 19 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
Allow iommufd to bind devices without an IOMMU (noiommu mode) by creating a dummy IOMMU group for such devices and skipping hwpt operations.
This enables noiommu devices to operate through the same iommufd API as IOMMU- capable devices.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/device.c | 113 ++++++++++++++++++++++----------- 1 file changed, 76 insertions(+), 37 deletions(-)drivers/iommu/iommufd/device.c +0
-0
drivers/iommu/iommufd/device.c struct iommufd_attach { · line 28
Quoted 6 lines
drivers/iommu/iommufd/device.c +3
-0@@ -23,6 +23,11 @@ struct iommufd_attach {struct xarray device_array; }; +static bool is_vfio_noiommu (struct iommufd_device *idev) +{ +return !device_iommu_mapped (idev->dev) || !idev->dev->iommu;
Do this need to check for CONFIG_VFIO_NOIOMMU and maybe the module param enable_unsafe_noiommu_mode similar to the legacy implemenation?
Quoted 84 lines
drivers/iommu/iommufd/device.c +32
-23+} +static void iommufd_group_release (struct kref *kref) { struct iommufd_group *igroup =@@ -205,32 +210,17 @@ void iommufd_device_destroy(struct iommufd_object *obj)struct iommufd_device *idev = container_of(obj, struct iommufd_device, obj); -iommu_device_release_dma_owner (idev->dev); +if (!is_vfio_noiommu (idev)) +iommu_device_release_dma_owner (idev->dev);iommufd_put_group (idev->igroup);if (!iommufd_selftest_is_mock_dev (idev->dev))iommufd_ctx_put (idev->ictx); } -/** -* iommufd_device_bind - Bind a physical device to an iommu fd -* @ictx: iommufd file descriptor -* @dev: Pointer to a physical device struct -* @id: Output ID number to return to userspace for this device -* -* A successful bind establishes an ownership over the device and returns -* struct iommufd_device pointer, otherwise returns error pointer. -* -* A driver using this API must set driver_managed_dma and must not touch -* the device until this routine succeeds and establishes ownership. -* -* Binding a PCI device places the entire RID under iommufd control. -* -* The caller must undo this with iommufd_device_unbind() -*/ -struct iommufd_device *iommufd_device_bind (struct iommufd_ctx *ictx, -struct device *dev,u32 *id) +static int iommufd_bind_iommu (struct iommufd_device *idev) { -struct iommufd_device *idev; +struct iommufd_ctx *ictx = idev->ictx; +struct device *dev = idev->dev;struct iommufd_group *igroup;int rc;@@ -239,11 +229,11 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx,* to restore cache coherency. */ if (!device_iommu_capable (dev,IOMMU_CAP_CACHE_COHERENCY )) -return ERR_PTR (-EINVAL ); +return -EINVAL ; - igroup =iommufd_get_group (ictx, dev); + igroup =iommufd_get_group (idev->ictx, dev);if (IS_ERR (igroup)) -return ERR_CAST (igroup); +return PTR_ERR (igroup);/* * For historical compat with VFIO the insecure interrupt path is @@ -269,21 +259,66 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx,if (rc) goto out_group_put; +/* igroup refcount moves into iommufd_device */ + idev->igroup = igroup; +return 0 ; + +out_group_put: +iommufd_put_group (igroup); +return rc; +} + +/** +* iommufd_device_bind - Bind a physical device to an iommu fd +* @ictx: iommufd file descriptor +* @dev: Pointer to a physical device struct +* @id: Output ID number to return to userspace for this device +* +* A successful bind establishes an ownership over the device and returns +* struct iommufd_device pointer, otherwise returns error pointer. +* +* A driver using this API must set driver_managed_dma and must not touch +* the device until this routine succeeds and establishes ownership. +* +* Binding a PCI device places the entire RID under iommufd control.
drivers/iommu/iommufd/device.c struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, · line 293
Quoted 10 lines
drivers/iommu/iommufd/device.c +9
-0+* +* The caller must undo this with iommufd_device_unbind() +*/ +struct iommufd_device *iommufd_device_bind (struct iommufd_ctx *ictx, +struct device *dev,u32 *id) +{ +struct iommufd_device *idev; +int rc; + idev =iommufd_object_alloc (ictx, idev,IOMMUFD_OBJ_DEVICE );
The next code introduces new error cases, do that need to be cleaned in that case by calling iommufd_object_abort_and_destroy()?
Thanks,
Mostafa
Quoted 93 lines
drivers/iommu/iommufd/device.c +32
-14-if (IS_ERR (idev)) { - rc =PTR_ERR (idev); - goto out_release_owner; - } +if (IS_ERR (idev)) +return idev; idev->ictx = ictx; -if (!iommufd_selftest_is_mock_dev (dev)) -iommufd_ctx_get (ictx); idev->dev = dev; idev->enforce_cache_coherency =device_iommu_capable (dev,IOMMU_CAP_ENFORCE_CACHE_COHERENCY ); + +if (!is_vfio_noiommu (idev)) { + rc =iommufd_bind_iommu (idev); +if (rc) +return ERR_PTR (rc); + }else { +struct iommufd_group *igroup; + +/* +* Create a dummy igroup, lots of stuff expects ths igroup to be +* present, but a NULL igroup->group is OK +*/ + igroup =iommufd_alloc_group (ictx,NULL ); +if (IS_ERR (igroup)) +return ERR_CAST (igroup); + idev->igroup = igroup; + } + +if (!iommufd_selftest_is_mock_dev (dev)) +iommufd_ctx_get (ictx);/* The calling driver is a user until iommufd_device_unbind() */ refcount_inc (&idev->obj.users); -/* igroup refcount moves into iommufd_device */ - idev->igroup = igroup;/* * If the caller fails after this success it must call @@ -295,11 +330,6 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, *id = idev->obj.id;return idev; -out_release_owner: -iommu_device_release_dma_owner (dev); -out_group_put: -iommufd_put_group (igroup); -return ERR_PTR (rc); }EXPORT_SYMBOL_NS_GPL (iommufd_device_bind,"IOMMUFD" );@@ -513,6 +543,9 @@ static int iommufd_hwpt_attach_device(struct iommufd_hw_pagetable *hwpt,struct iommufd_attach_handle *handle;int rc; +if (is_vfio_noiommu (idev)) +return 0 ; +if (!iommufd_hwpt_compatible_device (hwpt, idev))return -EINVAL ;@@ -560,6 +593,9 @@ static void iommufd_hwpt_detach_device(struct iommufd_hw_pagetable *hwpt, {struct iommufd_attach_handle *handle; +if (is_vfio_noiommu (idev)) +return ; + handle =iommufd_device_get_attach_handle (idev, pasid);if (pasid ==IOMMU_NO_PASID )iommu_detach_group_handle (hwpt->domain, idev->igroup->group);@@ -578,6 +614,9 @@ static int iommufd_hwpt_replace_device(struct iommufd_device *idev,struct iommufd_attach_handle *handle, *old_handle;int rc; +if (is_vfio_noiommu (idev)) +return 0 ; +if (!iommufd_hwpt_compatible_device (hwpt, idev))return -EINVAL ;@@ -653,7 +692,7 @@ int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt, goto err_release_devid; } -if (attach_resv) { +if (attach_resv && !is_vfio_noiommu (idev)) { rc =iommufd_device_attach_reserved_iova (idev, hwpt_paging);if (rc) goto err_release_devid;Signature · 1 line
2.34.1
Jason Gunthorpe
maintainerRe: [PATCH V2 03/11] iommufd: Allow binding to a noiommu device
Quoted 10 lines · Mostafa Saleh
drivers/iommu/iommufd/device.c struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, · line 293
Quoted 7 lines
drivers/iommu/iommufd/device.c +6
-0+struct iommufd_device *iommufd_device_bind (struct iommufd_ctx *ictx, +struct device *dev,u32 *id) +{ +struct iommufd_device *idev; +int rc; + idev =iommufd_object_alloc (ictx, idev,IOMMUFD_OBJ_DEVICE );The next code introduces new error cases, do that need to be cleaned in that case by calling
iommufd_object_abort_and_destroy()?
