1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-24 21:42:54 +02:00
llvm-mirror/lib/Transforms/Scalar/LoopDeletion.cpp
Chandler Carruth 4d1e1851a4 [PM] Port ScalarEvolution to the new pass manager.
This change makes ScalarEvolution a stand-alone object and just produces
one from a pass as needed. Making this work well requires making the
object movable, using references instead of overwritten pointers in
a number of places, and other refactorings.

I've also wired it up to the new pass manager and added a RUN line to
a test to exercise it under the new pass manager. This includes basic
printing support much like with other analyses.

But there is a big and somewhat scary change here. Prior to this patch
ScalarEvolution was never *actually* invalidated!!! Re-running the pass
just re-wired up the various other analyses and didn't remove any of the
existing entries in the SCEV caches or clear out anything at all. This
might seem OK as everything in SCEV that can uses ValueHandles to track
updates to the values that serve as SCEV keys. However, this still means
that as we ran SCEV over each function in the module, we kept
accumulating more and more SCEVs into the cache. At the end, we would
have a SCEV cache with every value that we ever needed a SCEV for in the
entire module!!! Yowzers. The releaseMemory routine would dump all of
this, but that isn't realy called during normal runs of the pipeline as
far as I can see.

To make matters worse, there *is* actually a key that we don't update
with value handles -- there is a map keyed off of Loop*s. Because
LoopInfo *does* release its memory from run to run, it is entirely
possible to run SCEV over one function, then over another function, and
then lookup a Loop* from the second function but find an entry inserted
for the first function! Ouch.

To make matters still worse, there are plenty of updates that *don't*
trip a value handle. It seems incredibly unlikely that today GVN or
another pass that invalidates SCEV can update values in *just* such
a way that a subsequent run of SCEV will incorrectly find lookups in
a cache, but it is theoretically possible and would be a nightmare to
debug.

With this refactoring, I've fixed all this by actually destroying and
recreating the ScalarEvolution object from run to run. Technically, this
could increase the amount of malloc traffic we see, but then again it is
also technically correct. ;] I don't actually think we're suffering from
tons of malloc traffic from SCEV because if we were, the fact that we
never clear the memory would seem more likely to have come up as an
actual problem before now. So, I've made the simple fix here. If in fact
there are serious issues with too much allocation and deallocation,
I can work on a clever fix that preserves the allocations (while
clearing the data) between each run, but I'd prefer to do that kind of
optimization with a test case / benchmark that shows why we need such
cleverness (and that can test that we actually make it faster). It's
possible that this will make some things faster by making the SCEV
caches have higher locality (due to being significantly smaller) so
until there is a clear benchmark, I think the simple change is best.

