mlir/lib/Dialect/GPU/Transforms/EliminateBarriers.cpp

//===- EliminateBarriers.cpp - Eliminate extra barriers --===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// Barrier elimination pattern and pass. If a barrier does not enforce any
// conflicting pair of memory effects, including a pair that is enforced by
// another barrier, it is unnecessary and can be removed. Adapted from
// "High-Performance GPU-to-CPU Transpilation and Optimization via High-Level
// Parallel Constructs" by Moses, Ivanov, Domke, Endo, Doerfert, and Zinenko in
// PPoPP 2023 and implementation in Polygeist.
//
//===----------------------------------------------------------------------===//

#include "mlir/Dialect/Func/IR/FuncOps.h"
#include "mlir/Dialect/GPU/IR/GPUDialect.h"
#include "mlir/Dialect/GPU/Transforms/Passes.h"
#include "mlir/Dialect/MemRef/IR/MemRef.h"
#include "mlir/Dialect/SCF/IR/SCF.h"
#include "mlir/Dialect/Vector/IR/VectorOps.h"
#include "mlir/IR/Operation.h"
#include "mlir/Transforms/GreedyPatternRewriteDriver.h"
#include "llvm/ADT/SetOperations.h"
#include "llvm/ADT/TypeSwitch.h"
#include "llvm/Support/Debug.h"
#include "llvm/Support/DebugLog.h"

namespace mlir {
#define GEN_PASS_DEF_GPUELIMINATEBARRIERS
#include "mlir/Dialect/GPU/Transforms/Passes.h.inc"
} // namespace mlir

using namespace mlir;
using namespace mlir::gpu;

#define DEBUG_TYPE "gpu-erase-barriers"
#define DEBUG_TYPE_ALIAS "gpu-erase-barries-alias"

// The functions below provide interface-like verification, but are too specific
// to barrier elimination to become interfaces.

/// Returns `true` if the op is defines the parallel region that is subject to
/// barrier synchronization.
static bool isParallelRegionBoundary(Operation *op) {
  if (op->hasAttr("__parallel_region_boundary_for_test"))
    return true;

  return isa<GPUFuncOp, LaunchOp>(op);
}

/// Returns `true` if the op behaves like a sequential loop, e.g., the control
/// flow "wraps around" from the end of the body region back to its start.
static bool isSequentialLoopLike(Operation *op) { return isa<scf::ForOp>(op); }

/// Returns `true` if the regions of the op are guaranteed to be executed at
/// most once. Thus, if an operation in one of the nested regions of `op` is
/// executed than so are all the other operations in this region.
static bool hasSingleExecutionBody(Operation *op) {
  return isa<FunctionOpInterface, scf::IfOp, memref::AllocaScopeOp>(op);
}

/// Returns `true` if the operation is known to produce a pointer-like object
/// distinct from any other object produced by a similar operation. For example,
/// an allocation produces such an object.
static bool producesDistinctBase(Operation *op) {
  return isa_and_nonnull<memref::AllocOp, memref::AllocaOp>(op);
}

/// Populates `effects` with all memory effects without associating them to a
/// specific value.
static void addAllValuelessEffects(
    SmallVectorImpl<MemoryEffects::EffectInstance> &effects) {
  effects.emplace_back(MemoryEffects::Effect::get<MemoryEffects::Read>());
  effects.emplace_back(MemoryEffects::Effect::get<MemoryEffects::Write>());
  effects.emplace_back(MemoryEffects::Effect::get<MemoryEffects::Allocate>());
  effects.emplace_back(MemoryEffects::Effect::get<MemoryEffects::Free>());
}

/// Looks through known "view-like" ops to find the base memref.
static Value getBase(Value v) {
  while (Operation *definingOp = v.getDefiningOp()) {
    if (auto viewLike = dyn_cast<ViewLikeOpInterface>(definingOp)) {
      v = viewLike.getViewSource();
      continue;
    }
    if (auto transposeOp = dyn_cast<memref::TransposeOp>(definingOp)) {
      v = transposeOp.getIn();
      continue;
    }
    break;
  }
  return v;
}

/// Returns `true` if accesses to the given memory space could potentially be
/// fenced by a barrier synchronizing on the given `fencedAddressSpaces`. If
/// the set of address spaces is not given, it is equal to all possible address
/// spaces. Memory spaces that are not `#gpu.address_space` are deemed to
/// overlap with all GPU address spaces.
static bool isAddressSpacePotentiallyFenced(
    Attribute memorySpace,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces) {
  if (!fencedAddressSpaces)
    return true;

  auto gpuMemSpace = dyn_cast_if_present<gpu::AddressSpaceAttr>(memorySpace);
  if (!gpuMemSpace)
    return true;

  // Check if this GPU address space is in the fenced set.
  return llvm::is_contained(*fencedAddressSpaces, gpuMemSpace);
}

/// Succeeds if the effect operates on a memref whose memory  space
/// could be one of the given fenced address spaces. This will both look at the
/// address space of the effect's operand and of the view-like operations that
/// define that memref, so as to inspect any memory-space casts or similar
/// operations (like amdgpu buffer casts) that may provide more information.
/// This assumes that directly-conflicting casts (that is, for example, casting
/// a memref in global memory to make it one in workspace memory) can't happen.
static LogicalResult effectMightAffectAddressSpaces(
    const MemoryEffects::EffectInstance &effect,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces) {
  if (!fencedAddressSpaces)
    return success();

  Value value = effect.getValue();
  if (!value)
    return success();

  auto mightMatch = [&](Value v) {
    auto memrefType = dyn_cast<BaseMemRefType>(v.getType());
    if (!memrefType)
      return true;
    return isAddressSpacePotentiallyFenced(memrefType.getMemorySpace(),
                                           fencedAddressSpaces);
  };

  if (!mightMatch(value))
    return failure();

  Value base = value;
  while (auto viewLike = base.getDefiningOp<ViewLikeOpInterface>()) {
    base = viewLike.getViewSource();
    // We assume that we won't see directly incompatible casts, like global =>
    // flat/null => workspace.
    if (!mightMatch(base))
      return failure();
  }

  return success();
}

/// Returns `true` if `op` is a `BarrierOp` that fences any address spaces that
/// could overlap with the given fenced address spaces.
static bool isBarrierWithCommonFencedMemory(
    Operation *op,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces) {
  auto barrier = dyn_cast<BarrierOp>(op);
  if (!barrier)
    return false;

  std::optional<ArrayAttr> otherFencedSpaces = barrier.getAddressSpaces();
  // Barriers with unspecified fencing fence everything.
  if (!otherFencedSpaces)
    return true;
  // While barriers that fence nothing can't close off our search.
  if (otherFencedSpaces->empty())
    return false;

  // If we fence all memory, we've got fencing in common with anything but the
  // non-merory barrier.
  if (!fencedAddressSpaces)
    return true;

  return llvm::any_of(
      otherFencedSpaces->getAsRange<gpu::AddressSpaceAttr>(),
      [&](auto a) { return llvm::is_contained(*fencedAddressSpaces, a); });
}

/// Collect the memory effects of the given op in 'effects'. Returns 'true' if
/// it could extract the effect information from the op, otherwise returns
/// 'false' and conservatively populates the list with all possible effects
/// associated with no particular value or symbol. `fencedAddressSpaces` is,
/// if given, the set of GPU memory spaces that are being synchronized by the
/// barrier being syrchronized - memory operations where the value being
/// impacted is known and either it or its base value have an address space that
/// is known to be distinct from the ones being synchronized on will not be
/// included in the effect set.
static bool collectEffects(
    Operation *op, SmallVectorImpl<MemoryEffects::EffectInstance> &effects,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces,
    bool ignoreBarriers = true) {
  // Skip over barriers to avoid infinite recursion (those barriers would ask
  // this barrier again).
  if (ignoreBarriers && isa<BarrierOp>(op))
    return true;

  // Collect effect instances the operation. Note that the implementation of
  // getEffects erases all effect instances that have the type other than the
  // template parameter so we collect them first in a local buffer and then
  // copy.
  if (auto iface = dyn_cast<MemoryEffectOpInterface>(op)) {
    SmallVector<MemoryEffects::EffectInstance> localEffects;
    iface.getEffects(localEffects);
    // Filter out effects that cannot affect the fenced address spaces.
    for (const MemoryEffects::EffectInstance &effect : localEffects) {
      if (succeeded(
              effectMightAffectAddressSpaces(effect, fencedAddressSpaces)))
        effects.push_back(effect);
    }
    return true;
  }
  if (op->hasTrait<OpTrait::HasRecursiveMemoryEffects>()) {
    for (auto &region : op->getRegions()) {
      for (auto &block : region) {
        for (auto &innerOp : block)
          if (!collectEffects(&innerOp, effects, fencedAddressSpaces,
                              ignoreBarriers))
            return false;
      }
    }
    return true;
  }

  // We need to be conservative here in case the op doesn't have the interface
  // and assume it can have any possible effect.
  addAllValuelessEffects(effects);
  return false;
}

/// Get all effects before the given operation caused by other operations in the
/// same block. That is, this will not consider operations beyond the block.
static bool getEffectsBeforeInBlock(
    Operation *op, SmallVectorImpl<MemoryEffects::EffectInstance> &effects,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces,
    bool stopAtBarrier) {
  if (op == &op->getBlock()->front())
    return true;

  for (Operation *it = op->getPrevNode(); it != nullptr;
       it = it->getPrevNode()) {
    if (isBarrierWithCommonFencedMemory(it, fencedAddressSpaces)) {
      if (stopAtBarrier)
        return true;
      continue;
    }

    if (!collectEffects(it, effects, fencedAddressSpaces))
      return false;
  }
  return true;
}

/// Collects memory effects from operations that may be executed before `op` in
/// a trivial structured control flow, e.g., without branches. Stops at the
/// parallel region boundary or at the barrier operation if `stopAtBarrier` is
/// set. Returns `true` if the memory effects added to `effects` are exact,
/// `false` if they are a conservative over-approximation. The latter means that
/// `effects` contain instances not associated with a specific value.
static bool getEffectsBefore(
    Operation *op, SmallVectorImpl<MemoryEffects::EffectInstance> &effects,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces,
    bool stopAtBarrier) {
  if (!op->getBlock())
    return true;

  // If there is a non-structured control flow, bail.
  Region *region = op->getBlock()->getParent();
  if (region && !region->hasOneBlock()) {
    addAllValuelessEffects(effects);
    return false;
  }

  // Collect all effects before the op.
  getEffectsBeforeInBlock(op, effects, fencedAddressSpaces, stopAtBarrier);

  // Stop if reached the parallel region boundary.
  if (isParallelRegionBoundary(op->getParentOp()))
    return true;

  Operation *parent = op->getParentOp();
  // Otherwise, keep collecting above the parent operation.
  if (!parent->hasTrait<OpTrait::IsIsolatedFromAbove>() &&
      !getEffectsBefore(parent, effects, fencedAddressSpaces, stopAtBarrier))
    return false;

  // If the op is loop-like, collect effects from the trailing operations until
  // we hit a barrier because they can executed before the current operation by
  // the previous iteration of this loop. For example, in the following loop
  //
  //   for i = ... {
  //     op1
  //     ...
  //     barrier
  //     op2
  //   }
  //
  // the operation `op2` at iteration `i` is known to be executed before the
  // operation `op1` at iteration `i+1` and the side effects must be ordered
  // appropriately.
  if (isSequentialLoopLike(parent)) {
    // Assuming loop terminators have no side effects.
    return getEffectsBeforeInBlock(op->getBlock()->getTerminator(), effects,
                                   fencedAddressSpaces, /*stopAtBarrier=*/true);
  }

  // If the parent operation is not guaranteed to execute its (single-block)
  // region once, walk the block.
  bool conservative = false;
  if (!hasSingleExecutionBody(op->getParentOp()))
    op->getParentOp()->walk([&](Operation *in) {
      if (conservative)
        return WalkResult::interrupt();
      if (!collectEffects(in, effects, fencedAddressSpaces)) {
        conservative = true;
        return WalkResult::interrupt();
      }
      return WalkResult::advance();
    });

  return !conservative;
}

/// Get all effects after the given operation caused by other operations in the
/// same block. That is, this will not consider operations beyond the block.
static bool getEffectsAfterInBlock(
    Operation *op, SmallVectorImpl<MemoryEffects::EffectInstance> &effects,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces,
    bool stopAtBarrier) {
  if (op == &op->getBlock()->back())
    return true;

  for (Operation *it = op->getNextNode(); it != nullptr;
       it = it->getNextNode()) {
    if (isBarrierWithCommonFencedMemory(it, fencedAddressSpaces)) {
      if (stopAtBarrier)
        return true;
      continue;
    }
    if (!collectEffects(it, effects, fencedAddressSpaces))
      return false;
  }
  return true;
}

/// Collects memory effects from operations that may be executed after `op` in
/// a trivial structured control flow, e.g., without branches. Stops at the
/// parallel region boundary or at the barrier operation if `stopAtBarrier` is
/// set. Returns `true` if the memory effects added to `effects` are exact,
/// `false` if they are a conservative over-approximation. The latter means that
/// `effects` contain instances not associated with a specific value.
static bool getEffectsAfter(
    Operation *op, SmallVectorImpl<MemoryEffects::EffectInstance> &effects,
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedAddressSpaces,
    bool stopAtBarrier) {
  if (!op->getBlock())
    return true;

  // If there is a non-structured control flow, bail.
  Region *region = op->getBlock()->getParent();
  if (region && !region->hasOneBlock()) {
    addAllValuelessEffects(effects);
    return false;
  }

  // Collect all effects after the op.
  getEffectsAfterInBlock(op, effects, fencedAddressSpaces, stopAtBarrier);

  Operation *parent = op->getParentOp();
  // Stop if reached the parallel region boundary.
  if (isParallelRegionBoundary(parent))
    return true;

  // Otherwise, keep collecting below the parent operation.
  // Don't look into, for example, neighboring functions
  if (!parent->hasTrait<OpTrait::IsIsolatedFromAbove>() &&
      !getEffectsAfter(parent, effects, fencedAddressSpaces, stopAtBarrier))
    return false;

  // If the op is loop-like, collect effects from the leading operations until
  // we hit a barrier because they can executed after the current operation by
  // the next iteration of this loop. For example, in the following loop
  //
  //   for i = ... {
  //     op1
  //     ...
  //     barrier
  //     op2
  //   }
  //
  // the operation `op1` at iteration `i` is known to be executed after the
  // operation `op2` at iteration `i-1` and the side effects must be ordered
  // appropriately.
  if (isSequentialLoopLike(parent)) {
    if (isBarrierWithCommonFencedMemory(&op->getBlock()->front(),
                                        fencedAddressSpaces))
      return true;

    bool exact =
        collectEffects(&op->getBlock()->front(), effects, fencedAddressSpaces);
    return getEffectsAfterInBlock(&op->getBlock()->front(), effects,
                                  fencedAddressSpaces,
                                  /*stopAtBarrier=*/true) &&
           exact;
  }

  // If the parent operation is not guaranteed to execute its (single-block)
  // region once, walk the block.
  bool conservative = false;
  if (!hasSingleExecutionBody(op->getParentOp()))
    op->getParentOp()->walk([&](Operation *in) {
      if (conservative)
        return WalkResult::interrupt();
      if (!collectEffects(in, effects, fencedAddressSpaces)) {
        conservative = true;
        return WalkResult::interrupt();
      }
      return WalkResult::advance();
    });

  return !conservative;
}

/// Returns `true` if the value is defined as a function argument.
static bool isFunctionArgument(Value v) {
  auto arg = dyn_cast<BlockArgument>(v);
  return arg && isa<FunctionOpInterface>(arg.getOwner()->getParentOp());
}

/// Returns the operand that the operation "propagates" through it for capture
/// purposes. That is, if the value produced by this operation is captured, then
/// so is the returned value.
static Value propagatesCapture(Operation *op) {
  return llvm::TypeSwitch<Operation *, Value>(op)
      .Case(
          [](ViewLikeOpInterface viewLike) { return viewLike.getViewSource(); })
      .Case([](CastOpInterface castLike) { return castLike->getOperand(0); })
      .Case([](memref::TransposeOp transpose) { return transpose.getIn(); })
      .Default(nullptr);
}

/// Returns `true` if the given operation is known to capture the given value,
/// `false` if it is known not to capture the given value, `nullopt` if neither
/// is known.
static std::optional<bool> getKnownCapturingStatus(Operation *op, Value v) {
  return llvm::TypeSwitch<Operation *, std::optional<bool>>(op)
      // Store-like operations don't capture the destination, but do capture
      // the value.
      .Case<memref::StoreOp, vector::TransferWriteOp>(
          [&](auto op) { return op.getValue() == v; })
      .Case<vector::StoreOp, vector::MaskedStoreOp>(
          [&](auto op) { return op.getValueToStore() == v; })
      // These operations are known not to capture.
      .Case([](memref::DeallocOp) { return false; })
      // By default, we don't know anything.
      .Default(std::nullopt);
}

/// Returns `true` if the value may be captured by any of its users, i.e., if
/// the user may be storing this value into memory. This makes aliasing analysis
/// more conservative as it cannot assume the pointer-like value is only passed
/// around through SSA use-def.
static bool maybeCaptured(Value v) {
  SmallVector<Value> todo = {v};
  while (!todo.empty()) {
    Value v = todo.pop_back_val();
    for (Operation *user : v.getUsers()) {
      // A user that is known to only read cannot capture.
      auto iface = dyn_cast<MemoryEffectOpInterface>(user);
      if (iface) {
        SmallVector<MemoryEffects::EffectInstance> effects;
        iface.getEffects(effects);
        if (llvm::all_of(effects,
                         [](const MemoryEffects::EffectInstance &effect) {
                           return isa<MemoryEffects::Read>(effect.getEffect());
                         })) {
          continue;
        }
      }

      // When an operation is known to create an alias, consider if the
      // source is captured as well.
      if (Value v = propagatesCapture(user)) {
        todo.push_back(v);
        continue;
      }

      std::optional<bool> knownCaptureStatus = getKnownCapturingStatus(user, v);
      if (!knownCaptureStatus || *knownCaptureStatus)
        return true;
    }
  }

  return false;
}

/// Returns true if two values may be referencing aliasing memory. This is a
/// rather naive and conservative analysis. Values defined by different
/// allocation-like operations as well as values derived from those by casts and
/// views cannot alias each other. Similarly, values defined by allocations
/// inside a function cannot alias function arguments. Global values cannot
/// alias each other or local allocations. Values that are captured, i.e.
/// themselves potentially stored in memory, are considered as aliasing with
/// everything. This seems sufficient to achieve barrier removal in structured
/// control flow, more complex cases would require a proper dataflow analysis.
static bool mayAlias(Value first, Value second) {
  LDBG(DEBUG_TYPE_ALIAS, 1)
      << "checking aliasing between " << first << " and " << second;

  first = getBase(first);
  second = getBase(second);

  LDBG(DEBUG_TYPE_ALIAS, 1) << "base " << first << " and " << second;

  // Values derived from the same base memref do alias (unless we do a more
  // advanced analysis to prove non-overlapping accesses).
  if (first == second) {
    LDBG(DEBUG_TYPE_ALIAS, 1) << "-> do alias!";
    return true;
  }

  // Different globals cannot alias.
  if (auto globFirst = first.getDefiningOp<memref::GetGlobalOp>()) {
    if (auto globSecond = second.getDefiningOp<memref::GetGlobalOp>()) {
      return globFirst.getNameAttr() == globSecond.getNameAttr();
    }
  }

  // Two function arguments marked as noalias do not alias.
  auto isNoaliasFuncArgument = [](Value value) {
    auto bbArg = dyn_cast<BlockArgument>(value);
    if (!bbArg)
      return false;
    auto iface = dyn_cast<FunctionOpInterface>(bbArg.getOwner()->getParentOp());
    if (!iface)
      return false;
    // TODO: we need a way to not depend on the LLVM dialect here.
    return iface.getArgAttr(bbArg.getArgNumber(), "llvm.noalias") != nullptr;
  };
  if (isNoaliasFuncArgument(first) && isNoaliasFuncArgument(second))
    return false;

  bool isDistinct[] = {producesDistinctBase(first.getDefiningOp()),
                       producesDistinctBase(second.getDefiningOp())};
  bool isGlobal[] = {first.getDefiningOp<memref::GetGlobalOp>() != nullptr,
                     second.getDefiningOp<memref::GetGlobalOp>() != nullptr};

  // Non-equivalent distinct bases and globals cannot alias. At this point, we
  // have already filtered out based on values being equal and global name being
  // equal.
  if ((isDistinct[0] || isGlobal[0]) && (isDistinct[1] || isGlobal[1]))
    return false;

  bool isArg[] = {isFunctionArgument(first), isFunctionArgument(second)};

  // Distinct bases (allocations) cannot have been passed as an argument.
  if ((isDistinct[0] && isArg[1]) || (isDistinct[1] && isArg[0]))
    return false;

  // Non-captured base distinct values cannot conflict with another base value.
  if (isDistinct[0] && !maybeCaptured(first))
    return false;
  if (isDistinct[1] && !maybeCaptured(second))
    return false;

  // Otherwise, conservatively assume aliasing.
  LDBG(DEBUG_TYPE_ALIAS, 1) << "-> may alias!";
  return true;
}

/// Returns `true` if the effect may be affecting memory aliasing the value. If
/// the effect is not associated with any value, it is assumed to affect all
/// memory and therefore aliases with everything.
static bool mayAlias(MemoryEffects::EffectInstance a, Value v2) {
  if (Value v = a.getValue()) {
    return mayAlias(v, v2);
  }
  return true;
}

/// Returns `true` if the two effects may be affecting aliasing memory. If
/// an effect is not associated with any value, it is assumed to affect all
/// memory and therefore aliases with everything. Effects on different resources
/// cannot alias.
static bool mayAlias(MemoryEffects::EffectInstance a,
                     MemoryEffects::EffectInstance b) {
  if (a.getResource()->getResourceID() != b.getResource()->getResourceID())
    return false;
  if (Value v2 = b.getValue()) {
    return mayAlias(a, v2);
  }
  if (Value v = a.getValue()) {
    return mayAlias(b, v);
  }
  return true;
}

/// Returns `true` if any of the "before" effect instances has a conflict with
/// any "after" instance for the purpose of barrier elimination. The effects are
/// supposed to be limited to a barrier synchronization scope. A conflict exists
/// if effects instances affect aliasing memory locations and at least on of
/// then as a write. As an exception, if the non-write effect is an allocation
/// effect, there is no conflict since we are only expected to see the
/// allocation happening in the same thread and it cannot be accessed from
/// another thread without capture (which we do handle in alias analysis).
static bool
haveConflictingEffects(ArrayRef<MemoryEffects::EffectInstance> beforeEffects,
                       ArrayRef<MemoryEffects::EffectInstance> afterEffects) {
  for (const MemoryEffects::EffectInstance &before : beforeEffects) {
    for (const MemoryEffects::EffectInstance &after : afterEffects) {
      // If cannot alias, definitely no conflict.
      if (!mayAlias(before, after))
        continue;

      // Read/read is not a conflict.
      if (isa<MemoryEffects::Read>(before.getEffect()) &&
          isa<MemoryEffects::Read>(after.getEffect())) {
        continue;
      }

      // Allocate/* is not a conflict since the allocation happens within the
      // thread context.
      // TODO: This is not the case for */Free unless the allocation happened in
      // the thread context, which we could also check for.
      if (isa<MemoryEffects::Allocate>(before.getEffect()) ||
          isa<MemoryEffects::Allocate>(after.getEffect())) {
        continue;
      }

      // In the particular case that the before effect is a free, we only have 2
      // possibilities:
      //   1. either the program is well-formed and there must be an interleaved
      //      alloc that must limit the scope of effect lookback and we can
      //      safely ignore the free -> read / free -> write and free -> free
      //      conflicts.
      //   2. either the program is ill-formed and we are in undefined behavior
      //      territory.
      if (isa<MemoryEffects::Free>(before.getEffect()))
        continue;

      // Other kinds of effects create a conflict, e.g. read-after-write.
      LDBG() << "found a conflict between (before): " << before.getValue()
             << " read:" << isa<MemoryEffects::Read>(before.getEffect())
             << " write:" << isa<MemoryEffects::Write>(before.getEffect())
             << " alloc:" << isa<MemoryEffects::Allocate>(before.getEffect())
             << " free:" << isa<MemoryEffects::Free>(before.getEffect());
      LDBG() << "and (after):                " << after.getValue()
             << " read:" << isa<MemoryEffects::Read>(after.getEffect())
             << " write:" << isa<MemoryEffects::Write>(after.getEffect())
             << " alloc:" << isa<MemoryEffects::Allocate>(after.getEffect())
             << " free:" << isa<MemoryEffects::Free>(after.getEffect());
      return true;
    }
  }

  return false;
}

namespace {
class BarrierElimination final : public OpRewritePattern<BarrierOp> {
public:
  using OpRewritePattern<BarrierOp>::OpRewritePattern;

  LogicalResult matchAndRewrite(BarrierOp barrier,
                                PatternRewriter &rewriter) const override {
    LDBG() << "checking the necessity of: " << barrier << " "
           << barrier.getLoc();

    std::optional<ArrayAttr> fencedMemSpaces = barrier.getAddressSpaces();
    if (fencedMemSpaces && fencedMemSpaces->empty()) {
      LDBG()
          << "barrier is not used to synchronize memory accesses, retain it\n";
      return failure();
    }

    // Convert the fenced address spaces to the proper type for passing through.
    SmallVector<gpu::AddressSpaceAttr> fencedSpacesStorage;
    std::optional<ArrayRef<gpu::AddressSpaceAttr>> fencedSpaces;
    if (fencedMemSpaces) {
      fencedSpacesStorage = llvm::map_to_vector(
          *fencedMemSpaces, llvm::CastTo<gpu::AddressSpaceAttr>);
      fencedSpaces = fencedSpacesStorage;
    }

    SmallVector<MemoryEffects::EffectInstance> beforeEffects;
    getEffectsBefore(barrier, beforeEffects, fencedSpaces,
                     /*stopAtBarrier=*/true);

    SmallVector<MemoryEffects::EffectInstance> afterEffects;
    getEffectsAfter(barrier, afterEffects, fencedSpaces,
                    /*stopAtBarrier=*/true);

    if (!haveConflictingEffects(beforeEffects, afterEffects)) {
      LDBG() << "the surrounding barriers are sufficient, removing " << barrier;
      rewriter.eraseOp(barrier);
      return success();
    }

    LDBG() << "barrier is necessary: " << barrier << " " << barrier.getLoc();
    return failure();
  }
};

class GpuEliminateBarriersPass
    : public impl::GpuEliminateBarriersBase<GpuEliminateBarriersPass> {
  void runOnOperation() override {
    auto funcOp = getOperation();
    RewritePatternSet patterns(&getContext());
    mlir::populateGpuEliminateBarriersPatterns(patterns);
    if (failed(applyPatternsGreedily(funcOp, std::move(patterns)))) {
      return signalPassFailure();
    }
  }
};

} // namespace

void mlir::populateGpuEliminateBarriersPatterns(RewritePatternSet &patterns) {
  patterns.insert<BarrierElimination>(patterns.getContext());
}