Are there methods to optimize these calls out of wasm (change code or use 
compiler options or add to webassembly)

(1) f64 to int - there is pattern (promote f64, and call 50 - f64 to int 
outside wasm)  - BINARYEN imprecise option is removed. I see this is the 
only asm-wasm imports getting called in my application.
https://github.com/kripken/emscripten/issues/4885
I am using default trap mode(not specifying anything).BINARYEN_IMPRECISE is 
not present in new toolchain versions.

(2) emscriptenmemcpy_big with 32768 bytes - is it some performance 
technique or necessary evil ?

(3) clock_get_time - perhaps there is no wasm instructions for this.

Chrome devtools profiler does not seem to profile inside wasm for 
performance(it just shows it as a monolithic blackbox) - only calls out to 
js are shown. Same with JS profiler tabs (although debugger is able to 
disassemble wasm). 
Has FF nightly started WASM debugging and profiling now ?
I remember seeing a snapshot of profiler with fast trampoline (calls from 
wasm to asm.js are shown) somewhere.

(4) Shadow stack - If application is taking address of local variable, wasm 
size can increase due to shadowing. Does it mean that even if I take 
address of a local variable in just one function, the application will 
always maintain the shadow stack ? Or does it increase proportionally with 
such instances ?
How can I spot such pattern in my application ?

Thanks

-- 
You received this message because you are subscribed to the Google Groups 
"emscripten-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to