this statement is bullshit. not only there are no issues writing a general purpose compiler targeting wasm so long as it doesn't create irreducible CFGs, but also convering irreducible CFGs to reducible is a perfectly normal transformation that has nothing to do with obfuscators
-
This Tweet is unavailable.
-
I know that it's how obfuscators work, but AFAICT it's just a coincidence in that it's one of the simplest CFG transformations you can do, and DRM isn't somewhere you go look for technical excellence
I know you don't need to do this for everything, I'm just amused that it comes down to having to do the same transformation in corner cases (and you need to support corner cases, so you need to write the algorithm into your compiler).
-
-
But I clearly over-dramatized it so I'll retract the tweet.
-
I appreciate it!
End of conversation
New conversation -
-
-
sort of, these corner cases only unavoidably arise from gotos and their equivalents, like tail calls. if you're compiling something with neither gotos nor tail calls and your passes don't produce irreducible CFGs you don't need to deal with it
-
I think originally the designers of wasm thought that restricting it to reducible CFGs only would provide some benefits to the backend but I think these benefits also never materialized, which is pretty unfortunate in retrospect
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.