From: Cong Wang <cwang@twopensource.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Tejun Heo <tj@kernel.org>, David Miller <davem@davemloft.net>,
Tom Herbert <tom@herbertland.com>,
kafai@fb.com, kernel-team <kernel-team@fb.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
netdev <netdev@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Jiri Pirko <jiri@resnulli.us>,
Nicolas Dichtel <nicolas.dichtel@6wind.com>,
Thomas Graf <tgraf@suug.ch>, Scott Feldman <sfeldma@gmail.com>
Subject: Re: Possible netlink autobind regression
Date: Wed, 16 Sep 2015 22:02:00 -0700 [thread overview]
Message-ID: <CAHA+R7OdJyZiidHarKhp-U6R5fkHj=5KjugK=XoWj=q4D8krxw@mail.gmail.com> (raw)
In-Reply-To: <20150917034134.GA19327@gondor.apana.org.au>
On Wed, Sep 16, 2015 at 8:41 PM, Herbert Xu <herbert@gondor.apana.org.au> wrote:
> On Thu, Sep 17, 2015 at 11:08:45AM +0800, Herbert Xu wrote:
>>
>> Good catch! I think your explanation makes perfect sense. Linus
>> ran into this previously too after suspend-and-resume.
>
> Unfortunately you can't just postpone the setting of portid because
> once you pass it onto rhashtable the portid must never change while
> it's in custody.
>
> So what I've done is essentially revert my previous fix and instead
> add a new boolean "bound" to indicate whether the socket has been
> bound.
>
> ---8<---
> netlink: Fix autobind race condition that leads to zero port ID
>
> The commit c0bb07df7d981e4091432754e30c9c720e2c0c78 ("netlink:
> Reset portid after netlink_insert failure") introduced a race
> condition where if two threads tried to autobind the same socket
> one of them may end up with a zero port ID.
>
> This patch reverts that commit and instead fixes it by introducing
> a separte "bound" variable to indicate whether a socket has been
> bound.
>
> Fixes: c0bb07df7d98 ("netlink: Reset portid after netlink_insert failure")
> Reported-by: Tejun Heo <tj@kernel.org>
> Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
We saw similar soft lockup with the one Tejun reported, in our data
center.
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Just one comment below.
[...]
> @@ -1285,7 +1287,7 @@ static int netlink_release(struct socket *sock)
>
> skb_queue_purge(&sk->sk_write_queue);
>
> - if (nlk->portid) {
> + if (nlk->bound) {
> struct netlink_notify n = {
> .net = sock_net(sk),
> .protocol = sk->sk_protocol,
This part doesn't look correct, seems it is checking if this is a kernel
netlink socket rather than if it is bound. But I am not sure...
Other than this, looks good to me:
Reviewed-by: Cong Wang <cwang@twopensource.com>
next prev parent reply other threads:[~2015-09-17 5:02 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-17 2:29 Possible netlink autobind regression Tejun Heo
2015-09-17 3:08 ` Herbert Xu
2015-09-17 3:41 ` Herbert Xu
2015-09-17 5:02 ` Cong Wang [this message]
2015-09-17 5:15 ` Herbert Xu
2015-09-17 11:25 ` Thomas Graf
2015-09-17 11:30 ` Tejun Heo
2015-09-18 6:36 ` [PATCH v3] netlink: Fix autobind race condition that leads to zero port ID Herbert Xu
2015-09-18 11:16 ` [PATCH v4] " Herbert Xu
2015-09-21 5:55 ` David Miller
2015-09-21 6:06 ` Herbert Xu
2015-09-21 6:11 ` David Miller
2015-09-21 13:34 ` netlink: Replace rhash_portid with bound Herbert Xu
2015-09-21 18:20 ` Tejun Heo
2015-09-22 3:38 ` [PATCH v2] " Herbert Xu
2015-09-22 16:10 ` Tejun Heo
2015-09-22 18:42 ` Linus Torvalds
2015-09-22 18:53 ` Tejun Heo
2015-09-22 19:28 ` Linus Torvalds
2015-09-22 19:50 ` Tejun Heo
2015-09-22 20:03 ` Linus Torvalds
2015-09-22 20:36 ` Bjørn Mork
2015-09-22 21:04 ` Linus Torvalds
2015-09-23 6:13 ` Herbert Xu
2015-09-23 15:54 ` Tejun Heo
2015-09-24 2:30 ` Herbert Xu
2015-09-24 2:46 ` Tejun Heo
2015-09-24 2:54 ` Herbert Xu
2015-09-24 3:06 ` Tejun Heo
2015-09-24 3:21 ` Herbert Xu
2015-09-24 3:29 ` Tejun Heo
2015-09-24 3:31 ` Herbert Xu
2015-09-24 3:41 ` Tejun Heo
2015-09-24 3:42 ` Herbert Xu
2015-09-24 3:43 ` Tejun Heo
2015-09-24 3:44 ` Herbert Xu
2015-09-24 19:11 ` David Miller
2015-09-24 20:05 ` Tejun Heo
2015-09-25 1:43 ` netlink: Add barrier to netlink_connect for theoretical case Herbert Xu
2015-09-25 3:24 ` Linus Torvalds
2015-09-25 3:39 ` Herbert Xu
2015-09-25 15:09 ` Tejun Heo
2015-09-25 15:01 ` Tejun Heo
2015-09-26 13:16 ` netlink: Add netlink_bound helper and use it in netlink_getname Herbert Xu
2015-09-26 18:09 ` Tejun Heo
2015-09-26 19:41 ` Herbert Xu
2015-09-26 19:45 ` Tejun Heo
2015-09-26 19:49 ` Herbert Xu
2015-09-26 19:52 ` Tejun Heo
2015-09-26 19:55 ` Herbert Xu
2015-09-26 20:05 ` Tejun Heo
2015-09-26 20:10 ` Herbert Xu
2015-09-26 20:17 ` Tejun Heo
2015-09-21 20:52 ` [PATCH] netlink: Replace rhash_portid with load_acquire protected boolean Tejun Heo
2015-09-18 13:37 ` [PATCH v3] netlink: Fix autobind race condition that leads to zero port ID Tejun Heo
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='CAHA+R7OdJyZiidHarKhp-U6R5fkHj=5KjugK=XoWj=q4D8krxw@mail.gmail.com' \
--to=cwang@twopensource.com \
--cc=davem@davemloft.net \
--cc=herbert@gondor.apana.org.au \
--cc=jiri@resnulli.us \
--cc=kafai@fb.com \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=nicolas.dichtel@6wind.com \
--cc=sfeldma@gmail.com \
--cc=tgraf@suug.ch \
--cc=tj@kernel.org \
--cc=tom@herbertland.com \
--cc=torvalds@linux-foundation.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).