MSP430FR2111: Timer B TBIV does not reset upon ISR

The breakpoint arrow is pointing at the switch statement, so TB0IV hasn't been read yet; thus it still shows 0x0E.

TB0R has continued to count after the ISR was entered. With ID=0 it is unremarkable that it has progressed some small amount past 0.

Thanks Bruce for your prompt reply

I understand your point but:
- For some reason I can't set the breakpoint on the case statemet (optimization ?)
- TB0R which shows 0x0007 on the screen cap, showed 0x0004 on the previous iteration and 0x000A on the next iteration, so apparently it is not a minute progress after IFG have occured but a consistent increase in value , this is why I suspected that TBIFG isn't cleared and the ISR is repeatedly triggered.
This behavior continues until it reaches TB0CCR0 and then CCR0 ISR is triggered as expoected (TIMER0_B0_VECTOR).

Another thought:
As MSP430FR2111 Timer B0 has only 3 CCR modules, maybe I have to use only 3 CCRx case statement (0/2/4) and then the TBIFG (so its value is 6 instead of 14) case statement ??

Thanks again

1) Yes, I usually blame the optimizer for an inability to set a breakpoint where I want. The disassembly might be interesting.
2) A small variation in TB0R (say <10 ticks) could be due to what kind of instruction was interrupted (the CPU won't interrupt in the middle of an instruction). If you're disabling interrupts elsewhere, of course, it could vary even more.
3) My guess (without checking the Book) is that the IV values are architected for an entire product line, even if some of the CCRs aren't implemented. If you're concerned, look in msp430fr2111.h (right click on TBIV__TBIFG and "go to definition").

You can try -O0 and/or if you have a scope try wiggling a GPIO to see the interrupt period. 60K ticks might even be long enough to see on an LED.

My copy of msp430fr2111.h (and 2311.h) says TBIV=(0xffff) -- i.e. a constant -- which would (presumably) cause your entire ISR to be optimized away -- thus your symptoms. If that's what that other example is doing, I don't know how it's working.

Hey Bruce,

You were right. TB0IV worked like a charm...

THANKS !

I managed to locate that example program, and when I compile it (even with -O0) the TIMERB1_ISR is empty -- just a RETI. When I find my Launchpad, I expect it to display the symptoms you were observing.

I suspect this one slipped through the cracks somehow. Good catch.