[tracing] Fix JSON format for memory allocator dumps.

[tracing] Fix allocator dump JSON format and roll trace-viewer.

This is to align the attributes used by memory allocator dump with
telemetry values [1] introducing types and units as attributes args.
Follow up patches will clean up the C++ interface to match the new
model.

This also rolls src/third_party/trace-viewer bd416ea:ba605a2 to pick up
the corresponding trace-viewer UI changes.

Summary of changes available at:
https://chromium.googlesource.com/external/trace-viewer/+log/bd416ea..ba605a2

BUG=466141
Committed: https://crrev.com/aeb8e8930e6a836dd77325ecef80818dfa1cb9f4
Cr-Commit-Position: refs/heads/master@{#327058}

[Primiano Tucci (use gerrit), 2015-04-27 10:05:10]

primiano@chromium.org changed reviewers:
+ petrcermak@chromium.org

[Primiano Tucci (use gerrit), 2015-04-27 10:05:17]

PTAL

[petrcermak, 2015-04-27 14:22:14]

LGTM with a couple of comments.

I tested the patch and it works fine with V8 instrumentation
(https://codereview.chromium.org/1088683003/) and the accordingly modified Trace
Viewer source code (https://codereview.appspot.com/232040043/).

Thanks,
Petr

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
File base/trace_event/memory_allocator_dump.cc (right):

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:80: // new generalized attribute
format. This code will be refactored by May 2015.
nit: "by May 2015" is ambiguous. Do you mean "by the end of May 2015"?

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:109: // on every event. The ability of
doing so depends on crbug.com/466121.
We're doing this any more - types will be dumped each time.

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:114: value->SetString("units",
attr_type);
Are you forcing all attributes to be scalars now? Or is this also temporarily
hard-coded?

[Primiano Tucci (use gerrit), 2015-04-27 15:03:18]

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
File base/trace_event/memory_allocator_dump.cc (right):

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:80: // new generalized attribute
format. This code will be refactored by May 2015.
On 2015/04/27 14:22:14, petrcermak wrote:
> nit: "by May 2015" is ambiguous. Do you mean "by the end of May 2015"?

Done.

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:109: // on every event. The ability of
doing so depends on crbug.com/466121.
On 2015/04/27 14:22:14, petrcermak wrote:
> We're doing this any more - types will be dumped each time.

Done.

https://codereview.chromium.org/1106943002/diff/20001/base/trace_event/memory...
base/trace_event/memory_allocator_dump.cc:114: value->SetString("units",
attr_type);
On 2015/04/27 14:22:14, petrcermak wrote:
> Are you forcing all attributes to be scalars now? Or is this also temporarily
> hard-coded?

The latter, all this function is to be refactored.

[Primiano Tucci (use gerrit), 2015-04-27 15:03:21]

The CQ bit was checked by primiano@chromium.org

[Primiano Tucci (use gerrit), 2015-04-27 15:03:21]

The patchset sent to the CQ was uploaded after l-g-t-m from
petrcermak@chromium.org
Link to the patchset: https://codereview.chromium.org/1106943002/#ps60001
(title: "Petr's nits")

[commit-bot: I haz the power, 2015-04-27 15:03:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1106943002/60001

[commit-bot: I haz the power, 2015-04-27 16:56:50]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-04-27 16:57:51]

Patchset 4 (id:??) landed as
https://crrev.com/aeb8e8930e6a836dd77325ecef80818dfa1cb9f4
Cr-Commit-Position: refs/heads/master@{#327058}