g+
g+ Communities
Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
<== Date ==> <== Thread ==>

Subject: Re: RTEMS-mvme5500 BSP
From: Kate Feng <feng1@bnl.gov>
To: ʦê»Àñ <shl@ihep.ac.cn>
Cc: rtems-users@rtems.org, tech-talk@aps.anl.gov
Date: Thu, 10 Jan 2008 09:03:05 -0500
Hi,

As what I have stated previously privately that
there are known bugs  unrelated to your problems, in the version
of the mvme5500 BSP you have that I discourage you from using it,
especially knowing that you are using it for the powerSupply.  This
is all I can say until the EPICS LICENSE issue is resolved between
the  mvme5500 BSP that I wrote and the beatnik BSP that Till
Straumann wrote after  mvme5500 BSP was released.

I can not send you the new kernel until this issue is resolved. I do
not think you can use beatnik leagally at thie point before it's resolved, either.
I will try to answer you the questions below as much as I  can, but
I do not see it as mvme5500 BSP related or  OS related. I recommend
that you  take an EPICS and RTEMS training and have all questions
solved there. I remember  that there is an EPICS workshop/training
to be held in China this Spring.

ʦê»Àñ wrote:

> My EPICS base version is 3.14.6, the same as you use.
>
> I locate sequencer source codes under base/src/sequence, and then
> modify the file configure/RELEASE(add the following lines)
> ******************************************************************
> EPICS_BASE=/home/rtems/base-3.14.6/base
> TEMPLATE_TOP=$(EPICS_BASE)/templates/makeBaseApp/top
> ******************************************************************
>
> modify the file configure/CONFIG_SITE
> ******************************************************************
> PVCA   = TRUE
> CROSS_COMPILER_TARGET_ARCHS = RTEMS-mvme5500
> ******************************************************************
>
> and then under the directory /home/rtems/base-3.14.6/base/src/sequence, then make.
>
> add the line to .cshrc
> ******************************************************************
> set path=(/home/rtems/base-3.14.6/base/sequence/bin/linux-x86 $path)
> ******************************************************************

I did not try sequence with RTEMS yet, but it should not be OS dependent.
Look at the linux-x86.
Obviously, the PATH is wrong if you intended to use RTEMS.  Depending
on the configuration of the Makefile, it should be something like
/home/rtems/base-3.14.6/base/sequence/bin/RTEMS-mvme5500  or
/home/rtems/base-3.14.6/bin/RTEMS-mvme5500  instead of
/home/rtems/base-3.14.6/base/sequence/bin/linux-x86.

Kate

>
>
> then add a line to the file /home/rtems/base-3.14.6/base/configure/RELEASE
> ***********************************
> SCNSEQ=/home/rtems/base-3.14.6/base/src/sequence
> ***********************************
>
> in my epics application, part of Makefile is as follows:
> *****************************************************
> ifneq ($(SNCSEQ),)
>     # This builds sncExample as a component of power_supply
>     power_supply_DBD += sncExample.dbd
>     power_supply_SRCS += sncExample.stt
>     power_supply_LIBS += seq pv
>
>     # The following builds myProgram as a standalone application
>     PROD_HOST += myProgram
>     myProgram_SNCFLAGS += +m
>     myProgram_SRCS += myProgram.st
>     myProgram_LIBS += seq pv
>     myProgram_LIBS += $(EPICS_BASE_HOST_LIBS)
> endif
>
> power_supply_LIBS += $(EPICS_BASE_IOC_LIBS)
> *****************************************************
> and then make, an error appears:
> *****************************************************
> make[1]: *** No rule to make target `myProgram.o', needed by `myProgram'.  Stop.
> *****************************************************
>
> Could you give me the latest kernel and try? any colleagues did not install
> snc under EPICS 3.14.X, so it's a problem to me.
>
> thanks again!
>
> Haoli Shi
>
> First, I think the version of the mvme5500 BSP that
> you use have a known bug at run-time, which,
> unfortunately,  I can not discolsed any further until
> I have resolved the issue of the EPICS license violation,
> which was caused by the beatnik BPS that is written
> by Till STraumann of SLAC.
>
> Thus, I do discourage you from using it with
> the power supply.
>
> 1) Unresolved symbol: pdevLibVirtualOS
> I do not have this problem.
> Which version of EPICS do you use ?
>
> 2) SNCSEQ should not have to do with any BSP.
> It depends on how you set the Makefile.
>
> Kate



Navigate by Date:
Prev: Re: VME Toaster - has anyone else experienced this? Korhonen Timo
Next: VXI11 GPIB and asynRecord Eric Williams
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
Navigate by Thread:
Prev: EDM Motif Widget Rendering John Dobbins
Next: VXI11 GPIB and asynRecord Eric Williams
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  <20082009  2010  2011  2012  2013  2014 
ANJ, 02 Sep 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· EPICSv4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·