[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1294340191.3074.96.camel@edumazet-laptop>
Date: Thu, 06 Jan 2011 19:56:31 +0100
From: Eric Dumazet <eric.dumazet@...il.com>
To: John Fastabend <john.r.fastabend@...el.com>
Cc: Ben Hutchings <bhutchings@...arflare.com>,
"davem@...emloft.net" <davem@...emloft.net>,
"jarkao2@...il.com" <jarkao2@...il.com>,
"hadi@...erus.ca" <hadi@...erus.ca>,
"shemminger@...tta.com" <shemminger@...tta.com>,
"tgraf@...radead.org" <tgraf@...radead.org>,
"nhorman@...driver.com" <nhorman@...driver.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [net-next-2.6 PATCH v5 1/2] net: implement mechanism for HW
based QOS
Le jeudi 06 janvier 2011 à 10:45 -0800, John Fastabend a écrit :
> It should still be 64 bytes.
>
> struct netdev_tc_txq {
> u16 count;
> u16 offset;
> };
>
> struct netdev_tc_txq (4 octets) and TC_MAX_QUEUE = 16, so 4*16 = 64
>
> + struct netdev_tc_txq tc_to_txq[TC_MAX_QUEUE]; => 64 octets
> + u8 prio_tc_map[TC_BITMASK+1]; => 16 octets
>
> All together 80 bytes this is certainly in the 512 unused bytes.
Good ;)
Considering a single TCP socket is larger than a net_device, I think its
OK to not worry too much ;)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists