Date: Mon, 12 Oct 2009 14:24:23 +0200 From: Bernd Walter <ticso@cicely7.cicely.de> To: Ivan Voras <ivoras@freebsd.org> Cc: freebsd-hackers@freebsd.org Subject: Re: "global" TCP_NODELAY? Message-ID: <20091012122423.GO48396@cicely7.cicely.de> In-Reply-To: <hav4ma$prm$1@ger.gmane.org> References: <hav3qp$n7n$1@ger.gmane.org> <20091012113619.GN48396@cicely7.cicely.de> <hav4ma$prm$1@ger.gmane.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Oct 12, 2009 at 01:42:08PM +0200, Ivan Voras wrote: > Bernd Walter wrote: > >On Mon, Oct 12, 2009 at 01:27:28PM +0200, Ivan Voras wrote: > >>I'm trying to work around some extreme brain damageness in PHP (yes, it > >>sucks) which doesn't have a way to set TCP_NODELAY on stream sockets so > >>I'm wondering what are my other options? Is there a way to set > >>TCP_NODELAY system-wide? > > > >net.inet.tcp.delayed_ack > >net.inet.tcp.delacktime > > > >Depending on your application it may be sufficient to just reduce the > >time. > > Really? Doesn't TCP_NODELAY (Nagle's algorithm) work on buffers to be > sent rather than on ACKs on received data? Good point. TCP_NODELAY disables both to my knowledge. And setting delacktime to 1 helped me in one buffer case. But I'm not sure anymore - a non delayed ack might piggypack payload much sooner of course. -- B.Walter <bernd@bwct.de> http://www.bwct.de Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20091012122423.GO48396>