Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1514053 > unrolled thread

[PATCH v2] USB hub_probe: remove ugly goto-into-compound-statement

Back to article view | Back to linux.kernel

Contents

  [PATCH v2] USB hub_probe: remove ugly goto-into-compound-statement Eugene Korenevsky <ekorenevsky@gmail.com> - 2016-11-02 19:10 +0100
    Re: [PATCH v2] USB hub_probe: remove ugly  goto-into-compound-statement Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-11-07 10:10 +0100
      Re: [PATCH v2] USB hub_probe: remove ugly  goto-into-compound-statement Eugene Korenevsky <ekorenevsky@gmail.com> - 2016-11-07 19:50 +0100

#1514053 — [PATCH v2] USB hub_probe: remove ugly goto-into-compound-statement

Rework smelling code (goto inside compound statement). Perhaps this is legacy.
Anyway such code is not appropriate for Linux kernel.

Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
---
 drivers/usb/core/hub.c | 24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index cbb1467..4081672 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1802,23 +1802,21 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id)
 
 	/* Some hubs have a subclass of 1, which AFAICT according to the */
 	/*  specs is not defined, but it works */
-	if ((desc->desc.bInterfaceSubClass != 0) &&
-	    (desc->desc.bInterfaceSubClass != 1)) {
-descriptor_error:
+
+	/* Reject in following cases:
+	 * - Interface subclass is not 0 or 1
+	 * - Multiple endpoints
+	 * - Not an interrupt in endpoint
+	 */
+	endpoint = &desc->endpoint[0].desc;
+	if ((desc->desc.bInterfaceSubClass != 0 &&
+	     desc->desc.bInterfaceSubClass != 1) ||
+	    desc->desc.bNumEndpoints != 1 ||
+	    !usb_endpoint_is_int_in(endpoint)) {
 		dev_err(&intf->dev, "bad descriptor, ignoring hub\n");
 		return -EIO;
 	}
 
-	/* Multiple endpoints? What kind of mutant ninja-hub is this? */
-	if (desc->desc.bNumEndpoints != 1)
-		goto descriptor_error;
-
-	endpoint = &desc->endpoint[0].desc;
-
-	/* If it's not an interrupt in endpoint, we'd better punt! */
-	if (!usb_endpoint_is_int_in(endpoint))
-		goto descriptor_error;
-
 	/* We found a hub */
 	dev_info(&intf->dev, "USB hub found\n");
 
-- 
2.10.2

[toc] | [next] | [standalone]

#1515919 — Re: [PATCH v2] USB hub_probe: remove ugly goto-into-compound-statement

On Wed, Nov 02, 2016 at 08:59:44PM +0300, Eugene Korenevsky wrote:
> Rework smelling code (goto inside compound statement). Perhaps this is legacy.
> Anyway such code is not appropriate for Linux kernel.
> 
> Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
> ---
>  drivers/usb/core/hub.c | 24 +++++++++++-------------
>  1 file changed, 11 insertions(+), 13 deletions(-)

What changed from v1?



> 
> diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> index cbb1467..4081672 100644
> --- a/drivers/usb/core/hub.c
> +++ b/drivers/usb/core/hub.c
> @@ -1802,23 +1802,21 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id)
>  
>  	/* Some hubs have a subclass of 1, which AFAICT according to the */
>  	/*  specs is not defined, but it works */
> -	if ((desc->desc.bInterfaceSubClass != 0) &&
> -	    (desc->desc.bInterfaceSubClass != 1)) {
> -descriptor_error:
> +
> +	/* Reject in following cases:
> +	 * - Interface subclass is not 0 or 1
> +	 * - Multiple endpoints
> +	 * - Not an interrupt in endpoint
> +	 */
> +	endpoint = &desc->endpoint[0].desc;
> +	if ((desc->desc.bInterfaceSubClass != 0 &&
> +	     desc->desc.bInterfaceSubClass != 1) ||
> +	    desc->desc.bNumEndpoints != 1 ||
> +	    !usb_endpoint_is_int_in(endpoint)) {
>  		dev_err(&intf->dev, "bad descriptor, ignoring hub\n");
>  		return -EIO;
>  	}
>  
> -	/* Multiple endpoints? What kind of mutant ninja-hub is this? */
> -	if (desc->desc.bNumEndpoints != 1)
> -		goto descriptor_error;
> -
> -	endpoint = &desc->endpoint[0].desc;
> -
> -	/* If it's not an interrupt in endpoint, we'd better punt! */
> -	if (!usb_endpoint_is_int_in(endpoint))
> -		goto descriptor_error;
> -

As "horrible" as the original code might be, it's much easier to read
and follow, which is the key thing here, right?  What's so bad about a
goto backwards?

thanks,

greg k-h

[toc] | [prev] | [next] | [standalone]

#1516464 — Re: [PATCH v2] USB hub_probe: remove ugly goto-into-compound-statement

On Mon, Nov 07, 2016 at 10:09:17AM +0100, Greg Kroah-Hartman wrote:

> > Rework smelling code (goto inside compound statement). Perhaps this is legacy.
> > Anyway such code is not appropriate for Linux kernel.
> > 
> > Signed-off-by: Eugene Korenevsky <ekorenevsky@gmail.com>
> > ---
> >  drivers/usb/core/hub.c | 24 +++++++++++-------------
> >  1 file changed, 11 insertions(+), 13 deletions(-)
> 
> What changed from v1?

Fixed faults: missed 'Signed-off-by', spaces instead of tab.


> > diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
> > index cbb1467..4081672 100644
> > --- a/drivers/usb/core/hub.c
> > +++ b/drivers/usb/core/hub.c
> > @@ -1802,23 +1802,21 @@ static int hub_probe(struct usb_interface *intf, const struct usb_device_id *id)
> >  
> >  	/* Some hubs have a subclass of 1, which AFAICT according to the */
> >  	/*  specs is not defined, but it works */
> > -	if ((desc->desc.bInterfaceSubClass != 0) &&
> > -	    (desc->desc.bInterfaceSubClass != 1)) {
> > -descriptor_error:
> > +
> > +	/* Reject in following cases:
> > +	 * - Interface subclass is not 0 or 1
> > +	 * - Multiple endpoints
> > +	 * - Not an interrupt in endpoint
> > +	 */
> > +	endpoint = &desc->endpoint[0].desc;
> > +	if ((desc->desc.bInterfaceSubClass != 0 &&
> > +	     desc->desc.bInterfaceSubClass != 1) ||
> > +	    desc->desc.bNumEndpoints != 1 ||
> > +	    !usb_endpoint_is_int_in(endpoint)) {
> >  		dev_err(&intf->dev, "bad descriptor, ignoring hub\n");
> >  		return -EIO;
> >  	}
> >  
> > -	/* Multiple endpoints? What kind of mutant ninja-hub is this? */
> > -	if (desc->desc.bNumEndpoints != 1)
> > -		goto descriptor_error;
> > -
> > -	endpoint = &desc->endpoint[0].desc;
> > -
> > -	/* If it's not an interrupt in endpoint, we'd better punt! */
> > -	if (!usb_endpoint_is_int_in(endpoint))
> > -		goto descriptor_error;
> > -
> 
> As "horrible" as the original code might be, it's much easier to read
> and follow, which is the key thing here, right?  What's so bad about a
> goto backwards?

OK, this patch is still not perfect. But jumping *back* into
*compound* *statement* hurts reader's eyes. Really. Maybe it's better to
extract this code to static function, compiler will inline it. See v3
patchset.

-- 
Eugene

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web