On Wed, Nov 11, 2015 at 05:51:52PM +0000, Russell King - ARM Linux wrote:
On Wed, Nov 11, 2015 at 03:34:32PM +0000, Liviu Dudau wrote:
While going through the code testing I've noticed an unbalanced .unbind missing drm_connector_unregister()
That actually doesn't matter, as DRM automatically tears them down anyway, so this isn't an urgent change. However, it's good practice to do so.
It looks like it doesn't, or at least not if the error code is -EPROBE_DEFER. On Juno, where the clocks are provided by SCPI and the load order is not guaranteed, the first bind will fail with -EPROBE_DEFER but the sysfs entry is not cleaned up, so on the next attempt the drm_connector_register() call will fail.
Best regards, Liviu
-- FTTC broadband for 0.8mile line: currently at 9.6Mbps down 400kbps up according to speedtest.net.