To paraphrase HN user amckinlay, C does not have an actual, formal semantic model. "Furthermore, we argue that the C standard does not allow Turing complete im

To paraphrase HN user amckinlay, C does not have an actual, formal semantic model. "Furthermore, we argue that the C standard does not allow Turing complete implementations, and that its evaluation semantics does not preserve typing. Finally, we claim that no strictly conforming programs exist. That is, there is no C program for which the standard can guarantee that it will not crash." [1] In some sense then, C is not an actual language, so no one can give definitive information about it without being totally wrong in one or two ways. I would argue this fragmentation of both the code and the user base is part of why there is almost no secondary documentation. Not even the specs are correct or complete. [1] https://www.semanticscholar.org/paper/Subtleties-of-the-ANSI-%2F-ISO-C-standard-Krebbers-Wiedijk/62376bcf5e1b55abcaa301b9c8d0f0062b2cba27 https://www.semanticscholar.org/paper/Subtleties-of-the-ANSI...