[Intel-gfx] [PATCH] drm/i915: Avoid null dereference if mst_port is unset.
I'm not sure if this is really the case and I don't believe this is the real fix for the bug mentioned here, but since I don't see a reliable path when mst_port is set and when mode_valid is requested I believe it is worth to have this protection here. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102022 Cc: Elizabeth Cc: Stefan Assmann Cc: Dhinakaran Pandiyan Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/i915/intel_dp_mst.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 58568559711a..93fc8ab9bb31 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -370,6 +370,9 @@ intel_dp_mst_mode_valid(struct drm_connector *connector, int bpp = 24; /* MST uses fixed bpp */ int max_rate, mode_rate, max_lanes, max_link_clock; + if (!intel_dp) + return MODE_ERROR; + max_link_clock = intel_dp_max_link_rate(intel_dp); max_lanes = intel_dp_max_lane_count(intel_dp); -- 2.13.2 ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH] drm/i915: Avoid null dereference if mst_port is unset.
Quoting Rodrigo Vivi (2017-08-10 15:50:43) > I'm not sure if this is really the case and I don't believe > this is the real fix for the bug mentioned here, but since > I don't see a reliable path when mst_port is set and when > mode_valid is requested I believe it is worth to have this > protection here. The guard looks correct. We defend ourselves against the async disconnect elsewhere through the post-detect callbacks, i.e. we cannot rely on intel_connector->mst_port remaining valid for a whole detect() cycle. Reviewed-by: Chris Wilson -Chris ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH] drm/i915: Avoid null dereference if mst_port is unset.
On 2017-08-10 07:50, Rodrigo Vivi wrote: > I'm not sure if this is really the case and I don't believe > this is the real fix for the bug mentioned here, but since > I don't see a reliable path when mst_port is set and when > mode_valid is requested I believe it is worth to have this > protection here. Hi Rodrigo, I've applied the patch but problem seems to be somewhere else. After switching display input the system is no longer reachable via ssh. I ran dmesg -w over ssh and captured the output as far as possible. Have a look: https://paste.fedoraproject.org/paste/mgzW~jAzpaS4A3rmfkTZKw Thanks! Stefan > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102022 > Cc: Elizabeth > Cc: Stefan Assmann > Cc: Dhinakaran Pandiyan > Signed-off-by: Rodrigo Vivi > --- > drivers/gpu/drm/i915/intel_dp_mst.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c > b/drivers/gpu/drm/i915/intel_dp_mst.c > index 58568559711a..93fc8ab9bb31 100644 > --- a/drivers/gpu/drm/i915/intel_dp_mst.c > +++ b/drivers/gpu/drm/i915/intel_dp_mst.c > @@ -370,6 +370,9 @@ intel_dp_mst_mode_valid(struct drm_connector *connector, > int bpp = 24; /* MST uses fixed bpp */ > int max_rate, mode_rate, max_lanes, max_link_clock; > > + if (!intel_dp) > + return MODE_ERROR; > + > max_link_clock = intel_dp_max_link_rate(intel_dp); > max_lanes = intel_dp_max_lane_count(intel_dp); > > -- > 2.13.2 > ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH] drm/i915: Avoid null dereference if mst_port is unset.
On Thu, 2017-08-10 at 19:45 +0200, Stefan Assmann wrote: > On 2017-08-10 07:50, Rodrigo Vivi wrote: > > I'm not sure if this is really the case and I don't believe > > this is the real fix for the bug mentioned here, but since > > I don't see a reliable path when mst_port is set and when > > mode_valid is requested I believe it is worth to have this > > protection here. > > Hi Rodrigo, > > I've applied the patch but problem seems to be somewhere else. After > switching display input the system is no longer reachable via ssh. > I ran dmesg -w over ssh and captured the output as far as possible. > Have a look: > https://paste.fedoraproject.org/paste/mgzW~jAzpaS4A3rmfkTZKw > > Thanks! > > Stefan > Looking at the logs, it appears you have a MST monitor connected to a MST-dock. DP-3 should have been branch device that always remains disconnected. But the log shows [3.812789] [drm] forcing DP-3 connector on [3.812809] [drm:drm_connector_init [drm]] cmdline mode for connector DP-3 0x0@60Hz And then when there is a long pulse (when you switched modes on the monitor), which should be marking all connectors as disconnected. [ 244.504884] [drm:intel_get_hpd_pins [i915]] hotplug event received, stat 0x0040, dig 0x00101210, pins 0x0040 This happens - [ 244.668328] [drm:drm_helper_probe_single_connector_modes [drm_kms_helper]] [CONNECTOR:79:DP-3] status updated from disconnected to connected Looks like this has something to do with the connector status being forced. -DK > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102022 > > Cc: Elizabeth > > Cc: Stefan Assmann > > Cc: Dhinakaran Pandiyan > > Signed-off-by: Rodrigo Vivi > > --- > > drivers/gpu/drm/i915/intel_dp_mst.c | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c > > b/drivers/gpu/drm/i915/intel_dp_mst.c > > index 58568559711a..93fc8ab9bb31 100644 > > --- a/drivers/gpu/drm/i915/intel_dp_mst.c > > +++ b/drivers/gpu/drm/i915/intel_dp_mst.c > > @@ -370,6 +370,9 @@ intel_dp_mst_mode_valid(struct drm_connector *connector, > > int bpp = 24; /* MST uses fixed bpp */ > > int max_rate, mode_rate, max_lanes, max_link_clock; > > > > + if (!intel_dp) > > + return MODE_ERROR; > > + > > max_link_clock = intel_dp_max_link_rate(intel_dp); > > max_lanes = intel_dp_max_lane_count(intel_dp); > > > > -- > > 2.13.2 > > ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH] drm/i915: Avoid null dereference if mst_port is unset.
On Thu, Aug 10, 2017 at 9:23 AM, Chris Wilson wrote: > Quoting Rodrigo Vivi (2017-08-10 15:50:43) >> I'm not sure if this is really the case and I don't believe >> this is the real fix for the bug mentioned here, but since >> I don't see a reliable path when mst_port is set and when >> mode_valid is requested I believe it is worth to have this >> protection here. > > The guard looks correct. We defend ourselves against the async > disconnect elsewhere through the post-detect callbacks, i.e. we cannot > rely on intel_connector->mst_port remaining valid for a whole detect() > cycle. > > Reviewed-by: Chris Wilson thanks, merged to dinq. the discussion around the actual bug should move back to 102022 > -Chris > ___ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx -- Rodrigo Vivi Blog: http://blog.vivi.eng.br ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
