Allow a vIOMMU object to allocate vIRQ Event Queues, with a condition that
each vIOMMU can only have one single vIRQ event queue per type.

Add iommufd_eventq_virq_alloc with an iommufd_eventq_virq_ops for this new
ioctl.

Signed-off-by: Nicolin Chen <nicol...@nvidia.com>
---
 drivers/iommu/iommufd/iommufd_private.h |  59 ++++++++++
 include/linux/iommufd.h                 |   3 +
 include/uapi/linux/iommufd.h            |  31 ++++++
 drivers/iommu/iommufd/eventq.c          | 138 ++++++++++++++++++++++++
 drivers/iommu/iommufd/main.c            |   6 ++
 drivers/iommu/iommufd/viommu.c          |   2 +
 6 files changed, 239 insertions(+)

diff --git a/drivers/iommu/iommufd/iommufd_private.h 
b/drivers/iommu/iommufd/iommufd_private.h
index 1c9a101cc435..fd0b87707967 100644
--- a/drivers/iommu/iommufd/iommufd_private.h
+++ b/drivers/iommu/iommufd/iommufd_private.h
@@ -548,6 +548,51 @@ static inline int iommufd_hwpt_replace_device(struct 
iommufd_device *idev,
        return iommu_group_replace_domain(idev->igroup->group, hwpt->domain);
 }
 
+/*
+ * An iommufd_eventq_virq object represents a queue to deliver vIOMMU 
interrupts
+ * to the user space. These objects are created/destroyed by the user space and
+ * associated with vIOMMU object(s) during the allocations.
+ */
+struct iommufd_eventq_virq {
+       struct iommufd_eventq common;
+       struct iommufd_viommu *viommu;
+       struct workqueue_struct *irq_wq;
+       struct list_head node;
+
+       unsigned int type;
+};
+
+static inline struct iommufd_eventq_virq *
+to_eventq_virq(struct iommufd_eventq *eventq)
+{
+       return container_of(eventq, struct iommufd_eventq_virq, common);
+}
+
+static inline struct iommufd_eventq_virq *
+iommufd_get_eventq_virq(struct iommufd_ucmd *ucmd, u32 id)
+{
+       return container_of(iommufd_get_object(ucmd->ictx, id,
+                                              IOMMUFD_OBJ_EVENTQ_VIRQ),
+                           struct iommufd_eventq_virq, common.obj);
+}
+
+int iommufd_eventq_virq_alloc(struct iommufd_ucmd *ucmd);
+void iommufd_eventq_virq_destroy(struct iommufd_object *obj);
+void iommufd_eventq_virq_abort(struct iommufd_object *obj);
+
+/* An iommufd_virq represents a vIOMMU interrupt in an eventq_virq queue */
+struct iommufd_virq {
+       struct iommufd_eventq_virq *eventq_virq;
+       struct list_head node;
+       ssize_t irq_len;
+       void *irq_data;
+};
+
+static inline int iommufd_eventq_virq_handler(struct iommufd_virq *virq)
+{
+       return iommufd_eventq_notify(&virq->eventq_virq->common, &virq->node);
+}
+
 static inline struct iommufd_viommu *
 iommufd_get_viommu(struct iommufd_ucmd *ucmd, u32 id)
 {
@@ -556,6 +601,20 @@ iommufd_get_viommu(struct iommufd_ucmd *ucmd, u32 id)
                            struct iommufd_viommu, obj);
 }
 
