From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH net-next 3/4] inet: Add a 2nd listener hashtable
 (port+addr)
Date: Fri, 01 Dec 2017 09:26:46 -0800
Message-ID: <1512149206.19682.37.camel@gmail.com>
References: <20171130232327.561108-1-kafai@fb.com>
         <20171130232327.561108-4-kafai@fb.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8bit
Cc: "David S . Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Kernel Team <kernel-team@fb.com>
To: Martin KaFai Lau <kafai@fb.com>, netdev <netdev@vger.kernel.org>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-pl0-f65.google.com ([209.85.160.65]:40321 "EHLO
        mail-pl0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751157AbdLAR0s (ORCPT
        <rfc822;netdev@vger.kernel.org>); Fri, 1 Dec 2017 12:26:48 -0500
Received: by mail-pl0-f65.google.com with SMTP id 1so6636138pla.7
        for <netdev@vger.kernel.org>; Fri, 01 Dec 2017 09:26:47 -0800 (PST)
In-Reply-To: <20171130232327.561108-4-kafai@fb.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Thu, 2017-11-30 at 15:23 -0800, Martin KaFai Lau wrote:
> The current listener hashtable is hashed by port only.
> When a process is listening at many IP addresses with the same port
> (e.g.
> [IP1]:443, [IP2]:443... [IPN]:443), the inet[6]_lookup_listener()
> performance is degraded to a link list.  It is prone to syn attack.
> 
> 

...

> diff --git a/include/net/inet_hashtables.h
> b/include/net/inet_hashtables.h
> index 4cce516c41ac..ebce55d694e7 100644
> --- a/include/net/inet_hashtables.h
> +++ b/include/net/inet_hashtables.h
> @@ -152,8 +152,19 @@ struct inet_hashinfo {
>  	 */
>  	struct inet_listen_hashbucket	listening_hash[INET_LHT
> ABLE_SIZE]
>  					____cacheline_aligned_in_smp
> ;
> +	struct inet_listen_hashbucket	*lhash2;
> +	unsigned int			lhash2_mask;
>  };
> 

Please move these two fields before listening_hash[], to avoid adding
another cache line to "struct inet_hashinfo"

( We have a 16-byte hole there, so lets use it ;) )