Re: [PATCH v8] tcp: move seq_to_tap update to when frame is queued

[Jon Maloy <jmaloy@redhat.com>]

On 2024-06-04 17:30, Stefano Brivio wrote:
> On Tue,  4 Jun 2024 14:29:08 -0400
> Jon Maloy <jmaloy@redhat.com> wrote:
>
>> +/**
>> + * tcp_revert_seq() - Revert affected conn->seq_to_tap after failed transmission
>> + * @conns:       Array of connection pointers corresponding to queued frames
>> + * @frames:      Two-dimensional array containing queued frames with sub-iovs
>> + * @num_frames:  Number of entries in the two arrays to be compared
>> + */
>> +static void tcp_revert_seq(struct tcp_tap_conn **conns, struct iovec (*frames)[TCP_NUM_IOVS],
>> +			   int num_frames)
>> +{
>> +	int i;
>> +
>> +	for (i = 0; i < num_frames; i++) {
>> +		struct tcp_tap_conn *conn = conns[i];
>> +		struct tcphdr *th = frames[i][TCP_IOV_PAYLOAD].iov_base;
>> +		uint32_t seq = ntohl(th->seq);
>> +
>> +		if (SEQ_LE(conn->seq_to_tap, seq))
>> +			continue;
> See:
>    https://archives.passt.top/passt-dev/Zkr_4LkjDImgFqSi@zatzit
>    https://archives.passt.top/passt-dev/ZlkrBFkxliCCT3st@zatzit
>
> about this if (SEQ_LE) ... continue;
>
My point is, again, that in the next patch in the series I do

static void tcp_revert_seq(struct tcp_tap_conn **conns, struct iovec 
(*frames)[T
                         continue;

                 conn->seq_to_tap = seq;
+               tcp_set_peek_offset(conn->sock, seq - 
conn->seq_ack_from_tap);
         }
  }

So, this still makes sense.

///jon