This is a wrapper to collect information needed to run WebAssembly modules generated by emscripten.
The idea is that we should be able to run any emscripten generated wasm module, and then post the results as a verified computation using TrueBit. First the script prepare.js adds the wrapper to the js file that was generated by emscripten. Then this file
can be ran using Node.js. This will collect the information about calls and memory changes that can be recorded, and used for verified computations. The script first uses the ocaml interpreter to generate a wasm file that has our runtime linked in.
Now this file can be ran using the ocaml off-chain interpreter.
Install npm dependencies
npm install
Install emscripten:
#You can place this wherever, root directory is recommended git clone https://github.com/juj/emsdk.git cd emsdk LLVM_CMAKE_ARGS="-DLLVM_EXPERIMENTAL_TARGETS_TO_BUILD=WebAssembly" ./emsdk install sdk-tag-1.37.28-64bit ./emsdk activate sdk-tag-1.37.28-64bit source ./emsdk_env.sh #this sets up necessary path variables for current terminal session
For more instructions for running emscripten to compile wasm see https://gist.github.com/nolash/910ac3892d48d2e70232c997ffa9d55e and Dockerfile at https://hub.docker.com/r/mrsmkl/coindrop/
Follow installation instructions for the ocaml-offchain interpreter here
If necessary edit prepare.js to include the correct path for wasm interpreter (default: ../ocaml-offchain/interpreter/wasm).
Also edit the IPFS host.
node prepare.js file.js
You can also follow along with this tutorial
Options:
--file fname: add the file to the IO block of the task--arg arg: add command line argument--float: add floating point emulation--out: the directory to store the results in, can be tmp if using ipfs.--upload-ipfs: upload to ipfs--ipfs-host: default is localhost
A successful output will look like this:
{
"ipfshash": "QmaDMzoEE51eNJCc1dMsxUKQGp41qvCsgbv3PHhcbKbAKe",
"codehash": "0xcc0d89e2f7ad0f720cdc6521ab698c1053dac534cd770fb6531d935975ee5d7e",
"info": {
"vm": {
"code": "0xcc0d89e2f7ad0f720cdc6521ab698c1053dac534cd770fb6531d935975ee5d7e",
"stack": "0xb4c11951957c6f8f642c4af61cd6b24640fec6dc7fc607ee8206a99e92410d30",
"memory": "0xb4c11951957c6f8f642c4af61cd6b24640fec6dc7fc607ee8206a99e92410d30",
"input_size": "0x593e5b969fbeac1646d534f40aeeb6d440f1b60353267ff7a67bb53a3a8f1125",
"input_name": "0x9f9a605ee9da9ebd0f0a58d289c2345d279c3e11baafdefe72bb5aa2ead36e38",
"input_data": "0x066ccee69369f2589250d208feef82cd3e06356124c01b9e9e8d56c9393e0e85",
"call_stack": "0xb4c11951957c6f8f642c4af61cd6b24640fec6dc7fc607ee8206a99e92410d30",
"globals": "0xb4c11951957c6f8f642c4af61cd6b24640fec6dc7fc607ee8206a99e92410d30",
"calltable": "0x7bf9aa8e0ce11d87877e8b7a304e8e7105531771dbff77d1b00366ecb1549624",
"calltypes": "0xb4c11951957c6f8f642c4af61cd6b24640fec6dc7fc607ee8206a99e92410d30",
"pc": 0,
"stack_ptr": 0,
"call_ptr": 0,
"memsize": 0
},
"hash": "0x9e10c72398fc17ba4fa0163ea8da8d1401abfe4ef93335b122df18498cac5da7"
},
"memsize": "25"
}