Date: Sun, 22 Jun 2014 14:34:57 +0200 From: Mateusz Guzik <mjguzik@gmail.com> To: Mindaugas Rasiukevicius <rmind@netbsd.org> Cc: Andrey Zonov <zont@FreeBSD.org>, freebsd-arch@FreeBSD.org Subject: Re: PoC: passive serialization Message-ID: <20140622123457.GA20525@dft-labs.eu> In-Reply-To: <20140621231853.394A914A2D0@mail.netbsd.org> References: <539FEBC1.5030501@FreeBSD.org> <20140621231853.394A914A2D0@mail.netbsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jun 22, 2014 at 12:18:52AM +0100, Mindaugas Rasiukevicius wrote: > Just a note on passive serialization in NetBSD: there is a lot of space for > optimisations, simplifications or improvements to that code, but it was a > deliberate choice to avoid them. The goal was to carefully implement the > logic described in the expired patent (or at least attempt to be as close as > our interpretation skills allow us to be). Any deviation from that logic > increases the risk of falling under some other technique, primarily RCU, > covered by other patent. > Cursory look suggests passive serialization got very limited use in NetBSD, I was also unable to find any benchmarks. I do understand and appreciate your cautious approach, but it also limits the potential. Do you have any materials which would show gains of using the concept as implemented in your code? Right now it is unclear if the implementation limited to this patent can yield benefits which would justify added complexity. Thanks, -- Mateusz Guzik <mjguzik gmail.com>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20140622123457.GA20525>