Lisa: Thank you for your response.
1. I don't have any "other" ccs5 environment to copy from. I was hoping that there was some WEB site where all the available CCS MSP430 compiler versions were available (and I thought I had found this before, but I can't seem to find it now) - so I could pick some newer ones to try. Can you point me to such an archive?
2. I have verified that this symptom does not occur if the optimizations are turned "off", and does occur on all the other 0-4 optimizations settings. I only did this for v4.12 of the compiler, but also verified that the "off" setting also works OK for the v4.21 compiler.(but the code still crashes in this situation also [as I mentioned before], for reasons I have yet to determine). However, the situation is a little more "grey" as I look at the disassembly a little closer. It appears that it did not actually ignore the line, it just moved the invocation of this functionality to a new location (optimizers do weird things) that I would not have "expected" (and I am not sure I agree with...). I tried implementing this with an inline assembler statement instead and it behaves the same way as far as the optimizer operations go!! It would thus appear that the compiler doesn't understand the context of these statements - treating them as it would for "any" asm() statment - which often have problems with the optimizer operation.
So I guess in my case, I either turn off the optimizer for the whole project, or move all the routines that I don't want optimized to a separte file and just turn off optimizations for this file (I didn't see a "pragma" that can specify/override optimizations for individual functions).