Re: [PATCH v9 01/10] btf: Add a new kfunc flag which allows to mark a function to be sleepable

[Benjamin Tissoires]

On Wed, Aug 10, 2022 at 3:44 PM Roberto Sassu <roberto.sassu@xxxxxxxxxx> wrote:
>
> > From: Jarkko Sakkinen [mailto:jarkko@xxxxxxxxxx]
> > Sent: Tuesday, August 9, 2022 6:55 PM
> > On Tue, Aug 09, 2022 at 03:45:54PM +0200, Roberto Sassu wrote:
> > > From: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx>
> > >
> > > From: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx>
> > >
> > > This allows to declare a kfunc as sleepable and prevents its use in
> > > a non sleepable program.
> >
> > Nit: "Declare a kfunc as sleepable and prevent its use in a
> > non-sleepable program."
> >
> > It's missing the part *how* the patch accomplishes its goals.
>
> I will add:
>
> If an eBPF program is going to call a kfunc declared as sleepable,
> eBPF will look at the eBPF program flags. If BPF_F_SLEEPABLE is
> not set, execution of that program is denied.

All those changes are looking good to me.

Thanks a lot for keeping pushing on this patch :)

Cheers,
Benjamin

>
> Roberto
>
> > > Signed-off-by: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx>
> > > Co-developed-by: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
> > > Signed-off-by: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
> > > Signed-off-by: Hao Luo <haoluo@xxxxxxxxxx>
> > > ---
> > >  Documentation/bpf/kfuncs.rst | 6 ++++++
> > >  include/linux/btf.h          | 1 +
> > >  kernel/bpf/btf.c             | 9 +++++++++
> > >  3 files changed, 16 insertions(+)
> > >
> > > diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
> > > index c0b7dae6dbf5..c8b21de1c772 100644
> > > --- a/Documentation/bpf/kfuncs.rst
> > > +++ b/Documentation/bpf/kfuncs.rst
> > > @@ -146,6 +146,12 @@ that operate (change some property, perform some
> > operation) on an object that
> > >  was obtained using an acquire kfunc. Such kfuncs need an unchanged pointer
> > to
> > >  ensure the integrity of the operation being performed on the expected object.
> > >
> > > +2.4.6 KF_SLEEPABLE flag
> > > +-----------------------
> > > +
> > > +The KF_SLEEPABLE flag is used for kfuncs that may sleep. Such kfuncs can
> > only
> > > +be called by sleepable BPF programs (BPF_F_SLEEPABLE).
> > > +
> > >  2.5 Registering the kfuncs
> > >  --------------------------
> > >
> > > diff --git a/include/linux/btf.h b/include/linux/btf.h
> > > index cdb376d53238..976cbdd2981f 100644
> > > --- a/include/linux/btf.h
> > > +++ b/include/linux/btf.h
> > > @@ -49,6 +49,7 @@
> > >   * for this case.
> > >   */
> > >  #define KF_TRUSTED_ARGS (1 << 4) /* kfunc only takes trusted pointer
> > arguments */
> > > +#define KF_SLEEPABLE   (1 << 5) /* kfunc may sleep */
> > >
> > >  struct btf;
> > >  struct btf_member;
> > > diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
> > > index 7e64447659f3..d3e4c86b8fcd 100644
> > > --- a/kernel/bpf/btf.c
> > > +++ b/kernel/bpf/btf.c
> > > @@ -6175,6 +6175,7 @@ static int btf_check_func_arg_match(struct
> > bpf_verifier_env *env,
> > >  {
> > >     enum bpf_prog_type prog_type = resolve_prog_type(env->prog);
> > >     bool rel = false, kptr_get = false, trusted_arg = false;
> > > +   bool sleepable = false;
> > >     struct bpf_verifier_log *log = &env->log;
> > >     u32 i, nargs, ref_id, ref_obj_id = 0;
> > >     bool is_kfunc = btf_is_kernel(btf);
> > > @@ -6212,6 +6213,7 @@ static int btf_check_func_arg_match(struct
> > bpf_verifier_env *env,
> > >             rel = kfunc_flags & KF_RELEASE;
> > >             kptr_get = kfunc_flags & KF_KPTR_GET;
> > >             trusted_arg = kfunc_flags & KF_TRUSTED_ARGS;
> > > +           sleepable = kfunc_flags & KF_SLEEPABLE;
> > >     }
> > >
> > >     /* check that BTF function arguments match actual types that the
> > > @@ -6419,6 +6421,13 @@ static int btf_check_func_arg_match(struct
> > bpf_verifier_env *env,
> > >                     func_name);
> > >             return -EINVAL;
> > >     }
> > > +
> > > +   if (sleepable && !env->prog->aux->sleepable) {
> > > +           bpf_log(log, "kernel function %s is sleepable but the program is
> > not\n",
> > > +                   func_name);
> > > +           return -EINVAL;
> > > +   }
> > > +
> > >     /* returns argument register number > 0 in case of reference release
> > kfunc */
> > >     return rel ? ref_regno : 0;
> > >  }
> > > --
> > > 2.25.1
> > >
> >
> > BR, Jarkko
>