doc: documented new clocksource module

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

doc/wrs-developer-manual.in

@@ -748,6 +748,15 @@ Currently, the package includes the following modules:
 @item @i{wr_rtu.ko}: the routing-table interface between the
       switching core and the associated user-space daemon.
 @item @i{wr_pstats.ko}: exports per-port statistics to /proc/sys.
+
+@item @i{wr_clocksource.ko}: uses WR time as a source for system time.
+This driver uses the @sc{wr} counters to make host time flow at the
+right speed. The system time is moved to the current value, with an
+error not bigger than one second, as soon as @i{ppsi} synchronizes,
+and the clocksource grants it won't move ever after, keeping the same
+offset. This is considered acceptable, because system time is only
+used for logging.
+
 @item @i{at91_softpwm.ko}: a driver that generates a PWM signal for the fan.
 @end itemize
 

doc/wrs-todo.in

@@ -159,16 +159,6 @@ like increasing @t{NR_IRQ} and exporting symbols for externally-loaded
 the counters are. This would make the @sc{snmp} code generic, so
 no change there would be needed if and when the counters change.
 
-@item We need RTC driver to use WR time from FPGA as a system clock in Linux
-running on WR Switch (or to synchronize the system clock). Setting initial time
-from NTP is good for Grand Master in the network, but then we should keep all
-system clocks in sync with WR timescale. It will become important when people
-start gathering logs and SNMP info on a central server and would like to
-correlate logs if something bad happens. Let's keep in mind that currently
-unplugging fiber from the Slave port resets WR counters (i.e. current time
-00:00 01/01/1970). System clock should not follow this, we need to keep
-reporting failures timestamped with correct time.
-
 @end itemize
 
 @c ##########################################################################