[PATCH 19/20] drm/i915/dp: Ack only the handled device service IRQs

[Luca Coelho]

On Thu, 2025-07-03 at 16:27 +0300, Imre Deak wrote:
> On Thu, Jul 03, 2025 at 04:18:21PM +0300, Imre Deak wrote:
> > On Thu, Jul 03, 2025 at 04:14:51PM +0300, Luca Coelho wrote:
> > > On Thu, 2025-06-26 at 11:20 +0300, Imre Deak wrote:
> > > > From: Imre Deak <imre.deak at gmail.com>
> > > > 
> > > > Only those IRQs should be acked that are handled, however for SST all
> > > > IRQs triggered by the sink are acked. This can be a problem for flags
> > > > that are reserved/reading zero at a given moment, but become used for
> > > > some purpose - with a side-effect if set - in a future DPCD revision.
> > > > 
> > > > Fix the above by acking only those device service IRQs that will be
> > > > handled. While at it add asserts that only the known/acked device
> > > > service IRQs are handled both in the MST and SST case.
> > > > 
> > > > Signed-off-by: Imre Deak <imre.deak at gmail.com>
> > > > ---
> > > >  drivers/gpu/drm/i915/display/intel_dp.c | 14 ++++++++++++++
> > > >  1 file changed, 14 insertions(+)
> > > > 
> > > > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> > > > index 52249fa5c8a6d..6f67fac9724e1 100644
> > > > --- a/drivers/gpu/drm/i915/display/intel_dp.c
> > > > +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> > > > @@ -4537,6 +4537,14 @@ intel_dp_mst_disconnect(struct intel_dp *intel_dp)
> > > >  	drm_dp_mst_topology_mgr_set_mst(&intel_dp->mst.mgr, intel_dp->is_mst);
> > > >  }
> > > >  
> > > > +#define INTEL_DP_DEVICE_SERVICE_IRQ_MASK_SST	(DP_AUTOMATED_TEST_REQUEST | \
> > > > +						 DP_CP_IRQ | \
> > > > +						 DP_SINK_SPECIFIC_IRQ)
> > > > +
> > > > +#define INTEL_DP_DEVICE_SERVICE_IRQ_MASK_MST	(DP_CP_IRQ | \
> > > > +						 DP_DOWN_REP_MSG_RDY | \
> > > > +						 DP_UP_REQ_MSG_RDY)
> > > > +
> > > >  static bool
> > > >  intel_dp_get_sink_irq_esi(struct intel_dp *intel_dp, u8 *esi)
> > > >  {
> > > > @@ -4628,6 +4636,8 @@ static bool intel_dp_get_and_ack_sink_irq_esi_sst(struct intel_dp *intel_dp, u8
> > > >  		    encoder->base.base.id, encoder->base.name,
> > > >  		    esi);
> > > >  
> > > > +	esi[1] &= INTEL_DP_DEVICE_SERVICE_IRQ_MASK_SST;
> > > > +
> > > >  	if (mem_is_zero(&esi[1], 3))
> > > >  		return true;
> > > >  
> > > > @@ -5172,6 +5182,8 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp)
> > > >  		if (mem_is_zero(ack, sizeof(ack)))
> > > >  			break;
> > > >  
> > > > +		drm_WARN_ON(display->drm, ack[1] & ~INTEL_DP_DEVICE_SERVICE_IRQ_MASK_MST);
> > > > +
> > > >  		if (!intel_dp_ack_sink_irq_esi(intel_dp, ack))
> > > >  			drm_dbg_kms(display->drm, "Failed to ack ESI\n");
> > > >  
> > > > @@ -5456,6 +5468,8 @@ static void intel_dp_handle_device_service_irq(struct intel_dp *intel_dp, u8 irq
> > > >  {
> > > >  	struct intel_display *display = to_intel_display(intel_dp);
> > > >  
> > > > +	drm_WARN_ON(display->drm, irq_mask & ~INTEL_DP_DEVICE_SERVICE_IRQ_MASK_SST);
> > > > +
> > > >  	if (irq_mask & DP_AUTOMATED_TEST_REQUEST)
> > > >  		intel_dp_test_request(intel_dp);
> > > >  
> > > 
> > > Evidently, this function is specific to SST irq.  Wouldn't it be better
> > > to have _sst in the name like in "intel_dp_check_mst_status()"? This
> > > function is probably in an earlier patch, though, so if handled it
> > > should be obviously be done there.
> > 
> > Yes, it used to be SST specific, but patch 18 reused it for MST as well.
> 
> Ah, sorry, yes this function is still only used for SST. However the
> device service IRQs handled here are not SST specific and imo the
> function should be used for MST as well. But until that I can rename it
> to intel_dp_handle_device_service_irq_sst().

No, no need to rename it temporarily.  I changed my mind when I read
the next patch (see my comment there).  The only difference is that one
is link service and the other one is device service, not MST vs SST, so
I think it's better to keep this function consistent with the next one.
It's just an artifact that it is not needed for MST in this case.

--
Cheers,
Luca.