linux-kernel - Re: [PATCH] net/ipv4: Fix circular deadlock in do_ip

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]

Message-ID: <CANn89iLxa6V7BZSzmj5A979M2ZObj-CcDD_izeKqtRCZj-+pmQ@mail.gmail.com>
Date: Sun, 22 Sep 2024 18:11:48 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: AnantaSrikar <srikarananta01@...il.com>
Cc: syzbot+e4c27043b9315839452d@...kaller.appspotmail.com, 
	alibuda@...ux.alibaba.com, davem@...emloft.net, dsahern@...nel.org, 
	dust.li@...ux.alibaba.com, kuba@...nel.org, linux-kernel@...r.kernel.org, 
	netdev@...r.kernel.org, pabeni@...hat.com, schnelle@...ux.ibm.com, 
	syzkaller-bugs@...glegroups.com, wenjia@...ux.ibm.com
Subject: Re: [PATCH] net/ipv4: Fix circular deadlock in do_ip_setsockop

On Wed, Sep 18, 2024 at 1:51 AM AnantaSrikar <srikarananta01@...il.com> wrote:
>
> Fixed the circular lock dependency reported by syzkaller.
>
> Signed-off-by: AnantaSrikar <srikarananta01@...il.com>
> Reported-by: syzbot+e4c27043b9315839452d@...kaller.appspotmail.com
> Closes: https://syzkaller.appspot.com/bug?extid=e4c27043b9315839452d
> Fixes: d2bafcf224f3 ("Merge tag 'cgroup-for-6.11-rc4-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup")
> ---
>  net/ipv4/ip_sockglue.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
> index cf377377b52d..a8f46d1ba62b 100644
> --- a/net/ipv4/ip_sockglue.c
> +++ b/net/ipv4/ip_sockglue.c
> @@ -1073,9 +1073,11 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
>         }
>
>         err = 0;
> +
> +       sockopt_lock_sock(sk);
> +
>         if (needs_rtnl)
>                 rtnl_lock();
> -       sockopt_lock_sock(sk);
>
>         switch (optname) {
>         case IP_OPTIONS:

I think you missed an earlier conversation about SMC being at fault here.

https://lore.kernel.org/netdev/CANn89iKcWmufo83xy-SwSrXYt6UpL2Pb+5pWuzyYjMva5F8bBQ@mail.gmail.com/