Date: Sat, 24 Aug 2013 18:47:08 +0100 From: "Robert N. M. Watson" <rwatson@FreeBSD.org> To: Alfred Perlstein <bright@mu.org> Cc: FreeBSD Net <freebsd-net@freebsd.org>, Adrian Chadd <adrian@freebsd.org>, freebsd-current <freebsd-current@freebsd.org>, "Alexander V. Chernikov" <melifaro@ipfw.ru> Subject: Re: [rfc] migrate lagg to an rmlock Message-ID: <A4BA90F3-AB8A-41A6-B931-20AF81F903E0@FreeBSD.org> In-Reply-To: <5218E108.6090901@mu.org> References: <CAJ-Vmo=VKVDEmmPrTbob6Ft%2B7FWypodNoL36Og=7p_CXBSfktg@mail.gmail.com> <5218AA36.1080807@ipfw.ru> <alpine.BSF.2.00.1308241511400.92711@fledge.watson.org> <5218E108.6090901@mu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 24 Aug 2013, at 17:36, Alfred Perlstein wrote: >> We should distinguish "lock contention" from "line contention". When = acquiring a rwlock on multiple CPUs concurrently, the cache lines used = to implement the lock are contended, as they must bounce between caches = via the cache coherence protocol, also referred to as "contention". In = the if_lagg code, I assume that the read-only acquire of the rwlock (and = perhaps now rmlock) is for data stability rather than mutual exclusion = -- e.g., to allow processing to completion against a stable version of = the lagg configuration. As such, indeed, there should be no lock = contention unless a configuration update takes place, and any line = contention is a property of the locking primitive rather than data = model. >>=20 >> There are a number of other places in the kernel where migration to = an rmlock makes sense -- however, some care must be taken for four = reasons: (1) while read locks don't experience line contention, write = locking becomes observably e.g., rmlocks might not be suitable for = tcbinfo; (2) rmlocks, unlike rwlocks, more expensive so is not suitable = for all rwlock line contention spots -- implement reader priority = propagation, so you must reason about; and (3) historically, rmlocks = have not fully implemented WITNESS so you may get less good debugging = output. if_lagg is a nice place to use rmlocks, as reconfigurations are = very rare, and it's really all about long-term data stability. >=20 > Robert, what do you think about a quick swap of the ifnet structures = to counter before 10.x? Could you be more specific about the proposal you're making? Robert=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?A4BA90F3-AB8A-41A6-B931-20AF81F903E0>