From owner-freebsd-arch@freebsd.org Mon Mar 13 12:33:17 2017 Return-Path: Delivered-To: freebsd-arch@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 7A2C8D05883 for ; Mon, 13 Mar 2017 12:33:17 +0000 (UTC) (envelope-from avg@FreeBSD.org) Received: from citapm.icyb.net.ua (citapm.icyb.net.ua [212.40.38.140]) by mx1.freebsd.org (Postfix) with ESMTP id CAF151A51 for ; Mon, 13 Mar 2017 12:33:16 +0000 (UTC) (envelope-from avg@FreeBSD.org) Received: from porto.starpoint.kiev.ua (porto-e.starpoint.kiev.ua [212.40.38.100]) by citapm.icyb.net.ua (8.8.8p3/ICyb-2.3exp) with ESMTP id OAA29703; Mon, 13 Mar 2017 14:33:07 +0200 (EET) (envelope-from avg@FreeBSD.org) Received: from localhost ([127.0.0.1]) by porto.starpoint.kiev.ua with esmtp (Exim 4.34 (FreeBSD)) id 1cnP9v-000NYd-KH; Mon, 13 Mar 2017 14:33:07 +0200 Subject: Re: UMA_ZONE_CACHESPREAD and uma_zsecond_add To: Andrew Gallatin , freebsd-arch@FreeBSD.org, Jeff Roberson References: <76a47d9a-da39-75f4-5794-24724d0befc7@FreeBSD.org> <035f48ea-8722-f697-cc71-cb8ee772e062@cs.duke.edu> From: Andriy Gapon Message-ID: Date: Mon, 13 Mar 2017 14:32:11 +0200 User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:45.0) Gecko/20100101 Thunderbird/45.7.1 MIME-Version: 1.0 In-Reply-To: <035f48ea-8722-f697-cc71-cb8ee772e062@cs.duke.edu> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Mar 2017 12:33:17 -0000 On 12/03/2017 20:55, Andrew Gallatin wrote: > On 03/08/2017 10:56, Andriy Gapon wrote: >> >> First, the history of UMA_ZONE_CACHESPREAD and uma_zsecond_add(): >> https://lists.freebsd.org/pipermail/freebsd-arch/2008-December/008800.html >> >> Now, more than 8 years after those features were introduced, we still don't have >> a single in-tree consumer for them. >> Does anyone use those features out of tree? >> Does anyone still have plans to make use of them? >> Will anyone get sad if those features get garbage collected? > > This is something that I keep getting suggestions to try at Netflix > on our 100G boxes. From the description, it really seems like > it might help us to have a few data types allocated like this. > > Can you give me, say, one month to look into this before axing > it? Sure! That's exactly why I wanted to ask everyone first. I am not itching to remove that code :) -- Andriy Gapon