2021-12-21 02:20:42

by Yang Li

[permalink] [raw]
Subject: [PATCH -next] usb: early: Fix an error code in xdbc_init()

When xdbc_trace("hardware not used anymore\n") is performed,
ret should be assigned -ENODEV to indicate this.

Clean up smatch warning:
drivers/usb/early/xhci-dbc.c:972 xdbc_init() warn: missing error code
'ret'

Reported-by: Abaci Robot <[email protected]>
Signed-off-by: Yang Li <[email protected]>
---
drivers/usb/early/xhci-dbc.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/drivers/usb/early/xhci-dbc.c b/drivers/usb/early/xhci-dbc.c
index 4502108069cd..fcb05a8948fa 100644
--- a/drivers/usb/early/xhci-dbc.c
+++ b/drivers/usb/early/xhci-dbc.c
@@ -969,6 +969,7 @@ static int __init xdbc_init(void)
if (early_xdbc_console.index == -1 ||
(early_xdbc_console.flags & CON_BOOT)) {
xdbc_trace("hardware not used anymore\n");
+ ret = -ENODEV;
goto free_and_quit;
}

--
2.20.1.7.g153144c



2021-12-21 06:37:37

by Greg Kroah-Hartman

[permalink] [raw]
Subject: Re: [PATCH -next] usb: early: Fix an error code in xdbc_init()

On Tue, Dec 21, 2021 at 10:20:36AM +0800, Yang Li wrote:
> When xdbc_trace("hardware not used anymore\n") is performed,
> ret should be assigned -ENODEV to indicate this.
>
> Clean up smatch warning:
> drivers/usb/early/xhci-dbc.c:972 xdbc_init() warn: missing error code
> 'ret'
>
> Reported-by: Abaci Robot <[email protected]>
> Signed-off-by: Yang Li <[email protected]>
> ---
> drivers/usb/early/xhci-dbc.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/usb/early/xhci-dbc.c b/drivers/usb/early/xhci-dbc.c
> index 4502108069cd..fcb05a8948fa 100644
> --- a/drivers/usb/early/xhci-dbc.c
> +++ b/drivers/usb/early/xhci-dbc.c
> @@ -969,6 +969,7 @@ static int __init xdbc_init(void)
> if (early_xdbc_console.index == -1 ||
> (early_xdbc_console.flags & CON_BOOT)) {
> xdbc_trace("hardware not used anymore\n");
> + ret = -ENODEV;
> goto free_and_quit;
> }

How did you test this?

I think you just broke this, the function should be returning 0 here,
like it currently is. Are you SURE smatch is correct?

I'll only accept this if you can prove it is tested. Do you have this
hardware?

thanks,

greg k-h