Spurious TCP retransmissions on ack vs kfree_skb reordering

Hi!

We see large (4-8x) increase of what looks like TCP RTOs after rising 
the Tx coalescing above Rx coalescing timeout.

Quick tracing of the events seems to indicate that the data has already
been acked when we enter tcp:tcp_retransmit_skb:

sk_state:      1
icsk_ca_state: 4

bytes_in:      0
bytes_out:   742
bytes_acked: 742

Is this a known condition? Is the recommendation to filter such events
out in tracing infra?