1. There isn't an especially good way to debug that, except for enabling assertions and looking at the errors that come up when the code is run. But generally the code size difference is very small, except for small programs where the function table size matters (usually code is far bigger anyhow).
2. Yes, as the error says, that option is no longer functional. In practice unaligned loads/stores are rare enough that they don't affect perf, except for special cases. I think that's why we never got around to updating the option for fastcomp. Both of those issues, btw, will go away in wasm (it allows unaligned accesses, and it has better ways to do function pointers). On Tue, Jan 3, 2017 at 6:14 AM, Gaurav Dewan <[email protected]> wrote: > I was investigation various optimization options by referring to here: > https://kripken.github.io/emscripten-site/docs/optimizing/Optimizing-Code. > html > https://github.com/kripken/emscripten/blob/master/src/settings.js > > (1) Is there any way to debug and find the culprits which cause failure > with ALIASING_FUNCTION_POINTERS option and fix those (that is, when not > using -s ALIASING_FUNCTION_POINTERS=0 with O3)? What is the gain generally > expected (size and performance) with this option ? > > > (2) Is FORCE_ALIGNED_MEMORY still supported? > I noticed FORCE_ALIGNED_MEMORY=1 on commandline causes toolchain error > immediately that it is not compatible with fastcomp(latest incoming). > Anyways, WARN_UNALIGNED=1 warns about number of unaligned store including > standard libraries like std::money_put etc - so I guess it would be very > less useful. > > > -- > 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. > -- 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.
