drm/amdkfd: use tlb_seq from the VM subsystem for SVM as well v2

Instead of hand rolling the table_freed parameter.

v2: add some changes suggested by Philip

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Philip Yang<Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

3 files changed, 12 insertions, 16 deletions

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 607f65ab39ac..60438193c0c1 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1146,7 +1146,6 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
 	long err = 0;
 	int i;
 	uint32_t *devices_arr = NULL;
-	bool table_freed = false;
 
 	if (!args->n_devices) {
 		pr_debug("Device IDs array empty\n");
@@ -1208,7 +1207,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
 
 		err = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(
 			peer_pdd->dev->adev, (struct kgd_mem *)mem,
-			peer_pdd->drm_priv, &table_freed);
+			peer_pdd->drm_priv);
 		if (err) {
 			struct pci_dev *pdev = peer_pdd->dev->adev->pdev;
 
@@ -1233,13 +1232,11 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
 	}
 
 	/* Flush TLBs after waiting for the page table updates to complete */
-	if (table_freed || !kfd_flush_tlb_after_unmap(dev)) {
-		for (i = 0; i < args->n_devices; i++) {
-			peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]);
-			if (WARN_ON_ONCE(!peer_pdd))
-				continue;
-			kfd_flush_tlb(peer_pdd, TLB_FLUSH_LEGACY);
-		}
+	for (i = 0; i < args->n_devices; i++) {
+		peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]);
+		if (WARN_ON_ONCE(!peer_pdd))
+			continue;
+		kfd_flush_tlb(peer_pdd, TLB_FLUSH_LEGACY);
 	}
 	kfree(devices_arr);
 
@@ -2206,8 +2203,8 @@ static int criu_restore_bo(struct kfd_process *p,
 		if (IS_ERR(peer_pdd))
 			return PTR_ERR(peer_pdd);
 
-		ret = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(peer->adev, kgd_mem, peer_pdd->drm_priv,
-							    NULL);
+		ret = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(peer->adev, kgd_mem,
+							    peer_pdd->drm_priv);
 		if (ret) {
 			pr_err("Failed to map to gpu %d/%d\n", j, p->n_pdds);
 			return ret;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 4a8a047b7593..ac8123c1ee8f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -722,7 +722,7 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd,
 		goto err_alloc_mem;
 
 	err = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(kdev->adev, *mem,
-			pdd->drm_priv, NULL);
+			pdd->drm_priv);
 	if (err)
 		goto err_map_mem;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 3b8856b4cece..bf6354f539d8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1243,7 +1243,6 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 {
 	struct amdgpu_device *adev = pdd->dev->adev;
 	struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv);
-	bool table_freed = false;
 	uint64_t pte_flags;
 	unsigned long last_start;
 	int last_domain;
@@ -1284,7 +1283,7 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 						last_start - prange->start,
 						NULL, dma_addr,
 						&vm->last_update,
-						&table_freed);
+						NULL);
 
 		for (j = last_start - prange->start; j <= i; j++)
 			dma_addr[j] |= last_domain;
@@ -1306,8 +1305,6 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 	if (fence)
 		*fence = dma_fence_get(vm->last_update);
 
-	if (table_freed)
-		kfd_flush_tlb(pdd, TLB_FLUSH_LEGACY);
 out:
 	return r;
 }
@@ -1363,6 +1360,8 @@ svm_range_map_to_gpus(struct svm_range *prange, unsigned long offset,
 				break;
 			}
 		}
+
+		kfd_flush_tlb(pdd, TLB_FLUSH_LEGACY);
 	}
 
 	return r;