Differential Revision: http://reviews.llvm.org/D12063

llvm-svn: 245193
2015-08-17 02:08:17 +00:00

254 lines
9.7 KiB
C++

//===- LoopDeletion.cpp - Dead Loop Deletion Pass ---------------===//
//
// 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 Dead Loop Deletion Pass. This pass is responsible
// for eliminating loops with non-infinite computable trip counts that have no
// side effects or volatile instructions, and do not contribute to the
// computation of the function's return value.
//
//===----------------------------------------------------------------------===//
#include "llvm/Transforms/Scalar.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/ADT/Statistic.h"
#include "llvm/Analysis/LoopPass.h"
#include "llvm/Analysis/ScalarEvolution.h"
#include "llvm/IR/Dominators.h"
using namespace llvm;
#define DEBUG_TYPE "loop-delete"
STATISTIC(NumDeleted, "Number of loops deleted");
namespace {
class LoopDeletion : public LoopPass {
public:
static char ID; // Pass ID, replacement for typeid
LoopDeletion() : LoopPass(ID) {
initializeLoopDeletionPass(*PassRegistry::getPassRegistry());
}
// Possibly eliminate loop L if it is dead.
bool runOnLoop(Loop *L, LPPassManager &LPM) override;
void getAnalysisUsage(AnalysisUsage &AU) const override {
AU.addRequired<DominatorTreeWrapperPass>();
AU.addRequired<LoopInfoWrapperPass>();
AU.addRequired<ScalarEvolutionWrapperPass>();
AU.addRequiredID(LoopSimplifyID);
AU.addRequiredID(LCSSAID);
AU.addPreserved<ScalarEvolutionWrapperPass>();
AU.addPreserved<DominatorTreeWrapperPass>();
AU.addPreserved<LoopInfoWrapperPass>();
AU.addPreservedID(LoopSimplifyID);
AU.addPreservedID(LCSSAID);
}
private:
bool isLoopDead(Loop *L, SmallVectorImpl<BasicBlock *> &exitingBlocks,
SmallVectorImpl<BasicBlock *> &exitBlocks,
bool &Changed, BasicBlock *Preheader);
};
}
char LoopDeletion::ID = 0;
INITIALIZE_PASS_BEGIN(LoopDeletion, "loop-deletion",
"Delete dead loops", false, false)
INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
INITIALIZE_PASS_DEPENDENCY(LoopInfoWrapperPass)
INITIALIZE_PASS_DEPENDENCY(ScalarEvolutionWrapperPass)
INITIALIZE_PASS_DEPENDENCY(LoopSimplify)
INITIALIZE_PASS_DEPENDENCY(LCSSA)
INITIALIZE_PASS_END(LoopDeletion, "loop-deletion",
"Delete dead loops", false, false)
Pass *llvm::createLoopDeletionPass() {
return new LoopDeletion();
}
/// isLoopDead - Determined if a loop is dead. This assumes that we've already
/// checked for unique exit and exiting blocks, and that the code is in LCSSA
/// form.
bool LoopDeletion::isLoopDead(Loop *L,
SmallVectorImpl<BasicBlock *> &exitingBlocks,
SmallVectorImpl<BasicBlock *> &exitBlocks,
bool &Changed, BasicBlock *Preheader) {
BasicBlock *exitBlock = exitBlocks[0];
// Make sure that all PHI entries coming from the loop are loop invariant.
// Because the code is in LCSSA form, any values used outside of the loop
// must pass through a PHI in the exit block, meaning that this check is
// sufficient to guarantee that no loop-variant values are used outside
// of the loop.
BasicBlock::iterator BI = exitBlock->begin();
while (PHINode *P = dyn_cast<PHINode>(BI)) {
Value *incoming = P->getIncomingValueForBlock(exitingBlocks[0]);
// Make sure all exiting blocks produce the same incoming value for the exit
// block. If there are different incoming values for different exiting
// blocks, then it is impossible to statically determine which value should
// be used.
for (unsigned i = 1, e = exitingBlocks.size(); i < e; ++i) {
if (incoming != P->getIncomingValueForBlock(exitingBlocks[i]))
return false;
}
if (Instruction *I = dyn_cast<Instruction>(incoming))
if (!L->makeLoopInvariant(I, Changed, Preheader->getTerminator()))
return false;
++BI;
}
// Make sure that no instructions in the block have potential side-effects.
// This includes instructions that could write to memory, and loads that are
// marked volatile. This could be made more aggressive by using aliasing
// information to identify readonly and readnone calls.
for (Loop::block_iterator LI = L->block_begin(), LE = L->block_end();
LI != LE; ++LI) {
for (BasicBlock::iterator BI = (*LI)->begin(), BE = (*LI)->end();
BI != BE; ++BI) {
if (BI->mayHaveSideEffects())
return false;
}
}
return true;
}
/// runOnLoop - Remove dead loops, by which we mean loops that do not impact the
/// observable behavior of the program other than finite running time. Note
/// we do ensure that this never remove a loop that might be infinite, as doing
/// so could change the halting/non-halting nature of a program.
/// NOTE: This entire process relies pretty heavily on LoopSimplify and LCSSA
