This allows the upper layers to set a domain to a PASID of a device
if the PASID feature is supported by the IOMMU hardware. The typical
use cases are, for example, kernel DMA with PASID and hardware
assisted mediated device drivers.
The attaching device and pasid information is tracked in a per-domain
list and is used for IOTLB and devTLB invalidation.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-8-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The domain_flush_pasid_iotlb() helper function is used to flush the IOTLB
entries for a given PASID. Previously, this function assumed that
RID2PASID was only used for the first-level DMA translation. However, with
the introduction of the set_dev_pasid callback, this assumption is no
longer valid.
Add a check before using the RID2PASID for PASID invalidation. This check
ensures that the domain has been attached to a physical device before
using RID2PASID.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-7-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Currently draining page requests and responses for a pasid is part of SVA
implementation. This is because the driver only supports attaching an SVA
domain to a device pasid. As we are about to support attaching other types
of domains to a device pasid, the prq draining code becomes generic.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-6-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The VT-d spec requires to use PASID-based-IOTLB invalidation descriptor
to invalidate IOTLB and the paging-structure caches for a first-stage
page table. Add a generic helper to do this.
RID2PASID is used if the domain has been attached to a physical device,
otherwise real PASIDs that the domain has been attached to will be used.
The 'real' PASID attachment is handled in the subsequent change.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-4-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
PCIe Process address space ID (PASID) is used to tag DMA traffic, it
provides finer grained isolation than requester ID (RID).
For each device/RID, 0 is a special PASID for the normal DMA (no
PASID). This is universal across all architectures that supports PASID,
therefore warranted to be reserved globally and declared in the common
header. Consequently, we can avoid the conflict between different PASID
use cases in the generic code. e.g. SVA and DMA API with PASIDs.
This paved away for device drivers to choose global PASID policy while
continue doing normal DMA.
Noting that VT-d could support none-zero RID/NO_PASID, but currently not
used.
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230802212427.1497170-2-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This is a step toward making __iommu_probe_device() self contained.
It should, under proper locking, check if the device is already associated
with an iommu driver and resolve parallel probes. All but one of the
callers open code this test using two different means, but they all
rely on dev->iommu_group.
Currently the bus_iommu_probe()/probe_iommu_group() and
probe_acpi_namespace_devices() rejects already probed devices with an
unlocked read of dev->iommu_group. The OF and ACPI "replay" functions use
device_iommu_mapped() which is the same read without the pointless
refcount.
Move this test into __iommu_probe_device() and put it under the
iommu_probe_device_lock. The store to dev->iommu_group is in
iommu_group_add_device() which is also called under this lock for iommu
driver devices, making it properly locked.
The only path that didn't have this check is the hotplug path triggered by
BUS_NOTIFY_ADD_DEVICE. The only way to get dev->iommu_group assigned
outside the probe path is via iommu_group_add_device(). Today the only
caller is VFIO no-iommu which never associates with an iommu driver. Thus
adding this additional check is safe.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/1-v3-328044aa278c+45e49-iommu_probe_jgg@nvidia.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
These lines of code were commented out when they were first added in commit
ba39592764 ("Intel IOMMU: Intel IOMMU driver"). We do not want to restore
them because the VT-d spec has deprecated the read/write draining hit.
VT-d spec (section 11.4.2):
"
Hardware implementation with Major Version 2 or higher (VER_REG), always
performs required drain without software explicitly requesting a drain in
IOTLB invalidation. This field is deprecated and hardware will always
report it as 1 to maintain backward compatibility with software.
"
Remove the code to make the code cleaner.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Link: https://lore.kernel.org/r/20230609060514.15154-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Remove the WARN_ON(did == 0) as the domain id 0 is reserved and
set once the domain_ids is allocated. So iommu_init_domains will
never return 0.
Remove the WARN_ON(!table) as this pointer will be accessed in
the following code, if empty "table" really happens, the kernel
will report a NULL pointer reference warning at the first place.
Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
Link: https://lore.kernel.org/r/20230605112659.308981-3-yanfei.xu@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
dmar_reenable_qi() may not succeed. Check and return when it fails.
Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
Link: https://lore.kernel.org/r/20230605112659.308981-2-yanfei.xu@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
It remains really handy to have distinct DMA domain types within core
code for the sake of default domain policy selection, but we can now
hide that detail from drivers by using the new capability instead.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Jerry Snitselaar <jsnitsel@redhat.com> # amd, intel, smmu-v3
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/1c552d99e8ba452bdac48209fa74c0bdd52fd9d9.1683233867.git.robin.murphy@arm.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Passing a special type to domain_alloc to indirectly query whether flush
queues are a worthwhile optimisation with the given driver is a bit
clunky, and looking increasingly anachronistic. Let's put that into an
explicit capability instead.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Tested-by: Jerry Snitselaar <jsnitsel@redhat.com> # amd, intel, smmu-v3
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/f0086a93dbccb92622e1ace775846d81c1c4b174.1683233867.git.robin.murphy@arm.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Domain map/unmap with invalid parameters shouldn't crash the kernel.
Therefore, using if() replaces the BUG_ON.
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406065944.2773296-6-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When performing domain_context_mapping or getting dma_pte of a pfn, the
availability of the domain page table directory is ensured. Therefore,
the domain->pgd checkings are unnecessary.
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406065944.2773296-5-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
VT-d iotlb cache invalidation request with unexpected type is considered
as a bug to developers, which can be fixed. So, when such kind of issue
comes out, it needs to be reported through the kernel log, instead of
halting the system. Replacing BUG_ON with warning reporting.
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406065944.2773296-4-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
When encountering an unexpected invalid pfn range, the kernel should
attempt recovery and proceed with execution. Therefore, using WARN_ON to
replace BUG_ON to avoid halting the machine.
Besides, one redundant checking is reduced.
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406065944.2773296-3-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This addresses the following issue reported by klocwork tool:
- operands of different size in bitwise operations
Suggested-by: Yongwei Ma <yongwei.ma@intel.com>
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20230406065944.2773296-2-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Supervisor Request Enable (SRE) bit in a PASID entry is for permission
checking on DMA requests. When SRE = 0, DMA with supervisor privilege
will be blocked. However, for in-kernel DMA this is not necessary in that
we are targeting kernel memory anyway. There's no need to differentiate
user and kernel for in-kernel DMA.
Let's use non-privileged (user) permission for all PASIDs used in kernel,
it will be consistent with DMA without PASID (RID_PASID) as well.
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230331231137.1947675-2-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
iommu_unregister_device_fault_handler() and iopf_queue_remove_device()
are called after device has stopped issuing new page falut requests and
all outstanding page requests have been drained. They should never fail.
Trigger a warning if it happens unfortunately.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-7-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
PRI is only used for IOPF. With this move, the PCI/PRI feature could be
controlled by the device driver through iommu_dev_enable/disable_feature()
interfaces.
Reviewed-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-6-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
They should be part of the per-device iommu private data initialization.
Reviewed-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-5-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Generally enabling IOMMU_DEV_FEAT_SVA requires IOMMU_DEV_FEAT_IOPF, but
some devices manage I/O Page Faults themselves instead of relying on the
IOMMU. Move IOPF related code from SVA to IOPF enabling path.
For the device drivers that relies on the IOMMU for IOPF through PCI/PRI,
IOMMU_DEV_FEAT_IOPF must be enabled before and disabled after
IOMMU_DEV_FEAT_SVA.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Currently enabling SVA requires IOPF support from the IOMMU and device
PCI PRI. However, some devices can handle IOPF by itself without ever
sending PCI page requests nor advertising PRI capability.
Allow SVA support with IOPF handled either by IOMMU (PCI PRI) or device
driver (device-specific IOPF). As long as IOPF could be handled, SVA
should continue to work.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230324120234.313643-3-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
INVALID_IOASID and IOMMU_PASID_INVALID are duplicated. Rename
INVALID_IOASID and consolidate since we are moving away from IOASID
infrastructure.
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230322200803.869130-7-jacob.jun.pan@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Use sysfs_emit() instead of the sprintf() for sysfs entries. sysfs_emit()
knows the maximum of the temporary buffer used for outputting sysfs
content and avoids overrunning the buffer length.
