From: Jiri Olsa <olsajiri@gmail.com>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: bpf@vger.kernel.org, Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Eduard Zingerman <eddyz87@gmail.com>, Hao Luo <haoluo@google.com>,
John Fastabend <john.fastabend@gmail.com>,
KP Singh <kpsingh@kernel.org>,
Martin KaFai Lau <martin.lau@linux.dev>,
Song Liu <song@kernel.org>, Stanislav Fomichev <sdf@fomichev.me>,
Thomas Gleixner <tglx@linutronix.de>,
Yonghong Song <yonghong.song@linux.dev>
Subject: Re: [PATCH v2 bpf-next 3/3] bpf: Implement bpf_check_basics_ok() as a macro.
Date: Tue, 2 Jul 2024 18:27:31 +0200 [thread overview]
Message-ID: <ZoQqc_dNjxF1-AR-@krava> (raw)
In-Reply-To: <20240702142542.179753-4-bigeasy@linutronix.de>
On Tue, Jul 02, 2024 at 04:21:43PM +0200, Sebastian Andrzej Siewior wrote:
> sparse complains about the argument type for filter that is passed to
> bpf_check_basics_ok(). There are two users of the function where the
> variable is with __user attribute one without. The pointer is only
> checked against NULL so there is no access to the content and so no need
> for any user-wrapper.
>
> Adding the __user to the declaration doesn't solve anything because
> there is one kernel user so it will be wrong again.
> Splitting the function in two seems an overkill because the function is
> small and simple.
could we just retype the __user argument? like
bpf_check_basics_ok((const struct sock_filter *) fprog->filter, ...)
>
> Make a macro based on the function which does not trigger a sparse
> warning. The change to a macro and "unsigned int" -> "u16" for `flen'
> alters gcc's code generation a bit.
>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
> net/core/filter.c | 24 ++++++++++++++----------
> 1 file changed, 14 insertions(+), 10 deletions(-)
>
> diff --git a/net/core/filter.c b/net/core/filter.c
> index 3f14c8019f26d..5747533ed5491 100644
> --- a/net/core/filter.c
> +++ b/net/core/filter.c
> @@ -1035,16 +1035,20 @@ static bool chk_code_allowed(u16 code_to_probe)
> return codes[code_to_probe];
> }
>
> -static bool bpf_check_basics_ok(const struct sock_filter *filter,
> - unsigned int flen)
> -{
> - if (filter == NULL)
> - return false;
> - if (flen == 0 || flen > BPF_MAXINSNS)
> - return false;
> -
> - return true;
> -}
> + /* macro instead of a function to avoid woring about _filter which might be a
> + * user or kernel pointer. It does not matter for the NULL check.
> + */
> +#define bpf_check_basics_ok(fprog_filter, fprog_flen) \
> +({ \
> + bool __ret = true; \
> + u16 __flen = fprog_flen; \
why not use fprog_flen directly? I'm not sure I get the changelog
explanation
thanks,
jirka
> + \
> + if (!(fprog_filter)) \
> + __ret = false; \
> + else if (__flen == 0 || __flen > BPF_MAXINSNS) \
> + __ret = false; \
> + __ret; \
> +})
>
> /**
> * bpf_check_classic - verify socket filter code
> --
> 2.45.2
>
next prev parent reply other threads:[~2024-07-02 16:27 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-02 14:21 [PATCH v2 bpf-next 0/3] bpf: sparse cleanup Sebastian Andrzej Siewior
2024-07-02 14:21 ` [PATCH v2 bpf-next 1/3] bpf: Add casts to keep sparse quiet Sebastian Andrzej Siewior
2024-07-03 21:39 ` Alexei Starovoitov
2024-07-04 8:00 ` Sebastian Andrzej Siewior
2024-07-07 0:42 ` Alexei Starovoitov
2024-07-02 14:21 ` [PATCH v2 bpf-next 2/3] bpf: Move a few bpf_func_proto declarations Sebastian Andrzej Siewior
2024-07-02 22:30 ` Andrii Nakryiko
2024-07-02 14:21 ` [PATCH v2 bpf-next 3/3] bpf: Implement bpf_check_basics_ok() as a macro Sebastian Andrzej Siewior
2024-07-02 16:27 ` Jiri Olsa [this message]
2024-07-02 18:12 ` Sebastian Andrzej Siewior
2024-07-02 22:26 ` Andrii Nakryiko
2024-07-03 12:39 ` Sebastian Andrzej Siewior
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZoQqc_dNjxF1-AR-@krava \
--to=olsajiri@gmail.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bigeasy@linutronix.de \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=haoluo@google.com \
--cc=john.fastabend@gmail.com \
--cc=kpsingh@kernel.org \
--cc=martin.lau@linux.dev \
--cc=sdf@fomichev.me \
--cc=song@kernel.org \
--cc=tglx@linutronix.de \
--cc=yonghong.song@linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox