EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  <20142015  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  <20142015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Future Minimum Requirements for IOCs
From: Andrew Johnson <[email protected]>
To: EPICS tech-talk <[email protected]>
Date: Mon, 18 Aug 2014 11:53:34 -0500
I would like to ask the EPICS community to discuss the minimum
requirements to be able to run IOCs in the future. I have two specific
areas, but feel free to discuss others if you know of any.

Memory
------
Currently (with Base 3.14.12.x and the up-coming 3.15.1 releases) an IOC
can be run on an embedded system (RTEMS or VxWorks) with as little as
8MB of RAM, although in practice you probably need more than that for
record instances, device support and driver code, and to allow for
client CA connections. Are any labs designing IOC systems to use small
amounts of RAM, or can we assume that future IOCs (3.16 and later) will
have at least 32MB available (for the OS and application)?

VxWorks
-------
Base 3.14.12.x releases can still be built against VxWorks 5.4.2, which
has a very old (pre-ANSI) C++ compiler. The next major release of Base
version 3.15.1 will require VxWorks 5.5.x. Who would be significantly
affected if later releases were to require VxWorks 6.6 (i.e. gcc 4.1.2)
or later?

This switch would preclude 680x0-based IOCs since Wind River dropped
support for the 68k family in moving to VxWorks 6.x. VxWorks 6.x also
provides a much better memory allocator, earlier versions fragmented
memory unless significant care was taken.


The EPICS core developers would like IOCs to be able to use more memory
and modern compilers, so please discuss the above requirement changes if
they will adversely affect your site.

- Andrew
-- 
Advertising may be described as the science of arresting the human
intelligence long enough to get money from it. -- Stephen Leacock

Navigate by Date:
Prev: Re: Recommended Control HWs for High Voltage Platform Jeong Han Lee
Next: Re: Building Cexp & GeSys with RTEMS 4.10.2 Till Straumann
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  <20142015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: RE: Areadetector IOC for Pilatus Mark Rivers
Next: New users - Channel Access Clients issues Bertrand de Boisdeffre
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  <20142015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 17 Dec 2015 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·