LICM: enable more stores to moved out of a loop

[eeckstein]

Even if a store is not dominating the loop exits, it makes sense to move it out of the loop if the pre-header also as a store to the same memory location.
When this is done, dead-store-elimination can then most likely remove the store in the pre-header.

[eeckstein]

@swift-ci test

[eeckstein]

@swift-ci benchmark

[swift-ci]

Performance: -O

Regression	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	7456	9273	+24.4%	0.80x (?)
FlattenListLoop	4092	4918	+20.2%	0.83x (?)
Improvement	OLD	NEW	DELTA	RATIO
PrefixWhileAnyCollectionLazy	33	28	-15.2%	1.18x (?)
PrefixWhileAnySeqCRangeIterLazy	33	28	-15.2%	1.18x
PrefixWhileAnySeqCntRangeLazy	33	28	-15.2%	1.18x

Code size: -O

Improvement	OLD	NEW	DELTA	RATIO
Prefix.o	18448	17664	-4.2%	1.04x

Performance: -Osize

Improvement	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	7654	6951	-9.2%	1.10x (?)
String.data.Medium	111	102	-8.1%	1.09x (?)
ObjectiveCBridgeFromNSArrayAnyObjectToStringForced	43600	40400	-7.3%	1.08x (?)

Code size: -Osize

Performance: -Onone

Regression	OLD	NEW	DELTA	RATIO
CharacterPropertiesFetch	5280	6390	+21.0%	0.83x (?)
ObjectiveCBridgeStubToNSDate2	630	710	+12.7%	0.89x (?)

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

[swift-ci]

Build failed
Swift Test OS X Platform
Git Sha - 9ac86dbb01ee6f55a87bcb2c6fb83426396a4602

[eeckstein]

@swift-ci smoke test macOS

[atrick]

If %cond was never true, then this is storing the wrong value ??

[eeckstein]

I should use a different name for %v2 here. It's actually the value which comes out of the SSAUpdater.

[atrick]

yeah, I would hope it's inserting a phi inside the loop

[atrick]

Would it be worth writing a unit test for the example from the documentation?

[eeckstein]

The test I added is exactly this. I added a check for the phi argument.

[eeckstein]

@swift-ci smoke test

[eeckstein]

@swift-ci smoke test linux

[atrick]

LGTM