On Mon, Jan 30, 2023 at 6:28 PM Maxime Ripard <maxime@xxxxxxxxxx> wrote: > > On Thu, Jan 26, 2023 at 08:48:48PM +0530, Jagan Teki wrote: > > On Thu, Jan 26, 2023 at 5:42 PM Maxime Ripard <maxime@xxxxxxxxxx> wrote: > > > > > > Hi, > > > > > > On Mon, Jan 23, 2023 at 08:41:56PM +0530, Jagan Teki wrote: > > > > Add devm OF helper to return the next DSI bridge in the chain. > > > > > > > > Unlike general bridge return helper devm_drm_of_get_bridge, this > > > > helper uses the dsi specific panel_or_bridge helper to find the > > > > next DSI device in the pipeline. > > > > > > > > Helper lookup a given child DSI node or a DT node's port and > > > > endpoint number, find the connected node and return either > > > > the associated struct drm_panel or drm_bridge device. > > > > > > I'm not sure that using a device managed helper is the right choice > > > here. The bridge will stay longer than the backing device so it will > > > create a use-after-free. You should probably use a DRM-managed action > > > here instead. > > > > Thanks for the comments. If I understand correctly we can use > > drmm_panel_bridge_add instead devm_drm_panel_bridge_add once we found > > the panel or bridge - am I correct? > > It's not that we can, it's that the devm_panel_bridge_add is unsafe: > when the module is removed the device will go away and all the devm > resources freed, but the DRM device sticks around until the last > application with a fd open closes that fd. Thanks for the details. I think this is the reason you have introduced this DRM-managed action helper - drmm_of_get_bridge. Initially, i thought of adding similar, but as you are aware it is not possible to call it from the host attach. Jagan.
