[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1386558263.30495.311.camel@edumazet-glaptop2.roam.corp.google.com>
Date: Sun, 08 Dec 2013 19:04:23 -0800
From: Eric Dumazet <eric.dumazet@...il.com>
To: Jerry Chu <hkchu@...gle.com>
Cc: Eric Dumazet <edumazet@...gle.com>,
Herbert Xu <herbert@...dor.apana.org.au>,
Or Gerlitz <ogerlitz@...lanox.com>,
David Miller <davem@...emloft.net>,
netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next] net-gro: Prepare GRO stack for the upcoming
tunneling support
On Mon, 2013-12-09 at 10:48 +0800, Jerry Chu wrote:
> Not sure why ingress qdisc - it seems that qdisc_pkt_len_init() is only called
> by the egress path (__dev_xmit_skb())
> .
Yeah, Qdisc was meant to be used for egress.
Then to get qdisc for ingress, we use IFB to reinject packets through
the qdisc layer.
Sample of script :
ETH=eth0
IFB=ifb0
RTT_HALF=50ms
REORDER=1500us
modprobe ifb
ip link set dev $IFB up
tc qdisc add dev $ETH ingress 2>/dev/null
tc filter add dev $ETH parent ffff: \
protocol ip u32 match u32 0 0 flowid 1:1 action mirred egress \
redirect dev $IFB
tc qdisc del dev $IFB root 2>/dev/null
tc qdisc add dev $IFB root netem limit 100000 delay $RTT_HALF $REORDER
> Strange why is it the business of *netif_receive_skb*() to set the
> transport header
> before IP has a chance to parse the IP hdr?
>
Maybe you should use git blame ;)
--
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