Handle begin_apply in TempRVO

[meg-gupta]

A tempobj passed to begin_apply instruction and cannot be modified by it
(is guaranteed and doesn't alias with other inout args) has potential to be optimized
away.

[meg-gupta]

@swift-ci please test

[meg-gupta]

@swift-ci benchmark

[swift-ci]

Build failed before running benchmark.

[meg-gupta]

@swift-ci please benchmark

[swift-ci]

Performance: -O

Regression	OLD	NEW	DELTA	RATIO
DictionaryBridgeToObjC_Access	881	1016	+15.3%	0.87x (?)
FloatingPointPrinting_Float_description_uniform	5300	5700	+7.5%	0.93x (?)
Improvement	OLD	NEW	DELTA	RATIO
RandomShuffleLCG2	816	208	-74.5%	3.92x
SortIntPyramid	1300	640	-50.8%	2.03x
SortAdjacentIntPyramids	1565	1175	-24.9%	1.33x
Breadcrumbs.MutatedUTF16ToIdx.ASCII	5	4	-20.0%	1.25x
Breadcrumbs.MutatedIdxToUTF16.ASCII	5	4	-20.0%	1.25x
FindString.Loop1.Substring	826	668	-19.1%	1.24x
FindString.Rec3.Substring	255	217	-14.9%	1.18x
FindString.Rec3.String	215	183	-14.9%	1.17x
ArrayAppendLazyMap	7100	6090	-14.2%	1.17x (?)
ReversedBidirectional	9918	8655	-12.7%	1.15x
LazilyFilteredArrayContains	41400	36600	-11.6%	1.13x (?)
String.replaceSubrange.RepChar	3325	2976	-10.5%	1.12x (?)
RemoveWhereSwapInts	69	62	-10.1%	1.11x
RemoveWhereFilterInts	51	46	-9.8%	1.11x (?)
Data.hash.Empty	80	74	-7.5%	1.08x (?)
Array2D	8112	7520	-7.3%	1.08x (?)
MapReduce	400	371	-7.2%	1.08x (?)
FlattenListLoop	4812	4466	-7.2%	1.08x (?)
ArrayPlusEqualFiveElementCollection	9065	8436	-6.9%	1.07x (?)
MapReduceAnyCollection	426	397	-6.8%	1.07x (?)

Code size: -O

Regression	OLD	NEW	DELTA	RATIO
RemoveWhere.o	17157	17845	+4.0%	0.96x
Improvement	OLD	NEW	DELTA	RATIO
RandomShuffle.o	3632	3368	-7.3%	1.08x
StringMatch.o	5304	5208	-1.8%	1.02x
ReversedCollections.o	9286	9126	-1.7%	1.02x

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
CharIteration_ascii_unicodeScalars_Backwards	5080	5640	+11.0%	0.90x (?)
CharIteration_tweet_unicodeScalars_Backwards	10080	11160	+10.7%	0.90x (?)
DropLastAnySequenceLazy	4096	4492	+9.7%	0.91x (?)
Improvement	OLD	NEW	DELTA	RATIO
RandomShuffleLCG2	816	192	-76.5%	4.25x
FlattenListLoop	5203	4005	-23.0%	1.30x
FindString.Loop1.Substring	846	681	-19.5%	1.24x
FlattenListFlatMap	7350	6011	-18.2%	1.22x (?)
ReversedBidirectional	13296	10958	-17.6%	1.21x
FindString.Rec3.String	216	185	-14.4%	1.17x
FindString.Rec3.Substring	257	221	-14.0%	1.16x
RemoveWhereFilterInts	50	44	-12.0%	1.14x (?)
ArrayAppendLazyMap	8130	7210	-11.3%	1.13x (?)
RemoveWhereSwapInts	67	60	-10.4%	1.12x (?)
CharIteration_russian_unicodeScalars	4040	3680	-8.9%	1.10x (?)
NSStringConversion.MutableCopy.Rebridge	955	878	-8.1%	1.09x (?)
Data.hash.Empty	77	71	-7.8%	1.08x
ArrayPlusEqualFiveElementCollection	8325	7696	-7.6%	1.08x (?)
String.replaceSubrange.RepChar	3243	2998	-7.6%	1.08x (?)
Array2D	7520	6960	-7.4%	1.08x (?)
MapReduce	435	406	-6.7%	1.07x (?)

Code size: -Osize

Improvement	OLD	NEW	DELTA	RATIO
RandomShuffle.o	3878	3614	-6.8%	1.07x
StringMatch.o	5138	5026	-2.2%	1.02x
RemoveWhere.o	16299	16091	-1.3%	1.01x

Performance: -Onone

Regression	OLD	NEW	DELTA	RATIO
Set.isDisjoint.Box25	2491	2835	+13.8%	0.88x (?)
Improvement	OLD	NEW	DELTA	RATIO
PrefixArrayLazy	88155	65766	-25.4%	1.34x
SuffixArrayLazy	28810	21874	-24.1%	1.32x
DropFirstArrayLazy	87077	66828	-23.3%	1.30x
DropLastArrayLazy	29019	22416	-22.8%	1.29x
SuffixCountableRangeLazy	38397	31162	-18.8%	1.23x (?)
DropLastCountableRangeLazy	38589	31578	-18.2%	1.22x
PrefixCountableRangeLazy	115551	94802	-18.0%	1.22x (?)
DropFirstCountableRangeLazy	115657	95543	-17.4%	1.21x
DropLastAnyCollectionLazy	69525	59474	-14.5%	1.17x (?)
PrefixAnyCollectionLazy	204819	178303	-12.9%	1.15x (?)
DropFirstAnyCollectionLazy	198077	176842	-10.7%	1.12x (?)
String.replaceSubrange.RepChar	3438	3074	-10.6%	1.12x (?)
SuffixAnyCollectionLazy	65850	58902	-10.6%	1.12x (?)
CharacterPropertiesStashed	2370	2150	-9.3%	1.10x (?)
RandomShuffleLCG2	56880	51920	-8.7%	1.10x (?)
CharacterPropertiesStashedMemo	3690	3390	-8.1%	1.09x (?)
DataSetCountSmall	154	142	-7.8%	1.08x
CharacterPropertiesPrecomputed	3240	3000	-7.4%	1.08x (?)
WordCountUniqueUTF16	13000	12050	-7.3%	1.08x (?)
WordCountUniqueASCII	12610	11720	-7.1%	1.08x (?)
CharacterPropertiesFetch	5420	5050	-6.8%	1.07x (?)
StringWordBuilderReservingCapacity	2730	2550	-6.6%	1.07x (?)

Code size: -swiftlibs

Improvement	OLD	NEW	DELTA	RATIO
libswiftCompression.dylib	45056	40960	-9.1%	1.10x

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

[atrick]

Thanks for fixing this. But remember that a begin_apply operand is used at the point of the end_apply. So the end_apply needs to be considered the use-point for things like checkNoSourceModification.

[meg-gupta]

@swift-ci test

[swift-ci]

Build failed
Swift Test Linux Platform
Git Sha - 7a6ed1a

[swift-ci]

Build failed
Swift Test OS X Platform
Git Sha - 7a6ed1a

[meg-gupta]

@swift-ci test linux

[meg-gupta]

@swift-ci please test

[swift-ci]

Build failed
Swift Test Linux Platform
Git Sha - c5c1330

[swift-ci]

Build failed
Swift Test OS X Platform
Git Sha - c5c1330

[atrick]

LGTM