[benchmark] bump minimum OS versions

[oxy]

https://ci.swift.org runs Xcode 14.2, and this PR brings the deployment versions in line with the official minimum supported versions.

[oxy]

@swift-ci please benchmark

[oxy]

x86_64 run: it turns out a lot of things were made randomly faster / slower by changing the build target, but its unclear how much this matters -

Performance (x86_64): -O

Regression	OLD	NEW	DELTA	RATIO
StringWithCString2	0.0	0.002	+200.0%	0.33x (?)
SIMDReduce.Int8x16.Cast	44.844	70.56	+57.3%	0.64x
CxxStringConversion.cxx.to.swift	142.5	181.0	+27.0%	0.79x (?)
BucketSort	86.958	97.522	+12.1%	0.89x (?)
CharIndexing_ascii_unicodeScalars	4175.385	4602.667	+10.2%	0.91x (?)
CharIndexing_korean_unicodeScalars	4816.0	5277.143	+9.6%	0.91x (?)
CharIndexing_japanese_unicodeScalars	5858.182	6403.333	+9.3%	0.91x (?)
CharIndexing_chinese_unicodeScalars	3623.529	3960.0	+9.3%	0.92x (?)
CharIndexing_russian_unicodeScalars	4133.333	4497.143	+8.8%	0.92x (?)
CharIndexing_punctuated_unicodeScalars	950.0	1032.5	+8.7%	0.92x (?)
CharIndexing_punctuatedJapanese_unicodeScalars	884.762	960.755	+8.6%	0.92x (?)
Improvement	OLD	NEW	DELTA	RATIO
SIMDReduce.Int32x16.Cast	56.167	17.42	-69.0%	3.22x
SIMDReduce.Int32	59.075	20.027	-66.1%	2.95x
SIMDReduce.Int32x4.Cast	56.185	28.641	-49.0%	1.96x
SIMDReduce.Int32x4.Initializer	56.176	34.133	-39.2%	1.65x
SequenceAlgosContiguousArray	2014.545	1603.571	-20.4%	1.26x (?)
SequenceAlgosArray	2011.667	1605.0	-20.2%	1.25x (?)
Dictionary4	188.125	151.5	-19.5%	1.24x (?)
Dictionary4OfObjects	214.333	182.571	-14.8%	1.17x (?)
CharIndexing_japanese_unicodeScalars_Backwards	7928.0	6854.545	-13.5%	1.16x (?)
CharIndexing_chinese_unicodeScalars_Backwards	4760.0	4145.714	-12.9%	1.15x (?)
CharIndexing_punctuatedJapanese_unicodeScalars_Backwards	1184.211	1032.0	-12.9%	1.15x (?)
CharIndexing_korean_unicodeScalars_Backwards	6250.909	5553.333	-11.2%	1.13x (?)
CharIndexing_punctuated_unicodeScalars_Backwards	1201.6	1073.103	-10.7%	1.12x (?)
CharIndexing_russian_unicodeScalars_Backwards	5136.667	4621.333	-10.0%	1.11x (?)
SIMDReduce.Int8x16.Initializer	57.286	51.875	-9.4%	1.10x (?)
FlattenListLoop	1620.0	1471.0	-9.2%	1.10x (?)
CharIndexing_tweet_unicodeScalars_Backwards	10208.889	9364.444	-8.3%	1.09x (?)
CharIndexing_ascii_unicodeScalars_Backwards	5152.0	4738.462	-8.0%	1.09x (?)

