Re: [PATCH] usb: misc: usbio: Handle negative error code from usb_control_msg()

[Greg KH]

On Sat, Oct 03, 2026 at 05:28:38PM +0300, Serhat Kumral wrote:
> The sizeof(*cpkt) is unsigned, so comparing "ret < sizeof(*cpkt)" promotes
> "ret" to unsigned and negative error codes from usb_control_msg() are
> not caught. Cast the sizeof to int so the error handling works.
> 
> Fixes: 121a0f839dbb ("usb: misc: Add Intel USBIO bridge driver")
> Assisted-by: LLM
> Signed-off-by: Serhat Kumral <serhatkumral1@xxxxxxxxx>
> ---
> found by smatch:
> - drivers/usb/misc/usbio.c:187 usbio_ctrl_msg() warn: error code type promoted to positive: 'ret'
> 
>  drivers/usb/misc/usbio.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/usb/misc/usbio.c b/drivers/usb/misc/usbio.c
> index fe093e7760d5..3f73f7c3d68a 100644
> --- a/drivers/usb/misc/usbio.c
> +++ b/drivers/usb/misc/usbio.c
> @@ -184,7 +184,7 @@ static int usbio_ctrl_msg(struct usbio_device *usbio, u8 type, u8 cmd,
>  	dev_dbg(usbio->dev, "control in %d hdr %*phN data %*phN\n", ret,
>  		(int)sizeof(*cpkt), cpkt, (int)cpkt->len, cpkt->data);
>  
> -	if (ret < sizeof(*cpkt)) {
> +	if (ret < (int)sizeof(*cpkt)) {

Ick, really?  There's no compiler check for this type of thing?

This whole function needs to be rewritten to use the "modern" control
message functions.  I gave it a go once, but it quickly got messy as
this is an abused function and it really should be massivly cleaned up.

Do you have the hardware for this to test changes?

thanks,

greg k-h