Cache the default malloc zone

[Catfish-Man]

I measured a 7% win on FilterEvenUsingReduceInto with fast dealloc enabled (which it won't be in CI)

[Catfish-Man]

@swift-ci please smoke test

[Catfish-Man]

@swift-ci please smoke benchmark

[swift-ci]

Performance: -O

Regression	OLD	NEW	DELTA	RATIO
StringComparison_nonBMPSlowestPrenormal	1390	1580	+13.7%	0.88x
ObjectiveCBridgeStringHash	70	76	+8.6%	0.92x (?)
EqualSubstringSubstring	39	42	+7.7%	0.93x (?)
LessSubstringSubstring	39	42	+7.7%	0.93x (?)
EqualSubstringSubstringGenericEquatable	39	42	+7.7%	0.93x (?)
EqualSubstringString	39	42	+7.7%	0.93x
LessSubstringSubstringGenericComparable	39	42	+7.7%	0.93x (?)

Code size: -O

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	5903	8048	+36.3%	0.73x (?)
FlattenListLoop	4452	5282	+18.6%	0.84x (?)
StringComparison_nonBMPSlowestPrenormal	1420	1570	+10.6%	0.90x (?)
EqualStringSubstring	40	44	+10.0%	0.91x
ObjectiveCBridgeStringHash	70	76	+8.6%	0.92x (?)
EqualSubstringString	39	42	+7.7%	0.93x
Improvement	OLD	NEW	DELTA	RATIO
ObjectiveCBridgeStubFromNSDateRef	4270	3800	-11.0%	1.12x (?)
ObjectiveCBridgeStubFromArrayOfNSString2	3560	3310	-7.0%	1.08x (?)

Code size: -Osize

Performance: -Onone

Regression	OLD	NEW	DELTA	RATIO
ObjectiveCBridgeStringHash	70	76	+8.6%	0.92x (?)
EqualStringSubstring	47	51	+8.5%	0.92x (?)
DictionaryBridgeToObjC_Access	969	1045	+7.8%	0.93x (?)
Improvement	OLD	NEW	DELTA	RATIO
StringBuilderWithLongSubstring	3200	2850	-10.9%	1.12x (?)
Breadcrumbs.CopyUTF16CodeUnits.ASCII	194	174	-10.3%	1.11x (?)
DictionaryOfAnyHashableStrings_lookup	5400	4968	-8.0%	1.09x (?)
ArrayOfPOD	1113	1033	-7.2%	1.08x (?)

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 Pro
  Model Identifier: MacPro6,1
  Processor Name: 12-Core Intel Xeon E5
  Processor Speed: 2.7 GHz
  Number of Processors: 1
  Total Number of Cores: 12
  L2 Cache (per Core): 256 KB
  L3 Cache: 30 MB
  Memory: 64 GB

[Catfish-Man]

Looks like with fast dealloc disabled the win is below the 7% threshold to be reported, so we just get noise.

Disturbing to see (?) on a 36% change though.

[gribozavr]

The static and SWIFT_LAZY_CONSTANT create two levels of lazy initialization; we only need one of them IIUC.

[mikeash]

Good catch! This shouldn't have the static.