It should probably use iommufd_object_alloc_ucmd() so the core code manages the lifecycle?
Jason
Re: [PATCH V2 03/11] iommufd: Allow binding to a noiommu device
In reply to Mostafa SalehHi Mostafa,
Quoted 38 lines · On Sun, 22 Mar 2026 09:54:15 +0000 Mostafa Saleh <[email protected]> wrote:
From: Mostafa Saleh <[email protected]>
To: Jacob Pan <[email protected]>
Cc: [email protected], "[email protected]" <[email protected]>, Jason Gunthorpe <[email protected]>, Alex Williamson <[email protected]>, Joerg Roedel <[email protected]>, David Matlack <[email protected]>, Robin Murphy <[email protected]>, Nicolin Chen <[email protected]>, "Tian, Kevin" <[email protected]>, Yi Liu <[email protected]>, [email protected], [email protected], Will Deacon <[email protected]>, Baolu Lu <[email protected]> Subject: Re:
[PATCH V2 03/11] iommufd: Allow binding to a noiommu device Date:
Sun, 22 Mar 2026 09:54:15 +0000Quoted 24 lines · Jacob Pan
From: Jason Gunthorpe <[email protected]>
Allow iommufd to bind devices without an IOMMU (noiommu mode) by creating a dummy IOMMU group for such devices and skipping hwpt operations.
This enables noiommu devices to operate through the same iommufd API as IOMMU- capable devices.
Signed-off-by Jason Gunthorpe <[email protected]>Signed-off-by Jacob Pan <[email protected]>--- drivers/iommu/iommufd/device.c | 113++++++++++++++++++++++----------- 1 file changed, 76 insertions(+), 37 deletions(-)
(quoted diff) +0
-0
b/drivers/iommu/iommufd/device.cindex54d73016468f..c38d3efa3d6f100644 ---a/drivers/iommu/iommufd/device.c
+++b/drivers/iommu/iommufd/device.c(quoted diff) +0
-0@@ -23,6 +23,11 @@ struct iommufd_attach { struct xarray device_array; };
Quoted 8 lines
Review struct iommufd_attach { · line 29
Quoted 4 lines
(quoted diff) +3
-0+static bool is_vfio_noiommu(struct iommufd_device *idev) +{ + return !device_iommu_mapped(idev->dev) || !idev->dev->iommu;Do this need to check for
CONFIG_VFIO_NOIOMMUand maybe the module paramenable_unsafe_noiommu_modesimilar to the legacy implemenation?
Checking for CONFIG_VFIO_NOIOMMU is not needed since all the conditions are not restricted by CONFIG_VFIO_NOIOMMU. I felt it is cleaner this way by not tying iommufd private code with vfio.
I guess we could do something like below but not necessary IMHO.
drivers/iommu/iommufd/device.c +0 -5
@@ -23,11 +23,6 @@ struct iommufd_attach {struct xarray device_array; }; -static bool is_vfio_noiommu (struct iommufd_device *idev) -{ -return !device_iommu_mapped (idev->dev) || !idev->dev->iommu; -} -static void iommufd_group_release (struct kref *kref) { struct iommufd_group *igroup =
(quoted diff) +0 -0
b/drivers/iommu/iommufd/iommufd_private.h index 3302c6a1f99e..cba5550e3f2b 100644 --- a/drivers/iommu/iommufd/iommufd_private.h +++ b/drivers/iommu/iommufd/iommufd_private.h @@ -711,6 +711,18 @@ iommufd_get_vdevice(struct iommufd_ctx *ictx, u32 id) struct iommufd_vdevice, obj); }
+#ifdef CONFIG_VFIO_NOIOMMU
+static inline bool is_vfio_noiommu(struct iommufd_device *idev) +{
+ return !device_iommu_mapped(idev->dev) || !idev->dev->iommu;
+}
+#else
+static inline bool is_vfio_noiommu(struct iommufd_device *idev) +{
+ return false;
+}
+#endif
+
PATCH v2 4/11 iommufd: Add an ioctl IOMMU_IOAS_GET_PA to query PA from IOVA
In reply to Jacob PanTo support no-IOMMU mode where userspace drivers perform unsafe DMA using physical addresses, introduce a new API to retrieve the physical address of a user-allocated DMA buffer that has been mapped to an IOVA via IOAS. The mapping is backed by mock I/O page tables maintained by generic IOMMUPT framework.
---
v2:
- Scan the contiguous physical-address span beyond the first page and return its length.
