llvm-project/llvm/lib/Transforms/Scalar/SimplifyCFGPass.cpp

235 lines
8.5 KiB
C++
Raw Normal View History

//===- SimplifyCFGPass.cpp - CFG Simplification Pass ----------------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
2007-12-04 03:43:18 +08:00
// This file implements dead code elimination and basic block merging, along
// with a collection of other peephole control flow optimizations. For example:
//
// * Removes basic blocks with no predecessors.
// * Merges a basic block into its predecessor if there is only one and the
// predecessor only has one successor.
// * Eliminates PHI nodes for basic blocks with a single predecessor.
// * Eliminates a basic block that only contains an unconditional branch.
2007-12-04 03:43:18 +08:00
// * Changes invoke instructions to nounwind functions to be calls.
// * Change things like "if (x) if (y)" into "if (x&y)".
// * etc..
//
//===----------------------------------------------------------------------===//
#include "llvm/Transforms/Scalar/SimplifyCFG.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/AssumptionCache.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/IR/Attributes.h"
#include "llvm/IR/CFG.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DataLayout.h"
#include "llvm/IR/Instructions.h"
#include "llvm/IR/IntrinsicInst.h"
#include "llvm/IR/Module.h"
#include "llvm/Pass.h"
[SimplifyCFG] threshold for folding branches with common destination Summary: This patch adds a threshold that controls the number of bonus instructions allowed for folding branches with common destination. The original code allows at most one bonus instruction. With this patch, users can customize the threshold to allow multiple bonus instructions. The default threshold is still 1, so that the code behaves the same as before when users do not specify this threshold. The motivation of this change is that tuning this threshold significantly (up to 25%) improves the performance of some CUDA programs in our internal code base. In general, branch instructions are very expensive for GPU programs. Therefore, it is sometimes worth trading more arithmetic computation for a more straightened control flow. Here's a reduced example: __global__ void foo(int a, int b, int c, int d, int e, int n, const int *input, int *output) { int sum = 0; for (int i = 0; i < n; ++i) sum += (((i ^ a) > b) && (((i | c ) ^ d) > e)) ? 0 : input[i]; *output = sum; } The select statement in the loop body translates to two branch instructions "if ((i ^ a) > b)" and "if (((i | c) ^ d) > e)" which share a common destination. With the default threshold, SimplifyCFG is unable to fold them, because computing the condition of the second branch "(i | c) ^ d > e" requires two bonus instructions. With the threshold increased, SimplifyCFG can fold the two branches so that the loop body contains only one branch, making the code conceptually look like: sum += (((i ^ a) > b) & (((i | c ) ^ d) > e)) ? 0 : input[i]; Increasing the threshold significantly improves the performance of this particular example. In the configuration where both conditions are guaranteed to be true, increasing the threshold from 1 to 2 improves the performance by 18.24%. Even in the configuration where the first condition is false and the second condition is true, which favors shortcuts, increasing the threshold from 1 to 2 still improves the performance by 4.35%. We are still looking for a good threshold and maybe a better cost model than just counting the number of bonus instructions. However, according to the above numbers, we think it is at least worth adding a threshold to enable more experiments and tuning. Let me know what you think. Thanks! Test Plan: Added one test case to check the threshold is in effect Reviewers: nadav, eliben, meheff, resistor, hfinkel Reviewed By: hfinkel Subscribers: hfinkel, llvm-commits Differential Revision: http://reviews.llvm.org/D5529 llvm-svn: 218711
2014-10-01 06:23:38 +08:00
#include "llvm/Support/CommandLine.h"
#include "llvm/Transforms/Utils/Local.h"
#include "llvm/Transforms/Scalar.h"
using namespace llvm;
#define DEBUG_TYPE "simplifycfg"
