Re: [PATCH v4 bpf-next 1/9] bpf: introduce bpf_spin_lock

From: Peter Zijlstra <peterz@infradead.org>
To: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Alexei Starovoitov <ast@kernel.org>,
	davem@davemloft.net, daniel@iogearbox.net,
	jakub.kicinski@netronome.com, netdev@vger.kernel.org,
	kernel-team@fb.com, mingo@redhat.com, will.deacon@arm.com,
	Paul McKenney <paulmck@linux.vnet.ibm.com>,
	jannh@google.com
Subject: Re: [PATCH v4 bpf-next 1/9] bpf: introduce bpf_spin_lock
Date: Mon, 28 Jan 2019 09:24:52 +0100	[thread overview]
Message-ID: <20190128082452.GA28467@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20190125234241.soomtkrgp2i7m7ul@ast-mbp.dhcp.thefacebook.com>

On Fri, Jan 25, 2019 at 03:42:43PM -0800, Alexei Starovoitov wrote:
> On Fri, Jan 25, 2019 at 10:10:57AM +0100, Peter Zijlstra wrote:

> > Do we want something like (the completely untested) below to avoid
> > having to manually audit this over and over?
> > 
> > ---
> >  include/linux/filter.h |  2 +-
> >  include/linux/kernel.h |  9 +++++++--
> >  kernel/sched/core.c    | 28 ++++++++++++++++++++++++++++
> >  3 files changed, 36 insertions(+), 3 deletions(-)
> > 
> > diff --git a/include/linux/filter.h b/include/linux/filter.h
> > index d531d4250bff..4ab51e78da36 100644
> > --- a/include/linux/filter.h
> > +++ b/include/linux/filter.h
> > @@ -513,7 +513,7 @@ struct sk_filter {
> >  	struct bpf_prog	*prog;
> >  };
> >  
> > -#define BPF_PROG_RUN(filter, ctx)  (*(filter)->bpf_func)(ctx, (filter)->insnsi)
> > +#define BPF_PROG_RUN(filter, ctx)  ({ cant_sleep(); (*(filter)->bpf_func)(ctx, (filter)->insnsi); })
> 
> That looks reasonable and I intent to apply this patch to bpf-next after testing.
> Can you pls reply with a sob ?

Sure; with the caveat that I didn't even hold it near a compiler, and it
probably should grow a comment to explain the interface (similar to
might_sleep):

Suggested-by: Jann Horn <jannh@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

> The easiest fix is to add preempt_disable/enable for socket filters.
> There is a concern that such fix will make classic bpf non-preemptable
> and classic bpf can be quite cpu expensive.

> Also on the receive side classic runs in bh, so 4k flow_dissector calls
> in classic has to be dealt with anyway.

Right and agreed; per that argument the worst case (legacy) BPF was
already present under non-preempt and thus making it consistently so
should not affect the worst case.