1. Patchew
  2. linux
  3. View series

[PATCH] net: ag71xx: remove dead code path

Qianqiang Liu posted 1 patch 2 months, 2 weeks ago
There is a newer version of this series
drivers/net/ethernet/atheros/ag71xx.c | 12 +++---------
1 file changed, 3 insertions(+), 9 deletions(-)
[PATCH] net: ag71xx: remove dead code path
Posted by Qianqiang Liu 2 months, 2 weeks ago 
The 'err' is always zero, so the following branch can never be executed:
if (err) {
	ndev->stats.rx_dropped++;
	kfree_skb(skb);
}
Therefore, the 'if' statement can be removed.

Signed-off-by: Qianqiang Liu <qianqiang.liu@163.com>
---
 drivers/net/ethernet/atheros/ag71xx.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
index 96a6189cc31e..5477f3f87e10 100644
--- a/drivers/net/ethernet/atheros/ag71xx.c
+++ b/drivers/net/ethernet/atheros/ag71xx.c
@@ -1616,7 +1616,6 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
 		unsigned int i = ring->curr & ring_mask;
 		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
 		int pktlen;
-		int err = 0;
 
 		if (ag71xx_desc_empty(desc))
 			break;
@@ -1646,14 +1645,9 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
 		skb_reserve(skb, offset);
 		skb_put(skb, pktlen);
 
-		if (err) {
-			ndev->stats.rx_dropped++;
-			kfree_skb(skb);
-		} else {
-			skb->dev = ndev;
-			skb->ip_summed = CHECKSUM_NONE;
-			list_add_tail(&skb->list, &rx_list);
-		}
+		skb->dev = ndev;
+		skb->ip_summed = CHECKSUM_NONE;
+		list_add_tail(&skb->list, &rx_list);
 
 next:
 		ring->buf[i].rx.rx_buf = NULL;
-- 
2.39.2
Re: [PATCH] net: ag71xx: remove dead code path
Posted by Qianqiang Liu 2 months, 2 weeks ago 
On Tue, Sep 10, 2024 at 11:22:54PM +0800, Qianqiang Liu wrote:
> The 'err' is always zero, so the following branch can never be executed:
> if (err) {
> 	ndev->stats.rx_dropped++;
> 	kfree_skb(skb);
> }
> Therefore, the 'if' statement can be removed.
> 
> Signed-off-by: Qianqiang Liu <qianqiang.liu@163.com>
> ---
>  drivers/net/ethernet/atheros/ag71xx.c | 12 +++---------
>  1 file changed, 3 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
> index 96a6189cc31e..5477f3f87e10 100644
> --- a/drivers/net/ethernet/atheros/ag71xx.c
> +++ b/drivers/net/ethernet/atheros/ag71xx.c
> @@ -1616,7 +1616,6 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>  		unsigned int i = ring->curr & ring_mask;
>  		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
>  		int pktlen;
> -		int err = 0;
>  
>  		if (ag71xx_desc_empty(desc))
>  			break;
> @@ -1646,14 +1645,9 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>  		skb_reserve(skb, offset);
>  		skb_put(skb, pktlen);
>  
> -		if (err) {
> -			ndev->stats.rx_dropped++;
> -			kfree_skb(skb);
> -		} else {
> -			skb->dev = ndev;
> -			skb->ip_summed = CHECKSUM_NONE;
> -			list_add_tail(&skb->list, &rx_list);
> -		}
> +		skb->dev = ndev;
> +		skb->ip_summed = CHECKSUM_NONE;
> +		list_add_tail(&skb->list, &rx_list);
>  
>  next:
>  		ring->buf[i].rx.rx_buf = NULL;
> -- 
> 2.39.2

Hi Jakub,

Could you please review this patch?

-- 
Best,
Qianqiang Liu
Re: [PATCH] net: ag71xx: remove dead code path
Posted by Jakub Kicinski 2 months, 2 weeks ago 
On Thu, 12 Sep 2024 23:46:18 +0800 Qianqiang Liu wrote:
> Could you please review this patch?

My preference is to combine the removal with Oleksij's suggestion
of adding the drop/error increment at the right spot.
-- 
pw-bot: cr
Re: [PATCH] net: ag71xx: remove dead code path
Posted by Rosen Penev 2 months, 2 weeks ago 
On Tue, Sep 10, 2024 at 8:30 AM Qianqiang Liu <qianqiang.liu@163.com> wrote:
>
> The 'err' is always zero, so the following branch can never be executed:
> if (err) {
>         ndev->stats.rx_dropped++;
>         kfree_skb(skb);
> }
> Therefore, the 'if' statement can be removed.
>
> Signed-off-by: Qianqiang Liu <qianqiang.liu@163.com>
Reviewed-by: Rosen Penev <rosenp@gmail.com>