[SimplifyCFG] threshold for folding branches with common destination Summary: This patch adds a threshold that controls the number of bonus instructions allowed for folding branches with common destination. The original code allows at most one bonus instruction. With this patch, users can customize the threshold to allow multiple bonus instructions. The default threshold is still 1, so that the code behaves the same as before when users do not specify this threshold. The motivation of this change is that tuning this threshold significantly (up to 25%) improves the performance of some CUDA programs in our internal code base. In general, branch instructions are very expensive for GPU programs. Therefore, it is sometimes worth trading more arithmetic computation for a more straightened control flow. Here's a reduced example: __global__ void foo(int a, int b, int c, int d, int e, int n, const int *input, int *output) { int sum = 0; for (int i = 0; i < n; ++i) sum += (((i ^ a) > b) && (((i | c ) ^ d) > e)) ? 0 : input[i]; *output = sum; } The select statement in the loop body translates to two branch instructions "if ((i ^ a) > b)" and "if (((i | c) ^ d) > e)" which share a common destination. With the default threshold, SimplifyCFG is unable to fold them, because computing the condition of the second branch "(i | c) ^ d > e" requires two bonus instructions. With the threshold increased, SimplifyCFG can fold the two branches so that the loop body contains only one branch, making the code conceptually look like: sum += (((i ^ a) > b) & (((i | c ) ^ d) > e)) ? 0 : input[i]; Increasing the threshold significantly improves the performance of this particular example. In the configuration where both conditions are guaranteed to be true, increasing the threshold from 1 to 2 improves the performance by 18.24%. Even in the configuration where the first condition is false and the second condition is true, which favors shortcuts, increasing the threshold from 1 to 2 still improves the performance by 4.35%. We are still looking for a good threshold and maybe a better cost model than just counting the number of bonus instructions. However, according to the above numbers, we think it is at least worth adding a threshold to enable more experiments and tuning. Let me know what you think. Thanks! Test Plan: Added one test case to check the threshold is in effect Reviewers: nadav, eliben, meheff, resistor, hfinkel Reviewed By: hfinkel Subscribers: hfinkel, llvm-commits Differential Revision: http://reviews.llvm.org/D5529 llvm-svn: 218711
2014-10-01 06:23:38 +08:00
static cl::opt<unsigned>
UserBonusInstThreshold("bonus-inst-threshold", cl::Hidden, cl::init(1),
cl::desc("Control the number of bonus instructions (default = 1)"));
STATISTIC(NumSimpl, "Number of blocks simplified");
/// mergeEmptyReturnBlocks - If we have more than one empty (other than phi
/// node) return blocks, merge them together to promote recursive block merging.
static bool mergeEmptyReturnBlocks(Function &F) {
bool Changed = false;
2012-07-24 18:51:42 +08:00
BasicBlock *RetBlock = nullptr;
2012-07-24 18:51:42 +08:00
// Scan all the blocks in the function, looking for empty return blocks.
for (Function::iterator BBI = F.begin(), E = F.end(); BBI != E; ) {
BasicBlock &BB = *BBI++;
2012-07-24 18:51:42 +08:00
// Only look at return blocks.
ReturnInst *Ret = dyn_cast<ReturnInst>(BB.getTerminator());
if (!Ret) continue;
2012-07-24 18:51:42 +08:00
// Only look at the block if it is empty or the only other thing in it is a
// single PHI node that is the operand to the return.
if (Ret != &BB.front()) {
// Check for something else in the block.
BasicBlock::iterator I = Ret;
--I;
// Skip over debug info.
while (isa<DbgInfoIntrinsic>(I) && I != BB.begin())
--I;
if (!isa<DbgInfoIntrinsic>(I) &&
(!isa<PHINode>(I) || I != BB.begin() ||
Ret->getNumOperands() == 0 ||
Ret->getOperand(0) != I))
continue;
}
// If this is the first returning block, remember it and keep going.
if (!RetBlock) {
RetBlock = &BB;
continue;
}
2012-07-24 18:51:42 +08:00
// Otherwise, we found a duplicate return block. Merge the two.
Changed = true;
2012-07-24 18:51:42 +08:00
// Case when there is no input to the return or when the returned values
// agree is trivial. Note that they can't agree if there are phis in the
// blocks.
if (Ret->getNumOperands() == 0 ||
2012-07-24 18:51:42 +08:00
Ret->getOperand(0) ==
cast<ReturnInst>(RetBlock->getTerminator())->getOperand(0)) {
BB.replaceAllUsesWith(RetBlock);
BB.eraseFromParent();
continue;
}
2012-07-24 18:51:42 +08:00
// If the canonical return block has no PHI node, create one now.
