[igt-dev] [PATCH i-g-t v2] tests/i915/gem_ctx_sseu: Skip gtt tests if mappable aperture unavailable

[Antonio Argenziano]

On 02/02/20 02:31, Chris Wilson wrote:
> Quoting Dixit, Ashutosh (2020-02-01 03:37:45)
>> On Fri, 31 Jan 2020 09:11:50 -0800, Antonio Argenziano wrote:
>>> The gtt tests try to verify that a gtt mapped area can be used as an
>>> sseu pointer so it wouldn't make sense to run those when gtt mapping is
>>> not available.
>>>
>>> v2:
>>>        - Move require in subtest definition. (Ashutosh)
>>>
>>
>> Reviewed-by: Ashutosh Dixit <ashutosh.dixit at intel.com>
>>
>>> Signed-off-by: Antonio Argenziano <antonio.argenziano at intel.com>
>>> Cc: Dixit Ashutosh <ashutosh.dixit at intel.com>
>>> ---
>>>   tests/i915/gem_ctx_sseu.c | 5 ++++-
>>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tests/i915/gem_ctx_sseu.c b/tests/i915/gem_ctx_sseu.c
>>> index 38dc584b..b5e9658e 100644
>>> --- a/tests/i915/gem_ctx_sseu.c
>>> +++ b/tests/i915/gem_ctx_sseu.c
>>> @@ -528,8 +528,11 @@ igt_main
>>>                igt_subtest("invalid-sseu")
>>>                        test_invalid_sseu(fd);
>>>
>>> -             igt_subtest("ggtt-args")
>>> +             igt_subtest("ggtt-args") {
>>> +                     gem_require_mappable_ggtt(fd);
>>> +
>>>                        test_ggtt_args(fd);
> 
> It pains me to say because of all the duplication, but we do need to
> cover the cases where args are inside strange mmaped objects.
> 
> Long ago we wanted a 'toolbox', as nearly all of these tests are the
> same except for the ioctl layout. Ideal candidates for annotated
> fuzzing (how many different types of mmap/faulthandlers can we think
> of?)
> 
> We need a better solution here.

How about looping across all mappings like we do for other tests?

Antonio

> -Chris
>