@BrendanEich It would be very easy to do on asm.js if there was a sync API for WebCrypto in workers, or if there was wait/async in JS.
-
-
Replying to @BRIAN_____
@BRIAN_____ I keep nagging@SickingJ about that, it is supposedly coming (wait in worker on promise).1 reply 0 retweets 1 like -
Replying to @BrendanEich
@BrendanEich@SickingJ A sync API would be better for performance in this case, at least for symmetric crypto. Wasteful to simulate sync.1 reply 0 retweets 1 like -
Replying to @BRIAN_____
@BRIAN_____@BrendanEich@SickingJ Is that objective or subjective? Just curious which part you see wasteful and if that's fundamental1 reply 0 retweets 0 likes -
Replying to @sleevi_
@BRIAN_____@BrendanEich@SickingJ I ask because I went back and forth with this with@slightlylate , and convinced me it's impl dependent2 replies 0 retweets 0 likes -
Replying to @sleevi_
@sleevi_@BrendanEich@SickingJ@slightlylate It's impl dependent, in theory. After all a really smart JIT could convert async to sync.3 replies 0 retweets 0 likes -
Replying to @BRIAN_____
@BRIAN_____ No, JIT = Just In Time. Whole-program analysis is AOT.@sleevi_@SickingJ@slightlylate2 replies 0 retweets 1 like -
Replying to @BrendanEich
@BrendanEich@sleevi_ Has to be done by the JIT/runtime because the code needs to call an internal sync. version the AoT compiler can't see1 reply 0 retweets 0 likes -
Replying to @BRIAN_____
@BRIAN_____ You're describing something different from async/await, here -- but what? Got a gist?@sleevi_1 reply 0 retweets 0 likes -
Replying to @BrendanEich
@BrendanEich@sleevi_ I mean transforming asnyc + await sequence into a synchronous execution of the body of the asynchronous function.1 reply 0 retweets 0 likes
@BrendanEich @sleevi_ e.g. write a synchronous SHA256 function in terms of the async WebCrypto digest API w/o any async overhead.
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.