The three bars show the individual status bits for each of the BIT-3 memories. If things are working, only the "Always on" bit is on. Other bits portend ill. If there is a problem, see below.
for a dump of the MPG header or queues to standard I/O.
And at the very bottom, the status of the MPG/BIC link along with the current MPG request insertion index and MPG process index are displayed.
The memory can be examined from the BIC side. The example below shows a problem reading the 16 bit (second line should really be all 555555555):
(login to the BIC)
MCCDEV::[RON]>rlogin bic06pep00
(look at the MPG/BIC shared memory)
-> d 0xf0000000,11,4
f0000000: 00010400 00000000 00002560 40450540 *..........%`@E.@*
f0000010: 45454545 45454545 45454545 45454545 *EEEEEEEEEEEEEEEE*
f0000020: 014d05c5 454542c5 c5462147 *.M..EEB..F!G....*
value = 21 = 0x15
(deposit some 5s and Fs in the second row)
-> m 0xf0000010
f0000010: 4545-5555
f0000012: 4545-ffff
f0000014: 4545-^D
value = 1 = 0x1
(look again - oops, still bad)
-> d 0xf0000000,11,4
f0000000: 00010400 00000000 00002560 40450540 *..........%`@E.@*
f0000010: 4545efef 45454545 45454545 45454545 *EE..EEEEEEEEEEEE*
f0000020: 014d05c5 454542c5 c5462147 *.M..EEB..F!G....*
value = 21 = 0x15
-> logout
The memory can also be examined from the MPG side via the BIC panel or Network-> network diagnostics -> BIT3 Diagnostics. The file describing the layout of memory is REF_C_INC:MPG_PEP2INJ_TYPES.H. The problem above was caused by a bad cable, with a loose connection, by the way.
The BxBCM VXI crates are (HER) above and (LER) to the left of the BIC VME crate. The BIT-3 connections are short cables between the crates. If you look at the BIT-3 cards in the VXI crate, the local light should be almost steady and the remote light will blink each second (when the BIC fetches data).
The BIC BxBCM addresses are 0xF0010000 for the HER and 0xF0020000 for the LER. For a quick check, a counter changing by about 100 at 1Hz is at offset 6004: