#include "llvm/Transforms/IPO/PassManagerBuilder.h"
-
#include "llvm-c/Transforms/PassManagerBuilder.h"
-
-#include "llvm/PassManager.h"
-#include "llvm/DefaultPasses.h"
-#include "llvm/PassManager.h"
+#include "llvm/ADT/SmallVector.h"
#include "llvm/Analysis/Passes.h"
-#include "llvm/Analysis/Verifier.h"
+#include "llvm/IR/Verifier.h"
+#include "llvm/PassManager.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/ManagedStatic.h"
#include "llvm/Target/TargetLibraryInfo.h"
-#include "llvm/Transforms/Scalar.h"
#include "llvm/Transforms/IPO.h"
-#include "llvm/ADT/SmallVector.h"
-#include "llvm/Support/ManagedStatic.h"
+#include "llvm/Transforms/Scalar.h"
+#include "llvm/Transforms/Vectorize.h"
using namespace llvm;
+static cl::opt<bool>
+RunLoopVectorization("vectorize-loops", cl::Hidden,
+ cl::desc("Run the Loop vectorization passes"));
+
+static cl::opt<bool>
+RunSLPVectorization("vectorize-slp", cl::Hidden,
+ cl::desc("Run the SLP vectorization passes"));
+
+static cl::opt<bool>
+RunBBVectorization("vectorize-slp-aggressive", cl::Hidden,
+ cl::desc("Run the BB vectorization passes"));
+
+static cl::opt<bool>
+UseGVNAfterVectorization("use-gvn-after-vectorization",
+ cl::init(false), cl::Hidden,
+ cl::desc("Run GVN instead of Early CSE after vectorization passes"));
+
+static cl::opt<bool> UseNewSROA("use-new-sroa",
+ cl::init(true), cl::Hidden,
+ cl::desc("Enable the new, experimental SROA pass"));
+
+static cl::opt<bool>
+RunLoopRerolling("reroll-loops", cl::Hidden,
+ cl::desc("Run the loop rerolling pass"));
+
+static cl::opt<bool> RunLoadCombine("combine-loads", cl::init(false),
+ cl::Hidden,
+ cl::desc("Run the load combining pass"));
+
PassManagerBuilder::PassManagerBuilder() {
OptLevel = 2;
SizeLevel = 0;
- LibraryInfo = 0;
- Inliner = 0;
- DisableSimplifyLibCalls = false;
+ LibraryInfo = nullptr;
+ Inliner = nullptr;
+ DisableTailCalls = false;
DisableUnitAtATime = false;
DisableUnrollLoops = false;
+ BBVectorize = RunBBVectorization;
+ SLPVectorize = RunSLPVectorization;
+ LoopVectorize = RunLoopVectorization;
+ RerollLoops = RunLoopRerolling;
+ LoadCombine = RunLoadCombine;
}
PassManagerBuilder::~PassManagerBuilder() {
addInitialAliasAnalysisPasses(FPM);
FPM.add(createCFGSimplificationPass());
- FPM.add(createScalarReplAggregatesPass());
+ if (UseNewSROA)
+ FPM.add(createSROAPass());
+ else
+ FPM.add(createScalarReplAggregatesPass());
FPM.add(createEarlyCSEPass());
FPM.add(createLowerExpectIntrinsicPass());
}
if (OptLevel == 0) {
if (Inliner) {
MPM.add(Inliner);
- Inliner = 0;
+ Inliner = nullptr;
}
+
+ // FIXME: This is a HACK! The inliner pass above implicitly creates a CGSCC
+ // pass manager, but we don't want to add extensions into that pass manager.
+ // To prevent this we must insert a no-op module pass to reset the pass
+ // manager to get the same behavior as EP_OptimizerLast in non-O0 builds.
+ if (!GlobalExtensions->empty() || !Extensions.empty())
+ MPM.add(createBarrierNoopPass());
+
+ addExtensionsToPM(EP_EnabledOnOptLevel0, MPM);
return;
}
addInitialAliasAnalysisPasses(MPM);
if (!DisableUnitAtATime) {
- MPM.add(createGlobalOptimizerPass()); // Optimize out global vars
+ addExtensionsToPM(EP_ModuleOptimizerEarly, MPM);
MPM.add(createIPSCCPPass()); // IP SCCP
+ MPM.add(createGlobalOptimizerPass()); // Optimize out global vars
+
MPM.add(createDeadArgEliminationPass()); // Dead argument elimination
MPM.add(createInstructionCombiningPass());// Clean up after IPCP & DAE
+ addExtensionsToPM(EP_Peephole, MPM);
MPM.add(createCFGSimplificationPass()); // Clean up after IPCP & DAE
}
MPM.add(createPruneEHPass()); // Remove dead EH info
if (Inliner) {
MPM.add(Inliner);
- Inliner = 0;
+ Inliner = nullptr;
}
if (!DisableUnitAtATime)
MPM.add(createFunctionAttrsPass()); // Set readonly/readnone attrs
// Start of function pass.
// Break up aggregate allocas, using SSAUpdater.
- MPM.add(createScalarReplAggregatesPass(-1, false));
+ if (UseNewSROA)
+ MPM.add(createSROAPass(/*RequiresDomTree*/ false));
+ else
+ MPM.add(createScalarReplAggregatesPass(-1, false));
MPM.add(createEarlyCSEPass()); // Catch trivial redundancies
- if (!DisableSimplifyLibCalls)
- MPM.add(createSimplifyLibCallsPass()); // Library Call Optimizations
MPM.add(createJumpThreadingPass()); // Thread jumps.
MPM.add(createCorrelatedValuePropagationPass()); // Propagate conditionals
MPM.add(createCFGSimplificationPass()); // Merge & remove BBs
MPM.add(createInstructionCombiningPass()); // Combine silly seq's
+ addExtensionsToPM(EP_Peephole, MPM);
- MPM.add(createTailCallEliminationPass()); // Eliminate tail calls
+ if (!DisableTailCalls)
+ MPM.add(createTailCallEliminationPass()); // Eliminate tail calls
MPM.add(createCFGSimplificationPass()); // Merge & remove BBs
MPM.add(createReassociatePass()); // Reassociate expressions
MPM.add(createLoopRotatePass()); // Rotate Loop
MPM.add(createIndVarSimplifyPass()); // Canonicalize indvars
MPM.add(createLoopIdiomPass()); // Recognize idioms like memset.
MPM.add(createLoopDeletionPass()); // Delete dead loops
+
if (!DisableUnrollLoops)
- MPM.add(createLoopUnrollPass()); // Unroll small loops
+ MPM.add(createSimpleLoopUnrollPass()); // Unroll small loops
addExtensionsToPM(EP_LoopOptimizerEnd, MPM);
if (OptLevel > 1)
// Run instcombine after redundancy elimination to exploit opportunities
// opened up by them.
MPM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, MPM);
MPM.add(createJumpThreadingPass()); // Thread jumps
MPM.add(createCorrelatedValuePropagationPass());
MPM.add(createDeadStoreEliminationPass()); // Delete dead stores
addExtensionsToPM(EP_ScalarOptimizerLate, MPM);
+ if (RerollLoops)
+ MPM.add(createLoopRerollPass());
+ if (SLPVectorize)
+ MPM.add(createSLPVectorizerPass()); // Vectorize parallel scalar chains.
+
+ if (BBVectorize) {
+ MPM.add(createBBVectorizePass());
+ MPM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, MPM);
+ if (OptLevel > 1 && UseGVNAfterVectorization)
+ MPM.add(createGVNPass()); // Remove redundancies
+ else
+ MPM.add(createEarlyCSEPass()); // Catch trivial redundancies
+
+ // BBVectorize may have significantly shortened a loop body; unroll again.
+ if (!DisableUnrollLoops)
+ MPM.add(createLoopUnrollPass());
+ }
+
+ if (LoadCombine)
+ MPM.add(createLoadCombinePass());
+
MPM.add(createAggressiveDCEPass()); // Delete dead instructions
- MPM.add(createCFGSimplificationPass()); // Merge & remove BBs
+ MPM.add(createCFGSimplificationPass()); // Merge & remove BBs
MPM.add(createInstructionCombiningPass()); // Clean up after everything.
+ addExtensionsToPM(EP_Peephole, MPM);
+
+ // FIXME: This is a HACK! The inliner pass above implicitly creates a CGSCC
+ // pass manager that we are specifically trying to avoid. To prevent this
+ // we must insert a no-op module pass to reset the pass manager.
+ MPM.add(createBarrierNoopPass());
+ MPM.add(createLoopVectorizePass(DisableUnrollLoops, LoopVectorize));
+ // FIXME: Because of #pragma vectorize enable, the passes below are always
+ // inserted in the pipeline, even when the vectorizer doesn't run (ex. when
+ // on -O1 and no #pragma is found). Would be good to have these two passes
+ // as function calls, so that we can only pass them when the vectorizer
+ // changed the code.
+ MPM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, MPM);
+ MPM.add(createCFGSimplificationPass());
+
+ if (!DisableUnrollLoops)
+ MPM.add(createLoopUnrollPass()); // Unroll small loops
if (!DisableUnitAtATime) {
// FIXME: We shouldn't bother with this anymore.
