From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by dpdk.org (Postfix) with ESMTP id 52CD22B82 for ; Mon, 23 Apr 2018 11:03:26 +0200 (CEST) Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 4502F8DC3E; Mon, 23 Apr 2018 09:03:25 +0000 (UTC) Received: from [10.36.112.58] (ovpn-112-58.ams2.redhat.com [10.36.112.58]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 53D0D1102E2B; Mon, 23 Apr 2018 09:03:24 +0000 (UTC) To: Yuanhan Liu Cc: Jianfeng Tan , dpdk stable References: <20180422150949.17523-1-yliu@fridaylinux.org> <20180422150949.17523-36-yliu@fridaylinux.org> From: Maxime Coquelin Message-ID: <6d78c0cd-08b1-2b57-ad96-b0b7a0d7063b@redhat.com> Date: Mon, 23 Apr 2018 11:03:22 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 MIME-Version: 1.0 In-Reply-To: <20180422150949.17523-36-yliu@fridaylinux.org> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.78 on 10.11.54.3 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.2]); Mon, 23 Apr 2018 09:03:25 +0000 (UTC) X-Greylist: inspected by milter-greylist-4.5.16 (mx1.redhat.com [10.11.55.2]); Mon, 23 Apr 2018 09:03:25 +0000 (UTC) for IP:'10.11.54.3' DOMAIN:'int-mx03.intmail.prod.int.rdu2.redhat.com' HELO:'smtp.corp.redhat.com' FROM:'maxime.coquelin@redhat.com' RCPT:'' Subject: Re: [dpdk-stable] patch 'vhost: avoid concurrency when logging dirty pages' has been queued to LTS release 17.11.2 X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 23 Apr 2018 09:03:27 -0000 Hi Yuanhan, On 04/22/2018 05:09 PM, Yuanhan Liu wrote: > Hi, > > FYI, your patch has been queued to LTS release 17.11.2 > > Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet. > It will be pushed if I get no objections before 04/29/18. So please > shout if anyone has objections. Please don't pick this patch, it cause a regression with multiqueue, I posted a revert of the patch on list last week. I'm working on an alternative solution, but while it is ready, better to keep performance penalty from initial patch. Thanks, Maxime > Thanks. > > --yliu > > --- > From acdf634b3533a5ccb1528ee546ac9dcfb57b8a36 Mon Sep 17 00:00:00 2001 > From: Maxime Coquelin > Date: Wed, 21 Mar 2018 16:44:13 +0100 > Subject: [PATCH] vhost: avoid concurrency when logging dirty pages > > [ upstream commit 394313fff39d0f994325c47f7eab39daf5dc9e11 ] > > This patch aims at fixing a migration performance regression > faced since atomic operation is used to log pages as dirty when > doing live migration. > > Instead of setting a single bit by doing an atomic read-modify-write > operation to log a page as dirty, this patch write 0xFF to the > corresponding byte, and so logs 8 page as dirty. > > The advantage is that it avoids concurrent atomic operations by > multiple PMD threads, the drawback is that some clean pages are > marked as dirty and so are transferred twice. > > Fixes: 897f13a1f726 ("vhost: make page logging atomic") > > Signed-off-by: Maxime Coquelin > Reviewed-by: Jianfeng Tan > --- > lib/librte_vhost/vhost.h | 12 ++++-------- > 1 file changed, 4 insertions(+), 8 deletions(-) > > diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h > index c8f2a8176..5c488fa57 100644 > --- a/lib/librte_vhost/vhost.h > +++ b/lib/librte_vhost/vhost.h > @@ -261,18 +261,14 @@ struct virtio_net { > #define VHOST_LOG_PAGE 4096 > > /* > - * Atomically set a bit in memory. > + * Mark all pages belonging to the same dirty log bitmap byte > + * as dirty. The goal is to avoid concurrency between different > + * threads doing atomic read-modify-writes on the same byte. > */ > static __rte_always_inline void > -vhost_set_bit(unsigned int nr, volatile uint8_t *addr) > -{ > - __sync_fetch_and_or_8(addr, (1U << nr)); > -} > - > -static __rte_always_inline void > vhost_log_page(uint8_t *log_base, uint64_t page) > { > - vhost_set_bit(page % 8, &log_base[page / 8]); > + log_base[page / 8] = 0xff; > } > > static __rte_always_inline void >