Re: [PATCH v3] cgroup/bpf: fast path skb BPF filtering

[Stanislav Fomichev <sdf@google.com>]

On Wed, Dec 15, 2021 at 11:55 AM Pavel Begunkov <asml.silence@gmail.com> wrote:
>
> On 12/15/21 19:15, Stanislav Fomichev wrote:
> > On Wed, Dec 15, 2021 at 10:54 AM Pavel Begunkov <asml.silence@gmail.com> wrote:
> >>
> >> On 12/15/21 18:24, sdf@google.com wrote:
> >>> On 12/15, Pavel Begunkov wrote:
> >>>> On 12/15/21 17:33, sdf@google.com wrote:
> >>>>> On 12/15, Pavel Begunkov wrote:
> >>>>>> On 12/15/21 16:51, sdf@google.com wrote:
> >>>>>>> On 12/15, Pavel Begunkov wrote:
> >>>>>>>> � /* Wrappers for __cgroup_bpf_run_filter_skb() guarded by cgroup_bpf_enabled. */
> >>>>>>>> � #define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk, skb)����������������� \
> >>>>>>>> � ({����������������������������������������� \
> >>>>>>>> ����� int __ret = 0;��������������������������������� \
> >>>>>>>> -��� if (cgroup_bpf_enabled(CGROUP_INET_INGRESS))������������� \
> >>>>>>>> +��� if (cgroup_bpf_enabled(CGROUP_INET_INGRESS) && sk &&������������� \
> >>>>>>>> +������� CGROUP_BPF_TYPE_ENABLED((sk), CGROUP_INET_INGRESS))���������� \
> >>>>>>>
> >>>>>>> Why not add this __cgroup_bpf_run_filter_skb check to
> >>>>>>> __cgroup_bpf_run_filter_skb? Result of sock_cgroup_ptr() is already there
> >>>>>>> and you can use it. Maybe move the things around if you want
> >>>>>>> it to happen earlier.
> >>>>>
> >>>>>> For inlining. Just wanted to get it done right, otherwise I'll likely be
> >>>>>> returning to it back in a few months complaining that I see measurable
> >>>>>> overhead from the function call :)
> >>>>>
> >>>>> Do you expect that direct call to bring any visible overhead?
> >>>>> Would be nice to compare that inlined case vs
> >>>>> __cgroup_bpf_prog_array_is_empty inside of __cgroup_bpf_run_filter_skb
> >>>>> while you're at it (plus move offset initialization down?).
> >>>
> >>>> Sorry but that would be waste of time. I naively hope it will be visible
> >>>> with net at some moment (if not already), that's how it was with io_uring,
> >>>> that's what I see in the block layer. And in anyway, if just one inlined
> >>>> won't make a difference, then 10 will.
> >>>
> >>> I can probably do more experiments on my side once your patch is
> >>> accepted. I'm mostly concerned with getsockopt(TCP_ZEROCOPY_RECEIVE).
> >>> If you claim there is visible overhead for a direct call then there
> >>> should be visible benefit to using CGROUP_BPF_TYPE_ENABLED there as
> >>> well.
> >>
> >> Interesting, sounds getsockopt might be performance sensitive to
> >> someone.
> >>
> >> FWIW, I forgot to mention that for testing tx I'm using io_uring
> >> (for both zc and not) with good submission batching.
> >
> > Yeah, last time I saw 2-3% as well, but it was due to kmalloc, see
> > more details in 9cacf81f8161, it was pretty visible under perf.
> > That's why I'm a bit skeptical of your claims of direct calls being
> > somehow visible in these 2-3% (even skb pulls/pushes are not 2-3%?).
>
> migrate_disable/enable together were taking somewhat in-between
> 1% and 1.5% in profiling, don't remember the exact number. The rest
> should be from rcu_read_lock/unlock() in BPF_PROG_RUN_ARRAY_CG_FLAGS()
> and other extra bits on the way.

You probably have a preemptiple kernel and preemptible rcu which most
likely explains why you see the overhead and I won't (non-preemptible
kernel in our env, rcu_read_lock is essentially a nop, just a compiler
barrier).

> I'm skeptical I'll be able to measure inlining one function,
> variability between boots/runs is usually greater and would hide it.

Right, that's why I suggested to mirror what we do in set/getsockopt
instead of the new extra CGROUP_BPF_TYPE_ENABLED. But I'll leave it up
to you, Martin and the rest.