Skip to contentSkip to conversation

Softirq:avoid large sched delay from the pending softirqs

PATCH Superseded [email protected] 2 participants 3 messages Jul 23, 2020

v3 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 Softirq:avoid large sched delay from the pending softirqs

From: jun qian <[email protected]>

When get the pending softirqs, it need to process all the pending softirqs in the while loop. If the processing time of each pending softirq is need more than 2 msec in this loop, or one of the softirq will running a long time, according to the original code logic, it will process all the pending softirqs without wakeuping ksoftirqd, which will cause a relatively large scheduling delay on the corresponding CPU, which we do not wish to see. The patch will check the total time to process pending softirq, if the time exceeds 2 ms we need to wakeup the ksofirqd to aviod large sched delay.

---
 kernel/softirq.c | 21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)
kernel/softirq.c +13 -8
index c4201b7f..8f47554 100644
@@ -200,17 +200,15 @@ void __local_bh_enable_ip(unsigned long ip, unsigned int cnt)
 /*
  * We restart softirq processing for at most MAX_SOFTIRQ_RESTART times,
  * but break the loop if need_resched() is set or after 2 ms.
- * The MAX_SOFTIRQ_TIME provides a nice upper bound in most cases, but in
- * certain cases, such as stop_machine(), jiffies may cease to
- * increment and so we need the MAX_SOFTIRQ_RESTART limit as
- * well to make sure we eventually return from this method.
+ * In the loop, if the processing time of the softirq has exceeded 2
+ * milliseconds, we also need to break the loop to wakeup the ksofirqd.
  *
  * These limits have been established via experimentation.
  * The two things to balance is latency against fairness -
  * we want to handle softirqs as soon as possible, but they
  * should not be able to lock up the box.
  */
-#define MAX_SOFTIRQ_TIME  msecs_to_jiffies(2)
+#define MAX_SOFTIRQ_TIME_NS 2000000
 #define MAX_SOFTIRQ_RESTART 10
 
 #ifdef CONFIG_TRACE_IRQFLAGS
@@ -248,7 +246,7 @@ static inline void lockdep_softirq_end(bool in_hardirq) { }
 
 asmlinkage __visible void __softirq_entry __do_softirq(void)
 {
-	unsigned long end = jiffies + MAX_SOFTIRQ_TIME;
+	ktime_t end = ktime_get() + MAX_SOFTIRQ_TIME_NS;
 	unsigned long old_flags = current->flags;
 	int max_restart = MAX_SOFTIRQ_RESTART;
 	struct softirq_action *h;
@@ -299,6 +297,13 @@ asmlinkage __visible void __softirq_entry __do_softirq(void)
 		}
 		h++;
 		pending >>= softirq_bit;
+
+		/*
+		 * the softirq's action has been running for too much time
+		 * so it may need to wakeup the ksoftirqd
+		 */
+		if (need_resched() && ktime_get() > end)
+			break;
 	}
 
 	if (__this_cpu_read(ksoftirqd) == current)
@@ -307,8 +312,8 @@ asmlinkage __visible void __softirq_entry __do_softirq(void)
 
 	pending = local_softirq_pending();
 	if (pending) {
-		if (time_before(jiffies, end) && !need_resched() &&
-		    --max_restart)
+		if (!need_resched() && --max_restart &&
+		    ktime_get() <= end)
 			goto restart;
 
 		wakeup_softirqd();
Signature · 1 line
1.8.3.1
#1 lore raw

kernel/softirq.c asmlinkage __visible void __softirq_entry __do_softirq(void) · line 306

Quoted 7 lines · [email protected] writes:

From: jun qian <[email protected]>

kernel/softirq.c +6 -0
+		/*
+		 * the softirq's action has been running for too much time
+		 * so it may need to wakeup the ksoftirqd
+		 */
+		if (need_resched() && ktime_get() > end)
+			break;

As per my reply on V2 this is leaking non handled pending bits. If you do a V4, can you please use sched_clock() instead of ktime_get()?

Thanks,

tglx

author
#2 lore raw
Quoted 10 lines · Thomas Gleixner

kernel/softirq.c asmlinkage __visible void __softirq_entry __do_softirq(void) · line 306

Quoted 7 lines · [email protected] writes:

From: jun qian <[email protected]>

kernel/softirq.c +6 -0
+             /*
+              * the softirq's action has been running for too much time
+              * so it may need to wakeup the ksoftirqd
+              */
+             if (need_resched() && ktime_get() > end)
+                     break;
Quoted 3 lines

As per my reply on V2 this is leaking non handled pending bits. If you do a V4, can you please use sched_clock() instead of ktime_get()?

The reason why the non handled pending bits leaked is set_softirq_pending(0) called in the start, if the loop is broken, the not handled bit will leak. This is my understanding, I am not sure if it is correct or not. Looking forward to your reply.

Thank you so much.

Quoted 3 lines

Thanks,

tglx