intel
target web assembly for hyperscan build
I'm interested in building hyperscan for web assembly (probably with emscripten). My primary goal is to enable a single code base that uses hyperscan to be compiled for either native or in-browser execution, understanding that many performance optimizations may be unavailable in wasm. With that in mind, a couple questions come to mind:
a. other than the expected grunt work involves such as navigating compiler option differences and/or replacing certain SIMD operations such as mm_crc32_u8 with plain C code, are there any major obstacles or show stoppers that I should be considering first?
b. has anyone already tried this? Searching on github and google suggests no, at least as of this writing, but maybe I'm missing something
Any thoughts / suggestions would be much appreciated. And of course, if anyone is interested to join me in this endeavor, please let me know
You might want to take a look at https://github.com/intel/hyperscan/pull/367
