[MLIR][RemoveDeadValues] Mark arguments of a public function Live #162038
+210
−19
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This diff also changes traversal order from forward to backward for region/block/ops. This order guanratees Liveness updates at a callsite can propagates to the defs of arguments. ``` ./bin/llvm-lit -v ../mlir/test/Transforms/remove-dead-values.mlir ```
|
@llvm/pr-subscribers-mlir Author: xin liu (navyxliu) ChangesThis diff is REDO of #160242. ProblemLiveness analysis is inter-procedural. If there are some unused arguments in a public function, they propagate to callers. From the perspective of RemoveDeadValues, the signature of a public function is immutable. It can't cope with this situation. One side, it deletes outgoing arguments, on the other side it keeps the function intact. SolutionWe deploy two methods to fix this bug.
Test planFull diff: https://github.com/llvm/llvm-project/pull/162038.diff 4 Files Affected:
diff --git a/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h b/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
index cf1fd6e2d48ca..be7e027b95f64 100644
--- a/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
+++ b/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
@@ -102,6 +102,9 @@ struct RunLivenessAnalysis {
const Liveness *getLiveness(Value val);
+ /// Return the configuration of the solver used for this analysis.
+ const DataFlowConfig &getSolverConfig() const { return solver.getConfig(); }
+
private:
/// Stores the result of the liveness analysis that was run.
DataFlowSolver solver;
diff --git a/mlir/include/mlir/IR/Visitors.h b/mlir/include/mlir/IR/Visitors.h
index 893f66ae33deb..5766d262796d6 100644
--- a/mlir/include/mlir/IR/Visitors.h
+++ b/mlir/include/mlir/IR/Visitors.h
@@ -39,6 +39,20 @@ struct ForwardIterator {
}
};
+/// This iterator enumerates the elements in "backward" order.
+struct BackwardIterator {
+ template <typename T>
+ static auto makeIterable(T &range) {
+ if constexpr (std::is_same<T, Operation>()) {
+ /// Make operations iterable: return the list of regions.
+ return llvm::reverse(range.getRegions());
+ } else {
+ /// Regions and block are already iterable.
+ return llvm::reverse(range);
+ }
+ }
+};
+
/// A utility class to encode the current walk stage for "generic" walkers.
/// When walking an operation, we can either choose a Pre/Post order walker
/// which invokes the callback on an operation before/after all its attached
diff --git a/mlir/lib/Transforms/RemoveDeadValues.cpp b/mlir/lib/Transforms/RemoveDeadValues.cpp
index e0c65b0e09774..ed5c6a8d2ead0 100644
--- a/mlir/lib/Transforms/RemoveDeadValues.cpp
+++ b/mlir/lib/Transforms/RemoveDeadValues.cpp
@@ -33,6 +33,7 @@
#include "mlir/Analysis/DataFlow/DeadCodeAnalysis.h"
#include "mlir/Analysis/DataFlow/LivenessAnalysis.h"
+#include "mlir/Dialect/Arith/IR/Arith.h"
#include "mlir/IR/Builders.h"
#include "mlir/IR/BuiltinAttributes.h"
#include "mlir/IR/Dialect.h"
@@ -118,8 +119,13 @@ struct RDVFinalCleanupList {
/// Return true iff at least one value in `values` is live, given the liveness
/// information in `la`.
static bool hasLive(ValueRange values, const DenseSet<Value> &nonLiveSet,
- RunLivenessAnalysis &la) {
+ const DenseSet<Value> &liveSet, RunLivenessAnalysis &la) {
for (Value value : values) {
+ if (liveSet.contains(value)) {
+ LDBG() << "Value " << value << " is marked live by CallOp";
+ return true;
+ }
+
if (nonLiveSet.contains(value)) {
LDBG() << "Value " << value << " is already marked non-live (dead)";
continue;
@@ -144,6 +150,7 @@ static bool hasLive(ValueRange values, const DenseSet<Value> &nonLiveSet,
/// Return a BitVector of size `values.size()` where its i-th bit is 1 iff the
/// i-th value in `values` is live, given the liveness information in `la`.
static BitVector markLives(ValueRange values, const DenseSet<Value> &nonLiveSet,
+ const DenseSet<Value> &liveSet,
RunLivenessAnalysis &la) {
BitVector lives(values.size(), true);
@@ -154,7 +161,9 @@ static BitVector markLives(ValueRange values, const DenseSet<Value> &nonLiveSet,
<< " is already marked non-live (dead) at index " << index;
continue;
}
-
+ if (liveSet.contains(value)) {
+ continue;
+ }
const Liveness *liveness = la.getLiveness(value);
// It is important to note that when `liveness` is null, we can't tell if
// `value` is live or not. So, the safe option is to consider it live. Also,
@@ -259,8 +268,19 @@ static SmallVector<OpOperand *> operandsToOpOperands(OperandRange operands) {
/// - Return-like
static void processSimpleOp(Operation *op, RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
- if (!isMemoryEffectFree(op) || hasLive(op->getResults(), nonLiveSet, la)) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
+ for (Value val : op->getResults()) {
+ if (liveSet.contains(val)) {
+ LDBG() << "Simple op is used by a public function, "
+ "preserving it: "
+ << OpWithFlags(op, OpPrintingFlags().skipRegions());
+ liveSet.insert_range(op->getOperands());
+ return;
+ }
+ }
+
+ if (!isMemoryEffectFree(op) ||
+ hasLive(op->getResults(), nonLiveSet, liveSet, la)) {
LDBG() << "Simple op is not memory effect free or has live results, "
"preserving it: "
<< OpWithFlags(op, OpPrintingFlags().skipRegions());
@@ -288,7 +308,7 @@ static void processSimpleOp(Operation *op, RunLivenessAnalysis &la,
/// (6) Marking all its results as non-live values.
