net: solve a NAPI race

[linux.git] / net / core / dev.c

diff --git a/net/core/dev.c b/net/core/dev.c
index 304f2deae5f9897e60a79ed8b69d6ef208295ded..e63bf61b19be029e30ac40443c0e2edb24de4a73 100644 (file)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4883,6 +4883,39 @@ void __napi_schedule(struct napi_struct *n)
 }
 EXPORT_SYMBOL(__napi_schedule);
 
+/**
+ *     napi_schedule_prep - check if napi can be scheduled
+ *     @n: napi context
+ *
+ * Test if NAPI routine is already running, and if not mark
+ * it as running.  This is used as a condition variable
+ * insure only one NAPI poll instance runs.  We also make
+ * sure there is no pending NAPI disable.
+ */
+bool napi_schedule_prep(struct napi_struct *n)
+{
+       unsigned long val, new;
+
+       do {
+               val = READ_ONCE(n->state);
+               if (unlikely(val & NAPIF_STATE_DISABLE))
+                       return false;
+               new = val | NAPIF_STATE_SCHED;
+
+               /* Sets STATE_MISSED bit if STATE_SCHED was already set
+                * This was suggested by Alexander Duyck, as compiler
+                * emits better code than :
+                * if (val & NAPIF_STATE_SCHED)
+                *     new |= NAPIF_STATE_MISSED;
+                */
+               new |= (val & NAPIF_STATE_SCHED) / NAPIF_STATE_SCHED *
+                                                  NAPIF_STATE_MISSED;
+       } while (cmpxchg(&n->state, val, new) != val);
+
+       return !(val & NAPIF_STATE_SCHED);
+}
+EXPORT_SYMBOL(napi_schedule_prep);
+
 /**
  * __napi_schedule_irqoff - schedule for receive
  * @n: entry to schedule
@@ -4897,7 +4930,7 @@ EXPORT_SYMBOL(__napi_schedule_irqoff);
 
 bool napi_complete_done(struct napi_struct *n, int work_done)
 {
-       unsigned long flags;
+       unsigned long flags, val, new;
 
        /*
         * 1) Don't let napi dequeue from the cpu poll list
@@ -4927,7 +4960,27 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
                list_del_init(&n->poll_list);
                local_irq_restore(flags);
        }
-       WARN_ON_ONCE(!test_and_clear_bit(NAPI_STATE_SCHED, &n->state));
+
+       do {
+               val = READ_ONCE(n->state);
+
+               WARN_ON_ONCE(!(val & NAPIF_STATE_SCHED));
+
+               new = val & ~(NAPIF_STATE_MISSED | NAPIF_STATE_SCHED);
+
+               /* If STATE_MISSED was set, leave STATE_SCHED set,
+                * because we will call napi->poll() one more time.
+                * This C code was suggested by Alexander Duyck to help gcc.
+                */
+               new |= (val & NAPIF_STATE_MISSED) / NAPIF_STATE_MISSED *
+                                                   NAPIF_STATE_SCHED;
+       } while (cmpxchg(&n->state, val, new) != val);
+
+       if (unlikely(val & NAPIF_STATE_MISSED)) {
+               __napi_schedule(n);
+               return false;
+       }
+
        return true;
 }
 EXPORT_SYMBOL(napi_complete_done);
@@ -4953,6 +5006,16 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock)
 {
        int rc;
 
+       /* Busy polling means there is a high chance device driver hard irq
+        * could not grab NAPI_STATE_SCHED, and that NAPI_STATE_MISSED was
+        * set in napi_schedule_prep().
+        * Since we are about to call napi->poll() once more, we can safely
+        * clear NAPI_STATE_MISSED.
+        *
+        * Note: x86 could use a single "lock and ..." instruction
+        * to perform these two clear_bit()
+        */
+       clear_bit(NAPI_STATE_MISSED, &napi->state);
        clear_bit(NAPI_STATE_IN_BUSY_POLL, &napi->state);
 
        local_bh_disable();
@@ -5088,8 +5151,13 @@ static enum hrtimer_restart napi_watchdog(struct hrtimer *timer)
        struct napi_struct *napi;
 
        napi = container_of(timer, struct napi_struct, timer);
-       if (napi->gro_list)
-               napi_schedule_irqoff(napi);
+
+       /* Note : we use a relaxed variant of napi_schedule_prep() not setting
+        * NAPI_STATE_MISSED, since we do not react to a device IRQ.
+        */
+       if (napi->gro_list && !napi_disable_pending(napi) &&
+           !test_and_set_bit(NAPI_STATE_SCHED, &napi->state))
+               __napi_schedule_irqoff(napi);
 
        return HRTIMER_NORESTART;
 }