TempRValueOpt: extend access scopes

[eeckstein]

Try to move an end_access down to extend the access scope over all uses of the temporary.
For example:

   %a = begin_access %src
   copy_addr %a to [initialization] %temp : $*T
   end_access %a
   use %temp

We must not replace %temp with %a after the end_access. Instead we try to move the end_access after use %temp.

This fixes generation of invalid SIL and/or the invalid removal of access checks.

The change also contains some small non-functional re-factorings/improvements.

[eeckstein]

@swift-ci test

[eeckstein]

@swift-ci benchmark

[swift-ci]

Performance: -O

Regression	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	4891	5644	+15.4%	0.87x (?)
Improvement	OLD	NEW	DELTA	RATIO
DictionaryKeysContainsCocoa	29	27	-6.9%	1.07x (?)

Code size: -O

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
DictionaryKeysContainsNative	26	29	+11.5%	0.90x (?)

Code size: -Osize

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 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 Linux Platform
Git Sha - bafc128336c532ccd39b2d9204da3d38899a6e39

[swift-ci]

Build failed
Swift Test OS X Platform
Git Sha - bafc128336c532ccd39b2d9204da3d38899a6e39

[eeckstein]

@swift-ci test

[eeckstein]

@swift-ci benchmark

[swift-ci]

Performance: -O

Improvement	OLD	NEW	DELTA	RATIO
FlattenListFlatMap	6048	4049	-33.1%	1.49x (?)
DataAppendDataSmallToSmall	4660	4240	-9.0%	1.10x (?)
NSStringConversion.MutableCopy.Medium	876	817	-6.7%	1.07x (?)
NSStringConversion.UTF8	992	926	-6.7%	1.07x (?)

Code size: -O

Performance: -Osize

Regression	OLD	NEW	DELTA	RATIO
String.data.LargeUnicode	110	126	+14.5%	0.87x (?)
String.data.Medium	111	124	+11.7%	0.90x (?)

Code size: -Osize

Performance: -Onone

Improvement	OLD	NEW	DELTA	RATIO
String.data.LargeUnicode	209	193	-7.7%	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

[swift-ci]

Build failed
Swift Test Linux Platform
Git Sha - 7c454e3

[atrick]

That looks safe enough. Thanks!

[eeckstein]

@swift-ci smoke test linux

[atrick]

@eeckstein We also still have an "unpaired" begin access, so strictly speaking, finding all the access scope beginnings means doing this:
isa<BeginAccessInst>(inst) || isa<BeginUnpairedAccessInst>(inst)
I'm surprised there's no helper for it yet.

[eeckstein]

The begin_unpaired_access is only used in the KeyPath runtime functions. And we compile the stdlib without access enforcement. So it should not be a problem.
But checking for the unpaired access does not harm. I'll add it.