Date: Fri, 27 Jan 2017 08:10:01 -0800 From: Gleb Smirnoff <glebius@FreeBSD.org> To: Julien Charbon <jch@freebsd.org> Cc: hiren@FreeBSD.org, Jason Eggleston <jeggleston@llnw.com>, jtl@FreeBSD.org, rrs@FreeBSD.org, net@FreeBSD.org Subject: Re: listening sockets as non sockets Message-ID: <20170127161001.GP2611@FreeBSD.org> In-Reply-To: <84f9c348-1e65-1b94-37a6-a4c67d195709@freebsd.org> References: <20170127005251.GM2611@FreeBSD.org> <84f9c348-1e65-1b94-37a6-a4c67d195709@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Julien, On Fri, Jan 27, 2017 at 03:37:57PM +0100, Julien Charbon wrote: J> I might be able to look at potential race conditions related to J> ACCEPT_LOCK and SO_ACCEPTFILTER usage (even if I am more used to J> INP_INFO lock), but I can certainly provide performance numbers and lock J> contention metrics using our setup. J> J> Do you think it is the right time to start performance testing with J> your change? Or it is a bit premature? I don't know of any race or bug in the patch right now, but of course I won't recommend start testing it in production. If you have performance testing environment that is safe (non production) I'd appreciate if you start testing. -- Totus tuus, Glebius.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20170127161001.GP2611>