Experimental Physics and
Industrial Control System

Folks,

In a new EPICS application I am talking to several GPIB devices each at 10 Hz, and I am getting swamped with "Bad VME interrupt 0" error messages.  The problem happens when using the SBS (Greenspring) IP-488 module, mpfGpib1-4, VIPC616 carrier, and an MVME2700 PPC CPU.  Andrew Johnson suggested removing the logMsg() call in target/config/mv2700/universe.c that generates the error.  While this would work, I am getting nearly 100 messages per second, and I worry that all these spurious interrupts could be hurting performance.

For testing purposes I created
- testApp that loads my normal libraries (mpf, mca, etc.)
- A single generic GPIB record is loaded into the crate, no other records
- No other drivers or device support are initialized from the startup script.
- My VIPC616 has the _01 modifications with new PAL/GAL chips and the wire-wrapped jumpers.
- The only IP module in the VIPC616 was the IP-488.
- A CA client that used the generic GPIB record to do lots of GPIB I/O to a Tektronix scope

Here is what I have found.

The "Bad VME interrupt" error messages only occur if the IP-488 module is plugged into IP slot B.  There are no errors if the IP-488 is plugged into slots A, C or D.

If the module is plugged into slot B there are no errors unless there are also several other VME modules plugged into the backplane.  These modules are all to the right of the VIPC616 in the backplane, so they should not be involved in interrupt grant to the VIPC616.  None of these other modules are actually generating any interrupts nor is there any VME bus activity directed to them.  I found that the errors did not seem to depend on what particular VME modules were plugged in.  Plugging in an OMS58, a Joerger scaler, and an SIS scaler would cause the errors to appear.  Unplugging any one of the modules would cause the errors to stop!

The bottom line is that this is a very strange error.  It only happens if the IP-488 is in slot B and it only happens if there are other modules on the VME bus.  These other modules can be to the right of the VIPC616, are not generating interrupts or being talked to, and it does not seem to matter what type of modules they are.

Has anyone else seen problems with an IP-488 or other interrupt-generating IP module in slot B of a VIPC616?

Thanks,
Mark

Replies:

Re: Error messages wuth IP-488, VIPC616 and PPC604 Maren Purves

Re: Error messages wuth IP-488, VIPC616 and PPC604 J. Frederick Bartlett ([email protected])

Navigate by Date:

Prev: Re: Software Licensing; the GNU GPL and EPICS Timo Korhonen

Next: Re: Software Licensing; the GNU GPL and EPICS john sinclair

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: Re: drvSerial/ioctl question Matthieu Bec

Next: Re: Error messages wuth IP-488, VIPC616 and PPC604 Maren Purves

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