Re: [RFC] tcp: race in receive part

From: Oleg Nesterov
Date: Tue Jun 23 2009 - 18:59:47 EST


On 06/23, Eric Dumazet wrote:
> The optimization we could do, defining
> a smp_mb_after_read_lock() that could be a nop on x86

Imho it would be nice to have smp_mb__before/after_lock.

insert_work(), __pte_alloc(), try_to_wake_up() could use it. Probably
something else.


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at