accel/amdxdna: Add BO import and export

Add amdxdna_gem_prime_export() and amdxdna_gem_prime_import() for BO
import and export. Register mmu notifier for imported BO as well. When
MMU_NOTIFIER_UNMAP event is received, queue work to remove the notifier.

The same BO could be mapped multiple times if it is exported and imported
by an application. Use a link list to track VMAs the BO been mapped.

v2: Rebased and call get_dma_buf() before dma_buf_attach()
v3: Removed import_attach usage

Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250325200105.2744079-1-lizhi.hou@amd.com

drivers/accel/amdxdna/TODO

@@ -1,3 +1,2 @@
-- Add import and export BO support
 - Add debugfs support
 - Add debug BO support

drivers/accel/amdxdna/aie2_ctx.c

@@ -758,27 +758,42 @@ int aie2_hwctx_config(struct amdxdna_hwctx *hwctx, u32 type, u64 value, void *bu
 static int aie2_populate_range(struct amdxdna_gem_obj *abo)
 {
 	struct amdxdna_dev *xdna = to_xdna_dev(to_gobj(abo)->dev);
-	struct mm_struct *mm = abo->mem.notifier.mm;
-	struct hmm_range range = { 0 };
+	struct amdxdna_umap *mapp;
 	unsigned long timeout;
+	struct mm_struct *mm;
+	bool found;
 	int ret;
 
-	XDNA_INFO_ONCE(xdna, "populate memory range %llx size %lx",
-		       abo->mem.userptr, abo->mem.size);
-	range.notifier = &abo->mem.notifier;
-	range.start = abo->mem.userptr;
-	range.end = abo->mem.userptr + abo->mem.size;
-	range.hmm_pfns = abo->mem.pfns;
-	range.default_flags = HMM_PFN_REQ_FAULT;
-
-	if (!mmget_not_zero(mm))
-		return -EFAULT;
-
 	timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
 again:
-	range.notifier_seq = mmu_interval_read_begin(&abo->mem.notifier);
+	found = false;
+	down_write(&xdna->notifier_lock);
+	list_for_each_entry(mapp, &abo->mem.umap_list, node) {
+		if (mapp->invalid) {
+			found = true;
+			break;
+		}
+	}
+
+	if (!found) {
+		abo->mem.map_invalid = false;
+		up_write(&xdna->notifier_lock);
+		return 0;
+	}
+	kref_get(&mapp->refcnt);
+	up_write(&xdna->notifier_lock);
+
+	XDNA_DBG(xdna, "populate memory range %lx %lx",
+		 mapp->vma->vm_start, mapp->vma->vm_end);
+	mm = mapp->notifier.mm;
+	if (!mmget_not_zero(mm)) {
+		amdxdna_umap_put(mapp);
+		return -EFAULT;
+	}
+
+	mapp->range.notifier_seq = mmu_interval_read_begin(&mapp->notifier);
 	mmap_read_lock(mm);
-	ret = hmm_range_fault(&range);
+	ret = hmm_range_fault(&mapp->range);
 	mmap_read_unlock(mm);
 	if (ret) {
 		if (time_after(jiffies, timeout)) {
@@ -786,21 +801,27 @@ again:
 			goto put_mm;
 		}
 
-		if (ret == -EBUSY)
+		if (ret == -EBUSY) {
+			amdxdna_umap_put(mapp);
 			goto again;
+		}
 
 		goto put_mm;
 	}
 
-	down_read(&xdna->notifier_lock);
-	if (mmu_interval_read_retry(&abo->mem.notifier, range.notifier_seq)) {
-		up_read(&xdna->notifier_lock);
+	down_write(&xdna->notifier_lock);
+	if (mmu_interval_read_retry(&mapp->notifier, mapp->range.notifier_seq)) {
+		up_write(&xdna->notifier_lock);
+		amdxdna_umap_put(mapp);
 		goto again;
 	}
-	abo->mem.map_invalid = false;
-	up_read(&xdna->notifier_lock);
+	mapp->invalid = false;
+	up_write(&xdna->notifier_lock);
+	amdxdna_umap_put(mapp);
+	goto again;
 
 put_mm:
+	amdxdna_umap_put(mapp);
 	mmput(mm);
 	return ret;
 }
@@ -908,10 +929,6 @@ void aie2_hmm_invalidate(struct amdxdna_gem_obj *abo,
 	struct drm_gem_object *gobj = to_gobj(abo);
 	long ret;
 
-	down_write(&xdna->notifier_lock);
-	abo->mem.map_invalid = true;
-	mmu_interval_set_seq(&abo->mem.notifier, cur_seq);
-	up_write(&xdna->notifier_lock);
 	ret = dma_resv_wait_timeout(gobj->resv, DMA_RESV_USAGE_BOOKKEEP,
 				    true, MAX_SCHEDULE_TIMEOUT);
 	if (!ret || ret == -ERESTARTSYS)

drivers/accel/amdxdna/amdxdna_gem.h

@@ -6,6 +6,20 @@
 #ifndef _AMDXDNA_GEM_H_
 #define _AMDXDNA_GEM_H_
 
+#include <linux/hmm.h>
+
+struct amdxdna_umap {
+	struct vm_area_struct		*vma;
+	struct mmu_interval_notifier	notifier;
+	struct hmm_range		range;
+	struct work_struct		hmm_unreg_work;
+	struct amdxdna_gem_obj		*abo;
+	struct list_head		node;
+	struct kref			refcnt;
+	bool				invalid;
+	bool				unmapped;
+};
+
 struct amdxdna_mem {
 	u64				userptr;
 	void				*kva;
@@ -13,8 +27,7 @@ struct amdxdna_mem {
 	size_t				size;
 	struct page			**pages;
 	u32				nr_pages;
-	struct mmu_interval_notifier	notifier;
-	unsigned long			*pfns;
+	struct list_head		umap_list;
 	bool				map_invalid;
 };
 
@@ -31,9 +44,12 @@ struct amdxdna_gem_obj {
 	struct amdxdna_gem_obj		*dev_heap; /* For AMDXDNA_BO_DEV */
 	struct drm_mm_node		mm_node; /* For AMDXDNA_BO_DEV */
 	u32				assigned_hwctx;
+	struct dma_buf			*dma_buf;
+	struct dma_buf_attachment	*attach;
 };
 
 #define to_gobj(obj)    (&(obj)->base.base)
+#define is_import_bo(obj) ((obj)->attach)
 
 static inline struct amdxdna_gem_obj *to_xdna_obj(struct drm_gem_object *gobj)
 {
@@ -47,8 +63,12 @@ static inline void amdxdna_gem_put_obj(struct amdxdna_gem_obj *abo)
 	drm_gem_object_put(to_gobj(abo));
 }
 
+void amdxdna_umap_put(struct amdxdna_umap *mapp);
+
 struct drm_gem_object *
 amdxdna_gem_create_object_cb(struct drm_device *dev, size_t size);
+struct drm_gem_object *
+amdxdna_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf);
 struct amdxdna_gem_obj *
 amdxdna_drm_alloc_dev_bo(struct drm_device *dev,
 			 struct amdxdna_drm_create_bo *args,

drivers/accel/amdxdna/amdxdna_pci_drv.c

@@ -226,6 +226,7 @@ const struct drm_driver amdxdna_drm_drv = {
 	.num_ioctls = ARRAY_SIZE(amdxdna_drm_ioctls),
 
 	.gem_create_object = amdxdna_gem_create_object_cb,
+	.gem_prime_import = amdxdna_gem_prime_import,
 };
 
 static const struct amdxdna_dev_info *
@@ -266,12 +267,16 @@ static int amdxdna_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		fs_reclaim_release(GFP_KERNEL);
 	}
 
+	xdna->notifier_wq = alloc_ordered_workqueue("notifier_wq", 0);
+	if (!xdna->notifier_wq)
+		return -ENOMEM;
+
 	mutex_lock(&xdna->dev_lock);
 	ret = xdna->dev_info->ops->init(xdna);
 	mutex_unlock(&xdna->dev_lock);
 	if (ret) {
 		XDNA_ERR(xdna, "Hardware init failed, ret %d", ret);
-		return ret;
+		goto destroy_notifier_wq;
 	}
 
 	ret = amdxdna_sysfs_init(xdna);
@@ -301,6 +306,8 @@ failed_dev_fini:
 	mutex_lock(&xdna->dev_lock);
 	xdna->dev_info->ops->fini(xdna);
 	mutex_unlock(&xdna->dev_lock);
+destroy_notifier_wq:
+	destroy_workqueue(xdna->notifier_wq);
 	return ret;
 }
 
@@ -310,6 +317,8 @@ static void amdxdna_remove(struct pci_dev *pdev)
 	struct device *dev = &pdev->dev;
 	struct amdxdna_client *client;
 
+	destroy_workqueue(xdna->notifier_wq);
+
 	pm_runtime_get_noresume(dev);
 	pm_runtime_forbid(dev);
 

drivers/accel/amdxdna/amdxdna_pci_drv.h

@@ -6,6 +6,7 @@
 #ifndef _AMDXDNA_PCI_DRV_H_
 #define _AMDXDNA_PCI_DRV_H_
 
+#include <linux/workqueue.h>
 #include <linux/xarray.h>
 
 #define XDNA_INFO(xdna, fmt, args...)	drm_info(&(xdna)->ddev, fmt, ##args)
@@ -98,6 +99,7 @@ struct amdxdna_dev {
 	struct list_head		client_list;
 	struct amdxdna_fw_ver		fw_ver;
 	struct rw_semaphore		notifier_lock; /* for mmu notifier*/
+	struct workqueue_struct		*notifier_wq;
 };
 
 /*