MPM.add(createStripDeadPrototypesPass()); // Get rid of dead prototypes
- // GlobalOpt already deletes dead functions and globals, at -O3 try a
+ // GlobalOpt already deletes dead functions and globals, at -O2 try a
// late pass of GlobalDCE. It is capable of deleting dead cycles.
- if (OptLevel > 2)
+ if (OptLevel > 1) {
MPM.add(createGlobalDCEPass()); // Remove dead fns and globals.
-
- if (OptLevel > 1)
MPM.add(createConstantMergePass()); // Merge dup global constants
+ }
}
+ addExtensionsToPM(EP_OptimizerLast, MPM);
}
void PassManagerBuilder::populateLTOPassManager(PassManagerBase &PM,
bool Internalize,
- bool RunInliner) {
+ bool RunInliner,
+ bool DisableGVNLoadPRE) {
// Provide AliasAnalysis services for optimizations.
addInitialAliasAnalysisPasses(PM);
// for a main function. If main is defined, mark all other functions
// internal.
if (Internalize)
- PM.add(createInternalizePass(true));
+ PM.add(createInternalizePass("main"));
// Propagate constants at call sites into the functions they call. This
// opens opportunities for globalopt (and inlining) by substituting function
// function pointers. When this happens, we often have to resolve varargs
// calls, etc, so let instcombine do this.
PM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, PM);
// Inline small functions
if (RunInliner)
// The IPO passes may leave cruft around. Clean up after them.
PM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, PM);
PM.add(createJumpThreadingPass());
+
// Break up allocas
- PM.add(createScalarReplAggregatesPass());
+ if (UseNewSROA)
+ PM.add(createSROAPass());
+ else
+ PM.add(createScalarReplAggregatesPass());
// Run a few AA driven optimizations here and now, to cleanup the code.
PM.add(createFunctionAttrsPass()); // Add nocapture.
PM.add(createGlobalsModRefPass()); // IP alias analysis.
- PM.add(createLICMPass()); // Hoist loop invariants.
- PM.add(createGVNPass()); // Remove redundancies.
- PM.add(createMemCpyOptPass()); // Remove dead memcpys.
+ PM.add(createLICMPass()); // Hoist loop invariants.
+ PM.add(createGVNPass(DisableGVNLoadPRE)); // Remove redundancies.
+ PM.add(createMemCpyOptPass()); // Remove dead memcpys.
+
// Nuke dead stores.
PM.add(createDeadStoreEliminationPass());
+ // More loops are countable; try to optimize them.
+ PM.add(createIndVarSimplifyPass());
+ PM.add(createLoopDeletionPass());
+ PM.add(createLoopVectorizePass(true, true));
+
+ // More scalar chains could be vectorized due to more alias information
+ PM.add(createSLPVectorizerPass()); // Vectorize parallel scalar chains.
+
+ if (LoadCombine)
+ PM.add(createLoadCombinePass());
+
// Cleanup and simplify the code after the scalar optimizations.
PM.add(createInstructionCombiningPass());
+ addExtensionsToPM(EP_Peephole, PM);
PM.add(createJumpThreadingPass());
PM.add(createGlobalDCEPass());
}
-LLVMPassManagerBuilderRef LLVMPassManagerBuilderCreate(void) {
+inline PassManagerBuilder *unwrap(LLVMPassManagerBuilderRef P) {
+ return reinterpret_cast<PassManagerBuilder*>(P);
+}
+
+inline LLVMPassManagerBuilderRef wrap(PassManagerBuilder *P) {
+ return reinterpret_cast<LLVMPassManagerBuilderRef>(P);
+}
+
+LLVMPassManagerBuilderRef LLVMPassManagerBuilderCreate() {
PassManagerBuilder *PMB = new PassManagerBuilder();
return wrap(PMB);
}
void
LLVMPassManagerBuilderSetDisableSimplifyLibCalls(LLVMPassManagerBuilderRef PMB,
LLVMBool Value) {
- PassManagerBuilder *Builder = unwrap(PMB);
- Builder->DisableSimplifyLibCalls = Value;
+ // NOTE: The simplify-libcalls pass has been removed.
}
void
void LLVMPassManagerBuilderPopulateLTOPassManager(LLVMPassManagerBuilderRef PMB,
LLVMPassManagerRef PM,
- bool Internalize,
- bool RunInliner) {
+ LLVMBool Internalize,
+ LLVMBool RunInliner) {
PassManagerBuilder *Builder = unwrap(PMB);
PassManagerBase *LPM = unwrap(PM);
- Builder->populateLTOPassManager(*LPM, Internalize, RunInliner);
+ Builder->populateLTOPassManager(*LPM, Internalize != 0, RunInliner != 0);
}
-
projects / oota-llvm.git / blobdiff