llvm-6502/lib/Transforms/IPO/Inliner.cpp

499 lines
19 KiB
C++
Raw Normal View History

//===- Inliner.cpp - Code common to all inliners --------------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file implements the mechanics required to implement inlining without
// missing any calls and updating the call graph. The decisions of which calls
// are profitable to inline are implemented elsewhere.
//
//===----------------------------------------------------------------------===//
#define DEBUG_TYPE "inline"
#include "llvm/Module.h"
#include "llvm/Instructions.h"
#include "llvm/IntrinsicInst.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/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 <set>
using namespace llvm;
STATISTIC(NumInlined, "Number of functions inlined");
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
STATISTIC(NumCallsDeleted, "Number of call sites deleted, not inlined");
STATISTIC(NumDeleted, "Number of functions deleted because all callers found");
STATISTIC(NumMergedAllocas, "Number of allocas merged together");
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)"));
static cl::opt<int>
HintThreshold("inlinehint-threshold", cl::Hidden, cl::init(325),
cl::desc("Threshold for inlining functions with inline hint"));
// Threshold to use when optsize is specified (and there is no -inline-limit).
const int OptSizeThreshold = 75;
Inliner::Inliner(void *ID)
: CallGraphSCCPass(ID), InlineThreshold(InlineLimit) {}
Inliner::Inliner(void *ID, int Threshold)
: CallGraphSCCPass(ID), InlineThreshold(Threshold) {}
/// 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);
}
typedef DenseMap<const ArrayType*, std::vector<AllocaInst*> >
InlinedArrayAllocasTy;
/// InlineCallIfPossible - If it is possible to inline the specified call site,
/// do so and update the CallGraph for this operation.
///
/// This function also does some basic book-keeping to update the IR. The
/// InlinedArrayAllocas map keeps track of any allocas that are already
/// available from other functions inlined into the caller. If we are able to
/// inline this call site we attempt to reuse already available allocas or add
/// any new allocas to the set if not possible.
static bool InlineCallIfPossible(CallSite CS, CallGraph &CG,
const TargetData *TD,
InlinedArrayAllocasTy &InlinedArrayAllocas) {
Function *Callee = CS.getCalledFunction();
Function *Caller = CS.getCaller();
// Try to inline the function. Get the list of static allocas that were
// inlined.
SmallVector<AllocaInst*, 16> StaticAllocas;
if (!InlineFunction(CS, &CG, TD, &StaticAllocas))
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);
// 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,
// then we know that they have disjoint lifetimes and that we can merge them.
//
// There are many heuristics possible for merging these allocas, and the
// different options have different tradeoffs. One thing that we *really*
// don't want to hurt is SRoA: once inlining happens, often allocas are no
// longer address taken and so they can be promoted.
//
// Our "solution" for that is to only merge allocas whose outermost type is an
// array type. These are usually not promoted because someone is using a
// variable index into them. These are also often the most important ones to
// merge.
//
// A better solution would be to have real memory lifetime markers in the IR
// and not have the inliner do any merging of allocas at all. This would
// allow the backend to do proper stack slot coloring of all allocas that
// *actually make it to the backend*, which is really what we want.
//
// Because we don't have this information, we do this simple and useful hack.
//
SmallPtrSet<AllocaInst*, 16> UsedAllocas;
// Loop over all the allocas we have so far and see if they can be merged with
// a previously inlined alloca. If not, remember that we had it.
for (unsigned AllocaNo = 0, e = StaticAllocas.size();
AllocaNo != e; ++AllocaNo) {
AllocaInst *AI = StaticAllocas[AllocaNo];
// Don't bother trying to merge array allocations (they will usually be
// canonicalized to be an allocation *of* an array), or allocations whose
// type is not itself an array (because we're afraid of pessimizing SRoA).
const ArrayType *ATy = dyn_cast<ArrayType>(AI->getAllocatedType());
if (ATy == 0 || AI->isArrayAllocation())
continue;
// Get the list of all available allocas for this array type.
std::vector<AllocaInst*> &AllocasForType = InlinedArrayAllocas[ATy];
// Loop over the allocas in AllocasForType to see if we can reuse one. Note
// that we have to be careful not to reuse the same "available" alloca for
// multiple different allocas that we just inlined, we use the 'UsedAllocas'
// set to keep track of which "available" allocas are being used by this
// function. Also, AllocasForType can be empty of course!
bool MergedAwayAlloca = false;
for (unsigned i = 0, e = AllocasForType.size(); i != e; ++i) {
AllocaInst *AvailableAlloca = AllocasForType[i];
// The available alloca has to be in the right function, not in some other
// function in this SCC.
if (AvailableAlloca->getParent() != AI->getParent())
continue;
// If the inlined function already uses this alloca then we can't reuse
// it.
if (!UsedAllocas.insert(AvailableAlloca))
continue;
// Otherwise, we *can* reuse it, RAUW AI into AvailableAlloca and declare
// success!
