Re: [PATCH] ptp_clock: future-proofing drivers against PTP subsystem becoming optional
From: Nicolas PitreDate: Tue, 20 Sep 2016 19:25:58 -0400 (EDT) > > Drivers must be ready to accept NULL from ptp_clock_register() if the > PTP clock subsystem is configured out. > > This patch documents that and ensures that all drivers cope well > with a NULL return. > > Signed-off-by: Nicolas Pitre > Reviewed-by: Eugenia Emantayev > > --- > > Let's have the basics merged now and work out the actual Kconfig issue > separately. Richard, if you agree with this patch, I think this could go > via the netdev tree. Applied to net-next, thanks.
Re: [PATCH] ptp_clock: future-proofing drivers against PTP subsystem becoming optional
On 21/09/16 00:25, Nicolas Pitre wrote: > Drivers must be ready to accept NULL from ptp_clock_register() if the > PTP clock subsystem is configured out. > > This patch documents that and ensures that all drivers cope well > with a NULL return. > > Signed-off-by: Nicolas Pitre> Reviewed-by: Eugenia Emantayev [...] > diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c > index c771e0af4e..f105a170b4 100644 > --- a/drivers/net/ethernet/sfc/ptp.c > +++ b/drivers/net/ethernet/sfc/ptp.c > @@ -1269,13 +1269,13 @@ int efx_ptp_probe(struct efx_nic *efx, struct > efx_channel *channel) > if (IS_ERR(ptp->phc_clock)) { > rc = PTR_ERR(ptp->phc_clock); > goto fail3; > - } > - > - INIT_WORK(>pps_work, efx_ptp_pps_worker); > - ptp->pps_workwq = create_singlethread_workqueue("sfc_pps"); > - if (!ptp->pps_workwq) { > - rc = -ENOMEM; > - goto fail4; > + } else if (ptp->phc_clock) { > + INIT_WORK(>pps_work, efx_ptp_pps_worker); > + ptp->pps_workwq = > create_singlethread_workqueue("sfc_pps"); > + if (!ptp->pps_workwq) { > + rc = -ENOMEM; > + goto fail4; > + } > } > } > ptp->nic_ts_enabled = false; For the sfc change: Acked-by: Edward Cree
Re: [PATCH] ptp_clock: future-proofing drivers against PTP subsystem becoming optional
On Tue, Sep 20, 2016 at 07:25:58PM -0400, Nicolas Pitre wrote: > > Drivers must be ready to accept NULL from ptp_clock_register() if the > PTP clock subsystem is configured out. > > This patch documents that and ensures that all drivers cope well > with a NULL return. > > Signed-off-by: Nicolas Pitre> Reviewed-by: Eugenia Emantayev > > --- > > Let's have the basics merged now and work out the actual Kconfig issue > separately. Richard, if you agree with this patch, I think this could go > via the netdev tree. It looks ok to me. Acked-by: Richard Cochran