| // Copyright (c) 2018 Google LLC. |
| // |
| // Licensed under the Apache License, Version 2.0 (the "License"); |
| // you may not use this file except in compliance with the License. |
| // You may obtain a copy of the License at |
| // |
| // http://www.apache.org/licenses/LICENSE-2.0 |
| // |
| // Unless required by applicable law or agreed to in writing, software |
| // distributed under the License is distributed on an "AS IS" BASIS, |
| // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| // See the License for the specific language governing permissions and |
| // limitations under the License. |
| |
| #include "source/opt/loop_unroller.h" |
| |
| #include <limits> |
| #include <map> |
| #include <memory> |
| #include <unordered_map> |
| #include <utility> |
| #include <vector> |
| |
| #include "source/opt/ir_builder.h" |
| #include "source/opt/loop_utils.h" |
| |
| // Implements loop util unrolling functionality for fully and partially |
| // unrolling loops. Given a factor it will duplicate the loop that many times, |
| // appending each one to the end of the old loop and removing backedges, to |
| // create a new unrolled loop. |
| // |
| // 1 - User calls LoopUtils::FullyUnroll or LoopUtils::PartiallyUnroll with a |
| // loop they wish to unroll. LoopUtils::CanPerformUnroll is used to |
| // validate that a given loop can be unrolled. That method (along with the |
| // constructor of loop) checks that the IR is in the expected canonicalised |
| // format. |
| // |
| // 2 - The LoopUtils methods create a LoopUnrollerUtilsImpl object to actually |
| // perform the unrolling. This implements helper methods to copy the loop basic |
| // blocks and remap the ids of instructions used inside them. |
| // |
| // 3 - The core of LoopUnrollerUtilsImpl is the Unroll method, this method |
| // actually performs the loop duplication. It does this by creating a |
| // LoopUnrollState object and then copying the loop as given by the factor |
| // parameter. The LoopUnrollState object retains the state of the unroller |
| // between the loop body copies as each iteration needs information on the last |
| // to adjust the phi induction variable, adjust the OpLoopMerge instruction in |
| // the main loop header, and change the previous continue block to point to the |
| // new header and the new continue block to the main loop header. |
| // |
| // 4 - If the loop is to be fully unrolled then it is simply closed after step |
| // 3, with the OpLoopMerge being deleted, the backedge removed, and the |
| // condition blocks folded. |
| // |
| // 5 - If it is being partially unrolled: if the unrolling factor leaves the |
| // loop with an even number of bodies with respect to the number of loop |
| // iterations then step 3 is all that is needed. If it is uneven then we need to |
| // duplicate the loop completely and unroll the duplicated loop to cover the |
| // residual part and adjust the first loop to cover only the "even" part. For |
| // instance if you request an unroll factor of 3 on a loop with 10 iterations |
| // then copying the body three times would leave you with three bodies in the |
| // loop |
| // where the loop still iterates over each 4 times. So we make two loops one |
| // iterating once then a second loop of three iterating 3 times. |
| |
| namespace spvtools { |
| namespace opt { |
| namespace { |
| |
| // Loop control constant value for DontUnroll flag. |
| static const uint32_t kLoopControlDontUnrollIndex = 2; |
| |
| // Operand index of the loop control parameter of the OpLoopMerge. |
| static const uint32_t kLoopControlIndex = 2; |
| |
| // This utility class encapsulates some of the state we need to maintain between |
| // loop unrolls. Specifically it maintains key blocks and the induction variable |
| // in the current loop duplication step and the blocks from the previous one. |
| // This is because each step of the unroll needs to use data from both the |
| // preceding step and the original loop. |
| struct LoopUnrollState { |
| LoopUnrollState() |
| : previous_phi_(nullptr), |
| previous_latch_block_(nullptr), |
| previous_condition_block_(nullptr), |
| new_phi(nullptr), |
| new_continue_block(nullptr), |
| new_condition_block(nullptr), |
| new_header_block(nullptr) {} |
| |
| // Initialize from the loop descriptor class. |
| LoopUnrollState(Instruction* induction, BasicBlock* latch_block, |
| BasicBlock* condition, std::vector<Instruction*>&& phis) |
| : previous_phi_(induction), |
| previous_latch_block_(latch_block), |
| previous_condition_block_(condition), |
| new_phi(nullptr), |
| new_continue_block(nullptr), |
| new_condition_block(nullptr), |
| new_header_block(nullptr) { |
| previous_phis_ = std::move(phis); |
| } |
| |
| // Swap the state so that the new nodes are now the previous nodes. |
| void NextIterationState() { |
| previous_phi_ = new_phi; |
| previous_latch_block_ = new_latch_block; |
| previous_condition_block_ = new_condition_block; |
| previous_phis_ = std::move(new_phis_); |
| |
| // Clear new nodes. |
| new_phi = nullptr; |
| new_continue_block = nullptr; |
| new_condition_block = nullptr; |
| new_header_block = nullptr; |
| new_latch_block = nullptr; |
| |
| // Clear new block/instruction maps. |
| new_blocks.clear(); |
| new_inst.clear(); |
| ids_to_new_inst.clear(); |
| } |
| |
| // The induction variable from the immediately preceding loop body. |
| Instruction* previous_phi_; |
| |
| // All the phi nodes from the previous loop iteration. |
| std::vector<Instruction*> previous_phis_; |
| |
| std::vector<Instruction*> new_phis_; |
| |
| // The previous latch block. The backedge will be removed from this and |
| // added to the new latch block. |
| BasicBlock* previous_latch_block_; |
| |
| // The previous condition block. This may be folded to flatten the loop. |
| BasicBlock* previous_condition_block_; |
| |
| // The new induction variable. |
| Instruction* new_phi; |
| |
| // The new continue block. |
| BasicBlock* new_continue_block; |
| |
| // The new condition block. |
| BasicBlock* new_condition_block; |
| |
| // The new header block. |
| BasicBlock* new_header_block; |
| |
| // The new latch block. |
| BasicBlock* new_latch_block; |
| |
| // A mapping of new block ids to the original blocks which they were copied |
| // from. |
| std::unordered_map<uint32_t, BasicBlock*> new_blocks; |
| |
| // A mapping of the original instruction ids to the instruction ids to their |
| // copies. |
| std::unordered_map<uint32_t, uint32_t> new_inst; |
| |
| std::unordered_map<uint32_t, Instruction*> ids_to_new_inst; |
| }; |
| |
| // This class implements the actual unrolling. It uses a LoopUnrollState to |
| // maintain the state of the unrolling inbetween steps. |
| class LoopUnrollerUtilsImpl { |
| public: |
| using BasicBlockListTy = std::vector<std::unique_ptr<BasicBlock>>; |
| |
| LoopUnrollerUtilsImpl(IRContext* c, Function* function) |
| : context_(c), |
| function_(*function), |
| loop_condition_block_(nullptr), |
| loop_induction_variable_(nullptr), |
| number_of_loop_iterations_(0), |
| loop_step_value_(0), |
| loop_init_value_(0) {} |
| |
| // Unroll the |loop| by given |factor| by copying the whole body |factor| |
| // times. The resulting basicblock structure will remain a loop. |
| void PartiallyUnroll(Loop*, size_t factor); |
| |
| // If partially unrolling the |loop| would leave the loop with too many bodies |
| // for its number of iterations then this method should be used. This method |
| // will duplicate the |loop| completely, making the duplicated loop the |
| // successor of the original's merge block. The original loop will have its |
| // condition changed to loop over the residual part and the duplicate will be |
| // partially unrolled. The resulting structure will be two loops. |
| void PartiallyUnrollResidualFactor(Loop* loop, size_t factor); |
| |
| // Fully unroll the |loop| by copying the full body by the total number of |
| // loop iterations, folding all conditions, and removing the backedge from the |
| // continue block to the header. |
| void FullyUnroll(Loop* loop); |
| |
| // Get the ID of the variable in the |phi| paired with |label|. |
| uint32_t GetPhiDefID(const Instruction* phi, uint32_t label) const; |
| |
| // Close the loop by removing the OpLoopMerge from the |loop| header block and |
| // making the backedge point to the merge block. |
| void CloseUnrolledLoop(Loop* loop); |
| |
| // Remove the OpConditionalBranch instruction inside |conditional_block| used |
| // to branch to either exit or continue the loop and replace it with an |
| // unconditional OpBranch to block |new_target|. |
| void FoldConditionBlock(BasicBlock* condtion_block, uint32_t new_target); |
| |
| // Add all blocks_to_add_ to function_ at the |insert_point|. |
| void AddBlocksToFunction(const BasicBlock* insert_point); |
| |
| // Duplicates the |old_loop|, cloning each body and remaping the ids without |
| // removing instructions or changing relative structure. Result will be stored |
| // in |new_loop|. |
| void DuplicateLoop(Loop* old_loop, Loop* new_loop); |
| |
| inline size_t GetLoopIterationCount() const { |
| return number_of_loop_iterations_; |
| } |
| |
| // Extracts the initial state information from the |loop|. |
| void Init(Loop* loop); |
| |
| // Replace the uses of each induction variable outside the loop with the final |
| // value of the induction variable before the loop exit. To reflect the proper |
| // state of a fully unrolled loop. |
| void ReplaceInductionUseWithFinalValue(Loop* loop); |
| |
| // Remove all the instructions in the invalidated_instructions_ vector. |
| void RemoveDeadInstructions(); |
| |
| // Replace any use of induction variables outwith the loop with the final |
| // value of the induction variable in the unrolled loop. |
| void ReplaceOutsideLoopUseWithFinalValue(Loop* loop); |
| |
| // Set the LoopControl operand of the OpLoopMerge instruction to be |
| // DontUnroll. |
| void MarkLoopControlAsDontUnroll(Loop* loop) const; |
| |
| private: |
| // Remap all the in |basic_block| to new IDs and keep the mapping of new ids |
| // to old |
| // ids. |loop| is used to identify special loop blocks (header, continue, |
| // ect). |
| void AssignNewResultIds(BasicBlock* basic_block); |
| |
| // Using the map built by AssignNewResultIds, replace the uses in |inst| |
| // by the id that the use maps to. |
| void RemapOperands(Instruction* inst); |
| |
| // Using the map built by AssignNewResultIds, for each instruction in |
| // |basic_block| use |
| // that map to substitute the IDs used by instructions (in the operands) with |
| // the new ids. |
| void RemapOperands(BasicBlock* basic_block); |
| |
| // Copy the whole body of the loop, all blocks dominated by the |loop| header |
| // and not dominated by the |loop| merge. The copied body will be linked to by |
| // the old |loop| continue block and the new body will link to the |loop| |
| // header via the new continue block. |eliminate_conditions| is used to decide |
| // whether or not to fold all the condition blocks other than the last one. |
| void CopyBody(Loop* loop, bool eliminate_conditions); |
| |
| // Copy a given |block_to_copy| in the |loop| and record the mapping of the |
| // old/new ids. |preserve_instructions| determines whether or not the method |
| // will modify (other than result_id) instructions which are copied. |
| void CopyBasicBlock(Loop* loop, const BasicBlock* block_to_copy, |
| bool preserve_instructions); |
| |
| // The actual implementation of the unroll step. Unrolls |loop| by given |
| // |factor| by copying the body by |factor| times. Also propagates the |
| // induction variable value throughout the copies. |
| void Unroll(Loop* loop, size_t factor); |
| |
| // Fills the loop_blocks_inorder_ field with the ordered list of basic blocks |
| // as computed by the method ComputeLoopOrderedBlocks. |
| void ComputeLoopOrderedBlocks(Loop* loop); |
| |
| // Adds the blocks_to_add_ to both the |loop| and to the parent of |loop| if |
| // the parent exists. |
| void AddBlocksToLoop(Loop* loop) const; |
| |
| // After the partially unroll step the phi instructions in the header block |
| // will be in an illegal format. This function makes the phis legal by making |
| // the edge from the latch block come from the new latch block and the value |
| // to be the actual value of the phi at that point. |
| void LinkLastPhisToStart(Loop* loop) const; |
| |
| // A pointer to the IRContext. Used to add/remove instructions and for usedef |
| // chains. |
| IRContext* context_; |
| |
| // A reference the function the loop is within. |
| Function& function_; |
| |
| // A list of basic blocks to be added to the loop at the end of an unroll |
| // step. |
| BasicBlockListTy blocks_to_add_; |
| |
| // List of instructions which are now dead and can be removed. |
| std::vector<Instruction*> invalidated_instructions_; |
| |
| // Maintains the current state of the transform between calls to unroll. |
| LoopUnrollState state_; |
| |
| // An ordered list containing the loop basic blocks. |
| std::vector<BasicBlock*> loop_blocks_inorder_; |
| |
| // The block containing the condition check which contains a conditional |
| // branch to the merge and continue block. |
| BasicBlock* loop_condition_block_; |
| |
| // The induction variable of the loop. |
| Instruction* loop_induction_variable_; |
| |
| // Phis used in the loop need to be remapped to use the actual result values |
| // and then be remapped at the end. |
| std::vector<Instruction*> loop_phi_instructions_; |
| |
| // The number of loop iterations that the loop would preform pre-unroll. |
| size_t number_of_loop_iterations_; |
| |
| // The amount that the loop steps each iteration. |
| int64_t loop_step_value_; |
| |
| // The value the loop starts stepping from. |
| int64_t loop_init_value_; |
| }; |
| |
| /* |
| * Static helper functions. |
| */ |
| |
| // Retrieve the index of the OpPhi instruction |phi| which corresponds to the |
| // incoming |block| id. |
| static uint32_t GetPhiIndexFromLabel(const BasicBlock* block, |
| const Instruction* phi) { |
| for (uint32_t i = 1; i < phi->NumInOperands(); i += 2) { |
| if (block->id() == phi->GetSingleWordInOperand(i)) { |
| return i; |
| } |
| } |
| assert(false && "Could not find operand in instruction."); |
| return 0; |
| } |
| |
| void LoopUnrollerUtilsImpl::Init(Loop* loop) { |
| loop_condition_block_ = loop->FindConditionBlock(); |
| |
| // When we reinit the second loop during PartiallyUnrollResidualFactor we need |
| // to use the cached value from the duplicate step as the dominator tree |
| // basded solution, loop->FindConditionBlock, requires all the nodes to be |
| // connected up with the correct branches. They won't be at this point. |
| if (!loop_condition_block_) { |
| loop_condition_block_ = state_.new_condition_block; |
| } |
| assert(loop_condition_block_); |
| |
| loop_induction_variable_ = loop->FindConditionVariable(loop_condition_block_); |
| assert(loop_induction_variable_); |
| |
| bool found = loop->FindNumberOfIterations( |
| loop_induction_variable_, &*loop_condition_block_->ctail(), |
| &number_of_loop_iterations_, &loop_step_value_, &loop_init_value_); |
| (void)found; // To silence unused variable warning on release builds. |
| assert(found); |
| |
| // Blocks are stored in an unordered set of ids in the loop class, we need to |
| // create the dominator ordered list. |
| ComputeLoopOrderedBlocks(loop); |
| } |
| |
| // This function is used to partially unroll the loop when the factor provided |
| // would normally lead to an illegal optimization. Instead of just unrolling the |
| // loop it creates two loops and unrolls one and adjusts the condition on the |
| // other. The end result being that the new loop pair iterates over the correct |
| // number of bodies. |
| void LoopUnrollerUtilsImpl::PartiallyUnrollResidualFactor(Loop* loop, |
| size_t factor) { |
| // TODO(1841): Handle id overflow. |
| std::unique_ptr<Instruction> new_label{new Instruction( |
| context_, SpvOp::SpvOpLabel, 0, context_->TakeNextId(), {})}; |
| std::unique_ptr<BasicBlock> new_exit_bb{new BasicBlock(std::move(new_label))}; |
| |
| // Save the id of the block before we move it. |
| uint32_t new_merge_id = new_exit_bb->id(); |
| |
| // Add the block the list of blocks to add, we want this merge block to be |
| // right at the start of the new blocks. |
| blocks_to_add_.push_back(std::move(new_exit_bb)); |
| BasicBlock* new_exit_bb_raw = blocks_to_add_[0].get(); |
| Instruction& original_conditional_branch = *loop_condition_block_->tail(); |
| // Duplicate the loop, providing access to the blocks of both loops. |
| // This is a naked new due to the VS2013 requirement of not having unique |
| // pointers in vectors, as it will be inserted into a vector with |
| // loop_descriptor.AddLoop. |
| std::unique_ptr<Loop> new_loop = MakeUnique<Loop>(*loop); |
| |
| // Clear the basic blocks of the new loop. |
| new_loop->ClearBlocks(); |
| |
| DuplicateLoop(loop, new_loop.get()); |
| |
| // Add the blocks to the function. |
| AddBlocksToFunction(loop->GetMergeBlock()); |
| blocks_to_add_.clear(); |
| |
| // Create a new merge block for the first loop. |
| InstructionBuilder builder{context_, new_exit_bb_raw}; |
| // Make the first loop branch to the second. |
| builder.AddBranch(new_loop->GetHeaderBlock()->id()); |
| |
| loop_condition_block_ = state_.new_condition_block; |
| loop_induction_variable_ = state_.new_phi; |
| // Unroll the new loop by the factor with the usual -1 to account for the |
| // existing block iteration. |
| Unroll(new_loop.get(), factor); |
| |
| LinkLastPhisToStart(new_loop.get()); |
| AddBlocksToLoop(new_loop.get()); |
| |
| // Add the new merge block to the back of the list of blocks to be added. It |
| // needs to be the last block added to maintain dominator order in the binary. |
| blocks_to_add_.push_back( |
| std::unique_ptr<BasicBlock>(new_loop->GetMergeBlock())); |
| |
| // Add the blocks to the function. |
| AddBlocksToFunction(loop->GetMergeBlock()); |
| |
| // Reset the usedef analysis. |
| context_->InvalidateAnalysesExceptFor( |
| IRContext::Analysis::kAnalysisLoopAnalysis); |
| analysis::DefUseManager* def_use_manager = context_->get_def_use_mgr(); |
| |
| // The loop condition. |
| Instruction* condition_check = def_use_manager->GetDef( |
| original_conditional_branch.GetSingleWordOperand(0)); |
| |
| // This should have been checked by the LoopUtils::CanPerformUnroll function |
| // before entering this. |
| assert(loop->IsSupportedCondition(condition_check->opcode())); |
| |
| // We need to account for the initial body when calculating the remainder. |
| int64_t remainder = Loop::GetResidualConditionValue( |
| condition_check->opcode(), loop_init_value_, loop_step_value_, |
| number_of_loop_iterations_, factor); |
| |
| assert(remainder > std::numeric_limits<int32_t>::min() && |
| remainder < std::numeric_limits<int32_t>::max()); |
| |
| Instruction* new_constant = nullptr; |
| |
| // If the remainder is negative then we add a signed constant, otherwise just |
| // add an unsigned constant. |
| if (remainder < 0) { |
| new_constant = builder.GetSintConstant(static_cast<int32_t>(remainder)); |
| } else { |
| new_constant = builder.GetUintConstant(static_cast<int32_t>(remainder)); |
| } |
| |
| uint32_t constant_id = new_constant->result_id(); |
| |
| // Update the condition check. |
| condition_check->SetInOperand(1, {constant_id}); |
| |
| // Update the next phi node. The phi will have a constant value coming in from |
| // the preheader block. For the duplicated loop we need to update the constant |
| // to be the amount of iterations covered by the first loop and the incoming |
| // block to be the first loops new merge block. |
| std::vector<Instruction*> new_inductions; |
| new_loop->GetInductionVariables(new_inductions); |
| |
| std::vector<Instruction*> old_inductions; |
| loop->GetInductionVariables(old_inductions); |
| for (size_t index = 0; index < new_inductions.size(); ++index) { |
| Instruction* new_induction = new_inductions[index]; |
| Instruction* old_induction = old_inductions[index]; |
| // Get the index of the loop initalizer, the value coming in from the |
| // preheader. |
| uint32_t initalizer_index = |
| GetPhiIndexFromLabel(new_loop->GetPreHeaderBlock(), old_induction); |
| |
| // Replace the second loop initalizer with the phi from the first |
| new_induction->SetInOperand(initalizer_index - 1, |
| {old_induction->result_id()}); |
| new_induction->SetInOperand(initalizer_index, {new_merge_id}); |
| |
| // If the use of the first loop induction variable is outside of the loop |
| // then replace that use with the second loop induction variable. |
| uint32_t second_loop_induction = new_induction->result_id(); |
| auto replace_use_outside_of_loop = [loop, second_loop_induction]( |
| Instruction* user, |
| uint32_t operand_index) { |
| if (!loop->IsInsideLoop(user)) { |
| user->SetOperand(operand_index, {second_loop_induction}); |
| } |
| }; |
| |
| context_->get_def_use_mgr()->ForEachUse(old_induction, |
| replace_use_outside_of_loop); |
| } |
| |
| context_->InvalidateAnalysesExceptFor( |
