On Thu, 2016-08-25 at 11:06 +0200, Florian Westphal wrote: > So I gave this a try and it does avoid this tcp_request_sock increase, > but I feel that getting boot_time_rnd is too easy.
Fair enough, I didn't think very hard about it. > > I tried a few other ideas but nothing satisfying/simpler came out of it > (e.g. i tried to also hash the isn but that gets scaled w. current > clock so it doesn't work). > > Are you more concerned wrt. complexity or the reqsk increase? > No, a reqsk increase is really fine. I guess I was simply worrying your work would make my future submission more tricky. Here at Google we have been using usec resolution in TCP timestamps for a while for all our DC traffic, and we have to upstream this at some point. It would be nice if the randomization was optional, because having usec timestamps with a common base (ie no per flow random base) helps a lot to understand the host delays at transmit time, and receive time. I will review your patch more in depth today, thanks.
