EPICS Home EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

Format page
for printing


Search Tech-talk
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  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 
<== Date ==> <== Thread ==>
Subject: Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules)
From: Eric Bjorklund <[email protected]>
To: EPICS tech-talk <[email protected]>
Date: Fri, 26 Mar 2004 13:10:28 -0700
Umm... I may be completely missing the boat here, but didn't the VME-MXI-1 problem mostly involve bad data returned because of "address rot" during the READ cycles? (c.f. Eric Norum's clarification post from 14 Nov. 2003).

We've seen a similar problem with one of our boards here at LANSCE. In our case, the "pipelined" address was not generated by the executing code. I don't think it was from a PPC or PCI bus "pre-fetch" either, as there was no read cycle generated for the new address. So unless there is some real cleverness going on between the Universe and the PCI controllers, it looks like the Universe chip is just trying to anticipate the next address I am going to want and putting out on the bus early (the new address was the logical successor to the addresses I had previously requested). I am not aware of any cure for this -- except that in our case it is our own board, so we can fix the problem at that level.

-Eric Bj.

On Mar 26, 2004, at 11:38 AM, Kate Feng wrote:

Kate Feng wrote:

I agree with you that one should enable the posted write by default.
However, it could be disabled, which means the address
pipelining on the UniverseII could be disabled as well on the Mvme5500.
I thought the origianl issue is "to disable the address pipelinig".
It looks like you have a reason to disable the posted write
for the NI VME-MXI-1. The trad off is to sacrifice the bandwidth
of the VME bus. However, I do not know if the decrease is
significant for your application. I hope this helps.


I  ran a   short test, which  showed  a decrease of  33% in
performance for the posted-write disabled mode.  It might not be worse
than the VMEchip2 that was used on the MVME167, which is in
your existing system.   I do not have  the datasheet for the
VMEchip2.  I guess it does not provide address pipeling,
which explained why it worked fine with the  MXI-1.


Again, hoping this helps. -- Kate





<><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Eric Bjorklund
Los Alamos Neutron Scattering Center (LANSCE)

phone: 505-667-6031 email: [email protected]
<><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Replies:
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules) Eric Norum
References:
mvme5500 (was National Instruments VME-MXI-1 modules vs. modern VME CPU modules) Kate Feng
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modern VME CPU modules) Andrew Johnson
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVME CPU modules) Kate Feng
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVME CPU modules) Andrew Johnson
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPU modules) Kate Feng
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPU modules) Andrew Johnson
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules) Kate Feng
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules) Kate Feng
Navigate by Date:
Prev: Job opening at APS John Maclean
Next: RE: Buffer problems Mark Rivers
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: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules) Kate Feng
Next: Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVMECPUmodules) Eric Norum
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 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·