[mcfig]

Great article. One real case I encountered that I find thought provoking, is where a bunch of test failures were bucketed into the same bucket because link-time code-generation had noticed that a bunch of C++ getter functions had the same output code and combined them all. So stack traces became confusing because the address-to-symbol mapping was more complicated than the logic we had in place was prepared for.

i.e. optimization had violated a rule we were implicitly relying on (that each non-inlined function should start at a distinct address, so that address-to-symbol mapping could be done easily). But that’s not an explicit guarantee and optimizers don’t seem to think about it much. (Well for inlining it seems to have had some thought, still sucks, but anyway this case doesn’t fit the pattern of inlining).

I find it hard to say anyone is dead wrong in this case… but I would turn off that LTCG optimization any time I could, except where proven necessary.