Skip to main content

Build on RISC-V 64

Prepare the Environment​

This tutorial is based on Ubuntu 22.04 host, and WasmEdge uses the RISCV-Lab, which provides Ubuntu 22.04 system with riscv64 architecture. Here users can use their riscv64 environment.

Install and run RISCV-Lab​

git clone https://gitee.com/tinylab/cloud-lab.git
cd cloud-lab
LOGIN=bash tools/docker/run riscv-lab

Note that it will take a long time to pull the image here.

Build WasmEdge​

Get Source code​

ubuntu@riscv-lab:/labs/riscv-lab$ git clone https://github.com/WasmEdge/WasmEdge.git
ubuntu@riscv-lab:/labs/riscv-lab$ cd WasmEdge

Dependencies​

WasmEdge requires LLVM 12 at least and you may need to install the following dependencies yourself.

ubuntu@riscv-lab:/labs/riscv-lab$ sudo apt-get update
ubuntu@riscv-lab:/labs/riscv-lab$ sudo apt install -y software-properties-common cmake libboost-all-dev
ubuntu@riscv-lab:/labs/riscv-lab$ sudo apt install -y llvm-12-dev liblld-12-dev

Compile​

Please refer to here for the descriptions of all CMake options.

ubuntu@riscv-lab:/labs/riscv-lab$ cd WasmEdge
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge$ mkdir -p build && cd build
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/build$ cmake -DCMAKE_BUILD_TYPE=Release .. && make -j

Test​

Execute the wasmedge tool​

For the pure WebAssembly, the wasmedge CLI tool will execute it in interpreter mode.

ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/build$ sudo make install
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/build$ cd ../examples/wasm
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ wasmedge -v
wasmedge version 0.12.0-alpha.1-13-g610cc21f
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ wasmedge --reactor fibonacci.wasm fib 10
89
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ wasmedge --reactor add.wasm add 2 2
4

Execute wasmedge compile​

To improve the performance, the wasmedge compile can compile WebAssembly into native machine code. After compiling with the wasmedge compile AOT compiler, the wasmedge tool can execute the WASM in AOT mode, which is much faster.

ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ wasmedge compile fibonacci.wasm fibonacci_aot.wasm
[2023-02-01 22:39:15.807] [info] compile start
[2023-02-01 22:39:15.857] [info] verify start
[2023-02-01 22:39:15.866] [info] optimize start
[2023-02-01 22:39:16.188] [info] codegen start
[2023-02-01 22:39:16.403] [info] output start
[2023-02-01 22:39:16.559] [info] compile done
[2023-02-01 22:39:16.565] [info] output start
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ time wasmedge --reactor fibonacci_aot.wasm fib 30
1346269
real    0m0.284s
user    0m0.282s
sys     0m0.005s
ubuntu@riscv-lab:/labs/riscv-lab/WasmEdge/examples/wasm$ time wasmedge --reactor fibonacci.wasm fib 30
1346269
real    0m1.814s
user    0m1.776s
sys     0m0.016s