Re: [PATCH bpf-next 1/6] i40e: introduce lazy Tx completions for AF_XDP zero-copy

From: Magnus Karlsson <magnus.karlsson@gmail.com>
To: Jakub Kicinski <kuba@kernel.org>
Cc: "Karlsson, Magnus" <magnus.karlsson@intel.com>,
	"Björn Töpel" <bjorn.topel@intel.com>,
	"Alexei Starovoitov" <ast@kernel.org>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	"Network Development" <netdev@vger.kernel.org>,
	"Jonathan Lemon" <jonathan.lemon@gmail.com>,
	bpf <bpf@vger.kernel.org>,
	jeffrey.t.kirsher@intel.com, anthony.l.nguyen@intel.com,
	"Fijalkowski, Maciej" <maciej.fijalkowski@intel.com>,
	"Maciej Fijalkowski" <maciejromanfijalkowski@gmail.com>,
	intel-wired-lan <intel-wired-lan@lists.osuosl.org>
Subject: Re: [PATCH bpf-next 1/6] i40e: introduce lazy Tx completions for AF_XDP zero-copy
Date: Thu, 5 Nov 2020 15:17:50 +0100	[thread overview]
Message-ID: <CAJ8uoz3-tjXekU=kR+HfMhGBcHtAFnKGq1ZvpFq99T_S-mknPg@mail.gmail.com> (raw)
In-Reply-To: <20201104153320.66cecba8@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>

On Thu, Nov 5, 2020 at 12:33 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Wed,  4 Nov 2020 15:08:57 +0100 Magnus Karlsson wrote:
> > From: Magnus Karlsson <magnus.karlsson@intel.com>
> >
> > Introduce lazy Tx completions when a queue is used for AF_XDP
> > zero-copy. In the current design, each time we get into the NAPI poll
> > loop we try to complete as many Tx packets as possible from the
> > NIC. This is performed by reading the head pointer register in the NIC
> > that tells us how many packets have been completed. Reading this
> > register is expensive as it is across PCIe, so let us try to limit the
> > number of times it is read by only completing Tx packets to user-space
> > when the number of available descriptors in the Tx HW ring is below
> > some threshold. This will decrease the number of reads issued to the
> > NIC and improves performance with 1.5% - 2% for the l2fwd xdpsock
> > microbenchmark.
> >
> > The threshold is set to the minimum possible size that the HW ring can
> > have. This so that we do not run into a scenario where the threshold
> > is higher than the configured number of descriptors in the HW ring.
> >
> > Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
>
> I feel like this needs a big fat warning somewhere.
>
> It's perfectly fine to never complete TCP packets, but AF_XDP could be
> used to implement protocols in user space. What if someone wants to
> implement something like TSQ?

I might misunderstand you, but with TSQ here (for something that
bypasses qdisk and any buffering and just goes straight to the driver)
you mean the ability to have just a few buffers outstanding and
continuously reuse these? If so, that is likely best achieved by
setting a low Tx queue size on the NIC. Note that even without this
patch, completions could be delayed. Though this patch makes that the
normal case. In any way, I think this calls for some improved
documentation.

I also discovered a corner case that will lead to a deadlock if the
completion ring size is half the size of the Tx NIC ring size. This
needs to be fixed, so I will spin a v2.

Thanks: Magnus