--- drivers/iommu/iommufd/io_pagetable.c | 60 +++++++++++++++++++++++++ drivers/iommu/iommufd/ioas.c | 28 ++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 3 ++ drivers/iommu/iommufd/main.c | 3 ++ include/uapi/linux/iommufd.h | 25 +++++++++++ 5 files changed, 119 insertions(+)
drivers/iommu/iommufd/io_pagetable.c +60 -0
@@ -849,6 +849,66 @@ int iopt_unmap_iova(struct io_pagetable *iopt, unsigned long iova,return iopt_unmap_iova_range (iopt, iova, iova_last, unmapped); } +int iopt_get_phys (struct io_pagetable *iopt,unsigned long iova,u64 *paddr, +u64 *length) +{ +struct iopt_area *area; +u64 tmp_length =0 ; +u64 tmp_paddr =0 ; +int rc =0 ; + +if (!IS_ENABLED (CONFIG_VFIO_NOIOMMU )) +return -EOPNOTSUPP ; + +down_read (&iopt->iova_rwsem); + area =iopt_area_iter_first (iopt, iova, iova); +if (!area || !area->pages) { + rc = -ENOENT ; + goto unlock_exit; + } + +if (!area->storage_domain || + area->storage_domain->owner != &iommufd_noiommu_ops) { + rc = -EOPNOTSUPP ; + goto unlock_exit; + } + + *paddr =iommu_iova_to_phys (area->storage_domain, iova); +if (!*paddr) { + rc = -EINVAL ; + goto unlock_exit; + } + + tmp_length =PAGE_SIZE ; + tmp_paddr = *paddr; +/* +* Scan the domain for the contiguous physical address length so that +* userspace search can be optimized for fewer ioctls. +*/ +while (iova <iopt_area_last_iova (area)) { +unsigned long next_iova; +u64 next_paddr; + +if (check_add_overflow (iova,PAGE_SIZE , &next_iova)) +break ; + + next_paddr =iommu_iova_to_phys (area->storage_domain, next_iova); + +if (!next_paddr || next_paddr != tmp_paddr +PAGE_SIZE ) +break ; + + iova = next_iova; + tmp_paddr +=PAGE_SIZE ; + tmp_length +=PAGE_SIZE ; + } + *length = tmp_length; + +unlock_exit: +up_read (&iopt->iova_rwsem); + +return rc; +} +int iopt_unmap_all (struct io_pagetable *iopt,unsigned long *unmapped) {/* If the IOVAs are empty then unmap all succeeds */
drivers/iommu/iommufd/ioas.c +28 -0
@@ -375,6 +375,34 @@ int iommufd_ioas_unmap(struct iommufd_ucmd *ucmd)return rc; } +int iommufd_ioas_get_pa (struct iommufd_ucmd *ucmd) +{ +struct iommu_ioas_get_pa *cmd = ucmd->cmd; +struct iommufd_ioas *ioas; +int rc; + +if (cmd->flags || cmd->__reserved) +return -EOPNOTSUPP ; + +if (!cmd->iova || cmd->iova >=ULONG_MAX ) +return -EINVAL ; + + ioas =iommufd_get_ioas (ucmd->ictx, cmd->ioas_id); +if (IS_ERR (ioas)) +return PTR_ERR (ioas); + + rc =iopt_get_phys (&ioas->iopt, cmd->iova, &cmd->out_phys, + &cmd->out_length); +if (rc) + goto out_put; + + rc =iommufd_ucmd_respond (ucmd,sizeof (*cmd)); +out_put: +iommufd_put_object (ucmd->ictx, &ioas->obj); + +return rc; +} +static void iommufd_release_all_iova_rwsem (struct iommufd_ctx *ictx,struct xarray *ioas_list) {
drivers/iommu/iommufd/iommufd_private.h +3 -0
@@ -118,6 +118,8 @@ int iopt_map_pages(struct io_pagetable *iopt, struct list_head *pages_list,int iopt_unmap_iova (struct io_pagetable *iopt,unsigned long iova,unsigned long length,unsigned long *unmapped);int iopt_unmap_all (struct io_pagetable *iopt,unsigned long *unmapped); +int iopt_get_phys (struct io_pagetable *iopt,unsigned long iova,u64 *paddr, +u64 *length); int iopt_read_and_clear_dirty_data(struct io_pagetable *iopt, struct iommu_domain *domain,
@@ -346,6 +348,7 @@ int iommufd_ioas_map_file(struct iommufd_ucmd *ucmd);int iommufd_ioas_change_process (struct iommufd_ucmd *ucmd);int iommufd_ioas_copy (struct iommufd_ucmd *ucmd);int iommufd_ioas_unmap (struct iommufd_ucmd *ucmd); +int iommufd_ioas_get_pa (struct iommufd_ucmd *ucmd);int iommufd_ioas_option (struct iommufd_ucmd *ucmd);int iommufd_option_rlimit_mode (struct iommu_option *cmd,struct iommufd_ctx *ictx);
drivers/iommu/iommufd/main.c +3 -0
@@ -432,6 +432,7 @@ union ucmd_buffer {struct iommu_veventq_alloc veventq;struct iommu_vfio_ioas vfio_ioas;struct iommu_viommu_alloc viommu; +struct iommu_ioas_get_pa get_pa;#ifdef CONFIG_IOMMUFD_TEST struct iommu_test_cmd test;#endif
@@ -484,6 +485,8 @@ static const struct iommufd_ioctl_op iommufd_ioctl_ops[] = { struct iommu_ioas_map_file, iova),IOCTL_OP (IOMMU_IOAS_UNMAP ,iommufd_ioas_unmap ,struct iommu_ioas_unmap ,length ), +IOCTL_OP (IOMMU_IOAS_GET_PA ,iommufd_ioas_get_pa ,struct iommu_ioas_get_pa , +out_phys ),IOCTL_OP (IOMMU_OPTION ,iommufd_option ,struct iommu_option ,val64 ),IOCTL_OP (IOMMU_VDEVICE_ALLOC ,iommufd_vdevice_alloc_ioctl , struct iommu_vdevice_alloc, virt_id),
include/uapi/linux/iommufd.h +25 -0
@@ -57,6 +57,7 @@ enum {IOMMUFD_CMD_IOAS_CHANGE_PROCESS =0x92 ,IOMMUFD_CMD_VEVENTQ_ALLOC =0x93 ,IOMMUFD_CMD_HW_QUEUE_ALLOC =0x94 , + IOMMUFD_CMD_IOAS_GET_PA = 0x95, };/**
@@ -219,6 +220,30 @@ struct iommu_ioas_map { };#define IOMMU_IOAS_MAP _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_MAP) +/** +* struct iommu_ioas_get_pa - ioctl(IOMMU_IOAS_GET_PA) +* @size: sizeof(struct iommu_ioas_get_pa) +* @flags: Reserved, must be 0 for now +* @ioas_id: IOAS ID to query IOVA to PA mapping from +* @__reserved: Must be 0 +* @iova: IOVA to query +* @out_length: Number of bytes contiguous physical address starting from phys +* @out_phys: Output physical address the IOVA maps to +* +* Query the physical address backing an IOVA range. The entire range must be +* mapped already. For noiommu devices doing unsafe DMA only. +*/ +struct iommu_ioas_get_pa { +__u32 size; +__u32 flags; +__u32 ioas_id; +__u32 __reserved; +__aligned_u64 iova; +__aligned_u64 out_length; +__aligned_u64 out_phys; +}; +#define IOMMU_IOAS_GET_PA _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_GET_PA) +/** * struct iommu_ioas_map_file - ioctl(IOMMU_IOAS_MAP_FILE) * @size: sizeof(struct iommu_ioas_map_file)
Signature · 1 line
2.34.1
PATCH v2 5/11 vfio: Allow null group for noiommu without containers
In reply to Jacob PanIn case of noiommu mode is enabled for VFIO cdev without VFIO container nor IOMMUFD provided compatibility container, there is no need to create a dummy group. Update the group operations to tolerate null group pointer.