+static inline struct iommufd_eventq_virq *
+iommufd_viommu_find_eventq_virq(struct iommufd_viommu *viommu, u32 type)
+{
+       struct iommufd_eventq_virq *eventq_virq, *next;
+
+       lockdep_assert_held(&viommu->virqs_rwsem);
+
+       list_for_each_entry_safe(eventq_virq, next, &viommu->virqs, node) {
+               if (eventq_virq->type == type)
+                       return eventq_virq;
+       }
+       return NULL;
+}
+
 int iommufd_viommu_alloc_ioctl(struct iommufd_ucmd *ucmd);
 void iommufd_viommu_destroy(struct iommufd_object *obj);
 int iommufd_vdevice_alloc_ioctl(struct iommufd_ucmd *ucmd);
diff --git a/include/linux/iommufd.h b/include/linux/iommufd.h
index 7ad105ab8090..40cc9bbb1d24 100644
--- a/include/linux/iommufd.h
+++ b/include/linux/iommufd.h
@@ -32,6 +32,7 @@ enum iommufd_object_type {
        IOMMUFD_OBJ_IOAS,
        IOMMUFD_OBJ_ACCESS,
        IOMMUFD_OBJ_EVENTQ_IOPF,
+       IOMMUFD_OBJ_EVENTQ_VIRQ,
        IOMMUFD_OBJ_VIOMMU,
        IOMMUFD_OBJ_VDEVICE,
 #ifdef CONFIG_IOMMUFD_TEST
@@ -93,6 +94,8 @@ struct iommufd_viommu {
        const struct iommufd_viommu_ops *ops;
 
        struct xarray vdevs;
+       struct list_head virqs;
+       struct rw_semaphore virqs_rwsem;
 
        unsigned int type;
 };
diff --git a/include/uapi/linux/iommufd.h b/include/uapi/linux/iommufd.h
index 34810f6ae2b5..d9319f5b7c69 100644
--- a/include/uapi/linux/iommufd.h
+++ b/include/uapi/linux/iommufd.h
@@ -55,6 +55,7 @@ enum {
        IOMMUFD_CMD_VIOMMU_ALLOC = 0x90,
        IOMMUFD_CMD_VDEVICE_ALLOC = 0x91,
        IOMMUFD_CMD_IOAS_CHANGE_PROCESS = 0x92,
+       IOMMUFD_CMD_VIRQ_ALLOC = 0x93,
 };
 
 /**
@@ -1012,4 +1013,34 @@ struct iommu_ioas_change_process {
 #define IOMMU_IOAS_CHANGE_PROCESS \
        _IO(IOMMUFD_TYPE, IOMMUFD_CMD_IOAS_CHANGE_PROCESS)
 
+/**
+ * enum iommu_virq_type - Virtual IRQ Type
+ * @IOMMU_VIRQ_TYPE_NONE: INVALID type
+ */
+enum iommu_virq_type {
+       IOMMU_VIRQ_TYPE_NONE = 0,
+};
+
+/**
+ * struct iommu_virq_alloc - ioctl(IOMMU_VIRQ_ALLOC)
+ * @size: sizeof(struct iommu_virq_alloc)
+ * @flags: Must be 0
+ * @viommu: virtual IOMMU ID to associate the virtual IRQ with
+ * @type: Type of the virtual IRQ. Must be defined in enum iommu_virq_type
+ * @out_virq_id: The ID of the new virtual IRQ
+ * @out_fault_fd: The fd of the new virtual IRQ. User space must close the
+ *                successfully returned fd after using it
+ *
+ * Explicitly allocate a virtual IRQ handler for a vIOMMU. A vIOMMU can have
+ * multiple FDs for different @type, but is confined to one FD per @type.
+ */
+struct iommu_virq_alloc {
+       __u32 size;
+       __u32 flags;
+       __u32 viommu_id;
+       __u32 type;
+       __u32 out_virq_id;
+       __u32 out_virq_fd;
+};
+#define IOMMU_VIRQ_ALLOC _IO(IOMMUFD_TYPE, IOMMUFD_CMD_VIRQ_ALLOC)
 #endif
diff --git a/drivers/iommu/iommufd/eventq.c b/drivers/iommu/iommufd/eventq.c
index 3674961a45c2..cf07b3c21f1e 100644
--- a/drivers/iommu/iommufd/eventq.c
+++ b/drivers/iommu/iommufd/eventq.c
@@ -346,6 +346,75 @@ static const struct iommufd_eventq_ops 
iommufd_eventq_iopf_ops = {
        .write = &iommufd_eventq_iopf_fops_write,
 };
 
