[PATCH] drm/amdgpu: Use drm_calloc_large for VM page_tables array

[Christian König]

Am 19.01.2016 um 09:59 schrieb Michel Dänzer:
> From: Michel Dänzer <michel.daenzer at amd.com>
>
> It can be big, depending on the VM address space size, which is tunable
> via the vm_size module parameter.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93721
> Signed-off-by: Michel Dänzer <michel.daenzer at amd.com>

I actually wanted to address this by reducing sizeof(struct 
amdgpu_vm_pt), but once more never got the time to do so.

Patch is Reviewed-by: Christian König <christian.koenig at amd.com>

> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 +++----
>   1 file changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> index aefc668..9599f75 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
> @@ -1282,7 +1282,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
>   {
>   	const unsigned align = min(AMDGPU_VM_PTB_ALIGN_SIZE,
>   		AMDGPU_VM_PTE_COUNT * 8);
> -	unsigned pd_size, pd_entries, pts_size;
> +	unsigned pd_size, pd_entries;
>   	int i, r;
>   
>   	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> @@ -1300,8 +1300,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
>   	pd_entries = amdgpu_vm_num_pdes(adev);
>   
>   	/* allocate page table array */
> -	pts_size = pd_entries * sizeof(struct amdgpu_vm_pt);
> -	vm->page_tables = kzalloc(pts_size, GFP_KERNEL);
> +	vm->page_tables = drm_calloc_large(pd_entries, sizeof(struct amdgpu_vm_pt));
>   	if (vm->page_tables == NULL) {
>   		DRM_ERROR("Cannot allocate memory for page table array\n");
>   		return -ENOMEM;
> @@ -1361,7 +1360,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
>   
>   	for (i = 0; i < amdgpu_vm_num_pdes(adev); i++)
>   		amdgpu_bo_unref(&vm->page_tables[i].entry.robj);
> -	kfree(vm->page_tables);
> +	drm_free_large(vm->page_tables);
>   
>   	amdgpu_bo_unref(&vm->page_directory);
>   	fence_put(vm->page_directory_fence);