Hi Jagan, On 2023-06-01 06:10, Jagan Teki wrote: > On Tue, May 30, 2023 at 3:56 PM Jonas Karlman <jo...@kwiboo.se> wrote: >> >> The call to device_find_first_child always return 0, change to return >> early when there is no child node. > > Can you explain a little more about this? Maybe adding device-tree > pipeline will give a better understanding of the issue.
The docs for device_find_first_child mention: @devp: Returns first child device, or NULL if none Return: 0 And the function does exactly this, always return 0. So, the current check for return value of this call is unnecessary. The following possible call to dwc3_glue_clk_init or dwc3_glue_reset_init may trigger a null pointer dereference further down the call chain. Looking closer this patch may have been more related to my rk3328 testing, where usb@ff600000 in rk3328.dtsi is missing child nodes and resets prop. For RK3568 this patch should not be needed, however it still fixes a possible null pointer dereference issue. Regards, Jonas > > Jagan.
