1. Delete prolog weak persistent handles when shutting down an isolate

1. Delete prolog weak persistent handles when shutting down an isolate
2. Remove the peer_ field from RawExternalTypedData as it is not used anywhere

R=iposva@google.com

Committed: https://code.google.com/p/dart/source/detail?r=31797

[Ivan Posva, 2014-01-14 02:08:27]

https://codereview.chromium.org/136853004/diff/1/runtime/vm/object.cc
File runtime/vm/object.cc (right):

https://codereview.chromium.org/136853004/diff/1/runtime/vm/object.cc#newcode...
runtime/vm/object.cc:16470: Isolate::Current()->heap()->SetPeer(this->raw(),
peer);
I don't think this is correct. Adding a VM internal finalizer should not have a
side-effect of setting the externally visible peer. For example the externally
visible peer for this type data could be some object that holds the data
internally.

[siva, 2014-01-14 16:44:14]

https://codereview.chromium.org/136853004/diff/1/runtime/vm/object.cc
File runtime/vm/object.cc (right):

https://codereview.chromium.org/136853004/diff/1/runtime/vm/object.cc#newcode...
runtime/vm/object.cc:16470: Isolate::Current()->heap()->SetPeer(this->raw(),
peer);
At least that is what the old code was doing. We seem to have a lot of
redundancy here, the peer is stored in multiple places.

On 2014/01/14 02:08:28, Ivan Posva wrote:
> I don't think this is correct. Adding a VM internal finalizer should not have
a
> side-effect of setting the externally visible peer. For example the externally
> visible peer for this type data could be some object that holds the data
> internally.

[siva, 2014-01-14 17:09:42]

removed setting of peer, PTAL.

[Ivan Posva, 2014-01-14 19:47:10]

LGTM -ip

[siva, 2014-01-14 19:55:51]

Committed patchset #2 manually as r31797 (presubmit successful).