From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH net-next] tun: replace get_cpu_ptr with this_cpu_ptr when bh disabled Date: Mon, 10 Dec 2018 13:04:47 -0500 Message-ID: <20181210130403-mutt-send-email-mst@kernel.org> References: <20181210050618.13300-1-bhole_prashant_q7@lab.ntt.co.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "David S . Miller" , Jason Wang , netdev@vger.kernel.org, Toshiaki Makita To: Prashant Bhole Return-path: Received: from mx1.redhat.com ([209.132.183.28]:21627 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726602AbeLJSEs (ORCPT ); Mon, 10 Dec 2018 13:04:48 -0500 Content-Disposition: inline In-Reply-To: <20181210050618.13300-1-bhole_prashant_q7@lab.ntt.co.jp> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Dec 10, 2018 at 02:06:18PM +0900, Prashant Bhole wrote: > tun_xdp_one() runs with local bh disabled. So there is no need to > disable preemption by calling get_cpu_ptr while updating stats. This > patch replaces the use of get_cpu_ptr() with this_cpu_ptr() as a > micro-optimization. Also removes related put_cpu_ptr call. > > Signed-off-by: Prashant Bhole Seems to make sense but how about a comment here explaining by it's safe? > --- > drivers/net/tun.c | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/drivers/net/tun.c b/drivers/net/tun.c > index 6760b86547df..abb750a9c9ca 100644 > --- a/drivers/net/tun.c > +++ b/drivers/net/tun.c > @@ -2494,12 +2494,11 @@ static int tun_xdp_one(struct tun_struct *tun, > skb_record_rx_queue(skb, tfile->queue_index); > netif_receive_skb(skb); > > - stats = get_cpu_ptr(tun->pcpu_stats); > + stats = this_cpu_ptr(tun->pcpu_stats); > u64_stats_update_begin(&stats->syncp); > stats->rx_packets++; > stats->rx_bytes += skb->len; > u64_stats_update_end(&stats->syncp); > - put_cpu_ptr(stats); > > if (rxhash) > tun_flow_update(tun, rxhash, tfile); > -- > 2.17.2 >