All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <edumazet@google.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: syzbot <syzbot+1e9af9185d8850e2c2fa@syzkaller.appspotmail.com>,
	davem@davemloft.net, kuba@kernel.org,
	linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
	pabeni@redhat.com, steffen.klassert@secunet.com,
	syzkaller-bugs@googlegroups.com
Subject: Re: [v2 PATCH] af_key: Fix send_acquire race with pfkey_register
Date: Sun, 23 Oct 2022 23:31:00 -0700	[thread overview]
Message-ID: <CANn89i+JY3PA_p5t3FrCeO+tAo3YuYOtnkeOeyYvBcqKhpgNZQ@mail.gmail.com> (raw)
In-Reply-To: <Y1YrVGP+5TP7V1/R@gondor.apana.org.au>

On Sun, Oct 23, 2022 at 11:06 PM Herbert Xu <herbert@gondor.apana.org.au> wrote:
>
> On Sun, Oct 23, 2022 at 10:21:05PM -0700, Eric Dumazet wrote:
> >
> > Are you sure we can sleep in mutex_lock() ?
> >
> > Use of GFP_ATOMIC would suggest otherwise :/
>
> Good point.  Acquires are triggered from the network stack so
> it may be in BH context.
>
> ---8<---
> With name space support, it is possible for a pfkey_register to
> occur in the middle of a send_acquire, thus changing the number
> of supported algorithms.
>
> This can be fixed by taking a lock to make it single-threaded
> again.  As this lock can be taken from both thread and softirq
> contexts, we need to take the necessary precausions with disabling
> BH and make it a spin lock.
>
> Reported-by: syzbot+1e9af9185d8850e2c2fa@syzkaller.appspotmail.com
> Fixes: 283bc9f35bbb ("xfrm: Namespacify xfrm state/policy locks")
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
>
> diff --git a/net/key/af_key.c b/net/key/af_key.c
> index c85df5b958d2..4e0d21e2045e 100644
> --- a/net/key/af_key.c
> +++ b/net/key/af_key.c
> @@ -23,6 +23,7 @@
>  #include <linux/proc_fs.h>
>  #include <linux/init.h>
>  #include <linux/slab.h>
> +#include <linux/spinlock.h>
>  #include <net/net_namespace.h>
>  #include <net/netns/generic.h>
>  #include <net/xfrm.h>
> @@ -39,6 +40,7 @@ struct netns_pfkey {
>         atomic_t socks_nr;
>  };
>  static DEFINE_MUTEX(pfkey_mutex);
> +static DEFINE_SPINLOCK(pfkey_alg_lock);
>
>  #define DUMMY_MARK 0
>  static const struct xfrm_mark dummy_mark = {0, 0};
> @@ -1697,11 +1699,11 @@ static int pfkey_register(struct sock *sk, struct sk_buff *skb, const struct sad
>                 pfk->registered |= (1<<hdr->sadb_msg_satype);
>         }
>
> -       mutex_lock(&pfkey_mutex);
> +       spin_lock_bh(&pfkey_alg_lock);
>         xfrm_probe_algs();
>
>         supp_skb = compose_sadb_supported(hdr, GFP_KERNEL | __GFP_ZERO);

s/GFP_KERNEL/GFP_ATOMIC/

> -       mutex_unlock(&pfkey_mutex);
> +       spin_unlock_bh(&pfkey_alg_lock);
>
>         if (!supp_skb) {
>                 if (hdr->sadb_msg_satype != SADB_SATYPE_UNSPEC)
> @@ -3160,6 +3162,7 @@ static int pfkey_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *t, struct
>                 (sockaddr_size * 2) +
>                 sizeof(struct sadb_x_policy);
>
> +       spin_lock_bh(&pfkey_alg_lock);
>         if (x->id.proto == IPPROTO_AH)
>                 size += count_ah_combs(t);
>         else if (x->id.proto == IPPROTO_ESP)
> @@ -3171,8 +3174,10 @@ static int pfkey_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *t, struct
>         }
>
>         skb =  alloc_skb(size + 16, GFP_ATOMIC);
> -       if (skb == NULL)
> +       if (skb == NULL) {
> +               spin_unlock_bh(&pfkey_alg_lock);
>                 return -ENOMEM;
> +       }
>
>         hdr = skb_put(skb, sizeof(struct sadb_msg));
>         hdr->sadb_msg_version = PF_KEY_V2;
> @@ -3228,6 +3233,7 @@ static int pfkey_send_acquire(struct xfrm_state *x, struct xfrm_tmpl *t, struct
>                 dump_ah_combs(skb, t);
>         else if (x->id.proto == IPPROTO_ESP)
>                 dump_esp_combs(skb, t);
> +       spin_unlock_bh(&pfkey_alg_lock);
>
>         /* security context */
>         if (xfrm_ctx) {
> --
> Email: Herbert Xu <herbert@gondor.apana.org.au>
> Home Page: http://gondor.apana.org.au/~herbert/
> PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

  reply	other threads:[~2022-10-24  6:31 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-24  2:37 [syzbot] kernel BUG in warn_crc32c_csum_combine syzbot
2022-10-24  2:49 ` Eric Dumazet
2022-10-24  5:01   ` Herbert Xu
2022-10-24  5:10 ` [PATCH] af_key: Fix send_acquire race with pfkey_register Herbert Xu
2022-10-24  5:21   ` Eric Dumazet
2022-10-24  6:06     ` [v2 PATCH] " Herbert Xu
2022-10-24  6:31       ` Eric Dumazet [this message]
2022-10-25  5:46         ` Herbert Xu
2022-10-24  7:20       ` Sabrina Dubroca
2022-10-25  6:06         ` [v3 " Herbert Xu
2022-10-26 13:38           ` Sabrina Dubroca
2022-10-27  3:42             ` Herbert Xu
2022-10-27  3:45               ` Eric Dumazet
2022-10-27  3:55                 ` Herbert Xu
2022-10-28 11:26                 ` Steffen Klassert

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=CANn89i+JY3PA_p5t3FrCeO+tAo3YuYOtnkeOeyYvBcqKhpgNZQ@mail.gmail.com \
    --to=edumazet@google.com \
    --cc=davem@davemloft.net \
    --cc=herbert@gondor.apana.org.au \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=steffen.klassert@secunet.com \
    --cc=syzbot+1e9af9185d8850e2c2fa@syzkaller.appspotmail.com \
    --cc=syzkaller-bugs@googlegroups.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.