[i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Bhanuprakash Modem]

Block KMS tests those are specific to (display_ver > 14)
are not applicable for i915.

Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
---
 tests/intel-ci/blacklist.txt | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
index f6e9bb474..d01ef5e16 100644
--- a/tests/intel-ci/blacklist.txt
+++ b/tests/intel-ci/blacklist.txt
@@ -140,3 +140,6 @@ igt@device_reset@reset-bound
 
 # not working for i915
 igt@kms_bw@.*
+
+# Tests specific to (Display_ver > 14) are not applicable for i915
+igt@kms_vrr@cmrr
-- 
2.43.2

Re: [i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Jani Nikula]

On Wed, 19 Jun 2024, Bhanuprakash Modem <bhanuprakash.modem@intel.com> wrote:
> Block KMS tests those are specific to (display_ver > 14)
> are not applicable for i915.

Feels a bit wrong to list features like this in a block list. IMO the
tests themselves should be aware of what's supported and what's not. I
imagine the rationale is performance? What's the impact?

BR,
Jani.

>
> Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
> ---
>  tests/intel-ci/blacklist.txt | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
> index f6e9bb474..d01ef5e16 100644
> --- a/tests/intel-ci/blacklist.txt
> +++ b/tests/intel-ci/blacklist.txt
> @@ -140,3 +140,6 @@ igt@device_reset@reset-bound
>  
>  # not working for i915
>  igt@kms_bw@.*
> +
> +# Tests specific to (Display_ver > 14) are not applicable for i915
> +igt@kms_vrr@cmrr

-- 
Jani Nikula, Intel

Re: [i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Modem, Bhanuprakash]

On 19-06-2024 03:33 pm, Jani Nikula wrote:
> On Wed, 19 Jun 2024, Bhanuprakash Modem <bhanuprakash.modem@intel.com> wrote:
>> Block KMS tests those are specific to (display_ver > 14)
>> are not applicable for i915.
> 
> Feels a bit wrong to list features like this in a block list. IMO the
> tests themselves should be aware of what's supported and what's not. I
> imagine the rationale is performance? What's the impact?

I believe there is no point in running this test for i915, as it will 
always be skipped. The test already knows which platforms it supports 
and which it doesn't.

- Bhanu

> 
> BR,
> Jani.
> 
>>
>> Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
>> ---
>>   tests/intel-ci/blacklist.txt | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
>> index f6e9bb474..d01ef5e16 100644
>> --- a/tests/intel-ci/blacklist.txt
>> +++ b/tests/intel-ci/blacklist.txt
>> @@ -140,3 +140,6 @@ igt@device_reset@reset-bound
>>   
>>   # not working for i915
>>   igt@kms_bw@.*
>> +
>> +# Tests specific to (Display_ver > 14) are not applicable for i915
>> +igt@kms_vrr@cmrr
> 

Re: [i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Jani Nikula]

On Wed, 19 Jun 2024, "Modem, Bhanuprakash" <bhanuprakash.modem@intel.com> wrote:
> On 19-06-2024 03:33 pm, Jani Nikula wrote:
>> On Wed, 19 Jun 2024, Bhanuprakash Modem <bhanuprakash.modem@intel.com> wrote:
>>> Block KMS tests those are specific to (display_ver > 14)
>>> are not applicable for i915.
>> 
>> Feels a bit wrong to list features like this in a block list. IMO the
>> tests themselves should be aware of what's supported and what's not. I
>> imagine the rationale is performance? What's the impact?
>
> I believe there is no point in running this test for i915, as it will 
> always be skipped. The test already knows which platforms it supports 
> and which it doesn't.

The question is whether it's worth it to start maintaining the info in
two places, the individual tests based on dynamic info obtained from the
driver, and a manually updated static block list.

Are you also filtering xe testing for tests applicable to old platforms
only? Is the manual work worth the effort?

BR,
Jani.


>
> - Bhanu
>
>> 
>> BR,
>> Jani.
>> 
>>>
>>> Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
>>> ---
>>>   tests/intel-ci/blacklist.txt | 3 +++
>>>   1 file changed, 3 insertions(+)
>>>
>>> diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
>>> index f6e9bb474..d01ef5e16 100644
>>> --- a/tests/intel-ci/blacklist.txt
>>> +++ b/tests/intel-ci/blacklist.txt
>>> @@ -140,3 +140,6 @@ igt@device_reset@reset-bound
>>>   
>>>   # not working for i915
>>>   igt@kms_bw@.*
>>> +
>>> +# Tests specific to (Display_ver > 14) are not applicable for i915
>>> +igt@kms_vrr@cmrr
>> 

-- 
Jani Nikula, Intel

Re: [i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Modem, Bhanuprakash]

On 19-06-2024 04:11 pm, Jani Nikula wrote:
> On Wed, 19 Jun 2024, "Modem, Bhanuprakash" <bhanuprakash.modem@intel.com> wrote:
>> On 19-06-2024 03:33 pm, Jani Nikula wrote:
>>> On Wed, 19 Jun 2024, Bhanuprakash Modem <bhanuprakash.modem@intel.com> wrote:
>>>> Block KMS tests those are specific to (display_ver > 14)
>>>> are not applicable for i915.
>>>
>>> Feels a bit wrong to list features like this in a block list. IMO the
>>> tests themselves should be aware of what's supported and what's not. I
>>> imagine the rationale is performance? What's the impact?
>>
>> I believe there is no point in running this test for i915, as it will
>> always be skipped. The test already knows which platforms it supports
>> and which it doesn't.
> 
> The question is whether it's worth it to start maintaining the info in
> two places, the individual tests based on dynamic info obtained from the
> driver, and a manually updated static block list. >
> Are you also filtering xe testing for tests applicable to old platforms
> only? Is the manual work worth the effort?