--- drivers/vfio/group.c | 14 ++++++++++++++ drivers/vfio/vfio.h | 17 +++++++++++++++++ 2 files changed, 31 insertions(+)
drivers/vfio/group.c +14 -0
@@ -381,6 +381,9 @@ int vfio_device_block_group(struct vfio_device *device)struct vfio_group *group = device->group;int ret =0 ; +if (vfio_null_group_allowed () && !group) +return 0 ; +mutex_lock (&group->group_lock);if (group->opened_file) { ret = -EBUSY ;
@@ -398,6 +401,9 @@ void vfio_device_unblock_group(struct vfio_device *device) {struct vfio_group *group = device->group; +if (vfio_null_group_allowed () && !group) +return ; +mutex_lock (&group->group_lock); group->cdev_device_open_cnt--;mutex_unlock (&group->group_lock);
@@ -589,6 +595,14 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev,struct vfio_group *group;int ret; +/* +* With noiommu enabled under cdev interface only, there is no need to +* create a vfio_group if the group based containers are not enabled. +* The cdev interface is exclusively used for iommufd. +*/ +if (vfio_null_group_allowed ()) +return NULL ; + iommu_group =iommu_group_alloc ();if (IS_ERR (iommu_group))return ERR_CAST (iommu_group);
drivers/vfio/vfio.h +17 -0
@@ -113,6 +113,18 @@ bool vfio_device_has_container(struct vfio_device *device);int __init vfio_group_init (void );void vfio_group_cleanup (void ); +/* +* With noiommu enabled and no containers are supported, allow devices that +* don't have a dummy group. +*/ +static inline bool vfio_null_group_allowed (void ) +{ +if (vfio_noiommu && (!IS_ENABLED (CONFIG_VFIO_CONTAINER ) && !IS_ENABLED (CONFIG_IOMMUFD_VFIO_CONTAINER ))) +return true; + +return false; +} +static inline bool vfio_device_is_noiommu (struct vfio_device *vdev) { return IS_ENABLED(CONFIG_VFIO_NOIOMMU) &&
@@ -189,6 +201,11 @@ static inline void vfio_group_cleanup(void) { } +static inline bool vfio_null_group_allowed (void ) +{ +return false; +} +static inline bool vfio_device_is_noiommu (struct vfio_device *vdev) {return false;
Signature · 1 line
2.34.1
Re: [PATCH V2 05/11] vfio: Allow null group for noiommu without containers
Quoted 37 lines · Jacob Pan
In case of noiommu mode is enabled for VFIO cdev without VFIO container nor IOMMUFD provided compatibility container, there is no need to create a dummy group. Update the group operations to tolerate null group pointer.
Signed-off-by Jacob Pan <[email protected]>--- drivers/vfio/group.c | 14 ++++++++++++++ drivers/vfio/vfio.h | 17 +++++++++++++++++ 2 files changed, 31 insertions(+)drivers/vfio/group.c +6
-0@@ -381,6 +381,9 @@ int vfio_device_block_group(struct vfio_device *device)struct vfio_group *group = device->group;int ret =0 ; +if (vfio_null_group_allowed () && !group) +return 0 ; +mutex_lock (&group->group_lock);if (group->opened_file) { ret = -EBUSY ;@@ -398,6 +401,9 @@ void vfio_device_unblock_group(struct vfio_device *device) {struct vfio_group *group = device->group; +if (vfio_null_group_allowed () && !group) +return ; +mutex_lock (&group->group_lock); group->cdev_device_open_cnt--;mutex_unlock (&group->group_lock);@@ -589,6 +595,14 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev,struct vfio_group *group;
drivers/vfio/group.c static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev, · line 605
Quoted 10 lines
drivers/vfio/group.c +8
-0int ret; +/* +* With noiommu enabled under cdev interface only, there is no need to +* create a vfio_group if the group based containers are not enabled. +* The cdev interface is exclusively used for iommufd. +*/ +if (vfio_null_group_allowed ()) +return NULL ; +
Now vfio_device_set_group() can return NULL when called from __vfio_register_dev() where the error path calls vfio_device_remove_group() which I believe would break.
But is that really needed, I feel like this optimization is not worth the extra effort to add those checks and the possiblity of missing some. what do you think?
Thanks,
Mostafa
Quoted 41 lines
drivers/vfio/group.c +0
-0iommu_group =iommu_group_alloc ();if (IS_ERR (iommu_group))return ERR_CAST (iommu_group);drivers/vfio/vfio.h +17
-0@@ -113,6 +113,18 @@ bool vfio_device_has_container(struct vfio_device *device);int __init vfio_group_init (void );void vfio_group_cleanup (void ); +/* +* With noiommu enabled and no containers are supported, allow devices that +* don't have a dummy group. +*/ +static inline bool vfio_null_group_allowed (void ) +{ +if (vfio_noiommu && (!IS_ENABLED (CONFIG_VFIO_CONTAINER ) && !IS_ENABLED (CONFIG_IOMMUFD_VFIO_CONTAINER ))) +return true; + +return false; +} +static inline bool vfio_device_is_noiommu (struct vfio_device *vdev) { return IS_ENABLED(CONFIG_VFIO_NOIOMMU) &&@@ -189,6 +201,11 @@ static inline void vfio_group_cleanup(void) { } +static inline bool vfio_null_group_allowed (void ) +{ +return false; +} +static inline bool vfio_device_is_noiommu (struct vfio_device *vdev) {return false;Signature · 1 line
2.34.1
Jason Gunthorpe
maintainerRe: [PATCH V2 05/11] vfio: Allow null group for noiommu without containers
Quoted 3 lines · Mostafa Saleh
But is that really needed, I feel like this optimization is not worth the extra effort to add those checks and the possiblity of missing some. what do you think?
I don't think it is an optimization, this is cleaning the uapi visible stuff not not require these hacks anymore.
Jason
PATCH v2 6/11 vfio: Introduce and set noiommu flag on vfio_device
In reply to Jacob PanWhen a VFIO device is added to a noiommu group, set the noiommu flag on the vfio_device structure to indicate that the device operates in noiommu mode.
Also update function signatures to pass vfio_device instead of device, which has the direct access to the noiommu flag.
--- drivers/vfio/group.c | 21 +++++++++++---------- include/linux/vfio.h | 1 + 2 files changed, 12 insertions(+), 10 deletions(-)
drivers/vfio/group.c +11 -10
@@ -588,7 +588,7 @@ static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group,return ret; } -static struct vfio_group *vfio_noiommu_group_alloc (struct device *dev, +static struct vfio_group *vfio_noiommu_group_alloc (struct vfio_device *vdev,enum vfio_group_type type) {struct iommu_group *iommu_group;
@@ -610,7 +610,7 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev, ret =iommu_group_set_name (iommu_group,"vfio-noiommu" );if (ret) goto out_put_group; - ret =iommu_group_add_device (iommu_group, dev); + ret =iommu_group_add_device (iommu_group, vdev->dev);if (ret) goto out_put_group;
@@ -625,7 +625,7 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev,return group; out_remove_device: -iommu_group_remove_device (dev); +iommu_group_remove_device (vdev->dev); out_put_group:iommu_group_put (iommu_group);return ERR_PTR (ret);
@@ -646,23 +646,24 @@ static bool vfio_group_has_device(struct vfio_group *group, struct device *dev)return false; } -static struct vfio_group *vfio_group_find_or_alloc (struct device *dev) +static struct vfio_group *vfio_group_find_or_alloc (struct vfio_device *vdev) {struct iommu_group *iommu_group;struct vfio_group *group; - iommu_group =iommu_group_get (dev); + iommu_group =iommu_group_get (vdev->dev);if (!iommu_group && vfio_noiommu) { + vdev->noiommu =1 ;/* * With noiommu enabled, create an IOMMU group for devices that * don't already have one, implying no IOMMU hardware/driver * exists. Taint the kernel because we're about to give a DMA * capable device to a user without IOMMU protection. */ - group =vfio_noiommu_group_alloc (dev,VFIO_NO_IOMMU ); + group =vfio_noiommu_group_alloc (vdev,VFIO_NO_IOMMU );if (!IS_ERR (group)) {add_taint (TAINT_USER ,LOCKDEP_STILL_OK ); -dev_warn (dev,"Adding kernel taint for vfio-noiommu group on device\n" ); +dev_warn (vdev->dev,"Adding kernel taint for vfio-noiommu group on device\n" ); }return group; }
@@ -673,7 +674,7 @@ static struct vfio_group *vfio_group_find_or_alloc(struct device *dev)mutex_lock (&vfio.group_lock); group =vfio_group_find_from_iommu (iommu_group);if (group) { -if (WARN_ON (vfio_group_has_device (group, dev))) +if (WARN_ON (vfio_group_has_device (group, vdev->dev))) group =ERR_PTR (-EINVAL );else refcount_inc (&group->drivers);
@@ -693,9 +694,9 @@ int vfio_device_set_group(struct vfio_device *device,struct vfio_group *group;if (type ==VFIO_IOMMU ) - group =vfio_group_find_or_alloc (device->dev); + group =vfio_group_find_or_alloc (device);else - group =vfio_noiommu_group_alloc (device->dev, type); + group =vfio_noiommu_group_alloc (device, type);if (IS_ERR (group))return PTR_ERR (group);
include/linux/vfio.h +1 -0
@@ -72,6 +72,7 @@ struct vfio_device { u8 iommufd_attached:1;#endif u8 cdev_opened:1; + u8 noiommu:1;#ifdef CONFIG_DEBUG_FS /* * debug_root is a static property of the vfio_device
Signature · 1 line
2.34.1
Re: [PATCH V2 06/11] vfio: Introduce and set noiommu flag on vfio_device
Quoted 8 lines · Jacob Pan
When a VFIO device is added to a noiommu group, set the noiommu flag on the
vfio_devicestructure to indicate that the device operates in noiommu mode.Also update function signatures to pass
vfio_deviceinstead of device, which has the direct access to the noiommu flag.Signed-off-by Jacob Pan <[email protected]>
Thanks,
Mostafa
Quoted 101 lines
--- drivers/vfio/group.c | 21 +++++++++++---------- include/linux/vfio.h | 1 + 2 files changed, 12 insertions(+), 10 deletions(-)drivers/vfio/group.c +11
-10@@ -588,7 +588,7 @@ static struct vfio_group *vfio_create_group(struct iommu_group *iommu_group,return ret; } -static struct vfio_group *vfio_noiommu_group_alloc (struct device *dev, +static struct vfio_group *vfio_noiommu_group_alloc (struct vfio_device *vdev,enum vfio_group_type type) {struct iommu_group *iommu_group;@@ -610,7 +610,7 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev, ret =iommu_group_set_name (iommu_group,"vfio-noiommu" );if (ret) goto out_put_group; - ret =iommu_group_add_device (iommu_group, dev); + ret =iommu_group_add_device (iommu_group, vdev->dev);if (ret) goto out_put_group;@@ -625,7 +625,7 @@ static struct vfio_group *vfio_noiommu_group_alloc(struct device *dev,return group; out_remove_device: -iommu_group_remove_device (dev); +iommu_group_remove_device (vdev->dev); out_put_group:iommu_group_put (iommu_group);return ERR_PTR (ret);@@ -646,23 +646,24 @@ static bool vfio_group_has_device(struct vfio_group *group, struct device *dev)return false; } -static struct vfio_group *vfio_group_find_or_alloc (struct device *dev) +static struct vfio_group *vfio_group_find_or_alloc (struct vfio_device *vdev) {struct iommu_group *iommu_group;struct vfio_group *group; - iommu_group =iommu_group_get (dev); + iommu_group =iommu_group_get (vdev->dev);if (!iommu_group && vfio_noiommu) { + vdev->noiommu =1 ;/* * With noiommu enabled, create an IOMMU group for devices that * don't already have one, implying no IOMMU hardware/driver * exists. Taint the kernel because we're about to give a DMA * capable device to a user without IOMMU protection. */ - group =vfio_noiommu_group_alloc (dev,VFIO_NO_IOMMU ); + group =vfio_noiommu_group_alloc (vdev,VFIO_NO_IOMMU );if (!IS_ERR (group)) {add_taint (TAINT_USER ,LOCKDEP_STILL_OK ); -dev_warn (dev,"Adding kernel taint for vfio-noiommu group on device\n" ); +dev_warn (vdev->dev,"Adding kernel taint for vfio-noiommu group on device\n" ); }return group; }@@ -673,7 +674,7 @@ static struct vfio_group *vfio_group_find_or_alloc(struct device *dev)mutex_lock (&vfio.group_lock); group =vfio_group_find_from_iommu (iommu_group);if (group) { -if (WARN_ON (vfio_group_has_device (group, dev))) +if (WARN_ON (vfio_group_has_device (group, vdev->dev))) group =ERR_PTR (-EINVAL );else refcount_inc (&group->drivers);@@ -693,9 +694,9 @@ int vfio_device_set_group(struct vfio_device *device,struct vfio_group *group;if (type ==VFIO_IOMMU ) - group =vfio_group_find_or_alloc (device->dev); + group =vfio_group_find_or_alloc (device);else - group =vfio_noiommu_group_alloc (device->dev, type); + group =vfio_noiommu_group_alloc (device, type);if (IS_ERR (group))return PTR_ERR (group);include/linux/vfio.h +1
-0@@ -72,6 +72,7 @@ struct vfio_device { u8 iommufd_attached:1;#endif u8 cdev_opened:1; + u8 noiommu:1;#ifdef CONFIG_DEBUG_FS /* * debug_root is a static property of the vfio_device Signature · 1 line
2.34.1
PATCH v2 7/11 vfio: Update noiommu device detection logic for cdev
In reply to Jacob PanRework vfio_device_is_noiommu() to derive noiommu mode based on device, group type, and configurations.
