Enabling augmenting callback with Location

Enabling augmenting callback with Location.
This is useful for instrumentations required for jankiness investigations. See the .h file comment for details.

In fact, one of my investigations is blocked on this feature.

BUG=401560
Committed: https://crrev.com/8ddfc1ef844ae617768b6b0cc5642c445a9ea9ff
Cr-Commit-Position: refs/heads/master@{#302324}

[vadimt, 2014-10-29 00:17:45]

vadimt@chromium.org changed reviewers:
+ isherman@chromium.org

[vadimt, 2014-10-29 00:19:06]

isherman@, please pre-review the change.
Thanks!

[Ilya Sherman, 2014-10-29 00:23:01]

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker.cc
File base/profiler/scoped_tracker.cc (right):

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:16: void ExecuteAndAugmentCallback(const
Location& location,
nit: I find the "augment" part of this name confusing.  WDYT of
"ExecuteAndTrackCallback"?

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:18: ScopedProfile tracking_profile(location);
Hmm, why not use a ScopedTracker here instead?

[vadimt, 2014-10-29 00:52:21]

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker.cc
File base/profiler/scoped_tracker.cc (right):

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:16: void ExecuteAndAugmentCallback(const
Location& location,
On 2014/10/29 00:23:00, Ilya Sherman wrote:
> nit: I find the "augment" part of this name confusing.  WDYT of
> "ExecuteAndTrackCallback"?

Done.

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:18: ScopedProfile tracking_profile(location);
ScopedTracker checks if tracking is enabled and calls ScopedProfile. Since
TrackCallback already checks g_scoped_profile_mode, we can use scopedProfile
directly, without checking once more.

[Ilya Sherman, 2014-10-29 00:57:12]

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker.cc
File base/profiler/scoped_tracker.cc (right):

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:18: ScopedProfile tracking_profile(location);
On 2014/10/29 00:52:21, vadimt wrote:
> ScopedTracker checks if tracking is enabled and calls ScopedProfile. Since
> TrackCallback already checks g_scoped_profile_mode, we can use scopedProfile
> directly, without checking once more.

My point is: Why does TrackCallback check g_scoped_profile_mode explicitly?  It
seems like it could skip that check.

[vadimt, 2014-10-29 17:47:29]

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker.cc
File base/profiler/scoped_tracker.cc (right):

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:18: ScopedProfile tracking_profile(location);
This allows me to totally skip creating/destroying ScopedProfile in Stable. Even
though it's very optimized, I don't see why we shouldn't use this opportunity,
given that it's technically feasible. We also save on additional invocation of a
callback doe to indirection.

[Ilya Sherman, 2014-10-30 00:51:35]

LGTM

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker.cc
File base/profiler/scoped_tracker.cc (right):

https://codereview.chromium.org/683283004/diff/1/base/profiler/scoped_tracker...
base/profiler/scoped_tracker.cc:18: ScopedProfile tracking_profile(location);
On 2014/10/29 17:47:29, vadimt wrote:
> This allows me to totally skip creating/destroying ScopedProfile in Stable.
Even
> though it's very optimized, I don't see why we shouldn't use this opportunity,
> given that it's technically feasible. We also save on additional invocation of
a
> callback doe to indirection.

I think this is unnecessary optimization that reduces readability.  However, if
you're really super attached to it, I don't feel strongly enough to try to
dissuade you.

[vadimt, 2014-10-30 00:57:34]

vadimt@chromium.org changed reviewers:
+ mark@chromium.org

[vadimt, 2014-10-30 00:57:34]

mark@chromium.org: Please provide OWNERS review.

[Mark Mentovai, 2014-10-31 17:47:44]

LGTM

[vadimt, 2014-10-31 19:10:03]

The CQ bit was checked by vadimt@chromium.org

[commit-bot: I haz the power, 2014-10-31 19:11:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/683283004/20001

[commit-bot: I haz the power, 2014-10-31 20:54:47]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2014-10-31 20:55:27]

Patchset 2 (id:??) landed as
https://crrev.com/8ddfc1ef844ae617768b6b0cc5642c445a9ea9ff
Cr-Commit-Position: refs/heads/master@{#302324}