[Pop_-]

Disclaimer: The title has been changed to "Rust std fs slower than Python!? No, it's hardware!" to avoid clickbait. However I'm not able to fix the title in HN.

[sharperguy]

"Works on contingency? No, money down!"

[pvg]

you can mail [e-mail] and they can change it for you to whatever.

[3cats-in-a-coat]

What's the TLDR on how... hardware performs differently on two software runtimes?

[pornel]

AMD's implementation of `rep movsb` instruction is surprisingly slow when addresses are page aligned. Python's allocator happens to add a 16-byte offset that avoids the hardware quirk/bug.

[sound1]

thank you, upvoted!

[lynndotpy]

One of the very first things in the article is a TLDR section that points you to the conclusion.

> In conclusion, the issue isn't software-related. Python outperforms C/Rust due to an AMD CPU bug.

[j16sdiz]

It is software-related. Just the CPU perform badly on some software instruction.

[xuanwo]

FSRM is a CPU feature embedded in the microcode (in this instance, amd-ucode) that software such as glibc cannot interact with. I refer to it as hardware because I consider microcode a part of the hardware.