On Wed, Jan 15, 2025 at 07:23:23AM +0000, Soumyadeep Hore wrote:
> From: Jun Zhang <xuejun.zh...@intel.com>
>
> When there are at least two VFs on a single adapter and both are used in
> the same VM, each of them will register its own PTP clock. However, every
> E810/E822 adapter has only one PHC clock that we use. In order to register
> only one PTP clock, VFs need to identify HW and make sure they come from
> the same board.
>
> This patch adds a @hardware_clock_id that helps in this identification.
> VFs from the same board should receive the same @hardware_clock_id and VFs
> from two different boards should receive different IDs.
>
> Signed-off-by: Jun Zhang <xuejun.zh...@intel.com>
> Signed-off-by: Soumyadeep Hore <soumyadeep.h...@intel.com>
> ---
> drivers/common/iavf/virtchnl.h | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/common/iavf/virtchnl.h b/drivers/common/iavf/virtchnl.h
> index c0d52ddd32..f4ca2cc9f0 100644
> --- a/drivers/common/iavf/virtchnl.h
> +++ b/drivers/common/iavf/virtchnl.h
> @@ -2224,6 +2224,7 @@ VIRTCHNL_CHECK_STRUCT_LEN(12, virtchnl_quanta_cfg);
> #define VIRTCHNL_1588_PTP_CAP_PHC_REGS BIT(4)
> #define VIRTCHNL_1588_PTP_CAP_SYNCE BIT(6)
> #define VIRTCHNL_1588_PTP_CAP_GNSS BIT(7)
> +#define VIRTCHNL_1588_PTP_CAP_HARDWARE_CLOCK_ID BIT(8)
>
> struct virtchnl_phc_regs {
> u32 clock_hi;
> @@ -2248,7 +2249,8 @@ struct virtchnl_ptp_caps {
> u8 n_per_out;
> u8 n_pins;
> u8 tx_tstamp_format;
> - u8 rsvd[11];
> + u8 hardware_clock_id;
> + u8 rsvd[10];
> };
>
These definitions look fine to me, but on the other hand it seems strange
adding them without any user of them. Can we defer this patch until such
time as we have a patch with code which uses these fields?
/Bruce
