Debugger: correctly report uncaught rejections in Promise.all and Promise.race.

Debugger: correctly report uncaught rejections in Promise.all and Promise.race.

The debugger calls PromiseHasUserDefinedRejectHandler to recursively search the
tree of dependent promises for user-defined reject handlers. If no such reject
handler exists, rejecting the promise is considered an uncaught exception.

Promise.race and Promise.all interupt the link of promise dependency wrt the
search. This change fixes that link.

R=rossberg@chromium.org
BUG=chromium:439585
LOG=N

Committed: https://crrev.com/8be20eee3ba63b1d78eab065d9b6787b624cfff2
Cr-Commit-Position: refs/heads/master@{#31392}

[rossberg, 2015-10-19 14:24:29]

lgtm

https://codereview.chromium.org/1411083003/diff/1/src/js/promise.js
File src/js/promise.js (right):

https://codereview.chromium.org/1411083003/diff/1/src/js/promise.js#newcode355
src/js/promise.js:355: if
(PromiseHasUserDefinedRejectHandlerRecursive(deferred.promise)) {
Why the extra conditional and not a tail call?

return PromiseHasUserDefinedRejectHandlerRecursive(deferred.promise);

Similarly below.

[Yang, 2015-10-19 16:24:09]

https://codereview.chromium.org/1411083003/diff/1/src/js/promise.js
File src/js/promise.js (right):

https://codereview.chromium.org/1411083003/diff/1/src/js/promise.js#newcode355
src/js/promise.js:355: if
(PromiseHasUserDefinedRejectHandlerRecursive(deferred.promise)) {
On 2015/10/19 14:24:29, rossberg wrote:
> Why the extra conditional and not a tail call?
> 
> return PromiseHasUserDefinedRejectHandlerRecursive(deferred.promise);
> 
> Similarly below.

A promise is considered to have a reject handler if one of its dependent
promises is.

So if it returns false here, we have to continue the loop, visiting other tree
branches (other dependent promises).

[Yang, 2015-10-19 16:24:23]

The CQ bit was checked by yangguo@chromium.org

[commit-bot: I haz the power, 2015-10-19 16:24:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1411083003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1411083003/1

[commit-bot: I haz the power, 2015-10-19 16:26:47]

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

[commit-bot: I haz the power, 2015-10-19 16:26:48]

Try jobs failed on following builders:
  v8_presubmit on tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_presubmit/builds/6810)

[Yang, 2015-10-19 16:30:57]

yangguo@chromium.org changed reviewers:
+ hpayer@chromium.org

[Yang, 2015-10-19 16:30:57]

Hannes, please look at the heap changes.

[Yang, 2015-10-20 05:15:49]

https://codereview.chromium.org/1411083003/diff/20001/src/js/promise.js
File src/js/promise.js (right):

https://codereview.chromium.org/1411083003/diff/20001/src/js/promise.js#newco...
src/js/promise.js:333: this.resolve(value).then(function(x) {
deferred.resolve(x) }, reject);
then() may not even return a Promise, if it's overridden (see failing test262
case). Instead of putting the deferred object on the return value of then(), we
put it onto the reject handler.

[Yang, 2015-10-20 05:15:54]

The CQ bit was checked by yangguo@chromium.org

[Yang, 2015-10-20 05:15:55]

The patchset sent to the CQ was uploaded after l-g-t-m from
rossberg@chromium.org
Link to the patchset: https://codereview.chromium.org/1411083003/#ps20001
(title: "move deferred from promise to handler")

[commit-bot: I haz the power, 2015-10-20 05:16:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1411083003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1411083003/20001

[Hannes Payer (out of office), 2015-10-20 05:17:17]

heap LGTM

[commit-bot: I haz the power, 2015-10-20 05:40:06]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-10-20 05:40:21]

Patchset 2 (id:??) landed as
https://crrev.com/8be20eee3ba63b1d78eab065d9b6787b624cfff2
Cr-Commit-Position: refs/heads/master@{#31392}