On Wed, Jul 24, 2024 at 07:21:53PM +0300, Jani Nikula wrote: > On Wed, 24 Jul 2024, Imre Deak <imre.deak@xxxxxxxxx> wrote: > > In some cases the sink can reset itself after it was configured into MST > > mode, without the driver noticing the disconnected state. For instance > > the reset may happen in the middle of a modeset, or the (long) HPD pulse > > generated may be not long enough for the encoder detect handler to > > observe the HPD's deasserted state. In this case the sink's DPCD > > register programmed to enable MST will be reset, while the driver still > > assumes MST is still enabled. Detect this condition, which will tear > > down and recreate/re-enable the MST topology. > > > > Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/11195 > > Signed-off-by: Imre Deak <imre.deak@xxxxxxxxx> > > --- > > drivers/gpu/drm/i915/display/intel_dp.c | 3 +++ > > drivers/gpu/drm/i915/display/intel_dp_mst.c | 30 +++++++++++++++++++++ > > drivers/gpu/drm/i915/display/intel_dp_mst.h | 1 + > > 3 files changed, 34 insertions(+) > > > > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c > > index 1e43e32e05199..c621f6daf8235 100644 > > --- a/drivers/gpu/drm/i915/display/intel_dp.c > > +++ b/drivers/gpu/drm/i915/display/intel_dp.c > > @@ -5878,6 +5878,9 @@ intel_dp_detect(struct drm_connector *connector, > > else > > status = connector_status_disconnected; > > > > + if (!intel_dp_mst_verify_dpcd_state(intel_dp)) > > + status = connector_status_disconnected; > > + > > if (status == connector_status_disconnected) { > > memset(&intel_dp->compliance, 0, sizeof(intel_dp->compliance)); > > memset(intel_connector->dp.dsc_dpcd, 0, sizeof(intel_connector->dp.dsc_dpcd)); > > diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c > > index 27ce5c3f5951e..89b147e37b400 100644 > > --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c > > +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c > > @@ -1998,3 +1998,33 @@ bool intel_dp_mst_crtc_needs_modeset(struct intel_atomic_state *state, > > > > return false; > > } > > + > > +/** > > + * intel_dp_mst_verify_dpcd_state - verify the MST SW enabled state wrt. the DPCD > > + * @intel_dp: DP port object > > + * > > + * Verify if @intel_dp's MST enabled SW state matches the corresponding DPCD > > + * state. A long HPD pulse -not long enough to be detected as a disconnected > > + * state - could've reset the DPCD state, which requires tearing > > + * down/recreating the MST topology. > > + * > > + * Returns %true if the SW MST enabled and DPCD states match, %false > > + * otherwise. > > + */ > > +bool intel_dp_mst_verify_dpcd_state(struct intel_dp *intel_dp) > > +{ > > + int ret; > > + u8 val; > > + > > + if (!intel_dp->is_mst) > > + return true; > > + > > + ret = drm_dp_dpcd_readb(intel_dp->mst_mgr.aux, DP_MSTM_CTRL, &val); > > + if (ret < 0) > > + return false; > > + > > + if (val != (DP_MST_EN | DP_UP_REQ_EN | DP_UPSTREAM_IS_SRC)) > > Hmm. This is not comparing the value to any software state, but some > hardcoded combo of values which may or may not match. The software state is intel_dp->is_mst. That's atm the way to check if MST was enabled previously and correspondingly DP_MSTM_CTRL programmed to the above value. There is also intel_dp->mst_detect, but that can't be used here, since that's reset after MST gets enabled. > In particular, this makes it harder to untangle MST from SST-with-sideband, > I think. After that's enabled this should be changed; there isn't a good way atm to tell full MST and SST-with-sideband apart. I don't see that as a problem as only full MST will be enabled. > > BR, > Jani. > > > + return false; > > + > > + return true; > > +} > > diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.h b/drivers/gpu/drm/i915/display/intel_dp_mst.h > > index 8ca1d599091c6..9e4c7679f1c3a 100644 > > --- a/drivers/gpu/drm/i915/display/intel_dp_mst.h > > +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.h > > @@ -27,5 +27,6 @@ int intel_dp_mst_atomic_check_link(struct intel_atomic_state *state, > > struct intel_link_bw_limits *limits); > > bool intel_dp_mst_crtc_needs_modeset(struct intel_atomic_state *state, > > struct intel_crtc *crtc); > > +bool intel_dp_mst_verify_dpcd_state(struct intel_dp *intel_dp); > > > > #endif /* __INTEL_DP_MST_H__ */ > > -- > Jani Nikula, Intel