[PATCH v2 08/11] drm/tests: Add test for drm_framebuffer_init()

Maxime Ripard mripard at kernel.org
Wed Oct 25 15:01:15 UTC 2023


On Tue, Oct 24, 2023 at 04:09:59PM -0300, Carlos Eduardo Gallo Filho wrote:
> Add a single KUnit test case for the drm_framebuffer_init function.
> 
> Signed-off-by: Carlos Eduardo Gallo Filho <gcarlos at disroot.org>
> ---
> v2:
>   - Reorder kunit cases alphabetically.
>   - Let fb1.dev unset instead of set it to wrong_drm to test mismatched
>     drm_device passed as drm_framebuffer_init() argument.
>   - Clean the framebuffer object.
> ---
>  drivers/gpu/drm/tests/drm_framebuffer_test.c | 52 ++++++++++++++++++++
>  1 file changed, 52 insertions(+)
> 
> diff --git a/drivers/gpu/drm/tests/drm_framebuffer_test.c b/drivers/gpu/drm/tests/drm_framebuffer_test.c
> index fb9589dd8aed..eedd5e920279 100644
> --- a/drivers/gpu/drm/tests/drm_framebuffer_test.c
> +++ b/drivers/gpu/drm/tests/drm_framebuffer_test.c
> @@ -551,10 +551,62 @@ static void drm_test_framebuffer_lookup(struct kunit *test)
>  	drm_framebuffer_cleanup(&fb1);
>  }
>  
> +static void drm_test_framebuffer_init(struct kunit *test)

Documentation

> +{
> +	struct drm_framebuffer_test_priv *priv = test->priv;
> +	struct drm_device *dev = &priv->dev;
> +	struct drm_format_info format = { };
> +	struct drm_framebuffer fb1 = { .format = &format };
> +	struct drm_framebuffer *fb2;
> +	struct drm_framebuffer_funcs funcs = { };
> +	int ret;
> +
> +	/* Fails if fb->dev doesn't point to the drm_device passed on first arg */
> +	ret = drm_framebuffer_init(dev, &fb1, &funcs);
> +	KUNIT_ASSERT_EQ(test, ret, -EINVAL);
> +	fb1.dev = dev;
> +
> +	/* Fails if fb.format isn't set */
> +	fb1.format = NULL;
> +	ret = drm_framebuffer_init(dev, &fb1, &funcs);
> +	KUNIT_ASSERT_EQ(test, ret, -EINVAL);
> +	fb1.format = &format;
> +
> +	ret = drm_framebuffer_init(dev, &fb1, &funcs);
> +	KUNIT_ASSERT_EQ(test, ret, 0);
> +
> +	/*
> +	 * Check if fb->funcs is actually set to the drm_framebuffer_funcs
> +	 * passed to it
> +	 */
> +	KUNIT_EXPECT_PTR_EQ(test, fb1.funcs, &funcs);
> +
> +	/* The fb->comm must be set to the current running process */
> +	KUNIT_EXPECT_STREQ(test, fb1.comm, current->comm);
> +
> +	/* The fb->base must be successfully initialized */
> +	KUNIT_EXPECT_NE(test, fb1.base.id, 0);
> +	KUNIT_EXPECT_EQ(test, fb1.base.type, DRM_MODE_OBJECT_FB);
> +	KUNIT_EXPECT_EQ(test, kref_read(&fb1.base.refcount), 1);
> +	KUNIT_EXPECT_PTR_EQ(test, fb1.base.free_cb, &drm_framebuffer_free);
> +
> +	/* Checks if the fb is really published and findable */
> +	fb2 = drm_framebuffer_lookup(dev, NULL, fb1.base.id);
> +	KUNIT_EXPECT_PTR_EQ(test, fb2, &fb1);
> +
> +	/* There must be just that one fb initialized */
> +	KUNIT_EXPECT_EQ(test, dev->mode_config.num_fb, 1);
> +	KUNIT_EXPECT_PTR_EQ(test, dev->mode_config.fb_list.prev, &fb1.head);
> +	KUNIT_EXPECT_PTR_EQ(test, dev->mode_config.fb_list.next, &fb1.head);
> +
> +	drm_framebuffer_cleanup(&fb1);
> +}

You're testing different failure cases, so these should all be their own
tests. Otherwise, you'll just get a single test failure that doesn't
really provide any feedback on what went wrong.

Maxime
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20231025/f5acf021/attachment.sig>


More information about the dri-devel mailing list