static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
RunLivenessAnalysis &la, DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
LDBG() << "Processing function op: "
<< OpWithFlags(funcOp, OpPrintingFlags().skipRegions());
if (funcOp.isPublic() || funcOp.isExternal()) {
@@ -299,7 +319,7 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
// Get the list of unnecessary (non-live) arguments in `nonLiveArgs`.
SmallVector<Value> arguments(funcOp.getArguments());
- BitVector nonLiveArgs = markLives(arguments, nonLiveSet, la);
+ BitVector nonLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
nonLiveArgs = nonLiveArgs.flip();
// Do (1).
@@ -352,7 +372,8 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
for (SymbolTable::SymbolUse use : uses) {
Operation *callOp = use.getUser();
assert(isa<CallOpInterface>(callOp) && "expected a call-like user");
- BitVector liveCallRets = markLives(callOp->getResults(), nonLiveSet, la);
+ BitVector liveCallRets =
+ markLives(callOp->getResults(), nonLiveSet, liveSet, la);
nonLiveRets &= liveCallRets.flip();
}
@@ -379,6 +400,56 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
}
}
+// Create a cheaper value with the same type of oldVal in front of CallOp.
+static Value createDummyArgument(CallOpInterface callOp, Value oldVal) {
+ OpBuilder builder(callOp.getOperation());
+ Type type = oldVal.getType();
+
+ // Create zero constant for any supported type
+ if (TypedAttr zeroAttr = builder.getZeroAttr(type)) {
+ return builder.create<arith::ConstantOp>(oldVal.getLoc(), type, zeroAttr);
+ }
+ return {};
+}
+
+static void processCallOp(CallOpInterface callOp, Operation *module,
+ RunLivenessAnalysis &la, DenseSet<Value> &nonLiveSet,
+ DenseSet<Value> &liveSet) {
+ if (!la.getSolverConfig().isInterprocedural())
+ return;
+
+ Operation *callableOp = callOp.resolveCallable();
+ auto funcOp = dyn_cast<FunctionOpInterface>(callableOp);
+ if (!funcOp || !funcOp.isPublic()) {
+ return;
+ }
+
+ LDBG() << "processCallOp to a public function: " << funcOp.getName();
+ // Get the list of unnecessary (non-live) arguments in `nonLiveArgs`.
+ SmallVector<Value> arguments(funcOp.getArguments());
+ BitVector nonLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
+ nonLiveArgs = nonLiveArgs.flip();
+
+ if (nonLiveArgs.count() > 0) {
+ LDBG() << funcOp.getName() << " contains NonLive arguments";
+ // The number of operands in the call op may not match the number of
+ // arguments in the func op.
+ SmallVector<OpOperand *> callOpOperands =
+ operandsToOpOperands(callOp.getArgOperands());
+
+ for (int index : nonLiveArgs.set_bits()) {
+ OpOperand *operand = callOpOperands[index];
+ Value oldVal = operand->get();
+ if (Value dummy = createDummyArgument(callOp, oldVal)) {
+ callOp->setOperand(operand->getOperandNumber(), dummy);
+ nonLiveSet.insert(oldVal);
+ } else {
+ liveSet.insert(oldVal);
+ }
+ }
+ }
+}
+
/// Process a region branch operation `regionBranchOp` using the liveness
/// information in `la`. The processing involves two scenarios:
///
@@ -411,12 +482,14 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
+ DenseSet<Value> &liveSet,
RDVFinalCleanupList &cl) {
LDBG() << "Processing region branch op: "
<< OpWithFlags(regionBranchOp, OpPrintingFlags().skipRegions());
// Mark live results of `regionBranchOp` in `liveResults`.
auto markLiveResults = [&](BitVector &liveResults) {
- liveResults = markLives(regionBranchOp->getResults(), nonLiveSet, la);
+ liveResults =
+ markLives(regionBranchOp->getResults(), nonLiveSet, liveSet, la);
};
// Mark live arguments in the regions of `regionBranchOp` in `liveArgs`.
@@ -425,7 +498,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
if (region.empty())
continue;
SmallVector<Value> arguments(region.front().getArguments());
- BitVector regionLiveArgs = markLives(arguments, nonLiveSet, la);
+ BitVector regionLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
liveArgs[®ion] = regionLiveArgs;
}
};
@@ -619,7 +692,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
// attributed to something else.
