On Wed, 2023-04-12 at 20:33 -0700, Richard Cochran wrote: > On Tue, Apr 04, 2023 at 12:45:46AM +0200, Richard Cochran wrote: > > On Fri, Mar 31, 2023 at 12:06:48AM -0700, Jakub Kicinski wrote: > > > On Thu, 30 Mar 2023 22:56:11 +0200 Johannes Berg wrote: > > > > * hardware timestamping support for some devices/firwmares > > > > > > Was Richard CCed on those patches? > > > Would have been good to see his acks there. > > > > > > Adding him here in case he wants to take a look 'post factum'. > > > > Timestamping on wifi is something I've spent a fair amount of time > > thinking about. I'll take a look but not this week as I am on > > vacation until April 10. > > This so-called "hardware timestamping support" appears to be a big > nothing burger. > > I took a quick look at the PR, and AFAICT this is only internal code > for iwlwifi/mvm, and the code is dead code, meaning not reachable by > user space for any practical purpose. > > Like in drivers/net/wireless/intel/iwlwifi/mvm/ops.c ... > > RX_HANDLER_GRP(LEGACY_GROUP, > WNM_80211V_TIMING_MEASUREMENT_NOTIFICATION, > iwl_mvm_time_sync_msmt_event, RX_HANDLER_SYNC, > struct iwl_time_msmt_notify), > RX_HANDLER_GRP(LEGACY_GROUP, > WNM_80211V_TIMING_MEASUREMENT_CONFIRM_NOTIFICATION, > iwl_mvm_time_sync_msmt_confirm_event, RX_HANDLER_SYNC, > struct iwl_time_msmt_cfm_notify), > > These aren't useful, or are they? > > I am aware of FTM etc, but I see no attempt to do anything with it in > this code. > > Thanks, > Richard Just a few clarifications. These two notifications are internal to iwlwifi, sent by the firmware to the driver. Then, the timestamps are added to the rx/tx status via mac80211 api. Actually, we already have a functional implementation of ptp4l over wifi using this driver support.
