Add 0xb demo bits

[lukewagner]

This PR adds a (currently candidate) .wasm binary that matches the 0xb branch binary encoding in design. The only thing missing (which I'll upload tomorrow assuming design/#664 is merged) is the br_table value. I confirmed locally that it runs in the three configurations: no wasm (which just has the message), FF nightly (0xa), FF with patches (0xb).

Happy to get any feedback and update the PR with new binary bits if any issues are found. I'll of course wait to confirm this decodes in at least one other engine before merging. (cc @binji @kripken @sunfishcode )

[kripken]

lgtm. feel free to ignore, but i would be happier with {} here

[lukewagner]

sure; the style in this file is all over

[lukewagner]

Now with br_table values. @kripken says SM bits run in Binaryen's 0xb interpreter. @binji / @titzer: lgty?

[titzer]

I haven't tried the demo bits yet but we are basically there functionally.
I'll take a look at the bits tomorrow.

On Fri, Apr 22, 2016 at 6:30 PM, Luke Wagner notifications@github.com
wrote:

Now with br_table values. @kripken https://github.com/kripken says SM
bits run in Binaryen's 0xb interpreter. @binji https://github.com/binji
/ @titzer https://github.com/titzer: lgty?

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
https://github.com/WebAssembly/webassembly.github.io/pull/5#issuecomment-213499963

[titzer]

Have you had a look at the burndown list recently?

On Fri, Apr 22, 2016 at 6:47 PM, Ben L. Titzer titzer@google.com wrote:

I haven't tried the demo bits yet but we are basically there functionally.
I'll take a look at the bits tomorrow.

On Fri, Apr 22, 2016 at 6:30 PM, Luke Wagner notifications@github.com
wrote:

Now with br_table values. @kripken https://github.com/kripken says SM
bits run in Binaryen's 0xb interpreter. @binji https://github.com/binji
/ @titzer https://github.com/titzer: lgty?

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
https://github.com/WebAssembly/webassembly.github.io/pull/5#issuecomment-213499963

[lukewagner]

@titzer Yes, it looks like the only diff from what is "titzer:CL" is the new form field in the types section. Once you add that, I expect V8 should decode these bits.

[titzer]

Looks like we never landed a design PR for arities in call, return, and break?

[titzer]

We need to rebase WebAssembly/design#595 onto binary_0xb and land that.

[lukewagner]

Alright, assuming @rossberg-chromium can get that rebased, here are the updated bits with arities (and, to be clear, the nop values removed from branches/return when arity=0).

[lukewagner]

How's it looking?

[s3ththompson]

@titzer It looks like we still need to land the experimentalVersion property?

[titzer]

LGTM.

I've verified that Chrome is able to run these demo bits with this patch (
https://codereview.chromium.org/1896863003/).

-B

On Thu, Apr 28, 2016 at 6:36 AM, Seth Thompson notifications@github.com
wrote:

@titzer https://github.com/titzer It looks like we still need to land
the experimentalVersion property?

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
https://github.com/WebAssembly/webassembly.github.io/pull/5#issuecomment-215308028

[titzer]

I also verified the backup path works (i.e. falling back to 0xa) in Chrome.

On Thu, Apr 28, 2016 at 2:32 PM, Ben L. Titzer titzer@google.com wrote:

LGTM.

I've verified that Chrome is able to run these demo bits with this patch (
https://codereview.chromium.org/1896863003/).

-B

On Thu, Apr 28, 2016 at 6:36 AM, Seth Thompson notifications@github.com
wrote:

@titzer https://github.com/titzer It looks like we still need to land
the experimentalVersion property?

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
https://github.com/WebAssembly/webassembly.github.io/pull/5#issuecomment-215308028

[lukewagner]

Commence landing!