[PATCH i-g-t v1] tests/kms_flip: Skip timestamp checks in Intel simulation environments

[Cavitt, Jonathan]

-----Original Message-----
From: igt-dev <igt-dev-bounces at lists.freedesktop.org> On Behalf Of Kamil Konieczny
Sent: Wednesday, July 2, 2025 6:06 AM
To: Naladala, Ramanaidu <ramanaidu.naladala at intel.com>
Cc: igt-dev at lists.freedesktop.org; Sharma, Swati2 <swati2.sharma at intel.com>; Golani, Mitulkumar Ajitkumar <mitulkumar.ajitkumar.golani at intel.com>
Subject: Re: [PATCH i-g-t v1] tests/kms_flip: Skip timestamp checks in Intel simulation environments
> 
> Hi Naladala,
> On 2025-06-24 at 14:22:17 +0530, Naladala Ramanaidu wrote:
> > Bypass timestamp and sequence validations when INTEL_SIMULATION is
> > active to accommodate non-deterministic timing in simulated test
> > environments.
> > 
> > Signed-off-by: Naladala Ramanaidu <ramanaidu.naladala at intel.com>
> > ---
> >  tests/kms_flip.c | 14 +++++++++++++-
> >  1 file changed, 13 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tests/kms_flip.c b/tests/kms_flip.c
> > index 65016c66d..db7ced8aa 100755
> > --- a/tests/kms_flip.c
> > +++ b/tests/kms_flip.c
> > @@ -318,7 +318,19 @@ static bool should_skip_ts_checks(void) {
> >  	 * timestamp to drift with a relatively larger standard deviation over a large sample.
> >  	 * As it's a known issue, skip any Timestamp or Sequence checks for MTK drivers.
> >  	 */
> > -	return is_mtk_device(drm_fd);
> > +	if (is_mtk_device(drm_fd))
> > +		return true;
> > +
> > +	/*
> > +	 * In simulation environments, hardware behavior may not accurately reflect real-world
> > +	 * timing characteristics. To avoid false negatives in tests due to simulated timing
> > +	 * artifacts, skip timestamp and sequence checks when the INTEL_SIMULATION environment
> > +	 * variable is set to a truthy value.
> > +	 */
> > +	if (igt_check_boolean_env_var("INTEL_SIMULATION", false))
> 
> Please use function: igt_run_in_simulation() so it will be:
> 
> 	if (igt_run_in_simulation())

I think we're looking for the opposite, actually:

"""
	if (!igt_run_in_simulation())
		return true;

	return false;
"""

Or, alternatively:

"""
	if (igt_run_in_simulation())
		return false;

	return true;
"""

Though, at that point, I think we'd may as well just return the value directly:

"""
	return !igt_run_in_simulation();
"""

-Jonathan Cavitt

> 
> Regards,
> Kamil
> 
> > +		return true;
> > +
> > +	return false;
> >  }
> >  
> >  static bool vblank_dependence(int flags)
> > -- 
> > 2.43.0
> > 
>