> Ah, so you meant that even all of them including specializing interpreter and copy-and-patch JIT may not give a reasonable speedup. But I think you have missed the fact that specializing interpreter has already landed on 3.11 and provided 10--60% speedup
No, I'm talking compared to the current default production state. Exactly what Brandt said in his talk at around 23:30, and what I observed when building his branch.
Then I'm not sure why that would refute the intermediate goal to "enable JIT codegen without sacrificing too much performance" as stated in my initial comment, since the proposed copy-and-patch JIT compiler can't make the impact by itself.
No, I'm talking compared to the current default production state. Exactly what Brandt said in his talk at around 23:30, and what I observed when building his branch.