err was used downstream in OpenWrt when platform data was used. In the
transition to OF, the function was removed. Which was back in 2019 at
this point.
> ---
>  drivers/net/ethernet/atheros/ag71xx.c | 12 +++---------
>  1 file changed, 3 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
> index 96a6189cc31e..5477f3f87e10 100644
> --- a/drivers/net/ethernet/atheros/ag71xx.c
> +++ b/drivers/net/ethernet/atheros/ag71xx.c
> @@ -1616,7 +1616,6 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>                 unsigned int i = ring->curr & ring_mask;
>                 struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
>                 int pktlen;
> -               int err = 0;
>
>                 if (ag71xx_desc_empty(desc))
>                         break;
> @@ -1646,14 +1645,9 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>                 skb_reserve(skb, offset);
>                 skb_put(skb, pktlen);
>
> -               if (err) {
> -                       ndev->stats.rx_dropped++;
> -                       kfree_skb(skb);
> -               } else {
> -                       skb->dev = ndev;
> -                       skb->ip_summed = CHECKSUM_NONE;
> -                       list_add_tail(&skb->list, &rx_list);
> -               }
> +               skb->dev = ndev;
> +               skb->ip_summed = CHECKSUM_NONE;
> +               list_add_tail(&skb->list, &rx_list);
>
>  next:
>                 ring->buf[i].rx.rx_buf = NULL;
> --
> 2.39.2
>
>
Re: [PATCH] net: ag71xx: remove dead code path
Posted by Andrew Lunn 2 months, 2 weeks ago 
On Tue, Sep 10, 2024 at 11:22:54PM +0800, Qianqiang Liu wrote:
> The 'err' is always zero, so the following branch can never be executed:
> if (err) {
> 	ndev->stats.rx_dropped++;
> 	kfree_skb(skb);
> }
> Therefore, the 'if' statement can be removed.

This code was added by Oleksij Rempel <o.rempel@pengutronix.de>. It is
good to Cc: him, he might have useful comments.

Your changed does look correct, but maybe ret was actually supposed to
be set somewhere? Is there an actual bug hiding here somewhere?

     Andrew

> 
> Signed-off-by: Qianqiang Liu <qianqiang.liu@163.com>
> ---
>  drivers/net/ethernet/atheros/ag71xx.c | 12 +++---------
>  1 file changed, 3 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c
> index 96a6189cc31e..5477f3f87e10 100644
> --- a/drivers/net/ethernet/atheros/ag71xx.c
> +++ b/drivers/net/ethernet/atheros/ag71xx.c
> @@ -1616,7 +1616,6 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>  		unsigned int i = ring->curr & ring_mask;
>  		struct ag71xx_desc *desc = ag71xx_ring_desc(ring, i);
>  		int pktlen;
> -		int err = 0;
>  
>  		if (ag71xx_desc_empty(desc))
>  			break;
> @@ -1646,14 +1645,9 @@ static int ag71xx_rx_packets(struct ag71xx *ag, int limit)
>  		skb_reserve(skb, offset);
>  		skb_put(skb, pktlen);
>  
> -		if (err) {
> -			ndev->stats.rx_dropped++;
> -			kfree_skb(skb);
> -		} else {
> -			skb->dev = ndev;
> -			skb->ip_summed = CHECKSUM_NONE;
> -			list_add_tail(&skb->list, &rx_list);
> -		}
> +		skb->dev = ndev;
> +		skb->ip_summed = CHECKSUM_NONE;
> +		list_add_tail(&skb->list, &rx_list);
>  
>  next:
>  		ring->buf[i].rx.rx_buf = NULL;
> -- 
> 2.39.2
> 
>
Re: [PATCH] net: ag71xx: remove dead code path
Posted by Oleksij Rempel 2 months, 2 weeks ago 
On Tue, Sep 10, 2024 at 05:58:22PM +0200, Andrew Lunn wrote:
> On Tue, Sep 10, 2024 at 11:22:54PM +0800, Qianqiang Liu wrote:
> > The 'err' is always zero, so the following branch can never be executed:
> > if (err) {
> > 	ndev->stats.rx_dropped++;
> > 	kfree_skb(skb);
> > }
> > Therefore, the 'if' statement can be removed.
> 
> This code was added by Oleksij Rempel <o.rempel@pengutronix.de>. It is
> good to Cc: him, he might have useful comments.

Yes, please.

> Your changed does look correct, but maybe ret was actually supposed to
> be set somewhere? Is there an actual bug hiding here somewhere?

Hm, let's see... this issue existed in the openwrt code and I didn't
spotted it by upstreaming.

The only place which may fail in this part is napi_build_skb(), we will
need to count it probably with rx_errors++.

I'm ok with this patch, it can be reworked to use rx_errors++ on
napi_build_skb() instead or this can be done in a separate patch.

If you like to keep this patch as is, here is my:
Reviewed-by: Oleksij Rempel <o.rempel@pengutronix.de>

Thank you!

Regards,
Oleksij
-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Patchew 2.1-devel-b67e4c2

© 2016 - 2024 Red Hat, Inc.