Experimental Physics and
Industrial Control System

"Thompson, David H." <[email protected]> · Wed, 19 Aug 2009 10:51:34 -0400

We have been running 6.2 on a 3100 for nearly 2 years and haven't had to
do this. What version is it that has the problem?

Main problem we had was hanging on to environment variables and I think
they fixed that.

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Andrew Johnson
Sent: Wednesday, August 19, 2009 10:34 AM
To: [email protected]
Cc: Dimitrios Tsifakis
Subject: Re: VME SBC suggestion?

Hi Dirk,

On Wednesday 19 August 2009 05:42:19 Dirk Zimoch wrote:
>
> Generally, vxWorks 6 has some new features (e.g. real time processes) 
> which are not used by EPICS. vxWorks 6 still supports running
applications, e.g.
> EPICS, in "kernel space" (which is all you had in vxWorks 5). However,

> the vxWorks shell has changed a bit and not everything that worked in 
> vxWorks 5 startup scripts might work in vxWork 6 startup scripts (I 
> had difficulties sourcing other scripts). I think I can tell more once

> I have the MVME4100 running.

I wonder if your sourcing scripts problem is the the same issue we found
when we upgraded to vxWorks 5.5.2 - if you use the construct
    ld < ioc.munch
in your startup script then all subsequent shell redirections to source
other scripts get ignored.  The solution is to replace the above line
with this
    ld 0, 0, "ioc.munch"
which permits later shell source commands to operate normally.  I think
it's highly likely that Wind River broke both versions of the target
shell at the same time...

- Andrew
--
The best FOSS code is written to be read by other humans -- Harold Welte

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System