On Tue, Jan 30, 2024 at 07:56:26AM +0000, Lin, Charlton wrote:
> On 1/27/2024 12:06 PM, Arun R Murthy wrote:
> > With DP2.1, multistream packetization and the underneth MST protocol
> > will be required for SST. So check for MSTM_CAP to see if MST is really
> > required and skip the MSTM_CTRL write so that we ensure that only the
> > underneth protocol and the multistream packetization will be enabled and
> > sink will not be confused by a corresponding dpcd write.
> >
> > Signed-off-by: Arun R Murthy <arun.r.murthy@xxxxxxxxx>
> > ---
> > drivers/gpu/drm/display/drm_dp_mst_topology.c | 26 +++++++++++--------
> > 1 file changed, 15 insertions(+), 11 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/display/drm_dp_mst_topology.c b/drivers/gpu/drm/display/drm_dp_mst_topology.c
> > index 8ca01a6bf645..22d81732a978 100644
> > --- a/drivers/gpu/drm/display/drm_dp_mst_topology.c
> > +++ b/drivers/gpu/drm/display/drm_dp_mst_topology.c
> > @@ -3666,10 +3666,11 @@ int drm_dp_mst_topology_mgr_set_mst(struct drm_dp_mst_topology_mgr *mgr, bool ms
> > mgr->mst_primary = mstb;
> > drm_dp_mst_topology_get_mstb(mgr->mst_primary);
> >
> > - ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
> > - DP_MST_EN |
> > - DP_UP_REQ_EN |
> > - DP_UPSTREAM_IS_SRC);
> > + if (drm_dp_read_mst_cap(mgr->aux, mgr->dpcd))
> > + ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
> > + DP_MST_EN |
> > + DP_UP_REQ_EN |
> > + DP_UPSTREAM_IS_SRC);
> > if (ret < 0)
> > goto out_unlock;
> >
> > @@ -3684,7 +3685,8 @@ int drm_dp_mst_topology_mgr_set_mst(struct drm_dp_mst_topology_mgr *mgr, bool ms
> > mstb = mgr->mst_primary;
> > mgr->mst_primary = NULL;
> > /* this can fail if the device is gone */
> > - drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, 0);
> > + if (drm_dp_read_mst_cap(mgr->aux, mgr->dpcd))
> > + drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, 0);
> > ret = 0;
> > mgr->payload_id_table_cleared = false;
> >
> > @@ -3724,8 +3726,9 @@ drm_dp_mst_topology_mgr_invalidate_mstb(struct drm_dp_mst_branch *mstb)
> > void drm_dp_mst_topology_mgr_suspend(struct drm_dp_mst_topology_mgr *mgr)
> > {
> > mutex_lock(&mgr->lock);
> > - drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
> > - DP_MST_EN | DP_UPSTREAM_IS_SRC);
> > + if (drm_dp_read_mst_cap(mgr->aux, mgr->dpcd))
> > + drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
> > + DP_MST_EN | DP_UPSTREAM_IS_SRC);
> > mutex_unlock(&mgr->lock);
> > flush_work(&mgr->up_req_work);
> > flush_work(&mgr->work);
> > @@ -3773,10 +3776,11 @@ int drm_dp_mst_topology_mgr_resume(struct drm_dp_mst_topology_mgr *mgr,
> > goto out_fail;
> > }
> >
> > - ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
> > - DP_MST_EN |
> > - DP_UP_REQ_EN |
> > - DP_UPSTREAM_IS_SRC);
> > + if (drm_dp_read_mst_cap(mgr->aux, mgr->dpcd))
> > + ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL,
>
> This did not compile in Chrome dev environemnt due to uninitialized variable ret.
>
> > + DP_MST_EN |
> > + DP_UP_REQ_EN |
> > + DP_UPSTREAM_IS_SRC);
> > if (ret < 0) {
> > drm_dbg_kms(mgr->dev, "mst write failed - undocked during suspend?\n");
> > goto out_fail;
>
> Tested-by: Charlton Lin <charlton.lin@xxxxxxxxx>
The mst code does the whole toplogy discovery over sideband,
so if we don't enable that we should not even get a connector
registered. How is anything at all working with this?
--
Ville Syrjälä
Intel