PHINode *RetBlockPHI = dyn_cast<PHINode>(RetBlock->begin());
if (!RetBlockPHI) {
Value *InVal = cast<ReturnInst>(RetBlock->getTerminator())->getOperand(0);
pred_iterator PB = pred_begin(RetBlock), PE = pred_end(RetBlock);
RetBlockPHI = PHINode::Create(Ret->getOperand(0)->getType(),
std::distance(PB, PE), "merge",
&RetBlock->front());
2012-07-24 18:51:42 +08:00
for (pred_iterator PI = PB; PI != PE; ++PI)
RetBlockPHI->addIncoming(InVal, *PI);
RetBlock->getTerminator()->setOperand(0, RetBlockPHI);
}
2012-07-24 18:51:42 +08:00
// Turn BB into a block that just unconditionally branches to the return
// block. This handles the case when the two return blocks have a common
// predecessor but that return different things.
RetBlockPHI->addIncoming(Ret->getOperand(0), &BB);
BB.getTerminator()->eraseFromParent();
BranchInst::Create(RetBlock, &BB);
}
2012-07-24 18:51:42 +08:00
return Changed;
}
/// iterativelySimplifyCFG - Call SimplifyCFG on all the blocks in the function,
/// iterating until no more changes are made.
static bool iterativelySimplifyCFG(Function &F, const TargetTransformInfo &TTI,
const DataLayout *DL, AssumptionCache *AC,
[SimplifyCFG] threshold for folding branches with common destination Summary: This patch adds a threshold that controls the number of bonus instructions allowed for folding branches with common destination. The original code allows at most one bonus instruction. With this patch, users can customize the threshold to allow multiple bonus instructions. The default threshold is still 1, so that the code behaves the same as before when users do not specify this threshold. The motivation of this change is that tuning this threshold significantly (up to 25%) improves the performance of some CUDA programs in our internal code base. In general, branch instructions are very expensive for GPU programs. Therefore, it is sometimes worth trading more arithmetic computation for a more straightened control flow. Here's a reduced example: __global__ void foo(int a, int b, int c, int d, int e, int n, const int *input, int *output) { int sum = 0; for (int i = 0; i < n; ++i) sum += (((i ^ a) > b) && (((i | c ) ^ d) > e)) ? 0 : input[i]; *output = sum; } The select statement in the loop body translates to two branch instructions "if ((i ^ a) > b)" and "if (((i | c) ^ d) > e)" which share a common destination. With the default threshold, SimplifyCFG is unable to fold them, because computing the condition of the second branch "(i | c) ^ d > e" requires two bonus instructions. With the threshold increased, SimplifyCFG can fold the two branches so that the loop body contains only one branch, making the code conceptually look like: sum += (((i ^ a) > b) & (((i | c ) ^ d) > e)) ? 0 : input[i]; Increasing the threshold significantly improves the performance of this particular example. In the configuration where both conditions are guaranteed to be true, increasing the threshold from 1 to 2 improves the performance by 18.24%. Even in the configuration where the first condition is false and the second condition is true, which favors shortcuts, increasing the threshold from 1 to 2 still improves the performance by 4.35%. We are still looking for a good threshold and maybe a better cost model than just counting the number of bonus instructions. However, according to the above numbers, we think it is at least worth adding a threshold to enable more experiments and tuning. Let me know what you think. Thanks! Test Plan: Added one test case to check the threshold is in effect Reviewers: nadav, eliben, meheff, resistor, hfinkel Reviewed By: hfinkel Subscribers: hfinkel, llvm-commits Differential Revision: http://reviews.llvm.org/D5529 llvm-svn: 218711
2014-10-01 06:23:38 +08:00
unsigned BonusInstThreshold) {
bool Changed = false;
bool LocalChange = true;
while (LocalChange) {
LocalChange = false;
2012-07-24 18:51:42 +08:00
// Loop over all of the basic blocks and remove them if they are unneeded...
