Performance issue related to operations on floating-point (FP) arrays

[ShuyaoJiang]

Summary

Hi, I ran the attached case (C program, compiled to Wasm by Emscripten) in different Wasm runtimes, and found abnormal performance in Wasmer compared with other runtimes. The execution time of this case (time interval from the start to the end of the execution of Wasm bytecode running command) on different runtimes is as follows:

• Wasmer: 981ms
• Wasmtime: 28ms
• WasmEdge (AOT): 14ms
• WAMR (AOT): 10ms

Emscripten

• emcc (Emscripten gcc/clang-like replacement + linker emulating GNU ld) 3.1.49 (04a0cad4d4c9f5d62876821274d78cd0a52427af)
  clang version 18.0.0 (https://github.com/llvm/llvm-project 269685545e439ad050b67740533c59f965cae955)
  Target: wasm32-unknown-emscripten
  Thread model: posix

Wasm Runtime Version

• Wasmer: 4.2.3
• Wasmtime: cli 15.0.0
• WasmEdge: 0.13.5
• WAMR: 1.2.3

Hardware & OS

• CPU: Intel(R) Xeon(R) E5-2686 v4 CPU @ 2.30GHz
• Memory: 16GB
• OS: Ubuntu 20.04.6 LTS

Additional details

The attached source program is synthesized by a Csmith seed and a code snippet from another program. The inserted code snippet is on lines 67-68 of the source program, which is an operation on floating-point (FP) arrays (the code snippet is attached below). So, we think that this abnormal performance may be caused by some improper handling of this operation. Could you please check this situation? Thank you!

for (print_hash_value = 0; print_hash_value < 1; print_hash_value++)
        float2_temp_3[print_hash_value] = 0x5.Ep+1;

1702343309202.zip

[linear]

RUN-34 Performance issue related to operations on floating-point (FP) arrays

[Arshia001]

Hello @ShuyaoJiang, thanks for reporting this! Can you please let us know which compiler backend you used when running the WASM file? For reference, wasmer has 3 backends: singlepass, cranelift and llvm.

[Arshia001]

Also, we included a number of performance related improvements in the recently released version 4.2.5. Can you try that version as well?

[Arshia001]

One more thing, was this the first time you were running the code? Wasmer compiles and caches modules the first time they are run.

[ShuyaoJiang]

Hi, I used cranelift backend. Also, I ran the test case multiple times and obtained similar results.

[Arshia001]

Thank you.

This is in part due to the fact that we're currently using an older version of cranelift, which we should update to the latest version. Tagging @theduke.