Just visit young array buffers during scavenge.

Just visit young array buffers during scavenge. Additionally keep the views in new space in a separate global list and move them to the corresponding array buffers when they get promoted.

This reduces young generation garbage collections when many array buffers are allocated.
BUG=

Committed: https://crrev.com/bd61a85fafd6461a40dc1e20252fd843f148e837
Cr-Commit-Position: refs/heads/master@{#26605}

[Hannes Payer (out of office), 2015-02-06 12:21:10]

hpayer@chromium.org changed reviewers:
+ dslomov@chromium.org, kbr@chromium.org, ulan@chromium.org

[Hannes Payer (out of office), 2015-02-06 12:21:11]

PTAL.

https://codereview.chromium.org/904633003/diff/40001/src/runtime/runtime-type...
File src/runtime/runtime-typedarray.cc (right):

https://codereview.chromium.org/904633003/diff/40001/src/runtime/runtime-type...
src/runtime/runtime-typedarray.cc:107: // that belong to the neutered array
buffer.
Dmitry, when we neuter, do we still have to keep the array buffer to view
relationship? I.e., would we be allowed to remove the views from the new space
list of views?

https://codereview.chromium.org/904633003/diff/60001/src/heap/heap.cc
File src/heap/heap.cc (right):

https://codereview.chromium.org/904633003/diff/60001/src/heap/heap.cc#newcode...
src/heap/heap.cc:1731: // and move them to the corresponding array buffer.
We could do this in one pass with a specialized Visitor.

[ulan, 2015-02-06 14:36:46]

> We could do this in one pass with a specialized Visitor.
+1, that seems much better.

https://codereview.chromium.org/904633003/diff/80001/src/heap/heap.cc
File src/heap/heap.cc (right):

https://codereview.chromium.org/904633003/diff/80001/src/heap/heap.cc#newcode...
src/heap/heap.cc:1754: }
Update new_array_buffer_views_list if the head was promoted?

https://codereview.chromium.org/904633003/diff/80001/src/heap/objects-visitin...
File src/heap/objects-visiting.cc (right):

https://codereview.chromium.org/904633003/diff/80001/src/heap/objects-visitin...
src/heap/objects-visiting.cc:237: !heap->InNewSpace(original_candidate)) {
Let's DCHECK that the remaining objects are in old space.

[Hannes Payer (out of office), 2015-02-10 14:41:31]

https://codereview.chromium.org/904633003/diff/80001/src/heap/heap.cc
File src/heap/heap.cc (right):

https://codereview.chromium.org/904633003/diff/80001/src/heap/heap.cc#newcode...
src/heap/heap.cc:1754: }
On 2015/02/06 14:36:46, ulan wrote:
> Update new_array_buffer_views_list if the head was promoted?

Done.

https://codereview.chromium.org/904633003/diff/80001/src/heap/objects-visitin...
File src/heap/objects-visiting.cc (right):

https://codereview.chromium.org/904633003/diff/80001/src/heap/objects-visitin...
src/heap/objects-visiting.cc:237: !heap->InNewSpace(original_candidate)) {
On 2015/02/06 14:36:46, ulan wrote:
> Let's DCHECK that the remaining objects are in old space.

I moved the DCHECK to the call site. The invariant is not true for pretenured
allocation sites.

[ulan, 2015-02-11 16:38:00]

lgtm

[Hannes Payer (out of office), 2015-02-12 09:13:58]

The CQ bit was checked by hpayer@chromium.org

[commit-bot: I haz the power, 2015-02-12 09:14:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/904633003/140001

[commit-bot: I haz the power, 2015-02-12 10:00:34]

Committed patchset #8 (id:140001)

[commit-bot: I haz the power, 2015-02-12 10:00:59]

Patchset 8 (id:??) landed as
https://crrev.com/bd61a85fafd6461a40dc1e20252fd843f148e837
Cr-Commit-Position: refs/heads/master@{#26605}

[Michael Achenbach, 2015-02-16 07:22:09]

A revert of this CL (patchset #8 id:140001) has been created in
https://codereview.chromium.org/929933002/ by machenbach@chromium.org.

The reason for reverting is: Introduces crashes..