From 14678219cf4093e897ab353fd78eab7994d1be7d Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:35 +0800 Subject: iommu: Introduce domain attachment handle Currently, when attaching a domain to a device or its PASID, domain is stored within the iommu group. It could be retrieved for use during the window between attachment and detachment. With new features introduced, there's a need to store more information than just a domain pointer. This information essentially represents the association between a domain and a device. For example, the SVA code already has a custom struct iommu_sva which represents a bond between sva domain and a PASID of a device. Looking forward, the IOMMUFD needs a place to store the iommufd_device pointer in the core, so that the device object ID could be quickly retrieved in the critical fault handling path. Introduce domain attachment handle that explicitly represents the attachment relationship between a domain and a device or its PASID. Co-developed-by: Jason Gunthorpe Signed-off-by: Jason Gunthorpe Signed-off-by: Lu Baolu Reviewed-by: Jason Gunthorpe Reviewed-by: Kevin Tian Link: https://lore.kernel.org/r/20240702063444.105814-2-baolu.lu@linux.intel.com Signed-off-by: Will Deacon --- drivers/iommu/iommu-sva.c | 13 ++++++++----- drivers/iommu/iommu.c | 26 ++++++++++++++++---------- 2 files changed, 24 insertions(+), 15 deletions(-) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c index 18a35e798b72..0fb923254062 100644 --- a/drivers/iommu/iommu-sva.c +++ b/drivers/iommu/iommu-sva.c @@ -99,7 +99,9 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm /* Search for an existing domain. */ list_for_each_entry(domain, &mm->iommu_mm->sva_domains, next) { - ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid); + handle->handle.domain = domain; + ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid, + &handle->handle); if (!ret) { domain->users++; goto out; @@ -113,7 +115,9 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm goto out_free_handle; } - ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid); + handle->handle.domain = domain; + ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid, + &handle->handle); if (ret) goto out_free_domain; domain->users = 1; @@ -124,7 +128,6 @@ out: list_add(&handle->handle_item, &mm->iommu_mm->sva_handles); mutex_unlock(&iommu_sva_lock); handle->dev = dev; - handle->domain = domain; return handle; out_free_domain: @@ -147,7 +150,7 @@ EXPORT_SYMBOL_GPL(iommu_sva_bind_device); */ void iommu_sva_unbind_device(struct iommu_sva *handle) { - struct iommu_domain *domain = handle->domain; + struct iommu_domain *domain = handle->handle.domain; struct iommu_mm_data *iommu_mm = domain->mm->iommu_mm; struct device *dev = handle->dev; @@ -170,7 +173,7 @@ EXPORT_SYMBOL_GPL(iommu_sva_unbind_device); u32 iommu_sva_get_pasid(struct iommu_sva *handle) { - struct iommu_domain *domain = handle->domain; + struct iommu_domain *domain = handle->handle.domain; return mm_get_enqcmd_pasid(domain->mm); } diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 9df7cc75c1bc..a712b0cc3a1d 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -3352,16 +3352,17 @@ static void __iommu_remove_group_pasid(struct iommu_group *group, * @domain: the iommu domain. * @dev: the attached device. * @pasid: the pasid of the device. + * @handle: the attach handle. * * Return: 0 on success, or an error. */ int iommu_attach_device_pasid(struct iommu_domain *domain, - struct device *dev, ioasid_t pasid) + struct device *dev, ioasid_t pasid, + struct iommu_attach_handle *handle) { /* Caller must be a probed driver on dev */ struct iommu_group *group = dev->iommu_group; struct group_device *device; - void *curr; int ret; if (!domain->ops->set_dev_pasid) @@ -3382,11 +3383,12 @@ int iommu_attach_device_pasid(struct iommu_domain *domain, } } - curr = xa_cmpxchg(&group->pasid_array, pasid, NULL, domain, GFP_KERNEL); - if (curr) { - ret = xa_err(curr) ? : -EBUSY; + if (handle) + handle->domain = domain; + + ret = xa_insert(&group->pasid_array, pasid, handle, GFP_KERNEL); + if (ret) goto out_unlock; - } ret = __iommu_set_group_pasid(domain, group, pasid); if (ret) @@ -3414,7 +3416,7 @@ void iommu_detach_device_pasid(struct iommu_domain *domain, struct device *dev, mutex_lock(&group->mutex); __iommu_remove_group_pasid(group, pasid, domain); - WARN_ON(xa_erase(&group->pasid_array, pasid) != domain); + xa_erase(&group->pasid_array, pasid); mutex_unlock(&group->mutex); } EXPORT_SYMBOL_GPL(iommu_detach_device_pasid); @@ -3439,15 +3441,19 @@ struct iommu_domain *iommu_get_domain_for_dev_pasid(struct device *dev, { /* Caller must be a probed driver on dev */ struct iommu_group *group = dev->iommu_group; - struct iommu_domain *domain; + struct iommu_attach_handle *handle; + struct iommu_domain *domain = NULL; if (!group) return NULL; xa_lock(&group->pasid_array); - domain = xa_load(&group->pasid_array, pasid); + handle = xa_load(&group->pasid_array, pasid); + if (handle) + domain = handle->domain; + if (type && domain && domain->type != type) - domain = ERR_PTR(-EBUSY); + domain = NULL; xa_unlock(&group->pasid_array); return domain; -- cgit v1.2.3 From 3e7f57d1ef3f5fbed58974fae38d35e430f57d35 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:36 +0800 Subject: iommu: Remove sva handle list The struct sva_iommu represents an association between an SVA domain and a PASID of a device. It's stored in the iommu group's pasid array and also tracked by a list in the per-mm data structure. Removes duplicate tracking of sva_iommu by eliminating the list. Signed-off-by: Lu Baolu Reviewed-by: Jason Gunthorpe Reviewed-by: Kevin Tian Link: https://lore.kernel.org/r/20240702063444.105814-3-baolu.lu@linux.intel.com Signed-off-by: Will Deacon --- drivers/iommu/iommu-priv.h | 3 +++ drivers/iommu/iommu-sva.c | 30 ++++++++++++++++++++---------- drivers/iommu/iommu.c | 31 +++++++++++++++++++++++++++++++ 3 files changed, 54 insertions(+), 10 deletions(-) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommu-priv.h b/drivers/iommu/iommu-priv.h index 5f731d994803..f1536a5ebb0d 100644 --- a/drivers/iommu/iommu-priv.h +++ b/drivers/iommu/iommu-priv.h @@ -28,4 +28,7 @@ void iommu_device_unregister_bus(struct iommu_device *iommu, const struct bus_type *bus, struct notifier_block *nb); +struct iommu_attach_handle *iommu_attach_handle_get(struct iommu_group *group, + ioasid_t pasid, + unsigned int type); #endif /* __LINUX_IOMMU_PRIV_H */ diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c index 0fb923254062..9b7f62517419 100644 --- a/drivers/iommu/iommu-sva.c +++ b/drivers/iommu/iommu-sva.c @@ -41,7 +41,6 @@ static struct