Re: [PATCH 4/6] tcp: Repair socket queues

[Pavel Emelyanov <xemul@parallels.com>]

On 05/02/2012 03:11 PM, Eric Dumazet wrote:
> On Thu, 2012-04-19 at 17:41 +0400, Pavel Emelyanov wrote:
>> Reading queues under repair mode is done with recvmsg call.
>> The queue-under-repair set by TCP_REPAIR_QUEUE option is used
>> to determine which queue should be read. Thus both send and
>> receive queue can be read with this.
>>
>> Caller must pass the MSG_PEEK flag.
>>
>> Writing to queues is done with sendmsg call and yet again --
>> the repair-queue option can be used to push data into the
>> receive queue.
>>
>> When putting an skb into receive queue a zero tcp header is
>> appented to its head to address the tcp_hdr(skb)->syn and
>> the ->fin checks by the (after repair) tcp_recvmsg. These
>> flags flags are both set to zero and that's why.
>>
>> The fin cannot be met in the queue while reading the source
>> socket, since the repair only works for closed/established
>> sockets and queueing fin packet always changes its state.
>>
>> The syn in the queue denotes that the respective skb's seq
>> is "off-by-one" as compared to the actual payload lenght. Thus,
>> at the rcv queue refill we can just drop this flag and set the
>> skb's sequences to precice values.
>>
>> When the repair mode is turned off, the write queue seqs are
>> updated so that the whole queue is considered to be 'already sent,
>> waiting for ACKs' (write_seq = snd_nxt <= snd_una). From the
>> protocol POV the send queue looks like it was sent, but the data
>> between the write_seq and snd_nxt is lost in the network.
>>
>> This helps to avoid another sockoption for setting the snd_nxt
>> sequence. Leaving the whole queue in a 'not yet sent' state (as
>> it will be after sendmsg-s) will not allow to receive any acks
>> from the peer since the ack_seq will be after the snd_nxt. Thus
>> even the ack for the window probe will be dropped and the
>> connection will be 'locked' with the zero peer window.
>>
>> Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
>> ---
>>  net/ipv4/tcp.c        |   89 +++++++++++++++++++++++++++++++++++++++++++++++--
>>  net/ipv4/tcp_output.c |    1 +
>>  2 files changed, 87 insertions(+), 3 deletions(-)
>>
>> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
>> index e38d6f2..47e2f49 100644
>> --- a/net/ipv4/tcp.c
>> +++ b/net/ipv4/tcp.c
>> @@ -912,6 +912,39 @@ static inline int select_size(const struct sock *sk, bool sg)
>>  	return tmp;
>>  }
>>  
>> +static int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size)
>> +{
>> +	struct sk_buff *skb;
>> +	struct tcp_skb_cb *cb;
>> +	struct tcphdr *th;
>> +
>> +	skb = alloc_skb(size + sizeof(*th), sk->sk_allocation);
> 
> I am not sure any check is performed on 'size' ?

No, no checks here.

> A caller might trigger OOM or wrap bug.

Well, yes, but this ability is given to CAP_SYS_NET_ADMIN users only.
Do you think it's nonetheless worth accounting this allocation into
the socket's rmem?

Thanks,
Pavel