[V4,net-next] tun/macvtap: use consume_skb() instead of kfree_skb() when needed
diff mbox series

Message ID 1417423995-4765-1-git-send-email-jasowang@redhat.com
State New, archived
Headers show
Series
  • [V4,net-next] tun/macvtap: use consume_skb() instead of kfree_skb() when needed
Related show

Commit Message

Jason Wang Dec. 1, 2014, 8:53 a.m. UTC
To be more friendly with drop monitor, we should only call kfree_skb() when
the packets were dropped and use consume_skb() in other cases.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
Changes from V3:
- rebase to net-next.git
Changes from V2:
- use unlikely() when necessary
Changes from V1:
- check the return value of tun/macvtap_put_user()
---
 drivers/net/macvtap.c | 5 ++++-
 drivers/net/tun.c     | 5 ++++-
 2 files changed, 8 insertions(+), 2 deletions(-)

Comments

Michael S. Tsirkin Dec. 1, 2014, 9:30 a.m. UTC | #1
On Mon, Dec 01, 2014 at 04:53:15PM +0800, Jason Wang wrote:
> To be more friendly with drop monitor, we should only call kfree_skb() when
> the packets were dropped and use consume_skb() in other cases.
> 
> Cc: Eric Dumazet <eric.dumazet@gmail.com>
> Signed-off-by: Jason Wang <jasowang@redhat.com>

Acked-by: Michael S. Tsirkin <mst@redhat.com>

> ---
> Changes from V3:
> - rebase to net-next.git
> Changes from V2:
> - use unlikely() when necessary
> Changes from V1:
> - check the return value of tun/macvtap_put_user()
> ---
>  drivers/net/macvtap.c | 5 ++++-
>  drivers/net/tun.c     | 5 ++++-
>  2 files changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
> index 22b4cf2..ba1e5db 100644
> --- a/drivers/net/macvtap.c
> +++ b/drivers/net/macvtap.c
> @@ -859,7 +859,10 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q,
>  	}
>  	if (skb) {
>  		ret = macvtap_put_user(q, skb, to);
> -		kfree_skb(skb);
> +		if (unlikely(ret < 0))
> +			kfree_skb(skb);
> +		else
> +			consume_skb(skb);
>  	}
>  	if (!noblock)
>  		finish_wait(sk_sleep(&q->sk), &wait);
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index 6d44da1..9c58286 100644
> --- a/drivers/net/tun.c
> +++ b/drivers/net/tun.c
> @@ -1362,7 +1362,10 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
>  		return 0;
>  
>  	ret = tun_put_user(tun, tfile, skb, to);
> -	kfree_skb(skb);
> +	if (unlikely(ret < 0))
> +		kfree_skb(skb);
> +	else
> +		consume_skb(skb);
>  
>  	return ret;
>  }
> -- 
> 1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
David Miller Dec. 6, 2014, 5:45 a.m. UTC | #2
From: Jason Wang <jasowang@redhat.com>
Date: Mon,  1 Dec 2014 16:53:15 +0800

> To be more friendly with drop monitor, we should only call kfree_skb() when
> the packets were dropped and use consume_skb() in other cases.
> 
> Cc: Eric Dumazet <eric.dumazet@gmail.com>
> Signed-off-by: Jason Wang <jasowang@redhat.com>

Applied, thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Patch
diff mbox series

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 22b4cf2..ba1e5db 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -859,7 +859,10 @@  static ssize_t macvtap_do_read(struct macvtap_queue *q,
 	}
 	if (skb) {
 		ret = macvtap_put_user(q, skb, to);
-		kfree_skb(skb);
+		if (unlikely(ret < 0))
+			kfree_skb(skb);
+		else
+			consume_skb(skb);
 	}
 	if (!noblock)
 		finish_wait(sk_sleep(&q->sk), &wait);
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 6d44da1..9c58286 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1362,7 +1362,10 @@  static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
 		return 0;
 
 	ret = tun_put_user(tun, tfile, skb, to);
-	kfree_skb(skb);
+	if (unlikely(ret < 0))
+		kfree_skb(skb);
+	else
+		consume_skb(skb);
 
 	return ret;
 }