CPUProfiler: Improve line numbers support in profiler.

CPUProfiler: Improve line numbers support in profiler.

1) report line number even if a script has no resource_name (evals);
  a) do that for already compiled functions in log.cc;
  b) do that for fresh evals in compiler.cc;

2) Implement the test for LineNumbers and make it fast and stable, otherwise we have to wait for tick samples;
  a) move processor_->Join() call into new Processor::StopSynchronously method;
  b) Process all the CodeEvents even if we are stopping Processor thread;
  c) make getters for generator and processor;

3) Fix the test for Jit that didn't expect line numbers;

4) Minor refactoring:
  a) in ProcessTicks;
  b) rename enqueue_order_ to last_code_event_id_ for better readability;
  c) rename dequeue_order_ to last_processed_code_event_id_ and make it a member for better readability;

BUG=
TEST=test-profile-generator/LineNumber
R=jkummerow@chromium.org, yurys@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=15530

[yurys, 2013-07-05 13:10:00]

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc
File src/cpu-profiler.cc (right):

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc#newcode81
src/cpu-profiler.cc:81: while (ProcessCodeEvent()) { }
Why do we have to do this processing on the main thread? It slows down the start
and adds access to the code map from one more thread.

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc#newcode86
src/cpu-profiler.cc:86: void ProfilerEventsProcessor::Stop() {
StopAndJoin or StopSynchronously.

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h
File src/cpu-profiler.h (right):

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h#newcode158
src/cpu-profiler.h:158: void Start();
There is already non-virtual method Start in Thread class.

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h#newcode185
src/cpu-profiler.h:185: unsigned last_recorded_code_event_id_;
last_code_event_id_

https://codereview.chromium.org/18058008/diff/1/src/log.cc
File src/log.cc (left):

https://codereview.chromium.org/18058008/diff/1/src/log.cc#oldcode1667
src/log.cc:1667: // Can't distinguish eval and script here, so always use
Script.
Revert this line?

https://codereview.chromium.org/18058008/diff/1/src/log.cc
File src/log.cc (right):

https://codereview.chromium.org/18058008/diff/1/src/log.cc#newcode1677
src/log.cc:1677: isolate_->heap()->empty_string(), line_num + 1));
You can do line_num = line_num + 1 once outside if block.

[loislo, 2013-07-05 13:15:37]

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc
File src/cpu-profiler.cc (right):

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc#newcode81
src/cpu-profiler.cc:81: while (ProcessCodeEvent()) { }
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> Why do we have to do this processing on the main thread? It slows down the
start
> and adds access to the code map from one more thread.

reverted

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc#newcode86
src/cpu-profiler.cc:86: void ProfilerEventsProcessor::Stop() {
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> StopAndJoin or StopSynchronously.

Done.

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.cc#newcode468
src/cpu-profiler.cc:468: processor_->Start();
reverted

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h
File src/cpu-profiler.h (right):

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h#newcode158
src/cpu-profiler.h:158: void Start();
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> There is already non-virtual method Start in Thread class.

reverted

https://codereview.chromium.org/18058008/diff/1/src/cpu-profiler.h#newcode185
src/cpu-profiler.h:185: unsigned last_recorded_code_event_id_;
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> last_code_event_id_

Done.

https://codereview.chromium.org/18058008/diff/1/src/log.cc
File src/log.cc (left):

https://codereview.chromium.org/18058008/diff/1/src/log.cc#oldcode1667
src/log.cc:1667: // Can't distinguish eval and script here, so always use
Script.
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> Revert this line?

Done.

https://codereview.chromium.org/18058008/diff/1/src/log.cc
File src/log.cc (right):

https://codereview.chromium.org/18058008/diff/1/src/log.cc#newcode1677
src/log.cc:1677: isolate_->heap()->empty_string(), line_num + 1));
On 2013/07/05 13:10:00, Yury Semikhatsky wrote:
> You can do line_num = line_num + 1 once outside if block.

Done.

[yurys, 2013-07-06 05:03:21]

lgtm

https://codereview.chromium.org/18058008/diff/10007/src/compiler.cc
File src/compiler.cc (right):

https://codereview.chromium.org/18058008/diff/10007/src/compiler.cc#newcode1204
src/compiler.cc:1204: USE(line_num);
It seems that you can drop this line as line_num is always used.

[Jakob Kummerow, 2013-07-06 07:52:53]

rubber-stamp LGTM

[loislo, 2013-07-07 11:42:42]

Committed patchset #3 manually as r15530 (presubmit successful).