--- drivers/vfio/vfio.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/vfio/vfio.h +7 -2
@@ -127,8 +127,13 @@ static inline bool vfio_null_group_allowed(void)static inline bool vfio_device_is_noiommu (struct vfio_device *vdev) { -return IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && - vdev->group->type ==VFIO_NO_IOMMU ; +if (!IS_ENABLED (CONFIG_VFIO_NOIOMMU )) +return false; + +if (vfio_null_group_allowed ()) +return vdev->noiommu; + +return vdev->group->type ==VFIO_NO_IOMMU ; }#else struct vfio_group ;
Signature · 1 line
2.34.1
Re: [PATCH V2 07/11] vfio: Update noiommu device detection logic for cdev
Quoted 15 lines · Jacob Pan
Rework
vfio_device_is_noiommu()to derive noiommu mode based on device, group type, and configurations.Signed-off-by Jacob Pan <[email protected]>--- drivers/vfio/vfio.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)drivers/vfio/vfio.h +0
-0@@ -127,8 +127,13 @@ static inline bool vfio_null_group_allowed(void)static inline bool vfio_device_is_noiommu (struct vfio_device *vdev)
drivers/vfio/vfio.h static inline bool vfio_null_group_allowed(void) · line 136
Quoted 10 lines
drivers/vfio/vfio.h +7
-2{ -return IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && - vdev->group->type ==VFIO_NO_IOMMU ; +if (!IS_ENABLED (CONFIG_VFIO_NOIOMMU )) +return false; + +if (vfio_null_group_allowed ()) +return vdev->noiommu; + +return vdev->group->type ==VFIO_NO_IOMMU ;
I see that noiommu is set for both, can this just be simplified to:
return IS_ENABLED(CONFIG_VFIO_NOIOMMU) && vdev->noiommu;
Thanks,
Mostafa
Quoted 6 lines
drivers/vfio/vfio.h +0
-0}#else struct vfio_group ;Signature · 1 line
2.34.1
Re: [PATCH V2 07/11] vfio: Update noiommu device detection logic for cdev
Hi Mostafa,
Quoted 29 lines · On Sun, 22 Mar 2026 10:04:35 +0000 Mostafa Saleh <[email protected]> wrote:
Quoted 15 lines · Jacob Pan
Rework
vfio_device_is_noiommu()to derive noiommu mode based on device, group type, and configurations.Signed-off-by Jacob Pan <[email protected]>--- drivers/vfio/vfio.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)drivers/vfio/vfio.h +0
-0@@ -127,8 +127,13 @@ static inline boolvfio_null_group_allowed (void ) static inline bool vfio_device_is_noiommu(struct vfio_device *vdev)drivers/vfio/vfio.h static inline bool · line 136
Quoted 10 lines
drivers/vfio/vfio.h +7
-2{ -return IS_ENABLED (CONFIG_VFIO_NOIOMMU ) && - vdev->group->type ==VFIO_NO_IOMMU ; +if (!IS_ENABLED (CONFIG_VFIO_NOIOMMU )) +return false; + +if (vfio_null_group_allowed ()) +return vdev->noiommu; + +return vdev->group->type ==VFIO_NO_IOMMU ;I see that noiommu is set for both, can this just be simplified to:
Quoted 2 lines
return IS_ENABLED(CONFIG_VFIO_NOIOMMU) && vdev->noiommu;
Good point, no need to check group now. will do.
Sorry for the late response.
Thanks,
Jacob
PATCH v2 8/11 vfio: Enable cdev noiommu mode under iommufd
In reply to Jacob PanNow that devices under noiommu mode can bind with IOMMUFD and perform IOAS operations, lift restrictions on cdev from VFIO side.
No IOMMU cdevs are explicitly named with noiommu prefix. e.g.
/dev/vfio/
|-- 7
|-- devices
| `-- noiommu-vfio0
`-- vfio---
v2:
- Fix build dependency on IOMMU_SUPPORT
--- drivers/vfio/Kconfig | 7 +++++-- drivers/vfio/iommufd.c | 7 ------- drivers/vfio/vfio.h | 8 +------- drivers/vfio/vfio_main.c | 22 +++++++++++++++++++--- include/linux/vfio.h | 9 +++++++++ 5 files changed, 34 insertions(+), 19 deletions(-)
drivers/vfio/Kconfig +5 -2
@@ -22,8 +22,7 @@ config VFIO_DEVICE_CDEV The VFIO device cdev is another way for userspace to get device access. Userspace gets device fd by opening device cdev under /dev/vfio/devices/vfioX, and then bind the device fd with an iommufd - to set up secure DMA context for device access. This interface does - not support noiommu. + to set up secure DMA context for device access. If you don't know what to do here, say N.
@@ -63,6 +62,10 @@ endif config VFIO_NOIOMMU bool "VFIO No-IOMMU support" depends on VFIO_GROUP + select GENERIC_PT + select IOMMU_PT + select IOMMU_PT_AMDV1 + depends on IOMMU_SUPPORT help VFIO is built on the ability to isolate devices using the IOMMU. Only with an IOMMU can userspace access to DMA capable devices be
drivers/vfio/iommufd.c +0 -7
@@ -25,10 +25,6 @@ int vfio_df_iommufd_bind(struct vfio_device_file *df)lockdep_assert_held (&vdev->dev_set->lock); -/* Returns 0 to permit device opening under noiommu mode */ -if (vfio_device_is_noiommu (vdev)) -return 0 ; -return vdev->ops->bind_iommufd (vdev, ictx, &df->devid); }
@@ -58,9 +54,6 @@ void vfio_df_iommufd_unbind(struct vfio_device_file *df)lockdep_assert_held (&vdev->dev_set->lock); -if (vfio_device_is_noiommu (vdev)) -return ; -if (vdev->ops->unbind_iommufd) vdev->ops->unbind_iommufd (vdev); }
drivers/vfio/vfio.h +1 -7
@@ -381,19 +381,13 @@ void vfio_init_device_cdev(struct vfio_device *device);static inline int vfio_device_add (struct vfio_device *device) { -/* cdev does not support noiommu device */ -if (vfio_device_is_noiommu (device)) -return device_add (&device->device);vfio_init_device_cdev (device);return cdev_device_add (&device->cdev, &device->device); }static inline void vfio_device_del (struct vfio_device *device) { -if (vfio_device_is_noiommu (device)) -device_del (&device->device); -else -cdev_device_del (&device->cdev, &device->device); +cdev_device_del (&device->cdev, &device->device); }int vfio_device_fops_cdev_open (struct inode *inode,struct file *filep);
drivers/vfio/vfio_main.c +19 -3
@@ -331,13 +331,15 @@ static int __vfio_register_dev(struct vfio_device *device,if (!device->dev_set)vfio_assign_device_set (device, device); - ret =dev_set_name (&device->device,"vfio%d" , device->index); + ret =vfio_device_set_group (device, type);if (ret)return ret; - ret =vfio_device_set_group (device, type); +/* Just to be safe, expose to user explicitly noiommu cdev node */ + ret =dev_set_name (&device->device,"%svfio%d" , + device->noiommu ?"noiommu-" :"" , device->index);if (ret) -return ret; + goto err_out;/* * VFIO always sets IOMMU_CACHE because we offer no way for userspace to
@@ -357,6 +359,10 @@ static int __vfio_register_dev(struct vfio_device *device,/* Refcounting can't start until the driver calls register */ refcount_set (&device->refcount,1 ); +/* noiommu device w/o container may have NULL group */ +if (vfio_device_is_noiommu (device) && !vfio_device_has_group (device)) +return 0 ; +vfio_device_group_register (device);vfio_device_debugfs_init (device);
@@ -391,6 +397,16 @@ void vfio_unregister_group_dev(struct vfio_device *device)bool interrupted = false;long rc; +/* +* For noiommu devices without a container, thus no dummy group, +* simply delete and unregister to balance refcount. +*/ +if (device->noiommu && !vfio_device_has_group (device)) { +vfio_device_del (device); +vfio_device_put_registration (device); +return ; + } +/* * Prevent new device opened by userspace via the * VFIO_GROUP_GET_DEVICE_FD in the group path.
include/linux/vfio.h +9 -0
@@ -327,6 +327,10 @@ struct iommu_group *vfio_file_iommu_group(struct file *file);#if IS_ENABLED (CONFIG_VFIO_GROUP )bool vfio_file_is_group (struct file *file);bool vfio_file_has_dev (struct file *file,struct vfio_device *device); +static inline bool vfio_device_has_group (struct vfio_device *device) +{ +return device->group; +}#else static inline bool vfio_file_is_group (struct file *file) {
@@ -337,6 +341,11 @@ static inline bool vfio_file_has_dev(struct file *file, struct vfio_device *devi {return false; } + +static inline bool vfio_device_has_group (struct vfio_device *device) +{ +return false; +}#endif bool vfio_file_is_valid (struct file *file);bool vfio_file_enforced_coherent (struct file *file);
Signature · 1 line
2.34.1
kernel test robot
build botRe: [PATCH V2 08/11] vfio: Enable cdev noiommu mode under iommufd
kernel test robot noticed the following build errors:
Hi Jacob,
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on v7.0-rc3 next-20260311]
[cannot apply to awilliam-vfio/next awilliam-vfio/for-linus] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Jacob-Pan/iommufd-Support-a-HWPT-without-an-iommu-driver-for-noiommu/20260313-182818 base: linus/master patch link: https://lore.kernel.org/r/20260312155637.376854-9-jacob.pan%40linux.microsoft.com patch subject: [PATCH V2 08/11] vfio: Enable cdev noiommu mode under iommufd config: m68k-allmodconfig (https://download.01.org/0day-ci/archive/20260314/[email protected]/config) compiler: m68k-linux-gcc (GCC) 15.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260314/[email protected]/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All errors (new ones prefixed by >>):
In file included from drivers/iommu/generic_pt/fmt/iommu_template.h:36,
from drivers/iommu/generic_pt/fmt/iommu_mock.c:10:
drivers/iommu/generic_pt/fmt/amdv1.h: In function 'amdv1pt_install_table':
>> drivers/iommu/generic_pt/fmt/amdv1.h:255:16: error: implicit declaration of function 'pt_table_install64'; did you mean 'pt_table_install32'? [-Wimplicit-function-declaration]
255 | return pt_table_install64(pts, entry);
| ^~~~~~~~~~~~~~~~~~
| pt_table_install32Kconfig warnings: (for reference only)
WARNING: unmet direct dependencies detected for IOMMU_PT_AMDV1 Depends on [n]: GENERIC_PT [=y] && IOMMU_PT [=m] && !GENERIC_ATOMIC64 [=y] Selected by [m]:
- VFIO_NOIOMMU [=y] && VFIO [=m] && VFIO_GROUP [=y] && IOMMU_SUPPORT [=y]
vim +255 drivers/iommu/generic_pt/fmt/amdv1.h
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 236
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 237 static inline bool amdv1pt_install_table(struct pt_state *pts,
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 238 pt_oaddr_t table_pa,
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 239 const struct pt_write_attrs *attrs)
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 240 {
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 241 u64 entry;
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 242
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 243 /*
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 244 * IR and IW are ANDed from the table levels along with the PTE. We
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 245 * always control permissions from the PTE, so always set IR and IW for
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 246 * tables.
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 247 */
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 248 entry = AMDV1PT_FMT_PR |
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 249 FIELD_PREP(AMDV1PT_FMT_NEXT_LEVEL, pts->level) |
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 250 FIELD_PREP(AMDV1PT_FMT_OA,
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 251 log2_div(table_pa, PT_GRANULE_LG2SZ)) |
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 252 AMDV1PT_FMT_IR | AMDV1PT_FMT_IW;
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 253 if (pts_feature(pts, PT_FEAT_AMDV1_ENCRYPT_TABLES))
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 254 entry = __sme_set(entry);
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 @255 return pt_table_install64(pts, entry);
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 256 }
879ced2bab1ba9 Jason Gunthorpe 2025-11-04 257 #define pt_install_table amdv1pt_install_table879ced2bab1ba9 Jason Gunthorpe 2025-11-04 258
Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki
PATCH v2 9/11 vfio:selftest: Handle VFIO noiommu cdev
In reply to Jacob PanWith unsafe DMA noiommu mode, the vfio devices are prefixed with noiommu-, e.g./dev/vfio/
|-- devices | `-- noiommu-vfio0 |-- noiommu-0
`-- vfio
Let vfio tests, such as luo kexec test, accommodate the noiommu device files.
--- .../selftests/vfio/lib/vfio_pci_device.c | 25 +++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-)
tools/testing/selftests/vfio/lib/vfio_pci_device.c +23 -2
@@ -290,6 +290,24 @@ static void vfio_pci_device_setup(struct vfio_pci_device *device) device->msi_eventfds[i] =-1 ; } + +static int is_unsafe_noiommu_mode_enabled (void ) +{ +const char *path ="/sys/module/vfio/parameters/enable_unsafe_noiommu_mode" ; +FILE *f; +int c; + + f =fopen (path,"re" ); +if (!f) +return 0 ; + + c =fgetc (f); +fclose (f); +if (c =='Y' || c =='y' ) +return 1 ; +return 0 ; +} +const char *vfio_pci_get_cdev_path (const char *bdf) {char dir_path[PATH_MAX ];
@@ -306,8 +324,11 @@ const char *vfio_pci_get_cdev_path(const char *bdf)VFIO_ASSERT_NOT_NULL (dir,"Failed to open directory %s\n" , dir_path);while ((entry =readdir (dir)) !=NULL ) { -/* Find the file that starts with "vfio" */ -if (strncmp ("vfio" , entry->d_name,4 )) +/* Find the file that starts with "noiommu-vfio" or "vfio" */ +if (is_unsafe_noiommu_mode_enabled ()) { +if (strncmp ("noiommu-vfio" , entry->d_name,strlen ("noiommu-vfio" ))) +continue ; + }else if (strncmp ("vfio" , entry->d_name,4 ))continue ;snprintf (cdev_path,PATH_MAX ,"/dev/vfio/devices/%s" , entry->d_name);
Signature · 1 line
2.34.1
PATCH v2 10/11 selftests/vfio: Add iommufd noiommu mode selftest for cdev
In reply to Jacob PanAdd comprehensive selftest for VFIO device operations with iommufd in noiommu mode. Tests cover:
- Device binding to iommufd
- IOAS (I/O Address Space) allocation, mapping with dummy IOVA
- Retrieve PA from dummy IOVA
- Device attach/detach operations as usual
---
v2:
- Use huge page ioas map to test GET_PA searching for contiguous PA range.
--- tools/testing/selftests/vfio/Makefile | 1 + .../vfio/vfio_iommufd_noiommu_test.c | 549 ++++++++++++++++++ 2 files changed, 550 insertions(+) create mode 100644 tools/testing/selftests/vfio/vfio_iommufd_noiommu_test.c
tools/testing/selftests/vfio/Makefile +1 -0
@@ -9,6 +9,7 @@ CFLAGS = $(KHDR_INCLUDES) TEST_GEN_PROGS += vfio_dma_mapping_test TEST_GEN_PROGS += vfio_dma_mapping_mmio_test TEST_GEN_PROGS += vfio_iommufd_setup_test +TEST_GEN_PROGS += vfio_iommufd_noiommu_test TEST_GEN_PROGS += vfio_pci_device_test TEST_GEN_PROGS += vfio_pci_device_init_perf_test TEST_GEN_PROGS += vfio_pci_driver_test
tools/testing/selftests/vfio/vfio_iommufd_noiommu_test.c new file +549 -0
549 lines not on this page · raw message
Signature · 1 line
2.34.1
PATCH v2 11/11 Doc: Update VFIO NOIOMMU mode
In reply to Jacob PanDocument the NOIOMMU mode with newly added cdev support under iommufd.
--- Documentation/driver-api/vfio.rst | 44 +++++++++++++++++++++++++++++-- 1 file changed, 42 insertions(+), 2 deletions(-)
Documentation/driver-api/vfio.rst +42 -2
@@ -275,8 +275,6 @@ in a VFIO group. With CONFIG_VFIO_DEVICE_CDEV=y the user can now acquire a device fd by directly opening a character device /dev/vfio/devices/vfioX where "X" is the number allocated uniquely by VFIO for registered devices. -cdev interface does not support noiommu devices, so user should use -the legacy group interface if noiommu is wanted. The cdev only works with IOMMUFD. Both VFIO drivers and applications must adapt to the new cdev security model which requires using
@@ -370,6 +368,48 @@ IOMMUFD IOAS/HWPT to enable userspace DMA:: /* Other device operations as stated in "VFIO Usage Example" */ +VFIO NOIOMMU mode +------------------------------------------------------------------------------- +VFIO also supports a no-IOMMU mode, intended for usages where unsafe DMA can +be performed by userspace drivers w/o physical IOMMU protection. This mode +is controlled by the parameter: + +/sys/module/vfio/parameters/enable_unsafe_noiommu_mode + +Upon enabling this mode, with an assigned device, the user will be presented +with a VFIO group and device file, e.g. + +/dev/vfio/ +|-- devices +| `-- noiommu-vfio0 /* VFIO device cdev */ +|-- noiommu-0 /* VFIO group */ +`-- vfio + +The capabilities vary depending on the device programming interface and kernel +configuration used. The following table summarizes the differences: + ++-------------------+---------------------+---------------------+ +| Feature | VFIO group | VFIO device cdev | ++===================+=====================+=====================+ +| VFIO device UAPI | Yes | Yes | ++-------------------+---------------------+---------------------+ +| VFIO container | No | No | ++-------------------+---------------------+---------------------+ +| IOMMUFD IOAS | No | Yes* | ++-------------------+---------------------+---------------------+ +Note that the VFIO container case includes IOMMUFD provided VFIO compatibility +interfaces when either CONFIG_VFIO_CONTAINER or CONFIG_IOMMUFD_VFIO_CONTAINER is +enabled. + +* IOMMUFD UAPI is available for VFIO device cdev to pin and map user memory with +the ability to retrieve physical addresses for DMA command submission. + +A new IOMMUFD ioctl IOMMU_IOAS_GET_PA is added to retrieve the physical address +for a given user virtual address. Note that IOMMU_IOAS_MAP_FIXED_IOVA flag is +ignored in no-IOMMU mode since there is no physical DMA remapping hardware. +tools/testing/selftests/vfio/vfio_iommufd_noiommu_test.c provides an example of +using this ioctl in no-IOMMU mode. + VFIO User API -------------------------------------------------------------------------------
Signature · 1 line
2.34.1
kernel test robot
build botRe: [PATCH V2 11/11] Doc: Update VFIO NOIOMMU mode
kernel test robot noticed the following build warnings:
Hi Jacob,
kernel test robot noticed the following build warnings:
[auto build test WARNING on linus/master]
[also build test WARNING on next-20260313]
[cannot apply to awilliam-vfio/next awilliam-vfio/for-linus v6.16-rc1] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Jacob-Pan/iommufd-Support-a-HWPT-without-an-iommu-driver-for-noiommu/20260313-182818 base: linus/master patch link: https://lore.kernel.org/r/20260312155637.376854-12-jacob.pan%40linux.microsoft.com patch subject: [PATCH V2 11/11] Doc: Update VFIO NOIOMMU mode compiler: clang version 20.1.8 (https://github.com/llvm/llvm-project 87f0227cb60147a26a1eeb4fb06e3b505e9c7261) docutils: docutils (Docutils 0.21.2, Python 3.13.5, on linux) reproduce: (https://download.01.org/0day-ci/archive/20260313/[email protected]/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All warnings (new ones prefixed by >>):
Ring Design
----------- [docutils]
WARNING: ./include/linux/usb/typec_altmode.h:44 struct member 'priority' not described in 'typec_altmode' WARNING: ./include/linux/usb/typec_altmode.h:44 struct member 'mode_selection' not described in 'typec_altmode' >> Documentation/driver-api/vfio.rst:382: WARNING: Inline substitution_reference start-string without end-string. [docutils] >> Documentation/driver-api/vfio.rst:382: WARNING: Inline interpreted text or phrase reference start-string without end-string. [docutils] >> Documentation/driver-api/vfio.rst:382: WARNING: Inline emphasis start-string without end-string. [docutils] >> Documentation/driver-api/vfio.rst:382: WARNING: Inline interpreted text or phrase reference start-string without end-string. [docutils] Documentation/driver-api/vfio.rst:392: ERROR: Malformed table. Right border not aligned or missing. -- +-------------------+---------------------+---------------------+ | VFIO container | No | No | +-------------------+---------------------+---------------------+ | IOMMUFD IOAS | No | Yes* | +-------------------+---------------------+---------------------+ [docutils] >> Documentation/driver-api/vfio.rst:400: WARNING: Blank line required after table. [docutils] >> Documentation/driver-api/vfio.rst:405: WARNING: Bullet list ends without a blank line; unexpected unindent. [docutils] WARNING: ./include/linux/virtio.h:188 struct member 'map' not described in 'virtio_device' WARNING: ./include/linux/virtio.h:188 struct member 'VIRTIO_DECLARE_FEATURES(features' not described in 'virtio_device' WARNING: ./include/linux/virtio.h:188 struct member 'vmap' not described in 'virtio_device' Documentation/gpu/amdgpu/display/display-manager:47: ./drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c:61: ERROR: Unexpected section title.
vim +382 Documentation/driver-api/vfio.rst
381
> 382 /dev/vfio/ 383 |-- devices 384 | `-- noiommu-vfio0 /* VFIO device cdev */ 385 |-- noiommu-0 /* VFIO group */ 386 `-- vfio 387 388 The capabilities vary depending on the device programming interface and kernel 389 configuration used. The following table summarizes the differences: 390 391 +-------------------+---------------------+---------------------+ 392 | Feature | VFIO group | VFIO device cdev | 393 +===================+=====================+=====================+ 394 | VFIO device UAPI | Yes | Yes | 395 +-------------------+---------------------+---------------------+ 396 | VFIO container | No | No | 397 +-------------------+---------------------+---------------------+ 398 | IOMMUFD IOAS | No | Yes* | 399 +-------------------+---------------------+---------------------+ > 400 Note that the VFIO container case includes IOMMUFD provided VFIO compatibility 401 interfaces when either CONFIG_VFIO_CONTAINER or CONFIG_IOMMUFD_VFIO_CONTAINER is 402 enabled. 403 404 * IOMMUFD UAPI is available for VFIO device cdev to pin and map user memory with > 405 the ability to retrieve physical addresses for DMA command submission.
406
Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki