Patch "drm/amdkfd: Page aligned memory reserve size" has been added to the 6.1-stable tree

[Sasha Levin <sashal@xxxxxxxxxx>]

This is a note to let you know that I've just added the patch titled

    drm/amdkfd: Page aligned memory reserve size

to the 6.1-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     drm-amdkfd-page-aligned-memory-reserve-size.patch
and it can be found in the queue-6.1 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.



commit ebd1db9baebf09a6df327dda6aa9c6c22ea06ae6
Author: Philip Yang <Philip.Yang@xxxxxxx>
Date:   Mon Jan 9 18:08:17 2023 -0500

    drm/amdkfd: Page aligned memory reserve size
    
    [ Upstream commit 0c2dece8fb541ab07b68c3312a1065fa9c927a81 ]
    
    Use page aligned size to reserve memory usage because page aligned TTM
    BO size is used to unreserve memory usage, otherwise no page aligned
    size causes memory usage accounting unbalanced.
    
    Change vram_used definition type to int64_t to be able to trigger
    WARN_ONCE(adev && adev->kfd.vram_used < 0, "..."), to help debug the
    accounting issue with warning and backtrace.
    
    Signed-off-by: Philip Yang <Philip.Yang@xxxxxxx>
    Reviewed-by: Felix Kuehling <Felix.Kuehling@xxxxxxx>
    Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
    Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 30f145dc8724e..dbc842590b253 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -95,7 +95,7 @@ struct amdgpu_amdkfd_fence {
 
 struct amdgpu_kfd_dev {
 	struct kfd_dev *dev;
-	uint64_t vram_used;
+	int64_t vram_used;
 	uint64_t vram_used_aligned;
 	bool init_complete;
 	struct work_struct reset_work;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 404c839683b1c..da01c1424b4ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1653,6 +1653,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	struct amdgpu_bo *bo;
 	struct drm_gem_object *gobj = NULL;
 	u32 domain, alloc_domain;
+	uint64_t aligned_size;
 	u64 alloc_flags;
 	int ret;
 
@@ -1703,22 +1704,23 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	 * the memory.
 	 */
 	if ((*mem)->aql_queue)
-		size = size >> 1;
+		size >>= 1;
+	aligned_size = PAGE_ALIGN(size);
 
 	(*mem)->alloc_flags = flags;
 
 	amdgpu_sync_create(&(*mem)->sync);
 
-	ret = amdgpu_amdkfd_reserve_mem_limit(adev, size, flags);
+	ret = amdgpu_amdkfd_reserve_mem_limit(adev, aligned_size, flags);
 	if (ret) {
 		pr_debug("Insufficient memory\n");
 		goto err_reserve_limit;
 	}
 
 	pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s\n",
-			va, size, domain_string(alloc_domain));
+			va, (*mem)->aql_queue ? size << 1 : size, domain_string(alloc_domain));
 
-	ret = amdgpu_gem_object_create(adev, size, 1, alloc_domain, alloc_flags,
+	ret = amdgpu_gem_object_create(adev, aligned_size, 1, alloc_domain, alloc_flags,
 				       bo_type, NULL, &gobj);
 	if (ret) {
 		pr_debug("Failed to create BO on domain %s. ret %d\n",
@@ -1775,7 +1777,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
 	/* Don't unreserve system mem limit twice */
 	goto err_reserve_limit;
 err_bo_create:
-	amdgpu_amdkfd_unreserve_mem_limit(adev, size, flags);
+	amdgpu_amdkfd_unreserve_mem_limit(adev, aligned_size, flags);
 err_reserve_limit:
 	mutex_destroy(&(*mem)->lock);
 	if (gobj)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 6d291aa6386bd..f79b8e964140e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1127,8 +1127,13 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 	}
 
 	/* Update the VRAM usage count */
-	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)
-		WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + args->size);
+	if (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) {
+		uint64_t size = args->size;
+
+		if (flags & KFD_IOC_ALLOC_MEM_FLAGS_AQL_QUEUE_MEM)
+			size >>= 1;
+		WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + PAGE_ALIGN(size));
+	}
 
 	mutex_unlock(&p->mutex);