[MLIR][SCF] Expose scf util functions in header file #138784
Closed
+25
−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
|
Thank you for submitting a Pull Request (PR) to the LLVM Project! This PR will be automatically labeled and the relevant teams will be notified. If you wish to, you can add reviewers by using the "Reviewers" section on this page. If this is not working for you, it is probably because you do not have write permissions for the repository. In which case you can instead tag reviewers by name in a comment by using If you have received no comments on your PR for a week, you can request a review by "ping"ing the PR by adding a comment “Ping”. The common courtesy "ping" rate is once a week. Please remember that you are asking for valuable time from other developers. If you have further questions, they may be answered by the LLVM GitHub User Guide. You can also ask questions in a comment on this PR, on the LLVM Discord or on the forums. |
|
@llvm/pr-subscribers-mlir-scf Author: Colin De Vlieghere (Cubevoid) ChangesThis patch exposes the Additionally, Full diff: https://github.com/llvm/llvm-project/pull/138784.diff 3 Files Affected:
diff --git a/mlir/include/mlir/Dialect/SCF/Utils/Utils.h b/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
index e620067c15be9..c7a3208a4176e 100644
--- a/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
+++ b/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
@@ -148,6 +148,23 @@ void denormalizeInductionVariable(RewriterBase &rewriter, Location loc,
Value normalizedIv, OpFoldResult origLb,
OpFoldResult origStep);
+/// For each original loop, the value of the
+/// induction variable can be obtained by dividing the induction variable of
+/// the linearized loop by the total number of iterations of the loops nested
+/// in it modulo the number of iterations in this loop (remove the values
+/// related to the outer loops):
+/// iv_i = floordiv(iv_linear, product-of-loop-ranges-until-i) mod range_i.
+/// Compute these iteratively from the innermost loop by creating a "running
+/// quotient" of division by the range.
+/// Returns the delinearized induction variables and the preserved users.
+std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
+delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
+ Value linearizedIv, ArrayRef<Value> ubs);
+
+/// Helper function to multiply a sequence of values.
+Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
+ ArrayRef<Value> values);
+
/// Tile a nest of standard for loops rooted at `rootForOp` by finding such
/// parametric tile sizes that the outer loops have a fixed number of iterations
/// as defined in `sizes`.
diff --git a/mlir/lib/Dialect/SCF/Utils/Utils.cpp b/mlir/lib/Dialect/SCF/Utils/Utils.cpp
index e9471c1dbd0b7..42a27e5ff783d 100644
--- a/mlir/lib/Dialect/SCF/Utils/Utils.cpp
+++ b/mlir/lib/Dialect/SCF/Utils/Utils.cpp
@@ -17,6 +17,7 @@
#include "mlir/Dialect/Arith/Utils/Utils.h"
#include "mlir/Dialect/Func/IR/FuncOps.h"
#include "mlir/Dialect/SCF/IR/SCF.h"
+#include "mlir/Dialect/Utils/StaticValueUtils.h"
#include "mlir/IR/BuiltinOps.h"
#include "mlir/IR/IRMapping.h"
#include "mlir/IR/OpDefinition.h"
@@ -807,7 +808,7 @@ void mlir::denormalizeInductionVariable(RewriterBase &rewriter, Location loc,
static OpFoldResult getProductOfIndexes(RewriterBase &rewriter, Location loc,
ArrayRef<OpFoldResult> values) {
- assert(!values.empty() && "unexecpted empty array");
+ assert(!values.empty() && "unexpected empty array");
AffineExpr s0, s1;
bindSymbols(rewriter.getContext(), s0, s1);
AffineExpr mul = s0 * s1;
@@ -819,9 +820,8 @@ static OpFoldResult getProductOfIndexes(RewriterBase &rewriter, Location loc,
return products;
}
-/// Helper function to multiply a sequence of values.
-static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
- ArrayRef<Value> values) {
+Value mlir::getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
+ ArrayRef<Value> values) {
assert(!values.empty() && "unexpected empty list");
if (getType(values.front()).isIndex()) {
SmallVector<OpFoldResult> ofrs = getAsOpFoldResult(values);
@@ -835,7 +835,7 @@ static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
continue;
if (productOf)
productOf =
- rewriter.create<arith::MulIOp>(loc, productOf.value(), v).getResult();
+ rewriter.createOrFold<arith::MulIOp>(loc, productOf.value(), v);
else
productOf = v;
}
@@ -848,17 +848,9 @@ static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
return productOf.value();
}
-/// For each original loop, the value of the
-/// induction variable can be obtained by dividing the induction variable of
-/// the linearized loop by the total number of iterations of the loops nested
-/// in it modulo the number of iterations in this loop (remove the values
-/// related to the outer loops):
-/// iv_i = floordiv(iv_linear, product-of-loop-ranges-until-i) mod range_i.
-/// Compute these iteratively from the innermost loop by creating a "running
-/// quotient" of division by the range.
-static std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
-delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
- Value linearizedIv, ArrayRef<Value> ubs) {
+std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
+mlir::delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
+ Value linearizedIv, ArrayRef<Value> ubs) {
if (linearizedIv.getType().isIndex()) {
Operation *delinearizedOp =
diff --git a/mlir/test/Dialect/SCF/transform-ops.mlir b/mlir/test/Dialect/SCF/transform-ops.mlir
index d9445182769e7..81462885dba4c 100644
--- a/mlir/test/Dialect/SCF/transform-ops.mlir
+++ b/mlir/test/Dialect/SCF/transform-ops.mlir
@@ -500,13 +500,11 @@ func.func @coalesce_i32_loops() {
%1 = arith.constant 128 : i32
%2 = arith.constant 2 : i32
%3 = arith.constant 64 : i32
- // CHECK: %[[VAL_4:.*]] = arith.constant 64 : i32
// CHECK: %[[ZERO:.*]] = arith.constant 0 : i32
// CHECK: %[[ONE:.*]] = arith.constant 1 : i32
- // CHECK: %[[VAL_7:.*]] = arith.constant 32 : i32
// CHECK: %[[VAL_8:.*]] = arith.constant 0 : i32
// CHECK: %[[VAL_9:.*]] = arith.constant 1 : i32
- // CHECK: %[[UB:.*]] = arith.muli %[[VAL_4]], %[[VAL_7]] : i32
+ // CHECK: %[[UB:.*]] = arith.constant 2048 : i32
// CHECK: scf.for %[[VAL_11:.*]] = %[[ZERO]] to %[[UB]] step %[[ONE]] : i32 {
scf.for %i = %0 to %1 step %2 : i32 {
scf.for %j = %0 to %3 step %2 : i32 {
|
|
@llvm/pr-subscribers-mlir Author: Colin De Vlieghere (Cubevoid) ChangesThis patch exposes the Additionally, Full diff: https://github.com/llvm/llvm-project/pull/138784.diff 3 Files Affected:
diff --git a/mlir/include/mlir/Dialect/SCF/Utils/Utils.h b/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
index e620067c15be9..c7a3208a4176e 100644
--- a/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
+++ b/mlir/include/mlir/Dialect/SCF/Utils/Utils.h
@@ -148,6 +148,23 @@ void denormalizeInductionVariable(RewriterBase &rewriter, Location loc,
Value normalizedIv, OpFoldResult origLb,
OpFoldResult origStep);
+/// For each original loop, the value of the
+/// induction variable can be obtained by dividing the induction variable of
+/// the linearized loop by the total number of iterations of the loops nested
+/// in it modulo the number of iterations in this loop (remove the values
+/// related to the outer loops):
+/// iv_i = floordiv(iv_linear, product-of-loop-ranges-until-i) mod range_i.
+/// Compute these iteratively from the innermost loop by creating a "running
+/// quotient" of division by the range.
+/// Returns the delinearized induction variables and the preserved users.
+std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
+delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
+ Value linearizedIv, ArrayRef<Value> ubs);
+
+/// Helper function to multiply a sequence of values.
+Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
+ ArrayRef<Value> values);
+
/// Tile a nest of standard for loops rooted at `rootForOp` by finding such
/// parametric tile sizes that the outer loops have a fixed number of iterations
/// as defined in `sizes`.
diff --git a/mlir/lib/Dialect/SCF/Utils/Utils.cpp b/mlir/lib/Dialect/SCF/Utils/Utils.cpp
index e9471c1dbd0b7..42a27e5ff783d 100644
--- a/mlir/lib/Dialect/SCF/Utils/Utils.cpp
+++ b/mlir/lib/Dialect/SCF/Utils/Utils.cpp
@@ -17,6 +17,7 @@
#include "mlir/Dialect/Arith/Utils/Utils.h"
#include "mlir/Dialect/Func/IR/FuncOps.h"
#include "mlir/Dialect/SCF/IR/SCF.h"
+#include "mlir/Dialect/Utils/StaticValueUtils.h"
#include "mlir/IR/BuiltinOps.h"
#include "mlir/IR/IRMapping.h"
#include "mlir/IR/OpDefinition.h"
@@ -807,7 +808,7 @@ void mlir::denormalizeInductionVariable(RewriterBase &rewriter, Location loc,
static OpFoldResult getProductOfIndexes(RewriterBase &rewriter, Location loc,
ArrayRef<OpFoldResult> values) {
- assert(!values.empty() && "unexecpted empty array");
+ assert(!values.empty() && "unexpected empty array");
AffineExpr s0, s1;
bindSymbols(rewriter.getContext(), s0, s1);
AffineExpr mul = s0 * s1;
@@ -819,9 +820,8 @@ static OpFoldResult getProductOfIndexes(RewriterBase &rewriter, Location loc,
return products;
}
-/// Helper function to multiply a sequence of values.
-static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
- ArrayRef<Value> values) {
+Value mlir::getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
+ ArrayRef<Value> values) {
assert(!values.empty() && "unexpected empty list");
if (getType(values.front()).isIndex()) {
SmallVector<OpFoldResult> ofrs = getAsOpFoldResult(values);
@@ -835,7 +835,7 @@ static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
continue;
if (productOf)
productOf =
- rewriter.create<arith::MulIOp>(loc, productOf.value(), v).getResult();
+ rewriter.createOrFold<arith::MulIOp>(loc, productOf.value(), v);
else
productOf = v;
}
@@ -848,17 +848,9 @@ static Value getProductOfIntsOrIndexes(RewriterBase &rewriter, Location loc,
return productOf.value();
}
-/// For each original loop, the value of the
-/// induction variable can be obtained by dividing the induction variable of
-/// the linearized loop by the total number of iterations of the loops nested
-/// in it modulo the number of iterations in this loop (remove the values
-/// related to the outer loops):
-/// iv_i = floordiv(iv_linear, product-of-loop-ranges-until-i) mod range_i.
-/// Compute these iteratively from the innermost loop by creating a "running
-/// quotient" of division by the range.
-static std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
-delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
- Value linearizedIv, ArrayRef<Value> ubs) {
+std::pair<SmallVector<Value>, SmallPtrSet<Operation *, 2>>
+mlir::delinearizeInductionVariable(RewriterBase &rewriter, Location loc,
+ Value linearizedIv, ArrayRef<Value> ubs) {
if (linearizedIv.getType().isIndex()) {
Operation *delinearizedOp =
diff --git a/mlir/test/Dialect/SCF/transform-ops.mlir b/mlir/test/Dialect/SCF/transform-ops.mlir
index d9445182769e7..81462885dba4c 100644
--- a/mlir/test/Dialect/SCF/transform-ops.mlir
+++ b/mlir/test/Dialect/SCF/transform-ops.mlir
@@ -500,13 +500,11 @@ func.func @coalesce_i32_loops() {
%1 = arith.constant 128 : i32
%2 = arith.constant 2 : i32
%3 = arith.constant 64 : i32
- // CHECK: %[[VAL_4:.*]] = arith.constant 64 : i32
// CHECK: %[[ZERO:.*]] = arith.constant 0 : i32
// CHECK: %[[ONE:.*]] = arith.constant 1 : i32
- // CHECK: %[[VAL_7:.*]] = arith.constant 32 : i32
// CHECK: %[[VAL_8:.*]] = arith.constant 0 : i32
// CHECK: %[[VAL_9:.*]] = arith.constant 1 : i32
- // CHECK: %[[UB:.*]] = arith.muli %[[VAL_4]], %[[VAL_7]] : i32
+ // CHECK: %[[UB:.*]] = arith.constant 2048 : i32
// CHECK: scf.for %[[VAL_11:.*]] = %[[ZERO]] to %[[UB]] step %[[ONE]] : i32 {
scf.for %i = %0 to %1 step %2 : i32 {
scf.for %j = %0 to %3 step %2 : i32 {
|
This patch exposes the `delinearizeInductionVariable` and `getProductOfIntsOrIndexes` helpers in the header file for the SCF utils, as these are useful for downstream users. Additionally, `getProductOfIntsOrIndexes` will now constant-fold the generated `arith::MulIOp`.
5d2e83b to
f0ecc8f
Compare
Good to know, thanks! |
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 patch exposes the
delinearizeInductionVariableandgetProductOfIntsOrIndexeshelpers in the header file for the SCF utils, as these are useful for downstream users.Additionally,
getProductOfIntsOrIndexeswill now constant-fold the generatedarith::MulIOp.