npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

@mlc-ai/web-runtime

v0.18.0-dev2

Published

TVM WASM/WebGPU runtime for JS/TS

Downloads

204

Readme

TVM WebAssembly Runtime

This folder contains TVM WebAssembly Runtime.

Installation

The LLVM main branch support webassembly as a target, we can directly build TVM with LLVM mainline to generate wasm modules. Note that, however, we still need emscripten to compile the runtime and provide system library support.

Note that so far we requires everything to be in the source and setup PYTHONPATH(instead of use setup.py install).

Setup Emscripten

We use emscripten to compile our runtime wasm library as well as a WASI variant that we can deploy to the browser environment.

Follow Emscripten to download emsdk and install emcc on your local environment.

Build TVM Wasm Runtime

After the emcc is setup correctly. We can build tvm's wasm runtime by typing make in the web folder.

make

This command will create the follow files:

  • dist/wasm/libtvm_runtime.bc bitcode library tvm.contrib.emcc will link into.
  • dist/wasm/tvmjs_runtime.wasm a standalone wasm runtime for testing purposes.
  • dist/wasm/tvmjs_runtime.wasi.js a WASI compatible library generated by emscripten that can be fed into runtime.

Build TVM Wasm JS Frontend

Type the following command in the web folder.

npm run bundle

This command will create the tvmjs library that we can use to interface with the wasm runtime.

Use TVM to Generate Wasm Library and Run it

Check code snippet in

  • tests/python/prepare_test_libs.py shows how to create a wasm library that links with tvm runtime.
    • Note that all wasm libraries have to created using the --system-lib option
    • emcc.create_wasm will automatically link the runtime library dist/wasm/libtvm_runtime.bc
  • tests/web/test_module_load.js demonstrate how to run the generated library through tvmjs API.

Run Wasm Remotely through WebSocket RPC.

We can now use js side to start an RPC server and connect to it from python side, making the testing flow easier.

The following is an example to reproduce this.

  • run python -m tvm.exec.rpc_proxy --example-rpc=1 to start proxy.
  • Start the WebSocket RPC
    • Browswer version: open https://localhost:8888, click connect to proxy
    • NodeJS version: npm run rpc
  • run python tests/python/websock_rpc_test.py to run the rpc test.

WebGPU Experiments

Web gpu is still experimental, so apis can change. Right now we use the SPIRV to generate shaders that can be accepted by Chrome and Firefox.

  • Obtain a browser that support webgpu.
    • So far only Chrome Canary on MacOS works
    • Firefox should be close pending the support of Fence.
  • Download vulkan SDK (1.1 or higher) that supports SPIRV 1.3
  • Start the WebSocket RPC
  • run python tests/python/webgpu_rpc_test.py