[dpdk-dev] [PATCH v9 1/8] ethdev: use locks to protect Rx/Tx callback lists

[Thomas Monjalon]

I agree this patch do not bring a new issue.
But the current status deserves to be discussed.

2016-06-15 09:54, Ananyev, Konstantin:
> It is safe to add/remove RX/TX callbacks while 
> another thread is doing simultaneously RX/TX burst over same queue.

You are probably right, but I don't why it is safe? On which CPU?
How can we be sure that read and write of the "next" pointer are atomic?