On Tue, 2026-08-25 at 11:22 +0530, Animesh Manna wrote: > intel_alpm_lobf_compute_config_late() selected the main-link wake > time > with the two cases inverted: when aux-less wake was supported it used > the > aux-wake io_wake_lines, and otherwise the aux_less_wake_lines. This > is the > opposite of the mapping used in intel_psr_compute_config_late() and > of the > underlying meaning of the two values. > > As a result the LOBF enable decision (scl + guardband > > first_sdp_position > + waketime_in_lines) was computed against the wrong wake time, so > LOBF > could be enabled or rejected incorrectly. > > Use aux_less_wake_lines when aux-less wake is supported and > io_wake_lines > otherwise, matching the PSR path. > > Cc: Jouni Högander <[email protected]> > Cc: Ankit Nautiyal <[email protected] > Cc: [email protected] > Fixes: 9090fa44a6d45 ("drm/i915/alpm: Compute LOBF late after > guardband is already determined") > Assisted-by: Claude:claude-opus-4-8 > Signed-off-by: Animesh Manna <[email protected]>
Reviewed-by: Jouni Högander <[email protected]> > --- > drivers/gpu/drm/i915/display/intel_alpm.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c > b/drivers/gpu/drm/i915/display/intel_alpm.c > index f1383764b702..559fe6d14c01 100644 > --- a/drivers/gpu/drm/i915/display/intel_alpm.c > +++ b/drivers/gpu/drm/i915/display/intel_alpm.c > @@ -314,9 +314,9 @@ void intel_alpm_lobf_compute_config_late(struct > intel_dp *intel_dp, > */ > first_sdp_position = adjusted_mode->crtc_vtotal - > adjusted_mode->crtc_vsync_start; > if (intel_alpm_aux_less_wake_supported(intel_dp)) > - waketime_in_lines = crtc_state- > >alpm_state.io_wake_lines; > - else > waketime_in_lines = crtc_state- > >alpm_state.aux_less_wake_lines; > + else > + waketime_in_lines = crtc_state- > >alpm_state.io_wake_lines; > > crtc_state->has_lobf = (crtc_state->set_context_latency + > crtc_state->vrr.guardband) > > (first_sdp_position + > waketime_in_lines);
