[igt-dev] [PATCH i-g-t 2/2] tests/kms_frontbuffer_tracking: Fix skips when PSR is not available.

Maarten Lankhorst maarten.lankhorst at linux.intel.com
Fri Feb 1 09:52:52 UTC 2019


Op 31-01-2019 om 21:58 schreef Souza, Jose:
> On Thu, 2019-01-31 at 15:18 +0100, Maarten Lankhorst wrote:
>> Use psr.can_test to determine we can disable PSR, and not skip
>> because
>> PSR is unavailable.
>>
>> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan at intel.com>
>> Cc: José Roberto de Souza <jose.souza at intel.com>
>> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
>> ---
>>  lib/igt_psr.c                    | 2 +-
>>  tests/kms_frontbuffer_tracking.c | 2 +-
>>  2 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/lib/igt_psr.c b/lib/igt_psr.c
>> index 5eb6d2c6ace5..6ad2c522f9a4 100644
>> --- a/lib/igt_psr.c
>> +++ b/lib/igt_psr.c
>> @@ -106,7 +106,7 @@ static bool psr_set(int debugfs_fd, int mode)
>>  
>>  	ret = has_psr_debugfs(debugfs_fd);
>>  	if (ret == -ENODEV) {
>> -		igt_skip_on_f(mode >= PSR_MODE_1, "PSR not
>> available\n");
>> +		igt_skip("PSR not available\n");
>>  		return false;
>>  	}
>>  
>> diff --git a/tests/kms_frontbuffer_tracking.c
>> b/tests/kms_frontbuffer_tracking.c
>> index 609f7b41a527..4d15ce1cf1d6 100644
>> --- a/tests/kms_frontbuffer_tracking.c
>> +++ b/tests/kms_frontbuffer_tracking.c
>> @@ -1134,7 +1134,7 @@ static bool disable_features(const struct
>> test_mode *t)
>>  
>>  	fbc_disable();
>>  	drrs_disable();
>> -	return psr_disable(drm.debugfs);
> Although the return true of this functions will cause a unecessary
> igt_display_commit() this changes looks good. In other patch we could
> just drop the return and remove the commit in prepare_subtest_data().
>
> Reviewed-by: José Roberto de Souza <jose.souza at intel.com>


Thanks, pushed.

This should have been documented, but the function must return true if and only if a modeset is required.

Specifically, It should return false for the debugfs case so we don't do unnecessary modesets,

but true for the modparam case.

I think the rework borked it.

~Maarten



More information about the igt-dev mailing list