[regexp] simplify unanchored advance for unicode regexps.

[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org

Committed: https://crrev.com/33c78c39d7ea624ca2635ce42a27c069d06d9da3
Cr-Commit-Position: refs/heads/master@{#33838}

[Yang, 2016-02-08 18:24:44]

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

[commit-bot: I haz the power, 2016-02-08 18:24:56]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1676293003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1676293003/1

[commit-bot: I haz the power, 2016-02-08 18:54:00]

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

[commit-bot: I haz the power, 2016-02-08 18:54:01]

Dry run: This issue passed the CQ dry run.

[Yang, 2016-02-09 06:43:10]

Description was changed from

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly and allows us to use
greedy loop for the unanchored advance, which avoids pushing
backtracking onto the stack, preventing stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org
==========

to

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org
==========

[Yang, 2016-02-09 06:43:19]

On 2016/02/08 18:54:01, commit-bot: I haz the power wrote:
> Dry run: This issue passed the CQ dry run.

Fixed CL description.

[erikcorry, 2016-02-09 08:04:46]

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

[erikcorry, 2016-02-09 08:04:47]

LGTM

https://codereview.chromium.org/1676293003/diff/1/test/mjsunit/harmony/unicod...
File test/mjsunit/harmony/unicode-regexp-unanchored-advance.js (right):

https://codereview.chromium.org/1676293003/diff/1/test/mjsunit/harmony/unicod...
test/mjsunit/harmony/unicode-regexp-unanchored-advance.js:7: var s =
"a".repeat(1E8) + "\u1234";
I think this will time out on ARM.

[Yang, 2016-02-09 08:26:26]

On 2016/02/09 08:04:47, erikcorry wrote:
> LGTM
> 
>
https://codereview.chromium.org/1676293003/diff/1/test/mjsunit/harmony/unicod...
> File test/mjsunit/harmony/unicode-regexp-unanchored-advance.js (right):
> 
>
https://codereview.chromium.org/1676293003/diff/1/test/mjsunit/harmony/unicod...
> test/mjsunit/harmony/unicode-regexp-unanchored-advance.js:7: var s =
> "a".repeat(1E8) + "\u1234";
> I think this will time out on ARM.

1E7 works to reproduce the issue.

[Yang, 2016-02-09 08:26:30]

The CQ bit was checked by yangguo@chromium.org

[Yang, 2016-02-09 08:26:30]

The patchset sent to the CQ was uploaded after l-g-t-m from erikcorry@google.com
Link to the patchset: https://codereview.chromium.org/1676293003/#ps20001
(title: "shorten test")

[commit-bot: I haz the power, 2016-02-09 08:26:34]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1676293003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1676293003/20001

[commit-bot: I haz the power, 2016-02-09 09:11:04]

Description was changed from

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org
==========

to

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org
==========

[commit-bot: I haz the power, 2016-02-09 09:11:05]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-02-09 09:11:34]

Description was changed from

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org
==========

to

==========
[regexp] simplify unanchored advance for unicode regexps.

When doing advance at the start of an unanchored unicode regexp,
we do not have to care about surrogate pairs. If we actually advance
into the middle of a surrogate pair, the only choice is to also
consume trail surrogate as nothing else can match from there.

This reduces the emitted code slightly. By not having choice in the
loop, we do not have to push backtrack onto the stack, preventing
stack overflow.

R=erik.corry@gmail.com, erikcorry@chromium.org

Committed: https://crrev.com/33c78c39d7ea624ca2635ce42a27c069d06d9da3
Cr-Commit-Position: refs/heads/master@{#33838}
==========

[commit-bot: I haz the power, 2016-02-09 09:11:35]

Patchset 2 (id:??) landed as
https://crrev.com/33c78c39d7ea624ca2635ce42a27c069d06d9da3
Cr-Commit-Position: refs/heads/master@{#33838}