DEBUG(dbgs() << " ***MERGED ALLOCA: " << *AI);
AI->replaceAllUsesWith(AvailableAlloca);
AI->eraseFromParent();
MergedAwayAlloca = true;
++NumMergedAllocas;
break;
}
// If we already nuked the alloca, we're done with it.
if (MergedAwayAlloca)
continue;
// If we were unable to merge away the alloca either because there are no
// allocas of the right type available or because we reused them all
// already, remember that this alloca came from an inlined function and mark
// it used so we don't reuse it for other allocas from this inline
// operation.
AllocasForType.push_back(AI);
UsedAllocas.insert(AI);
}
return true;
}
unsigned Inliner::getInlineThreshold(CallSite CS) const {
int thres = InlineThreshold;
// Listen to optsize when -inline-limit is not given.
Function *Caller = CS.getCaller();
if (Caller && !Caller->isDeclaration() &&
Caller->hasFnAttr(Attribute::OptimizeForSize) &&
InlineLimit.getNumOccurrences() == 0)
thres = OptSizeThreshold;
// Listen to inlinehint when it would increase the threshold.
Function *Callee = CS.getCalledFunction();
if (HintThreshold > thres && Callee && !Callee->isDeclaration() &&
Callee->hasFnAttr(Attribute::InlineHint))
thres = HintThreshold;
return thres;
}
/// shouldInline - Return true if the inliner should attempt to inline
/// at the given CallSite.
bool Inliner::shouldInline(CallSite CS) {
InlineCost IC = getInlineCost(CS);
if (IC.isAlways()) {
DEBUG(dbgs() << " Inlining: cost=always"
<< ", Call: " << *CS.getInstruction() << "\n");
return true;
}
if (IC.isNever()) {
DEBUG(dbgs() << " NOT Inlining: cost=never"
<< ", Call: " << *CS.getInstruction() << "\n");
return false;
}
int Cost = IC.getValue();
Function *Caller = CS.getCaller();
int CurrentThreshold = getInlineThreshold(CS);
float FudgeFactor = getInlineFudgeFactor(CS);
int AdjThreshold = (int)(CurrentThreshold * FudgeFactor);
if (Cost >= AdjThreshold) {
DEBUG(dbgs() << " NOT Inlining: cost=" << Cost
<< ", thres=" << AdjThreshold
<< ", Call: " << *CS.getInstruction() << "\n");
return false;
}
// Try to detect the case where the current inlining candidate caller
// (call it B) is a static function and is an inlining candidate elsewhere,
// and the current candidate callee (call it C) is large enough that
// inlining it into B would make B too big to inline later. In these
// circumstances it may be best not to inline C into B, but to inline B
// into its callers.
if (Caller->hasLocalLinkage()) {
int TotalSecondaryCost = 0;
bool outerCallsFound = false;
bool allOuterCallsWillBeInlined = true;
bool someOuterCallWouldNotBeInlined = false;
for (Value::use_iterator I = Caller->use_begin(), E =Caller->use_end();
I != E; ++I) {
CallSite CS2 = CallSite::get(*I);
// If this isn't a call to Caller (it could be some other sort
// of reference) skip it.
if (CS2.getInstruction() == 0 || CS2.getCalledFunction() != Caller)
continue;
InlineCost IC2 = getInlineCost(CS2);
if (IC2.isNever())
allOuterCallsWillBeInlined = false;
if (IC2.isAlways() || IC2.isNever())
continue;
outerCallsFound = true;
int Cost2 = IC2.getValue();
int CurrentThreshold2 = getInlineThreshold(CS2);
float FudgeFactor2 = getInlineFudgeFactor(CS2);
if (Cost2 >= (int)(CurrentThreshold2 * FudgeFactor2))
allOuterCallsWillBeInlined = false;
// See if we have this case. We subtract off the penalty
// for the call instruction, which we would be deleting.
if (Cost2 < (int)(CurrentThreshold2 * FudgeFactor2) &&
Cost2 + Cost - (InlineConstants::CallPenalty + 1) >=
(int)(CurrentThreshold2 * FudgeFactor2)) {
someOuterCallWouldNotBeInlined = true;
TotalSecondaryCost += Cost2;
}
}
// If all outer calls to Caller would get inlined, the cost for the last
// one is set very low by getInlineCost, in anticipation that Caller will
// be removed entirely. We did not account for this above unless there
// is only one caller of Caller.
