From: Martin Karsten <mkarsten@xxxxxxxxxxxx>
Setting prefer_busy_poll now leads to an effectively nonblocking
iteration though napi_busy_loop, even when busy_poll_usecs is 0.
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>
---
fs/eventpoll.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index f53ca4f7fced..cc47f72005ed 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -420,7 +420,9 @@ static bool busy_loop_ep_timeout(unsigned long start_time,
static bool ep_busy_loop_on(struct eventpoll *ep)
{
- return !!ep->busy_poll_usecs || net_busy_loop_on();
+ return !!READ_ONCE(ep->busy_poll_usecs) ||
+ READ_ONCE(ep->prefer_busy_poll) ||
+ net_busy_loop_on();
}
static bool ep_busy_loop_end(void *p, unsigned long start_time)
--
2.25.1
