Re: [RFC PATCH bpf-next v5 2/2] bpf: Call rcu_momentary_dyntick_idle() in task work periodically

["Paul E. McKenney" <paulmck@kernel.org>]

On Tue, Jun 20, 2023 at 09:15:03AM -0700, Alexei Starovoitov wrote:
> On Mon, Jun 19, 2023 at 7:00 AM Hou Tao <houtao@huaweicloud.com> wrote:
> >
> > +static void bpf_rcu_gp_acc_work(struct callback_head *head)
> > +{
> > +       struct bpf_rcu_gp_acc_ctx *ctx = container_of(head, struct bpf_rcu_gp_acc_ctx, work);
> > +
> > +       local_irq_disable();
> > +       rcu_momentary_dyntick_idle();
> > +       local_irq_enable();
> 
> We discussed this with Paul off-line and decided to go a different route.
> Paul prepared a patch for us to expose rcu_request_urgent_qs_task().
> I'll be sending the series later this week.

Just for completeness, this patch is in -rcu here and as shown below:

b1edaa1e6364 ("rcu: Export rcu_request_urgent_qs_task()")

							Thanx, Paul

------------------------------------------------------------------------

commit b1edaa1e6364caf9833a30b5dd7599080b6806b8
Author: Paul E. McKenney <paulmck@kernel.org>
Date:   Thu Jun 8 16:31:49 2023 -0700

    rcu: Export rcu_request_urgent_qs_task()
    
    If a CPU is executing a long series of non-sleeping system calls,
    RCU grace periods can be delayed for on the order of a couple hundred
    milliseconds.  This is normally not a problem, but if each system call
    does a call_rcu(), those callbacks can stack up.  RCU will eventually
    notice this callback storm, but use of rcu_request_urgent_qs_task()
    allows the code invoking call_rcu() to give RCU a heads up.
    
    This function is not for general use, not yet, anyway.
    
    Reported-by: Alexei Starovoitov <ast@kernel.org>
    Signed-off-by: Paul E. McKenney <paulmck@kernel.org>

diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 7f17acf29dda..7b949292908a 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -138,6 +138,8 @@ static inline int rcu_needs_cpu(void)
 	return 0;
 }
 
+static inline void rcu_request_urgent_qs_task(struct task_struct *t) { }
+
 /*
  * Take advantage of the fact that there is only one CPU, which
  * allows us to ignore virtualization-based context switches.
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 56bccb5a8fde..126f6b418f6a 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -21,6 +21,7 @@ void rcu_softirq_qs(void);
 void rcu_note_context_switch(bool preempt);
 int rcu_needs_cpu(void);
 void rcu_cpu_stall_reset(void);
+void rcu_request_urgent_qs_task(struct task_struct *t);
 
 /*
  * Note a virtualization-based context switch.  This is simply a
diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
index 9829d8161b21..0c99ec19c8e0 100644
--- a/kernel/rcu/rcu.h
+++ b/kernel/rcu/rcu.h
@@ -493,7 +493,6 @@ static inline void rcu_expedite_gp(void) { }
 static inline void rcu_unexpedite_gp(void) { }
 static inline void rcu_async_hurry(void) { }
 static inline void rcu_async_relax(void) { }
-static inline void rcu_request_urgent_qs_task(struct task_struct *t) { }
 #else /* #ifdef CONFIG_TINY_RCU */
 bool rcu_gp_is_normal(void);     /* Internal RCU use. */
 bool rcu_gp_is_expedited(void);  /* Internal RCU use. */
@@ -514,7 +513,6 @@ struct task_struct *get_rcu_tasks_rude_gp_kthread(void);
 #else /* #ifdef CONFIG_TASKS_RCU_GENERIC */
 static inline void show_rcu_tasks_gp_kthreads(void) {}
 #endif /* #else #ifdef CONFIG_TASKS_RCU_GENERIC */
-void rcu_request_urgent_qs_task(struct task_struct *t);
 #endif /* #else #ifdef CONFIG_TINY_RCU */
 
 #define RCU_SCHEDULER_INACTIVE	0