From: Eric Dumazet <edumazet@google.com>
Date: Mon, 14 Mar 2016 22:40:00 +0000 (-0700)
Subject: net: diag: add a scheduling point in inet_diag_dump_icsk()
X-Git-Tag: v4.6-rc1~91^2~36
X-Git-Url: https://asedeno.scripts.mit.edu/gitweb/?a=commitdiff_plain;h=acffb584cda7069b0c2c83045503ccd07516a891;p=linux.git

net: diag: add a scheduling point in inet_diag_dump_icsk()

On loaded TCP servers, looking at millions of sockets can hold
cpu for many seconds, if the lookup condition is very narrow.

(eg : ss dst 1.2.3.4 )

Better add a cond_resched() to allow other processes to access
the cpu.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---

diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 50c0d96b8441..5fdb02f5598e 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -879,6 +879,7 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb,
 		}
 
 		spin_unlock_bh(lock);
+		cond_resched();
 	}
 
 done: