> -----Original Message----- > From: Hogander, Jouni <jouni.hogander@xxxxxxxxx> > Sent: Thursday, June 6, 2024 9:18 PM > To: Manna, Animesh <animesh.manna@xxxxxxxxx>; intel- > gfx@xxxxxxxxxxxxxxxxxxxxx > Cc: Kahola, Mika <mika.kahola@xxxxxxxxx> > Subject: Re: [PATCH v6 14/26] dmr/i915/psr: Wake time is aux less wake time > for Panel Replay > > On Thu, 2024-06-06 at 13:26 +0000, Manna, Animesh wrote: > > > > > > > -----Original Message----- > > > From: Hogander, Jouni <jouni.hogander@xxxxxxxxx> > > > Sent: Wednesday, June 5, 2024 3:56 PM > > > To: intel-gfx@xxxxxxxxxxxxxxxxxxxxx > > > Cc: Manna, Animesh <animesh.manna@xxxxxxxxx>; Kahola, Mika > > > <mika.kahola@xxxxxxxxx>; Hogander, Jouni <jouni.hogander@xxxxxxxxx> > > > Subject: [PATCH v6 14/26] dmr/i915/psr: Wake time is aux less wake > > > time for Panel Replay > > > > > > When checking vblank length used wake time is aux less wake time for > > > eDP Panel Replay (vblank length is not checked for DP2.0 Panel > > > Replay). > > > > > > Signed-off-by: Jouni Högander <jouni.hogander@xxxxxxxxx> > > > --- > > > drivers/gpu/drm/i915/display/intel_psr.c | 8 ++++++-- > > > 1 file changed, 6 insertions(+), 2 deletions(-) > > > > > > diff --git a/drivers/gpu/drm/i915/display/intel_psr.c > > > b/drivers/gpu/drm/i915/display/intel_psr.c > > > index c9fcb25184ee..ccb7c022e364 100644 > > > --- a/drivers/gpu/drm/i915/display/intel_psr.c > > > +++ b/drivers/gpu/drm/i915/display/intel_psr.c > > > @@ -1251,8 +1251,12 @@ static bool vblank_length_valid(struct > > > intel_dp *intel_dp, > > > crtc_state->hw.adjusted_mode.crtc_vblank_start; > > > int wake_lines; > > > > > > - wake_lines = DISPLAY_VER(i915) < 20 ? > > > psr2_block_count_lines(intel_dp) : > > > - intel_dp->alpm_parameters.io_wake_lines; > > > + if (crtc_state->has_panel_replay) > > > + wake_lines = intel_dp- > > > > alpm_parameters.aux_less_wake_lines; > > > > Existing code can take care for EDP panel replay as EDP panel replay > > is supported from LNL onwards. > > Not sure why this change is needed. > > Panel Replay eDP --> aux less wake lines is used for the check > > LunarLake PSR2 --> io wake lines is used for the check > > Prior LunarLake PSR2 --> psr2_block_count_lines is used for the check > > Existing code is taking care of last two cases. Change is taking care of the first > one. Got it, thanks for clarifying. Reviewed-by: Animesh Manna <animesh.manna@xxxxxxxxx> > > BR, > > Jouni Högander > > > > > Regards, > > Animesh > > > > > + else > > > + wake_lines = DISPLAY_VER(i915) < 20 ? > > > + psr2_block_count_lines(intel_dp) : > > > + intel_dp->alpm_parameters.io_wake_lines; > > > > > > if (crtc_state->req_psr2_sdp_prior_scanline) > > > vblank -= 1; > > > -- > > > 2.34.1 > >