// Do (1') and (2').
if (isMemoryEffectFree(regionBranchOp.getOperation()) &&
- !hasLive(regionBranchOp->getResults(), nonLiveSet, la)) {
+ !hasLive(regionBranchOp->getResults(), nonLiveSet, liveSet, la)) {
cl.operations.push_back(regionBranchOp.getOperation());
return;
}
@@ -698,7 +771,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
static void processBranchOp(BranchOpInterface branchOp, RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
LDBG() << "Processing branch op: " << *branchOp;
unsigned numSuccessors = branchOp->getNumSuccessors();
@@ -716,7 +789,7 @@ static void processBranchOp(BranchOpInterface branchOp, RunLivenessAnalysis &la,
// Do (2)
BitVector successorNonLive =
- markLives(operandValues, nonLiveSet, la).flip();
+ markLives(operandValues, nonLiveSet, liveSet, la).flip();
collectNonLiveValues(nonLiveSet, successorBlock->getArguments(),
successorNonLive);
@@ -876,26 +949,29 @@ void RemoveDeadValues::runOnOperation() {
// Tracks values eligible for erasure - complements liveness analysis to
// identify "droppable" values.
DenseSet<Value> deadVals;
+ // mark outgoing arguments to a public function LIVE. We also propagate
+ // liveness backward.
+ DenseSet<Value> liveVals;
// Maintains a list of Ops, values, branches, etc., slated for cleanup at the
// end of this pass.
RDVFinalCleanupList finalCleanupList;
- module->walk([&](Operation *op) {
+ module->walk<WalkOrder::PostOrder, BackwardIterator>([&](Operation *op) {
if (auto funcOp = dyn_cast<FunctionOpInterface>(op)) {
- processFuncOp(funcOp, module, la, deadVals, finalCleanupList);
+ processFuncOp(funcOp, module, la, deadVals, liveVals, finalCleanupList);
} else if (auto regionBranchOp = dyn_cast<RegionBranchOpInterface>(op)) {
- processRegionBranchOp(regionBranchOp, la, deadVals, finalCleanupList);
+ processRegionBranchOp(regionBranchOp, la, deadVals, liveVals,
+ finalCleanupList);
} else if (auto branchOp = dyn_cast<BranchOpInterface>(op)) {
- processBranchOp(branchOp, la, deadVals, finalCleanupList);
+ processBranchOp(branchOp, la, deadVals, liveVals, finalCleanupList);
} else if (op->hasTrait<::mlir::OpTrait::IsTerminator>()) {
// Nothing to do here because this is a terminator op and it should be
// honored with respect to its parent
} else if (isa<CallOpInterface>(op)) {
- // Nothing to do because this op is associated with a function op and gets
- // cleaned when the latter is cleaned.
+ processCallOp(cast<CallOpInterface>(op), module, la, deadVals, liveVals);
} else {
- processSimpleOp(op, la, deadVals, finalCleanupList);
+ processSimpleOp(op, la, deadVals, liveVals, finalCleanupList);
}
});
diff --git a/mlir/test/Transforms/remove-dead-values.mlir b/mlir/test/Transforms/remove-dead-values.mlir
index 56449469dc29f..ebb76a8835ceb 100644
--- a/mlir/test/Transforms/remove-dead-values.mlir
+++ b/mlir/test/Transforms/remove-dead-values.mlir
@@ -569,6 +569,24 @@ module @return_void_with_unused_argument {
call @fn_return_void_with_unused_argument(%arg0, %unused) : (i32, memref<4xi32>) -> ()
return %unused : memref<4xi32>
}
+
+ // the function signature is immutable because it is public.
+ func.func public @immutable_fn_with_unused_argument(%arg0: i32, %arg1: memref<4xf32>) -> () {
+ return
+ }
+
+ // CHECK-LABEL: func.func @main2
+ // CHECK: %[[MEM:.*]] = memref.alloc() : memref<4xf32>
+ // CHECK: %[[UNUSED:.*]] = arith.constant 0 : i32
+ // CHECK: call @immutable_fn_with_unused_argument(%[[UNUSED]], %[[MEM]]) : (i32, memref<4xf32>) -> ()
+ func.func @main2() -> () {
+ %one = arith.constant 1 : i32
+ %scalar = arith.addi %one, %one: i32
+ %mem = memref.alloc() : memref<4xf32>
+
+ call @immutable_fn_with_unused_argument(%scalar, %mem) : (i32, memref<4xf32>) -> ()
+ return
+ }
}
// -----
|
|
@llvm/pr-subscribers-mlir-core Author: xin liu (navyxliu) ChangesThis diff is REDO of #160242. ProblemLiveness analysis is inter-procedural. If there are some unused arguments in a public function, they propagate to callers. From the perspective of RemoveDeadValues, the signature of a public function is immutable. It can't cope with this situation. One side, it deletes outgoing arguments, on the other side it keeps the function intact. SolutionWe deploy two methods to fix this bug.
