[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240724195822.6f0388f1@kernel.org>
Date: Wed, 24 Jul 2024 19:58:22 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Eric Dumazet <edumazet@...gle.com>
Cc: syzbot <syzbot+90fd70a665715bd10258@...kaller.appspotmail.com>,
davem@...emloft.net, linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
pabeni@...hat.com, syzkaller-bugs@...glegroups.com
Subject: Re: [syzbot] [net?] INFO: task hung in nsim_create (2)
On Wed, 24 Jul 2024 20:28:09 +0200 Eric Dumazet wrote:
> The fact that netdevsim uses a sysfs file to add/delete netdevice
> seems unfortunate,
Let me not point fingers but agreed, and the original implementation
didn't :)
> because of the classical sysfs vs rtnl issue (see rtnl_trylock() calls
> in net/core/net-sysfs.c for reference)
Is that any different than binding a physical device via sysfs?
Modules have "bind" and "unbind" files, which lead to probe/remove
AFAIU completely inline.
Powered by blists - more mailing lists