Interrupt Being Disabled while Debugging?

Hi,

I'm in the process of porting our machine control code from CE 3.00 to CE
4.2. Our platform is a CEPC.One of our modules is an ActiveX object that
provides control of servo motors. This object, when created, sets up an
interrupt service thread that handles communication between the CEPC and the
Servo hardware via dual port RAM.

On the 3.00 platform, we could debug through the application program that
creates the instance of the ActiveX object with no problem. With the 4.2
platform, it appears that the interrupt is being held off whenever a break
point in the application is encountered. Any suggestions as to what might
be causing this? I've tried the testing on two systems to try to eliminate
hardware, but both exhibit the same problem. Both the 3.00 and 4.2 systems
use platforms we built. There are only small differences between these
platforms and the "Standard" CEPC platform. The hardware is identical on
both systems.

Any ideas would be greatly appreciated!

Thanks,
Jim Frazer

David
Fri Jul 23 16:02:58 CDT 2004

I think this is by design.
After hit breaking point, the Interrupt is desabled. Otherwise ISR can run
and modify the content that we don't want to happen.
David Liao

"Jim Frazer" <jtfrazer@nospam.net> wrote in message
news:eQNkRIPcEHA.644@tk2msftngp13.phx.gbl...
> Hi,
>
> I'm in the process of porting our machine control code from CE 3.00 to CE
> 4.2. Our platform is a CEPC.One of our modules is an ActiveX object that
> provides control of servo motors. This object, when created, sets up an
> interrupt service thread that handles communication between the CEPC and
the
> Servo hardware via dual port RAM.
>
> On the 3.00 platform, we could debug through the application program that
> creates the instance of the ActiveX object with no problem. With the 4.2
> platform, it appears that the interrupt is being held off whenever a break
> point in the application is encountered. Any suggestions as to what might
> be causing this? I've tried the testing on two systems to try to
eliminate
> hardware, but both exhibit the same problem. Both the 3.00 and 4.2
systems
> use platforms we built. There are only small differences between these
> platforms and the "Standard" CEPC platform. The hardware is identical on
> both systems.
>
> Any ideas would be greatly appreciated!
>
> Thanks,
> Jim Frazer
>
>
>

Seung
Fri Jul 23 17:01:36 CDT 2004

David Liao (MS) wrote:

> I think this is by design.
> After hit breaking point, the Interrupt is desabled. Otherwise ISR can run
> and modify the content that we don't want to happen.
> David Liao
>
> "Jim Frazer" <jtfrazer@nospam.net> wrote in message
> news:eQNkRIPcEHA.644@tk2msftngp13.phx.gbl...
>
>>Hi,
>>
>>I'm in the process of porting our machine control code from CE 3.00 to CE
>>4.2. Our platform is a CEPC.One of our modules is an ActiveX object that
>>provides control of servo motors. This object, when created, sets up an
>>interrupt service thread that handles communication between the CEPC and
>
> the
>
>>Servo hardware via dual port RAM.
>>
>>On the 3.00 platform, we could debug through the application program that
>>creates the instance of the ActiveX object with no problem. With the 4.2
>>platform, it appears that the interrupt is being held off whenever a break
>>point in the application is encountered. Any suggestions as to what might
>>be causing this? I've tried the testing on two systems to try to
>
> eliminate
>
>>hardware, but both exhibit the same problem. Both the 3.00 and 4.2
>
> systems
>
>>use platforms we built. There are only small differences between these
>>platforms and the "Standard" CEPC platform. The hardware is identical on
>>both systems.
>>
>>Any ideas would be greatly appreciated!
>>
>>Thanks,
>>Jim Frazer
>>
>>
>>
>
>
>
Try debugging the application using an application debugger like EVC
rather than using the Platform builder debugger. The application
debugger will not halt the system (since it can operate over product
ethernet or active sync transports).

--
Seung Yi
Senior Engineer
Codetelligence, Inc.
Web : www.codetelligence.com

Jim
Sun Jul 25 12:52:18 CDT 2004

I should have stated that I was using eVC ++ 4.0. When using eVC++ 3.0 on a
CE 3.00 platform it is possible to place a break point in one thread of the
application and have all the other threads continue to run when the break
point is reached. This is not what I saw in a CE 4.2 platform using eVC++
4.0. What I see in the 4.2 platform with eVC++ 4.0 and an application with
two threads is that when one thread reaches a break point, the other thread
stops execution. If I place a break point in both threads I can get into a
situation where a message that says "All Threads are Suspended" when I try
to continue step-by-step execution. I have to believe that I have something
set up incorrectly.

Jim


"Seung Yi" <seungyi@codetelligence_NOSPAMMERS_.com> wrote in message
news:uwnmpBQcEHA.3420@TK2MSFTNGP10.phx.gbl...
> David Liao (MS) wrote:
>
> > I think this is by design.
> > After hit breaking point, the Interrupt is desabled. Otherwise ISR can
run
> > and modify the content that we don't want to happen.
> > David Liao
> >
> > "Jim Frazer" <jtfrazer@nospam.net> wrote in message
> > news:eQNkRIPcEHA.644@tk2msftngp13.phx.gbl...
> >
> >>Hi,
> >>
> >>I'm in the process of porting our machine control code from CE 3.00 to
CE
> >>4.2. Our platform is a CEPC.One of our modules is an ActiveX object
that
> >>provides control of servo motors. This object, when created, sets up an
> >>interrupt service thread that handles communication between the CEPC and
> >
> > the
> >
> >>Servo hardware via dual port RAM.
> >>
> >>On the 3.00 platform, we could debug through the application program
that
> >>creates the instance of the ActiveX object with no problem. With the
4.2
> >>platform, it appears that the interrupt is being held off whenever a
break
> >>point in the application is encountered. Any suggestions as to what
might
> >>be causing this? I've tried the testing on two systems to try to
> >
> > eliminate
> >
> >>hardware, but both exhibit the same problem. Both the 3.00 and 4.2
> >
> > systems
> >
> >>use platforms we built. There are only small differences between these
> >>platforms and the "Standard" CEPC platform. The hardware is identical
on
> >>both systems.
> >>
> >>Any ideas would be greatly appreciated!
> >>
> >>Thanks,
> >>Jim Frazer
> >>
> >>
> >>
> >
> >
> >
> Try debugging the application using an application debugger like EVC
> rather than using the Platform builder debugger. The application
> debugger will not halt the system (since it can operate over product
> ethernet or active sync transports).
>
> --
> Seung Yi
> Senior Engineer
> Codetelligence, Inc.
> Web : www.codetelligence.com
>