[PATCH 2/3] epoll: loosen irq safety in ep_poll_callback()

From: Roman Penyaev
Date: Wed Dec 12 2018 - 06:04:51 EST


Callers of the ep_poll_callback() (all set of wake_up_*poll()) disable
interrupts, so no need to save/restore irq flags.

Signed-off-by: Roman Penyaev <rpenyaev@xxxxxxx>
Cc: Davidlohr Bueso <dbueso@xxxxxxx>
Cc: Jason Baron <jbaron@xxxxxxxxxx>
Cc: Al Viro <viro@xxxxxxxxxxxxxxxxxx>
Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: linux-fsdevel@xxxxxxxxxxxxxxx
Cc: linux-kernel@xxxxxxxxxxxxxxx
---
fs/eventpoll.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 3627c2e07149..ea0025e77519 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1124,13 +1124,15 @@ struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd,
static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, void *key)
{
int pwake = 0;
- unsigned long flags;
struct epitem *epi = ep_item_from_wait(wait);
struct eventpoll *ep = epi->ep;
__poll_t pollflags = key_to_poll(key);
int ewake = 0;

- spin_lock_irqsave(&ep->wq.lock, flags);
+ /* Interrupts are disabled by the wake_up_*poll() callers */
+ lockdep_assert_irqs_disabled();
+
+ spin_lock(&ep->wq.lock);

ep_set_busy_poll_napi_id(epi);

@@ -1207,7 +1209,7 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v
pwake++;

out_unlock:
- spin_unlock_irqrestore(&ep->wq.lock, flags);
+ spin_unlock(&ep->wq.lock);

/* We have to call this outside the lock */
if (pwake)
--
2.19.1