On the subject of CAMAC serial drivers, I note the following:
1) the original CEBAF/TJNAF version supports multiple IOC's sharing a
single serial highway, and no LAM support. Single threaded operation
is guaranteed by disabling pre-emption during hardware access, and
by using the hardware mutex for multi-IOC operation. On hard errors
on the serial highway a short delay (~100 usec) is incurred.
Statistics are kept on the total number of operations, noX, and noQ.
2) the LANL version supports only a single IOC, but multiple serial
highways, and with LAM support. Single threaded operation is
guaranteed by the use of semaphores. On hard errors on the serial
highway, a full 1 second delay is seen by the scan task processing
the record if the expected interrupt is not delivered! We saw this
problem this week -- the IOC appears to run 10,000 times slower than
normal :) Different statistics are kept, with some errors being
tracked on a slot by slot basis, and other statistics kept by the
original version no longer tracked. A few useful convenience routines
are included.
CEBAF/TJNAF is working on merging these two versions. The choice between
semaphores and pre-emption disable will be made on the basis of performance,
although I suppose it could be a compile time switch for those who dislike
disabling pre-emption.
Chip
- References:
- [Summary] CAMAC Serial Driver Noboru Yamamoto
- Navigate by Date:
- Prev:
[Summary] CAMAC Serial Driver Noboru Yamamoto
- Next:
Re: [Summary] CAMAC Serial Driver Eric Bjorklund, NPSM
- 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:
[Summary] CAMAC Serial Driver Noboru Yamamoto
- Next:
Re: [Summary] CAMAC Serial Driver Eric Bjorklund, NPSM
- 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
|