+/* IOMMUFD_OBJ_EVENTQ_VIRQ Functions */
+
+void iommufd_eventq_virq_abort(struct iommufd_object *obj)
+{
+       struct iommufd_eventq *eventq =
+               container_of(obj, struct iommufd_eventq, obj);
+       struct iommufd_eventq_virq *eventq_virq = to_eventq_virq(eventq);
+       struct iommufd_viommu *viommu = eventq_virq->viommu;
+       struct iommufd_virq *virq, *next;
+
+       lockdep_assert_held_write(&viommu->virqs_rwsem);
+
+       list_for_each_entry_safe(virq, next, &eventq->deliver, node) {
+               list_del(&virq->node);
+               kfree(virq);
+       }
+
+       if (eventq_virq->irq_wq)
+               destroy_workqueue(eventq_virq->irq_wq);
+       refcount_dec(&viommu->obj.users);
+       mutex_destroy(&eventq->mutex);
+       list_del(&eventq_virq->node);
+}
+
+void iommufd_eventq_virq_destroy(struct iommufd_object *obj)
+{
+       struct iommufd_eventq_virq *eventq_virq =
+               to_eventq_virq(container_of(obj, struct iommufd_eventq, obj));
+
+       down_write(&eventq_virq->viommu->virqs_rwsem);
+       iommufd_eventq_virq_abort(obj);
+       up_write(&eventq_virq->viommu->virqs_rwsem);
+}
+
+static ssize_t iommufd_eventq_virq_fops_read(struct iommufd_eventq *eventq,
+                                            char __user *buf, size_t count,
+                                            loff_t *ppos)
+{
+       size_t done = 0;
+       int rc = 0;
+
+       if (*ppos)
+               return -ESPIPE;
+
+       mutex_lock(&eventq->mutex);
+       while (!list_empty(&eventq->deliver) && count > done) {
+               struct iommufd_virq *virq = list_first_entry(
+                       &eventq->deliver, struct iommufd_virq, node);
+
+               if (virq->irq_len > count - done)
+                       break;
+
+               if (copy_to_user(buf + done, virq->irq_data, virq->irq_len)) {
+                       rc = -EFAULT;
+                       break;
+               }
+               done += virq->irq_len;
+               list_del(&virq->node);
+               kfree(virq);
+       }
+       mutex_unlock(&eventq->mutex);
+
+       return done == 0 ? rc : done;
+}
+
+static const struct iommufd_eventq_ops iommufd_eventq_virq_ops = {
+       .read = &iommufd_eventq_virq_fops_read,
+};
+
 /* Common Event Queue Functions */
 
 static ssize_t iommufd_eventq_fops_read(struct file *filep, char __user *buf,
@@ -472,3 +541,72 @@ int iommufd_eventq_iopf_alloc(struct iommufd_ucmd *ucmd)
 
        return rc;
 }
+
+int iommufd_eventq_virq_alloc(struct iommufd_ucmd *ucmd)
+{
+       struct iommu_virq_alloc *cmd = ucmd->cmd;
+       struct iommufd_eventq_virq *eventq_virq;
+       struct iommufd_viommu *viommu;
+       int fdno;
+       int rc;
+
+       if (cmd->flags || cmd->type == IOMMU_VIRQ_TYPE_NONE)
+               return -EOPNOTSUPP;
+
+       viommu = iommufd_get_viommu(ucmd, cmd->viommu_id);
+       if (IS_ERR(viommu))
+               return PTR_ERR(viommu);
+       down_write(&viommu->virqs_rwsem);
+
+       if (iommufd_viommu_find_eventq_virq(viommu, cmd->type)) {
+               rc = -EEXIST;
+               goto out_unlock_virqs;
+       }
+
+       eventq_virq = __iommufd_object_alloc(
+               ucmd->ictx, eventq_virq, IOMMUFD_OBJ_EVENTQ_VIRQ, common.obj);
+       if (IS_ERR(eventq_virq)) {
+               rc = PTR_ERR(eventq_virq);
+               goto out_unlock_virqs;
+       }
+
+       eventq_virq->type = cmd->type;
+       eventq_virq->viommu = viommu;
+       refcount_inc(&viommu->obj.users);
+       list_add_tail(&eventq_virq->node, &viommu->virqs);
+
+       fdno = iommufd_eventq_init(&eventq_virq->common, "[iommufd-viommu-irq]",
+                                ucmd->ictx, &iommufd_eventq_virq_ops);
+       if (fdno < 0) {
+               rc = fdno;
+               goto out_abort;
+       }
+
+       eventq_virq->irq_wq = alloc_workqueue("viommu_irq/%d", WQ_UNBOUND, 0,
+                                             eventq_virq->common.obj.id);
+       if (!eventq_virq->irq_wq) {
+               rc = -ENOMEM;
+               goto out_put_fdno;
+       }
+
+       cmd->out_virq_id = eventq_virq->common.obj.id;
+       cmd->out_virq_fd = fdno;
+
+       rc = iommufd_ucmd_respond(ucmd, sizeof(*cmd));
+       if (rc)
+               goto out_put_fdno;
+
+       iommufd_object_finalize(ucmd->ictx, &eventq_virq->common.obj);
+       fd_install(fdno, eventq_virq->common.filep);
+       goto out_unlock_virqs;
+
+out_put_fdno:
+       put_unused_fd(fdno);
+       fput(eventq_virq->common.filep);
+out_abort:
+       iommufd_object_abort_and_destroy(ucmd->ictx, &eventq_virq->common.obj);
+out_unlock_virqs:
+       up_write(&viommu->virqs_rwsem);
+       iommufd_put_object(ucmd->ictx, &viommu->obj);
+       return rc;
+}
diff --git a/drivers/iommu/iommufd/main.c b/drivers/iommu/iommufd/main.c
index 539c24ada6d0..89e8ac56f4ce 100644
--- a/drivers/iommu/iommufd/main.c
+++ b/drivers/iommu/iommufd/main.c
@@ -335,6 +335,8 @@ 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_eventq_iopf_alloc,
                 struct iommu_fault_alloc, out_fault_fd),
+       IOCTL_OP(IOMMU_VIRQ_ALLOC, iommufd_eventq_virq_alloc,
+                struct iommu_virq_alloc, out_virq_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,
@@ -504,6 +506,10 @@ static const struct iommufd_object_ops 
iommufd_object_ops[] = {
        [IOMMUFD_OBJ_EVENTQ_IOPF] = {
                .destroy = iommufd_eventq_iopf_destroy,
        },
+       [IOMMUFD_OBJ_EVENTQ_VIRQ] = {
+               .destroy = iommufd_eventq_virq_destroy,
+               .abort = iommufd_eventq_virq_abort,
+       },
        [IOMMUFD_OBJ_VIOMMU] = {
                .destroy = iommufd_viommu_destroy,
        },
diff --git a/drivers/iommu/iommufd/viommu.c b/drivers/iommu/iommufd/viommu.c
index 69b88e8c7c26..075b6aed79bc 100644
--- a/drivers/iommu/iommufd/viommu.c
+++ b/drivers/iommu/iommufd/viommu.c
@@ -59,6 +59,8 @@ int iommufd_viommu_alloc_ioctl(struct iommufd_ucmd *ucmd)
        viommu->ictx = ucmd->ictx;
        viommu->hwpt = hwpt_paging;
        refcount_inc(&viommu->hwpt->common.obj.users);
+       INIT_LIST_HEAD(&viommu->virqs);
+       init_rwsem(&viommu->virqs_rwsem);
        /*
         * It is the most likely case that a physical IOMMU is unpluggable. A
         * pluggable IOMMU instance (if exists) is responsible for refcounting
-- 
2.43.0


Reply via email to