Test planFull diff: https://github.com/llvm/llvm-project/pull/162038.diff 4 Files Affected:
diff --git a/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h b/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
index cf1fd6e2d48ca..be7e027b95f64 100644
--- a/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
+++ b/mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h
@@ -102,6 +102,9 @@ struct RunLivenessAnalysis {
const Liveness *getLiveness(Value val);
+ /// Return the configuration of the solver used for this analysis.
+ const DataFlowConfig &getSolverConfig() const { return solver.getConfig(); }
+
private:
/// Stores the result of the liveness analysis that was run.
DataFlowSolver solver;
diff --git a/mlir/include/mlir/IR/Visitors.h b/mlir/include/mlir/IR/Visitors.h
index 893f66ae33deb..5766d262796d6 100644
--- a/mlir/include/mlir/IR/Visitors.h
+++ b/mlir/include/mlir/IR/Visitors.h
@@ -39,6 +39,20 @@ struct ForwardIterator {
}
};
+/// This iterator enumerates the elements in "backward" order.
+struct BackwardIterator {
+ template <typename T>
+ static auto makeIterable(T &range) {
+ if constexpr (std::is_same<T, Operation>()) {
+ /// Make operations iterable: return the list of regions.
+ return llvm::reverse(range.getRegions());
+ } else {
+ /// Regions and block are already iterable.
+ return llvm::reverse(range);
+ }
+ }
+};
+
/// A utility class to encode the current walk stage for "generic" walkers.
/// When walking an operation, we can either choose a Pre/Post order walker
/// which invokes the callback on an operation before/after all its attached
diff --git a/mlir/lib/Transforms/RemoveDeadValues.cpp b/mlir/lib/Transforms/RemoveDeadValues.cpp
index e0c65b0e09774..ed5c6a8d2ead0 100644
--- a/mlir/lib/Transforms/RemoveDeadValues.cpp
+++ b/mlir/lib/Transforms/RemoveDeadValues.cpp
@@ -33,6 +33,7 @@
#include "mlir/Analysis/DataFlow/DeadCodeAnalysis.h"
#include "mlir/Analysis/DataFlow/LivenessAnalysis.h"
+#include "mlir/Dialect/Arith/IR/Arith.h"
#include "mlir/IR/Builders.h"
#include "mlir/IR/BuiltinAttributes.h"
#include "mlir/IR/Dialect.h"
@@ -118,8 +119,13 @@ struct RDVFinalCleanupList {
/// Return true iff at least one value in `values` is live, given the liveness
/// information in `la`.
static bool hasLive(ValueRange values, const DenseSet<Value> &nonLiveSet,
- RunLivenessAnalysis &la) {
+ const DenseSet<Value> &liveSet, RunLivenessAnalysis &la) {
for (Value value : values) {
+ if (liveSet.contains(value)) {
+ LDBG() << "Value " << value << " is marked live by CallOp";
+ return true;
+ }
+
if (nonLiveSet.contains(value)) {
LDBG() << "Value " << value << " is already marked non-live (dead)";
continue;
@@ -144,6 +150,7 @@ static bool hasLive(ValueRange values, const DenseSet<Value> &nonLiveSet,
/// Return a BitVector of size `values.size()` where its i-th bit is 1 iff the
/// i-th value in `values` is live, given the liveness information in `la`.
static BitVector markLives(ValueRange values, const DenseSet<Value> &nonLiveSet,
+ const DenseSet<Value> &liveSet,
RunLivenessAnalysis &la) {
BitVector lives(values.size(), true);
@@ -154,7 +161,9 @@ static BitVector markLives(ValueRange values, const DenseSet<Value> &nonLiveSet,
<< " is already marked non-live (dead) at index " << index;
continue;
}
-
+ if (liveSet.contains(value)) {
+ continue;
+ }
const Liveness *liveness = la.getLiveness(value);
// It is important to note that when `liveness` is null, we can't tell if
// `value` is live or not. So, the safe option is to consider it live. Also,
@@ -259,8 +268,19 @@ static SmallVector<OpOperand *> operandsToOpOperands(OperandRange operands) {
/// - Return-like
static void processSimpleOp(Operation *op, RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
- if (!isMemoryEffectFree(op) || hasLive(op->getResults(), nonLiveSet, la)) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
+ for (Value val : op->getResults()) {
+ if (liveSet.contains(val)) {
+ LDBG() << "Simple op is used by a public function, "
+ "preserving it: "
+ << OpWithFlags(op, OpPrintingFlags().skipRegions());
+ liveSet.insert_range(op->getOperands());
+ return;
+ }
+ }
+
+ if (!isMemoryEffectFree(op) ||
+ hasLive(op->getResults(), nonLiveSet, liveSet, la)) {
LDBG() << "Simple op is not memory effect free or has live results, "
"preserving it: "
<< OpWithFlags(op, OpPrintingFlags().skipRegions());
@@ -288,7 +308,7 @@ static void processSimpleOp(Operation *op, RunLivenessAnalysis &la,
/// (6) Marking all its results as non-live values.
static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
RunLivenessAnalysis &la, DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
LDBG() << "Processing function op: "
<< OpWithFlags(funcOp, OpPrintingFlags().skipRegions());
if (funcOp.isPublic() || funcOp.isExternal()) {
@@ -299,7 +319,7 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
// Get the list of unnecessary (non-live) arguments in `nonLiveArgs`.