Code size: -O

Performance (x86_64): -Osize

Regression	OLD	NEW	DELTA	RATIO
NaiveRRC.append.largeContiguous	0.0	0.419	+41900.0%	0.00x (?)
FlattenListLoop	1412.0	1620.0	+14.7%	0.87x (?)
Set.isStrictSubset.Box0	78.438	86.036	+9.7%	0.91x (?)
CharIndexing_russian_unicodeScalars	4694.545	5080.0	+8.2%	0.92x (?)
Improvement	OLD	NEW	DELTA	RATIO
SIMDReduce.Int32x16.Cast	56.414	17.904	-68.3%	3.15x
Data.init.Sequence.64kB.Count	86.563	29.5	-65.9%	2.93x
Data.init.Sequence.64kB.Count.I	86.571	29.563	-65.9%	2.93x
Data.append.Sequence.64kB.Count.I	87.0	30.828	-64.6%	2.82x
Data.append.Sequence.64kB.Count	87.0	31.385	-63.9%	2.77x
Data.init.Sequence.809B.Count.I	124.7	53.8	-56.9%	2.32x
Data.init.Sequence.809B.Count	124.643	53.941	-56.7%	2.31x
Data.init.Sequence.2049B.Count.I	144.25	62.455	-56.7%	2.31x
Data.init.Sequence.2047B.Count.I	144.091	62.412	-56.7%	2.31x
Data.init.Sequence.513B.Count.I	130.545	66.7	-48.9%	1.96x
Data.init.Sequence.511B.Count.I	126.909	67.294	-47.0%	1.89x
Data.append.Sequence.809B.Count.I	141.615	76.632	-45.9%	1.85x
Data.append.Sequence.809B.Count	141.636	76.647	-45.9%	1.85x
RemoveWhereMoveInts	13.7	7.688	-43.9%	1.78x
SIMDReduce.Int32x4.Cast	56.621	42.704	-24.6%	1.33x
Array.removeAll.keepingCapacity.Object	6.918	5.296	-23.4%	1.31x (?)
StringWalk	1391.724	1110.909	-20.2%	1.25x
CharIteration_tweet_unicodeScalars_Backwards	9377.778	8155.0	-13.0%	1.15x (?)
CharIteration_ascii_unicodeScalars_Backwards	4741.538	4140.0	-12.7%	1.15x
CharIteration_punctuated_unicodeScalars_Backwards	1103.784	979.487	-11.3%	1.13x (?)
RemoveWhereSwapInts	17.588	15.694	-10.8%	1.12x (?)
ObjectiveCBridgeStubNSDateRefAccess	609.667	544.75	-10.6%	1.12x (?)
CharIteration_punctuatedJapanese_unicodeScalars_Backwards	1112.727	1000.0	-10.1%	1.11x (?)
ArrayInClass	217.857	196.328	-9.9%	1.11x (?)
CharIteration_chinese_unicodeScalars_Backwards	4382.857	3960.0	-9.6%	1.11x (?)
CharIteration_korean_unicodeScalars_Backwards	5860.0	5302.857	-9.5%	1.11x (?)
CharIteration_russian_unicodeScalars_Backwards	4772.308	4334.118	-9.2%	1.10x
CharIteration_japanese_unicodeScalars_Backwards	7328.889	6686.154	-8.8%	1.10x (?)
ObjectiveCBridgeStubFromNSStringRef	121.0	112.0	-7.4%	1.08x (?)
FlattenListFlatMap	4141.0	3850.0	-7.0%	1.08x (?)
CharIndexing_chinese_unicodeScalars_Backwards	7340.0	6831.111	-6.9%	1.07x (?)
DataSubscriptMedium	44.409	41.364	-6.9%	1.07x (?)

Code size: -Osize

Performance (x86_64): -Onone

Regression	OLD	NEW	DELTA	RATIO
CxxStringConversion.swift.to.cxx	95.0	103.923	+9.4%	0.91x (?)
Improvement	OLD	NEW	DELTA	RATIO
ArrayAppendGenericStructs	1783.333	1170.0	-34.4%	1.52x (?)
Array.removeAll.keepingCapacity.Object	8.046	6.31	-21.6%	1.28x (?)

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: 32 GB

[oxy]

Condensing the above results: it appears that SIMD (or anything vectorizable) got faster on Intel because of finally gaining SSE4 as a minimum requirement (10.9 supported Merom / 65nm Core 2 Duo (no SSE4), while 10.13 bumped minimums to Penryn / 45nm Core 2 Duo (with SSE4)).

[oxy]

@swift-ci please smoke test

[oxy]

@swift-ci please merge