Re: [PATCH net-next] netdevsim: Register and unregister devlink traps on probe/remove device

[Leon Romanovsky <leon@kernel.org>]

On Mon, Oct 25, 2021 at 04:19:07PM -0700, Edwin Peer wrote:
> On Sun, Oct 24, 2021 at 3:35 PM Ido Schimmel <idosch@idosch.org> wrote:
> 
> > On Sun, Oct 24, 2021 at 11:42:11AM +0300, Leon Romanovsky wrote:
> > > From: Leon Romanovsky <leonro@nvidia.com>
> > >
> > > Align netdevsim to be like all other physical devices that register and
> > > unregister devlink traps during their probe and removal respectively.
> >
> > No, this is incorrect. Out of the three drivers that support both reload
> > and traps, both netdevsim and mlxsw unregister the traps during reload.
> > Here is another report from syzkaller about mlxsw [1].
> >
> > Please revert both 22849b5ea595 ("devlink: Remove not-executed trap
> > policer notifications") and 8bbeed485823 ("devlink: Remove not-executed
> > trap group notifications").
> 
> Could we also revert 82465bec3e97 ("devlink: Delete reload
> enable/disable interface")? 

Absolutely not.

> This interface is needed because bnxt_en cannot reorder devlink last.
> If Leon had fully carried out the re-ordering in our driver he would
> have introduced a udev phys_port_name regression because of:
> 
> cda2cab0771 ("bnxt_en: Move devlink_register before registering netdev")
> 
> and:
> 
> ab178b058c4 ("bnxt: remove ndo_get_phys_port_name implementation")

devlink_register() doesn't do anything except performing as a barrier.

In a nutshell, latest devlink_register() implementation is better
implementation of previously existed "reload enable/disable" boolean.

You don't need to reorder whole devlink logic, just put a call to
devlink_register() in the place where you wanted to put your
devlink_reload_enable().

> 
> I think this went unnoticed for bnxt_en, because Michael had not yet
> posted our devlink reload patches, which presently rely on the reload
> enable/disable API. Absent horrible kludges in reload down/up which
> currently depends on the netdev, there doesn't appear to be a clean
> way to resolve the circular dependency without the interlocks this API
> provides.

You was supposed to update and retest your out-of-tree implementation
of devlink reload before posting it to the ML. However, if you use
devlink_*() API correctly, such dependency won't exist.

> 
> I imagine other subtle regressions are lying in wait.

Sorry, but we don't have crystal ball and can't guess what else is
broken in your out-of-tree driver.

Thanks

> 
> Regards,
> Edwin Peer