| |||||||||||||
Google's Native Client forges aheadGoogle's Native Client forges aheadPosted Jul 28, 2011 7:55 UTC (Thu) by ikm (subscriber, #493)Parent article: Google's Native Client forges ahead
> As for PNaCl's promise to eliminate the architecture problem by targeting an intermediate byte-code representation instead that platform starts to sound more and more like client-side Java
This comparison is incorrect. LLVM uses bytecode too, but that doesn't make it resemble Java.
(Log in to post comments)
Google's Native Client forges ahead Posted Jul 28, 2011 8:46 UTC (Thu) by kragilkragil2 (guest, #76172) [Link]
AFAIK they use what they call it bitcode and it translate much more directly to native machine language. I can't be bothered to really check, but I guess it looks more like an machine indepentent assembly with more primitive idioms.
More like for 1+2 = move.l 1,d0; move.l 2,d1; add.l d0,d1; (my 68k assembly days are gone 20 years, not sure that is still correct code or if they changed it ;-)
But I guess someelse around here knows better.
Google's Native Client forges ahead Posted Jul 28, 2011 18:18 UTC (Thu) by n8willis (editor, #43041) [Link]
It doesn't say that they are the same, just that the need to compile to an intermediate representation which is then translated again is going to make the sales pitch more difficult thanks to Java's storied history.
Nate
|
|||||||||||||