I've found compiler bugs before, but this is the first compiler-compiler bug I've ever seen – incredible stuff:
3 posts total
I've found compiler bugs before, but this is the first compiler-compiler bug I've ever seen – incredible stuff: New blog post: "Why is Rosetta 2 fast?" https://dougallj.wordpress.com/2022/11/09/why-is-rosetta-2-fast/ The Rosetta 2 instruction size expansion factor for an sqlite3 binary is ~1.64x (1.05MB of x86 instructions vs 1.72MB of ARM instructions). Surprisingly good, especially given Firestorm cores have six-times the instruction cache of Ice Lake (192KiB vs 32KiB). Something I'm not sure I said is that the goal is to have a single, equivalent ARM instruction for each x86 instruction. And, in real-world code, combining all those tricks allows Rosetta 2 to achieve that surprisingly often. @dougall I see you mention Windows on ARM's emulator, did you look at another high performance emulator like FEX ? |
@dougall after reading through all that I can see why it might not fall under their VRP, but my *goodness* it's hard not to wish they'd throw the initial reporter a few bucks for the effort and persistence to find something that turned out so *weird*
@dougall clearly, the answer to the classic question from Trusting Trust is "asan will catch it, i guess"
@ytvwld ^^