Experimental Physics and
Industrial Control System

"Denison, PN \(Peter\)" <[email protected]> · Wed, 28 Mar 2007 11:53:36 +0100

> From: Korhonen Timo [mailto:[email protected]] 
> 
> Denison, PN (Peter) wrote:
> > I was searching for drvTS yesterday, and couldn't find it for
versions 
> > of base after 3.14.8
> >
> > I know that it was removed from base, and I found this section of
the 
> > release notes: "These are removed from EPICS base. The version that 
> > previously came with base was the version that worked with the APS 
> > event system. It is available from APS. The version that works with 
> > the newer event systems that evolved from the APS system is
available 
> > from the Swiss Light Source."
> >
> > As we are using the "newer event systems", I guess we'll need the
SLS 
> > version. Can anyone help me get hold of it - Timo?
> >   
> If you _really_ want it, we can give you 'our' version of 
> drvTS. However, if you are using 3.14., I would recommend 
> moving to GeneralTime. We have a GeneralTime time driver for 
> the MicroResearch EVR/EVG( although it may need some minor 
> modifications for the 200-series EVRs.). We have no 
> intentions to support drvTS further. GeneralTime is a big 
> improvement over drvTS.

I was planning to move to generalTime eventually, and I'm pleased to
hear that it's ready to try.

> > I assume that this is orthogonal to generalTime, and I'll still need
a 
> > version of drvTS to support the MicroResearch EVR, even if I were to

> > get hold of generalTime (which I have managed to find a link for)?
We 
> > have the drivers for the EVR, but they don't compile against 
> > base-3.14.8 because drvTS is missing.
> >
> >   
> I would like to state it once again:
> the drvTS will not be supported anymore and will be replaced 
> by General Time.
>  (as discussed in the last EPICS meeting. It took us a while to 
> implement the replacement but at least a first version exists 
> since a few months.)
> 
> There are still a few issues to be sorted out, but we have tested the 
> GeneralTime (event) driver extensively.
> 
> We have sent our GeneralTime package (GT base + our eventTime 
> driver) to Andrew and a few other people for them to take a look 
> at it and to do the necessary preparations for integrating it.
> We can of course send you the same package, and try to help you
> to get started. There will also be a talk in the Epics meeting in 
> Hamburg (by Babak) and Babak will be there for detail discussions.

I've got hold of the 1-2-0 release from a link recently posted to
tech-talk by David.

By "our eventTime driver", do you mean the MicroResearch EVR driver, or
another intermediate driver that is also required?

> I do not immediately remember what the dependency between the EVR 
> drivers and drvTS is, but Babak could probably tell it right away. 
> Let us take a look at it (however, this week there is a SLS shutdown 
> and people are pretty busy with maintenance and testing new stuff.)

I've done a bit more digging and now understand a little more. The only
dependency seemed to be on TSepochUnixToEpics(), which I've replaced
with a little rearrangement and epicsTimeFromTimespec(). There were also
trivially some header files that had been included by drvTS.h as well.

It now builds, but I haven't yet done the calls to iocClockRegister to
make it supply the IOC time. Is that sufficient to make it work in the
absence of drvTS and generalTime, or do I need at least one of them?
(provided of course that the timing system stays alive - I know that
I'll have no fallback until I implement generalTime)

-- 
Peter Denison, Senior Software Engineer
Diamond Light Source Ltd., Diamond House, Chilton, Didcot, Oxon, OX11
0DE
Tel: +44 1235 778511

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System