After moving unreachable weak global handles only process harmony collections

After moving unreachable weak global handles only process harmony collections

Groups and implicit references are no longer relevant at this point.

Also add tests that fail if the first or second round of ephemeral
marking is omitted

BUG=none
R=mstarzinger@chromium.org,hpayer@chromium.org
LOG=n

[Hannes Payer (out of office), 2014-12-01 09:16:50]

lgtm

https://codereview.chromium.org/761343004/diff/1/src/heap/mark-compact.cc
File src/heap/mark-compact.cc (right):

https://codereview.chromium.org/761343004/diff/1/src/heap/mark-compact.cc#new...
src/heap/mark-compact.cc:2219: ProcessEphemeralMarking(&root_visitor, false);
I thought you can skip the object groups here?

[Michael Starzinger, 2014-12-02 10:22:42]

LGTM, and congratulations, you just inherited all bugs that have to do with the
weak transitive closure not being complete. :)

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc
File src/heap/mark-compact.cc (right):

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc...
src/heap/mark-compact.cc:2208: // reachable only from weak global handles.
nit: This dangling part of the sentence looks like a left-over, can we drop it?

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc...
src/heap/mark-compact.cc:2218: // the weak roots we just marked as pending
destruction.
This requires more explanation as to why we can skip other embedder-specific
marking here, e.g. all objects groups have been fully processed and no weakly
reachable node can discover new object groups.

[jochen (gone - plz use gerrit), 2014-12-02 10:37:18]

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc
File src/heap/mark-compact.cc (right):

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc...
src/heap/mark-compact.cc:2208: // reachable only from weak global handles.
On 2014/12/02 at 10:22:42, Michael Starzinger wrote:
> nit: This dangling part of the sentence looks like a left-over, can we drop
it?

done

https://codereview.chromium.org/761343004/diff/20001/src/heap/mark-compact.cc...
src/heap/mark-compact.cc:2218: // the weak roots we just marked as pending
destruction.
On 2014/12/02 at 10:22:42, Michael Starzinger wrote:
> This requires more explanation as to why we can skip other embedder-specific
marking here, e.g. all objects groups have been fully processed and no weakly
reachable node can discover new object groups.

done

[jochen (gone - plz use gerrit), 2014-12-02 10:37:30]

The CQ bit was checked by jochen@chromium.org

[commit-bot: I haz the power, 2014-12-02 10:38:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/761343004/40001

[commit-bot: I haz the power, 2014-12-02 11:23:56]

Committed patchset #3 (id:40001)