Correctly handle clearing of deprecated field types.

Correctly handle clearing of deprecated field types.

BUG=v8:4027
LOG=NO

Committed: https://crrev.com/68a7773e0f218def0f39a1ee37972695f09a7c4e
Cr-Commit-Position: refs/heads/master@{#27837}

[ulan, 2015-04-15 09:06:56]

ulan@chromium.org changed reviewers:
+ jkummerow@chromium.org, verwaest@chromium.org

[ulan, 2015-04-15 09:06:56]

PTAL

[Jakob Kummerow, 2015-04-15 09:25:16]

LGTM

https://codereview.chromium.org/1086063003/diff/1/test/mjsunit/regress/regres...
File test/mjsunit/regress/regress-4027.js (right):

https://codereview.chromium.org/1086063003/diff/1/test/mjsunit/regress/regres...
test/mjsunit/regress/regress-4027.js:55: for (var i = 0; i < 3; i++) gc();
Isn't a single GC sufficient?

[ulan, 2015-04-15 09:31:44]

https://codereview.chromium.org/1086063003/diff/1/test/mjsunit/regress/regres...
File test/mjsunit/regress/regress-4027.js (right):

https://codereview.chromium.org/1086063003/diff/1/test/mjsunit/regress/regres...
test/mjsunit/regress/regress-4027.js:55: for (var i = 0; i < 3; i++) gc();
On 2015/04/15 09:25:16, Jakob wrote:
> Isn't a single GC sufficient?

Done. Yes, assert triggers with single GC.

[ulan, 2015-04-15 09:31:52]

The CQ bit was checked by ulan@chromium.org

[ulan, 2015-04-15 09:31:53]

The patchset sent to the CQ was uploaded after l-g-t-m from
jkummerow@chromium.org
Link to the patchset: https://codereview.chromium.org/1086063003/#ps20001
(title: "just do one GC in test")

[commit-bot: I haz the power, 2015-04-15 09:32:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1086063003/20001

[commit-bot: I haz the power, 2015-04-15 09:55:45]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-15 09:55:58]

Patchset 2 (id:??) landed as
https://crrev.com/68a7773e0f218def0f39a1ee37972695f09a7c4e
Cr-Commit-Position: refs/heads/master@{#27837}