Alexey Kuznetsov wrote: > It is workaround for a bug, noticed in win2000/XP?, when > overoptimized TCP receiver in http client delivers data only > when it sees a PSH. Normally, TCP may send the whole multigigabyte > stream of data not doing even single PSH because it always has more > data to transmit. > > So, forced_push() means that we make at least one PSH per window, > not depending on any other things sort of "amount of queued data". If the problem is win2000/xp not waking up its http client, why one PSH per window? Why not one PSH per N kilobytes or one per second? (One per second would be better for web browers over dialup, wouldn't it?) If you don't do once per window, does it mean the TCP session locks up due to the client window filling completely without the client app seeing any data? Can you point me to a web page with more information about this bug? I searched but didn't find anything. I would like to know more about it. Thanks. -- Jamie - : 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
