On 07/06/2017 09:45 PM, Vivien Didelot wrote: > Hi Arkadi, > > Florian Fainelli <f.faine...@gmail.com> writes: > >> On 07/05/2017 08:36 AM, Arkadi Sharshevsky wrote: >>> This workqueue will be used for FDB add/del processing. It should >>> be destroyed after all devices unregistered successfully. >>> >>> Signed-off-by: Arkadi Sharshevsky <arka...@mellanox.com> >>> --- >>> include/net/dsa.h | 1 + >>> net/dsa/dsa.c | 13 +++++++++++++ >>> 2 files changed, 14 insertions(+) >>> >>> diff --git a/include/net/dsa.h b/include/net/dsa.h >>> index f054d41..4835b0e 100644 >>> --- a/include/net/dsa.h >>> +++ b/include/net/dsa.h >>> @@ -451,6 +451,7 @@ void unregister_switch_driver(struct dsa_switch_driver >>> *type); >>> struct mii_bus *dsa_host_dev_to_mii_bus(struct device *dev); >>> >>> struct net_device *dsa_dev_to_net_device(struct device *dev); >>> +bool dsa_schedule_work(struct work_struct *work); >> >> I'd move this to dsa_priv.h in net/dsa/ because it is not supposed to be >> used by DSA drivers. >> >> You may also consider squashing this into the next patch since in itself >> it's not used just yet. > > I agree with Florian on this. Even though it'll make the next patch a > bit bigger, it'll make it easier to understand the whole workqueue > processing logic. > > Thanks, > > Vivien >
Ok no problem, will squash. Thanks
