Re: [RFC] tcp: race in receive part

On 06/24, Oleg Nesterov wrote:

On 06/24, Jiri Olsa wrote:

quoted

+/* The read_lock() on x86 is a full memory barrier. */
+#define smp_mb__after_read_lock() barrier()

Just curious, why do we need barrier() ?

I must admit, personally I dislike _read_lock part. Because I think we
need a "more generic" smp_mb__{before,after}_lock() or whatever which
work for spin_lock/read_lock/write_lock.

In that case it can have more users. Btw, in fs/select.c too, see
__pollwake().

And surprise,

quoted

--- a/fs/select.c
+++ b/fs/select.c

@@ -219,6 +219,10 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address,
 	init_waitqueue_func_entry(&entry->wait, pollwake);
 	entry->wait.private = pwq;
 	add_wait_queue(wait_address, &entry->wait);
+
+	/* This memory barrier is paired with the smp_mb__after_read_lock
+	 * in the sk_has_sleeper. */
+	smp_mb();

This could be smp_mb__after_lock() too.

Cough. this needs mb__after_UNlock(), sorry.

Oleg.