The domain ID allocator is currently protected by a spin_lock. However,
ida_alloc_range can potentially block if it needs to allocate memory to
grow its internal structures.
Replace the spin_lock with a mutex which allows sleep on block. Thus,
the memory allocation flags can be updated from GFP_ATOMIC to GFP_KERNEL
to allow blocking memory allocations if necessary.
Introduce a new mutex, did_lock, specifically for protecting the domain
ida. The existing spinlock will remain for protecting other intel_iommu
fields.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20250430021135.2370244-3-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Switch the intel iommu driver to use the ida mechanism for managing domain
IDs, replacing the previous fixed-size bitmap.
The previous approach allocated a bitmap large enough to cover the maximum
number of domain IDs supported by the hardware, regardless of the actual
number of domains in use. This led to unnecessary memory consumption,
especially on systems supporting a large number of iommu units but only
utilizing a small number of domain IDs.
The ida allocator dynamically manages the allocation and freeing of integer
IDs, only consuming memory for the IDs that are currently in use. This
significantly optimizes memory usage compared to the fixed-size bitmap.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20250430021135.2370244-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Update iopf enablement in the driver to use the new method, similar to
the arm-smmu-v3 driver. Enable iopf support when any domain with an
iopf_handler is attached, and disable it when the domain is removed.
Place all the logic for controlling the PRI and iopf queue in the domain
set/remove/replace paths. Keep track of the number of domains set to the
device and PASIDs that require iopf. When the first domain requiring iopf
is attached, add the device to the iopf queue and enable PRI. When the
last domain is removed, remove it from the iopf queue and disable PRI.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Yi Liu <yi.l.liu@intel.com>
Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Link: https://lore.kernel.org/r/20250418080130.1844424-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Update PRI enablement to use the new method, similar to the amd iommu
driver. Enable PRI in the device probe path and disable it when the device
is released. PRI is enabled throughout the device's iommu lifecycle. The
infrastructure for the iommu subsystem to handle iopf requests is created
during iopf enablement and released during iopf disablement. All invalid
page requests from the device are automatically handled by the iommu
subsystem if iopf is not enabled. Add iopf_refcount to track the iopf
enablement.
Convert the return type of intel_iommu_disable_iopf() to void, as there
is no way to handle a failure when disabling this feature. Make
intel_iommu_enable/disable_iopf() helpers global, as they will be used
beyond the current file in the subsequent patch.
The iopf_refcount is not protected by any lock. This is acceptable, as
there is no concurrent access to it in the current code. The following
patch will address this by moving it to the domain attach/detach paths,
which are protected by the iommu group mutex.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Link: https://lore.kernel.org/r/20250228092631.3425464-6-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
If all the inlines are unwound virt_to_dma_pfn() is simply:
return page_to_pfn(virt_to_page(p)) << (PAGE_SHIFT - VTD_PAGE_SHIFT);
Which can be re-arranged to:
(page_to_pfn(virt_to_page(p)) << PAGE_SHIFT) >> VTD_PAGE_SHIFT
The only caller is:
((uint64_t)virt_to_dma_pfn(tmp_page) << VTD_PAGE_SHIFT)
re-arranged to:
((page_to_pfn(virt_to_page(tmp_page)) << PAGE_SHIFT) >> VTD_PAGE_SHIFT)
<< VTD_PAGE_SHIFT
Which simplifies to:
page_to_pfn(virt_to_page(tmp_page)) << PAGE_SHIFT
That is the same as virt_to_phys(tmp_page), so just remove all of this.
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/8-v3-e797f4dc6918+93057-iommu_pages_jgg@nvidia.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
It turns out all the drivers that are using this immediately call into
another function, so just make that function directly into the op. This
makes paging=NULL for domain_alloc_user and we can remove the argument in
the next patch.
The function mirrors the similar op in the viommu that allocates a nested
domain on top of the viommu's nesting parent. This version supports cases
where a viommu is not being used.
Link: https://patch.msgid.link/r/1-v1-c252ebdeb57b+329-iommu_paging_flags_jgg@nvidia.com
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The domain_add_dev_pasid() and domain_remove_dev_pasid() are added to
consolidate the adding/removing of the struct dev_pasid_info. Besides,
it includes the cache tag assign/unassign as well.
This also prepares for adding domain replacement for pasid. The
set_dev_pasid callbacks need to deal with the dev_pasid_info for both old
and new domain. These two helpers make the life easier.
intel_iommu_set_dev_pasid() and intel_svm_set_dev_pasid() are updated to
use the helpers.
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Link: https://lore.kernel.org/r/20241107122234.7424-6-yi.l.liu@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
IO page faults are no longer dependent on CONFIG_INTEL_IOMMU_SVM. Move
all Page Request Queue (PRQ) functions that handle prq events to a new
file in drivers/iommu/intel/prq.c. The page_req_des struct is now
declared in drivers/iommu/intel/prq.c.
No functional changes are intended. This is a preparation patch to
enable the use of IO page faults outside the SVM/PASID use cases.
Signed-off-by: Joel Granados <joel.granados@kernel.org>
Link: https://lore.kernel.org/r/20241015-jag-iopfv8-v4-1-b696ca89ba29@kernel.org
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
GCC is not happy with the current code, e.g.:
.../iommu/intel/dmar.c:1063:9: note: ‘sprintf’ output between 6 and 15 bytes into a destination of size 13
1063 | sprintf(iommu->name, "dmar%d", iommu->seq_id);
When `make W=1` is supplied, this prevents kernel building. Fix it by
increasing the buffer size for device name and use sizeoF() instead of
hard coded constants.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20241014104529.4025937-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The driver now supports domain_alloc_paging, ensuring that a valid device
pointer is provided whenever a paging domain is allocated. Additionally,
the dmar_domain attributes are set up at the time of allocation.
Consistent with the established semantics in the IOMMU core, if a domain is
attached to a device and found to be incompatible with the IOMMU hardware
capabilities, the operation will return an -EINVAL error. This implicitly
advises the caller to allocate a new domain for the device and attempt the
domain attachment again.
Rename prepare_domain_attach_device() to a more meaningful name.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20241021085125.192333-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Pull perf events updates from Ingo Molnar:
- Implement per-PMU context rescheduling to significantly improve
single-PMU performance, and related cleanups/fixes (Peter Zijlstra
and Namhyung Kim)
- Fix ancient bug resulting in a lot of events being dropped
erroneously at higher sampling frequencies (Luo Gengkun)
- uprobes enhancements:
- Implement RCU-protected hot path optimizations for better
performance:
"For baseline vs SRCU, peak througput increased from 3.7 M/s
(million uprobe triggerings per second) up to about 8 M/s. For
uretprobes it's a bit more modest with bump from 2.4 M/s to
5 M/s.
For SRCU vs RCU Tasks Trace, peak throughput for uprobes
increases further from 8 M/s to 10.3 M/s (+28%!), and for
uretprobes from 5.3 M/s to 5.8 M/s (+11%), as we have more
work to do on uretprobes side.
Even single-thread (no contention) performance is slightly
better: 3.276 M/s to 3.396 M/s (+3.5%) for uprobes, and 2.055
M/s to 2.174 M/s (+5.8%) for uretprobes."
(Andrii Nakryiko et al)
- Document mmap_lock, don't abuse get_user_pages_remote() (Oleg
Nesterov)
- Cleanups & fixes to prepare for future work:
- Remove uprobe_register_refctr()
- Simplify error handling for alloc_uprobe()
- Make uprobe_register() return struct uprobe *
- Fold __uprobe_unregister() into uprobe_unregister()
- Shift put_uprobe() from delete_uprobe() to uprobe_unregister()
- BPF: Fix use-after-free in bpf_uprobe_multi_link_attach()
(Oleg Nesterov)
- New feature & ABI extension: allow events to use PERF_SAMPLE READ
with inheritance, enabling sample based profiling of a group of
counters over a hierarchy of processes or threads (Ben Gainey)
- Intel uncore & power events updates:
- Add Arrow Lake and Lunar Lake support
- Add PERF_EV_CAP_READ_SCOPE
- Clean up and enhance cpumask and hotplug support
(Kan Liang)
- Add LNL uncore iMC freerunning support
- Use D0:F0 as a default device
(Zhenyu Wang)
- Intel PT: fix AUX snapshot handling race (Adrian Hunter)
- Misc fixes and cleanups (James Clark, Jiri Olsa, Oleg Nesterov and
Peter Zijlstra)
* tag 'perf-core-2024-09-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
dmaengine: idxd: Clean up cpumask and hotplug for perfmon
iommu/vt-d: Clean up cpumask and hotplug for perfmon
perf/x86/intel/cstate: Clean up cpumask and hotplug
perf: Add PERF_EV_CAP_READ_SCOPE
perf: Generic hotplug support for a PMU with a scope
uprobes: perform lockless SRCU-protected uprobes_tree lookup
rbtree: provide rb_find_rcu() / rb_find_add_rcu()
perf/uprobe: split uprobe_unregister()
uprobes: travers uprobe's consumer list locklessly under SRCU protection
uprobes: get rid of enum uprobe_filter_ctx in uprobe filter callbacks
uprobes: protected uprobe lifetime with SRCU
uprobes: revamp uprobe refcounting and lifetime management
bpf: Fix use-after-free in bpf_uprobe_multi_link_attach()
perf/core: Fix small negative period being ignored
perf: Really fix event_function_call() locking
perf: Optimize __pmu_ctx_sched_out()
perf: Add context time freeze
perf: Fix event_function_call() locking
perf: Extract a few helpers
perf: Optimize context reschedule for single PMU cases
...
