3 ms·

It's more a comment about the syntax of the C switch statement than about micro-optimization.

klyrs 6y ago
There's two kinds of C programmers: those familiar with Duff's device, and those who wouldn't expect Duff's device to even compile.
- AdmiralAsshat 6y ago
  I'm familiar with it and I still don't think it should compile...
  - burfog 6y ago
    I'm disappointed that gcc's nested function extension doesn't let me use a switch to have a second entry point for the inner function. I also can't seem to get there with a plain goto. I can't jump out either. Probably the computed goto extension will do the job, but that is cheating.
    - marvy 6y ago
      Using computed goto would (probably) make the code compile, but it wouldn't actually work. That's would be heading directly into undefined behavior territory (aka land of the nasal demons). To give just one or two examples of why this is doomed: imagine you want to jump "inward". If the inner function takes any arguments, you haven't passed them! And if you jump inward, and the inner function then hits a return statement, do you return to the outer function, or to its caller? And after you resolve all such semantic issues, you need to actually implement all this, without slowing down code which doesn't use this feature. And all this complexity doesn't buy you much because you have two other options: 1. just call the function to jump inwards, and simply return to jump out. 2. Don't bother with nested functions, and just inline the code.
      - marvy 6y ago
        So it turns out I'm kind of wrong! My arguments for jumping "inward" still seem pretty valid, but they don't generalize to jumping "outward". And indeed, GCC actually allows this, but you need to use the right syntax: https://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html https://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html
- enriquto 6y ago
  > There's two kinds of C programmers: those familiar with Duff's device, and those who wouldn't expect Duff's device to even compile. There's also the C programmers that come from assembler for whom the construction expressed by Duff's device is very natural. They would have a hard time believing that somebody find it surprising. For those people, it is obvious that the code should compile and its meaning is clear, as can be mapped directly to the compiled code.

There's two kinds of C programmers: those familiar with Duff's device, and those who wouldn't expect Duff's device to even compile.

AdmiralAsshat 6y ago
I'm familiar with it and I still don't think it should compile...
- burfog 6y ago
  I'm disappointed that gcc's nested function extension doesn't let me use a switch to have a second entry point for the inner function. I also can't seem to get there with a plain goto. I can't jump out either. Probably the computed goto extension will do the job, but that is cheating.
  - marvy 6y ago
    Using computed goto would (probably) make the code compile, but it wouldn't actually work. That's would be heading directly into undefined behavior territory (aka land of the nasal demons). To give just one or two examples of why this is doomed: imagine you want to jump "inward". If the inner function takes any arguments, you haven't passed them! And if you jump inward, and the inner function then hits a return statement, do you return to the outer function, or to its caller? And after you resolve all such semantic issues, you need to actually implement all this, without slowing down code which doesn't use this feature. And all this complexity doesn't buy you much because you have two other options: 1. just call the function to jump inwards, and simply return to jump out. 2. Don't bother with nested functions, and just inline the code.
    - marvy 6y ago
      So it turns out I'm kind of wrong! My arguments for jumping "inward" still seem pretty valid, but they don't generalize to jumping "outward". And indeed, GCC actually allows this, but you need to use the right syntax: https://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html https://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html
enriquto 6y ago
> There's two kinds of C programmers: those familiar with Duff's device, and those who wouldn't expect Duff's device to even compile. There's also the C programmers that come from assembler for whom the construction expressed by Duff's device is very natural. They would have a hard time believing that somebody find it surprising. For those people, it is obvious that the code should compile and its meaning is clear, as can be mapped directly to the compiled code.