Re: [net-next rfc 1/3] net: avoid high order memory allocation forqueues by using flex array

From: Eric Dumazet
Date: Thu Jun 20 2013 - 02:05:26 EST


On Thu, 2013-06-20 at 13:14 +0800, Jason Wang wrote:

> A drawback of NETIF_F_LLTX is that we may contend on qdisc lock
> especially when we have a huge number of tx queues.

For your information loopback driver is LLTX, and there is no qdisc on
it, unless you specifically add one.

Once you add a qdisc on a device, you hit the typical contention on a
spinlock. But its hard to design a parallel qdisc. So far nobody did
that.



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/