Re: longout record issue

[Emmanuel Mayssat <emmanuel_mayssat@lynceantech.com>]

Andrew,

Note that the values reported in the previous email are read from a
'text monitor' widget.
I noticed that in a text area, the longout do not update the same way.

In the TEXT AREA widget, I read
-1143013366 when 10 is entered
-1143013367 when 9 ...
etc.

--
E



On Tue, 2009-04-07 at 16:07 -0700, Emmanuel Mayssat wrote:
> Andrew,
> 
> I am on linux-x86 client and server.
> VAL is as follow
> -114301336 for entered values between 7 and 16   (10 being in that
> range)
> -114301335 for values between 17 ad 26
> -114301334 for ... 27 ... 36
> etc.
> It seems to be an encoding problem.
> 
> caput/caget from a terminal prompt work correctly.
> The val of the longout is incorrectly updated prior to the process
> routine of the longout record.
> 
> Regards,
> 
> --
> E
> 
> 
> 
> 
> On Tue, 2009-04-07 at 15:06 -0500, Andrew Johnson wrote:
> > On Monday 06 April 2009 19:37:30 Emmanuel Mayssat wrote:
> > > I have noticed oddities with my longout records.
> > > Particularly when I use the soft Channel record.
> > >
> > > record(longout, "myLO") {
> > > 	field(VAL ,"100")
> > > 	field(SCAN, "Passive")
> > > 	field(DTYP, "Soft Channel")
> > > }
> > >
> > > Now if I create a medm text entry attached to this PV, and attempt to
> > > modify the PV, it updates but not as expected. For example, if I enter
> > > the value 10, the PV is updated to -11100023 (or something like that)
> > >
> > > I am using medm 3.1.1 compiled against EPICS 3.14.9.
> > >
> > > I have other longout records which are not "Soft Channel".
> > > They correctly update with the same medm widget.
> > 
> > What is your target architecture, for both the client and server?  It is known 
> > that IOCs built against Base before 3.14.10 do not work properly on 64-bit 
> > architectures, so if you're running this on linux-x86_64 you should switch to 
> > the 32-bit linux-x86 target.  The CA client code does work properly on 64-bit 
> > systems though, it's just the IOC that's faulty.
> > 
> > With 32-bit linux-x86 clients I only see Dirk's issue when using old versions 
> > of caget and caput (not from Base, the ones from extensions/src/EzcaScan) and 
> > I suspect that's a due to a bug in ezca or in the caget and caput programs 
> > since all the other clients I've tried (including MEDM and Probe) do show the 
> > correct value for a longout record.
> > 
> > Is the "-11100023" number you gave above correct?  If not, please post the 
> > exact number that you get when you put 10 into the PV.
> > 
> > - Andrew
>