[PATCH v3 0/2] sched: Add cond_resched_rcu_lock() helper

[PATCH v3 0/2] sched: Add cond_resched_rcu_lock() helper

[Simon Horman]

Add a helper that for use in loops which read data protected by RCU and
may have a large number of iterations.  Such an example is dumping the list
of connections known to IPVS: ip_vs_conn_array() and ip_vs_conn_seq_next().

This series also updates the two ip_vs functions mentioned above
to use the helper.

Changes since v1 and v2 noted in the changelog of each patch.

Simon Horman (2):
  sched: Add cond_resched_rcu_lock() helper
  ipvs: Use cond_resched_rcu_lock() helper when dumping connections

 include/linux/sched.h           | 11 +++++++++++
 net/netfilter/ipvs/ip_vs_conn.c |  6 ++----
 2 files changed, 13 insertions(+), 4 deletions(-)

-- 
1.8.2.1

[PATCH v3 1/2] sched: Add cond_resched_rcu_lock() helper

[Simon Horman]

This is intended for use in loops which read data protected by RCU and may
have a large number of iterations.  Such an example is dumping the list of
connections known to IPVS: ip_vs_conn_array() and ip_vs_conn_seq_next().

The call to cond_resched() is guarded by #ifndef CONFIG_PREEMPT_RCU as in
the case of CONFIG_PREEMPT_RCU _rcu_read_unlock() will check to see if the
RCU core needs to be informed, so there is no need to invoke cond_resched()
in that case. Thanks to Paul E. McKenney for explaining this.

cond_resched_rcu_lock() suggested by Eric Dumazet.
ifndef guard suggested by Paul E. McKenney and Julian Anastasov.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Julian Anastasov <ja@ssi.bg>
Acked-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Simon Horman <horms@verge.net.au>

---

v3
* Remove if (need_resched()), it is no longer needed with
  the guard added in v2.

v2
* Guard the call to cond_resched() with #ifndef CONFIG_PREEMPT_RCU
---
 include/linux/sched.h | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index e692a02..79cfe6d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2787,3 +2787,12 @@ static inline unsigned long rlimit_max(unsigned int limit)
 }
 
 #endif
+
+static void inline cond_resched_rcu_lock(void)
+{
+	rcu_read_unlock();
+#ifndef CONFIG_PREEMPT_RCU
+	cond_resched();
+#endif
+	rcu_read_lock();
+}
-- 
1.8.2.1

[PATCH v3 2/2] ipvs: Use cond_resched_rcu_lock() helper when dumping connections

[Simon Horman]

This avoids the situation where a dump of a large number of connections
may prevent scheduling for a long time while also avoiding excessive
calls to rcu_read_unlock() and rcu_read_lock().

Note that in the case of !CONFIG_PREEMPT_RCU this will
add a call to cond_resched().

Compile tested only.

Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Julian Anastasov <ja@ssi.bg>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
---
 net/netfilter/ipvs/ip_vs_conn.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c
index a083bda..42a7b33 100644
--- a/net/netfilter/ipvs/ip_vs_conn.c
+++ b/net/netfilter/ipvs/ip_vs_conn.c
@@ -975,8 +975,7 @@ static void *ip_vs_conn_array(struct seq_file *seq, loff_t pos)
 				return cp;
 			}
 		}
-		rcu_read_unlock();
-		rcu_read_lock();
+		cond_resched_rcu_lock();
 	}
 
 	return NULL;
@@ -1015,8 +1014,7 @@ static void *ip_vs_conn_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 			iter->l = &ip_vs_conn_tab[idx];
 			return cp;
 		}
-		rcu_read_unlock();
-		rcu_read_lock();
+		cond_resched_rcu_lock();
 	}
 	iter->l = NULL;
 	return NULL;
-- 
1.8.2.1