Extracts IOTLB and Dev-IOTLB invalidation logic from cache tag flush
interfaces into dedicated helper functions. It prepares the codebase
for upcoming changes to support batched cache invalidations.
To enable direct use of qi_flush helpers in the new functions,
iommu->flush.flush_iotlb and quirk_extra_dev_tlb_flush() are opened up.
No functional changes are intended.
Co-developed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20240815065221.50328-3-tina.zhang@intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Separate the logic for constructing IOTLB and device TLB invalidation
descriptors from the qi_flush interfaces. New helpers, qi_desc(), are
introduced to encapsulate this common functionality.
Moving descriptor composition code to new helpers enables its reuse in
the upcoming qi_batch interfaces.
No functional changes are intended.
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Link: https://lore.kernel.org/r/20240815065221.50328-2-tina.zhang@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The has_iotlb_device flag was used to indicate if a domain had attached
devices with ATS enabled. Domains without this flag didn't require device
TLB invalidation during unmap operations, optimizing performance by
avoiding unnecessary device iteration.
With the introduction of cache tags, this flag is no longer needed. The
code to iterate over attached devices was removed by commit 06792d0679
("iommu/vt-d: Cleanup use of iommu_flush_iotlb_psi()").
Remove has_iotlb_device to avoid unnecessary code.
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240809055431.36513-5-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The struct intel_svm was used for keeping attached devices info for sva
domain. Since sva domain is a kind of iommu_domain, the struct
dmar_domain should centralize all info of a sva domain, including the
info of attached devices. Therefore, retire struct intel_svm and clean up
the code.
Besides, register mmu notifier callback in domain_alloc_sva() callback
which allows the memory management notifier lifetime to follow the lifetime
of the iommu_domain. Call mmu_notifier_put() in the domain free and defer
the real free to the mmu free_notifier callback.
Co-developed-by: Tina Zhang <tina.zhang@intel.com>
Signed-off-by: Tina Zhang <tina.zhang@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240416080656.60968-13-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The arch_invalidate_secondary_tlbs callback is called in the SVA mm
notification path. It invalidates all or a range of caches after the
CPU page table is modified. Use the cache tag helps in this path.
The mm_types defines vm_end as the first byte after the end address
which is different from the iommu gather API, hence convert the end
parameter from mm_types to iommu gather scheme before calling the
cache_tag helper.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240416080656.60968-10-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Add several helpers to invalidate the caches after mappings in the
affected domain are changed.
- cache_tag_flush_range() invalidates a range of caches after mappings
within this range are changed. It uses the page-selective cache
invalidation methods.
- cache_tag_flush_all() invalidates all caches tagged by a domain ID.
It uses the domain-selective cache invalidation methods.
- cache_tag_flush_range_np() invalidates a range of caches when new
mappings are created in the domain and the corresponding page table
entries change from non-present to present.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240416080656.60968-3-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Caching tag is a combination of tags used by the hardware to cache various
translations. Whenever a mapping in a domain is changed, the IOMMU driver
should invalidate the caches with the caching tags. The VT-d specification
describes caching tags in section 6.2.1, Tagging of Cached Translations.
Add interface to assign caching tags to an IOMMU domain when attached to a
RID or PASID, and unassign caching tags when a domain is detached from a
RID or PASID. All caching tags are listed in the per-domain tag list and
are protected by a dedicated lock.
In addition to the basic IOTLB and devTLB caching tag types, NESTING_IOTLB
and NESTING_DEVTLB tag types are also introduced. These tags are used for
caches that store translations for DMA accesses through a nested user
domain. They are affected by changes to mappings in the parent domain.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240416080656.60968-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
According to Intel VT-d specification revision 4.0, "Private Data"
field has been removed from Page Request/Response.
Since the private data field is not used in fault message, remove the
related definitions in page request descriptor and remove the related
code in page request/response handler, as Intel hasn't shipped any
products which support private data in the page request message.
Signed-off-by: Jingqi Liu <Jingqi.liu@intel.com>
Link: https://lore.kernel.org/r/20240308103811.76744-3-Jingqi.liu@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
In order to improve observability and accountability of IOMMU layer, we
must account the number of pages that are allocated by functions that
are calling directly into buddy allocator.
This is achieved by first wrapping the allocation related functions into a
separate inline functions in new file:
drivers/iommu/iommu-pages.h
Convert all page allocation calls under iommu/intel to use these new
functions.
Signed-off-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Acked-by: David Rientjes <rientjes@google.com>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Link: https://lore.kernel.org/r/20240413002522.1101315-2-pasha.tatashin@soleen.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The existing I/O page fault handler currently locates the PCI device by
calling pci_get_domain_bus_and_slot(). This function searches the list
of all PCI devices until the desired device is found. To improve lookup
efficiency, replace it with device_rbtree_find() to search the device
within the probed device rbtree.
The I/O page fault is initiated by the device, which does not have any
synchronization mechanism with the software to ensure that the device
stays in the probed device tree. Theoretically, a device could be released
by the IOMMU subsystem after device_rbtree_find() and before
iopf_get_dev_fault_param(), which would cause a use-after-free problem.
Add a mutex to synchronize the I/O page fault reporting path and the IOMMU
release device path. This lock doesn't introduce any performance overhead,
as the conflict between I/O page fault reporting and device releasing is
very rare.
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20240220065939.121116-3-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Use a red-black tree(rbtree) to track devices probed by the driver's
probe_device callback. These devices need to be looked up quickly by
a source ID when the hardware reports a fault, either recoverable or
unrecoverable.
Fault reporting paths are critical. Searching a list in this scenario
is inefficient, with an algorithm complexity of O(n). An rbtree is a
self-balancing binary search tree, offering an average search time
complexity of O(log(n)). This significant performance improvement
makes rbtrees a better choice.
Furthermore, rbtrees are implemented on a per-iommu basis, eliminating
the need for global searches and further enhancing efficiency in
critical fault paths. The rbtree is protected by a spin lock with
interrupts disabled to ensure thread-safe access even within interrupt
contexts.
Co-developed-by: Huang Jiaqing <jiaqing.huang@intel.com>
Signed-off-by: Huang Jiaqing <jiaqing.huang@intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20240220065939.121116-2-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Today the parent domain (s2_domain) is unaware of which DID's are
used by and which devices are attached to nested domains (s1_domain)
nested on it. This leads to a problem that some operations (flush
iotlb/devtlb and enable dirty tracking) on parent domain only apply to
DID's and devices directly tracked in the parent domain hence are
incomplete.
This tracks the nested domains in list in parent domain. With this,
operations on parent domain can loop the nested domains and refer to
the devices and iommu_array to ensure the operations on parent domain
take effect on all the affected devices and iommus.
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240208082307.15759-2-yi.l.liu@intel.com
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The iopf_group_response() should return void, as nothing can do anything
with the failure. This implies that ops->page_response() must also return
void; this is consistent with what the drivers do. The failure paths,
which are all integrity validations of the fault, should be WARN_ON'd,
not return codes.
If the iommu core fails to enqueue the fault, it should respond the fault
directly by calling ops->page_response() instead of returning an error
number and relying on the iommu drivers to do so. Consolidate the error
fault handling code in the core.
Co-developed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20240212012227.119381-16-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
