[PATCH net-next v9 4/6] eventpoll: Control irq suspension for prefer_busy_poll

[Joe Damato <jdamato@xxxxxxxxxx>]

From: Martin Karsten <mkarsten@xxxxxxxxxxxx>

When events are reported to userland and prefer_busy_poll is set, irqs
are temporarily suspended using napi_suspend_irqs.

If no events are found and ep_poll would go to sleep, irq suspension is
cancelled using napi_resume_irqs.

Signed-off-by: Martin Karsten <mkarsten@xxxxxxxxxxxx>
Co-developed-by: Joe Damato <jdamato@xxxxxxxxxx>
Signed-off-by: Joe Damato <jdamato@xxxxxxxxxx>
Tested-by: Joe Damato <jdamato@xxxxxxxxxx>
Tested-by: Martin Karsten <mkarsten@xxxxxxxxxxxx>
Acked-by: Stanislav Fomichev <sdf@xxxxxxxxxxx>
Reviewed-by: Sridhar Samudrala <sridhar.samudrala@xxxxxxxxx>
---
 v5:
   - Only call ep_suspend_napi_irqs when ep_send_events returns a
     positive value. IRQs are not suspended in error (e.g. EINTR)
     cases. This issue was pointed out by Hillf Danton.

 rfc -> v1:
   - move irq resume code from ep_free to a helper which either resumes
     IRQs or does nothing if !defined(CONFIG_NET_RX_BUSY_POLL).

 fs/eventpoll.c | 32 +++++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index f9e0d9307dad..83bcb559b89f 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -457,6 +457,8 @@ static bool ep_busy_loop(struct eventpoll *ep, int nonblock)
 		 * it back in when we have moved a socket with a valid NAPI
 		 * ID onto the ready list.
 		 */
+		if (prefer_busy_poll)
+			napi_resume_irqs(napi_id);
 		ep->napi_id = 0;
 		return false;
 	}
@@ -540,6 +542,22 @@ static long ep_eventpoll_bp_ioctl(struct file *file, unsigned int cmd,
 	}
 }
 
+static void ep_suspend_napi_irqs(struct eventpoll *ep)
+{
+	unsigned int napi_id = READ_ONCE(ep->napi_id);
+
+	if (napi_id >= MIN_NAPI_ID && READ_ONCE(ep->prefer_busy_poll))
+		napi_suspend_irqs(napi_id);
+}
+
+static void ep_resume_napi_irqs(struct eventpoll *ep)
+{
+	unsigned int napi_id = READ_ONCE(ep->napi_id);
+
+	if (napi_id >= MIN_NAPI_ID && READ_ONCE(ep->prefer_busy_poll))
+		napi_resume_irqs(napi_id);
+}
+
 #else
 
 static inline bool ep_busy_loop(struct eventpoll *ep, int nonblock)
@@ -557,6 +575,14 @@ static long ep_eventpoll_bp_ioctl(struct file *file, unsigned int cmd,
 	return -EOPNOTSUPP;
 }
 
+static void ep_suspend_napi_irqs(struct eventpoll *ep)
+{
+}
+
+static void ep_resume_napi_irqs(struct eventpoll *ep)
+{
+}
+
 #endif /* CONFIG_NET_RX_BUSY_POLL */
 
 /*
@@ -788,6 +814,7 @@ static bool ep_refcount_dec_and_test(struct eventpoll *ep)
 
 static void ep_free(struct eventpoll *ep)
 {
+	ep_resume_napi_irqs(ep);
 	mutex_destroy(&ep->mtx);
 	free_uid(ep->user);
 	wakeup_source_unregister(ep->ws);
@@ -2005,8 +2032,11 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
 			 * trying again in search of more luck.
 			 */
 			res = ep_send_events(ep, events, maxevents);
-			if (res)
+			if (res) {
+				if (res > 0)
+					ep_suspend_napi_irqs(ep);
 				return res;
+			}
 		}
 
 		if (timed_out)
-- 
2.25.1