@dwillems42@rootme_org Oh root-me j'avais oublie ce site, je l'adorais ! A l'epoque j'avais fini a 100% la categorie "cracking", mais je vois qu'ils ont rajouter pleins de challenges depuis !
https://t.co/aObzzPzNU3
Interesting benchmark on pointers vs. indices, relevant to Rust code which for borrow checking reasons tends to use indices a lot: https://t.co/wiWg9iUQAS
tl;dr you pay 1 cycle penalty for address computation, but indices can be 32-bit instead of 64-bit, which can be a big win.
It's been a while since I wrote anything so I wrote an article on how to discover the entire x86-64 instruction set in seconds including any hidden instructions and learn their basic properties while on it. There were some pretty interesting results!
https://t.co/t7vMwYDYas
4/ So Apple simply cheated. They added Intel's memory-ordering to their CPU. When running translated x86 code, they switch the mode of the CPU to conform to Intel's memory ordering.
The guy who maintains Helm, the most important package in Emacs, is a 57-year old alpine mountain guide who learned to program when he was 42, as a hobby. https://t.co/Za8IHb1EeU
"0 overhead" wait-free hazard pointers (: https://t.co/IrtlHToFHU
I thought there might be something nice in https://t.co/gCsVbkY6Vj, but I didn't expect such a nice result.