1 files changed, 133 insertions, 35 deletions

diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c
index eea23c630f7c..d343af4fa60e 100644
--- a/drivers/vdpa/vdpa_sim/vdpa_sim.c
+++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c
@@ -11,8 +11,8 @@
 #include <linux/module.h>
 #include <linux/device.h>
 #include <linux/kernel.h>
+#include <linux/kthread.h>
 #include <linux/slab.h>
-#include <linux/sched.h>
 #include <linux/dma-map-ops.h>
 #include <linux/vringh.h>
 #include <linux/vdpa.h>
@@ -35,10 +35,44 @@ module_param(max_iotlb_entries, int, 0444);
 MODULE_PARM_DESC(max_iotlb_entries,
 		 "Maximum number of iotlb entries for each address space. 0 means unlimited. (default: 2048)");
 
+static bool use_va = true;
+module_param(use_va, bool, 0444);
+MODULE_PARM_DESC(use_va, "Enable/disable the device's ability to use VA");
+
 #define VDPASIM_QUEUE_ALIGN PAGE_SIZE
 #define VDPASIM_QUEUE_MAX 256
 #define VDPASIM_VENDOR_ID 0
 
+struct vdpasim_mm_work {
+	struct kthread_work work;
+	struct vdpasim *vdpasim;
+	struct mm_struct *mm_to_bind;
+	int ret;
+};
+
+static void vdpasim_mm_work_fn(struct kthread_work *work)
+{
+	struct vdpasim_mm_work *mm_work =
+		container_of(work, struct vdpasim_mm_work, work);
+	struct vdpasim *vdpasim = mm_work->vdpasim;
+
+	mm_work->ret = 0;
+
+	//TODO: should we attach the cgroup of the mm owner?
+	vdpasim->mm_bound = mm_work->mm_to_bind;
+}
+
+static void vdpasim_worker_change_mm_sync(struct vdpasim *vdpasim,
+					  struct vdpasim_mm_work *mm_work)
+{
+	struct kthread_work *work = &mm_work->work;
+
+	kthread_init_work(work, vdpasim_mm_work_fn);
+	kthread_queue_work(vdpasim->worker, work);
+
+	kthread_flush_work(work);
+}
+
 static struct vdpasim *vdpa_to_sim(struct vdpa_device *vdpa)
 {
 	return container_of(vdpa, struct vdpasim, vdpa);
@@ -59,13 +93,20 @@ static void vdpasim_queue_ready(struct vdpasim *vdpasim, unsigned int idx)
 {
 	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
 	uint16_t last_avail_idx = vq->vring.last_avail_idx;
-
-	vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num, true,
-			  (struct vring_desc *)(uintptr_t)vq->desc_addr,
-			  (struct vring_avail *)
-			  (uintptr_t)vq->driver_addr,
-			  (struct vring_used *)
-			  (uintptr_t)vq->device_addr);
+	struct vring_desc *desc = (struct vring_desc *)
+				  (uintptr_t)vq->desc_addr;
+	struct vring_avail *avail = (struct vring_avail *)
+				    (uintptr_t)vq->driver_addr;
+	struct vring_used *used = (struct vring_used *)
+				  (uintptr_t)vq->device_addr;
+
+	if (use_va && vdpasim->mm_bound) {
+		vringh_init_iotlb_va(&vq->vring, vdpasim->features, vq->num,
+				     true, desc, avail, used);
+	} else {
+		vringh_init_iotlb(&vq->vring, vdpasim->features, vq->num,
+				  true, desc, avail, used);
+	}
 
 	vq->vring.last_avail_idx = last_avail_idx;
 
@@ -127,6 +168,25 @@ static void vdpasim_do_reset(struct vdpasim *vdpasim)
 static const struct vdpa_config_ops vdpasim_config_ops;
 static const struct vdpa_config_ops vdpasim_batch_config_ops;
 
+static void vdpasim_work_fn(struct kthread_work *work)
+{
+	struct vdpasim *vdpasim = container_of(work, struct vdpasim, work);
+	struct mm_struct *mm = vdpasim->mm_bound;
+
+	if (use_va && mm) {
+		if (!mmget_not_zero(mm))
+			return;
+		kthread_use_mm(mm);
+	}
+
+	vdpasim->dev_attr.work_fn(vdpasim);
+
+	if (use_va && mm) {
+		kthread_unuse_mm(mm);
+		mmput(mm);
+	}
+}
+
 struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
 			       const struct vdpa_dev_set_config *config)
 {
@@ -155,7 +215,7 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
 	vdpa = __vdpa_alloc_device(NULL, ops,
 				   dev_attr->ngroups, dev_attr->nas,
 				   dev_attr->alloc_size,
-				   dev_attr->name, false);
+				   dev_attr->name, use_va);
 	if (IS_ERR(vdpa)) {
 		ret = PTR_ERR(vdpa);
 		goto err_alloc;
@@ -163,11 +223,17 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
 
 	vdpasim = vdpa_to_sim(vdpa);
 	vdpasim->dev_attr = *dev_attr;
-	INIT_WORK(&vdpasim->work, dev_attr->work_fn);
-	spin_lock_init(&vdpasim->lock);
+	dev = &vdpasim->vdpa.dev;
+
+	kthread_init_work(&vdpasim->work, vdpasim_work_fn);
+	vdpasim->worker = kthread_create_worker(0, "vDPA sim worker: %s",
+						dev_attr->name);
+	if (IS_ERR(vdpasim->worker))
+		goto err_iommu;
+
+	mutex_init(&vdpasim->mutex);
 	spin_lock_init(&vdpasim->iommu_lock);
 
-	dev = &vdpasim->vdpa.dev;
 	dev->dma_mask = &dev->coherent_dma_mask;
 	if (dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)))
 		goto err_iommu;
@@ -195,10 +261,6 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr,
 	for (i = 0; i < vdpasim->dev_attr.nas; i++)
 		vhost_iotlb_init(&vdpasim->iommu[i], max_iotlb_entries, 0);
 
-	vdpasim->buffer = kvmalloc(dev_attr->buffer_size, GFP_KERNEL);
-	if (!vdpasim->buffer)
-		goto err_iommu;
-
 	for (i = 0; i < dev_attr->nvqs; i++)
 		vringh_set_iotlb(&vdpasim->vqs[i].vring, &vdpasim->iommu[0],
 				 &vdpasim->iommu_lock);
@@ -214,6 +276,12 @@ err_alloc:
 }
 EXPORT_SYMBOL_GPL(vdpasim_create);
 
+void vdpasim_schedule_work(struct vdpasim *vdpasim)
+{
+	kthread_queue_work(vdpasim->worker, &vdpasim->work);
+}
+EXPORT_SYMBOL_GPL(vdpasim_schedule_work);
+
 static int vdpasim_set_vq_address(struct vdpa_device *vdpa, u16 idx,
 				  u64 desc_area, u64 driver_area,
 				  u64 device_area)
@@ -248,7 +316,7 @@ static void vdpasim_kick_vq(struct vdpa_device *vdpa, u16 idx)
 	}
 
 	if (vq->ready)
-		schedule_work(&vdpasim->work);
+		vdpasim_schedule_work(vdpasim);
 }
 
 static void vdpasim_set_vq_cb(struct vdpa_device *vdpa, u16 idx,
@@ -267,13 +335,13 @@ static void vdpasim_set_vq_ready(struct vdpa_device *vdpa, u16 idx, bool ready)
 	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
 	bool old_ready;
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	old_ready = vq->ready;
 	vq->ready = ready;
 	if (vq->ready && !old_ready) {
 		vdpasim_queue_ready(vdpasim, idx);
 	}
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 }
 
 static bool vdpasim_get_vq_ready(struct vdpa_device *vdpa, u16 idx)
@@ -291,9 +359,9 @@ static int vdpasim_set_vq_state(struct vdpa_device *vdpa, u16 idx,
 	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
 	struct vringh *vrh = &vq->vring;
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	vrh->last_avail_idx = state->split.avail_index;
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return 0;
 }
@@ -390,9 +458,9 @@ static u8 vdpasim_get_status(struct vdpa_device *vdpa)
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 	u8 status;
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	status = vdpasim->status;
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return status;
 }
@@ -401,19 +469,19 @@ static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status)
 {
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	vdpasim->status = status;
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 }
 
