| Message ID | 20181129110921.5832-1-stanislav.lisovskiy@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v3] Return only active connectors for get_resources ioctl | expand |
On Thu, Nov 29, 2018 at 01:09:21PM +0200, Stanislav Lisovskiy wrote: > Currently kernel might allocate different connector ids > for the same outputs in case of DP MST, which seems to > confuse userspace. There are can be different connector > ids in the list, which could be assigned to the same > output, while being in different states. > This results in issues, like external displays staying > blank after quick unplugging and plugging back(bug #106250). > Returning only active DP connectors fixes the issue. > > v2: Removed caps from the title > > v3: Refactored initial condition to look more compact. > Thanks to Chris Wilson for giving a hint. > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106250 > Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> > --- > drivers/gpu/drm/drm_mode_config.c | 6 ++++++ > 1 file changed, 6 insertions(+) <formletter> This is not the correct way to submit patches for inclusion in the stable kernel tree. Please read: https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html for how to do this properly. </formletter>
On Thu, 2018-11-29 at 13:09 +0200, Stanislav Lisovskiy wrote: > Currently kernel might allocate different connector ids > for the same outputs in case of DP MST, which seems to > confuse userspace. There are can be different connector > ids in the list, which could be assigned to the same > output, while being in different states. > This results in issues, like external displays staying > blank after quick unplugging and plugging back(bug #106250). > Returning only active DP connectors fixes the issue. > > v2: Removed caps from the title > > v3: Refactored initial condition to look more compact. > Thanks to Chris Wilson for giving a hint. > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106250 > Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> > --- > drivers/gpu/drm/drm_mode_config.c | 6 ++++++ > 1 file changed, 6 insertions(+) > > diff --git a/drivers/gpu/drm/drm_mode_config.c > b/drivers/gpu/drm/drm_mode_config.c > index ee80788f2c40..3e2cd959a66a 100644 > --- a/drivers/gpu/drm/drm_mode_config.c > +++ b/drivers/gpu/drm/drm_mode_config.c > @@ -143,22 +143,28 @@ int drm_mode_getresources(struct drm_device *dev, void > *data, > drm_connector_list_iter_begin(dev, &conn_iter); > count = 0; > connector_id = u64_to_user_ptr(card_res->connector_id_ptr); > + DRM_DEBUG_KMS("GetResources: writing connectors start"); > drm_for_each_connector_iter(connector, &conn_iter) { > /* only expose writeback connectors if userspace understands > them */ > if (!file_priv->writeback_connectors && > (connector->connector_type == > DRM_MODE_CONNECTOR_WRITEBACK)) > continue; > > + if (READ_ONCE(connector->registration_state) != > DRM_CONNECTOR_REGISTERED) > + continue; > + This conditional should just use drm_connector_is_unregistered() instead I'm fairly sure. > if (drm_lease_held(file_priv, connector->base.id)) { > if (count < card_res->count_connectors && > put_user(connector->base.id, connector_id + > count)) { > drm_connector_list_iter_end(&conn_iter); > return -EFAULT; > } > + DRM_DEBUG_KMS("GetResources: connector %s", connector- > >name); > count++; > } > } > card_res->count_connectors = count; > + DRM_DEBUG_KMS("GetResources: writing connectors end - count %d", > count); > drm_connector_list_iter_end(&conn_iter); > > return ret;
On Thu, 29 Nov 2018, Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> wrote: > Currently kernel might allocate different connector ids > for the same outputs in case of DP MST, which seems to > confuse userspace. There are can be different connector > ids in the list, which could be assigned to the same > output, while being in different states. > This results in issues, like external displays staying > blank after quick unplugging and plugging back(bug #106250). > Returning only active DP connectors fixes the issue. Some nitpicks here... The patch subject lacks a prefix. > > v2: Removed caps from the title > > v3: Refactored initial condition to look more compact. > Thanks to Chris Wilson for giving a hint. > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106250 > Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> > --- > drivers/gpu/drm/drm_mode_config.c | 6 ++++++ > 1 file changed, 6 insertions(+) > > diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c > index ee80788f2c40..3e2cd959a66a 100644 > --- a/drivers/gpu/drm/drm_mode_config.c > +++ b/drivers/gpu/drm/drm_mode_config.c > @@ -143,22 +143,28 @@ int drm_mode_getresources(struct drm_device *dev, void *data, > drm_connector_list_iter_begin(dev, &conn_iter); > count = 0; > connector_id = u64_to_user_ptr(card_res->connector_id_ptr); > + DRM_DEBUG_KMS("GetResources: writing connectors start"); I think the added debugs here are too verbose. You should probably separate the debugs from the actual change here. Note also that the logging functions already add the function name, so "GetResources:" is redundant. Also "\n" is missing. > drm_for_each_connector_iter(connector, &conn_iter) { > /* only expose writeback connectors if userspace understands them */ > if (!file_priv->writeback_connectors && > (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK)) > continue; > > + if (READ_ONCE(connector->registration_state) != DRM_CONNECTOR_REGISTERED) > + continue; > + > if (drm_lease_held(file_priv, connector->base.id)) { > if (count < card_res->count_connectors && > put_user(connector->base.id, connector_id + count)) { > drm_connector_list_iter_end(&conn_iter); > return -EFAULT; > } > + DRM_DEBUG_KMS("GetResources: connector %s", connector->name); Please look around, the connector logging is fairly consistent throughout. BR, Jani. > count++; > } > } > card_res->count_connectors = count; > + DRM_DEBUG_KMS("GetResources: writing connectors end - count %d", count); > drm_connector_list_iter_end(&conn_iter); > > return ret;
diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c index ee80788f2c40..3e2cd959a66a 100644 --- a/drivers/gpu/drm/drm_mode_config.c +++ b/drivers/gpu/drm/drm_mode_config.c @@ -143,22 +143,28 @@ int drm_mode_getresources(struct drm_device *dev, void *data, drm_connector_list_iter_begin(dev, &conn_iter); count = 0; connector_id = u64_to_user_ptr(card_res->connector_id_ptr); + DRM_DEBUG_KMS("GetResources: writing connectors start"); drm_for_each_connector_iter(connector, &conn_iter) { /* only expose writeback connectors if userspace understands them */ if (!file_priv->writeback_connectors && (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK)) continue; + if (READ_ONCE(connector->registration_state) != DRM_CONNECTOR_REGISTERED) + continue; + if (drm_lease_held(file_priv, connector->base.id)) { if (count < card_res->count_connectors && put_user(connector->base.id, connector_id + count)) { drm_connector_list_iter_end(&conn_iter); return -EFAULT; } + DRM_DEBUG_KMS("GetResources: connector %s", connector->name); count++; } } card_res->count_connectors = count; + DRM_DEBUG_KMS("GetResources: writing connectors end - count %d", count); drm_connector_list_iter_end(&conn_iter); return ret;
Currently kernel might allocate different connector ids for the same outputs in case of DP MST, which seems to confuse userspace. There are can be different connector ids in the list, which could be assigned to the same output, while being in different states. This results in issues, like external displays staying blank after quick unplugging and plugging back(bug #106250). Returning only active DP connectors fixes the issue. v2: Removed caps from the title v3: Refactored initial condition to look more compact. Thanks to Chris Wilson for giving a hint. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106250 Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> --- drivers/gpu/drm/drm_mode_config.c | 6 ++++++ 1 file changed, 6 insertions(+)