| IRContext::Analysis::kAnalysisLoopAnalysis); |
| |
| context_->ReplaceAllUsesWith(loop->GetMergeBlock()->id(), new_merge_id); |
| |
| LoopDescriptor& loop_descriptor = *context_->GetLoopDescriptor(&function_); |
| |
| loop_descriptor.AddLoop(std::move(new_loop), loop->GetParent()); |
| |
| RemoveDeadInstructions(); |
| } |
| |
| // Mark this loop as DontUnroll as it will already be unrolled and it may not |
| // be safe to unroll a previously partially unrolled loop. |
| void LoopUnrollerUtilsImpl::MarkLoopControlAsDontUnroll(Loop* loop) const { |
| Instruction* loop_merge_inst = loop->GetHeaderBlock()->GetLoopMergeInst(); |
| assert(loop_merge_inst && |
| "Loop merge instruction could not be found after entering unroller " |
| "(should have exited before this)"); |
| loop_merge_inst->SetInOperand(kLoopControlIndex, |
| {kLoopControlDontUnrollIndex}); |
| } |
| |
| // Duplicate the |loop| body |factor| - 1 number of times while keeping the loop |
| // backedge intact. This will leave the loop with |factor| number of bodies |
| // after accounting for the initial body. |
| void LoopUnrollerUtilsImpl::Unroll(Loop* loop, size_t factor) { |
| // If we unroll a loop partially it will not be safe to unroll it further. |
| // This is due to the current method of calculating the number of loop |
| // iterations. |
| MarkLoopControlAsDontUnroll(loop); |
| |
| std::vector<Instruction*> inductions; |
| loop->GetInductionVariables(inductions); |
| state_ = LoopUnrollState{loop_induction_variable_, loop->GetLatchBlock(), |
| loop_condition_block_, std::move(inductions)}; |
| for (size_t i = 0; i < factor - 1; ++i) { |
| CopyBody(loop, true); |
| } |
| } |
| |
| void LoopUnrollerUtilsImpl::RemoveDeadInstructions() { |
| // Remove the dead instructions. |
| for (Instruction* inst : invalidated_instructions_) { |
| context_->KillInst(inst); |
| } |
| } |
| |
| void LoopUnrollerUtilsImpl::ReplaceInductionUseWithFinalValue(Loop* loop) { |
| context_->InvalidateAnalysesExceptFor( |
| IRContext::Analysis::kAnalysisLoopAnalysis | |
| IRContext::Analysis::kAnalysisDefUse | |
| IRContext::Analysis::kAnalysisInstrToBlockMapping); |
| |
| std::vector<Instruction*> inductions; |
| loop->GetInductionVariables(inductions); |
| |
| for (size_t index = 0; index < inductions.size(); ++index) { |
| uint32_t trip_step_id = GetPhiDefID(state_.previous_phis_[index], |
| state_.previous_latch_block_->id()); |
| context_->ReplaceAllUsesWith(inductions[index]->result_id(), trip_step_id); |
| invalidated_instructions_.push_back(inductions[index]); |
| } |
| } |
| |
| // Fully unroll the loop by partially unrolling it by the number of loop |
| // iterations minus one for the body already accounted for. |
| void LoopUnrollerUtilsImpl::FullyUnroll(Loop* loop) { |
| // We unroll the loop by number of iterations in the loop. |
| Unroll(loop, number_of_loop_iterations_); |
| |
| // The first condition block is preserved until now so it can be copied. |
| FoldConditionBlock(loop_condition_block_, 1); |
| |
| // Delete the OpLoopMerge and remove the backedge to the header. |
| CloseUnrolledLoop(loop); |
| |
| // Mark the loop for later deletion. This allows us to preserve the loop |
| // iterators but still disregard dead loops. |
| loop->MarkLoopForRemoval(); |
| |
| // If the loop has a parent add the new blocks to the parent. |
| if (loop->GetParent()) { |
| AddBlocksToLoop(loop->GetParent()); |
| } |
| |
| // Add the blocks to the function. |
| AddBlocksToFunction(loop->GetMergeBlock()); |
| |
| ReplaceInductionUseWithFinalValue(loop); |
| |
| RemoveDeadInstructions(); |
| // Invalidate all analyses. |
| context_->InvalidateAnalysesExceptFor( |
| IRContext::Analysis::kAnalysisLoopAnalysis | |
| IRContext::Analysis::kAnalysisDefUse); |
| } |
| |
| // Copy a given basic block, give it a new result_id, and store the new block |
| // and the id mapping in the state. |preserve_instructions| is used to determine |
| // whether or not this function should edit instructions other than the |
| // |result_id|. |
| void LoopUnrollerUtilsImpl::CopyBasicBlock(Loop* loop, const BasicBlock* itr, |
| bool preserve_instructions) { |
| // Clone the block exactly, including the IDs. |
| BasicBlock* basic_block = itr->Clone(context_); |
| basic_block->SetParent(itr->GetParent()); |
| |
| // Assign each result a new unique ID and keep a mapping of the old ids to |
| // the new ones. |
| AssignNewResultIds(basic_block); |
| |
| // If this is the continue block we are copying. |
| if (itr == loop->GetContinueBlock()) { |
| // Make the OpLoopMerge point to this block for the continue. |
| if (!preserve_instructions) { |
| Instruction* merge_inst = loop->GetHeaderBlock()->GetLoopMergeInst(); |
| merge_inst->SetInOperand(1, {basic_block->id()}); |
| context_->UpdateDefUse(merge_inst); |
| } |
| |
| state_.new_continue_block = basic_block; |
| } |
| |
| // If this is the header block we are copying. |
| if (itr == loop->GetHeaderBlock()) { |
| state_.new_header_block = basic_block; |
| |
| if (!preserve_instructions) { |
| // Remove the loop merge instruction if it exists. |
| Instruction* merge_inst = basic_block->GetLoopMergeInst(); |
| if (merge_inst) invalidated_instructions_.push_back(merge_inst); |
| } |
| } |
| |
| // If this is the latch block being copied, record it in the state. |
| if (itr == loop->GetLatchBlock()) state_.new_latch_block = basic_block; |
| |
| // If this is the condition block we are copying. |
| if (itr == loop_condition_block_) { |
| state_.new_condition_block = basic_block; |
| } |
| |
| // Add this block to the list of blocks to add to the function at the end of |
| // the unrolling process. |
| blocks_to_add_.push_back(std::unique_ptr<BasicBlock>(basic_block)); |
| |
| // Keep tracking the old block via a map. |
| state_.new_blocks[itr->id()] = basic_block; |
| } |
| |
| void LoopUnrollerUtilsImpl::CopyBody(Loop* loop, bool eliminate_conditions) { |
| // Copy each basic block in the loop, give them new ids, and save state |
| // information. |
| for (const BasicBlock* itr : loop_blocks_inorder_) { |
| CopyBasicBlock(loop, itr, false); |
| } |
| |
| // Set the previous latch block to point to the new header. |
| Instruction* latch_branch = state_.previous_latch_block_->terminator(); |
| latch_branch->SetInOperand(0, {state_.new_header_block->id()}); |
| context_->UpdateDefUse(latch_branch); |
| |
| // As the algorithm copies the original loop blocks exactly, the tail of the |
| // latch block on iterations after the first one will be a branch to the new |
| // header and not the actual loop header. The last continue block in the loop |
| // should always be a backedge to the global header. |
| Instruction* new_latch_branch = state_.new_latch_block->terminator(); |
| new_latch_branch->SetInOperand(0, {loop->GetHeaderBlock()->id()}); |
| context_->AnalyzeUses(new_latch_branch); |
| |
| std::vector<Instruction*> inductions; |
| loop->GetInductionVariables(inductions); |
| for (size_t index = 0; index < inductions.size(); ++index) { |
| Instruction* master_copy = inductions[index]; |
| |
| assert(master_copy->result_id() != 0); |
| Instruction* induction_clone = |
| state_.ids_to_new_inst[state_.new_inst[master_copy->result_id()]]; |
| |
| state_.new_phis_.push_back(induction_clone); |
| assert(induction_clone->result_id() != 0); |
| |
| if (!state_.previous_phis_.empty()) { |
| state_.new_inst[master_copy->result_id()] = GetPhiDefID( |
| state_.previous_phis_[index], state_.previous_latch_block_->id()); |
| } else { |
| // Do not replace the first phi block ids. |
| state_.new_inst[master_copy->result_id()] = master_copy->result_id(); |
| } |
| } |
| |
| if (eliminate_conditions && |
| state_.new_condition_block != loop_condition_block_) { |
| FoldConditionBlock(state_.new_condition_block, 1); |
| } |
| |
| // Only reference to the header block is the backedge in the latch block, |
| // don't change this. |
| state_.new_inst[loop->GetHeaderBlock()->id()] = loop->GetHeaderBlock()->id(); |
| |
| for (auto& pair : state_.new_blocks) { |
| RemapOperands(pair.second); |
| } |
| |
| for (Instruction* dead_phi : state_.new_phis_) |
| invalidated_instructions_.push_back(dead_phi); |
| |
| // Swap the state so the new is now the previous. |
| state_.NextIterationState(); |
| } |
| |
| uint32_t LoopUnrollerUtilsImpl::GetPhiDefID(const Instruction* phi, |
| uint32_t label) const { |
| for (uint32_t operand = 3; operand < phi->NumOperands(); operand += 2) { |
| if (phi->GetSingleWordOperand(operand) == label) { |
| return phi->GetSingleWordOperand(operand - 1); |
| } |
| } |
| assert(false && "Could not find a phi index matching the provided label"); |
| return 0; |
| } |
| |
| void LoopUnrollerUtilsImpl::FoldConditionBlock(BasicBlock* condition_block, |
| uint32_t operand_label) { |
| // Remove the old conditional branch to the merge and continue blocks. |
| Instruction& old_branch = *condition_block->tail(); |
| uint32_t new_target = old_branch.GetSingleWordOperand(operand_label); |
| |
| context_->KillInst(&old_branch); |
| // Add the new unconditional branch to the merge block. |
| InstructionBuilder builder( |
| context_, condition_block, |
| IRContext::Analysis::kAnalysisDefUse | |
| IRContext::Analysis::kAnalysisInstrToBlockMapping); |
| builder.AddBranch(new_target); |
| } |
| |
| void LoopUnrollerUtilsImpl::CloseUnrolledLoop(Loop* loop) { |
| // Remove the OpLoopMerge instruction from the function. |
| Instruction* merge_inst = loop->GetHeaderBlock()->GetLoopMergeInst(); |
| invalidated_instructions_.push_back(merge_inst); |
| |
| // Remove the final backedge to the header and make it point instead to the |
| // merge block. |
| Instruction* latch_instruction = state_.previous_latch_block_->terminator(); |
| latch_instruction->SetInOperand(0, {loop->GetMergeBlock()->id()}); |
| context_->UpdateDefUse(latch_instruction); |
| |
| // Remove all induction variables as the phis will now be invalid. Replace all |
| // uses with the constant initializer value (all uses of phis will be in |
| // the first iteration with the subsequent phis already having been removed). |
| std::vector<Instruction*> inductions; |
| loop->GetInductionVariables(inductions); |
| |
| // We can use the state instruction mechanism to replace all internal loop |
| // values within the first loop trip (as the subsequent ones will be updated |
| // by the copy function) with the value coming in from the preheader and then |
| // use context ReplaceAllUsesWith for the uses outside the loop with the final |
| // trip phi value. |
| state_.new_inst.clear(); |
| for (Instruction* induction : inductions) { |
| uint32_t initalizer_id = |
| GetPhiDefID(induction, loop->GetPreHeaderBlock()->id()); |
| |
| state_.new_inst[induction->result_id()] = initalizer_id; |
| } |
| |
| for (BasicBlock* block : loop_blocks_inorder_) { |
| RemapOperands(block); |
| } |
| |
| // Rewrite the last phis, since they may still reference the original phi. |
| for (Instruction* last_phi : state_.previous_phis_) { |
| RemapOperands(last_phi); |
| } |
| } |
| |
| // Uses the first loop to create a copy of the loop with new IDs. |
| void LoopUnrollerUtilsImpl::DuplicateLoop(Loop* old_loop, Loop* new_loop) { |
| std::vector<BasicBlock*> new_block_order; |
| |
| // Copy every block in the old loop. |
| for (const BasicBlock* itr : loop_blocks_inorder_) { |
| CopyBasicBlock(old_loop, itr, true); |
| new_block_order.push_back(blocks_to_add_.back().get()); |
| } |
| |
| // Clone the merge block, give it a new id and record it in the state. |
| BasicBlock* new_merge = old_loop->GetMergeBlock()->Clone(context_); |
| new_merge->SetParent(old_loop->GetMergeBlock()->GetParent()); |
| AssignNewResultIds(new_merge); |
| state_.new_blocks[old_loop->GetMergeBlock()->id()] = new_merge; |
| |
| // Remap the operands of every instruction in the loop to point to the new |
| // copies. |
| for (auto& pair : state_.new_blocks) { |
| RemapOperands(pair.second); |
| } |
| |
| loop_blocks_inorder_ = std::move(new_block_order); |
| |
| AddBlocksToLoop(new_loop); |
| |
| new_loop->SetHeaderBlock(state_.new_header_block); |
| new_loop->SetContinueBlock(state_.new_continue_block); |
| new_loop->SetLatchBlock(state_.new_latch_block); |
| new_loop->SetMergeBlock(new_merge); |
| } |
| |
| // Whenever the utility copies a block it stores it in a tempory buffer, this |
| // function adds the buffer into the Function. The blocks will be inserted |
| // after the block |insert_point|. |
| void LoopUnrollerUtilsImpl::AddBlocksToFunction( |
| const BasicBlock* insert_point) { |
| for (auto basic_block_iterator = function_.begin(); |
| basic_block_iterator != function_.end(); ++basic_block_iterator) { |
| if (basic_block_iterator->id() == insert_point->id()) { |
| basic_block_iterator.InsertBefore(&blocks_to_add_); |
| return; |
| } |
| } |
| |
| assert( |
| false && |
| "Could not add basic blocks to function as insert point was not found."); |
| } |
| |
| // Assign all result_ids in |basic_block| instructions to new IDs and preserve |
| // the mapping of new ids to old ones. |
| void LoopUnrollerUtilsImpl::AssignNewResultIds(BasicBlock* basic_block) { |
| analysis::DefUseManager* def_use_mgr = context_->get_def_use_mgr(); |
| |
| // Label instructions aren't covered by normal traversal of the |
| // instructions. |
| // TODO(1841): Handle id overflow. |
| uint32_t new_label_id = context_->TakeNextId(); |
| |
| // Assign a new id to the label. |
| state_.new_inst[basic_block->GetLabelInst()->result_id()] = new_label_id; |
| basic_block->GetLabelInst()->SetResultId(new_label_id); |
| def_use_mgr->AnalyzeInstDefUse(basic_block->GetLabelInst()); |
| |
| for (Instruction& inst : *basic_block) { |
| uint32_t old_id = inst.result_id(); |
| |
| // Ignore stores etc. |
| if (old_id == 0) { |
| continue; |
| } |
| |
| // Give the instruction a new id. |
| // TODO(1841): Handle id overflow. |
| inst.SetResultId(context_->TakeNextId()); |
| def_use_mgr->AnalyzeInstDef(&inst); |
| |
| // Save the mapping of old_id -> new_id. |
| state_.new_inst[old_id] = inst.result_id(); |
| // Check if this instruction is the induction variable. |
| if (loop_induction_variable_->result_id() == old_id) { |
| // Save a pointer to the new copy of it. |
| state_.new_phi = &inst; |
| } |
| state_.ids_to_new_inst[inst.result_id()] = &inst; |
| } |
| } |
| |
| void LoopUnrollerUtilsImpl::RemapOperands(Instruction* inst) { |
| auto remap_operands_to_new_ids = [this](uint32_t* id) { |
| auto itr = state_.new_inst.find(*id); |
| |
| if (itr != state_.new_inst.end()) { |
| *id = itr->second; |
| } |
| }; |
| |
| inst->ForEachInId(remap_operands_to_new_ids); |
| context_->AnalyzeUses(inst); |
| } |
| |
| void LoopUnrollerUtilsImpl::RemapOperands(BasicBlock* basic_block) { |
| for (Instruction& inst : *basic_block) { |
| RemapOperands(&inst); |
| } |
| } |
| |
| // Generate the ordered list of basic blocks in the |loop| and cache it for |
| // later use. |
| void LoopUnrollerUtilsImpl::ComputeLoopOrderedBlocks(Loop* loop) { |
| loop_blocks_inorder_.clear(); |
| loop->ComputeLoopStructuredOrder(&loop_blocks_inorder_); |
| } |
| |
| // Adds the blocks_to_add_ to both the loop and to the parent. |
| void LoopUnrollerUtilsImpl::AddBlocksToLoop(Loop* loop) const { |
| // Add the blocks to this loop. |
| for (auto& block_itr : blocks_to_add_) { |
| loop->AddBasicBlock(block_itr.get()); |
| } |
| |
| // Add the blocks to the parent as well. |
| if (loop->GetParent()) AddBlocksToLoop(loop->GetParent()); |
| } |
| |
| void LoopUnrollerUtilsImpl::LinkLastPhisToStart(Loop* loop) const { |
| std::vector<Instruction*> inductions; |
| loop->GetInductionVariables(inductions); |
| |
| for (size_t i = 0; i < inductions.size(); ++i) { |
| Instruction* last_phi_in_block = state_.previous_phis_[i]; |
| |
| uint32_t phi_index = |
| GetPhiIndexFromLabel(state_.previous_latch_block_, last_phi_in_block); |
