Added benchmarks for full program execution

[Razican]

This adds benchmarks for full program execution.

Sometimes we have benchmarks that tell us things like "the parser is now 50% slower", but what does that mean for the whole program execution? Maybe it's just a 1% speed reduction.

Also, we have no benchmarks to check improvements if for example, we create the realm in a new thread in parallel to lexing + parsing. This should reduce at least a bit the full program execution time, but there is no way to measure it.

With this PR, all execution benchmarks get duplicated by using our exec() function, that creates a realm, lexes and parses and finally executes everything. This should give us insights on how each change affects to a full workflow.

I also took the opportunity to remove some non-needed black boxes from old benchmarks. They were just preventing optimizations in pre-benchmark code, so they were not useful.

Ideally, this should land before #486/#559 in order to see the change in that PR.

[codecov]

Codecov Report

Merging #560 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #560   +/-   ##
=======================================
  Coverage   68.19%   68.19%           
=======================================
  Files         172      172           
  Lines       10573    10573           
=======================================
  Hits         7210     7210           
  Misses       3363     3363           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8b3d52b...5646d21. Read the comment docs.

[Lan2u]

Is it potentially also worth adding some more examples of 'less abstract' javascript?

[github-actions]

Benchmark for 84a457a

Click to view benchmark

Test	PR Benchmark	Master Benchmark	%
Arithmetic operations (Execution)	440.3±33.57ns	514.5±30.52ns	-14.42%
Arithmetic operations (Full)	188.1±9.11µs	N/A	N/A
Array access (Execution)	15.3±0.81µs	16.2±1.02µs	-5.56%
Array access (Full)	213.6±9.02µs	N/A	N/A
Array creation (Execution)	4.0±0.13ms	4.3±0.18ms	-6.98%
Array creation (Full)	4.5±0.15ms	N/A	N/A
Array pop (Execution)	1514.4±64.51µs	1667.7±1001.09µs	-9.19%
Array pop (Full)	1911.0±114.57µs	N/A	N/A
Boolean Object Access (Execution)	6.1±0.38µs	6.0±0.36µs	+1.67%
Boolean Object Access (Full)	200.6±12.15µs	N/A	N/A
Create Realm	168.9±10.99µs	175.0±11.32µs	-3.49%
Dynamic Object Property Access (Execution)	8.1±0.49µs	8.2±0.53µs	-1.22%
Dynamic Object Property Access (Full)	203.0±8.80µs	N/A	N/A
Expression (Lexer)	2.3±0.14µs	2.4±0.12µs	-4.17%
Expression (Parser)	5.4±0.37µs	5.6±0.29µs	-3.57%
Fibonacci (Execution)	1213.5±49.36µs	1319.8±125.56µs	-8.05%
Fibonacci (Full)	1390.3±72.11µs	N/A	N/A
For loop (Execution)	26.8±1.38µs	28.6±2.34µs	-6.29%
For loop (Full)	219.6±10.78µs	N/A	N/A
For loop (Lexer)	6.0±0.36µs	5.9±0.26µs	+1.69%
For loop (Parser)	15.6±0.61µs	15.5±0.79µs	+0.65%
Goal Symbols (Parser)	9.6±0.36µs	9.7±0.55µs	-1.03%
Hello World (Lexer)	1060.1±127.68ns	1046.9±35.47ns	+1.26%
Hello World (Parser)	2.4±0.10µs	2.6±0.19µs	-7.69%
Long file (Parser)	7.0±0.19ms	7.1±0.33ms	-1.41%
Number Object Access (Execution)	4.7±0.23µs	4.9±0.30µs	-4.08%
Number Object Access (Full)	244.4±10.60µs	N/A	N/A
Object Creation (Execution)	7.1±0.26µs	7.3±0.73µs	-2.74%
Object Creation (Full)	224.8±16.39µs	N/A	N/A
RegExp (Execution)	85.8±4.66µs	87.4±6.27µs	-1.83%
RegExp (Full)	344.5±15.72µs	N/A	N/A
RegExp Literal (Execution)	91.1±6.97µs	92.5±6.05µs	-1.51%
RegExp Literal (Full)	295.7±12.08µs	N/A	N/A
RegExp Literal Creation (Execution)	85.0±4.20µs	86.9±4.50µs	-2.19%
RegExp Literal Creation (Full)	345.0±15.20µs	N/A	N/A
Static Object Property Access (Execution)	7.4±0.38µs	7.6±0.52µs	-2.63%
Static Object Property Access (Full)	200.7±11.28µs	N/A	N/A
String Object Access (Execution)	10.2±0.56µs	10.3±0.51µs	-0.97%
String Object Access (Full)	203.1±10.14µs	N/A	N/A
String comparison (Execution)	7.9±0.45µs	8.4±0.31µs	-5.95%
String comparison (Full)	212.8±16.13µs	N/A	N/A
String concatenation (Execution)	6.6±0.29µs	7.0±0.84µs	-5.71%
String concatenation (Full)	197.3±15.18µs	N/A	N/A
String copy (Execution)	5.3±0.29µs	5.9±0.41µs	-10.17%
String copy (Full)	190.6±6.08µs	N/A	N/A
Symbols (Execution)	4.7±0.23µs	5.0±0.36µs	-6.00%
Symbols (Full)	190.7±10.82µs	N/A	N/A

[Razican]

Is it potentially also worth adding some more examples of 'less abstract' javascript?

What do you mean? If you find interesting stuff to benchmark, feel free to add it :)

[Lan2u]

Is it potentially also worth adding some more examples of 'less abstract' javascript?

What do you mean? If you find interesting stuff to benchmark, feel free to add it :)

I'll see if I can find some good examples