Re: [RFC V2 PATCH 17/25] net/netpolicy: introduce netpolicy_pick_queue

On Fri, Aug 5, 2016 at 6:55 AM, Liang, Kan [off-list ref] wrote:

quoted

On Thu, Aug 4, 2016 at 12:36 PM,  [off-list ref] wrote:

quoted

From: Kan Liang <redacted>

To achieve better network performance, the key step is to distribute
the packets to dedicated queues according to policy and system run
time status.

This patch provides an interface which can return the proper dedicated
queue for socket/task. Then the packets of the socket/task will be
redirect to the dedicated queue for better network performance.

For selecting the proper queue, currently it uses round-robin
algorithm to find the available object from the given policy object
list. The algorithm is good enough for now. But it could be improved
by some adaptive algorithm later.

Seriously? You want to all of this code so we revert to TX queue selection by
round robin?

I agree that the round robin is not an optimal algorithm.
For this series, we intends to provide a generic infrastructure for review.
For the algorithm parts, it's already in our TODO list. We will replace it later.

Kan,

The justification for this patch is that to achieve to network
performance to steer TX distribute the packets to according to policy
and system runtime status. But the patch doesn't remotely implement
that, there's no data provided that these do anything useful or ever
will do anything useful, and it seems like this is completely ignoring
existing mechanisms like XPS that have proven to improve performance.

Tom