[PATCH v5 5/6] drm/i915/dp: Queue a link check after link training is complete

[Imre Deak]

On Tue, Mar 04, 2025 at 07:56:28PM +0200, Jani Nikula wrote:
> On Tue, 04 Mar 2025, Imre Deak <imre.deak at intel.com> wrote:
> > After link training - both in case of a passing and failing LT result -
> > a work is scheduled to check the link state. This check should take
> > place after the link training is completed by disabling the link
> > training pattern and setting intel_dp::link_trained=true. Atm, the work
> > is scheduled before these steps, which may result in checking the link
> > state too early (and thus not retraining the link as expected).
> >
> > Fix the above by scheduling the link check work after link training is
> > complete.
> >
> > Signed-off-by: Imre Deak <imre.deak at intel.com>
> > ---
> >  .../gpu/drm/i915/display/intel_dp_link_training.c   | 13 +++++++++----
> >  1 file changed, 9 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
> > index 3906c11acc968..c1be073b9fc48 100644
> > --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
> > +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
> > @@ -1110,6 +1110,7 @@ intel_dp_128b132b_intra_hop(struct intel_dp *intel_dp,
> >  void intel_dp_stop_link_train(struct intel_dp *intel_dp,
> >  			      const struct intel_crtc_state *crtc_state)
> >  {
> > +	struct intel_display *display = to_intel_display(intel_dp);
> >  	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
> >  
> >  	intel_dp->link_trained = true;
> 
> Side note, it's a bit misleading that we set this even if link training
> failed... but we probably depend on it.

Yes, it's used in the sense that the link is active. I guess it could
be renamed to link_active accordingly (can follow up with that).

> 
> > @@ -1124,6 +1125,13 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
> >  	}
> >  
> >  	intel_hpd_unblock(encoder);
> > +
> > +	if (!display->hotplug.ignore_long_hpd &&
> 
> We already poke at this directly elsewhere in the file, but I'm not a
> fan of doing so. We should have a function and use that.

Agreed, can follow up with this separately.

> The fact that we can do anything doesn't mean we should.
> 
> I guess that's for another patch, another time.
> 
> > +	    intel_dp->link.seq_train_failures < 2) {
> 
> It's not great that the magic 2 is duplicated. It was fine in one place.

Ok, will add a macro for it.

> Regardless,
> 
> Reviewed-by: Jani Nikula <jani.nikula at intel.com>
> 
> > +		int delay_ms = intel_dp->link.seq_train_failures ? 0 : 2000;
> > +
> > +		intel_encoder_link_check_queue_work(encoder, delay_ms);
> > +	}
> >  }
> >  
> >  static bool
> > @@ -1628,7 +1636,6 @@ void intel_dp_start_link_train(struct intel_atomic_state *state,
> >  		lt_dbg(intel_dp, DP_PHY_DPRX, "Forcing link training failure\n");
> >  	} else if (passed) {
> >  		intel_dp->link.seq_train_failures = 0;
> > -		intel_encoder_link_check_queue_work(encoder, 2000);
> >  		return;
> >  	}
> >  
> > @@ -1651,10 +1658,8 @@ void intel_dp_start_link_train(struct intel_atomic_state *state,
> >  		return;
> >  	}
> >  
> > -	if (intel_dp->link.seq_train_failures < 2) {
> > -		intel_encoder_link_check_queue_work(encoder, 0);
> > +	if (intel_dp->link.seq_train_failures < 2)
> >  		return;
> > -	}
> >  
> >  	if (intel_dp_schedule_fallback_link_training(state, intel_dp, crtc_state))
> 
> 
> 
> >  		return;
> 
> -- 
> Jani Nikula, Intel