I view the null check stuff as being more of an example of the compilers/even kernel Devs not bothering to try and properly express their desires i.e. in this case the check is has wording in the standard covering it, so it should be very explicitly desired to remain in the binary (volatility or similar, although there are limits to what you can ergonomically express in C)

Similarly strict aliasing can change the behaviour of code but if you're genuinely relying on that your code is probably bad - either in the standards view, or in my view in that you can write the same code in a manner that won't cause any mischief (i.e. there are standard-friendly ways to do ugly pointer crap even if they mean memcpying pointers - which will then be eliminated by the optimizer)