cfallin

Oh goodness, that name is so good!

(And thanks!)

Thanks! I haven't studied Graal's IR in detail, no. I'll add it to my reading list...

OK, cool. I was assuming "escape analysis and type inference" implied a JS JIT -- straight from your comment, no other assumptions intended. But you've got a lot of interesting experience here and thanks for your thoughts.

All the best!

Hi Fil -- thanks for the comment!

I think we may be playing in slightly different spaces: unlike a JS JIT, Cranelift doesn't have "super fancy escape/type analysis". We're really targeting the core optimizations (GVN, LICM, cprop, RLE, STLF, various algebraic rules) in a fast compile pass. The RLE+GVN interaction was pretty real in our case, as are interactions between the algebraic rewrites and GVN.

You'll note that my main point is that the single fixpoint loop for all of the core rewrites is what we wanted, and what the sea-of-nodes-with-CFG gets us; the egraph (multiple versions of one value) is kind of an aside. One could say: well sure but I could just do a single pass with that fixpoint loop without all the egraph stuff; and, yes, that's what our single rewrite pass is.

We have two: fuel and epochs. Fuel (analogous to "gas" as it appears in many VM platforms) is a deterministic count and epochs check an always-increasing counter in shared memory meant to be periodically bumped by another thread. In either case, hitting the limit can be configured to either async-yield back to the event loop (in async mode) or to trap/terminate the Wasm instance. Both are based on instrumentation, i.e., extra code inserted during Wasm-to-machine-code compilation to do these checks at the start of loops and at the top of each function. Epoch instrumentation is cheaper because it's checking a mostly-read-only value in memory (so usually cached) while fuel loads and stores that value constantly from the VMContext.

(Core Wasmtime maintainer here, and I built our epochs mechanism when I realized we could do better than fuel if one doesn't need the determinism, only periodic yields)

> It'll be interesting to see what the second non-browser-based WASM runtime to fully support 3.0 will be (I'm guessing wasmtime will be first; ...)

Wasmtime already supports every major feature in the Wasm 3.0 release, I believe. Of the big ones: garbage collection was implemented by my colleague Nick Fitzgerald a few years ago; tail calls by Jamey Sharp and Trevor Elliott last year (with full generality, any signature to any signature, no trampolines required!); and I built our exceptions support which merged last month and is about to go out in Wasmtime 37 in 3 days.

The "3.0" release of the Wasm spec is meant to show progress and provide a shorthand for a level of features, I think, but the individual proposals have been in progress for a long time so all the engine maintainers have known about them, given their feedback, and built their implementations for the most part already.

(Obligatory: I'm a core maintainer of Wasmtime and its compiler Cranelift)

They're hitting another design point on the compile time vs. code-quality tradeoff curve, which is interesting. They compile 4.27x faster than Cranelift with default (higher quality) regalloc, but Cranelift produces code that runs 1.64x faster (section 6.2.2).

This isn't too surprising to me, as the person who wrote Cranelift's current regalloc (hi!) -- regalloc is super important to run-time perf, so for Wasmtime's use-case at least, we've judged that it's worth the compile time.

TPDE is pretty cool and it's great to see more exploration in compiler architectures!

Indeed; the tl;dr is "yes it works on SpiderMonkey, and was designed to do so". Blog post in a few months, I promise!

> That is, if x and y are determined only at runtime for power(x, y) then I don't see what can be optimized.

Yes, the example in Max's post is specifically assuming one wants to generate a specialized version of `power` where `y` is fixed.

To take it back to weval: we can know what the bytecode input to the interpreter is; we provide an intrinsic (part of the "wevaling" request) to indicate that some function argument is a pointer to memory with constant, guaranteed-not-to-change content. That, together with context specialization on PC (another intrinsic), allows us to unroll the interpreter loop and branch-fold it so we get the equivalent of a template method compiler that reconstitutes the CFG embedded in the bytecode.

This is indeed a good point and something I want to write about when I eventually do a blog post on weval.

A few counterpoints that I'd offer (and what led me to still take this approach):

- If the target has sub-par debugging infrastructure, it can be easier to debug an interpreter (which is portable) then apply the semantics-preserving PE. In particular when targeting Wasm outside the browser, there is... not really a good debug experience, anywhere, for that. It was way easier to get an interpreter right by developing on native with gdb/rr/whatever, and then separately ensure weval preserves semantics (which I tested with lockstep differential execution).

- Maintenance: if one is going to have an interpreter and a compiler anyway (and one often wants this or needs this e.g. to handle eval()), easier for them both to come from the same source.

- Amortized cost: in the Wasm world we want AOT compilers for many languages eventually; there are interpreter ports with no Wasm backends; developing weval was a one-time cost and we can eventually apply it multiple times.

- If the semantics of the existing interpreter are quite nontrivial, that can push the balance the other way. I designed weval as part of my work on SpiderMonkey; extremely nontrivial interpreter with all sorts of edge cases; replicating that in a from-scratch compiler seemed a far harder path. (It's since been done by someone else and you can find the "wasm32 codegen" patchset in Bugzilla but there are other phasing issues with it from our use-case's PoV; it's not true AOT, it requires codegen at runtime.)

I don't think the tradeoff is always clear and if one is building a language from scratch, and targeting a simple ISA, by all means write a direct compiler! But other interesting use-cases do exist.