Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [High] Missing OF node reference acquisition for explicitly provided device_node `np`, leading to a reference count underflow when the device is destroyed or initialization fails. --
commit e57ae2f012dd4cdc820d51259b2a37d588d5c31d Author: Chaoyi Chen <[email protected]> drm/bridge: aux: Add drm_aux_bridge_register_from_node() This commit adds a new function, drm_aux_bridge_register_from_node(), which allows specifying the of_node corresponding to the bridge when creating a simple DRM bridge device. > diff --git a/drivers/gpu/drm/bridge/aux-bridge.c > b/drivers/gpu/drm/bridge/aux-bridge.c > index 1ed21a8713bfd..06a1466f49bd2 100644 > --- a/drivers/gpu/drm/bridge/aux-bridge.c > +++ b/drivers/gpu/drm/bridge/aux-bridge.c [ ... ] > @@ -62,7 +63,10 @@ int drm_aux_bridge_register(struct device *parent) > adev->dev.parent = parent; > adev->dev.release = drm_aux_bridge_release; > > - device_set_of_node_from_dev(&adev->dev, parent); > + if (np) > + device_set_node(&adev->dev, of_fwnode_handle(np)); [Severity: High] Does this miss a reference increment for the device node? Looking at drivers/base/core.c, device_set_node() simply assigns the pointer without taking a reference, whereas device_set_of_node_from_dev() internally calls of_node_get(dev2->of_node). > + else > + device_set_of_node_from_dev(&adev->dev, parent); > > ret = auxiliary_device_init(adev); > if (ret) { > of_node_put(adev->dev.of_node); [Severity: High] If np was provided and auxiliary_device_init() fails, won't this drop a reference that was never acquired? The same reference count underflow appears to happen during normal teardown in drm_aux_bridge_release(), which will also unconditionally call of_node_put() on a node that didn't have its reference count incremented here. Should an explicit of_node_get(np) be used before assigning it? [ ... ] -- Sashiko AI review ยท https://sashiko.dev/#/patchset/[email protected]?part=4
