Hi Eric, > waitpid() distinguishes between normal exits > 128 and a > signal exit, and all existing shells are good about only reporting core > dumps or other signal exits when waitpid() says the child exits via > signal. It's just the ambiguity that results when dealing with $? after > the fact, if you are trying to analyze the failure; and for this > particular .m4 snippet, the chance of a signal is so slim that we are > safe analyzing values > 128.
Thanks for explaining. Since our test programs are invoked from a shell and all we see is $?, not from a C program where we could distinguish raise(SIGSEGV) from exit(139), I think we should continue to use the rule "all exit codes should be < 128", both in m4/* and tests/*. Bruno
