[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <OS3P286MB229533DAC2CF9CE98C54B428F5D89@OS3P286MB2295.JPNP286.PROD.OUTLOOK.COM>
Date:   Wed, 8 Feb 2023 11:08:44 +0800
From:   Eddy Tao <taoyuan_eddy@...mail.com>
To:     Simon Horman <simon.horman@...igine.com>
Cc:     netdev@...r.kernel.org, Pravin B Shelar <pshelar@....org>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>, dev@...nvswitch.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v1 1/1] net: openvswitch: remove unnecessary vlan
 init in key_extract
Hi, Simon:
     Thanks for looking into this. by looking into the compiled 
instructions, i am convinced the patch is logically correct but not help 
in performance, and it adds complexity,
Below function is actually optimized to one movq instruction with no 
special -O settings on my centos8, which seems good enough for me
static void clear_vlan(struct sw_flow_key *key)
{
     key->eth.vlan.tci = 0;
     key->eth.vlan.tpid = 0;
     key->eth.cvlan.tci = 0;
     key->eth.cvlan.tpid = 0;
}
optimized to movq   $0x0,0x158(%rbx)
I am withdrawing this patch
About your comment below, since it is after the parse_vlan handling, the 
code looks fine and does not affect the logic of the patch
 >>I think you missed the following case further down: >>>>if 
(unlikely(key->eth.type == htons(0))) Have a great day
eddy
Powered by blists - more mailing lists
 
