netdev - Re: [patch net-next v2 3/3] net: devlink: add WARN_ON to check return value of unregister_netdevice_notifier

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] [day] [month] [year] [list]

Date:   Thu, 10 Nov 2022 10:07:16 -0800
From:   Eric Dumazet <edumazet@...gle.com>
To:     Ido Schimmel <idosch@...sch.org>
Cc:     Jiri Pirko <jiri@...nulli.us>, Jakub Kicinski <kuba@...nel.org>,
        netdev@...r.kernel.org, davem@...emloft.net, pabeni@...hat.com,
        bigeasy@...utronix.de, imagedong@...cent.com, kuniyu@...zon.com,
        petrm@...dia.com
Subject: Re: [patch net-next v2 3/3] net: devlink: add WARN_ON to check return
 value of unregister_netdevice_notifier_net() call

On Thu, Nov 10, 2022 at 10:04 AM Ido Schimmel <idosch@...sch.org> wrote:
>
> On Thu, Nov 10, 2022 at 09:21:23AM -0800, Eric Dumazet wrote:
> > On Wed, Nov 9, 2022 at 11:54 PM Jiri Pirko <jiri@...nulli.us> wrote:
> > >
> > > Wed, Nov 09, 2022 at 10:45:36PM CET, kuba@...nel.org wrote:
> > > >On Wed, 9 Nov 2022 08:26:10 -0800 Eric Dumazet wrote:
> > > >> > On Tue, Nov 08, 2022 at 02:22:08PM +0100, Jiri Pirko wrote:
> > > >> > > From: Jiri Pirko <jiri@...dia.com>
> > > >> > >
> > > >> > > As the return value is not 0 only in case there is no such notifier
> > > >> > > block registered, add a WARN_ON() to yell about it.
> > > >> > >
> > > >> > > Suggested-by: Ido Schimmel <idosch@...sch.org>
> > > >> > > Signed-off-by: Jiri Pirko <jiri@...dia.com>
> > > >> >
> > > >> > Reviewed-by: Ido Schimmel <idosch@...dia.com>
> > > >>
> > > >> Please consider WARN_ON_ONCE(), or DEBUG_NET_WARN_ON_ONCE()
> > > >
> > > >Do you have any general guidance on when to pick WARN() vs WARN_ONCE()?
> > > >Or should we always prefer _ONCE() going forward?
> > >
> > > Good question. If so, it should be documented or spotted by checkpatch.
> > >
> > > >
> > > >Let me take the first 2 in, to lower the syzbot volume.
> >
> > Well, I am not sure what you call 'lower syzbot volume'
> >
> > netdevsim netdevsim0 netdevsim3 (unregistering): unset [1, 0] type 2
> > family 0 port 6081 - 0
> > ------------[ cut here ]------------
> > WARNING: CPU: 1 PID: 41 at net/core/devlink.c:10001
> > devl_port_unregister+0x2f6/0x390 net/core/devlink.c:10001
>
> Hi Eric,
>
> That's a different bug than the one fixed by this patchset. Should be
> fixed by this patch:
>
> https://patchwork.kernel.org/project/netdevbpf/patch/20221110085150.520800-1-idosch@nvidia.com/

OK, I will dup the new syzbot report, thanks.