Noboru Yamamoto wrote:
>
> Ever since the KEKB accelerators became operational, the VME/MXI/VXI
> connection has been a problem. Recently Mr. Akiyama of our group
> studied the problem carefully and identified the source of problem.
> After the discussion with National Instrument, the maker of VME-MXI
> interface board, we applied a modification to the VME-MXI board
> following the suggestion by NI.
This was interesting. I would like to add that APS has experienced a
different problem when trying to use a VME-MXI-1 card with a PowerPC CPU.
Our symptoms cannot be explained by a glitch on the DTACK line as
described, and we do not experience lock-up of the VMEbus in the way that
KEKB have, therefore ours is definitely a different problem with the
VME-MXI-1 module.
Our experience with an MVME2700 was a problem reading some of the VXIbus
configuration registers. When using a PowerPC the bits on the lowest 4
data lines D0-D3 were not reliably read, although an MVME167 in the same
crate showed the registers to be completely stable. This only affected a
very few registers, the most important being the VXIbus Status Register
VSR, which has a self-test Pass/Fail bit in bit 2. By telling the VXI
Resource Manager to ignore the value returned in this bit we are able to
use a VXI-MXI-1 board with an MVME2700, although we haven't done extensive
testing of this configuration.
We have not contacted National Instruments about this, and will probably
replace our VME-MXI-1 boards with VME-MXI-2 as we come to install PowerPC
CPU boards as the older VME-MXI-1 module is now obsolete. I would be
interested to hear whether KEKB have seen the problem I described above,
and if so did you make any changes to the drvEpvxi.c VXI Resource Manager
to avoid it or are you using a different Resource Manager anyway.
- Andrew
--
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away.
- Antoine de Saint-Exupery
- References:
- A VME-MXI problem in the KEKB control system. Noboru Yamamoto
- Navigate by Date:
- Prev:
A VME-MXI problem in the KEKB control system. Noboru Yamamoto
- Next:
undefined symbol: _rsrvCheckPut Graham Waters
- Index:
1994
1995
1996
1997
1998
1999
2000
<2001>
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
A VME-MXI problem in the KEKB control system. Noboru Yamamoto
- Next:
undefined symbol: _rsrvCheckPut Graham Waters
- Index:
1994
1995
1996
1997
1998
1999
2000
<2001>
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|