Eagerly declare eval scopes, even for sloppy scopes

Eagerly declare eval scopes, even for sloppy scopes

R=mstarzinger@chromium.org
LOG=N
BUG=N

Committed: https://crrev.com/9510a9c3ef55489d38ab0adbe3898efa2a734c95
Cr-Commit-Position: refs/heads/master@{#28072}

[wingo, 2015-04-27 10:21:53]

This is a reapplication of https://codereview.chromium.org/1085263003.  Patchset
1 is the original patchset as reviewed, which caused a regression on CodeLoad. 
Patchset 2 fixes the regression by restoring lazy parsing of sloppy eval inside
script scopes.  I didn't realize that lazy parsing of eval code was important --
I guess to allow nested functions to themselves be lazily parsed?  Anyway, PTAL
:-)

[Michael Starzinger, 2015-04-27 10:44:03]

LGTM (only reviewed diff between patch set #1 and #2).

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc
File src/parser.cc (right):

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc#newcode993
src/parser.cc:993: // Enter 'scope' with the given parsing mode..
nit: One dot oughta be enough for everyone.

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc#newcode1273
src/parser.cc:1273: if (scope_->is_eval_scope()) mode_ = PARSE_EAGERLY;
nit: Needs a comment.

[wingo, 2015-04-27 11:43:06]

Patchset #3 (id:40001) has been deleted

[wingo, 2015-04-27 11:44:25]

Thanks for the review :)  I've tested the perf locally on code-load and it looks
good, but let's keep an eye on the perf bots too.

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc
File src/parser.cc (right):

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc#newcode993
src/parser.cc:993: // Enter 'scope' with the given parsing mode..
On 2015/04/27 10:44:03, Michael Starzinger wrote:
> nit: One dot oughta be enough for everyone.

Done.

https://codereview.chromium.org/1110473003/diff/20001/src/parser.cc#newcode1273
src/parser.cc:1273: if (scope_->is_eval_scope()) mode_ = PARSE_EAGERLY;
On 2015/04/27 10:44:03, Michael Starzinger wrote:
> nit: Needs a comment.

Done, though it's speculation to an extent; it replicates the original logic but
that didn't have a comment either.  Commenting out line 1280 causes the
mentioned test to fail though, so there is that.

[wingo, 2015-04-27 11:45:02]

The CQ bit was checked by wingo@igalia.com

[wingo, 2015-04-27 11:45:03]

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/1110473003/#ps60001
(title: "Fix nits")

[commit-bot: I haz the power, 2015-04-27 11:45:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1110473003/60001

[commit-bot: I haz the power, 2015-04-27 12:13:50]

Committed patchset #3 (id:60001)

[commit-bot: I haz the power, 2015-04-27 12:14:05]

Patchset 3 (id:??) landed as
https://crrev.com/9510a9c3ef55489d38ab0adbe3898efa2a734c95
Cr-Commit-Position: refs/heads/master@{#28072}