//===----------------------------------------------------------------------===//
#define DEBUG_TYPE "inline"
-#include "llvm/Module.h"
-#include "llvm/Instructions.h"
-#include "llvm/IntrinsicInst.h"
+#include "llvm/Transforms/IPO/InlinerPass.h"
+#include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/CallGraph.h"
#include "llvm/Analysis/InlineCost.h"
-#include "llvm/Target/TargetData.h"
-#include "llvm/Transforms/IPO/InlinerPass.h"
-#include "llvm/Transforms/Utils/Cloning.h"
-#include "llvm/Transforms/Utils/Local.h"
+#include "llvm/IR/DataLayout.h"
+#include "llvm/IR/Instructions.h"
+#include "llvm/IR/IntrinsicInst.h"
+#include "llvm/IR/Module.h"
#include "llvm/Support/CallSite.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/raw_ostream.h"
-#include "llvm/ADT/SmallPtrSet.h"
-#include "llvm/ADT/Statistic.h"
+#include "llvm/Target/TargetLibraryInfo.h"
+#include "llvm/Transforms/Utils/Cloning.h"
+#include "llvm/Transforms/Utils/Local.h"
using namespace llvm;
STATISTIC(NumInlined, "Number of functions inlined");
STATISTIC(NumDeleted, "Number of functions deleted because all callers found");
STATISTIC(NumMergedAllocas, "Number of allocas merged together");
+// This weirdly named statistic tracks the number of times that, when attempting
+// to inline a function A into B, we analyze the callers of B in order to see
+// if those would be more profitable and blocked inline steps.
+STATISTIC(NumCallerCallersAnalyzed, "Number of caller-callers analyzed");
+
static cl::opt<int>
InlineLimit("inline-threshold", cl::Hidden, cl::init(225), cl::ZeroOrMore,
cl::desc("Control the amount of inlining to perform (default = 225)"));
/// getAnalysisUsage - For this class, we declare that we require and preserve
/// the call graph. If the derived class implements this method, it should
/// always explicitly call the implementation here.
-void Inliner::getAnalysisUsage(AnalysisUsage &Info) const {
- CallGraphSCCPass::getAnalysisUsage(Info);
+void Inliner::getAnalysisUsage(AnalysisUsage &AU) const {
+ CallGraphSCCPass::getAnalysisUsage(AU);
}
typedef DenseMap<ArrayType*, std::vector<AllocaInst*> >
InlinedArrayAllocasTy;
+/// \brief If the inlined function had a higher stack protection level than the
+/// calling function, then bump up the caller's stack protection level.
+static void AdjustCallerSSPLevel(Function *Caller, Function *Callee) {
+ // If upgrading the SSP attribute, clear out the old SSP Attributes first.
+ // Having multiple SSP attributes doesn't actually hurt, but it adds useless
+ // clutter to the IR.
+ AttrBuilder B;
+ B.addAttribute(Attribute::StackProtect)
+ .addAttribute(Attribute::StackProtectStrong);
+ AttributeSet OldSSPAttr = AttributeSet::get(Caller->getContext(),
+ AttributeSet::FunctionIndex,
+ B);
+ AttributeSet CallerAttr = Caller->getAttributes(),
+ CalleeAttr = Callee->getAttributes();
+
+ if (CalleeAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtectReq)) {
+ Caller->removeAttributes(AttributeSet::FunctionIndex, OldSSPAttr);
+ Caller->addFnAttr(Attribute::StackProtectReq);
+ } else if (CalleeAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtectStrong) &&
+ !CallerAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtectReq)) {
+ Caller->removeAttributes(AttributeSet::FunctionIndex, OldSSPAttr);
+ Caller->addFnAttr(Attribute::StackProtectStrong);
+ } else if (CalleeAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtect) &&
+ !CallerAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtectReq) &&
+ !CallerAttr.hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::StackProtectStrong))
+ Caller->addFnAttr(Attribute::StackProtect);
+}
+
/// InlineCallIfPossible - If it is possible to inline the specified call site,
/// do so and update the CallGraph for this operation.
///
/// any new allocas to the set if not possible.
static bool InlineCallIfPossible(CallSite CS, InlineFunctionInfo &IFI,
InlinedArrayAllocasTy &InlinedArrayAllocas,
- int InlineHistory, bool InsertLifetime) {
+ int InlineHistory, bool InsertLifetime,
+ const DataLayout *TD) {
Function *Callee = CS.getCalledFunction();
Function *Caller = CS.getCaller();
if (!InlineFunction(CS, IFI, InsertLifetime))
return false;
- // If the inlined function had a higher stack protection level than the
- // calling function, then bump up the caller's stack protection level.
- if (Callee->hasFnAttr(Attribute::StackProtectReq))
- Caller->addFnAttr(Attribute::StackProtectReq);
- else if (Callee->hasFnAttr(Attribute::StackProtect) &&
- !Caller->hasFnAttr(Attribute::StackProtectReq))
- Caller->addFnAttr(Attribute::StackProtect);
+ AdjustCallerSSPLevel(Caller, Callee);
// Look at all of the allocas that we inlined through this call site. If we
// have already inlined other allocas through other calls into this function,
bool MergedAwayAlloca = false;
for (unsigned i = 0, e = AllocasForType.size(); i != e; ++i) {
AllocaInst *AvailableAlloca = AllocasForType[i];
+
+ unsigned Align1 = AI->getAlignment(),
+ Align2 = AvailableAlloca->getAlignment();
+ // If we don't have data layout information, and only one alloca is using
+ // the target default, then we can't safely merge them because we can't
+ // pick the greater alignment.
+ if (!TD && (!Align1 || !Align2) && Align1 != Align2)
+ continue;
// The available alloca has to be in the right function, not in some other
// function in this SCC.
<< *AvailableAlloca << '\n');
AI->replaceAllUsesWith(AvailableAlloca);
+
+ if (Align1 != Align2) {
+ if (!Align1 || !Align2) {
+ assert(TD && "DataLayout required to compare default alignments");
+ unsigned TypeAlign = TD->getABITypeAlignment(AI->getAllocatedType());
+
+ Align1 = Align1 ? Align1 : TypeAlign;
+ Align2 = Align2 ? Align2 : TypeAlign;
+ }
+
+ if (Align1 > Align2)
+ AvailableAlloca->setAlignment(AI->getAlignment());
+ }
+
AI->eraseFromParent();
MergedAwayAlloca = true;
++NumMergedAllocas;
}
unsigned Inliner::getInlineThreshold(CallSite CS) const {
- int thres = InlineThreshold;
+ int thres = InlineThreshold; // -inline-threshold or else selected by
+ // overall opt level
- // Listen to optsize when -inline-limit is not given.
+ // If -inline-threshold is not given, listen to the optsize attribute when it
+ // would decrease the threshold.
Function *Caller = CS.getCaller();
- if (Caller && !Caller->isDeclaration() &&
- Caller->hasFnAttr(Attribute::OptimizeForSize) &&
- InlineLimit.getNumOccurrences() == 0)
+ bool OptSize = Caller && !Caller->isDeclaration() &&
+ Caller->getAttributes().hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::OptimizeForSize);
+ if (!(InlineLimit.getNumOccurrences() > 0) && OptSize &&
+ OptSizeThreshold < thres)
thres = OptSizeThreshold;
- // Listen to inlinehint when it would increase the threshold.
+ // Listen to the inlinehint attribute when it would increase the threshold
+ // and the caller does not need to minimize its size.
Function *Callee = CS.getCalledFunction();
- if (HintThreshold > thres && Callee && !Callee->isDeclaration() &&
- Callee->hasFnAttr(Attribute::InlineHint))
+ bool InlineHint = Callee && !Callee->isDeclaration() &&
+ Callee->getAttributes().hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::InlineHint);
+ if (InlineHint && HintThreshold > thres
+ && !Caller->getAttributes().hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::MinSize))
thres = HintThreshold;
return thres;
}
InlineCost IC2 = getInlineCost(CS2);
+ ++NumCallerCallersAnalyzed;
if (!IC2) {
callerWillBeRemoved = false;
continue;
bool Inliner::runOnSCC(CallGraphSCC &SCC) {
CallGraph &CG = getAnalysis<CallGraph>();
- const TargetData *TD = getAnalysisIfAvailable<TargetData>();
+ const DataLayout *TD = getAnalysisIfAvailable<DataLayout>();
+ const TargetLibraryInfo *TLI = getAnalysisIfAvailable<TargetLibraryInfo>();
SmallPtrSet<Function*, 8> SCCFunctions;
DEBUG(dbgs() << "Inliner visiting SCC:");
// just delete the call instead of trying to inline it, regardless of
// size. This happens because IPSCCP propagates the result out of the
// call and then we're left with the dead call.
- if (isInstructionTriviallyDead(CS.getInstruction())) {
+ if (isInstructionTriviallyDead(CS.getInstruction(), TLI)) {
DEBUG(dbgs() << " -> Deleting dead call: "
<< *CS.getInstruction() << "\n");
// Update the call graph by deleting the edge from Callee to Caller.
// Attempt to inline the function.
if (!InlineCallIfPossible(CS, InlineInfo, InlinedArrayAllocas,
- InlineHistoryID, InsertLifetime))
+ InlineHistoryID, InsertLifetime, TD))
continue;
++NumInlined;
// Handle the case when this function is called and we only want to care
// about always-inline functions. This is a bit of a hack to share code
// between here and the InlineAlways pass.
- if (AlwaysInlineOnly && !F->hasFnAttr(Attribute::AlwaysInline))
+ if (AlwaysInlineOnly &&
+ !F->getAttributes().hasAttribute(AttributeSet::FunctionIndex,
+ Attribute::AlwaysInline))
continue;
// If the only remaining users of the function are dead constants, remove
// Note that it doesn't matter that we are iterating over a non-stable order
// here to do this, it doesn't matter which order the functions are deleted
// in.
- std::sort(FunctionsToRemove.begin(), FunctionsToRemove.end());
+ array_pod_sort(FunctionsToRemove.begin(), FunctionsToRemove.end());
FunctionsToRemove.erase(std::unique(FunctionsToRemove.begin(),
FunctionsToRemove.end()),
FunctionsToRemove.end());