News for hackers. Stuff that matters.
Re:Introducing SIMD.js by legulere on Thursday October 30, 2014 @05:55PM (#8536054)
Well LLVM heavily relies on not having forward compatibility (new versions spew out LLVM IR that can't be used by older versions) and backward compatibility is merely an afterthought.
Re:Introducing SIMD.js by hajile on Thursday October 30, 2014 @06:20PM (#8536174)
That's a standards issue. If w3c froze a particular version as the "one true version", then whatever experimentation the llvm guys want to do with whatever other versions doesn't matter.

Another solution is to version your code <script type="llvm" src="foo.ll" version="3.6.1" />

I'm already of the belief that w3c needs to require versioning of js code instead of stuff like 'use strict' (if it's bad, just remove it in new versions and move on). For backwards compatibility, if there's no version, assume ECMAScript3.

Re:Introducing SIMD.js by sanxiyn on Friday October 31, 2014 @12:39AM (#8537347)
Khronos Group standardized LLVM IR 3.2 as SPIR(Standard Portable Intermediate Representation). It is an existence proof that standardizing LLVM IR is possible.

https://www.khronos.org/spir/

Re:Introducing SIMD.js by legulere on Sunday November 02, 2014 @08:42AM (#8546555)
I actually work a bit with SPIR. They standardized a subset of LLVM IR 3.2 as SPIR 1.2

SPIR 2.0 on the other hand already is based off LLVM IR 3.4