[compiler] Bytecode preparation fails for asm.js modules.

[compiler] Bytecode preparation fails for asm.js modules.

This handles the case where preparation of bytecode might fail inside
Compiler::EnsureBytecode due to the underlying function being a fully
validated asm.js module. We simply bailout of bytecode preparation.

R=bradnelson@chromium.org
TEST=mjsunit/regress/regress-crbug-644111
BUG=chromium:644111
Committed: https://crrev.com/cc1249b779fae9fbd019f6c6861bf1d1534f0a8c
Cr-Commit-Position: refs/heads/master@{#39187}

[Michael Starzinger, 2016-09-05 15:44:54]

The CQ bit was checked by mstarzinger@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-09-05 15:44:58]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-05 16:21:14]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-09-05 16:21:15]

Dry run: This issue passed the CQ dry run.

[bradn, 2016-09-05 23:00:45]

bradnelson@google.com changed reviewers:
+ bradnelson@google.com

[bradn, 2016-09-05 23:00:46]

lgtm

Sharing these fields seems error prone.
Wonder if there's a better way to avoid clashing.

[bradn, 2016-09-05 23:00:59]

Thanks for fixing though!

[bradn, 2016-09-05 23:01:03]

The CQ bit was checked by bradnelson@google.com

[commit-bot: I haz the power, 2016-09-05 23:01:12]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-09-05 23:03:02]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-09-05 23:03:15]

Description was changed from

==========
[compiler] Bytecode preparation fails for asm.js modules.

This handles the case where preparation of bytecode might fail inside
Compiler::EnsureBytecode due to the underlying function being a fully
validated asm.js module. We simply bailout of bytecode preparation.

R=bradnelson@chromium.org
TEST=mjsunit/regress/regress-crbug-644111
BUG=chromium:644111
==========

to

==========
[compiler] Bytecode preparation fails for asm.js modules.

This handles the case where preparation of bytecode might fail inside
Compiler::EnsureBytecode due to the underlying function being a fully
validated asm.js module. We simply bailout of bytecode preparation.

R=bradnelson@chromium.org
TEST=mjsunit/regress/regress-crbug-644111
BUG=chromium:644111

Committed: https://crrev.com/cc1249b779fae9fbd019f6c6861bf1d1534f0a8c
Cr-Commit-Position: refs/heads/master@{#39187}
==========

[commit-bot: I haz the power, 2016-09-05 23:03:16]

Patchset 1 (id:??) landed as
https://crrev.com/cc1249b779fae9fbd019f6c6861bf1d1534f0a8c
Cr-Commit-Position: refs/heads/master@{#39187}

[Michael Starzinger, 2016-09-06 07:37:30]

On 2016/09/05 23:00:46, bradn wrote:
> lgtm
> 
> Sharing these fields seems error prone.
> Wonder if there's a better way to avoid clashing.

Note that the actual bug is that {GetUnoptimizedCode} does not produce bytecode
at all. Even if the fields were not shared, this "fix" would still be necessary.
I actually think that the overlaying of the bytecode and the asm-wasm-data field
itself works out nicely.

[bradn, 2016-09-07 05:43:51]

Ah true, fair point!