Date: Fri, 15 Sep 2017 13:50:49 +0200
From: Michal Hocko <mhocko@kernel.org>
To: Kemi Wang <kemi.wang@intel.com>
Cc: "Luis R . Rodriguez" <mcgrof@kernel.org>,
        Kees Cook <keescook@chromium.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Jonathan Corbet <corbet@lwn.net>,
        Mel Gorman <mgorman@techsingularity.net>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Christopher Lameter <cl@linux.com>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Vlastimil Babka <vbabka@suse.cz>,
        Hillf Danton <hillf.zj@alibaba-inc.com>,
        Dave <dave.hansen@linux.intel.com>, Tim Chen <tim.c.chen@intel.com>,
        Andi Kleen <andi.kleen@intel.com>,
        Jesper Dangaard Brouer <brouer@redhat.com>,
        Ying Huang <ying.huang@intel.com>, Aaron Lu <aaron.lu@intel.com>,
        Proc sysctl <linux-fsdevel@vger.kernel.org>,
        Linux MM <linux-mm@kvack.org>,
        Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/3] mm: Handle numa statistics distinctively based-on
 different VM stats modes
Message-ID: <20170915115049.vqthfawg3y4r6ogh@dhcp22.suse.cz>
References: <1505467406-9945-1-git-send-email-kemi.wang@intel.com>
 <1505467406-9945-3-git-send-email-kemi.wang@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1505467406-9945-3-git-send-email-kemi.wang@intel.com>
User-Agent: NeoMutt/20170609 (1.8.3)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 696
Lines: 25

On Fri 15-09-17 17:23:25, Kemi Wang wrote:
[...]
> @@ -2743,6 +2745,17 @@ static inline void zone_statistics(struct zone *preferred_zone, struct zone *z)
>  #ifdef CONFIG_NUMA
>  	enum numa_stat_item local_stat = NUMA_LOCAL;
>  
> +	/*
> +	 * skip zone_statistics() if vmstat is a coarse mode or zone statistics
> +	 * is inactive in auto vmstat mode
> +	 */
> +
> +	if (vmstat_mode) {
> +		if (vmstat_mode == VMSTAT_COARSE_MODE)
> +			return;
> +	} else if (disable_zone_statistics)
> +		return;
> +
>  	if (z->node != numa_node_id())
>  		local_stat = NUMA_OTHER;

A jump label could make this completely out of the way for the case
where every single cycle matters.
-- 
Michal Hocko
SUSE Labs