Re: [PATCH v2 1/2] sched: Add cond_resched_rcu_lock() helper
From: Paul E. McKenney <hidden>
Date: 2013-05-02 17:35:27
Also in:
lkml, lvs-devel, netfilter-devel
On Thu, May 02, 2013 at 06:54:05PM +0300, Julian Anastasov wrote:
Hello, On Thu, 2 May 2013, Peter Zijlstra wrote:quoted
On Wed, May 01, 2013 at 09:22:08PM +0300, Julian Anastasov wrote:quoted
quoted
+extern int __cond_resched_rcu(void); + +#define cond_resched_rcu() ({ \ + __might_sleep(__FILE__, __LINE__, 0); \I see your goal. But digging into __might_sleep() I see that rcu_sleep_check() will scream for the non-preempt case because we are under rcu_read_lock.#ifdef CONFIG_PREEMPT_RCU #define PREEMPT_RCU_OFFSET 0 #else #define PREEMPT_RCU_OFFSET 1 #endif #define cond_resched_rcu() ({ \ __might_sleep(__FILE__, __LINE__, PREEMPT_RCU_OFFSET); \ __cond_resched_rcu(); \ }) Should work I think..I implemented your idea. I tested the following patch in 2 variants, TINY_RCU and CONFIG_TREE_PREEMPT_RCU. I see the
Could you please also try CONFIG_TREE_RCU?
quoted hunk ↗ jump to hunk
error if extra rcu_read_lock is added for testing. I'm using the PREEMPT_ACTIVE flag to indicate that we are already under lock. It should work because __might_sleep is not called with such bit. I also tried to add new flag in include/linux/hardirq.h but PREEMPT_ACTIVE depends on the arch, so this alternative looked difficult to implement. include/linux/rcupdate.h | 7 ++++--- include/linux/sched.h | 14 ++++++++++++++ kernel/sched/core.c | 20 ++++++++++++++++++-- 3 files changed, 36 insertions(+), 5 deletions(-)diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index b758ce1..b594759 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h@@ -480,9 +480,10 @@ static inline void rcu_preempt_sleep_check(void) } #endif /* #else #ifdef CONFIG_PROVE_RCU */ -#define rcu_sleep_check() \ +#define rcu_sleep_check(locked) \ do { \ - rcu_preempt_sleep_check(); \ + if (!(locked)) \ + rcu_preempt_sleep_check(); \ rcu_lockdep_assert(!lock_is_held(&rcu_bh_lock_map), \ "Illegal context switch in RCU-bh" \ " read-side critical section"); \@@ -494,7 +495,7 @@ static inline void rcu_preempt_sleep_check(void) #else /* #ifdef CONFIG_PROVE_RCU */ #define rcu_lockdep_assert(c, s) do { } while (0) -#define rcu_sleep_check() do { } while (0) +#define rcu_sleep_check(locked) do { } while (0) #endif /* #else #ifdef CONFIG_PROVE_RCU */diff --git a/include/linux/sched.h b/include/linux/sched.h index e692a02..027deea 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h@@ -2608,6 +2608,20 @@ extern int __cond_resched_softirq(void); __cond_resched_softirq(); \ }) +#ifdef CONFIG_PREEMPT_RCU +#define PREEMPT_RCU_OFFSET 1 +#else +#define PREEMPT_RCU_OFFSET PREEMPT_CHECK_OFFSET +#endif + +extern int __cond_resched_rcu(void); + +#define cond_resched_rcu() ({ \ + __might_sleep(__FILE__, __LINE__, PREEMPT_ACTIVE | \ + PREEMPT_RCU_OFFSET); \ + __cond_resched_rcu(); \ +}) + /* * Does a critical section need to be broken due to another * task waiting?: (technically does not depend on CONFIG_PREEMPT,diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 67d0465..2724be7 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c@@ -2793,7 +2793,7 @@ static inline void schedule_debug(struct task_struct *prev) */ if (unlikely(in_atomic_preempt_off() && !prev->exit_state)) __schedule_bug(prev); - rcu_sleep_check(); + rcu_sleep_check(0); profile_hit(SCHED_PROFILING, __builtin_return_address(0));@@ -4364,6 +4364,20 @@ int __sched __cond_resched_softirq(void) } EXPORT_SYMBOL(__cond_resched_softirq); +int __sched __cond_resched_rcu(void) +{ +#ifndef CONFIG_PREEMPT_RCU + if (should_resched()) { + rcu_read_unlock(); + __cond_resched(); + rcu_read_lock(); + return 1; + } +#endif + return 0; +} +EXPORT_SYMBOL(__cond_resched_rcu); + /** * yield - yield the current processor to other threads. *@@ -7062,7 +7076,9 @@ void __might_sleep(const char *file, int line, int preempt_offset) { static unsigned long prev_jiffy; /* ratelimiting */ - rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */ + /* WARN_ON_ONCE() by default, no rate limit reqd. */ + rcu_sleep_check(preempt_offset & PREEMPT_ACTIVE);
Color me confused.
From what I can see, the two values passed in through preempt_offset
are PREEMPT_LOCK_OFFSET and SOFTIRQ_DISABLE_OFFSET. PREEMPT_ACTIVE is normally a high-order bit, above PREEMPT_MASK, SOFTIRQ_MASK, and HARDIRQ_MASK. PREEMPT_LOCK_OFFSET and SOFTIRQ_DISABLE_OFFSET have only low-order bits, so I don't see how rcu_sleep_check() is passed anything other than zero. Am I going blind, or what? Thanx, Paul
+ preempt_offset &= ~PREEMPT_ACTIVE; if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || system_state != SYSTEM_RUNNING || oops_in_progress) return; -- 1.7.3.4 Regards -- Julian Anastasov [off-list ref] -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html