The only advantage I see in maintaining an additional list is improved 
reporting.

Yes, we maintain an XE specific blocklist[*]. This manual effort is 
worthy for someone looking different execution reports. In fact, some 
efforts are going on to develop a platform-specific blocklist.

@kamil, Can you please add some details about upstreaming 
platform-specific blocklist?

[*]: 
https://cgit.freedesktop.org/drm/igt-gpu-tools/tree/tests/intel-ci/xe.blocklist.txt#n65

- Bhanu

> 
> BR,
> Jani.
> 
> 
>>
>> - Bhanu
>>
>>>
>>> BR,
>>> Jani.
>>>
>>>>
>>>> Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
>>>> ---
>>>>    tests/intel-ci/blacklist.txt | 3 +++
>>>>    1 file changed, 3 insertions(+)
>>>>
>>>> diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
>>>> index f6e9bb474..d01ef5e16 100644
>>>> --- a/tests/intel-ci/blacklist.txt
>>>> +++ b/tests/intel-ci/blacklist.txt
>>>> @@ -140,3 +140,6 @@ igt@device_reset@reset-bound
>>>>    
>>>>    # not working for i915
>>>>    igt@kms_bw@.*
>>>> +
>>>> +# Tests specific to (Display_ver > 14) are not applicable for i915
>>>> +igt@kms_vrr@cmrr
>>>
> 

Re: [i-g-t] tests/intel-ci: Block unsupported KMS tests for i915

[Kamil Konieczny]

Hi Modem,,
On 2024-06-20 at 10:35:34 +0530, Modem, Bhanuprakash wrote:
> 
> On 19-06-2024 04:11 pm, Jani Nikula wrote:
> > On Wed, 19 Jun 2024, "Modem, Bhanuprakash" <bhanuprakash.modem@intel.com> wrote:
> > > On 19-06-2024 03:33 pm, Jani Nikula wrote:
> > > > On Wed, 19 Jun 2024, Bhanuprakash Modem <bhanuprakash.modem@intel.com> wrote:
> > > > > Block KMS tests those are specific to (display_ver > 14)
> > > > > are not applicable for i915.
> > > > 
> > > > Feels a bit wrong to list features like this in a block list. IMO the
> > > > tests themselves should be aware of what's supported and what's not. I
> > > > imagine the rationale is performance? What's the impact?
> > > 
> > > I believe there is no point in running this test for i915, as it will
> > > always be skipped. The test already knows which platforms it supports
> > > and which it doesn't.
> > 
> > The question is whether it's worth it to start maintaining the info in
> > two places, the individual tests based on dynamic info obtained from the
> > driver, and a manually updated static block list. >
> > Are you also filtering xe testing for tests applicable to old platforms
> > only? Is the manual work worth the effort?
> 
> The only advantage I see in maintaining an additional list is improved
> reporting.
> 

Also saving CI time, the quickest run is no run. It also saves DB space
and man-hour when someone will need to look into reports and scroll all
those skips.

> Yes, we maintain an XE specific blocklist[*]. This manual effort is worthy
> for someone looking different execution reports. In fact, some efforts are
> going on to develop a platform-specific blocklist.
> 
> @kamil, Can you please add some details about upstreaming platform-specific
> blocklist?
> 
> [*]: https://cgit.freedesktop.org/drm/igt-gpu-tools/tree/tests/intel-ci/xe.blocklist.txt#n65
> 
> - Bhanu
> 
> > 
> > BR,
> > Jani.
> > 
> > 
> > > 
> > > - Bhanu
> > > 
> > > > 
> > > > BR,
> > > > Jani.
> > > > 
> > > > > 
> > > > > Signed-off-by: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
> > > > > ---
> > > > >    tests/intel-ci/blacklist.txt | 3 +++
> > > > >    1 file changed, 3 insertions(+)
> > > > > 
> > > > > diff --git a/tests/intel-ci/blacklist.txt b/tests/intel-ci/blacklist.txt
> > > > > index f6e9bb474..d01ef5e16 100644
> > > > > --- a/tests/intel-ci/blacklist.txt
> > > > > +++ b/tests/intel-ci/blacklist.txt
> > > > > @@ -140,3 +140,6 @@ igt@device_reset@reset-bound
> > > > >    # not working for i915
> > > > >    igt@kms_bw@.*
> > > > > +
> > > > > +# Tests specific to (Display_ver > 14) are not applicable for i915
> > > > > +igt@kms_vrr@cmrr

Acked-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>

Regards,
Kamil

> > > > 
> >