Add full TurboFan support for accessing SeqString contents.

Add full TurboFan support for accessing SeqString contents.

LOG=n

Committed: https://crrev.com/4518e927067055035d14a8b777fbac4401fd5c7d
Cr-Commit-Position: refs/heads/master@{#27465}

[Sven Panne, 2015-03-25 13:52:46]

svenpanne@chromium.org changed reviewers:
+ mstarzinger@chromium.org

[Michael Starzinger, 2015-03-25 17:13:21]

LGTM to this approach (modulo failing tests due to return value).

[Michael Starzinger, 2015-03-25 17:41:55]

The more I think about it ...

https://codereview.chromium.org/1013753016/diff/1/src/compiler/js-intrinsic-l...
File src/compiler/js-intrinsic-lowering.cc (right):

https://codereview.chromium.org/1013753016/diff/1/src/compiler/js-intrinsic-l...
src/compiler/js-intrinsic-lowering.cc:263:
NodeProperties::ReplaceWithValue(node, node, node);
Let's just use "string" as the second argument to ReplaceWithValue here, this is
more correct and robust in general and also solves the problem with the missing
return value in the tests. Because it would be way to easy to write such a
construct by accident in the builtin JavaScript code. The more I think about it,
the more it feels like a good rule-of-thumb to let every intrinsic have a
default return value just to keep the graph structurally intact.

[Sven Panne, 2015-03-26 08:02:34]

PTAL...

https://codereview.chromium.org/1013753016/diff/1/src/compiler/js-intrinsic-l...
File src/compiler/js-intrinsic-lowering.cc (right):

https://codereview.chromium.org/1013753016/diff/1/src/compiler/js-intrinsic-l...
src/compiler/js-intrinsic-lowering.cc:263:
NodeProperties::ReplaceWithValue(node, node, node);
On 2015/03/25 17:41:55, Michael Starzinger wrote:
> Let's just use "string" as the second argument to ReplaceWithValue here, this
is
> more correct and robust in general and also solves the problem with the
missing
> return value in the tests. Because it would be way to easy to write such a
> construct by accident in the builtin JavaScript code. The more I think about
it,
> the more it feels like a good rule-of-thumb to let every intrinsic have a
> default return value just to keep the graph structurally intact.

Done.

[Sven Panne, 2015-03-26 08:29:56]

The CQ bit was checked by svenpanne@chromium.org

[Sven Panne, 2015-03-26 08:29:57]

The patchset sent to the CQ was uploaded after l-g-t-m from
mstarzinger@chromium.org
Link to the patchset: https://codereview.chromium.org/1013753016/#ps20001
(title: "Fixed return value")

[commit-bot: I haz the power, 2015-03-26 08:30:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1013753016/20001

[commit-bot: I haz the power, 2015-03-26 08:36:33]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-03-26 08:36:45]

Patchset 2 (id:??) landed as
https://crrev.com/4518e927067055035d14a8b777fbac4401fd5c7d
Cr-Commit-Position: refs/heads/master@{#27465}