Prefer 'long long' over 'long long int' as suggested by checkpatch.pl.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20230322123421.278852-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Some polishing and small fixes for iommufd:
- Remove IOMMU_CAP_INTR_REMAP, instead rely on the interrupt subsystem
- Use GFP_KERNEL_ACCOUNT inside the iommu_domains
- Support VFIO_NOIOMMU mode with iommufd
- Various typos
- A list corruption bug if HWPTs are used for attach
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY/TgzQAKCRCFwuHvBreF
Ya3AAP4/WxTJIbDvtTyH3Fae3NxTdO8j8gsUvU1vrRYG83zdnAEAxd1yii7GEO8D
crkeq9D4FUiPAkFnJ64Exw2FHb060Qg=
=RABK
-----END PGP SIGNATURE-----
Merge tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd
Pull iommufd updates from Jason Gunthorpe:
"Some polishing and small fixes for iommufd:
- Remove IOMMU_CAP_INTR_REMAP, instead rely on the interrupt
subsystem
- Use GFP_KERNEL_ACCOUNT inside the iommu_domains
- Support VFIO_NOIOMMU mode with iommufd
- Various typos
- A list corruption bug if HWPTs are used for attach"
* tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd:
iommufd: Do not add the same hwpt to the ioas->hwpt_list twice
iommufd: Make sure to zero vfio_iommu_type1_info before copying to user
vfio: Support VFIO_NOIOMMU with iommufd
iommufd: Add three missing structures in ucmd_buffer
selftests: iommu: Fix test_cmd_destroy_access() call in user_copy
iommu: Remove IOMMU_CAP_INTR_REMAP
irq/s390: Add arch_is_isolated_msi() for s390
iommu/x86: Replace IOMMU_CAP_INTR_REMAP with IRQ_DOMAIN_FLAG_ISOLATED_MSI
genirq/msi: Rename IRQ_DOMAIN_MSI_REMAP to IRQ_DOMAIN_ISOLATED_MSI
genirq/irqdomain: Remove unused irq_domain_check_msi_remap() code
iommufd: Convert to msi_device_has_isolated_msi()
vfio/type1: Convert to iommu_group_has_isolated_msi()
iommu: Add iommu_group_has_isolated_msi()
genirq/msi: Add msi_device_has_isolated_msi()
Commit 29b3283972 ("iommu/vt-d: Do not use flush-queue when caching-mode
is on") forced default domains to be strict mode as long as IOMMU
caching-mode is flagged. The reason for doing this is that when vIOMMU
uses VT-d caching mode to synchronize shadowing page tables, the strict
mode shows better performance.
However, this optimization is orthogonal to the first-level page table
because the Intel VT-d architecture does not define the caching mode of
the first-level page table. Refer to VT-d spec, section 6.1, "When the
CM field is reported as Set, any software updates to remapping
structures other than first-stage mapping (including updates to not-
present entries or present entries whose programming resulted in
translation faults) requires explicit invalidation of the caches."
Exclude the first-level page table from this optimization.
Generally using first-stage translation in vIOMMU implies nested
translation enabled in the physical IOMMU. In this case the first-stage
page table is wholly captured by the guest. The vIOMMU only needs to
transfer the cache invalidations on vIOMMU to the physical IOMMU.
Forcing the default domain to strict mode will cause more frequent
cache invalidations, resulting in performance degradation. In a real
performance benchmark test measured by iperf receive, the performance
result on Sapphire Rapids 100Gb NIC shows:
w/ this fix ~51 Gbits/s, w/o this fix ~39.3 Gbits/s.
Theoretically a first-stage IOMMU page table can still be shadowed
in absence of the caching mode, e.g. with host write-protecting guest
IOMMU page table to synchronize changed PTEs with the physical
IOMMU page table. In this case the shadowing overhead is decoupled
from emulating IOTLB invalidation then the overhead of the latter part
is solely decided by the frequency of IOTLB invalidations. Hence
allowing guest default dma domain to be lazy can also benefit the
overall performance by reducing the total VM-exit numbers.
Fixes: 29b3283972 ("iommu/vt-d: Do not use flush-queue when caching-mode is on")
Reported-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Suggested-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20230214025618.2292889-1-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Intel IOMMU driver implements IOTLB flush queue with domain selective
or PASID selective invalidations. In this case there's no need to track
IOVA page range and sync IOTLBs, which may cause significant performance
hit.
This patch adds a check to avoid IOVA gather page and IOTLB sync for
the lazy path.
The performance difference on Sapphire Rapids 100Gb NIC is improved by
the following (as measured by iperf send):
w/o this fix~48 Gbits/s. with this fix ~54 Gbits/s
Cc: <stable@vger.kernel.org>
Fixes: 2a2b8eaa5b ("iommu: Handle freelists when using deferred flushing in iommu drivers")
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Tested-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Sanjay Kumar <sanjay.k.kumar@intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20230209175330.1783556-1-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Roll back all previous actions in error paths of intel_iommu_enable_sva()
and intel_iommu_disable_sva().
Fixes: d5b9e4bfe0 ("iommu/vt-d: Report prq to io-pgfault framework")
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230208051559.700109-1-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Register and enable an IOMMU perfmon for each active IOMMU device.
The failure of IOMMU perfmon registration doesn't impact other
functionalities of an IOMMU device.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20230128200428.1459118-8-kan.liang@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The Enhanced Command Register is to submit command and operand of
enhanced commands to DMA Remapping hardware. It can supports up to 256
enhanced commands.
There is a HW register to indicate the availability of all 256 enhanced
commands. Each bit stands for each command. But there isn't an existing
interface to read/write all 256 bits. Introduce the u64 ecmdcap[4] to
store the existence of each enhanced command. Read 4 times to get all of
them in map_iommu().
Add a helper to facilitate an enhanced command launch. Make sure hardware
complete the command. Also add a helper to facilitate the check of PMU
essentials. These helpers will be used later.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20230128200428.1459118-4-kan.liang@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
There's no need to have a public header for Intel SVA implementation.
The device driver should interact with Intel SVA implementation via
the IOMMU generic APIs.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Link: https://lore.kernel.org/r/20230109014955.147068-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Jason Gunthorpe says:
====================
iommufd follows the same design as KVM and uses memory cgroups to limit
the amount of kernel memory a iommufd file descriptor can pin down. The
various internal data structures already use GFP_KERNEL_ACCOUNT to charge
its own memory.
However, one of the biggest consumers of kernel memory is the IOPTEs
stored under the iommu_domain and these allocations are not tracked.
This series is the first step in fixing it.
The iommu driver contract already includes a 'gfp' argument to the
map_pages op, allowing iommufd to specify GFP_KERNEL_ACCOUNT and then
having the driver allocate the IOPTE tables with that flag will capture a
significant amount of the allocations.
Update the iommu_map() API to pass in the GFP argument, and fix all call
sites. Replace iommu_map_atomic().
Audit the "enterprise" iommu drivers to make sure they do the right thing.
Intel and S390 ignore the GFP argument and always use GFP_ATOMIC. This is
problematic for iommufd anyhow, so fix it. AMD and ARM SMMUv2/3 are
already correct.
A follow up series will be needed to capture the allocations made when the
iommu_domain itself is allocated, which will complete the job.
====================
* 'iommu-memory-accounting' of ssh://gitolite.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
iommu/s390: Use GFP_KERNEL in sleepable contexts
iommu/s390: Push the gfp parameter to the kmem_cache_alloc()'s
iommu/intel: Use GFP_KERNEL in sleepable contexts
iommu/intel: Support the gfp argument to the map_pages op
iommu/intel: Add a gfp parameter to alloc_pgtable_page()
iommufd: Use GFP_KERNEL_ACCOUNT for iommu_map()
iommu/dma: Use the gfp parameter in __iommu_dma_alloc_noncontiguous()
iommu: Add a gfp parameter to iommu_map_sg()
iommu: Remove iommu_map_atomic()
iommu: Add a gfp parameter to iommu_map()
Link: https://lore.kernel.org/linux-iommu/0-v3-76b587fe28df+6e3-iommu_map_gfp_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
These contexts are sleepable, so use the proper annotation. The GFP_ATOMIC
was added mechanically in the prior patches.
