Hello! > Good to know that, but then, does that mean that we don't do an explicit > sender-side SWS check. We do. It is just combined to Nagle according to advice of rfc1122. I guess, you are bothered with missing link to PUSHED ((2) in rfc 1121 terms) and to max_window (3). (1) and (2) are Nagle algorithm, PUSHED is not related directly to PSH in existing TCP implementations, it is each end of write() or an explicit uncork on CORKed connections. Note, that while long sendpage() transfers this just never happens. The only implicit thing is (3). It is hidden inside calculation of mss: our mss is min(rfc_mss, max_window*Fs), so (3) is guaranteed as soon as Nagle requirements are held. Alexey - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
