Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1695138
| From | "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH tip/core/rcu 09/15] rcu: use idle versions of swait to make idle-hack clear |
| Date | 2017-07-24 23:50 +0200 |
| Message-ID | <u6ToK-2fn-19@gated-at.bofh.it> (permalink) |
| References | <u6ToJ-2fn-3@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
From: "Luis R. Rodriguez" <mcgrof@kernel.org>
These RCU waits were set to use interruptible waits to avoid the kthreads
contributing to system load average, even though they are not interruptible
as they are spawned from a kthread. Use the new TASK_IDLE swaits which makes
our goal clear, and removes confusion about these paths possibly being
interruptible -- they are not.
When the system is idle the RCU grace-period kthread will spend all its time
blocked inside the swait_event_interruptible(). If the interruptible() was
not used, then this kthread would contribute to the load average. This means
that an idle system would have a load average of 2 (or 3 if PREEMPT=y),
rather than the load average of 0 that almost fifty years of UNIX has
conditioned sysadmins to expect.
The same argument applies to swait_event_interruptible_timeout() use. The
RCU grace-period kthread spends its time blocked inside this call while
waiting for grace periods to complete. In particular, if there was only one
busy CPU, but that CPU was frequently invoking call_rcu(), then the RCU
grace-period kthread would spend almost all its time blocked inside the
swait_event_interruptible_timeout(). This would mean that the load average
would be 2 rather than the expected 1 for the single busy CPU.
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
kernel/rcu/tree.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index ab4c2cea208d..9a90d9b1dc04 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -2052,8 +2052,8 @@ static bool rcu_gp_init(struct rcu_state *rsp)
}
/*
- * Helper function for wait_event_interruptible_timeout() wakeup
- * at force-quiescent-state time.
+ * Helper function for swait_event_idle() wakeup at force-quiescent-state
+ * time.
*/
static bool rcu_gp_fqs_check_wake(struct rcu_state *rsp, int *gfp)
{
@@ -2191,9 +2191,8 @@ static int __noreturn rcu_gp_kthread(void *arg)
READ_ONCE(rsp->gpnum),
TPS("reqwait"));
rsp->gp_state = RCU_GP_WAIT_GPS;
- swait_event_interruptible(rsp->gp_wq,
- READ_ONCE(rsp->gp_flags) &
- RCU_GP_FLAG_INIT);
+ swait_event_idle(rsp->gp_wq, READ_ONCE(rsp->gp_flags) &
+ RCU_GP_FLAG_INIT);
rsp->gp_state = RCU_GP_DONE_GPS;
/* Locking provides needed memory barrier. */
if (rcu_gp_init(rsp))
@@ -2224,7 +2223,7 @@ static int __noreturn rcu_gp_kthread(void *arg)
READ_ONCE(rsp->gpnum),
TPS("fqswait"));
rsp->gp_state = RCU_GP_WAIT_FQS;
- ret = swait_event_interruptible_timeout(rsp->gp_wq,
+ ret = swait_event_idle_timeout(rsp->gp_wq,
rcu_gp_fqs_check_wake(rsp, &gf), j);
rsp->gp_state = RCU_GP_DOING_FQS;
/* Locking provides needed memory barriers. */
--
2.5.2
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH tip/core/rcu 0/15] General fixes "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 04/15] rcu: Create reasonable API for do_exit() TASKS_RCU processing "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 09/15] rcu: use idle versions of swait to make idle-hack clear "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups Steven Rostedt <rostedt@goodmis.org> - 2017-07-25 20:20 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-25 21:20 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups Steven Rostedt <rostedt@goodmis.org> - 2017-07-26 00:20 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-26 02:10 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups Steven Rostedt <rostedt@goodmis.org> - 2017-07-26 23:20 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-26 23:50 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups Steven Rostedt <rostedt@goodmis.org> - 2017-07-27 01:10 +0200
Re: [PATCH tip/core/rcu 02/15] rcu: Use timer as backstop for NOCB deferred wakeups "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-27 19:40 +0200
[PATCH tip/core/rcu 10/15] rcu: Add TPS() protection for _rcu_barrier_trace strings "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
Re: [PATCH tip/core/rcu 10/15] rcu: Add TPS() protection for _rcu_barrier_trace strings Steven Rostedt <rostedt@goodmis.org> - 2017-07-28 03:50 +0200
[PATCH tip/core/rcu 14/15] rcu: Add warning to rcu_idle_enter() for irqs enabled "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 01/15] sched,rcu: Make cond_resched() provide RCU quiescent state "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 08/15] swait: add idle variants which don't contribute to load average "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 15/15] rcu: Remove exports from rcu_idle_exit() and rcu_idle_enter() "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 03/15] rcu: Drive TASKS_RCU directly off of PREEMPT "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
Re: [PATCH tip/core/rcu 03/15] rcu: Drive TASKS_RCU directly off of PREEMPT Steven Rostedt <rostedt@goodmis.org> - 2017-07-25 20:20 +0200
Re: [PATCH tip/core/rcu 03/15] rcu: Drive TASKS_RCU directly off of PREEMPT "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-25 21:20 +0200
[PATCH tip/core/rcu 12/15] rcu: Add assertions verifying blocked-tasks list "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
[PATCH tip/core/rcu 05/15] rcu: Add TPS() to event-traced strings "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
Re: [PATCH tip/core/rcu 05/15] rcu: Add TPS() to event-traced strings Steven Rostedt <rostedt@goodmis.org> - 2017-07-28 03:40 +0200
[PATCH tip/core/rcu 06/15] rcu: Move rcu.h to new trivial-function style "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2017-07-24 23:50 +0200
csiph-web