top | item 26929610 (no title) Others | 4 years ago I worked on one as an undergrad: https://github.com/gwsystems/aWsmFull AoT compilation, C programs run within 10% of native discuss order hn newest heavyset_go|4 years ago 10% difference from native, or 10% of native performance? Others|4 years ago 10% worse performance. Not 10x worse :D moonchild|4 years ago Since it compiles using llvm, what's the reason for the discrepancy?Lost information about e.g. aliasing? Others|4 years ago Bounds checking, dynamic dispatch is more expensive, wasm semantics have no undefined behavior to exploit for optimizations. And there is bloat in the process. I’m sure losing information effects things as well
heavyset_go|4 years ago 10% difference from native, or 10% of native performance? Others|4 years ago 10% worse performance. Not 10x worse :D
moonchild|4 years ago Since it compiles using llvm, what's the reason for the discrepancy?Lost information about e.g. aliasing? Others|4 years ago Bounds checking, dynamic dispatch is more expensive, wasm semantics have no undefined behavior to exploit for optimizations. And there is bloat in the process. I’m sure losing information effects things as well
Others|4 years ago Bounds checking, dynamic dispatch is more expensive, wasm semantics have no undefined behavior to exploit for optimizations. And there is bloat in the process. I’m sure losing information effects things as well
heavyset_go|4 years ago
Others|4 years ago
moonchild|4 years ago
Lost information about e.g. aliasing?
Others|4 years ago