FP16 Hardware Feature Detection
Nessuno ha ancora preso questa issue.
Valutazione
- Difficoltà
- 5/5
- Tempo stimato
- Più di una settimana
- Idoneità per principianti
- 25/100
- Tipo di issue
- Funzionalità
- Chiarezza
- Abbastanza chiara
- Stato di attività
- Attiva
- Stack tecnologico
- wasm
- Ambito
- operating-systems
Direzione di ricerca
Start with the linked Lowering.md and compare the three host-detection options and open questions in this issue. The issue does not identify proposal files, tests, or a settled design; first determine the intended interface and how it fits WebAssembly semantics. Done means an agreed, specified way to report FP16 support tiers, including forward compatibility and the stated stability and privacy constraints.
Scritto dal modello di indicizzazione a partire dal testo della issue.
Descrizione
Motivation
Host CPUs fall into three tiers of FP16 hardware support (see Lowering.md):
| Tier | Hardware support | Examples |
|---|---|---|
| 0 | None. All FP16 ops are emulated in software | x86 without F16C (e.g. Jasper Lake), RISC-V without Zfhmin/Zvfhmin, interpreters |
| 1 | f16 <-> f32 conversions only. Other f16x8 ops lower via wider floats or software |
Client x86-64 with F16C (AMD Zen 1–5, Intel Core through Arrow/Lunar Lake), AArch64 (ARMv8.0-A), RISC-V Zfhmin+Zvfhmin |
| 2 | Native f16x8 SIMD |
ARMv8.2+ FEAT_FP16, server x86 AVX512-FP16+VL / AVX10, RISC-V Zvfh |
"Conversions" means f32.load_f16, f32.store_f16, f32x4.promote_low_f16x8, f16x8.demote_f32x4_zero, and f16x8.{splat,extract_lane,replace_lane} (which convert lanes to/from f32).
Every engine supporting FP16 validates all instructions on all three tiers, so a single .wasm binary runs everywhere and can bundle both f16x8 and f32x4 code paths. Because everything validates, WebAssembly.validate only reports engine support, not whether the host CPU accelerates FP16.
Libraries such as XNNPACK need the hardware tier to pick the right kernel:
- Tier 2: compute directly in
f16x8. - Tier 1: store weights in
f16(halving memory bandwidth), promote on load, and compute inf32x4. Running a puref16x8kernel on tier 1 (nearly all consumer x86-64 CPUs) is slower thanf32x4because each arithmetic op unpacks to wider floats, computes, and repacks (f16x8.madd/nmaddcannot even usef32FMA without double rounding). - Tier 0: compute in
f32x4(unpacking anyf16weights once up front rather than in inner loops).
Without feature detection, the only way to choose at runtime is a microbenchmark at startup, which is slow, noisy, and adds to startup time.
Requirements
All options below share these properties:
- Hint only: the result does not affect validation or instruction semantics. Engines may conservatively return
0(e.g. in anti-fingerprinting/jitless modes or the deterministic profile). - Stable: the value is constant for at least a compiled module or instance lifetime.
- Low fingerprinting surface: at most a couple of coarse bits, which are already observable via timing.
Options
Option 1: Dedicated instruction (f16.feat)
f16.feat : [] -> [i32] returns the tier from the table above.
- Pros: Self-contained and works in any embedding (Web, WASI) without glue code. Engines can constant-fold it at compile time and eliminate dead inline branches. JS loaders (e.g.
wasm-feature-detect) can still probe it before fetching large bundles via a tiny module. - Cons: Adds a host-dependent instruction to the core spec (though
relaxed-simdsets precedent for embedder-chosen hardware behavior). Specific to FP16, so other proposals would need their own.
Option 2: Opcode-parameterized query (instr.support <opcode>)
instr.support <opcode> : [] -> [i32] returns 1 if the given opcode is natively accelerated on the host, 0 otherwise.
- Pros: General. Also covers relaxed SIMD (e.g. FMA, dot product) and future proposals without new opcodes. Can be constant-folded like Option 1.
- Cons: "Accelerated" is ill-defined per opcode. For example,
f16x8.abs/negare a single bitwise op (sign-bit mask) on any SIMD hardware, yet say nothing about whetherf16x8.addis fast. Larger fingerprinting surface. Broadens the scope of the FP16 proposal. Encoding an opcode as an immediate will be awkward.
Option 3: JS API (WebAssembly.getHardwareSupport(name))
const tier = WebAssembly.getHardwareSupport("fp16"); // 0, 1, 2, or undefined if unsupported
- Pros: Keeps host introspection out of the core spec. Extensible to other features via new names.
- Cons: Web-only unless mirrored in WASI/C APIs. Pure-Wasm libraries can't self-detect, so the embedder must plumb the result in. Constant folding requires pairing with a compile-time import (e.g. via Wasm 3.0
builtins).
Open questions
- Single bit vs. 3 tiers? A single bit ("is native
f16x8available?") merges tiers 0 and 1, even though tier 0 still occurs on recent low-power x86 (e.g. Jasper Lake Chromebooks), non-ZfhminRISC-V, VMs, interpreters, and privacy modes. Without distinguishing them, libraries can't choose between the tier 1 promote-on-load path and the tier 0 unpack-up-front path above. - Ordered tier (
0, 1, 2) vs. bitmask? An ordered tier requires callers to checktier >= 2to stay forward-compatible, whereas a bitmask (1 << 0= conversions,1 << 1=f16x8SIMD, upper bits0) also handles non-linear extensions (e.g. follow-on widening FP16->FP32 FMA such as ARMFEAT_FHM).
- Lingua principale
- WebAssembly
- Stelle
- 14
- Fork
- 0
- Metriche di merge delle PR
- Nessuna PR unita negli ultimi 30g
Preparare l'ambiente
- Nessun Dockerfile né file Docker Compose
- Nessun modello di pull request
- Leggi la guida per i contributori
Come iniziare
- Leggi tutta la issue e poi la guida ai contributi del progetto.
- Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
- Fai un fork del repository e lavora su un branch.
- Apri una pull request che faccia riferimento al numero della issue.
Altre issue di WebAssembly/fp16
-
Difficoltà 5/5 Più di una settimana Idoneità per principianti 38/100
WebAssembly/fp16#10 · 6 commenti ·
-
Difficoltà 4/5 3-5 giorni Idoneità per principianti 35/100
WebAssembly/fp16#8 · 1 commento ·
-
Hardware supportAperta
Difficoltà 5/5 Più di una settimana Idoneità per principianti 25/100
WebAssembly/fp16#7 · 2 commenti ·
-
Difficoltà 4/5 3-5 giorni Idoneità per principianti 45/100
WebAssembly/fp16#6 · 1 commento · 1 reazione ·
-
Difficoltà 5/5 Più di una settimana Idoneità per principianti 20/100
WebAssembly/fp16#4 · 22 commenti ·
Tutte le issue di WebAssembly/fp16
Issue simili
-
flaky(kvm): mounted_static_x86_poll_stdio_matches_native native Linux oracle inherits open fd 42Apertaflaky
Difficoltà 2/5 1-3 ore Idoneità per principianti 78/100
carrick-sh/carrick#2550 ·
I maintainer di solito rispondono entro 1 giorno
-
feature wsl2
Difficoltà 2/5 1-3 ore Idoneità per principianti 80/100
microsoft/WSL#42306 · 3 commenti ·
I maintainer di solito rispondono entro 1 giorno
-
Difficoltà 2/5 1-3 ore Idoneità per principianti 60/100
srid/nixos-unified#164 ·
-
arch-based hyprland
Difficoltà 2/5 1-3 ore Idoneità per principianti 77/100
noctalia-dev/noctalia#5423 ·
I maintainer di solito rispondono entro 1 giorno
-
agent:WSL bug code-review linux networking pri:low sev:low
Difficoltà 2/5 1-3 ore Idoneità per principianti 70/100
I maintainer di solito rispondono entro 1 giorno