From owner-freebsd-current@freebsd.org Mon Aug 1 11:25:09 2016 Return-Path: Delivered-To: freebsd-current@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id A8117BA322D for ; Mon, 1 Aug 2016 11:25:09 +0000 (UTC) (envelope-from slw@zxy.spb.ru) Received: from zxy.spb.ru (zxy.spb.ru [195.70.199.98]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 6EB571A4B; Mon, 1 Aug 2016 11:25:09 +0000 (UTC) (envelope-from slw@zxy.spb.ru) Received: from slw by zxy.spb.ru with local (Exim 4.86 (FreeBSD)) (envelope-from ) id 1bUBLG-000PfX-Ai; Mon, 01 Aug 2016 14:25:06 +0300 Date: Mon, 1 Aug 2016 14:25:06 +0300 From: Slawa Olhovchenkov To: Andriy Gapon Cc: Mateusz Guzik , freebsd-current@FreeBSD.org Subject: Re: [PATCH] randomized delay in locking primitives, take 2 Message-ID: <20160801112506.GB22212@zxy.spb.ru> References: <20160731095706.GB9408@dft-labs.eu> <32aa8f7f-8096-740c-13e3-d51e0cb57c23@FreeBSD.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <32aa8f7f-8096-740c-13e3-d51e0cb57c23@FreeBSD.org> User-Agent: Mutt/1.5.24 (2015-08-30) X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: slw@zxy.spb.ru X-SA-Exim-Scanned: No (on zxy.spb.ru); SAEximRunCond expanded to false X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 01 Aug 2016 11:25:09 -0000 On Mon, Aug 01, 2016 at 02:16:59PM +0300, Andriy Gapon wrote: > > Mateusz, > > just out of curiosity, have you tried to explore alternative spinlock > implementations like a ticket lock? It would be interesting to see if > there are any improvements to be gained there. Effective ticket lock implementation as I see limited to 256 (255?) concurent locks. And anyway may be need randomize delay (QPI bus on x86 have very limited performance).