Code (Performance)

After getting in an extended discussion about the supposed performance tradeoff between #pragma once and #ifndef guards vs. the argument of correctness or not (I was taking the side of #pragma once based on some relatively recent indoctrination to that end), I decided to finally test the theory that #pragma once is faster because the compiler doesn’t have to try to re-#include a file that had already been included.

For the test, I automatically generated 500 header files with complex interdependencies, and had a .c file that #includes them all. I ran the test three ways, once with just #ifndef, once with just #pragma once, and once with both. I performed the test on a fairly modern system (a 2014 MacBook Pro running OSX, using XCode’s bundled Clang, with the internal SSD).

Today I got into a discussion regarding embedded programmers and their tendency to avoid exceptions in C++ (even after moving to new-school “embedded” systems such as smartphones, game consoles, and set-top boxes). The argument basically boils down to three things: code size, memory usage, and execution speed. So, rather than continue on without actually putting my beliefs to the test (the discussion basically centered around whether engineers are making assumptions based on how things were 15+ years ago), I decided to construct a minimal test which I think shows where the tradeoffs may be.

Note: The timing and memory analysis has been updated as of May 24, 2018.