netdev - Re: [PATCH 1/1] net: ipv4: refine the logic for determining daddr in ip_queue

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20130115101229.GA27222@gmail.com>
Date:	Tue, 15 Jan 2013 10:12:30 +0000
From:	Cong Ding <dinggnu@...il.com>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	Harvey Yang <harvey.huawei.yang@...il.com>,
	"David S. Miller" <davem@...emloft.net>,
	Alexey Kuznetsov <kuznet@....inr.ac.ru>,
	James Morris <jmorris@...ei.org>,
	Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/1] net: ipv4: refine the logic for determining daddr in
 ip_queue_xmit.

On Mon, Jan 14, 2013 at 08:57:09PM -0800, Eric Dumazet wrote:
> On Tue, 2013-01-15 at 12:47 +0800, Harvey Yang wrote:
> > The destination address daddr is faddr if source route option is set, otherwise it is inet_daddr. So use if-else to assign the value.
> > 
> > Signed-off-by: Harvey Yang <harvey.huawei.yang@...il.com>
> > ---
> >  net/ipv4/ip_output.c |    4 ++--
> >  1 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
> > index 3e98ed2..22b738c 100644
> > --- a/net/ipv4/ip_output.c
> > +++ b/net/ipv4/ip_output.c
> > @@ -349,10 +349,10 @@ int ip_queue_xmit(struct sk_buff *skb, struct flowi *fl)
> >  		__be32 daddr;
> >  
> >  		/* Use correct destination address if we have options. */
> > -		daddr = inet->inet_daddr;
> >  		if (inet_opt && inet_opt->opt.srr)
> >  			daddr = inet_opt->opt.faddr;
> > -
> > +		else
> > +			daddr = inet->inet_daddr;
> >  		/* If this fails, retransmit mechanism of transport layer will
> >  		 * keep trying until route appears or the connection times
> >  		 * itself out.
> 
> Current code is faster, srr is not often used.
If you turn on O2 option in GCC, both versions behave the same (generate same
binary code). If srr is not often used, why not use unlikely() macro?
- cong

--
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