iommu_mm_data *iommu_alloc_mm_data(struct mm_struct *mm, struct de } iommu_mm->pasid = pasid; INIT_LIST_HEAD(&iommu_mm->sva_domains); - INIT_LIST_HEAD(&iommu_mm->sva_handles); /* * Make sure the write to mm->iommu_mm is not reordered in front of * initialization to iommu_mm fields. If it does, readers may see a @@ -69,11 +68,16 @@ static struct iommu_mm_data *iommu_alloc_mm_data(struct mm_struct *mm, struct de */ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm) { + struct iommu_group *group = dev->iommu_group; + struct iommu_attach_handle *attach_handle; struct iommu_mm_data *iommu_mm; struct iommu_domain *domain; struct iommu_sva *handle; int ret; + if (!group) + return ERR_PTR(-ENODEV); + mutex_lock(&iommu_sva_lock); /* Allocate mm->pasid if necessary. */ @@ -83,12 +87,22 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm goto out_unlock; } - list_for_each_entry(handle, &mm->iommu_mm->sva_handles, handle_item) { - if (handle->dev == dev) { - refcount_inc(&handle->users); - mutex_unlock(&iommu_sva_lock); - return handle; + /* A bond already exists, just take a reference`. */ + attach_handle = iommu_attach_handle_get(group, iommu_mm->pasid, IOMMU_DOMAIN_SVA); + if (!IS_ERR(attach_handle)) { + handle = container_of(attach_handle, struct iommu_sva, handle); + if (attach_handle->domain->mm != mm) { + ret = -EBUSY; + goto out_unlock; } + refcount_inc(&handle->users); + mutex_unlock(&iommu_sva_lock); + return handle; + } + + if (PTR_ERR(attach_handle) != -ENOENT) { + ret = PTR_ERR(attach_handle); + goto out_unlock; } handle = kzalloc(sizeof(*handle), GFP_KERNEL); @@ -99,7 +113,6 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm /* Search for an existing domain. */ list_for_each_entry(domain, &mm->iommu_mm->sva_domains, next) { - handle->handle.domain = domain; ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid, &handle->handle); if (!ret) { @@ -115,7 +128,6 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm goto out_free_handle; } - handle->handle.domain = domain; ret = iommu_attach_device_pasid(domain, dev, iommu_mm->pasid, &handle->handle); if (ret) @@ -125,7 +137,6 @@ struct iommu_sva *iommu_sva_bind_device(struct device *dev, struct mm_struct *mm out: refcount_set(&handle->users, 1); - list_add(&handle->handle_item, &mm->iommu_mm->sva_handles); mutex_unlock(&iommu_sva_lock); handle->dev = dev; return handle; @@ -159,7 +170,6 @@ void iommu_sva_unbind_device(struct iommu_sva *handle) mutex_unlock(&iommu_sva_lock); return; } - list_del(&handle->handle_item); iommu_detach_device_pasid(domain, dev, iommu_mm->pasid); if (--domain->users == 0) { diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index a712b0cc3a1d..7890bd21dff6 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -3486,3 +3486,34 @@ void iommu_free_global_pasid(ioasid_t pasid) ida_free(&iommu_global_pasid_ida, pasid); } EXPORT_SYMBOL_GPL(iommu_free_global_pasid); + +/** + * iommu_attach_handle_get - Return the attach handle + * @group: the iommu group that domain was attached to + * @pasid: the pasid within the group + * @type: matched domain type, 0 for any match + * + * Return handle or ERR_PTR(-ENOENT) on none, ERR_PTR(-EBUSY) on mismatch. + * + * Return the attach handle to the caller. The life cycle of an iommu attach + * handle is from the time when the domain is attached to the time when the + * domain is detached. Callers are required to synchronize the call of + * iommu_attach_handle_get() with domain attachment and detachment. The attach + * handle can only be used during its life cycle. + */ +struct iommu_attach_handle * +iommu_attach_handle_get(struct iommu_group *group, ioasid_t pasid, unsigned int type) +{ + struct iommu_attach_handle *handle; + + xa_lock(&group->pasid_array); + handle = xa_load(&group->pasid_array, pasid); + if (!handle) + handle = ERR_PTR(-ENOENT); + else if (type && handle->domain->type != type) + handle = ERR_PTR(-EBUSY); + xa_unlock(&group->pasid_array); + + return handle; +} +EXPORT_SYMBOL_NS_GPL(iommu_attach_handle_get, IOMMUFD_INTERNAL); -- cgit v1.2.3 From 06cdcc32d65759d42c6340700796e2906045b6a5 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:37 +0800 Subject: iommu: Add attach handle to struct iopf_group Previously, the domain that a page fault targets is stored in an iopf_group, which represents a minimal set of page faults. With the introduction of attach handle, replace the domain with the handle so that the fault handler can obtain more information as needed when handling the faults. iommu_report_device_fault() is currently used for SVA page faults, which handles the page fault in an internal cycle. The domain is retrieved with iommu_get_domain_for_dev_pasid() if the pasid in the fault message is valid. This doesn't work in IOMMUFD case, where if the pasid table of a device is wholly managed by user space, there is no domain attached to the PASID of the device, and all page faults are forwarded through a NESTING domain attaching to RID. Add a static flag in iommu ops, which indicates if the IOMMU driver supports user-managed PASID tables. In the iopf deliver path, if no attach handle found for the iopf PASID, roll back to RID domain when the IOMMU driver supports this capability. iommu_get_domain_for_dev_pasid() is no longer used and can be removed. Signed-off-by: Lu Baolu Reviewed-by: Jason Gunthorpe Reviewed-by: Kevin Tian Link: https://lore.kernel.org/r/20240702063444.105814-4-baolu.lu@linux.intel.com Signed-off-by: Will Deacon --- drivers/iommu/io-pgfault.c | 61 ++++++++++++++++++++++++++-------------------- drivers/iommu/iommu-sva.c | 3 ++- drivers/iommu/iommu.c | 39 ----------------------------- 3 files changed, 36 insertions(+), 67 deletions(-) (limited to 'drivers/iommu') diff --git a/drivers/iommu/io-pgfault.c b/drivers/iommu/io-pgfault.c index 06d78fcc79fd..7c9011992d3f 100644 --- a/drivers/iommu/io-pgfault.c +++ b/drivers/iommu/io-pgfault.c @@ -59,30 +59,6 @@ void iopf_free_group(struct iopf_group *group) } EXPORT_SYMBOL_GPL(iopf_free_group); -static struct iommu_domain *get_domain_for_iopf(struct device *dev, - struct iommu_fault *fault) -{ - struct iommu_domain *domain; - - if (fault->prm.flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID) { - domain = iommu_get_domain_for_dev_pasid(dev, fault->prm.pasid, 0); - if (IS_ERR(domain)) - domain = NULL; - } else { - domain = iommu_get_domain_for_dev(dev); - } - - if (!domain || !domain->iopf_handler) { - dev_warn_ratelimited(dev, - "iopf (pasid %d) without domain attached or handler installed\n", - fault->prm.pasid); - - return NULL; - } - - return domain; -} - /* Non-last request of a group. Postpone until the last one. */ static int report_partial_fault(struct iommu_fault_param *fault_param, struct iommu_fault *fault) @@ -206,20 +182,51 @@ void iommu_report_device_fault(struct device *dev, struct iopf_fault *evt) if (group == &abort_group) goto err_abort; - group->domain = get_domain_for_iopf(dev, fault); - if (!group->domain) + if (fault->prm.flags & IOMMU_FAULT_PAGE_REQUEST_PASID_VALID) { + group->attach_handle = iommu_attach_handle_get(dev->iommu_group, + fault->prm.pasid, + 0); + if (IS_ERR(group->attach_handle)) { + const struct iommu_ops *ops = dev_iommu_ops(dev); + + if (!ops->user_pasid_table) + goto err_abort; + + /* + * The iommu driver for this device supports user- + * managed PASID table. Therefore page faults for + * any PASID should go through the NESTING domain + * attached to the device RID. + */ + group->attach_handle = + iommu_attach_handle_get(dev->iommu_group, + IOMMU_NO_PASID, + IOMMU_DOMAIN_NESTED); + if (IS_ERR(group->attach_handle)) + goto err_abort; + } + } else { + group->attach_handle = + iommu_attach_handle_get(dev->iommu_group, IOMMU_NO_PASID, 0); + if (IS_ERR(group->attach_handle)) + goto err_abort; + } + + if (!group->attach_handle->domain->iopf_handler) goto err_abort; /* * On success iopf_handler must call iopf_group_response() and * iopf_free_group() */ - if (group->domain->iopf_handler(group)) + if (group->attach_handle->domain->iopf_handler(group)) goto err_abort; return; err_abort: + dev_warn_ratelimited(dev, "iopf with pasid %d aborted\n", + fault->prm.pasid); iopf_group_response(group, IOMMU_PAGE_RESP_FAILURE); if (group == &abort_group) __iopf_free_group(group); diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c index 9b7f62517419..69cde094440e 100644 --- a/drivers/iommu/iommu-sva.c +++ b/drivers/iommu/iommu-sva.c @@ -272,7 +272,8 @@ static void iommu_sva_handle_iopf(struct work_struct *work) if (status != IOMMU_PAGE_RESP_SUCCESS) break; - status = iommu_sva_handle_mm(&iopf->fault, group->domain->mm); + status = iommu_sva_handle_mm(&iopf->fault, + group->attach_handle->domain->mm); } iopf_group_response(group, status); diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 7890bd21dff6..5a7e874abb36 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -3421,45 +3421,6 @@ void iommu_detach_device_pasid(struct iommu_domain *domain, struct device *dev, } EXPORT_SYMBOL_GPL(iommu_detach_device_pasid); -/* - * iommu_get_domain_for_dev_pasid() - Retrieve domain for @pasid of @dev - * @dev: the queried device - * @pasid: the pasid of the device - * @type: matched domain type, 0 for any match - * - * This is a variant of iommu_get_domain_for_dev(). It returns the existing - * domain attached to pasid of a device. Callers must hold a lock around this - * function, and both iommu_attach/detach_dev_pasid() whenever a domain of - * type is being manipulated. This API does not internally resolve races with - * attach/detach. - * - * Return: attached domain on success, NULL otherwise. - */ -struct iommu_domain *iommu_get_domain_for_dev_pasid(struct device *dev, - ioasid_t pasid, - unsigned int type) -{ - /* Caller must be a probed driver on dev */ - struct iommu_group *group = dev->iommu_group; - struct iommu_attach_handle *handle; - struct iommu_domain *domain = NULL; - - if (!group) - return NULL; - - xa_lock(&group->pasid_array); - handle = xa_load(&group->pasid_array, pasid); - if (handle) - domain = handle->domain; - - if (type && domain && domain->type != type) - domain = NULL; - xa_unlock(&group->pasid_array); - - return domain; -} -EXPORT_SYMBOL_GPL(iommu_get_domain_for_dev_pasid); - ioasid_t iommu_alloc_global_pasid(struct device *dev) { int ret; -- cgit v1.2.3 From 8519e689834a3ecf9a36332a9abb1844bd34e459 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:38 +0800 Subject: iommu: Extend domain attach group with handle support Unlike the SVA case where each PASID of a device has an SVA domain attached to it, the I/O page faults are handled by the fault handler of the SVA domain. The I/O page faults for a user page table might be handled by the domain attached to RID or the domain attached to the PASID, depending on whether the PASID table is managed by user space or kernel. As a result, there is a need for the domain attach group interfaces to have attach handle support. The attach handle will be forwarded to the fault handler of the user domain. Add some variants of the domain attaching group interfaces so that they could support the attach handle and export them for use in IOMMUFD. Signed-off-by: Lu Baolu Reviewed-by: Kevin Tian Link: https://lore.kernel.org/r/20240702063444.105814-5-baolu.lu@linux.intel.com Signed-off-by: Will Deacon --- drivers/iommu/iommu-priv.h | 8 ++++ drivers/iommu/iommu.c | 103 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 111 insertions(+) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommu-priv.h b/drivers/iommu/iommu-priv.h index f1536a5ebb0d..c37801c32f33 100644 --- a/drivers/iommu/iommu-priv.h +++ b/drivers/iommu/iommu-priv.h @@ -31,4 +31,12 @@ void iommu_device_unregister_bus(struct iommu_device *iommu, struct iommu_attach_handle *iommu_attach_handle_get(struct iommu_group *group, ioasid_t pasid, unsigned int type); +int iommu_attach_group_handle(struct iommu_domain *domain, + struct iommu_group *group, + struct iommu_attach_handle *handle); +void iommu_detach_group_handle(struct iommu_domain *domain, + struct iommu_group *group); +int iommu_replace_group_handle(struct iommu_group *group, + struct iommu_domain *new_domain, + struct iommu_attach_handle *handle); #endif /* __LINUX_IOMMU_PRIV_H */ diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 5a7e874abb36..8484285fbaa8 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -3478,3 +3478,106 @@ iommu_attach_handle_get(struct iommu_group *group, ioasid_t pasid, unsigned int return handle; } EXPORT_SYMBOL_NS_GPL(iommu_attach_handle_get, IOMMUFD_INTERNAL); + +/** + * iommu_attach_group_handle - Attach an IOMMU domain to an IOMMU group + * @domain: IOMMU domain to attach + * @group: IOMMU group that will be attached + * @handle: attach handle + * + * Returns 0 on success and error code on failure. + * + * This is a variant of iommu_attach_group(). It allows the caller to provide + * an attach handle and use it when the domain is attached. This is currently + * used by IOMMUFD to deliver the I/O page faults. + */ +int iommu_attach_group_handle(struct iommu_domain *domain, + struct iommu_group *group, + struct iommu_attach_handle *handle) +{ + int ret; + + if (handle) + handle->domain = domain; + + mutex_lock(&group->mutex); + ret = xa_insert(&group->pasid_array, IOMMU_NO_PASID, handle, GFP_KERNEL); + if (ret) + goto err_unlock; + + ret = __iommu_attach_group(domain, group); + if (ret) + goto err_erase; + mutex_unlock(&group->mutex); + + return 0; +err_erase: + xa_erase(&group->pasid_array, IOMMU_NO_PASID); +err_unlock: + mutex_unlock(&group->mutex); + return ret; +} +EXPORT_SYMBOL_NS_GPL(iommu_attach_group_handle, IOMMUFD_INTERNAL); + +/** + * iommu_detach_group_handle - Detach an IOMMU domain from an IOMMU group + * @domain: IOMMU domain to attach + * @group: IOMMU group that will be attached + * + * Detach the specified IOMMU domain from the specified IOMMU group. + * It must be used in conjunction with iommu_attach_group_handle(). + */ +void iommu_detach_group_handle(struct iommu_domain *domain, + struct iommu_group *group) +{ + mutex_lock(&group->mutex); + __iommu_group_set_core_domain(group); + xa_erase(&group->pasid_array, IOMMU_NO_PASID); + mutex_unlock(&group->mutex); +} +EXPORT_SYMBOL_NS_GPL(iommu_detach_group_handle, IOMMUFD_INTERNAL); + +/** + * iommu_replace_group_handle - replace the domain that a group is attached to + * @group: IOMMU group that will be attached to the new domain + * @new_domain: new IOMMU domain to replace with + * @handle: attach handle + * + * This is a variant of iommu_group_replace_domain(). It allows the caller to + * provide an attach handle for the new domain and use it when the domain is + * attached. + */ +int iommu_replace_group_handle(struct iommu_group *group, + struct iommu_domain *new_domain, + struct iommu_attach_handle *handle) +{ + void *curr; + int ret; + + if (!new_domain) + return -EINVAL; + + mutex_lock(&group->mutex); + if (handle) { + ret = xa_reserve(&group->pasid_array, IOMMU_NO_PASID, GFP_KERNEL); + if (ret) + goto err_unlock; + } + + ret = __iommu_group_set_domain(group, new_domain); + if (ret) + goto err_release; + + curr = xa_store(&group->pasid_array, IOMMU_NO_PASID, handle, GFP_KERNEL); + WARN_ON(xa_is_err(curr)); + + mutex_unlock(&group->mutex); + + return 0; +err_release: + xa_release(&group->pasid_array, IOMMU_NO_PASID); +err_unlock: + mutex_unlock(&group->mutex); + return ret; +} +EXPORT_SYMBOL_NS_GPL(iommu_replace_group_handle, IOMMUFD_INTERNAL); -- cgit v1.2.3 From 07838f7fd529c8a6de44b601d4b7057e6c8d36ed Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:40 +0800 Subject: iommufd: Add iommufd fault object An iommufd fault object provides an interface for delivering I/O page faults to user space. These objects are created and destroyed by user space, and they can be associated with or dissociated from hardware page table objects during page table allocation or destruction. User space interacts with the fault object through a file interface. This interface offers a straightforward and efficient way for user space to handle page faults. It allows user space to read fault messages sequentially and respond to them by writing to the same file. The file interface supports reading messages in poll mode, so it's recommended that user space applications use io_uring to enhance read and write efficiency. A fault object can be associated with any iopf-capable iommufd_hw_pgtable during the pgtable's allocation. All I/O page faults triggered by devices when accessing the I/O addresses of an iommufd_hw_pgtable are routed through the fault object to user space. Similarly, user space's responses to these page faults are routed back to the iommu device driver through the same fault object. Link: https://lore.kernel.org/r/20240702063444.105814-7-baolu.lu@linux.intel.com Signed-off-by: Lu Baolu Reviewed-by: Jason Gunthorpe Reviewed-by: Kevin Tian Signed-off-by: Jason Gunthorpe --- drivers/iommu/io-pgfault.c | 2 + drivers/iommu/iommufd/Makefile | 1 + drivers/iommu/iommufd/fault.c | 226 ++++++++++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 30 +++++ drivers/iommu/iommufd/main.c | 6 + 5 files changed, 265 insertions(+) create mode 100644 drivers/iommu/iommufd/fault.c (limited to 'drivers/iommu') diff --git a/drivers/iommu/io-pgfault.c b/drivers/iommu/io-pgfault.c index 7c9011992d3f..cd679c13752e 100644 --- a/drivers/iommu/io-pgfault.c +++ b/drivers/iommu/io-pgfault.c @@ -110,6 +110,8 @@ static struct iopf_group *iopf_group_alloc(struct iommu_fault_param *iopf_param, list_add(&group->pending_node, &iopf_param->faults); mutex_unlock(&iopf_param->lock); + group->fault_count = list_count_nodes(&group->faults); + return group; } diff --git a/drivers/iommu/iommufd/Makefile b/drivers/iommu/iommufd/Makefile index 34b446146961..cf4605962bea 100644 --- a/drivers/iommu/iommufd/Makefile +++ b/drivers/iommu/iommufd/Makefile @@ -1,6 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only iommufd-y := \ device.o \ + fault.o \ hw_pagetable.o \ io_pagetable.o \ ioas.o \ diff --git a/drivers/iommu/iommufd/fault.c b/drivers/iommu/iommufd/fault.c new file mode 100644 index 000000000000..68ff94671d48 --- /dev/null +++ b/drivers/iommu/iommufd/fault.c @@ -0,0 +1,226 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (C) 2024 Intel Corporation + */ +#define pr_fmt(fmt) "iommufd: " fmt + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../iommu-priv.h" +#include "iommufd_private.h" + +void iommufd_fault_destroy(struct iommufd_object *obj) +{ + struct iommufd_fault *fault = container_of(obj, struct iommufd_fault, obj); + struct iopf_group *group, *next; + + /* + * The iommufd object's reference count is zero at this point. + * We can be confident that no other threads are currently + * accessing this pointer. Therefore, acquiring the mutex here + * is unnecessary. + */ + list_for_each_entry_safe(group, next, &fault->deliver, node) { + list_del(&group->node); + iopf_group_response(group, IOMMU_PAGE_RESP_INVALID); + iopf_free_group(group); + } +} + +static void iommufd_compose_fault_message(struct iommu_fault *fault, + struct iommu_hwpt_pgfault *hwpt_fault, + struct iommufd_device *idev, + u32 cookie) +{ + hwpt_fault->flags = fault->prm.flags; + hwpt_fault->dev_id = idev->obj.id; + hwpt_fault->pasid = fault->prm.pasid; + hwpt_fault->grpid = fault->prm.grpid; + hwpt_fault->perm = fault->prm.perm; + hwpt_fault->addr = fault->prm.addr; + hwpt_fault->length = 0; + hwpt_fault->cookie = cookie; +} + +static ssize_t iommufd_fault_fops_read(struct file *filep, char __user *buf, + size_t count, loff_t *ppos) +{ + size_t fault_size = sizeof(struct iommu_hwpt_pgfault); + struct iommufd_fault *fault = filep->private_data; + struct iommu_hwpt_pgfault data; + struct iommufd_device *idev; + struct iopf_group *group; + struct iopf_fault *iopf; + size_t done = 0; + int rc = 0; + + if (*ppos || count % fault_size) + return -ESPIPE; + + mutex_lock(&fault->mutex); + while (!list_empty(&fault->deliver) && count > done) { + group = list_first_entry(&fault->deliver, + struct iopf_group, node); + + if (group->fault_count * fault_size > count - done) + break; + + rc = xa_alloc(&fault->response, &group->cookie, group, + xa_limit_32b, GFP_KERNEL); + if (rc) + break; + + idev = to_iommufd_handle(group->attach_handle)->idev; + list_for_each_entry(iopf, &group->faults, list) { + iommufd_compose_fault_message(&iopf->fault, + &data, idev, + group->cookie); + if (copy_to_user(buf + done, &data, fault_size)) { + xa_erase(&fault->response, group->cookie); + rc = -EFAULT; + break; + } + done += fault_size; + } + + list_del(&group->node); + } + mutex_unlock(&fault->mutex); + + return done == 0 ? rc : done; +} + +static ssize_t iommufd_fault_fops_write(struct file *filep, const char __user *buf, + size_t count, loff_t *ppos) +{ + size_t response_size = sizeof(struct iommu_hwpt_page_response); + struct iommufd_fault *fault = filep->private_data; + struct iommu_hwpt_page_response response; + struct iopf_group *group; + size_t done = 0; + int rc = 0; + + if (*ppos || count % response_size) + return -ESPIPE; + + mutex_lock(&fault->mutex); + while (count > done) { + rc = copy_from_user(&response, buf + done, response_size); + if (rc) + break; + + group = xa_erase(&fault->response, response.cookie); + if (!group) { + rc = -EINVAL; + break; + } + + iopf_group_response(group, response.code); + iopf_free_group(group); + done += response_size; + } + mutex_unlock(&fault->mutex); + + return done == 0 ? rc : done; +} + +static __poll_t iommufd_fault_fops_poll(struct file *filep, + struct poll_table_struct *wait) +{ + struct iommufd_fault *fault = filep->private_data; + __poll_t pollflags = EPOLLOUT; + + poll_wait(filep, &fault->wait_queue, wait); + mutex_lock(&fault->mutex); + if (!list_empty(&fault->deliver)) + pollflags |= EPOLLIN | EPOLLRDNORM; + mutex_unlock(&fault->mutex); + + return pollflags; +} + +static int iommufd_fault_fops_release(struct inode *inode, struct file *filep) +{ + struct iommufd_fault *fault = filep->private_data; + + refcount_dec(&fault->obj.users); + iommufd_ctx_put(fault->ictx); + return 0; +} + +static const struct file_operations iommufd_fault_fops = { + .owner = THIS_MODULE, + .open = nonseekable_open, + .read = iommufd_fault_fops_read, + .write = iommufd_fault_fops_write, + .poll = iommufd_fault_fops_poll, + .release = iommufd_fault_fops_release, + .llseek = no_llseek, +}; + +int iommufd_fault_alloc(struct iommufd_ucmd *ucmd) +{ + struct iommu_fault_alloc *cmd = ucmd->cmd; + struct iommufd_fault *fault; + struct file *filep; + int fdno; + int rc; + + if (cmd->flags) + return -EOPNOTSUPP; + + fault = iommufd_object_alloc(ucmd->ictx, fault, IOMMUFD_OBJ_FAULT); + if (IS_ERR(fault)) + return PTR_ERR(fault); + + fault->ictx = ucmd->ictx; + INIT_LIST_HEAD(&fault->deliver); + xa_init_flags(&fault->response, XA_FLAGS_ALLOC1); + mutex_init(&fault->mutex); + init_waitqueue_head(&fault->wait_queue); + + filep = anon_inode_getfile("[iommufd-pgfault]", &iommufd_fault_fops, + fault, O_RDWR); + if (IS_ERR(filep)) { + rc = PTR_ERR(filep); + goto out_abort; + } + + refcount_inc(&fault->obj.users); + iommufd_ctx_get(fault->ictx); + fault->filep = filep; + + fdno = get_unused_fd_flags(O_CLOEXEC); + if (fdno < 0) { + rc = fdno; + goto out_fput; + } + + cmd->out_fault_id = fault->obj.id; + cmd->out_fault_fd = fdno; + + rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd)); + if (rc) + goto out_put_fdno; + iommufd_object_finalize(ucmd->ictx, &fault->obj); + + fd_install(fdno, fault->filep); + + return 0; +out_put_fdno: + put_unused_fd(fdno); +out_fput: + fput(filep); + refcount_dec(&fault->obj.users); + iommufd_ctx_put(fault->ictx); +out_abort: + iommufd_object_abort_and_destroy(ucmd->ictx, &fault->obj); + + return rc; +} diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h index 991f864d1f9b..c8a4519f1405 100644 --- a/drivers/iommu/iommufd/iommufd_private.h +++ b/drivers/iommu/iommufd/iommufd_private.h @@ -128,6 +128,7 @@ enum iommufd_object_type { IOMMUFD_OBJ_HWPT_NESTED, IOMMUFD_OBJ_IOAS, IOMMUFD_OBJ_ACCESS, + IOMMUFD_OBJ_FAULT, #ifdef CONFIG_IOMMUFD_TEST IOMMUFD_OBJ_SELFTEST, #endif @@ -426,6 +427,35 @@ void iopt_remove_access(struct io_pagetable *iopt, u32 iopt_access_list_id); void iommufd_access_destroy_object(struct iommufd_object *obj); +/* + * An iommufd_fault object represents an interface to deliver I/O page faults + * to the user space. These objects are created/destroyed by the user space and + * associated with hardware page table objects during page-table allocation. + */ +struct iommufd_fault { + struct iommufd_object obj; + struct iommufd_ctx *ictx; + struct file *filep; + + /* The lists of outstanding faults protected by below mutex. */ + struct mutex mutex; + struct list_head deliver; + struct xarray response; + + struct wait_queue_head wait_queue; +}; + +struct iommufd_attach_handle { + struct iommu_attach_handle handle; + struct iommufd_device *idev; +}; + +/* Convert an iommu attach handle to iommufd handle. */ +#define to_iommufd_handle(hdl) container_of(hdl, struct iommufd_attach_handle, handle) + +int iommufd_fault_alloc(struct iommufd_ucmd *ucmd); +void iommufd_fault_destroy(struct iommufd_object *obj); + #ifdef CONFIG_IOMMUFD_TEST int iommufd_test(struct iommufd_ucmd *ucmd); void iommufd_selftest_destroy(struct iommufd_object *obj); diff --git a/drivers/iommu/iommufd/main.c b/drivers/iommu/iommufd/main.c index 39b32932c61e..83bbd7c5d160 100644 --- a/drivers/iommu/iommufd/main.c +++ b/drivers/iommu/iommufd/main.c @@ -319,6 +319,7 @@ static int iommufd_option(struct iommufd_ucmd *ucmd) union ucmd_buffer { struct iommu_destroy destroy; + struct iommu_fault_alloc fault; struct iommu_hw_info info; struct iommu_hwpt_alloc hwpt; struct iommu_hwpt_get_dirty_bitmap get_dirty_bitmap; @@ -355,6 +356,8 @@ struct iommufd_ioctl_op { } static const struct iommufd_ioctl_op iommufd_ioctl_ops[] = { IOCTL_OP(IOMMU_DESTROY, iommufd_destroy, struct iommu_destroy, id), + IOCTL_OP(IOMMU_FAULT_QUEUE_ALLOC, iommufd_fault_alloc, struct iommu_fault_alloc, + out_fault_fd), IOCTL_OP(IOMMU_GET_HW_INFO, iommufd_get_hw_info, struct iommu_hw_info, __reserved), IOCTL_OP(IOMMU_HWPT_ALLOC, iommufd_hwpt_alloc, struct iommu_hwpt_alloc, @@ -513,6 +516,9 @@ static const struct iommufd_object_ops iommufd_object_ops[] = { .destroy = iommufd_hwpt_nested_destroy, .abort = iommufd_hwpt_nested_abort, }, + [IOMMUFD_OBJ_FAULT] = { + .destroy = iommufd_fault_destroy, + }, #ifdef CONFIG_IOMMUFD_TEST [IOMMUFD_OBJ_SELFTEST] = { .destroy = iommufd_selftest_destroy, -- cgit v1.2.3 From b7d8833677baad8c80ed1aac8c396d687e64a376 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:41 +0800 Subject: iommufd: Fault-capable hwpt attach/detach/replace Add iopf-capable hw page table attach/detach/replace helpers. The pointer to iommufd_device is stored in the domain attachment handle, so that it can be echo'ed back in the iopf_group. The iopf-capable hw page tables can only be attached to devices that support the IOMMU_DEV_FEAT_IOPF feature. On the first attachment of an iopf-capable hw_pagetable to the device, the IOPF feature is enabled on the device. Similarly, after the last iopf-capable hwpt is detached from the device, the IOPF feature is disabled on the device. The current implementation allows a replacement between iopf-capable and non-iopf-capable hw page tables. This matches the nested translation use case, where a parent domain is attached by default and can then be replaced with a nested user domain with iopf support. Link: https://lore.kernel.org/r/20240702063444.105814-8-baolu.lu@linux.intel.com Signed-off-by: Lu Baolu Reviewed-by: Kevin Tian Signed-off-by: Jason Gunthorpe --- drivers/iommu/iommufd/device.c | 7 +- drivers/iommu/iommufd/fault.c | 190 ++++++++++++++++++++++++++++++++ drivers/iommu/iommufd/iommufd_private.h | 41 +++++++ 3 files changed, 235 insertions(+), 3 deletions(-) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommufd/device.c b/drivers/iommu/iommufd/device.c index 873630c111c1..9a7ec5997c61 100644 --- a/drivers/iommu/iommufd/device.c +++ b/drivers/iommu/iommufd/device.c @@ -215,6 +215,7 @@ struct iommufd_device *iommufd_device_bind(struct iommufd_ctx *ictx, refcount_inc(&idev->obj.users); /* igroup refcount moves into iommufd_device */ idev->igroup = igroup; + mutex_init(&idev->iopf_lock); /* * If the caller fails after this success it must call @@ -376,7 +377,7 @@ int iommufd_hw_pagetable_attach(struct iommufd_hw_pagetable *hwpt, * attachment. */ if (list_empty(&idev->igroup->device_list)) { - rc = iommu_attach_group(hwpt->domain, idev->igroup->group); + rc = iommufd_hwpt_attach_device(hwpt, idev); if (rc) goto err_unresv; idev->igroup->hwpt = hwpt; @@ -402,7 +403,7 @@ iommufd_hw_pagetable_detach(struct iommufd_device *idev) mutex_lock(&idev->igroup->lock); list_del(&idev->group_item); if (list_empty(&idev->igroup->device_list)) { - iommu_detach_group(hwpt->domain, idev->igroup->group); + iommufd_hwpt_detach_device(hwpt, idev); idev->igroup->hwpt = NULL; } if (hwpt_is_paging(hwpt)) @@ -497,7 +498,7 @@ iommufd_device_do_replace(struct iommufd_device *idev, goto err_unlock; } - rc = iommu_group_replace_domain(igroup->group, hwpt->domain); + rc = iommufd_hwpt_replace_device(idev, hwpt, old_hwpt); if (rc) goto err_unresv; diff --git a/drivers/iommu/iommufd/fault.c b/drivers/iommu/iommufd/fault.c index 68ff94671d48..4934ae572638 100644 --- a/drivers/iommu/iommufd/fault.c +++ b/drivers/iommu/iommufd/fault.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -15,6 +16,195 @@ #include "../iommu-priv.h" #include "iommufd_private.h" +static int iommufd_fault_iopf_enable(struct iommufd_device *idev) +{ + struct device *dev = idev->dev; + int ret; + + /* + * Once we turn on PCI/PRI support for VF, the response failure code + * should not be forwarded to the hardware due to PRI being a shared + * resource between PF and VFs. There is no coordination for this + * shared capability. This waits for a vPRI reset to recover. + */ + if (dev_is_pci(dev) && to_pci_dev(dev)->is_virtfn) + return -EINVAL; + + mutex_lock(&idev->iopf_lock); + /* Device iopf has already been on. */ + if (++idev->iopf_enabled > 1) { + mutex_unlock(&idev->iopf_lock); + return 0; + } + + ret = iommu_dev_enable_feature(dev, IOMMU_DEV_FEAT_IOPF); + if (ret) + --idev->iopf_enabled; + mutex_unlock(&idev->iopf_lock); + + return ret; +} + +static void iommufd_fault_iopf_disable(struct iommufd_device *idev) +{ + mutex_lock(&idev->iopf_lock); + if (!WARN_ON(idev->iopf_enabled == 0)) { + if (--idev->iopf_enabled == 0) + iommu_dev_disable_feature(idev->dev, IOMMU_DEV_FEAT_IOPF); + } + mutex_unlock(&idev->iopf_lock); +} + +static int __fault_domain_attach_dev(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev) +{ + struct iommufd_attach_handle *handle; + int ret; + + handle = kzalloc(sizeof(*handle), GFP_KERNEL); + if (!handle) + return -ENOMEM; + + handle->idev = idev; + ret = iommu_attach_group_handle(hwpt->domain, idev->igroup->group, + &handle->handle); + if (ret) + kfree(handle); + + return ret; +} + +int iommufd_fault_domain_attach_dev(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev) +{ + int ret; + + if (!hwpt->fault) + return -EINVAL; + + ret = iommufd_fault_iopf_enable(idev); + if (ret) + return ret; + + ret = __fault_domain_attach_dev(hwpt, idev); + if (ret) + iommufd_fault_iopf_disable(idev); + + return ret; +} + +static void iommufd_auto_response_faults(struct iommufd_hw_pagetable *hwpt, + struct iommufd_attach_handle *handle) +{ + struct iommufd_fault *fault = hwpt->fault; + struct iopf_group *group, *next; + unsigned long index; + + if (!fault) + return; + + mutex_lock(&fault->mutex); + list_for_each_entry_safe(group, next, &fault->deliver, node) { + if (group->attach_handle != &handle->handle) + continue; + list_del(&group->node); + iopf_group_response(group, IOMMU_PAGE_RESP_INVALID); + iopf_free_group(group); + } + + xa_for_each(&fault->response, index, group) { + if (group->attach_handle != &handle->handle) + continue; + xa_erase(&fault->response, index); + iopf_group_response(group, IOMMU_PAGE_RESP_INVALID); + iopf_free_group(group); + } + mutex_unlock(&fault->mutex); +} + +static struct iommufd_attach_handle * +iommufd_device_get_attach_handle(struct iommufd_device *idev) +{ + struct iommu_attach_handle *handle; + + handle = iommu_attach_handle_get(idev->igroup->group, IOMMU_NO_PASID, 0); + if (!handle) + return NULL; + + return to_iommufd_handle(handle); +} + +void iommufd_fault_domain_detach_dev(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev) +{ + struct iommufd_attach_handle *handle; + + handle = iommufd_device_get_attach_handle(idev); + iommu_detach_group_handle(hwpt->domain, idev->igroup->group); + iommufd_auto_response_faults(hwpt, handle); + iommufd_fault_iopf_disable(idev); + kfree(handle); +} + +static int __fault_domain_replace_dev(struct iommufd_device *idev, + struct iommufd_hw_pagetable *hwpt, + struct iommufd_hw_pagetable *old) +{ + struct iommufd_attach_handle *handle, *curr = NULL; + int ret; + + if (old->fault) + curr = iommufd_device_get_attach_handle(idev); + + if (hwpt->fault) { + handle = kzalloc(sizeof(*handle), GFP_KERNEL); + if (!handle) + return -ENOMEM; + + handle->handle.domain = hwpt->domain; + handle->idev = idev; + ret = iommu_replace_group_handle(idev->igroup->group, + hwpt->domain, &handle->handle); + } else { + ret = iommu_replace_group_handle(idev->igroup->group, + hwpt->domain, NULL); + } + + if (!ret && curr) { + iommufd_auto_response_faults(old, curr); + kfree(curr); + } + + return ret; +} + +int iommufd_fault_domain_replace_dev(struct iommufd_device *idev, + struct iommufd_hw_pagetable *hwpt, + struct iommufd_hw_pagetable *old) +{ + bool iopf_off = !hwpt->fault && old->fault; + bool iopf_on = hwpt->fault && !old->fault; + int ret; + + if (iopf_on) { + ret = iommufd_fault_iopf_enable(idev); + if (ret) + return ret; + } + + ret = __fault_domain_replace_dev(idev, hwpt, old); + if (ret) { + if (iopf_on) + iommufd_fault_iopf_disable(idev); + return ret; + } + + if (iopf_off) + iommufd_fault_iopf_disable(idev); + + return 0; +} + void iommufd_fault_destroy(struct iommufd_object *obj) { struct iommufd_fault *fault = container_of(obj, struct iommufd_fault, obj); diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h index c8a4519f1405..aa4c26c87cb9 100644 --- a/drivers/iommu/iommufd/iommufd_private.h +++ b/drivers/iommu/iommufd/iommufd_private.h @@ -11,6 +11,7 @@ #include #include #include +#include "../iommu-priv.h" struct iommu_domain; struct iommu_group; @@ -293,6 +294,7 @@ int iommufd_check_iova_range(struct io_pagetable *iopt, struct iommufd_hw_pagetable { struct iommufd_object obj; struct iommu_domain *domain; + struct iommufd_fault *fault; }; struct iommufd_hwpt_paging { @@ -396,6 +398,9 @@ struct iommufd_device { /* always the physical device */ struct device *dev; bool enforce_cache_coherency; + /* protect iopf_enabled counter */ + struct mutex iopf_lock; + unsigned int iopf_enabled; }; static inline struct iommufd_device * @@ -456,6 +461,42 @@ struct iommufd_attach_handle { int iommufd_fault_alloc(struct iommufd_ucmd *ucmd); void iommufd_fault_destroy(struct iommufd_object *obj); +int iommufd_fault_domain_attach_dev(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev); +void iommufd_fault_domain_detach_dev(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev); +int iommufd_fault_domain_replace_dev(struct iommufd_device *idev, + struct iommufd_hw_pagetable *hwpt, + struct iommufd_hw_pagetable *old); + +static inline int iommufd_hwpt_attach_device(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev) +{ + if (hwpt->fault) + return iommufd_fault_domain_attach_dev(hwpt, idev); + + return iommu_attach_group(hwpt->domain, idev->igroup->group); +} + +static inline void iommufd_hwpt_detach_device(struct iommufd_hw_pagetable *hwpt, + struct iommufd_device *idev) +{ + if (hwpt->fault) + iommufd_fault_domain_detach_dev(hwpt, idev); + + iommu_detach_group(hwpt->domain, idev->igroup->group); +} + +static inline int iommufd_hwpt_replace_device(struct iommufd_device *idev, + struct iommufd_hw_pagetable *hwpt, + struct iommufd_hw_pagetable *old) +{ + if (old->fault || hwpt->fault) + return iommufd_fault_domain_replace_dev(idev, hwpt, old); + + return iommu_group_replace_domain(idev->igroup->group, hwpt->domain); +} + #ifdef CONFIG_IOMMUFD_TEST int iommufd_test(struct iommufd_ucmd *ucmd); void iommufd_selftest_destroy(struct iommufd_object *obj); -- cgit v1.2.3 From 34765cbc679c59ea5d952d738d2d16bf4aadc497 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:42 +0800 Subject: iommufd: Associate fault object with iommufd_hw_pgtable When allocating a user iommufd_hw_pagetable, the user space is allowed to associate a fault object with the hw_pagetable by specifying the fault object ID in the page table allocation data and setting the IOMMU_HWPT_FAULT_ID_VALID flag bit. On a successful return of hwpt allocation, the user can retrieve and respond to page faults by reading and writing the file interface of the fault object. Once a fault object has been associated with a hwpt, the hwpt is iopf-capable, indicated by hwpt->fault is non NULL. Attaching, detaching, or replacing an iopf-capable hwpt to an RID or PASID will differ from those that are not iopf-capable. Link: https://lore.kernel.org/r/20240702063444.105814-9-baolu.lu@linux.intel.com Signed-off-by: Lu Baolu Reviewed-by: Kevin Tian Signed-off-by: Jason Gunthorpe --- drivers/iommu/iommufd/fault.c | 17 +++++++++++++++ drivers/iommu/iommufd/hw_pagetable.c | 38 ++++++++++++++++++++++++++------- drivers/iommu/iommufd/iommufd_private.h | 9 ++++++++ 3 files changed, 56 insertions(+), 8 deletions(-) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommufd/fault.c b/drivers/iommu/iommufd/fault.c index 4934ae572638..54d6cd20a673 100644 --- a/drivers/iommu/iommufd/fault.c +++ b/drivers/iommu/iommufd/fault.c @@ -414,3 +414,20 @@ out_abort: return rc; } + +int iommufd_fault_iopf_handler(struct iopf_group *group) +{ + struct iommufd_hw_pagetable *hwpt; + struct iommufd_fault *fault; + + hwpt = group->attach_handle->domain->fault_data; + fault = hwpt->fault; + + mutex_lock(&fault->mutex); + list_add_tail(&group->node, &fault->deliver); + mutex_unlock(&fault->mutex); + + wake_up_interruptible(&fault->wait_queue); + + return 0; +} diff --git a/drivers/iommu/iommufd/hw_pagetable.c b/drivers/iommu/iommufd/hw_pagetable.c index 33d142f8057d..e63f80f087d1 100644 --- a/drivers/iommu/iommufd/hw_pagetable.c +++ b/drivers/iommu/iommufd/hw_pagetable.c @@ -8,6 +8,15 @@ #include "../iommu-priv.h" #include "iommufd_private.h" +static void __iommufd_hwpt_destroy(struct iommufd_hw_pagetable *hwpt) +{ + if (hwpt->domain) + iommu_domain_free(hwpt->domain); + + if (hwpt->fault) + refcount_dec(&hwpt->fault->obj.users); +} + void iommufd_hwpt_paging_destroy(struct iommufd_object *obj) { struct iommufd_hwpt_paging *hwpt_paging = @@ -22,9 +31,7 @@ void iommufd_hwpt_paging_destroy(struct iommufd_object *obj) hwpt_paging->common.domain); } - if (hwpt_paging->common.domain) - iommu_domain_free(hwpt_paging->common.domain); - + __iommufd_hwpt_destroy(&hwpt_paging->common); refcount_dec(&hwpt_paging->ioas->obj.users); } @@ -49,9 +56,7 @@ void iommufd_hwpt_nested_destroy(struct iommufd_object *obj) struct iommufd_hwpt_nested *hwpt_nested = container_of(obj, struct iommufd_hwpt_nested, common.obj); - if (hwpt_nested->common.domain) - iommu_domain_free(hwpt_nested->common.domain); - + __iommufd_hwpt_destroy(&hwpt_nested->common); refcount_dec(&hwpt_nested->parent->common.obj.users); } @@ -213,7 +218,8 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, struct iommufd_hw_pagetable *hwpt; int rc; - if (flags || !user_data->len || !ops->domain_alloc_user) + if ((flags & ~IOMMU_HWPT_FAULT_ID_VALID) || + !user_data->len || !ops->domain_alloc_user) return ERR_PTR(-EOPNOTSUPP); if (parent->auto_domain || !parent->nest_parent) return ERR_PTR(-EINVAL); @@ -227,7 +233,8 @@ iommufd_hwpt_nested_alloc(struct iommufd_ctx *ictx, refcount_inc(&parent->common.obj.users); hwpt_nested->parent = parent; - hwpt->domain = ops->domain_alloc_user(idev->dev, flags, + hwpt->domain = ops->domain_alloc_user(idev->dev, + flags & ~IOMMU_HWPT_FAULT_ID_VALID, parent->common.domain, user_data); if (IS_ERR(hwpt->domain)) { rc = PTR_ERR(hwpt->domain); @@ -308,6 +315,21 @@ int iommufd_hwpt_alloc(struct iommufd_ucmd *ucmd) goto out_put_pt; } + if (cmd->flags & IOMMU_HWPT_FAULT_ID_VALID) { + struct iommufd_fault *fault; + + fault = iommufd_get_fault(ucmd, cmd->fault_id); + if (IS_ERR(fault)) { + rc = PTR_ERR(fault); + goto out_hwpt; + } + hwpt->fault = fault; + hwpt->domain->iopf_handler = iommufd_fault_iopf_handler; + hwpt->domain->fault_data = hwpt; + refcount_inc(&fault->obj.users); + iommufd_put_object(ucmd->ictx, &fault->obj); + } + cmd->out_hwpt_id = hwpt->obj.id; rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd)); if (rc) diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h index aa4c26c87cb9..92efe30a8f0d 100644 --- a/drivers/iommu/iommufd/iommufd_private.h +++ b/drivers/iommu/iommufd/iommufd_private.h @@ -458,8 +458,17 @@ struct iommufd_attach_handle { /* Convert an iommu attach handle to iommufd handle. */ #define to_iommufd_handle(hdl) container_of(hdl, struct iommufd_attach_handle, handle) +static inline struct iommufd_fault * +iommufd_get_fault(struct iommufd_ucmd *ucmd, u32 id) +{ + return container_of(iommufd_get_object(ucmd->ictx, id, + IOMMUFD_OBJ_FAULT), + struct iommufd_fault, obj); +} + int iommufd_fault_alloc(struct iommufd_ucmd *ucmd); void iommufd_fault_destroy(struct iommufd_object *obj); +int iommufd_fault_iopf_handler(struct iopf_group *group); int iommufd_fault_domain_attach_dev(struct iommufd_hw_pagetable *hwpt, struct iommufd_device *idev); -- cgit v1.2.3 From ddee19971081b42615d62f4fdada21274708ed4d Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Tue, 2 Jul 2024 14:34:43 +0800 Subject: iommufd/selftest: Add IOPF support for mock device Extend the selftest mock device to support generating and responding to an IOPF. Also add an ioctl interface to userspace applications to trigger the IOPF on the mock device. This would allow userspace applications to test the IOMMUFD's handling of IOPFs without having to rely on any real hardware. Link: https://lore.kernel.org/r/20240702063444.105814-10-baolu.lu@linux.intel.com Signed-off-by: Lu Baolu Reviewed-by: Kevin Tian Signed-off-by: Jason Gunthorpe --- drivers/iommu/iommufd/iommufd_test.h | 8 +++++ drivers/iommu/iommufd/selftest.c | 64 ++++++++++++++++++++++++++++++++++++ 2 files changed, 72 insertions(+) (limited to 'drivers/iommu') diff --git a/drivers/iommu/iommufd/iommufd_test.h b/drivers/iommu/iommufd/iommufd_test.h index e854d3f67205..acbbba1c6671 100644 --- a/drivers/iommu/iommufd/iommufd_test.h +++ b/drivers/iommu/iommufd/iommufd_test.h @@ -22,6 +22,7 @@ enum { IOMMU_TEST_OP_MOCK_DOMAIN_FLAGS, IOMMU_TEST_OP_DIRTY, IOMMU_TEST_OP_MD_CHECK_IOTLB, + IOMMU_TEST_OP_TRIGGER_IOPF, }; enum { @@ -127,6 +128,13 @@ struct iommu_test_cmd { __u32 id; __u32 iotlb; } check_iotlb; + struct { + __u32 dev_id; + __u32 pasid; + __u32 grpid; + __u32 perm; + __u64 addr; + } trigger_iopf; }; __u32 last; }; diff --git a/drivers/iommu/iommufd/selftest.c b/drivers/iommu/iommufd/selftest.c index 7a2199470f31..1133f1b2362f 100644 --- a/drivers/iommu/iommufd/selftest.c +++ b/drivers/iommu/iommufd/selftest.c @@ -504,6 +504,8 @@ static bool mock_domain_capable(struct device *dev, enum iommu_cap cap) return false; } +static struct iopf_queue *mock_iommu_iopf_queue; + static struct iommu_device mock_iommu_device = { }; @@ -514,6 +516,29 @@ static struct iommu_device *mock_probe_device(struct device *dev) return &mock_iommu_device; } +static void mock_domain_page_response(struct device *dev, struct iopf_fault *evt, + struct iommu_page_response *msg) +{ +} + +static int mock_dev_enable_feat(struct device *dev, enum iommu_dev_features feat) +{ + if (feat != IOMMU_DEV_FEAT_IOPF || !mock_iommu_iopf_queue) + return -ENODEV; + + return iopf_queue_add_device(mock_iommu_iopf_queue, dev); +} + +static int mock_dev_disable_feat(struct device *dev, enum iommu_dev_features feat) +{ + if (feat != IOMMU_DEV_FEAT_IOPF || !mock_iommu_iopf_queue) + return -ENODEV; + + iopf_queue_remove_device(mock_iommu_iopf_queue, dev); + + return 0; +} + static const struct iommu_ops mock_ops = { /* * IOMMU_DOMAIN_BLOCKED cannot be returned from def_domain_type() @@ -529,6 +554,10 @@ static const struct iommu_ops mock_ops = { .capable = mock_domain_capable, .device_group = generic_device_group, .probe_device = mock_probe_device, + .page_response = mock_domain_page_response, + .dev_enable_feat = mock_dev_enable_feat, + .dev_disable_feat = mock_dev_disable_feat, + .user_pasid_table = true, .default_domain_ops = &(struct iommu_domain_ops){ .free = mock_domain_free, @@ -1375,6 +1404,31 @@ out_put: return rc; } +static int iommufd_test_trigger_iopf(struct iommufd_ucmd *ucmd, + struct iommu_test_cmd *cmd) +{ + struct iopf_fault event = { }; + struct iommufd_device *idev; + + idev = iommufd_get_device(ucmd, cmd->trigger_iopf.dev_id); + if (IS_ERR(idev)) + return PTR_ERR(idev); + + event.fault.prm.flags = IOMMU_FAULT_PAGE_REQUEST_LAST_PAGE; + if (cmd->trigger_iopf.pasid != IOMMU_NO_PASID) + event.fault.prm.flags |= IOMMU_FAULT_PAGE_REQUEST_PASID_VALID; + event.fault.type = IOMMU_FAULT_PAGE_REQ; + event.fault.prm.addr = cmd->trigger_iopf.addr; + event.fault.prm.pasid = cmd->trigger_iopf.pasid; + event.fault.prm.grpid = cmd->trigger_iopf.grpid; + event.fault.prm.perm = cmd->trigger_iopf.perm; + + iommu_report_device_fault(idev->dev, &event); + iommufd_put_object(ucmd->ictx, &idev->obj); + + return 0; +} + void iommufd_selftest_destroy(struct iommufd_object *obj) { struct selftest_obj *sobj = container_of(obj, struct selftest_obj, obj); @@ -1450,6 +1504,8 @@ int iommufd_test(struct iommufd_ucmd *ucmd) cmd->dirty.page_size, u64_to_user_ptr(cmd->dirty.uptr), cmd->dirty.flags); + case IOMMU_TEST_OP_TRIGGER_IOPF: + return iommufd_test_trigger_iopf(ucmd, cmd); default: return -EOPNOTSUPP; } @@ -1491,6 +1547,9 @@ int __init iommufd_test_init(void) &iommufd_mock_bus_type.nb); if (rc) goto err_sysfs; + + mock_iommu_iopf_queue = iopf_queue_alloc("mock-iopfq"); + return 0; err_sysfs: @@ -1506,6 +1565,11 @@ err_dbgfs: void iommufd_test_exit(void) { + if (mock_iommu_iopf_queue) { + iopf_queue_free(mock_iommu_iopf_queue); + mock_iommu_iopf_queue = NULL; + } + iommu_device_sysfs_remove(&mock_iommu_device); iommu_device_unregister_bus(&mock_iommu_device, &iommufd_mock_bus_type.bus, -- cgit v1.2.3