Date: Thu, 23 Feb 2006 17:44:05 +0100 (CET) From: "Arne H. Juul" <arnej@pvv.ntnu.no> To: Kurt Miller <kurt@intricatesoftware.com> Cc: Nate Williams <nate@yogotech.com>, freebsd-java@freebsd.org Subject: Re: SO_REUSEADDR should not also mean SO_REUSEPORT Message-ID: <Pine.LNX.4.62.0602231735500.8287@decibel.pvv.ntnu.no> In-Reply-To: <200602211011.39387.kurt@intricatesoftware.com> References: <43F4F22F.1060402@europe.yahoo-inc.com> <17399.20883.741021.688682@caddis.yogotech.com> <Pine.LNX.4.62.0602182107440.31913@decibel.pvv.ntnu.no> <200602211011.39387.kurt@intricatesoftware.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 21 Feb 2006, Kurt Miller wrote: > Thanks for the explanation and also to Nate for the Multicast > history. I've looked into this a bit more over the weekend and > found that the network stack promotes SO_REUSEADDR to include > SO_REUSEPORT for multicast addresses, so I believe that case is > covered already. I ran the network jck's on the 1.5 jvm with your > patch and found that SO_REUSEPORT is still needed to pass the jck's > but for datagram sockets only. I guess we have to emulate Solaris to pass these, even if I think the Solaris behaviour is a bit bad. > Could you try this patch and test it with the program you referred > to above? I'll see if I can try this tomorrow; it *should* work for the applications I'm worried about since those are all using TCP sockets only. Please update the comment to say why the "bug-compatibility" is there. Something like: /* * On Solaris, SO_REUSEADDR will allow multiple datagram * sockets to bind to the same port. The network jck tests * for this "feature", so we need to emulate it by turning on * SO_REUSEPORT as well for that combination. */ - Arne H. J.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.62.0602231735500.8287>