So, next year's Macbook will be based on Apple's own ARM chip

randomx benchmarks will be interesting

hyc did you know that these days code optimizations involve _adding_ more instructions to the code?

I literally added one instruction to the generated RandomX code to get 1% speedup on Ryzen

What insn ?

"interrupts off" :D

No, "jmp $+2"

which is essentially a no-op

but it speeds up things mystically

Alignment to cache ?

my theory it has something to do with uOp cache

but I have no clue what actually happens and what it triggers inside the CPU

the trick doesn't work on Intel btw

I'd be more inclined to believe cache alignment

It's odd since if it's that easy you'd expdct the cpu to do it itself.

(if not alignment)

I theorized that ldmxcsr switched CPU from uOp cache to the regular x86 decoder

and jmp instruction was supposed to force it back to uOp cache

interesting

but I'm not sure this is what actually happens there

I guess there's no perf counters to expose uOp cache operation

And a 2 byte nop doesn't speed up I assume ?

(if jmp $+2 takes two bytes)

No, it has something to do with what jmp does

Because I found in the AMD optimization manual that switch between uOp cache and x86 decoder can only happen after a completed jump

sech1: btw I tried to compile randomx for iOS, do you know why it doesn’t recognise the ARM asm syntax in jit_compiler_a64_static.S?

maybe you need some special compiler options, no idea

first it didn’t like the `/*` style of comments and then `.global DECL(randomx_program_aarch64)` stuff

hmm ok

There are no /* comments in that file

selsta: i think the /* comments depend on if you call the assembler directly, or if you get gcc to call the assembler for you

ah, you compiled RandomX repo

xmrig's version should compile fine

all I did was `cmake .. -GXcode -DCMAKE_SYSTEM_NAME=iOS -DARCH_ID=armv8-a` to create a Xcode project and then compile to my phone

but no idea if that should work :) will try xmrig

the xmrig version uses `#` for comments which Xcode also doesn’t like

I give up lol

should be using ; for asm comments

It wouldn’t be too interesting anyway until the ARM macbooks get released

that Ryzen optimization is some µarch black magic