if (allOuterCallsWillBeInlined && Caller->use_begin() != Caller->use_end())
TotalSecondaryCost += InlineConstants::LastCallToStaticBonus;
if (outerCallsFound && someOuterCallWouldNotBeInlined &&
TotalSecondaryCost < Cost) {
DEBUG(dbgs() << " NOT Inlining: " << *CS.getInstruction() <<
" Cost = " << Cost <<
", outer Cost = " << TotalSecondaryCost << '\n');
return false;
}
}
DEBUG(dbgs() << " Inlining: cost=" << Cost
<< ", thres=" << AdjThreshold
<< ", Call: " << *CS.getInstruction() << '\n');
return true;
}
bool Inliner::runOnSCC(std::vector<CallGraphNode*> &SCC) {
CallGraph &CG = getAnalysis<CallGraph>();
const TargetData *TD = getAnalysisIfAvailable<TargetData>();
SmallPtrSet<Function*, 8> SCCFunctions;
DEBUG(dbgs() << "Inliner visiting SCC:");
for (unsigned i = 0, e = SCC.size(); i != e; ++i) {
Function *F = SCC[i]->getFunction();
if (F) SCCFunctions.insert(F);
DEBUG(dbgs() << " " << (F ? F->getName() : "INDIRECTNODE"));
}
// Scan through and identify all call sites ahead of time so that we only
// inline call sites in the original functions, not call sites that result
// from inlining other functions.
SmallVector<CallSite, 16> CallSites;
for (unsigned i = 0, e = SCC.size(); i != e; ++i) {
Function *F = SCC[i]->getFunction();
if (!F) continue;
for (Function::iterator BB = F->begin(), E = F->end(); BB != E; ++BB)
for (BasicBlock::iterator I = BB->begin(), E = BB->end(); I != E; ++I) {
CallSite CS = CallSite::get(I);
// If this isn't a call, or it is a call to an intrinsic, it can
// never be inlined.
if (CS.getInstruction() == 0 || isa<IntrinsicInst>(I))
continue;
// If this is a direct call to an external function, we can never inline
// it. If it is an indirect call, inlining may resolve it to be a
// direct call, so we keep it.
if (CS.getCalledFunction() && CS.getCalledFunction()->isDeclaration())
continue;
CallSites.push_back(CS);
}
}
DEBUG(dbgs() << ": " << CallSites.size() << " call sites.\n");
// Now that we have all of the call sites, move the ones to functions in the
// current SCC to the end of the list.
unsigned FirstCallInSCC = CallSites.size();
for (unsigned i = 0; i < FirstCallInSCC; ++i)
if (Function *F = CallSites[i].getCalledFunction())
if (SCCFunctions.count(F))
std::swap(CallSites[i--], CallSites[--FirstCallInSCC]);
InlinedArrayAllocasTy InlinedArrayAllocas;
// Now that we have all of the call sites, loop over them and inline them if
// it looks profitable to do so.
bool Changed = false;
bool LocalChange;
do {
LocalChange = false;
// Iterate over the outer loop because inlining functions can cause indirect
// calls to become direct calls.
for (unsigned CSi = 0; CSi != CallSites.size(); ++CSi) {
CallSite CS = CallSites[CSi];
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
Function *Caller = CS.getCaller();
Function *Callee = CS.getCalledFunction();
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
// If this call site is dead and it is to a readonly function, we should
// 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())) {
DEBUG(dbgs() << " -> Deleting dead call: "
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
<< *CS.getInstruction() << "\n");
// Update the call graph by deleting the edge from Callee to Caller.
CG[Caller]->removeCallEdgeFor(CS);
CS.getInstruction()->eraseFromParent();
++NumCallsDeleted;
// Update the cached cost info with the missing call
growCachedCostInfo(Caller, NULL);
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
} else {
// We can only inline direct calls to non-declarations.
if (Callee == 0 || Callee->isDeclaration()) continue;
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
// If the policy determines that we should inline this function,
// try to do so.
if (!shouldInline(CS))
continue;
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
// Attempt to inline the function...
if (!InlineCallIfPossible(CS, CG, TD, InlinedArrayAllocas))
continue;
++NumInlined;
// Update the cached cost info with the inlined call.
