When the timer doesn't increment it shouldn't be concerning the
interrupt problem. I don't have experience with the particular msp430
series, but am thinking about the possibilities:
1) gdbproxy (jtag commands), eventually silicon contain bug which
disables some functionalities of the chip after issuing some particular
commands.
2) Some other hardware bug, which is extremely hard to track, but can
cause some unspecified behaviour inside your PCB - volages, signal
stabilities etc.
3) gdb hooks, but now it seems to me not so probable. You can look for
.gdbinit files or gdb.ini(t) depending on O/S in the source directories
of your project to check it, or better using the console with 'show
user' command.
In the first place I would check electrical signals in PCB (VCC and
_GND_ are in 99% of all cases the most important source of various
probles, also it's better to check it in JTAG iface !!) and initial
setting of all the timer A and B registers (and system clock signals
generators) if it remains the same after reset, second to check if
halting/starting timer helps when it's OK try to check for silicon
erratas and look for similar problems with mspgcc and other tools for
the solution, eventually contact TI support if you can to check the
behaviour if you're 100% sure that it's not your fault. If you have
access to some tools as periferal configurator ( Hi Tech compiler
provides the functionality I've read) you can also check if with
generated configuration the behaviour is still the same.
I think I cannot help you anymore,
regards,
W liście z pon, 20-09-2004, godz. 16:22, k.sm...@realtimeengineers.com
pisze:
> More information:
>
> Hardware -> Softbaugh ES449 with MSP430F449
> Interface -> Softbaugh FETP (parallel port)
> GCC version -> msp430-gcc (GCC) 3.2.3
> GDB version -> GNU gdb 6.0
>
> also Insight version as alternative -> GNU gdb 5.1.1
>
> The insight 5.1.1 and GDB 6.0 both exhibit the same behaviour.
>
> I have tried using timer_a and timer_b but the timer does not increment. I
> presumed this was my setup code but have now seen that at least one other
> person has the same problem.
>
> With both the watchdog timer set to interval mode and the basic timer I find
> that I can set a break point in the interrupt service routine and that the
> routine will fire once. After that nowt unless I manually poke the
> interrupt flag in the memory. The global interrupt and timer interrupt
> enable flags remain set correctly.
>
> Thanks for your assistance.
>
>
> Tadeusz Szczyrba writes:
>
> > Maybe you are relaying on register initial values after reset - i don't
> > know, if gdbproxy performs full hardware reset after disconnection, but
> > if not it can be the source of your problem. But none of you gives the
> > exact information: what kind of hardware, S/W version etc. it's
> > sometimes really important.
> >
> > W li.cie z pon, 20-09-2004, godz. 15:10, Frederic Beaulieu pisze:
> >> We have a similar problem, but it seems a little bit
> >> different from yours!
> >>
> >> After a hardware reset (power off - power on), the
> >> code execute without problem in debug mode (after many
> >> steps and breakpoints). If a hardware reset is done
> >> each time we restart the program in debug mode, all
> >> works fine. However, without hardware reset and just
> >> restart GDB, the TBR counter (Timer B) don't start
> >> (all Timer B register and interrupt enabler (GIE and
> >> TBIE) seem ok).
> >>
> >> Any help appreciated.
> >>
> >> --- k.sm...@realtimeengineers.com wrote:
> >>
> >> > Hi,
> >> >
> >> > I am using a Softbaugh FETP and having the following
> >> > problem.
> >> >
> >> > If I build my code and execute it without the
> >> > debugger everything is fine.
> >> > If I step through the code using GDB I find that the
> >> > timer interrupts only
> >> > occur once. After that nothing. I can manually set
> >> > the interrupt flag
> >> > again by poking into memory and this causes the ISR
> >> > to execute, but the
> >> > interrupt flag will never get set automatically
> >> > after the first time. If I
> >> > kill the debug session then the code again operates
> >> > without error.
> >> >
> >> > I have seen a post on the MSP430 Yahoo group with
> >> > the same problem but there
> >> > was no resolution given.
> >> >
> >> > Any help appreciated.
> >> >
> >> >
> >> >
> >> >
> >> >
> >> -------------------------------------------------------
> >> > This SF.Net email is sponsored by: YOU BE THE JUDGE.
> >> > Be one of 170
> >> > Project Admins to receive an Apple iPod Mini FREE
> >> > for your judgement on
> >> > who ports your project to Linux PPC the best.
> >> > Sponsored by IBM.
> >> > Deadline: Sept. 24. Go here:
> >> > http://sf.net/ppc_contest.php
> >> > _______________________________________________
> >> > Mspgcc-users mailing list
> >> > Mspgcc-users@lists.sourceforge.net
> >> >
> >> https://lists.sourceforge.net/lists/listinfo/mspgcc-users
> >> >
> >>
> >>
> >>
> >>
> >> _______________________________
> >> Do you Yahoo!?
> >> Declare Yourself - Register online to vote today!
> >> http://vote.yahoo.com
> >>
> >>
> >> -------------------------------------------------------
> >> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> >> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> >> who ports your project to Linux PPC the best. Sponsored by IBM.
> >> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> >> _______________________________________________
> >> Mspgcc-users mailing list
> >> Mspgcc-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/mspgcc-users
> >
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> > Project Admins to receive an Apple iPod Mini FREE for your judgement on
> > who ports your project to Linux PPC the best. Sponsored by IBM.
> > Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> > _______________________________________________
> > Mspgcc-users mailing list
> > Mspgcc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/mspgcc-users
> >
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Mspgcc-users mailing list
> Mspgcc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mspgcc-users
