Netdev List
 help / color / mirror / Atom feed
From: Jiayuan Chen <jiayuan.chen@linux.dev>
To: Eric Dumazet <edumazet@google.com>,
	Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: netdev@vger.kernel.org,
	syzbot+d3f912a2e30e2079a512@syzkaller.appspotmail.com,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Simon Horman <horms@kernel.org>,
	Christian Brauner <brauner@kernel.org>,
	Kuniyuki Iwashima <kuniyu@google.com>, Jan Kara <jack@suse.cz>,
	Qiliang Yuan <realwujing@gmail.com>,
	Jeff Layton <jlayton@kernel.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH net] net: fix refcnt_tracker use-after-free on netns creation failure
Date: Thu, 27 Aug 2026 15:05:09 +0800	[thread overview]
Message-ID: <512af244-e4af-4d85-9be6-b3bd8141e092@linux.dev> (raw)
In-Reply-To: <CANn89iKK5jxVa34F7QMPjp5wormMQrZgoK4THJCqVp76Tw1NNA@mail.gmail.com>


在 8/27/26 2:56 PM, Eric Dumazet 写道:
> On Thu, Aug 27, 2026 at 7:22 AM Jiayuan Chen <jiayuan.chen@linux.dev> wrote:
>> The copy_net_ns() error path tears the net down via net_passive_dec(),
>> which only exits notrefcnt_tracker. refcnt_tracker is exited in
>> __put_net(), which this path never reaches, so it is left alive.
>>
>> Since ref_tracker_dir_init() started registering a debugfs file per
>> dir, that file must be dropped by ref_tracker_dir_exit(). The missing
>> exit leaves the file pointing at the freed net, so reading it walks
>> net->refcnt_tracker.lock and hits a use-after-free.
>>
>> The missing exit has been there all along, but before the debugfs file
>> refcnt_tracker was just an in-memory ledger and empty on this path, so
>> skipping it leaked nothing and did no harm. Only the commit below, which
>> gives every dir a debugfs file, turned it into a use-after-free, so
>> Fixes points there.
>>
>> Exit refcnt_tracker on the failure path.
>>
>> Fixes: 65b584f53611 ("ref_tracker: automatically register a file in debugfs for a ref_tracker_dir")
>> Reported-by: syzbot+d3f912a2e30e2079a512@syzkaller.appspotmail.com
> Missing `Closes:` tag makes analysis cumbersome.
>
>> Signed-off-by: Jiayuan Chen <jiayuan.chen@linux.dev>
>> ---
>>   net/core/net_namespace.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>
>> diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
>> index a91d2b58aadd..0c3f169d9afb 100644
>> --- a/net/core/net_namespace.c
>> +++ b/net/core/net_namespace.c
>> @@ -591,6 +591,10 @@ struct net *copy_net_ns(u64 flags,
>>                  key_remove_domain(net->key_domain);
>>   #endif
>>                  put_user_ns(user_ns);
>> +               /* setup_net() failure is not routed through __put_net(), so the
>> +                * refcnt_tracker (and its debugfs file) has to be released here.
>> +                */
>> +               ref_tracker_dir_exit(&net->refcnt_tracker);
>>                  net_passive_dec(net);
>>   dec_ucounts:
>>                  dec_net_namespaces(ucounts);
> I am a bit confused. I thought this was already discussed and fixed by
> Tetsuo Handa?
>
> f85dc137aabf ("net: add missing ref_tracker_dir_exit() to net_passive_dec()")
> should be fixing this?


Thanks Eric, now I find it it in net tree. I should anylyze it base net 
tree not net-next.

Please ignore my patch.


pw-bot:cr


>
> Tetsuo also followed up with 3220b62fbb8a  ("net: fix a resource leak
> in copy_net_ns() error handling path").

  reply	other threads:[~2026-08-27  7:05 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-27  5:22 [PATCH net] net: fix refcnt_tracker use-after-free on netns creation failure Jiayuan Chen
2026-08-27  6:56 ` Eric Dumazet
2026-08-27  7:05   ` Jiayuan Chen [this message]
2026-08-27  7:06   ` Jiayuan Chen
2026-08-30 10:57   ` Tetsuo Handa

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=512af244-e4af-4d85-9be6-b3bd8141e092@linux.dev \
    --to=jiayuan.chen@linux.dev \
    --cc=brauner@kernel.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=horms@kernel.org \
    --cc=jack@suse.cz \
    --cc=jlayton@kernel.org \
    --cc=kuba@kernel.org \
    --cc=kuniyu@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=penguin-kernel@i-love.sakura.ne.jp \
    --cc=realwujing@gmail.com \
    --cc=syzbot+d3f912a2e30e2079a512@syzkaller.appspotmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox