Re: [PATCH] bpf: verifier: avoid fall-through warnings

[Andrii Nakryiko]

On Tue, Jun 11, 2019 at 10:41 AM Gustavo A. R. Silva
<gustavo@xxxxxxxxxxxxxx> wrote:
>
>
>
> On 6/11/19 12:27 PM, Gustavo A. R. Silva wrote:
> >
> >
> > On 6/11/19 12:22 PM, Andrii Nakryiko wrote:
> >> On Tue, Jun 11, 2019 at 7:05 AM Gustavo A. R. Silva
> >> <gustavo@xxxxxxxxxxxxxx> wrote:
> >>>
> >>> In preparation to enabling -Wimplicit-fallthrough, this patch silences
> >>> the following warning:
> >>
> >> Your patch doesn't apply cleanly to neither bpf nor bpf-next tree.
> >> Could you please rebase and re-submit? Please also include which tree
> >> (probably bpf-next) you are designating this patch to in subject
> >> prefix.
> >>
> >
> > This patch applies cleanly to linux-next (tag next-20190611).
> >
>
> It seems that this commit hasn't been merged into bpf/bpf-next yet:
>
> 983695fa676568fc0fe5ddd995c7267aabc24632
>
> --
> Gustavo
>
> >>>
> >>> kernel/bpf/verifier.c: In function âcheck_return_codeâ:
> >>> kernel/bpf/verifier.c:5509:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
> >>>    if (env->prog->expected_attach_type == BPF_CGROUP_UDP4_RECVMSG ||
> >>>       ^
> >>> kernel/bpf/verifier.c:5512:2: note: here
> >>>   case BPF_PROG_TYPE_CGROUP_SKB:
> >>>   ^~~~
> >>>
> >>> Warning level 3 was used: -Wimplicit-fallthrough=3
> >>>
> >>> Notice that it's much clearer to explicitly add breaks in each case
> >>> (that actually contains some code), rather than letting the code to
> >>> fall through.
> >>>
> >>> This patch is part of the ongoing efforts to enable
> >>> -Wimplicit-fallthrough.
> >>>
> >>> Signed-off-by: Gustavo A. R. Silva <gustavo@xxxxxxxxxxxxxx>
> >>> ---
> >>>  kernel/bpf/verifier.c | 2 ++
> >>>  1 file changed, 2 insertions(+)
> >>>
> >>> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> >>> index 1e9d10b32984..e9fc28991548 100644
> >>> --- a/kernel/bpf/verifier.c
> >>> +++ b/kernel/bpf/verifier.c
> >>> @@ -5509,11 +5509,13 @@ static int check_return_code(struct bpf_verifier_env *env)
> >>>                 if (env->prog->expected_attach_type == BPF_CGROUP_UDP4_RECVMSG ||
> >>>                     env->prog->expected_attach_type == BPF_CGROUP_UDP6_RECVMSG)
> >>>                         range = tnum_range(1, 1);
> >>> +               break;

So this part is in bpf tree only...

> >>>         case BPF_PROG_TYPE_CGROUP_SKB:
> >>>                 if (env->prog->expected_attach_type == BPF_CGROUP_INET_EGRESS) {
> >>>                         range = tnum_range(0, 3);
> >>>                         enforce_attach_type_range = tnum_range(2, 3);
> >>>                 }
> >>> +               break;

... while this one is in bpf-next only.

Maybe just split this into two separate patches, one targeting bpf
tree and another for bpf-next tree? Unless you are willing to wait
till bpf is merged into bpf-next.

> >>>         case BPF_PROG_TYPE_CGROUP_SOCK:
> >>>         case BPF_PROG_TYPE_SOCK_OPS:
> >>>         case BPF_PROG_TYPE_CGROUP_DEVICE:
> >>> --
> >>> 2.21.0
> >>>