Andrew Johnson wrote:
> Kate Feng wrote:
> >
> > Measured 60x bus bandwidth (MPC7450 + GT64260) :
> > Pipelining disabled : 177 MBytes/sec
> > Pipelining enabled : 589 MBytes/sec
> >
> > Thus, theoretically, one should be able to disable the
> > address pipelining of the mvme5500 (MPC7455 + GT64260).
>
> There's no way of getting 589MB/s through a regular VMEbus.
>
Perhaps I forgot to mention that the paper was based on work done
on an evaluation board without the VME bus. However, I did not
say one can get that 589 MB/s through a regular VME bus.
It is just a reference number to describe the possibility of
disabling the address pipelining on the mvme5500.
>
> Address pipelining on the VMEbus (which was causing the problems with the
> VME-MXI-1 modules) is not related to address pipelining by the CPU on its
> local bus, which is presumably what the Motorola paper is quoting.
I did not say it was quoted by Motorola. I said it was brought up by
someone in the tech-talk discussion. Now I looked back at the archiver.
Oh, it was quoted by someone from APS submitted to tech-talk
on Fri, 14 Nov 2003 10:52:45 -060.
> The
> VMEbus cycle timings are completely controlled by the Tundra Universe-2
> chip on Motorola PowerPC boards, and VMEbus address pipelining cannot be
> disabled.
>
This is yet to be verified. However, in the previous tech-talk it was
quoted :
> The APS has a rather large number of these modules. The flaw in the
VME-MXI-1
> has not caused problems because the modules have been used only
> with MVME167 CPU modules which do not perform address pipelining. We now
> want to upgrade some of these MVME167 IOCs to more modern CPU
> modules, but we've run into a problem. CPU modules with a UNIVERSE-II VME
interface
> chip (all Motorola 2100, 5100 and 5500 series) do perform
> address pipelining which results in erratic data transfer from VME-MXI-1
modules.
> My guess is that it is likely that any modern CPU is going to cause the
same problem.
Now you are telling me a different story.
Thanks,
Kate
- 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
- Navigate by Date:
- Prev:
modtcp channel status Eric Williams
- Next:
RE: EPICS on MVME2100 series? Sibley III, Coles
- 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. modern VME CPU modules) Andrew Johnson
- Next:
Re: mvme5500 (was National Instruments VME-MXI-1 modules vs. modernVME CPU modules) Kate Feng
- 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
|