Add result checking to the ChaCha benchmark.

[atrick]

This benchmark was being miscompiled for some time and
we did not catch it. I only noticed because a correcteness
fix caused a regression.

[atrick]

@swift-ci test

[atrick]

@swift-ci benchmark

[swift-ci]

Performance: -O

Regression	OLD	NEW	DELTA	RATIO
ExclusivityGlobal	3	4	+33.3%	0.75x
ChaCha	87	96	+10.3%	0.91x (?)

Code size: -O

Regression	OLD	NEW	DELTA	RATIO
ChaCha.o	14209	15619	+9.9%	0.91x

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	2615	2954	+13.0%	0.89x (?)
DictionaryBridgeToObjC_BulkAccess	106	114	+7.5%	0.93x (?)

Code size: -Osize

Regression	OLD	NEW	DELTA	RATIO
ChaCha.o	11004	12278	+11.6%	0.90x

Performance: -Onone

Regression	OLD	NEW	DELTA	RATIO
ChaCha	36480	79295	+117.4%	0.46x

Code size: -swiftlibs

How to read the data

The tables contain differences in performance which are larger than 8% and differences in code size which are larger than 1%.

If you see any unexpected regressions, you should consider fixing the
regressions before you merge the PR.

Noise: Sometimes the performance results (not code size!) contain false
alarms. Unexpected regressions which are marked with '(?)' are probably noise.
If you see regressions which you cannot explain you can try to run the
benchmarks again. If regressions still show up, please consult with the
performance team (@eeckstein).

Hardware Overview

  Model Name: Mac mini
  Model Identifier: Macmini8,1
  Processor Name: 6-Core Intel Core i7
  Processor Speed: 3.2 GHz
  Number of Processors: 1
  Total Number of Cores: 6
  L2 Cache (per Core): 256 KB
  L3 Cache: 12 MB
  Memory: 64 GB

[atrick]

The results are inexplicable because I only saw a small performance regression at -Onone when building locally. Nonetheless, I reduced the verification so it definitely won't take a significant portion of the run time at -Onone.

[atrick]

@swift-ci benchmark

[atrick]

@swift-ci smoke test

[swift-ci]

Build failed before running benchmark.

[atrick]

@swift-ci smoke test osx

[atrick]

@swift-ci benchmark

[atrick]

@swift-ci smoke test osx platform

[atrick]

@swift-ci smoke test OS X platform

[atrick]

@swift-ci smoke test OS X platform

[atrick]

@swift-ci benchmark

[atrick]

@swift-ci smoke test

[atrick]

@swift-ci benchmark

[swift-ci]

Performance: -O

Improvement	OLD	NEW	DELTA	RATIO
DictionaryBridgeToObjC_BulkAccess	111	102	-8.1%	1.09x (?)

Code size: -O

Regression	OLD	NEW	DELTA	RATIO
ChaCha.o	14209	15667	+10.3%	0.91x

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	2969	3220	+8.5%	0.92x (?)

Code size: -Osize

Regression	OLD	NEW	DELTA	RATIO
ChaCha.o	11004	12316	+11.9%	0.89x

Performance: -Onone

Code size: -swiftlibs

How to read the data

The tables contain differences in performance which are larger than 8% and differences in code size which are larger than 1%.

If you see any unexpected regressions, you should consider fixing the
regressions before you merge the PR.

Noise: Sometimes the performance results (not code size!) contain false
alarms. Unexpected regressions which are marked with '(?)' are probably noise.
If you see regressions which you cannot explain you can try to run the
benchmarks again. If regressions still show up, please consult with the
performance team (@eeckstein).

Hardware Overview

  Model Name: Mac mini
  Model Identifier: Macmini8,1
  Processor Name: 6-Core Intel Core i7
  Processor Speed: 3.2 GHz
  Number of Processors: 1
  Total Number of Cores: 6
  L2 Cache (per Core): 256 KB
  L3 Cache: 12 MB
  Memory: 64 GB