Re: [PATCH RFC v3 0/8] xdp: Infrastructure to generalize XDP

On Tue, Feb 21, 2017 at 7:34 PM, David Miller [off-list ref] wrote:

From: Tom Herbert <redacted>
Date: Tue, 21 Feb 2017 18:54:53 -0800

quoted

It is part of the direction to take XDP beyond the first use case of
BPF and leverage the high performance processing model in a much
broader context.

And I've stated repeatedly that it's too early to be looking
that far into the future.

Look Tom, if all you want to do is create infrastructure so that
you can very easily slither XDP for modules in somewhere, please
just stop now.  Right now I am completely not interested in even
entertaining patches which facilitate that.

If instead, you are genuinely interested in making the process of
writing XDP support for drivers easier, that extremely useful right
now so please just stick to that specific focus.

I really see no other use case for the XDP hook abstraction other
than to support XDP module support, which I've said is in no way
proven to be necessary.

We don't even know what eBPF XDP itself is fully capable of yet,
so please stop reaching like this.

I took out the parts about allowing non-BPF hooks in the patchset as
you requested.I believe code that is cleaner than what is currently
there, and the fact that the API is extensible to allow other uses is
a hallmark of good design.

Tom