netdev - Re: [PATCH RFC net-next 2/3] Expose tc block ports to the datapath

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]

Date: Wed, 16 Aug 2023 07:06:31 -0400
From: Jamal Hadi Salim <jhs@...atatu.com>
To: Stephen Hemminger <stephen@...workplumber.org>
Cc: jiri@...nulli.us, xiyou.wangcong@...il.com, netdev@...r.kernel.org, 
	vladbu@...dia.com, mleitner@...hat.com, Victor Nogueira <victor@...atatu.com>, 
	Pedro Tammela <pctammela@...atatu.com>
Subject: Re: [PATCH RFC net-next 2/3] Expose tc block ports to the datapath

On Tue, Aug 15, 2023 at 1:52 PM Stephen Hemminger
<stephen@...workplumber.org> wrote:
>
> On Tue, 15 Aug 2023 12:25:29 -0400
> Jamal Hadi Salim <jhs@...atatu.com> wrote:
>
> > +struct tcf_block *tcf_block_lookup(struct net *net, u32 block_index)
> >  {
> >       struct tcf_net *tn = net_generic(net, tcf_net_id);
> >
> >       return idr_find(&tn->idr, block_index);
> >  }
> > +EXPORT_SYMBOL(tcf_block_lookup)
>
> Use EXPORT_SYMBOL_GPL?

Sure.

On Wed, Aug 16, 2023 at 5:04 AM Simon Horman <horms@...nel.org> wrote:
>
On Wed, Aug 16, 2023 at 5:04 AM Simon Horman <horms@...nel.org> wrote:
>
> On Tue, Aug 15, 2023 at 12:25:29PM -0400, Jamal Hadi Salim wrote:
> > The datapath can now find the block of the port in which the packet arrived at.
> > It can then use it for various activities.
> >
> > In the next patch we show a simple action that multicast to all ports except for
> > the port in which the packet arrived on.
> >
> > Co-developed-by: Victor Nogueira <victor@...atatu.com>
> > Signed-off-by: Victor Nogueira <victor@...atatu.com>
> > Co-developed-by: Pedro Tammela <pctammela@...atatu.com>
> > Signed-off-by: Pedro Tammela <pctammela@...atatu.com>
> > Signed-off-by: Jamal Hadi Salim <jhs@...atatu.com>
>
> ...
>
> > diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
> > index a976792ef02f..be4555714519 100644
> > --- a/net/sched/cls_api.c
> > +++ b/net/sched/cls_api.c
>
> ...
>
> > @@ -1737,9 +1738,12 @@ int tcf_classify(struct sk_buff *skb,
> >                const struct tcf_proto *tp,
> >                struct tcf_result *res, bool compat_mode)
> >  {
> > +     struct qdisc_skb_cb *qdisc_cb = qdisc_skb_cb(skb);
>
> Hi Jamal,
>
> Does the line above belong inside the condition immediately below?
> It seems potentially unused otherwise.

Indeed.
The Intel bot also complained about this. I guess we'll need an
additional patch and move up "last_executed_chain" variable which is
repeated twice. Then i can add the assignment on top of this.
Something like:
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 4af48f76f..5d9959381 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -1736,13 +1736,12 @@ int tcf_classify(struct sk_buff *skb,
                 const struct tcf_proto *tp,
                 struct tcf_result *res, bool compat_mode)
 {
-#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
        u32 last_executed_chain = 0;
-
+#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
        return __tcf_classify(skb, tp, tp, res, compat_mode, NULL, 0,
                              &last_executed_chain);
 #else
-       u32 last_executed_chain = tp ? tp->chain->index : 0;
+       last_executed_chain = tp ? tp->chain->index : 0;
        struct tcf_exts_miss_cookie_node *n = NULL;
        const struct tcf_proto *orig_tp = tp;
        struct tc_skb_ext *ext;

cheers,
jamal