[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YYgSzEHppKY3oYTb@unreal>
Date: Sun, 7 Nov 2021 19:54:20 +0200
From: Leon Romanovsky <leon@...nel.org>
To: Ido Schimmel <idosch@...sch.org>
Cc: Jakub Kicinski <kuba@...nel.org>, Jiri Pirko <jiri@...nulli.us>,
"David S . Miller" <davem@...emloft.net>,
Jiri Pirko <jiri@...dia.com>, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org, edwin.peer@...adcom.com
Subject: Re: [PATCH net-next] devlink: Require devlink lock during device
reload
On Sun, Nov 07, 2021 at 07:16:05PM +0200, Ido Schimmel wrote:
> On Mon, Nov 01, 2021 at 04:11:22PM -0700, Jakub Kicinski wrote:
> > On Mon, 1 Nov 2021 22:52:19 +0200 Ido Schimmel wrote:
> > > > >Signed-off-by: Leon Romanovsky <leonro@...dia.com>
> > > >
> > > > Looks fine to me.
> > > >
> > > > Reviewed-by: Jiri Pirko <jiri@...dia.com>
> > >
> > > Traces from mlxsw / netdevsim below:
> >
> > Thanks a lot for the testing Ido!
> >
> > Would you mind giving my RFC a spin as well on your syzbot machinery?
<...>
> >
> > (3) should we let drivers take refs on the devlink instance?
>
> I think it's fine mainly because I don't expect it to be used by too
> many drivers other than netdevsim which is somewhat special. Looking at
> the call sites of devlink_get() in netdevsim, it is only called from
> places (debugfs and trap workqueue) that shouldn't be present in real
> drivers.
Sorry, I'm obligated to ask. In which universe is it ok to create new
set of API that no real driver should use?
Thanks
Powered by blists - more mailing lists