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/8-v3-76b587fe28df+6e3-iommu_map_gfp_jgg@nvidia.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Flow it down to alloc_pgtable_page() via pfn_to_dma_pte() and
__domain_mapping().
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/7-v3-76b587fe28df+6e3-iommu_map_gfp_jgg@nvidia.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This is eventually called by iommufd through intel_iommu_map_pages() and
it should not be forced to atomic. Push the GFP_ATOMIC to all callers.
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/6-v3-76b587fe28df+6e3-iommu_map_gfp_jgg@nvidia.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
On x86 platforms when the HW can support interrupt remapping the iommu
driver creates an irq_domain for the IR hardware and creates a child MSI
irq_domain.
When the global irq_remapping_enabled is set, the IR MSI domain is
assigned to the PCI devices (by intel_irq_remap_add_device(), or
amd_iommu_set_pci_msi_domain()) making those devices have the isolated MSI
property.
Due to how interrupt domains work, setting IRQ_DOMAIN_FLAG_ISOLATED_MSI on
the parent IR domain will cause all struct devices attached to it to
return true from msi_device_has_isolated_msi(). This replaces the
IOMMU_CAP_INTR_REMAP flag as all places using IOMMU_CAP_INTR_REMAP also
call msi_device_has_isolated_msi()
Set the flag and delete the cap.
Link: https://lore.kernel.org/r/7-v3-3313bb5dd3a3+10f11-secure_msi_jgg@nvidia.com
Tested-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Including:
- Core code:
- map/unmap_pages() cleanup
- SVA and IOPF refactoring
- Clean up and document return codes from device/domain
attachment code
- AMD driver:
- Rework and extend parsing code for ivrs_ioapic, ivrs_hpet
and ivrs_acpihid command line options
- Some smaller cleanups
- Intel driver:
- Blocking domain support
- Cleanups
- S390 driver:
- Fixes and improvements for attach and aperture handling
- PAMU driver:
- Resource leak fix and cleanup
- Rockchip driver:
- Page table permission bit fix
- Mediatek driver:
- Improve safety from invalid dts input
- Smaller fixes and improvements
- Exynos driver:
- Fix driver initialization sequence
- Sun50i driver:
- Remove IOMMU_DOMAIN_IDENTITY as it has not been working
forever
- Various other fixes
-----BEGIN PGP SIGNATURE-----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=ZM/m
-----END PGP SIGNATURE-----
Merge tag 'iommu-updates-v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu
Pull iommu updates from Joerg Roedel:
"Core code:
- map/unmap_pages() cleanup
- SVA and IOPF refactoring
- Clean up and document return codes from device/domain attachment
AMD driver:
- Rework and extend parsing code for ivrs_ioapic, ivrs_hpet and
ivrs_acpihid command line options
- Some smaller cleanups
Intel driver:
- Blocking domain support
- Cleanups
S390 driver:
- Fixes and improvements for attach and aperture handling
PAMU driver:
- Resource leak fix and cleanup
Rockchip driver:
- Page table permission bit fix
Mediatek driver:
- Improve safety from invalid dts input
- Smaller fixes and improvements
Exynos driver:
- Fix driver initialization sequence
Sun50i driver:
- Remove IOMMU_DOMAIN_IDENTITY as it has not been working forever
- Various other fixes"
* tag 'iommu-updates-v6.2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (74 commits)
iommu/mediatek: Fix forever loop in error handling
iommu/mediatek: Fix crash on isr after kexec()
iommu/sun50i: Remove IOMMU_DOMAIN_IDENTITY
iommu/amd: Fix typo in macro parameter name
iommu/mediatek: Remove unused "mapping" member from mtk_iommu_data
iommu/mediatek: Improve safety for mediatek,smi property in larb nodes
iommu/mediatek: Validate number of phandles associated with "mediatek,larbs"
iommu/mediatek: Add error path for loop of mm_dts_parse
iommu/mediatek: Use component_match_add
iommu/mediatek: Add platform_device_put for recovering the device refcnt
iommu/fsl_pamu: Fix resource leak in fsl_pamu_probe()
iommu/vt-d: Use real field for indication of first level
iommu/vt-d: Remove unnecessary domain_context_mapped()
iommu/vt-d: Rename domain_add_dev_info()
iommu/vt-d: Rename iommu_disable_dev_iotlb()
iommu/vt-d: Add blocking domain support
iommu/vt-d: Add device_block_translation() helper
iommu/vt-d: Allocate pasid table in device probe path
iommu/amd: Check return value of mmu_notifier_register()
iommu/amd: Fix pci device refcount leak in ppr_notifier()
...