| uint32_t phi_variable = |
| last_phi_in_block->GetSingleWordInOperand(phi_index - 1); |
| uint32_t phi_label = last_phi_in_block->GetSingleWordInOperand(phi_index); |
| |
| Instruction* phi = inductions[i]; |
| phi->SetInOperand(phi_index - 1, {phi_variable}); |
| phi->SetInOperand(phi_index, {phi_label}); |
| } |
| } |
| |
| // Duplicate the |loop| body |factor| number of times while keeping the loop |
| // backedge intact. |
| void LoopUnrollerUtilsImpl::PartiallyUnroll(Loop* loop, size_t factor) { |
| Unroll(loop, factor); |
| LinkLastPhisToStart(loop); |
| AddBlocksToLoop(loop); |
| AddBlocksToFunction(loop->GetMergeBlock()); |
| RemoveDeadInstructions(); |
| } |
| |
| /* |
| * End LoopUtilsImpl. |
| */ |
| |
| } // namespace |
| |
| /* |
| * |
| * Begin Utils. |
| * |
| * */ |
| |
| bool LoopUtils::CanPerformUnroll() { |
| // The loop is expected to be in structured order. |
| if (!loop_->GetHeaderBlock()->GetMergeInst()) { |
| return false; |
| } |
| |
| // Find check the loop has a condition we can find and evaluate. |
| const BasicBlock* condition = loop_->FindConditionBlock(); |
| if (!condition) return false; |
| |
| // Check that we can find and process the induction variable. |
| const Instruction* induction = loop_->FindConditionVariable(condition); |
| if (!induction || induction->opcode() != SpvOpPhi) return false; |
| |
| // Check that we can find the number of loop iterations. |
| if (!loop_->FindNumberOfIterations(induction, &*condition->ctail(), nullptr)) |
| return false; |
| |
| // Make sure the latch block is a unconditional branch to the header |
| // block. |
| const Instruction& branch = *loop_->GetLatchBlock()->ctail(); |
| bool branching_assumption = |
| branch.opcode() == SpvOpBranch && |
| branch.GetSingleWordInOperand(0) == loop_->GetHeaderBlock()->id(); |
| if (!branching_assumption) { |
| return false; |
| } |
| |
| std::vector<Instruction*> inductions; |
| loop_->GetInductionVariables(inductions); |
| |
| // Ban breaks within the loop. |
| const std::vector<uint32_t>& merge_block_preds = |
| context_->cfg()->preds(loop_->GetMergeBlock()->id()); |
| if (merge_block_preds.size() != 1) { |
| return false; |
| } |
| |
| // Ban continues within the loop. |
| const std::vector<uint32_t>& continue_block_preds = |
| context_->cfg()->preds(loop_->GetContinueBlock()->id()); |
| if (continue_block_preds.size() != 1) { |
| return false; |
| } |
| |
| // Ban returns in the loop. |
| // Iterate over all the blocks within the loop and check that none of them |
| // exit the loop. |
| for (uint32_t label_id : loop_->GetBlocks()) { |
| const BasicBlock* block = context_->cfg()->block(label_id); |
| if (block->ctail()->opcode() == SpvOp::SpvOpKill || |
| block->ctail()->opcode() == SpvOp::SpvOpReturn || |
| block->ctail()->opcode() == SpvOp::SpvOpReturnValue) { |
| return false; |
| } |
| } |
| // Can only unroll inner loops. |
| if (!loop_->AreAllChildrenMarkedForRemoval()) { |
| return false; |
| } |
| |
| return true; |
| } |
| |
| bool LoopUtils::PartiallyUnroll(size_t factor) { |
| if (factor == 1 || !CanPerformUnroll()) return false; |
| |
| // Create the unroller utility. |
| LoopUnrollerUtilsImpl unroller{context_, |
| loop_->GetHeaderBlock()->GetParent()}; |
| unroller.Init(loop_); |
| |
| // If the unrolling factor is larger than or the same size as the loop just |
| // fully unroll the loop. |
| if (factor >= unroller.GetLoopIterationCount()) { |
| unroller.FullyUnroll(loop_); |
| return true; |
| } |
| |
| // If the loop unrolling factor is an residual number of iterations we need to |
| // let run the loop for the residual part then let it branch into the unrolled |
| // remaining part. We add one when calucating the remainder to take into |
| // account the one iteration already in the loop. |
| if (unroller.GetLoopIterationCount() % factor != 0) { |
| unroller.PartiallyUnrollResidualFactor(loop_, factor); |
| } else { |
| unroller.PartiallyUnroll(loop_, factor); |
| } |
| |
| return true; |
| } |
| |
| bool LoopUtils::FullyUnroll() { |
| if (!CanPerformUnroll()) return false; |
| |
| std::vector<Instruction*> inductions; |
| loop_->GetInductionVariables(inductions); |
| |
| LoopUnrollerUtilsImpl unroller{context_, |
| loop_->GetHeaderBlock()->GetParent()}; |
| |
| unroller.Init(loop_); |
| unroller.FullyUnroll(loop_); |
| |
| return true; |
| } |
| |
| void LoopUtils::Finalize() { |
| // Clean up the loop descriptor to preserve the analysis. |
| |
| LoopDescriptor* LD = context_->GetLoopDescriptor(&function_); |
| LD->PostModificationCleanup(); |
| } |
| |
| /* |
| * |
| * Begin Pass. |
| * |
| */ |
| |
| Pass::Status LoopUnroller::Process() { |
| bool changed = false; |
| for (Function& f : *context()->module()) { |
| LoopDescriptor* LD = context()->GetLoopDescriptor(&f); |
| for (Loop& loop : *LD) { |
| LoopUtils loop_utils{context(), &loop}; |
| if (!loop.HasUnrollLoopControl() || !loop_utils.CanPerformUnroll()) { |
| continue; |
| } |
| |
| if (fully_unroll_) { |
| loop_utils.FullyUnroll(); |
| } else { |
| loop_utils.PartiallyUnroll(unroll_factor_); |
| } |
| changed = true; |
| } |
| LD->PostModificationCleanup(); |
| } |
| |
| return changed ? Status::SuccessWithChange : Status::SuccessWithoutChange; |
| } |
| |
| } // namespace opt |
| } // namespace spvtools |