Re: [PATCH 0/2] HID: alps: fix DualPoint Stick input device lifecycle

[Jiri Kosina]

On Fri, 14 Aug 2026, Chen Changcheng wrote:

> This series fixes two issues in the hid-alps driver related to the
> separate input device ("DualPoint Stick", input2) that is allocated in
> alps_input_configured() but not registered in hdev->inputs:
> 
>   Patch 1: input2 is never unregistered on remove.
>            The driver struct has no .remove handler, and input2 is not
>            tracked in hdev->inputs, so the default remove path
>            (hid_hw_stop -> hidinput_disconnect) skips it.  Result: every
>            device removal leaks one struct input_dev.
> 
>   Patch 2: data->input2 is stored before registration, and on
>            registration failure the dangling pointer is read by
>            u1_raw_event() (URBs are already active because
>            alps_input_configured() calls hid_hw_open() before
>            allocating input2) -> use-after-free.
> 
> Both patches apply against linux.git master at 7.2-rc7.

Applied, thank you.

-- 
Jiri Kosina
SUSE Labs