I don’t know about the “low-level” part. I have a feeling you’d get just as much of a win from a HLL statically-typed language, like Haskell.

It’s the static-typing, not the low-level-ness, doing most of the heavy lifting in making code JITable/WPOable. You don’t need to manually inline a comparison function, if the JIT knows how to inline the larger class of thing that a comparison function happens to fall into, and if the code is amenable to that particular WPO transformation.

I would compare this to SQL: you don’t optimize a SQL query plan by dropping to some lower level where you directly do the query planner’s job for it. Instead, you just add information to the shape of your query such that it becomes amenable to the optimization the query planner knows how to do. That will almost always get you 100% of the wins it’s possible to get on the given query engine anyway, such that there’d be nothing to gain by writing a query plan yourself.

So, you'd think static typing was a major win but it actually isn't (surprisingly) for a JITed language. Most of the benefits of statically typed languages comes from memory layout optimizations. However, that sort of layout optimization is something that every fast JITed language ends up doing.

This is why, for example, javascript in Graal ends up running nearly as fast as Java in the same VM. The reason it isn't just as fast is the VM has to insert constraint checks to deopt when an assumption about the type shape is violated.

https://www.graalvm.org/javascript/

not really, Haskell is slower than C++/Rust. You get faster performance by being mechanical sympathetic, i.e. you care about how to laid out your data in memory efficiently for the CPU to process, when and how much to allocate, which code path to fold together (inlining) to create a smallest possible set of instructions.

JIT is theoretically possible to figure out all that and transform your code into optimized form, but practically? we don't have Sufficiently Smart Compiler[1] yet. Usually, JIT is worse in restructure your data layout than figure out which part to inline.

[1] https://wiki.c2.com/?SufficientlySmartCompiler

There are certainly nasty edges to javascript programming and I'm not trying to say that it will always be near the same performance as a WASM implementation or a statically compiled binary.

What I'm saying is that you'll find more often than not that it gets close enough to not matter. You'll see javascript at or below 2->3x the runtime of a C++ or Rust statically compiled implementation in most cases. That is pretty close as far as languages go. Java is right in the same range (maybe a little lower).

Isn't Java's JIT is better than static optimizations of C++ because of runtime data is also very important for optimizations? Also inlining of comparison functions usually already done in V8...

Depends. Java's JIT can handle cases of dynamic dispatch better than C++ because of runtime information. However, C++ has the advantage of time. It can do more heavy optimizations because the time budget to do such opts is much longer than what the JIT has available to it. C++ compilers can be much more aggressive about things like inlining.

That said, yes, the runtime information provides really valuable information. That's why PGO is a thing for C++.