Set option default for enabling memory ssa for new pass manager loop sink pass to true.

[jamieschmeiser]

Summary

Set the default for the option enabling memory ssa use in the loop sink
pass to true for the new pass manager.

[nikic]

Compile-time: https://llvm-compile-time-tracker.com/compare.php?from=54eab293f523956bdc4b1a98b6cf5abc0bd1ef3f&to=b023c8adc9eb545ec5b8aa1a8d6f77a46d622006&stat=instructions

The problem here is that LoopSink is only relevant for PGO, but MemorySSA gets computed unconditionally in the legacy PM. This problem does not affect the new pass manager, where it's possible to compute MemorySSA only if useful. Supporting this properly in LegacyPM would require something like LazyMemorySSA.

I think in this instance it might be best to do this change only for NewPM. It's best to keep both pass managers in sync, but here there is a technical reason for different treatment.

[fhahn]

In D92486#2428821, @nikic wrote:

Compile-time: https://llvm-compile-time-tracker.com/compare.php?from=54eab293f523956bdc4b1a98b6cf5abc0bd1ef3f&to=b023c8adc9eb545ec5b8aa1a8d6f77a46d622006&stat=instructions

The problem here is that LoopSink is only relevant for PGO, but MemorySSA gets computed unconditionally in the legacy PM. This problem does not affect the new pass manager, where it's possible to compute MemorySSA only if useful. Supporting this properly in LegacyPM would require something like LazyMemorySSA.

I think in this instance it might be best to do this change only for NewPM. It's best to keep both pass managers in sync, but here there is a technical reason for different treatment.

Perhaps it would also be possible to directly construct MemorySSA in the pass, if PGO is enabled with the legacy PM?

[nikic]

In D92486#2428865, @fhahn wrote:

In D92486#2428821, @nikic wrote:

Compile-time: https://llvm-compile-time-tracker.com/compare.php?from=54eab293f523956bdc4b1a98b6cf5abc0bd1ef3f&to=b023c8adc9eb545ec5b8aa1a8d6f77a46d622006&stat=instructions

The problem here is that LoopSink is only relevant for PGO, but MemorySSA gets computed unconditionally in the legacy PM. This problem does not affect the new pass manager, where it's possible to compute MemorySSA only if useful. Supporting this properly in LegacyPM would require something like LazyMemorySSA.

I think in this instance it might be best to do this change only for NewPM. It's best to keep both pass managers in sync, but here there is a technical reason for different treatment.

Perhaps it would also be possible to directly construct MemorySSA in the pass, if PGO is enabled with the legacy PM?

I think that would require constructing MemorySSA anew for each loop. I'm not really familiar with pass manager limitations though, maybe this is not the case.

[jamieschmeiser]

Remove changes pertaining to legacy pass manager.

[asbirlea]

I believe for the NPM this should work fine.
@nikic @fhahn Any objections?

llvm/lib/Transforms/Scalar/LoopSink.cpp

static cl::opt<unsigned> SinkFrequencyPercentThreshold(
     cl::desc("Do not sink instructions that require cloning unless they "
              "execute less than this percent of the time."));
 
 static cl::opt<unsigned> MaxNumberOfUseBBsForSinking(
     "max-uses-for-sinking", cl::Hidden, cl::init(30),
     cl::desc("Do not sink instructions that have too many uses."));
 
 static cl::opt<bool> EnableMSSAInLoopSink(
-    "enable-mssa-in-loop-sink", cl::Hidden, cl::init(false),
+    "enable-mssa-in-loop-sink", cl::Hidden, cl::init(true),
     cl::desc("Enable MemorySSA for LoopSink in new pass manager"));
 
 static cl::opt<bool> EnableMSSAInLegacyLoopSink(
     "enable-mssa-in-legacy-loop-sink", cl::Hidden, cl::init(false),
     cl::desc("Enable MemorySSA for LoopSink in legacy pass manager"));
 
 /// Return adjusted total frequency of \p BBs.
 ///