Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 4 Aug 2023 23:01:24 +0200
From:   Simon Horman <horms@kernel.org>
To:     "huangjie.albert" <huangjie.albert@bytedance.com>
Cc:     davem@davemloft.net, edumazet@google.com, kuba@kernel.org,
        pabeni@redhat.com, Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Jesper Dangaard Brouer <hawk@kernel.org>,
        John Fastabend <john.fastabend@gmail.com>,
        =?utf-8?B?QmrDtnJuIFTDtnBlbA==?= <bjorn@kernel.org>,
        Magnus Karlsson <magnus.karlsson@intel.com>,
        Maciej Fijalkowski <maciej.fijalkowski@intel.com>,
        Jonathan Lemon <jonathan.lemon@gmail.com>,
        Pavel Begunkov <asml.silence@gmail.com>,
        Shmulik Ladkani <shmulik.ladkani@gmail.com>,
        Kees Cook <keescook@chromium.org>,
        Richard Gobert <richardbgobert@gmail.com>,
        Yunsheng Lin <linyunsheng@huawei.com>,
        "open list:NETWORKING DRIVERS" <netdev@vger.kernel.org>,
        open list <linux-kernel@vger.kernel.org>,
        "open list:XDP (eXpress Data Path)" <bpf@vger.kernel.org>
Subject: Re: [RFC Optimizing veth xsk performance 06/10] veth: add
 ndo_xsk_wakeup callback for veth
Message-ID: <ZM1nJPEAKLjM1YN/@vergenet.net>
References: <20230803140441.53596-1-huangjie.albert@bytedance.com>
 <20230803140441.53596-7-huangjie.albert@bytedance.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230803140441.53596-7-huangjie.albert@bytedance.com>
Precedence: bulk

On Thu, Aug 03, 2023 at 10:04:32PM +0800, huangjie.albert wrote:
> Signed-off-by: huangjie.albert <huangjie.albert@bytedance.com>
> ---
>  drivers/net/veth.c | 40 ++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 40 insertions(+)
> 
> diff --git a/drivers/net/veth.c b/drivers/net/veth.c
> index 944761807ca4..600225e27e9e 100644
> --- a/drivers/net/veth.c
> +++ b/drivers/net/veth.c
> @@ -1840,6 +1840,45 @@ static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
>  	rcu_read_unlock();
>  }
>  
> +static void veth_xsk_remote_trigger_napi(void *info)
> +{
> +	struct veth_sq *sq = info;
> +
> +	napi_schedule(&sq->xdp_napi);
> +}
> +
> +static int veth_xsk_wakeup(struct net_device *dev, u32 qid, u32 flag)
> +{
> +	struct veth_priv *priv;
> +	struct veth_sq *sq;
> +	u32 last_cpu, cur_cpu;
> +
> +	if (!netif_running(dev))
> +		return -ENETDOWN;
> +
> +	if (qid >= dev->real_num_rx_queues)
> +		return -EINVAL;
> +
> +	priv = netdev_priv(dev);
> +	sq = &priv->sq[qid];
> +
> +	if (napi_if_scheduled_mark_missed(&sq->xdp_napi))
> +		return 0;
> +
> +	last_cpu = sq->xsk.last_cpu;
> +	cur_cpu = get_cpu();
> +
> +	/*  raise a napi */
> +	if (last_cpu == cur_cpu) {
> +		napi_schedule(&sq->xdp_napi);
> +	} else {
> +		smp_call_function_single(last_cpu, veth_xsk_remote_trigger_napi, sq, true);
> +	}

nit: no need for braces in the above.

	if (last_cpu == cur_cpu)
		napi_schedule(&sq->xdp_napi);
	else
		smp_call_function_single(last_cpu, veth_xsk_remote_trigger_napi, sq, true);

...