Re: [PATCH 0/5] bpf: Add user-space-publisher ringbuffer map type

[David Vernet]

On Tue, Aug 16, 2022 at 10:01:38AM -0700, Hao Luo wrote:
> On Tue, Aug 16, 2022 at 8:10 AM David Vernet <void@xxxxxxxxxxxxx> wrote:
> >
> > On Mon, Aug 15, 2022 at 02:13:13PM -0700, Hao Luo wrote:
> > > >
> > > > Iters allow userspace to kick the kernel, but IMO they're meant to enable
> > > > data extraction from the kernel, and dumping kernel data into user-space.
> > >
> > > Not necessarily extracting data and dumping data. It could be used to
> > > do operations on a set of objects, the operation could be
> > > notification. Iterating and notifying are orthogonal IMHO.
> > >
> > > > What I'm proposing is a more generalizable way of driving logic in the
> > > > kernel from user-space.
> > > > Does that make sense? Looking forward to hearing your thoughts.
> > >
> > > Yes, sort of. I see the difference between iter and the proposed
> > > interface. But I am not clear about the motivation of a new APis for
> > > kicking callbacks from userspace. I guess maybe it will become clear,
> > > when you publish a concerte RFC of that interface and integrates with
> > > your userspace publisher.
> >
> > Fair enough -- let me remove this from the cover letter in future
> > versions of the patch-set. To your point, there's probably little to be
> > gained in debating the merits of adding such APIs until there's a
> > concrete use-case.
> >
> 
> Yep, sounds good. I don't mean to debate :) I would like to help. If
> we could build on top of existing infra and make improvements, IMHO it
> would be easier to maintain. Anyway, I'm looking forward to your
> proposed APIs.

Don't worry, I did not take it that you were debating. I very much
appreciate your thoughts and help. If and when I send out that RFC
patchset, I'll be sure to cc you (if not reach out beforehand as well to
discuss).