iommufd is the user API to control the IOMMU subsystem as it relates to
managing IO page tables that point at user space memory.
It takes over from drivers/vfio/vfio_iommu_type1.c (aka the VFIO
container) which is the VFIO specific interface for a similar idea.
We see a broad need for extended features, some being highly IOMMU device
specific:
- Binding iommu_domain's to PASID/SSID
- Userspace IO page tables, for ARM, x86 and S390
- Kernel bypassed invalidation of user page tables
- Re-use of the KVM page table in the IOMMU
- Dirty page tracking in the IOMMU
- Runtime Increase/Decrease of IOPTE size
- PRI support with faults resolved in userspace
Many of these HW features exist to support VM use cases - for instance the
combination of PASID, PRI and Userspace IO Page Tables allows an
implementation of DMA Shared Virtual Addressing (vSVA) within a
guest. Dirty tracking enables VM live migration with SRIOV devices and
PASID support allow creating "scalable IOV" devices, among other things.
As these features are fundamental to a VM platform they need to be
uniformly exposed to all the driver families that do DMA into VMs, which
is currently VFIO and VDPA.
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY5ct7wAKCRCFwuHvBreF
YZZ5AQDciXfcgXLt0UBEmWupNb0f/asT6tk717pdsKm8kAZMNAEAsIyLiKT5HqGl
s7fAu+CQ1pr9+9NKGevD+frw8Solsw4=
=jJkd
-----END PGP SIGNATURE-----
Merge tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd
Pull iommufd implementation from Jason Gunthorpe:
"iommufd is the user API to control the IOMMU subsystem as it relates
to managing IO page tables that point at user space memory.
It takes over from drivers/vfio/vfio_iommu_type1.c (aka the VFIO
container) which is the VFIO specific interface for a similar idea.
We see a broad need for extended features, some being highly IOMMU
device specific:
- Binding iommu_domain's to PASID/SSID
- Userspace IO page tables, for ARM, x86 and S390
- Kernel bypassed invalidation of user page tables
- Re-use of the KVM page table in the IOMMU
- Dirty page tracking in the IOMMU
- Runtime Increase/Decrease of IOPTE size
- PRI support with faults resolved in userspace
Many of these HW features exist to support VM use cases - for instance
the combination of PASID, PRI and Userspace IO Page Tables allows an
implementation of DMA Shared Virtual Addressing (vSVA) within a guest.
Dirty tracking enables VM live migration with SRIOV devices and PASID
support allow creating "scalable IOV" devices, among other things.
As these features are fundamental to a VM platform they need to be
uniformly exposed to all the driver families that do DMA into VMs,
which is currently VFIO and VDPA"
For more background, see the extended explanations in Jason's pull request:
https://lore.kernel.org/lkml/Y5dzTU8dlmXTbzoJ@nvidia.com/
* tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd: (62 commits)
iommufd: Change the order of MSI setup
iommufd: Improve a few unclear bits of code
iommufd: Fix comment typos
vfio: Move vfio group specific code into group.c
vfio: Refactor dma APIs for emulated devices
vfio: Wrap vfio group module init/clean code into helpers
vfio: Refactor vfio_device open and close
vfio: Make vfio_device_open() truly device specific
vfio: Swap order of vfio_device_container_register() and open_device()
vfio: Set device->group in helper function
vfio: Create wrappers for group register/unregister
vfio: Move the sanity check of the group to vfio_create_group()
vfio: Simplify vfio_create_group()
iommufd: Allow iommufd to supply /dev/vfio/vfio
vfio: Make vfio_container optionally compiled
vfio: Move container related MODULE_ALIAS statements into container.c
vfio-iommufd: Support iommufd for emulated VFIO devices
vfio-iommufd: Support iommufd for physical VFIO devices
vfio-iommufd: Allow iommufd to be used in place of a container fd
vfio: Use IOMMU_CAP_ENFORCE_CACHE_COHERENCY for vfio_file_enforced_coherent()
...
Impacted QAT device IDs that need extra dtlb flush quirk is ranging
from 0x4940 to 0x4943. After bitwise AND device ID with 0xfffc the
result should be 0x4940 instead of 0x494c to identify these devices.
Fixes: e65a6897be ("iommu/vt-d: Add a fix for devices need extra dtlb flush")
Reported-by: Raghunathan Srinivasan <raghunathan.srinivasan@intel.com>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20221203005610.2927487-1-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Resolve conflicts in drivers/vfio/vfio_main.c by using the iommfd version.
The rc fix was done a different way when iommufd patches reworked this
code.
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
for_each_pci_dev() is implemented by pci_get_device(). The comment of
pci_get_device() says that it will increase the reference count for the
returned pci_dev and also decrease the reference count for the input
pci_dev @from if it is not NULL.
If we break for_each_pci_dev() loop with pdev not NULL, we need to call
pci_dev_put() to decrease the reference count. Add the missing
pci_dev_put() before 'return true' to avoid reference count leak.
Fixes: 89a6079df7 ("iommu/vt-d: Force IOMMU on for platform opt in hint")
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Link: https://lore.kernel.org/r/20221121113649.190393-2-wangxiongfeng2@huawei.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
QAT devices on Intel Sapphire Rapids and Emerald Rapids have a defect in
address translation service (ATS). These devices may inadvertently issue
ATS invalidation completion before posted writes initiated with
translated address that utilized translations matching the invalidation
address range, violating the invalidation completion ordering.
This patch adds an extra device TLB invalidation for the affected devices,
it is needed to ensure no more posted writes with translated address
following the invalidation completion. Therefore, the ordering is
preserved and data-corruption is prevented.
Device TLBs are invalidated under the following six conditions:
1. Device driver does DMA API unmap IOVA
2. Device driver unbind a PASID from a process, sva_unbind_device()
3. PASID is torn down, after PASID cache is flushed. e.g. process
exit_mmap() due to crash
4. Under SVA usage, called by mmu_notifier.invalidate_range() where
VM has to free pages that were unmapped
5. userspace driver unmaps a DMA buffer
6. Cache invalidation in vSVA usage (upcoming)
For #1 and #2, device drivers are responsible for stopping DMA traffic
before unmap/unbind. For #3, iommu driver gets mmu_notifier to
invalidate TLB the same way as normal user unmap which will do an extra
invalidation. The dTLB invalidation after PASID cache flush does not
need an extra invalidation.
Therefore, we only need to deal with #4 and #5 in this patch. #1 is also
covered by this patch due to common code path with #5.
Tested-by: Yuzhang Luo <yuzhang.luo@intel.com>
Reviewed-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Link: https://lore.kernel.org/r/20221130062449.1360063-1-jacob.jun.pan@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
This queries if a domain linked to a device should expect to support
enforce_cache_coherency() so iommufd can negotiate the rules for when a
domain should be shared or not.
For iommufd a device that declares IOMMU_CAP_ENFORCE_CACHE_COHERENCY will
not be attached to a domain that does not support it.
Link: https://lore.kernel.org/r/1-v6-a196d26f289e+11787-iommufd_jgg@nvidia.com
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Tested-by: Yi Liu <yi.l.liu@intel.com>
Tested-by: Lixiao Yang <lixiao.yang@intel.com>
Tested-by: Matthew Rosato <mjrosato@linux.ibm.com>
Tested-by: Yu He <yu.he@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The dmar_domain uses bit field members to indicate the behaviors. Add
a bit field for using first level and remove the flags member to avoid
duplication.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20221118132451.114406-8-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
