netdev - Re: [PATCH net-next 2/4] net: use skb_get

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1395365518.9114.116.camel@edumazet-glaptop2.roam.corp.google.com>
Date:	Thu, 20 Mar 2014 18:31:58 -0700
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	Tom Herbert <therbert@...gle.com>
Cc:	davem@...emloft.net, netdev@...r.kernel.org
Subject: Re: [PATCH net-next 2/4] net: use skb_get_hash in tx queue selection

On Thu, 2014-03-20 at 17:09 -0700, Tom Herbert wrote:
> __skb_get_hash tries to get the skb->hash from a connected socket
> associated with an skb. __skb_tx_hash and get_xps_queue just call
> skb_get_hash().
> 
> Signed-off-by: Tom Herbert <therbert@...gle.com>
> ---
>  include/linux/netdevice.h |  2 +-
>  include/linux/skbuff.h    |  2 +-
>  net/core/flow_dissector.c | 22 +++++++++-------------
>  3 files changed, 11 insertions(+), 15 deletions(-)
> 
> diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> index 4b6d12c..a1fa2e4 100644
> --- a/include/linux/netdevice.h
> +++ b/include/linux/netdevice.h
> @@ -2481,7 +2481,7 @@ static inline int netif_set_xps_queue(struct net_device *dev,
>   * as a distribution range limit for the returned value.
>   */
>  static inline u16 skb_tx_hash(const struct net_device *dev,
> -			      const struct sk_buff *skb)
> +			      struct sk_buff *skb)
>  {
>  	return __skb_tx_hash(dev, skb, dev->real_num_tx_queues);
>  }
> diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
> index aa2c22c..f392b74 100644
> --- a/include/linux/skbuff.h
> +++ b/include/linux/skbuff.h
> @@ -2863,7 +2863,7 @@ static inline bool skb_rx_queue_recorded(const struct sk_buff *skb)
>  	return skb->queue_mapping != 0;
>  }
>  
> -u16 __skb_tx_hash(const struct net_device *dev, const struct sk_buff *skb,
> +u16 __skb_tx_hash(const struct net_device *dev, struct sk_buff *skb,
>  		  unsigned int num_tx_queues);
>  
>  static inline struct sec_path *skb_sec_path(struct sk_buff *skb)
> diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
> index 107ed12..432d35a 100644
> --- a/net/core/flow_dissector.c
> +++ b/net/core/flow_dissector.c
> @@ -212,6 +212,12 @@ void __skb_get_hash(struct sk_buff *skb)
>  	struct flow_keys keys;
>  	u32 hash;
>  
> +	if (skb->sk && skb->sk->sk_hash) {
> +		skb->hash = skb->sk->sk_hash;
> +		skb->l4_hash = 1;
> +		return;
> +	}

Wouldn't it be faster to simply set skb->hash in providers (TCP, UDP),
to that we do not add this test here ?

After packet has been a while in a qdisc, access to skb->sk_hash will
likely trigger a cache line miss.



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html