Use -enable-ossa-modules for all of stdlib/public

[meg-gupta]

This will pass -enable-ossa-modules while compiling all of stdlib/public not just stdlib/public/core.
With this, even non-transparent functions will have their ownership eliminated later in the pipeline.

[meg-gupta]

@swift-ci test

[meg-gupta]

@swift-ci benchmark

[swift-ci]

Build failed
Swift Test Linux Platform
Git Sha - 1b2e60791eaf971c758505429b3d03e975b2760a

[meg-gupta]

@swift-ci test

[meg-gupta]

@swift-ci benchmark

[swift-ci]

Performance (x86_64): -O

Regression	OLD	NEW	DELTA	RATIO
ArrayAppendGenericStructs	1340	1540	+14.9%	0.87x (?)
Improvement	OLD	NEW	DELTA	RATIO
Data.init.Sequence.64kB.Count.RE	2	1	-50.0%	2.00x
StringInterpolationManySmallSegments	8700	7300	-16.1%	1.19x (?)
FlattenListLoop	1570	1386	-11.7%	1.13x (?)
DataToStringEmpty	600	550	-8.3%	1.09x (?)
FlattenListFlatMap	4299	3961	-7.9%	1.09x (?)

Code size: -O

Performance (x86_64): -Osize

Regression	OLD	NEW	DELTA	RATIO
ArrayAppendGenericStructs	610	1250	+104.9%	0.49x (?)
Improvement	OLD	NEW	DELTA	RATIO
FlattenListLoop	1565	935	-40.3%	1.67x (?)

Code size: -Osize

Performance (x86_64): -Onone

Improvement	OLD	NEW	DELTA	RATIO
NSStringConversion.MutableCopy.Rebridge.UTF8	486	454	-6.6%	1.07x (?)

Code size: -swiftlibs

Improvement	OLD	NEW	DELTA	RATIO
libswiftStdlibUnittest.dylib	327680	311296	-5.0%	1.05x

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: 64 GB

[meg-gupta]

@swift-ci test

[atrick]

I like the idea. If this sticks, can we get rid of all the special case modes for transparent functions and stdlib builds?

[atrick]

list(APPEND swift_stdlib_compile_flags "-Xfrontend" "-enable-ossa-modules")

Do we still need this? Will the flag be passed twice to the stldib build now?

[meg-gupta]

list(APPEND swift_stdlib_compile_flags "-Xfrontend" "-enable-ossa-modules")

Do we still need this? Will the flag be passed twice to the stldib build now?

@atrick Thanks. I got rid of it in the latest update.

list(APPEND swift_stdlib_compile_flags "-Xfrontend" "-enable-ossa-modules")

Do we still need this? Will the flag be passed twice to the stldib build now?

Good point. I need to test this, but I think we still need these modes for the deserialization hooks we install in OME.

[meg-gupta]

@swift-ci test