/// in order to make various safety checks work.
bool LoopDeletion::runOnLoop(Loop *L, LPPassManager &LPM) {
if (skipOptnoneFunction(L))
return false;
// We can only remove the loop if there is a preheader that we can
// branch from after removing it.
BasicBlock *preheader = L->getLoopPreheader();
if (!preheader)
return false;
// If LoopSimplify form is not available, stay out of trouble.
if (!L->hasDedicatedExits())
return false;
// We can't remove loops that contain subloops. If the subloops were dead,
// they would already have been removed in earlier executions of this pass.
if (L->begin() != L->end())
return false;
SmallVector<BasicBlock*, 4> exitingBlocks;
L->getExitingBlocks(exitingBlocks);
SmallVector<BasicBlock*, 4> exitBlocks;
L->getUniqueExitBlocks(exitBlocks);
// We require that the loop only have a single exit block. Otherwise, we'd
// be in the situation of needing to be able to solve statically which exit
// block will be branched to, or trying to preserve the branching logic in
// a loop invariant manner.
if (exitBlocks.size() != 1)
return false;
// Finally, we have to check that the loop really is dead.
bool Changed = false;
if (!isLoopDead(L, exitingBlocks, exitBlocks, Changed, preheader))
return Changed;
// Don't remove loops for which we can't solve the trip count.
// They could be infinite, in which case we'd be changing program behavior.
ScalarEvolution &SE = getAnalysis<ScalarEvolutionWrapperPass>().getSE();
const SCEV *S = SE.getMaxBackedgeTakenCount(L);
if (isa<SCEVCouldNotCompute>(S))
return Changed;
// Now that we know the removal is safe, remove the loop by changing the
// branch from the preheader to go to the single exit block.
BasicBlock *exitBlock = exitBlocks[0];
// Because we're deleting a large chunk of code at once, the sequence in which
// we remove things is very important to avoid invalidation issues. Don't
// mess with this unless you have good reason and know what you're doing.
// Tell ScalarEvolution that the loop is deleted. Do this before
// deleting the loop so that ScalarEvolution can look at the loop
// to determine what it needs to clean up.
SE.forgetLoop(L);
// Connect the preheader directly to the exit block.
TerminatorInst *TI = preheader->getTerminator();
TI->replaceUsesOfWith(L->getHeader(), exitBlock);
// Rewrite phis in the exit block to get their inputs from
// the preheader instead of the exiting block.
BasicBlock *exitingBlock = exitingBlocks[0];
BasicBlock::iterator BI = exitBlock->begin();
while (PHINode *P = dyn_cast<PHINode>(BI)) {
int j = P->getBasicBlockIndex(exitingBlock);
assert(j >= 0 && "Can't find exiting block in exit block's phi node!");
P->setIncomingBlock(j, preheader);
for (unsigned i = 1; i < exitingBlocks.size(); ++i)
P->removeIncomingValue(exitingBlocks[i]);
++BI;
}
// Update the dominator tree and remove the instructions and blocks that will
// be deleted from the reference counting scheme.
DominatorTree &DT = getAnalysis<DominatorTreeWrapperPass>().getDomTree();
SmallVector<DomTreeNode*, 8> ChildNodes;
for (Loop::block_iterator LI = L->block_begin(), LE = L->block_end();
LI != LE; ++LI) {
// Move all of the block's children to be children of the preheader, which
// allows us to remove the domtree entry for the block.
ChildNodes.insert(ChildNodes.begin(), DT[*LI]->begin(), DT[*LI]->end());
for (SmallVectorImpl<DomTreeNode *>::iterator DI = ChildNodes.begin(),
DE = ChildNodes.end(); DI != DE; ++DI) {
DT.changeImmediateDominator(*DI, DT[preheader]);
}
ChildNodes.clear();
DT.eraseNode(*LI);
// Remove the block from the reference counting scheme, so that we can
// delete it freely later.
(*LI)->dropAllReferences();
}
// Erase the instructions and the blocks without having to worry
// about ordering because we already dropped the references.
// NOTE: This iteration is safe because erasing the block does not remove its
// entry from the loop's block list. We do that in the next section.
for (Loop::block_iterator LI = L->block_begin(), LE = L->block_end();
LI != LE; ++LI)
(*LI)->eraseFromParent();
// Finally, the blocks from loopinfo. This has to happen late because
// otherwise our loop iterators won't work.
LoopInfo &loopInfo = getAnalysis<LoopInfoWrapperPass>().getLoopInfo();
SmallPtrSet<BasicBlock*, 8> blocks;
blocks.insert(L->block_begin(), L->block_end());
for (BasicBlock *BB : blocks)
loopInfo.removeBlock(BB);
// The last step is to inform the loop pass manager that we've
// eliminated this loop.
LPM.deleteLoopFromQueue(L);
Changed = true;
++NumDeleted;
return Changed;
}