netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Daniel Borkmann <daniel@iogearbox.net>
To: Cong Wang <cwang@twopensource.com>
Cc: David Miller <davem@davemloft.net>,
	Alexei Starovoitov <ast@plumgrid.com>,
	Jamal Hadi Salim <jhs@mojatatu.com>,
	netdev <netdev@vger.kernel.org>
Subject: Re: [PATCH net] net: sched: fix refcount imbalance in actions
Date: Thu, 30 Jul 2015 02:55:39 +0200	[thread overview]
Message-ID: <55B9760B.10004@iogearbox.net> (raw)
In-Reply-To: <CAHA+R7MVFhv1_J=0QE6zjCiVtCCKsaU=WgX6eH-u3op7GLZW5w@mail.gmail.com>

On 07/30/2015 02:33 AM, Cong Wang wrote:
...
> diff --git a/net/sched/act_api.c b/net/sched/act_api.c
> index af427a3..bd63a39 100644
> --- a/net/sched/act_api.c
> +++ b/net/sched/act_api.c
> @@ -53,8 +53,11 @@ int tcf_hash_release(struct tc_action *a, int bind)
>          if (p) {
>                  if (bind)
>                          p->tcfc_bindcnt--;
> -               else if (p->tcfc_bindcnt > 0)
> -                       return -EPERM;
> +               else {
> +                       if (p->tcfc_bindcnt > 0)
> +                               return -EPERM;
> +                       return ret;
> +               }

Hm, so this seems not correct: if we only ever increase tcfc_refcnt
when there's bind=1, and only ever decrease when bind=1, then we
will never free the action as we do start out from ref=1 in case
it has been added without initial binding, if I see this correctly.

>                  p->tcfc_refcnt--;
>                  if (p->tcfc_bindcnt <= 0 && p->tcfc_refcnt <= 0) {
> @@ -214,9 +217,10 @@ int tcf_hash_check(u32 index, struct tc_action
> *a, int bind)
>          struct tcf_hashinfo *hinfo = a->ops->hinfo;
>          struct tcf_common *p = NULL;
>          if (index && (p = tcf_hash_lookup(index, hinfo)) != NULL) {
> -               if (bind)
> +               if (bind) {
>                          p->tcfc_bindcnt++;
> -               p->tcfc_refcnt++;
> +                       p->tcfc_refcnt++;
> +               }
>                  a->priv = p;
>                  return 1;
>          }
> diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
> index a42a3b2..2685450 100644
> --- a/net/sched/act_mirred.c
> +++ b/net/sched/act_mirred.c
> @@ -98,6 +98,8 @@ static int tcf_mirred_init(struct net *net, struct
> nlattr *nla,
>                          return ret;
>                  ret = ACT_P_CREATED;
>          } else {
> +               if (bind)
> +                       return 0;
>                  if (!ovr) {
>                          tcf_hash_release(a, bind);
>                          return -EEXIST;
>

  reply	other threads:[~2015-07-30  0:55 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-29 21:35 [PATCH net] net: sched: fix refcount imbalance in actions Daniel Borkmann
2015-07-30  0:33 ` Cong Wang
2015-07-30  0:55   ` Daniel Borkmann [this message]
2015-07-30 18:48     ` Cong Wang
2015-07-30 20:01       ` Daniel Borkmann
2015-07-30 21:21 ` David Miller

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=55B9760B.10004@iogearbox.net \
    --to=daniel@iogearbox.net \
    --cc=ast@plumgrid.com \
    --cc=cwang@twopensource.com \
    --cc=davem@davemloft.net \
    --cc=jhs@mojatatu.com \
    --cc=netdev@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).