 static int vdpasim_reset(struct vdpa_device *vdpa)
 {
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	vdpasim->status = 0;
 	vdpasim_do_reset(vdpasim);
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return 0;
 }
@@ -422,9 +490,9 @@ static int vdpasim_suspend(struct vdpa_device *vdpa)
 {
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	vdpasim->running = false;
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return 0;
 }
@@ -434,7 +502,7 @@ static int vdpasim_resume(struct vdpa_device *vdpa)
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 	int i;
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 	vdpasim->running = true;
 
 	if (vdpasim->pending_kick) {
@@ -445,7 +513,7 @@ static int vdpasim_resume(struct vdpa_device *vdpa)
 		vdpasim->pending_kick = false;
 	}
 
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return 0;
 }
@@ -517,14 +585,14 @@ static int vdpasim_set_group_asid(struct vdpa_device *vdpa, unsigned int group,
 
 	iommu = &vdpasim->iommu[asid];
 
-	spin_lock(&vdpasim->lock);
+	mutex_lock(&vdpasim->mutex);
 
 	for (i = 0; i < vdpasim->dev_attr.nvqs; i++)
 		if (vdpasim_get_vq_group(vdpa, i) == group)
 			vringh_set_iotlb(&vdpasim->vqs[i].vring, iommu,
 					 &vdpasim->iommu_lock);
 
-	spin_unlock(&vdpasim->lock);
+	mutex_unlock(&vdpasim->mutex);
 
 	return 0;
 }
@@ -563,6 +631,30 @@ err:
 	return ret;
 }
 
+static int vdpasim_bind_mm(struct vdpa_device *vdpa, struct mm_struct *mm)
+{
+	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
+	struct vdpasim_mm_work mm_work;
+
+	mm_work.vdpasim = vdpasim;
+	mm_work.mm_to_bind = mm;
+
+	vdpasim_worker_change_mm_sync(vdpasim, &mm_work);
+
+	return mm_work.ret;
+}
+
+static void vdpasim_unbind_mm(struct vdpa_device *vdpa)
+{
+	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
+	struct vdpasim_mm_work mm_work;
+
+	mm_work.vdpasim = vdpasim;
+	mm_work.mm_to_bind = NULL;
+
+	vdpasim_worker_change_mm_sync(vdpasim, &mm_work);
+}
+
 static int vdpasim_dma_map(struct vdpa_device *vdpa, unsigned int asid,
 			   u64 iova, u64 size,
 			   u64 pa, u32 perm, void *opaque)
@@ -610,14 +702,16 @@ static void vdpasim_free(struct vdpa_device *vdpa)
 	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
 	int i;
 
-	cancel_work_sync(&vdpasim->work);
+	kthread_cancel_work_sync(&vdpasim->work);
+	kthread_destroy_worker(vdpasim->worker);
 
 	for (i = 0; i < vdpasim->dev_attr.nvqs; i++) {
 		vringh_kiov_cleanup(&vdpasim->vqs[i].out_iov);
 		vringh_kiov_cleanup(&vdpasim->vqs[i].in_iov);
 	}
 
-	kvfree(vdpasim->buffer);
+	vdpasim->dev_attr.free(vdpasim);
+
 	for (i = 0; i < vdpasim->dev_attr.nas; i++)
 		vhost_iotlb_reset(&vdpasim->iommu[i]);
 	kfree(vdpasim->iommu);
@@ -658,6 +752,8 @@ static const struct vdpa_config_ops vdpasim_config_ops = {
 	.set_group_asid         = vdpasim_set_group_asid,
 	.dma_map                = vdpasim_dma_map,
 	.dma_unmap              = vdpasim_dma_unmap,
+	.bind_mm		= vdpasim_bind_mm,
+	.unbind_mm		= vdpasim_unbind_mm,
 	.free                   = vdpasim_free,
 };
 
@@ -692,6 +788,8 @@ static const struct vdpa_config_ops vdpasim_batch_config_ops = {
 	.get_iova_range         = vdpasim_get_iova_range,
 	.set_group_asid         = vdpasim_set_group_asid,
 	.set_map                = vdpasim_set_map,
+	.bind_mm		= vdpasim_bind_mm,
+	.unbind_mm		= vdpasim_unbind_mm,
 	.free                   = vdpasim_free,
 };