[Patch net-next] net: add doc for ip_early_demux sysctl

[Patch net-next] net: add doc for ip_early_demux sysctl

[Cong Wang]

From: Cong Wang <amwang@redhat.com>

commit 6648bd7e0e62c0c8c03b (ipv4: Add sysctl knob to control
early socket demux) introduced such sysctl, but forgot to add
doc into Documentation/networking/ip-sysctl.txt. This patch adds it.

Basically I grab the doc from the description of commit 41063e9dd11956f2d285
(ipv4: Early TCP socket demux.) and the above commit.

Cc: Eric Dumazet <edumazet@google.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>

---
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 398d0fb..79d0483 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -685,6 +685,15 @@ ip_dynaddr - BOOLEAN
 	occurs.
 	Default: 0
 
+ip_early_demux - BOOLEAN
+	Optimize input packet processing down to one demux for
+	certain kinds of local sockets.  Currently we only do this
+	for established TCP sockets.
+
+	It may add an additional cost to routing that reduces overall
+	throughput, in such case you should disable it.
+	Default: 1
+
 icmp_echo_ignore_all - BOOLEAN
 	If set non-zero, then the kernel will ignore all ICMP ECHO
 	requests sent to it.

Re: [Patch net-next] net: add doc for ip_early_demux sysctl

[David Miller]

From: Cong Wang <amwang@redhat.com>
Date: Thu,  6 Jun 2013 13:10:23 +0800

> +	It may add an additional cost to routing that reduces overall
> +	throughput, in such case you should disable it.

I would state this as "additional cost for pure routing workloads" to
emphasize the fact that the potential bad case is when we are not
delivering to a local socket.

The way you have it worded, it makes it sound like route lookups (in
general) will perform more poorly.  This is simply not the case.