growCachedCostInfo(Caller, Callee);
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
}
implement a nice little efficiency hack in the inliner. Since we're now running IPSCCP early, and we run functionattrs interlaced with the inliner, we often (particularly for small or noop functions) completely propagate all of the information about a call to its call site in IPSSCP (making a call dead) and functionattrs is smart enough to realize that the function is readonly (because it is interlaced with inliner). To improve compile time and make the inliner threshold more accurate, realize that we don't have to inline dead readonly function calls. Instead, just delete the call. This happens all the time for C++ codes, here are some counters from opt/llvm-ld counting the number of times calls were deleted vs inlined on various apps: Tramp3d opt: 5033 inline - Number of call sites deleted, not inlined 24596 inline - Number of functions inlined llvm-ld: 667 inline - Number of functions deleted because all callers found 699 inline - Number of functions inlined 483.xalancbmk opt: 8096 inline - Number of call sites deleted, not inlined 62528 inline - Number of functions inlined llvm-ld: 217 inline - Number of allocas merged together 2158 inline - Number of functions inlined 471.omnetpp: 331 inline - Number of call sites deleted, not inlined 8981 inline - Number of functions inlined llvm-ld: 171 inline - Number of functions deleted because all callers found 629 inline - Number of functions inlined Deleting a call is much faster than inlining it, and is insensitive to the size of the callee. :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86975 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-12 07:56:08 +00:00
// If we inlined or deleted the last possible call site to the function,
// delete the function body now.
if (Callee && Callee->use_empty() && Callee->hasLocalLinkage() &&
// TODO: Can remove if in SCC now.
!SCCFunctions.count(Callee) &&
// The function may be apparently dead, but if there are indirect
// callgraph references to the node, we cannot delete it yet, this
// could invalidate the CGSCC iterator.
CG[Callee]->getNumReferences() == 0) {
DEBUG(dbgs() << " -> Deleting dead function: "
<< Callee->getName() << "\n");
CallGraphNode *CalleeNode = CG[Callee];
// Remove any call graph edges from the callee to its callees.
CalleeNode->removeAllCalledFunctions();
resetCachedCostInfo(Callee);
// Removing the node for callee from the call graph and delete it.
delete CG.removeFunctionFromModule(CalleeNode);
++NumDeleted;
}
// Remove this call site from the list. If possible, use
// swap/pop_back for efficiency, but do not use it if doing so would
// move a call site to a function in this SCC before the
// 'FirstCallInSCC' barrier.
if (SCC.size() == 1) {
std::swap(CallSites[CSi], CallSites.back());
CallSites.pop_back();
} else {
CallSites.erase(CallSites.begin()+CSi);
}
--CSi;
Changed = true;
LocalChange = true;
}
} while (LocalChange);
return Changed;
}
// doFinalization - Remove now-dead linkonce functions at the end of
// processing to avoid breaking the SCC traversal.
bool Inliner::doFinalization(CallGraph &CG) {
return removeDeadFunctions(CG);
}
/// removeDeadFunctions - Remove dead functions that are not included in
/// DNR (Do Not Remove) list.
bool Inliner::removeDeadFunctions(CallGraph &CG,
SmallPtrSet<const Function *, 16> *DNR) {
SmallPtrSet<CallGraphNode*, 16> FunctionsToRemove;
// Scan for all of the functions, looking for ones that should now be removed
// from the program. Insert the dead ones in the FunctionsToRemove set.
for (CallGraph::iterator I = CG.begin(), E = CG.end(); I != E; ++I) {
CallGraphNode *CGN = I->second;
if (CGN->getFunction() == 0)
continue;
Function *F = CGN->getFunction();
// If the only remaining users of the function are dead constants, remove
// them.
F->removeDeadConstantUsers();
if (DNR && DNR->count(F))
continue;
if (!F->hasLinkOnceLinkage() && !F->hasLocalLinkage() &&
!F->hasAvailableExternallyLinkage())
continue;
if (!F->use_empty())
continue;
// Remove any call graph edges from the function to its callees.
CGN->removeAllCalledFunctions();
// Remove any edges from the external node to the function's call graph
// node. These edges might have been made irrelegant due to
// optimization of the program.
CG.getExternalCallingNode()->removeAnyCallEdgeTo(CGN);
// Removing the node for callee from the call graph and delete it.
FunctionsToRemove.insert(CGN);
}
// Now that we know which functions to delete, do so. We didn't want to do
// this inline, because that would invalidate our CallGraph::iterator
// objects. :(
//
// Note that it doesn't matter that we are iterating over a non-stable set
// here to do this, it doesn't matter which order the functions are deleted
// in.
bool Changed = false;
for (SmallPtrSet<CallGraphNode*, 16>::iterator I = FunctionsToRemove.begin(),
E = FunctionsToRemove.end(); I != E; ++I) {
resetCachedCostInfo((*I)->getFunction());
delete CG.removeFunctionFromModule(*I);
++NumDeleted;
Changed = true;
}
return Changed;
}