Comments by "⃠" (@U20E0) on "Core Dumped" channel.

  1. GPTpasta is now an official term
    2
  2.  @Null-h6c  Constexpr, AI, and manual calculations all don't have any relation to heap memory management. Youtube has a bug where it sometimes loads and posts comments from and to the wrong video for no explicable reason.
    2
  3. The stack's size is fixed. As you add more elements, they get lower addresses, but as far as i know the only reason for that is convention. The memory layout in reality is also nowhere near as clean. There are usually multiple heaps, for example.
    1
  4. In pretty much every language, and for good reason, all structures and arrays must always be contiguous. Additionally in C, the fields of a structure must appear in the exact order they were written in Also this would accomplish nothing. You'd need to keep track of where the pieces went, but that metadata has a good chance of ending up larger than the initial structure (because pointers are big), which means you're back at square -1.
    1
  5. That is called an arena allocation, and is indeed faster, but might also be more complicated to use.
    1
  6.  @cheesepie4ever x86 does have a separate version of every instruction for every (relevant) size
    1
  7. i knew dynamic RAM was sketchy, but i was not suspecting it this sketchy. I hope they also have error correction.
    1
  8. You can also do that using VLAs. But what he meant is that you can't ever change the size of something allocated on the stack.
    1
  9. is this comment on the right video?
    1