Re: [PATCH] usb: hub: Fix unhandled return value of usb_autopm_get_interface()

[Alan Stern]

On Tue, 25 Feb 2020, Eugeniu Rosca wrote:

> Address below Coverity complaint (Feb 25, 2020, 8:06 AM CET):
> 
> *** CID 1458999:  Error handling issues  (CHECKED_RETURN)
> /drivers/usb/core/hub.c: 1869 in hub_probe()
> 1863
> 1864            if (id->driver_info & HUB_QUIRK_CHECK_PORT_AUTOSUSPEND)
> 1865                    hub->quirk_check_port_auto_suspend = 1;
> 1866
> 1867            if (id->driver_info & HUB_QUIRK_DISABLE_AUTOSUSPEND) {
> 1868                    hub->quirk_disable_autosuspend = 1;
>  >>>     CID 1458999:  Error handling issues  (CHECKED_RETURN)
>  >>>     Calling "usb_autopm_get_interface" without checking return value (as is done elsewhere 97 out of 111 times).
> 1869                    usb_autopm_get_interface(intf);
> 1870            }
> 1871
> 1872            if (hub_configure(hub, &desc->endpoint[0].desc) >= 0)
> 1873                    return 0;
> 1874
> 
> Fixes: 1208f9e1d758c9 ("USB: hub: Fix the broken detection of USB3 device in SMSC hub")
> Cc: Hardik Gajjar <hgajjar@xxxxxxxxxxxxxx>
> Cc: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
> Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> Reported-by: scan-admin@xxxxxxxxxxxx
> Signed-off-by: Eugeniu Rosca <erosca@xxxxxxxxxxxxxx>
> ---
>  drivers/usb/core/hub.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> index 1d212f82c69b..ff04ca28970d 100644
> --- a/drivers/usb/core/hub.c
> +++ b/drivers/usb/core/hub.c
> @@ -1865,8 +1865,12 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id)
>  		hub->quirk_check_port_auto_suspend = 1;
>  
>  	if (id->driver_info & HUB_QUIRK_DISABLE_AUTOSUSPEND) {
> -		hub->quirk_disable_autosuspend = 1;
> -		usb_autopm_get_interface(intf);
> +		int r = usb_autopm_get_interface(intf);
> +
> +		if (!r)
> +			hub->quirk_disable_autosuspend = 1;
> +		else
> +			dev_dbg(&intf->dev, "disable autosuspend err=%d\n", r);
>  	}
>  
>  	if (hub_configure(hub, &desc->endpoint[0].desc) >= 0)

This change is not necessary, because the resume operation cannot fail
at this point (interfaces are always powered-up during probe).  A
better solution would be to call usb_autpm_get_interface_no_resume()
instead.

On the other hand, the other places that call
usb_autopm_get_interface() without checking should be audited.  Some of
them almost certainly need to be fixed.

Alan Stern