On Tue, Mar 24, 2020 at 01:14:29PM -0700, José Roberto de Souza wrote:
> As now the cost to lock and use a TC port is higher due the
> implementation of the TCCOLD sequences it is worty to hold a reference
> of the TC port to avoid all this locking at every aux transaction
> part of the DisplayPort detection.
The problem with locking the port for detection is that it would block
modesets on the port, which we should avoid. By blocking tc-cold
whenever enabling an AUX power well you would avoid the overhead of the
PCODE requests for each AUX transfer, since the AUX power refs are
dropped asynchronously with a delay.
>
> Cc: Imre Deak <imre.deak@xxxxxxxxx>
> Cc: Cooper Chiou <cooper.chiou@xxxxxxxxx>
> Cc: Kai-Heng Feng <kai.heng.feng@xxxxxxxxxxxxx>
> Signed-off-by: José Roberto de Souza <jose.souza@xxxxxxxxx>
> ---
> drivers/gpu/drm/i915/display/intel_dp.c | 19 ++++++++++++++-----
> 1 file changed, 14 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> index 7f1a4e55cda1..6fbf3beee544 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> @@ -6041,6 +6041,7 @@ intel_dp_detect(struct drm_connector *connector,
> struct intel_dp *intel_dp = intel_attached_dp(to_intel_connector(connector));
> struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
> struct intel_encoder *encoder = &dig_port->base;
> + enum phy phy = intel_port_to_phy(dev_priv, encoder->port);
> enum drm_connector_status status;
>
> drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s]\n",
> @@ -6049,12 +6050,17 @@ intel_dp_detect(struct drm_connector *connector,
> !drm_modeset_is_locked(&dev_priv->drm.mode_config.connection_mutex));
>
> /* Can't disconnect eDP */
> - if (intel_dp_is_edp(intel_dp))
> + if (intel_dp_is_edp(intel_dp)) {
> status = edp_detect(intel_dp);
> - else if (intel_digital_port_connected(encoder))
> - status = intel_dp_detect_dpcd(intel_dp);
> - else
> - status = connector_status_disconnected;
> + } else {
> + if (intel_phy_is_tc(dev_priv, phy))
> + intel_tc_port_get_link(dig_port, 1);
> +
> + if (intel_digital_port_connected(encoder))
> + status = intel_dp_detect_dpcd(intel_dp);
> + else
> + status = connector_status_disconnected;
> + }
>
> if (status == connector_status_disconnected) {
> memset(&intel_dp->compliance, 0, sizeof(intel_dp->compliance));
> @@ -6132,6 +6138,9 @@ intel_dp_detect(struct drm_connector *connector,
> if (status != connector_status_connected && !intel_dp->is_mst)
> intel_dp_unset_edid(intel_dp);
>
> + if (intel_phy_is_tc(dev_priv, phy))
> + intel_tc_port_put_link(dig_port);
> +
> /*
> * Make sure the refs for power wells enabled during detect are
> * dropped to avoid a new detect cycle triggered by HPD polling.
> --
> 2.26.0
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
