From e6753f23d961d601dbae50a2fc2a3975c9715b14 Mon Sep 17 00:00:00 2001 From: "Joel Fernandes (Google)" Date: Mon, 30 Jul 2018 15:24:22 -0700 Subject: tracepoint: Make rcuidle tracepoint callers use SRCU In recent tests with IRQ on/off tracepoints, a large performance overhead ~10% is noticed when running hackbench. This is root caused to calls to rcu_irq_enter_irqson and rcu_irq_exit_irqson from the tracepoint code. Following a long discussion on the list [1] about this, we concluded that srcu is a better alternative for use during rcu idle. Although it does involve extra barriers, its lighter than the sched-rcu version which has to do additional RCU calls to notify RCU idle about entry into RCU sections. In this patch, we change the underlying implementation of the trace_*_rcuidle API to use SRCU. This has shown to improve performance alot for the high frequency irq enable/disable tracepoints. Test: Tested idle and preempt/irq tracepoints. Here are some performance numbers: With a run of the following 30 times on a single core x86 Qemu instance with 1GB memory: hackbench -g 4 -f 2 -l 3000 Completion times in seconds. CONFIG_PROVE_LOCKING=y. No patches (without this series) Mean: 3.048 Median: 3.025 Std Dev: 0.064 With Lockdep using irq tracepoints with RCU implementation: Mean: 3.451 (-11.66 %) Median: 3.447 (-12.22%) Std Dev: 0.049 With Lockdep using irq tracepoints with SRCU implementation (this series): Mean: 3.020 (I would consider the improvement against the "without this series" case as just noise). Median: 3.013 Std Dev: 0.033 [1] https://patchwork.kernel.org/patch/10344297/ [remove rcu_read_lock_sched_notrace as its the equivalent of preempt_disable_notrace and is unnecessary to call in tracepoint code] Link: http://lkml.kernel.org/r/20180730222423.196630-3-joel@joelfernandes.org Cleaned-up-by: Peter Zijlstra Acked-by: Peter Zijlstra Reviewed-by: Mathieu Desnoyers Signed-off-by: Joel Fernandes (Google) [ Simplified WARN_ON_ONCE() ] Signed-off-by: Steven Rostedt (VMware) --- kernel/tracepoint.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'kernel/tracepoint.c') diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 6dc6356c3327..955148d91b74 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -31,6 +31,9 @@ extern struct tracepoint * const __start___tracepoints_ptrs[]; extern struct tracepoint * const __stop___tracepoints_ptrs[]; +DEFINE_SRCU(tracepoint_srcu); +EXPORT_SYMBOL_GPL(tracepoint_srcu); + /* Set to 1 to enable tracepoint debug output */ static const int tracepoint_debug; @@ -67,16 +70,27 @@ static inline void *allocate_probes(int count) return p == NULL ? NULL : p->probes; } -static void rcu_free_old_probes(struct rcu_head *head) +static void srcu_free_old_probes(struct rcu_head *head) { kfree(container_of(head, struct tp_probes, rcu)); } +static void rcu_free_old_probes(struct rcu_head *head) +{ + call_srcu(&tracepoint_srcu, head, srcu_free_old_probes); +} + static inline void release_probes(struct tracepoint_func *old) { if (old) { struct tp_probes *tp_probes = container_of(old, struct tp_probes, probes[0]); + /* + * Tracepoint probes are protected by both sched RCU and SRCU, + * by calling the SRCU callback in the sched RCU callback we + * cover both cases. So let us chain the SRCU and sched RCU + * callbacks to wait for both grace periods. + */ call_rcu_sched(&tp_probes->rcu, rcu_free_old_probes); } } -- cgit