SmallVector<Value> arguments(funcOp.getArguments());
- BitVector nonLiveArgs = markLives(arguments, nonLiveSet, la);
+ BitVector nonLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
nonLiveArgs = nonLiveArgs.flip();
// Do (1).
@@ -352,7 +372,8 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
for (SymbolTable::SymbolUse use : uses) {
Operation *callOp = use.getUser();
assert(isa<CallOpInterface>(callOp) && "expected a call-like user");
- BitVector liveCallRets = markLives(callOp->getResults(), nonLiveSet, la);
+ BitVector liveCallRets =
+ markLives(callOp->getResults(), nonLiveSet, liveSet, la);
nonLiveRets &= liveCallRets.flip();
}
@@ -379,6 +400,56 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
}
}
+// Create a cheaper value with the same type of oldVal in front of CallOp.
+static Value createDummyArgument(CallOpInterface callOp, Value oldVal) {
+ OpBuilder builder(callOp.getOperation());
+ Type type = oldVal.getType();
+
+ // Create zero constant for any supported type
+ if (TypedAttr zeroAttr = builder.getZeroAttr(type)) {
+ return builder.create<arith::ConstantOp>(oldVal.getLoc(), type, zeroAttr);
+ }
+ return {};
+}
+
+static void processCallOp(CallOpInterface callOp, Operation *module,
+ RunLivenessAnalysis &la, DenseSet<Value> &nonLiveSet,
+ DenseSet<Value> &liveSet) {
+ if (!la.getSolverConfig().isInterprocedural())
+ return;
+
+ Operation *callableOp = callOp.resolveCallable();
+ auto funcOp = dyn_cast<FunctionOpInterface>(callableOp);
+ if (!funcOp || !funcOp.isPublic()) {
+ return;
+ }
+
+ LDBG() << "processCallOp to a public function: " << funcOp.getName();
+ // Get the list of unnecessary (non-live) arguments in `nonLiveArgs`.
+ SmallVector<Value> arguments(funcOp.getArguments());
+ BitVector nonLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
+ nonLiveArgs = nonLiveArgs.flip();
+
+ if (nonLiveArgs.count() > 0) {
+ LDBG() << funcOp.getName() << " contains NonLive arguments";
+ // The number of operands in the call op may not match the number of
+ // arguments in the func op.
+ SmallVector<OpOperand *> callOpOperands =
+ operandsToOpOperands(callOp.getArgOperands());
+
+ for (int index : nonLiveArgs.set_bits()) {
+ OpOperand *operand = callOpOperands[index];
+ Value oldVal = operand->get();
+ if (Value dummy = createDummyArgument(callOp, oldVal)) {
+ callOp->setOperand(operand->getOperandNumber(), dummy);
+ nonLiveSet.insert(oldVal);
+ } else {
+ liveSet.insert(oldVal);
+ }
+ }
+ }
+}
+
/// Process a region branch operation `regionBranchOp` using the liveness
/// information in `la`. The processing involves two scenarios:
///
@@ -411,12 +482,14 @@ static void processFuncOp(FunctionOpInterface funcOp, Operation *module,
static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
+ DenseSet<Value> &liveSet,
RDVFinalCleanupList &cl) {
LDBG() << "Processing region branch op: "
<< OpWithFlags(regionBranchOp, OpPrintingFlags().skipRegions());
// Mark live results of `regionBranchOp` in `liveResults`.
auto markLiveResults = [&](BitVector &liveResults) {
- liveResults = markLives(regionBranchOp->getResults(), nonLiveSet, la);
+ liveResults =
+ markLives(regionBranchOp->getResults(), nonLiveSet, liveSet, la);
};
// Mark live arguments in the regions of `regionBranchOp` in `liveArgs`.
@@ -425,7 +498,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
if (region.empty())
continue;
SmallVector<Value> arguments(region.front().getArguments());
- BitVector regionLiveArgs = markLives(arguments, nonLiveSet, la);
+ BitVector regionLiveArgs = markLives(arguments, nonLiveSet, liveSet, la);
liveArgs[®ion] = regionLiveArgs;
}
};
@@ -619,7 +692,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
// attributed to something else.
// Do (1') and (2').
if (isMemoryEffectFree(regionBranchOp.getOperation()) &&
- !hasLive(regionBranchOp->getResults(), nonLiveSet, la)) {
+ !hasLive(regionBranchOp->getResults(), nonLiveSet, liveSet, la)) {
cl.operations.push_back(regionBranchOp.getOperation());
return;
}
@@ -698,7 +771,7 @@ static void processRegionBranchOp(RegionBranchOpInterface regionBranchOp,
static void processBranchOp(BranchOpInterface branchOp, RunLivenessAnalysis &la,
DenseSet<Value> &nonLiveSet,
- RDVFinalCleanupList &cl) {
+ DenseSet<Value> &liveSet, RDVFinalCleanupList &cl) {
LDBG() << "Processing branch op: " << *branchOp;
unsigned numSuccessors = branchOp->getNumSuccessors();
@@ -716,7 +789,7 @@ static void processBranchOp(BranchOpInterface branchOp, RunLivenessAnalysis &la,
// Do (2)
BitVector successorNonLive =
- markLives(operandValues, nonLiveSet, la).flip();
+ markLives(operandValues, nonLiveSet, liveSet, la).flip();
collectNonLiveValues(nonLiveSet, successorBlock->getArguments(),
successorNonLive);
@@ -876,26 +949,29 @@ void RemoveDeadValues::runOnOperation() {
// Tracks values eligible for erasure - complements liveness analysis to
// identify "droppable" values.
DenseSet<Value> deadVals;
+ // mark outgoing arguments to a public function LIVE. We also propagate
+ // liveness backward.
+ DenseSet<Value> liveVals;
// Maintains a list of Ops, values, branches, etc., slated for cleanup at the
// end of this pass.
RDVFinalCleanupList finalCleanupList;
- module->walk([&](Operation *op) {
+ module->walk<WalkOrder::PostOrder, BackwardIterator>([&](Operation *op) {
if (auto funcOp = dyn_cast<FunctionOpInterface>(op)) {
- processFuncOp(funcOp, module, la, deadVals, finalCleanupList);
+ processFuncOp(funcOp, module, la, deadVals, liveVals, finalCleanupList);
} else if (auto regionBranchOp = dyn_cast<RegionBranchOpInterface>(op)) {
- processRegionBranchOp(regionBranchOp, la, deadVals, finalCleanupList);
+ processRegionBranchOp(regionBranchOp, la, deadVals, liveVals,
+ finalCleanupList);
} else if (auto branchOp = dyn_cast<BranchOpInterface>(op)) {
- processBranchOp(branchOp, la, deadVals, finalCleanupList);
+ processBranchOp(branchOp, la, deadVals, liveVals, finalCleanupList);
} else if (op->hasTrait<::mlir::OpTrait::IsTerminator>()) {
// Nothing to do here because this is a terminator op and it should be
// honored with respect to its parent
} else if (isa<CallOpInterface>(op)) {
- // Nothing to do because this op is associated with a function op and gets
- // cleaned when the latter is cleaned.
+ processCallOp(cast<CallOpInterface>(op), module, la, deadVals, liveVals);
} else {
- processSimpleOp(op, la, deadVals, finalCleanupList);
+ processSimpleOp(op, la, deadVals, liveVals, finalCleanupList);
}
});
diff --git a/mlir/test/Transforms/remove-dead-values.mlir b/mlir/test/Transforms/remove-dead-values.mlir
index 56449469dc29f..ebb76a8835ceb 100644
--- a/mlir/test/Transforms/remove-dead-values.mlir
+++ b/mlir/test/Transforms/remove-dead-values.mlir
@@ -569,6 +569,24 @@ module @return_void_with_unused_argument {
call @fn_return_void_with_unused_argument(%arg0, %unused) : (i32, memref<4xi32>) -> ()
return %unused : memref<4xi32>
}
+
+ // the function signature is immutable because it is public.
+ func.func public @immutable_fn_with_unused_argument(%arg0: i32, %arg1: memref<4xf32>) -> () {
+ return
+ }
+
+ // CHECK-LABEL: func.func @main2
+ // CHECK: %[[MEM:.*]] = memref.alloc() : memref<4xf32>
+ // CHECK: %[[UNUSED:.*]] = arith.constant 0 : i32
+ // CHECK: call @immutable_fn_with_unused_argument(%[[UNUSED]], %[[MEM]]) : (i32, memref<4xf32>) -> ()
+ func.func @main2() -> () {
+ %one = arith.constant 1 : i32
+ %scalar = arith.addi %one, %one: i32
+ %mem = memref.alloc() : memref<4xf32>
+
+ call @immutable_fn_with_unused_argument(%scalar, %mem) : (i32, memref<4xf32>) -> ()
+ return
+ }
}
// -----
|
|
hi, @joker-eph |
joker-eph
reviewed
Oct 6, 2025
| call @immutable_fn_with_unused_argument(%scalar, %mem) : (i32, memref<4xf32>) -> () | ||
| return | ||
| } | ||
| } |
There was a problem hiding this comment.
Can you add two CFG examples where the blocks are listed in different order to ensure you're not sensitive to the order the blocks are in-memory.
There was a problem hiding this comment.
hi, @joker-eph
I did add a testcase as you said. Then I realize that it's non-trivial to propagate liveness in RemoveDeadValues.
Here is a testcase only for RegionBranchOpinterface. From %0 is live at line 11, we need to mark %0 is live at line 2. After that, we need to mark %c1_i32 at line 4 and c0_i32 at line 7 live as well. In order words, we need to walk function @main3 preorder + backward.
1 func.func @main3(%arg0: i1) {
2 %0 = scf.if %arg0 -> (i32) {
3 %c1_i32 = arith.constant 1 : i32
4 scf.yield %c1_i32 : i32
5 } else {
6 %c0_i32 = arith.constant 0 : i32
7 scf.yield %c0_i32 : i32
8 }
9 %mem = memref.alloc() : memref<4xf32>
10
11 call @immutable_fn_with_unused_argument(%0, %mem) : (i32, memref<4xf32>) -> ()
12 return
13 }
I manage to fix this in propagateBackward. It pretty much redo what liveness analysis has done. TBH, I don't think this is the right way to proceed. RemoveDeadValues should keep its own single responsibility.
I take a step back and think about why we end up here. The very reason we try to propagate liveness in it because:
- liveness is immutable.
- We somehow need to update the NonLive arguments of a public function.
How about we just introduce a new pass: 'privatize-public-function' right before 'remove-dead-values'.
- It deploys separation of interface and implementation.
- If nothing changes, we preserve liveness. Otherwise, we invalidate it and let remove-dead-value recompute.
Here is a demo what this pass transforms.
i think we can waive cost model because we don't clone function body. We just create a
thin wrapper.
public void foo(int unused){...}
void main() {
arg = compute();
call foo(arg);
}
=>
public void foo(int unused) { // interface
return __foo_impl(unused);
}
private void __foo_impl(int unused) { //implementation, new function.
... // the function body of the original foo.
}
void main() {
arg = compute();
call __foo_impl(arg);
}
This is my prototype here. do you think it's more feasible solution?
navyxliu@b73f537#diff-904855c22d662d8afbc11c40fe2906259836ba53e907e1cc899e6355358ec482
There was a problem hiding this comment.
Seems reasonable, but this needs to be callable from RemoveDeadValues itself (the pass can't crash itself here)
| RunLivenessAnalysis &la, DenseSet<Value> &nonLiveSet, | ||
| DenseSet<Value> &liveSet) { | ||
| if (!la.getSolverConfig().isInterprocedural()) | ||
| return; |
There was a problem hiding this comment.
That deserves a comment.
| return {}; | ||
| } | ||
|
|
||
| static void processCallOp(CallOpInterface callOp, Operation *module, |
There was a problem hiding this comment.
Can you document the API?
Comment on lines
+423
to
+425
| if (!funcOp || !funcOp.isPublic()) { | ||
| return; | ||
| } |
There was a problem hiding this comment.
Suggested change
| if (!funcOp || !funcOp.isPublic()) { | |
| return; | |
| } | |
| if (!funcOp || !funcOp.isPublic()) | |
| return; |
Nit: no trivial braces.
| return llvm::reverse(range); | ||
| } | ||
| } | ||
| }; |
There was a problem hiding this comment.
Which do you need to do this? Are you somehow trying to make an assumption about the order in which the regions are executed at runtime?
The order of the region on the op are not indicative of anything related to this.
There was a problem hiding this comment.
hi, @joker-eph,
Which do you need to do this?
I change iterator here. My intention is to visit basic block and CFG in reverse order.
module->walk<WalkOrder::PostOrder, BackwardIterator>
The order of the region on the op are not indicative of anything related to this.
TBH, this is the part of MLIR I don't understand. Should we also reverse Operation::getRegions()? yes, my understanding is that the order doesn't matter for regions.
There was a problem hiding this comment.
My intention is to visit basic block and CFG in reverse order.
Reverse order from what? The order of basic blocks isn't indicative of any execution order. The only constraint is on the first basic block to be the "entry" into the region.
Comment on lines
+165
to
+166
| continue; | ||
| } |
There was a problem hiding this comment.
Suggested change
| continue; | |
| } | |
| LDBG() << "Value " << value | |
| << " is already marked live at index " << index; | |
| continue; | |
| } |
Should we align the logging with above?
(otherwise remove the braces)
| return builder.create<arith::ConstantOp>(oldVal.getLoc(), type, zeroAttr); | ||
| } | ||
| return {}; | ||
| } |
There was a problem hiding this comment.
Can we split this out in a follow-up PR: keep this PR about fixing the bugs without introducing an "aggressive" optimization, and introduce the optimization on its own afterward?
This diff also tries to propagate liveness recursively.
You can test this locally with the following command:git-clang-format --diff origin/main HEAD --extensions cpp,h -- mlir/include/mlir/Analysis/DataFlow/LivenessAnalysis.h mlir/include/mlir/IR/Visitors.h mlir/lib/Transforms/RemoveDeadValues.cpp
View the diff from clang-format here.diff --git a/mlir/lib/Transforms/RemoveDeadValues.cpp b/mlir/lib/Transforms/RemoveDeadValues.cpp
index ba2c56e5d..ce2168668 100644
--- a/mlir/lib/Transforms/RemoveDeadValues.cpp
+++ b/mlir/lib/Transforms/RemoveDeadValues.cpp
@@ -402,10 +402,12 @@ static Value createDummyArgument(CallOpInterface callOp, Value oldVal) {
return {};
}
-// When you mark a call operand as live, also mark its definition chain, recursively.
-// We handle RegionBranchOpInterface here. I think we should handle BranchOpInterface as well.
+// When you mark a call operand as live, also mark its definition chain,
+// recursively. We handle RegionBranchOpInterface here. I think we should handle
+// BranchOpInterface as well.
void propagateBackward(Value val, DenseSet<Value> &liveSet) {
- if (liveSet.contains(val)) return;
+ if (liveSet.contains(val))
+ return;
liveSet.insert(val);
if (auto defOp = val.getDefiningOp()) {
@@ -417,7 +419,8 @@ void propagateBackward(Value val, DenseSet<Value> &liveSet) {
// Handle RegionBranchOpInterface specially
if (auto regionBranchOp = dyn_cast<RegionBranchOpInterface>(defOp)) {
// If this is a result of a RegionBranchOpInterface, we need to trace back
- // through the control flow to find the sources that contribute to this result
+ // through the control flow to find the sources that contribute to this
+ // result
OpResult result = cast<OpResult>(val);
unsigned resultIndex = result.getResultNumber();
@@ -425,11 +428,13 @@ void propagateBackward(Value val, DenseSet<Value> &liveSet) {
// Find all possible sources that can contribute to this result
// by examining all regions and their terminators
for (Region ®ion : regionBranchOp->getRegions()) {
- if (region.empty()) continue;
+ if (region.empty())
+ continue;
// Get the successors from this region
SmallVector<RegionSuccessor> successors;
- regionBranchOp.getSuccessorRegions(RegionBranchPoint(®ion), successors);
+ regionBranchOp.getSuccessorRegions(RegionBranchPoint(®ion),
+ successors);
// Check if any successor can produce this result
for (const RegionSuccessor &successor : successors) {
@@ -440,9 +445,10 @@ void propagateBackward(Value val, DenseSet<Value> &liveSet) {
// Find the terminator that contributes to this result
Operation *terminator = region.back().getTerminator();
if (auto regionBranchTerm =
- dyn_cast<RegionBranchTerminatorOpInterface>(terminator)) {
+ dyn_cast<RegionBranchTerminatorOpInterface>(terminator)) {
OperandRange terminatorOperands =
- regionBranchTerm.getSuccessorOperands(RegionBranchPoint::parent());
+ regionBranchTerm.getSuccessorOperands(
+ RegionBranchPoint::parent());
if (resultIndex < terminatorOperands.size()) {
// This terminator operand contributes to our result
propagateBackward(terminatorOperands[resultIndex], liveSet);
@@ -452,20 +458,23 @@ void propagateBackward(Value val, DenseSet<Value> &liveSet) {
}
}
- // Also mark region arguments as live if they might contribute to this result
- // Find which operand of the parent operation corresponds to region arguments
+ // Also mark region arguments as live if they might contribute to this
+ // result Find which operand of the parent operation corresponds to
+ // region arguments
Block &entryBlock = region.front();
for (BlockArgument arg : entryBlock.getArguments()) {
// Get entry successor operands - these are the operands that flow
// from the parent operation to this region
SmallVector<RegionSuccessor> entrySuccessors;
- regionBranchOp.getSuccessorRegions(RegionBranchPoint::parent(), entrySuccessors);
+ regionBranchOp.getSuccessorRegions(RegionBranchPoint::parent(),
+ entrySuccessors);
for (const RegionSuccessor &entrySuccessor : entrySuccessors) {
if (entrySuccessor.getSuccessor() == ®ion) {
// Get the operands that are forwarded to this region
OperandRange entryOperands =
- regionBranchOp.getEntrySuccessorOperands(RegionBranchPoint::parent());
+ regionBranchOp.getEntrySuccessorOperands(
+ RegionBranchPoint::parent());
unsigned argIndex = arg.getArgNumber();
if (argIndex < entryOperands.size()) {
propagateBackward(entryOperands[argIndex], liveSet);
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This diff is REDO of #160242.
Integrated feedbacks from the previous feedbacks and also try to substitute arguments with dummy values first.
Problem
Liveness analysis is inter-procedural. If there are some unused arguments in a public function, they propagate to callers. From the perspective of RemoveDeadValues, the signature of a public function is immutable. It can't cope with this situation. One side, it deletes outgoing arguments, on the other side it keeps the function intact.
Solution
We deploy two methods to fix this bug.
createDummyArgument replaces arguments with dummies. eg. If the type can create a value with zero initializer, such as 'T x{0}'.
As a fallback, we add another DenseSet called 'liveSet'. The initial values are the arguments of public functions. It propagates liveness backward just like Liveness analysis.
Test plan