[PATCH 03/10] drm/tests: Add test case for drm_internal_framebuffer_create()

[Maira Canal]

Hi Carlos,

On 9/4/23 13:57, Carlos wrote:
> Hi Maíra,
> 
> On 8/26/23 10:58, Maíra Canal wrote:
>> Hi Carlos,
>>
>> On 8/25/23 13:07, Carlos Eduardo Gallo Filho wrote:
>>> Introduce a test to cover the creation of framebuffer with
>>> modifier on a device that doesn't support it.
>>>
>>> Signed-off-by: Carlos Eduardo Gallo Filho <gcarlos at disroot.org>
>>> ---
>>>   drivers/gpu/drm/tests/drm_framebuffer_test.c | 28 ++++++++++++++++++++
>>>   1 file changed, 28 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/tests/drm_framebuffer_test.c 
>>> b/drivers/gpu/drm/tests/drm_framebuffer_test.c
>>> index aeaf2331f9cc..b20871e88995 100644
>>> --- a/drivers/gpu/drm/tests/drm_framebuffer_test.c
>>> +++ b/drivers/gpu/drm/tests/drm_framebuffer_test.c
>>> @@ -396,7 +396,35 @@ static void drm_framebuffer_test_to_desc(const 
>>> struct drm_framebuffer_test *t, c
>>>   KUNIT_ARRAY_PARAM(drm_framebuffer_create, 
>>> drm_framebuffer_create_cases,
>>>             drm_framebuffer_test_to_desc);
>>>   +/*
>>> + * This test is very similar to drm_test_framebuffer_create, except 
>>> that it
>>> + * set mock->mode_config.fb_modifiers_not_supported member to 1, 
>>> covering
>>> + * the case of trying to create a framebuffer with modifiers without 
>>> the
>>> + * device really supporting it.
>>> + */
>>> +static void drm_test_framebuffer_modifiers_not_supported(struct 
>>> kunit *test)
>>> +{
>>> +    struct drm_mock *mock = test->priv;
>>> +    struct drm_device *dev = &mock->dev;
>>> +    int buffer_created = 0;
>>> +
>>> +    /* A valid cmd with modifier */
>>> +    struct drm_mode_fb_cmd2 cmd = {
>>> +        .width = MAX_WIDTH, .height = MAX_HEIGHT,
>>> +        .pixel_format = DRM_FORMAT_ABGR8888, .handles = { 1, 0, 0 },
>>> +        .offsets = { UINT_MAX / 2, 0, 0 }, .pitches = { 4 * 
>>> MAX_WIDTH, 0, 0 },
>>> +        .flags = DRM_MODE_FB_MODIFIERS,
>>> +    };
>>> +
>>> +    mock->private = &buffer_created;
>>> +    dev->mode_config.fb_modifiers_not_supported = 1;
>>> +
>>> +    drm_internal_framebuffer_create(dev, &cmd, NULL);
>>> +    KUNIT_EXPECT_EQ(test, 0, buffer_created);
>>> +}
>>> +
>>>   static struct kunit_case drm_framebuffer_tests[] = {
>>> +    KUNIT_CASE(drm_test_framebuffer_modifiers_not_supported),
>>
>> Could we preserve alphabetical order?
>>
> I've see a lot of other tests files with this ordered by every KUNIT_CASE()
> coming before KUNIT_CASE_PARAM(), with each set ordered among themselves.
> Did younoticed that or are you suggesting ordering it even so? Or maybe
> you're referring about another unordered thing that I didn't noticed?

Actually, I was suggesting to keep the alphabetical order related to the
tests naming. So, drm_test_framebuffer_create would come ahead of
drm_test_framebuffer_modifiers_not_supported.


Best Regards,
- Maíra

> 
> Thanks,
> Carlos
> 
>> Best Regards,
>> - Maíra
>>
>>> KUNIT_CASE_PARAM(drm_test_framebuffer_create, 
>>> drm_framebuffer_create_gen_params),
>>>       { }
>>>   };