//
for (Function::iterator BBIt = F.begin(); BBIt != F.end(); ) {
if (SimplifyCFG(BBIt++, TTI, BonusInstThreshold, DL, AC)) {
LocalChange = true;
++NumSimpl;
}
}
Changed |= LocalChange;
}
return Changed;
}
static bool simplifyFunctionCFG(Function &F, const TargetTransformInfo &TTI,
const DataLayout *DL, AssumptionCache *AC,
int BonusInstThreshold) {
bool EverChanged = removeUnreachableBlocks(F);
EverChanged |= mergeEmptyReturnBlocks(F);
EverChanged |= iterativelySimplifyCFG(F, TTI, DL, AC, BonusInstThreshold);
// If neither pass changed anything, we're done.
if (!EverChanged) return false;
// iterativelySimplifyCFG can (rarely) make some loops dead. If this happens,
// removeUnreachableBlocks is needed to nuke them, which means we should
// iterate between the two optimizations. We structure the code like this to
// avoid reruning iterativelySimplifyCFG if the second pass of
// removeUnreachableBlocks doesn't do anything.
if (!removeUnreachableBlocks(F))
return true;
do {
EverChanged = iterativelySimplifyCFG(F, TTI, DL, AC, BonusInstThreshold);
EverChanged |= removeUnreachableBlocks(F);
} while (EverChanged);
return true;
}
SimplifyCFGPass::SimplifyCFGPass()
: BonusInstThreshold(UserBonusInstThreshold) {}
SimplifyCFGPass::SimplifyCFGPass(int BonusInstThreshold)
: BonusInstThreshold(BonusInstThreshold) {}
PreservedAnalyses SimplifyCFGPass::run(Function &F,
AnalysisManager<Function> *AM) {
auto *DL = F.getParent()->getDataLayout();
auto &TTI = AM->getResult<TargetIRAnalysis>(F);
auto &AC = AM->getResult<AssumptionAnalysis>(F);
if (!simplifyFunctionCFG(F, TTI, DL, &AC, BonusInstThreshold))
return PreservedAnalyses::none();
return PreservedAnalyses::all();
}
namespace {
struct CFGSimplifyPass : public FunctionPass {
static char ID; // Pass identification, replacement for typeid
unsigned BonusInstThreshold;
CFGSimplifyPass(int T = -1) : FunctionPass(ID) {
BonusInstThreshold = (T == -1) ? UserBonusInstThreshold : unsigned(T);
initializeCFGSimplifyPassPass(*PassRegistry::getPassRegistry());
}
bool runOnFunction(Function &F) override {
if (skipOptnoneFunction(F))
return false;
AssumptionCache *AC =
&getAnalysis<AssumptionCacheTracker>().getAssumptionCache(F);
const TargetTransformInfo &TTI =
getAnalysis<TargetTransformInfoWrapperPass>().getTTI();
DataLayoutPass *DLP = getAnalysisIfAvailable<DataLayoutPass>();
const DataLayout *DL = DLP ? &DLP->getDataLayout() : nullptr;
return simplifyFunctionCFG(F, TTI, DL, AC, BonusInstThreshold);
}
void getAnalysisUsage(AnalysisUsage &AU) const override {
AU.addRequired<AssumptionCacheTracker>();
AU.addRequired<TargetTransformInfoWrapperPass>();
}
};
}
char CFGSimplifyPass::ID = 0;
INITIALIZE_PASS_BEGIN(CFGSimplifyPass, "simplifycfg", "Simplify the CFG", false,
false)
INITIALIZE_PASS_DEPENDENCY(TargetTransformInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(AssumptionCacheTracker)
INITIALIZE_PASS_END(CFGSimplifyPass, "simplifycfg", "Simplify the CFG", false,
false)
// Public interface to the CFGSimplification pass
FunctionPass *llvm::createCFGSimplificationPass(int Threshold) {
return new CFGSimplifyPass(Threshold);
}