Re: [PATCH 10/12] drm/i915/hdcp: Don't enable HDCP directly from check_link

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




On 2/15/2024 10:35 AM, Suraj Kandpal wrote:
Whenever LIC fails instead of moving from ENABLED to DESIRED
CP property we directly enable HDCP without informing the userspace
of this failure in link integrity check.
Now we will just update the value to DESIRED send the event to
userspace and then continue with the normal flow of HDCP enablement.

Signed-off-by: Suraj Kandpal <suraj.kandpal@xxxxxxxxx>
---
  drivers/gpu/drm/i915/display/intel_hdcp.c | 31 ++++-------------------
  1 file changed, 5 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 16b2b180563f..b22dbc6494e0 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -30,7 +30,7 @@
  #define KEY_LOAD_TRIES	5
  #define HDCP2_LC_RETRY_CNT			3
-static int intel_conn_to_vcpi(struct drm_atomic_state *state,
+static int intel_conn_to_vcpi(struct intel_atomic_state *state,
  			      struct intel_connector *connector)


This change does not belong to this patch. Perhaps should be moved to the next patch.


  {
  	struct drm_dp_mst_topology_mgr *mgr;
@@ -43,7 +43,7 @@ static int intel_conn_to_vcpi(struct drm_atomic_state *state,
  		return 0;
  	mgr = connector->port->mgr;
- drm_modeset_lock(&mgr->base.lock, state->acquire_ctx);
+	drm_modeset_lock(&mgr->base.lock, state->base.acquire_ctx);

Same as above.


  	mst_state = to_drm_dp_mst_topology_state(mgr->base.state);
  	payload = drm_atomic_get_mst_payload_state(mst_state, connector->port);
  	if (drm_WARN_ON(mgr->dev, !payload))
@@ -2068,17 +2068,6 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
drm_dbg_kms(&i915->drm,
  			    "HDCP2.2 Downstream topology change\n");
-		ret = hdcp2_authenticate_repeater_topology(connector);
-		if (!ret) {
-			intel_hdcp_update_value(connector,
-					DRM_MODE_CONTENT_PROTECTION_ENABLED,
-					true);
-			goto out;
-		}

I think we still need these. If repeater topology is authenticated, we can go ahead with hdcp_update_value.

Otherwise we disable hdcp first.


Also, we need to make similar change for hdcp_check_link.


Regards,

Ankit



-		drm_dbg_kms(&i915->drm,
-			    "[CONNECTOR:%d:%s] Repeater topology auth failed.(%d)\n",
-			    connector->base.base.id, connector->base.name,
-			    ret);
  	} else {
  		drm_dbg_kms(&i915->drm,
  			    "[CONNECTOR:%d:%s] HDCP2.2 link failed, retrying auth\n",
@@ -2095,18 +2084,8 @@ static int intel_hdcp2_check_link(struct intel_connector *connector)
  		goto out;
  	}
- ret = _intel_hdcp2_enable(connector);
-	if (ret) {
-		drm_dbg_kms(&i915->drm,
-			    "[CONNECTOR:%d:%s] Failed to enable hdcp2.2 (%d)\n",
-			    connector->base.base.id, connector->base.name,
-			    ret);
-		intel_hdcp_update_value(connector,
-					DRM_MODE_CONTENT_PROTECTION_DESIRED,
-					true);
-		goto out;
-	}
-
+	intel_hdcp_update_value(connector,
+				DRM_MODE_CONTENT_PROTECTION_DESIRED, true);
  out:
  	mutex_unlock(&dig_port->hdcp_mutex);
  	mutex_unlock(&hdcp->mutex);
@@ -2345,7 +2324,7 @@ intel_hdcp_set_streams(struct intel_digital_port *dig_port,
  			continue;
data->streams[data->k].stream_id =
-			intel_conn_to_vcpi(&state->base, connector);
+			intel_conn_to_vcpi(state, connector);
  		data->k++;
/* if there is only one active stream */



[Index of Archives]     [AMD Graphics]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux