Why Is It Fast?

Why standard libc functions might be faster than handwritten loops - loop unrolling and optimizations in the C programming language

#c #clang #clanguage #cli #programming #arch #linux #unix #technology #tutorial #loop #strcpy #string #memcmp #memcpy

[stargazer]

It's too bad the end of the video is completely blocked with pages for other videos overlaid over top of it.

[redcrafterlppa]

The compiler does craft a better memcpy than the manually optimized versions of glibc. Especially in cases where the number of bytes in memcpy is a compile time known value.
I rested memcpy on different compilers and with and without a fixed bytes length and the results show that more often than not memcpy was completely unrolled out of existence.

[drd]

Even 8088 assembly had a memory region copy operation. Even -o3 can't compete with a single assembly instruction.

[yuankai]

Great video! what's the vim colorscheme ?