netdev - Re: [PATCH net-next v3 2/4] netdevsim: forward skbs from one connected port to another

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]

Date: Sat, 16 Dec 2023 10:22:31 +0100
From: Jiri Pirko <jiri@...nulli.us>
To: David Wei <dw@...idwei.uk>
Cc: Jakub Kicinski <kuba@...nel.org>, Sabrina Dubroca <sd@...asysnail.net>,
	netdev@...r.kernel.org, "David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>, Paolo Abeni <pabeni@...hat.com>
Subject: Re: [PATCH net-next v3 2/4] netdevsim: forward skbs from one
 connected port to another

Fri, Dec 15, 2023 at 07:31:42PM CET, dw@...idwei.uk wrote:
>On 2023-12-15 02:45, Jiri Pirko wrote:
>> Thu, Dec 14, 2023 at 10:24:41PM CET, dw@...idwei.uk wrote:
>>> Forward skbs sent from one netdevsim port to its connected netdevsim
>>> port using dev_forward_skb, in a spirit similar to veth.
>> 
>> Perhaps better to write "dev_forward_skb()" to make obvious you talk
>> about function.
>
>Sorry, it's a bad habit at this point :)
>
>> 
>> 
>>>
>>> Signed-off-by: David Wei <dw@...idwei.uk>
>>> ---
>>> drivers/net/netdevsim/netdev.c | 23 ++++++++++++++++++-----
>>> 1 file changed, 18 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c
>>> index e290c54b0e70..c5f53b1dbdcc 100644
>>> --- a/drivers/net/netdevsim/netdev.c
>>> +++ b/drivers/net/netdevsim/netdev.c
>>> @@ -29,19 +29,33 @@
>>> static netdev_tx_t nsim_start_xmit(struct sk_buff *skb, struct net_device *dev)
>>> {
>>> 	struct netdevsim *ns = netdev_priv(dev);
>>> +	struct netdevsim *peer_ns;
>>> +	int ret = NETDEV_TX_OK;
>>>
>>> +	rcu_read_lock();
>> 
>> Why do you need to be in rcu read locked section here?
>
>So the RCU protected pointer `peer` does not change during the critical
>section. Veth does something similar in its xmit() for its peer.

RCU does not work like this. Please check out the documentation.


>
>> 
>> 
>>> 	if (!nsim_ipsec_tx(ns, skb))
>>> -		goto out;
>>> +		goto err;
>> 
>> Not sure why you need to rename the label. Why "out" is not okay?
>> 
>>>
>>> 	u64_stats_update_begin(&ns->syncp);
>>> 	ns->tx_packets++;
>>> 	ns->tx_bytes += skb->len;
>>> 	u64_stats_update_end(&ns->syncp);
>>>
>>> -out:
>>> -	dev_kfree_skb(skb);
>>> +	peer_ns = rcu_dereference(ns->peer);
>>> +	if (!peer_ns)
>>> +		goto err;
>> 
>> This is definitelly not an error path, "err" label name is misleading.
>
>That's fair, I can change it back. Lots has changed since my original
>intentions.
>
>> 
>> 
>>> +
>>> +	skb_tx_timestamp(skb);
>>> +	if (unlikely(dev_forward_skb(peer_ns->netdev, skb) == NET_RX_DROP))
>>> +		ret = NET_XMIT_DROP;
>> 
>> Hmm, can't you track dropped packets in ns->tx_dropped and expose in
>> nsim_get_stats64() ?
>
>I can add this.
>
>> 
>> 
>>>
>>> -	return NETDEV_TX_OK;
>>> +	rcu_read_unlock();
>>> +	return ret;
>>> +
>>> +err:
>>> +	rcu_read_unlock();
>>> +	dev_kfree_skb(skb);
>>> +	return ret;
>>> }
>>>
>>> static void nsim_set_rx_mode(struct net_device *dev)
>>> @@ -302,7 +316,6 @@ static void nsim_setup(struct net_device *dev)
>>> 	eth_hw_addr_random(dev);
>>>
>>> 	dev->tx_queue_len = 0;
>>> -	dev->flags |= IFF_NOARP;
>>> 	dev->flags &= ~IFF_MULTICAST;
>>> 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE |
>>> 			   IFF_NO_QUEUE;
>>> -- 
>>> 2.39.3
>>>