[PATCH 08/12] drm/i915/display: Add guardband check for feature latencies

[Jani Nikula]

On Wed, 20 Aug 2025, Ankit Nautiyal <ankit.k.nautiyal at intel.com> wrote:
> +	if (HAS_VRR(display) && intel_vrr_possible(crtc_state)) {

Nitpick, and a tangential to designing stuff:

intel_vrr_possible() never returns true for !HAS_VRR(). The HAS_VRR()
check is redundant. Adding redundant checks adds uncertainty about what
intel_vrr_possible() can return. "Whoa, can it return true even for
!HAS_VRR()? Why?" And then it reinforces the mentality that everything
needs redundancy and double checking.

This is not about just that one check and one line. The idea is that for
most "has feature" checks that enable something in the crtc state, you
do that check in very few places, and the fields in crtc state dictate
the rest. You're not supposed to have to second guess what crtc state
has.

Food for though.


BR,
Jani.


-- 
Jani Nikula, Intel