[Intel-gfx] [igt-dev] [PATCH i-g-t] igt/prime_vgem: Ask the shrinker to purge a vgem bo from inside i915

[Tvrtko Ursulin]

On 07/08/2018 16:44, Chris Wilson wrote:
> Link a vgem dmabuf into an i915 bo and then ask the i915 shrinker to
> purge/invalidate its pages. This should establish the lockdep link from
> the fs_reclaim shrinker section to whatever locks are used to
> acquire/release dmabuf mappings; if any are required ofc.
> 
> Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> Cc: Daniel Vetter <daniel at ffwll.ch>
> ---
>   tests/prime_vgem.c | 29 +++++++++++++++++++++++++++++
>   1 file changed, 29 insertions(+)
> 
> diff --git a/tests/prime_vgem.c b/tests/prime_vgem.c
> index d886044a4..3b3dcc910 100644
> --- a/tests/prime_vgem.c
> +++ b/tests/prime_vgem.c
> @@ -237,6 +237,32 @@ static void test_gtt(int vgem, int i915)
>   	gem_close(vgem, scratch.handle);
>   }
>   
> +static void test_shrink(int vgem, int i915)
> +{
> +	struct vgem_bo scratch = {
> +		.width = 1024,
> +		.height = 1024,
> +		.bpp = 32
> +	};
> +	int dmabuf;
> +
> +	vgem_create(vgem, &scratch);
> +
> +	dmabuf = prime_handle_to_fd(vgem, scratch.handle);
> +	gem_close(vgem, scratch.handle);
> +
> +	scratch.handle = prime_fd_to_handle(i915, dmabuf);
> +	close(dmabuf);
> +
> +	/* Populate the i915_bo->pages. */
> +	gem_set_domain(i915, scratch.handle, I915_GEM_DOMAIN_GTT, 0);
> +
> +	/* Now evict them, establising the link from i915:shrinker to vgem. */
> +	igt_drop_caches_set(i915, DROP_SHRINK_ALL);
> +
> +	gem_close(i915, scratch.handle);
> +}
> +
>   static bool is_coherent(int i915)
>   {
>   	int val = 1; /* by default, we assume GTT is coherent, hence the test */
> @@ -794,6 +820,9 @@ igt_main
>   	igt_subtest("basic-gtt")
>   		test_gtt(vgem, i915);
>   
> +	igt_subtest("shrink")
> +		test_shrink(vgem, i915);
> +
>   	igt_subtest("coherency-gtt")
>   		test_gtt_interleaved(vgem, i915);
>   
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>

Regards,

Tvrtko