>
> That concern is where events involve multiple supermodules and it
> is necessary to correlate WFD stops across supermodule or uVAX
> boundries. Is there a better way to do this than using the
> cSPAM stop times? If so, I'd like to hear about it.
It is true that the CSPAM does adjacent SMs within a microvax
at the same time (at least withing the .1usec Stop Master resolution).
Across microvaxes there can be a few microseconds delay (you can
confirm this by comparing the SM+1/SM-1 times latched for tohm events
by the stop master to the TOHM trigger times relative to CSPAM times
AND/OR by looking at relative trigger times latched by the stop master
for the ERP, CSPAM, and LIP).
Unfortunately the answer to your question is that multi-SM
timing is only designed into the stop master for the tohm through the
"SM+1" and "SM-1" inputs. These inputs take the "OR" of all TOHM
triggers for adjacent supermodules. I guess Rongzhi felt that the
inter-ERP would take care of the fast stuff.
You CAN get timing on 2 SMs within a microvax for all triggers
into the Stop Master because of the cross-Stop scheme. One of the
spare latches on each Stop Master records the time that a stop was
sent by the other Stop Master on that microvax. You can work
backwards and resolve timings of up to 2ms with a nominal .1usec
resolution.
-Chris