netdev - Re: [PATCH net-next 2/2] net/sched: cls_flower: add support for matching tunnel control flags

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Zbt9JmpDq3cAbq1b@dcaratti.users.ipa.redhat.com>
Date: Thu, 1 Feb 2024 12:14:46 +0100
From: Davide Caratti <dcaratti@...hat.com>
To: Jamal Hadi Salim <jhs@...atatu.com>
Cc: Cong Wang <xiyou.wangcong@...il.com>, Jiri Pirko <jiri@...nulli.us>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	netdev@...r.kernel.org,
	Marcelo Ricardo Leitner <marcelo.leitner@...il.com>,
	Xin Long <lucien.xin@...il.com>, Ilya Maximets <i.maximets@....org>
Subject: Re: [PATCH net-next 2/2] net/sched: cls_flower: add support for
 matching tunnel control flags

hello Jamal, thanks for looking at this!

On Wed, Jan 31, 2024 at 04:13:25PM -0500, Jamal Hadi Salim wrote:
> On Wed, Jan 31, 2024 at 11:16 AM Davide Caratti <dcaratti@...hat.com> wrote:
> >
> > extend cls_flower to match flags belonging to 'TUNNEL_FLAGS_PRESENT' mask
> > inside skb tunnel metadata.
> >
> > Suggested-by: Ilya Maximets <i.maximets@....org>
> > Signed-off-by: Davide Caratti <dcaratti@...hat.com>

[...]

> > @@ -1748,6 +1753,21 @@ static int fl_set_key_cfm(struct nlattr **tb,
> >         return 0;
> >  }
> >
> > +static int fl_set_key_enc_flags(struct nlattr **tb, __be16 *flags_key,
> > +                               __be16 *flags_mask, struct netlink_ext_ack *extack)
> > +{
> > +       /* mask is mandatory for flags */
> > +       if (!tb[TCA_FLOWER_KEY_ENC_FLAGS_MASK]) {
> 
> if (NL_REQ_ATTR_CHECK(extack,...))
> 
> > +               NL_SET_ERR_MSG(extack, "missing enc_flags mask");
> > +               return -EINVAL;
> > +       }

right, I will change it in the v2.

[...]

> > @@ -1986,6 +2006,10 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
> >                 ret = fl_set_key_flags(tb, &key->control.flags,
> >                                        &mask->control.flags, extack);
> >
> > +       if (tb[TCA_FLOWER_KEY_ENC_FLAGS])
> 
> And here..
> 
> cheers,
> jamal
> 
> > +               ret = fl_set_key_enc_flags(tb, &key->enc_flags.flags,
> > +                                          &mask->enc_flags.flags, extack);
> > +
> >         return ret;

here I don't see any advantage in doing

if (!NL_REQ_ATTR_CHECK(extack, NULL, tb, TCA_FLOWER_KEY_ENC_FLAGS))
	ret = fl_set_key_enc_flags(tb, ... );

return ret;

the attribute is not mandatory, so a call to NL_SET_ERR_ATTR_MISS()
would do a useless/misleading assignment in extack->miss_type.

However, thanks for bringing the attention here :) At a second look,
this hunk introduces a bug: in case the parsing of TCA_FLOWER_KEY_FLAGS
fails, 'ret' is -EINVAL. If attributes TCA_FLOWER_KEY_ENC_FLAGS +
TCA_FLOWER_KEY_ENC_FLAGS_MASK are good to go, 'ret' will be overwritten
with 0 and flower will accept the rule... this is not intentional :)
will fix this in the v2.

-- 
davide