Files
llvm/mlir/lib/Transforms/ComposeAffineMaps.cpp

104 lines
3.6 KiB
C++
Raw Normal View History

//===- ComposeAffineMaps.cpp - MLIR Affine Transform Class-----*- C++ -*-===//
//
// Copyright 2019 The MLIR Authors.
//
// 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.
// =============================================================================
//
// This file implements a testing pass which composes affine maps from
// AffineApplyOps in a Function, by forward subtituting results from an
// AffineApplyOp into any of its users which are also AffineApplyOps.
//
//===----------------------------------------------------------------------===//
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
#include "mlir/Analysis/AffineAnalysis.h"
#include "mlir/Analysis/NestedMatcher.h"
#include "mlir/IR/AffineMap.h"
#include "mlir/IR/Attributes.h"
#include "mlir/IR/Builders.h"
#include "mlir/IR/BuiltinOps.h"
#include "mlir/IR/InstVisitor.h"
Introduce memref bound checking. Introduce analysis to check memref accesses (in MLFunctions) for out of bound ones. It works as follows: $ mlir-opt -memref-bound-check test/Transforms/memref-bound-check.mlir /tmp/single.mlir:10:12: error: 'load' op memref out of upper bound access along dimension tensorflow/mlir#1 %x = load %A[%idxtensorflow/mlir#0, %idxtensorflow/mlir#1] : memref<9 x 9 x i32> ^ /tmp/single.mlir:10:12: error: 'load' op memref out of lower bound access along dimension tensorflow/mlir#1 %x = load %A[%idxtensorflow/mlir#0, %idxtensorflow/mlir#1] : memref<9 x 9 x i32> ^ /tmp/single.mlir:10:12: error: 'load' op memref out of upper bound access along dimension tensorflow/mlir#2 %x = load %A[%idxtensorflow/mlir#0, %idxtensorflow/mlir#1] : memref<9 x 9 x i32> ^ /tmp/single.mlir:10:12: error: 'load' op memref out of lower bound access along dimension tensorflow/mlir#2 %x = load %A[%idxtensorflow/mlir#0, %idxtensorflow/mlir#1] : memref<9 x 9 x i32> ^ /tmp/single.mlir:12:12: error: 'load' op memref out of upper bound access along dimension tensorflow/mlir#1 %y = load %B[%idy] : memref<128 x i32> ^ /tmp/single.mlir:12:12: error: 'load' op memref out of lower bound access along dimension tensorflow/mlir#1 %y = load %B[%idy] : memref<128 x i32> ^ #map0 = (d0, d1) -> (d0, d1) #map1 = (d0, d1) -> (d0 * 128 - d1) mlfunc @test() { %0 = alloc() : memref<9x9xi32> %1 = alloc() : memref<128xi32> for %i0 = -1 to 9 { for %i1 = -1 to 9 { %2 = affine_apply #map0(%i0, %i1) %3 = load %0[%2tensorflow/mlir#0, %2tensorflow/mlir#1] : memref<9x9xi32> %4 = affine_apply #map1(%i0, %i1) %5 = load %1[%4] : memref<128xi32> } } return } - Improves productivity while manually / semi-automatically developing MLIR for testing / prototyping; also provides an indirect way to catch errors in transformations. - This pass is an easy way to test the underlying affine analysis machinery including low level routines. Some code (in getMemoryRegion()) borrowed from @andydavis cl/218263256. While on this: - create mlir/Analysis/Passes.h; move Pass.h up from mlir/Transforms/ to mlir/ - fix a bug in AffineAnalysis.cpp::toAffineExpr TODO: extend to non-constant loop bounds (straightforward). Will transparently work for all accesses once floordiv, mod, ceildiv are supported in the AffineMap -> FlatAffineConstraints conversion. PiperOrigin-RevId: 219397961
2018-10-30 17:43:06 -07:00
#include "mlir/Pass.h"
#include "mlir/StandardOps/StandardOps.h"
#include "mlir/Transforms/Passes.h"
#include "mlir/Transforms/Utils.h"
#include "llvm/Support/CommandLine.h"
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
#include "llvm/Support/raw_ostream.h"
using namespace mlir;
namespace {
// ComposeAffineMaps walks inst blocks in a Function, and for each
// AffineApplyOp, forward substitutes its results into any users which are
// also AffineApplyOps. After forward subtituting its results, AffineApplyOps
// with no remaining uses are collected and erased after the walk.
// TODO(andydavis) Remove this when Chris adds instruction combiner pass.
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
struct ComposeAffineMaps : public FunctionPass {
explicit ComposeAffineMaps() : FunctionPass(&ComposeAffineMaps::passID) {}
PassResult runOnFunction(Function *f) override;
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
// Thread-safe RAII contexts local to pass, BumpPtrAllocator freed on exit.
NestedPatternContext MLContext;
static char passID;
};
} // end anonymous namespace
char ComposeAffineMaps::passID = 0;
FunctionPass *mlir::createComposeAffineMapsPass() {
return new ComposeAffineMaps();
}
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
static bool affineApplyOp(const Instruction &inst) {
const auto &opInst = cast<OperationInst>(inst);
return opInst.isa<AffineApplyOp>();
}
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
PassResult ComposeAffineMaps::runOnFunction(Function *f) {
using matcher::Op;
auto pattern = Op(affineApplyOp);
auto apps = pattern.match(f);
for (auto m : apps) {
auto app = cast<OperationInst>(m.first)->cast<AffineApplyOp>();
SmallVector<Value *, 8> operands(app->getOperands());
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
FuncBuilder b(m.first);
auto newApp = makeComposedAffineApply(&b, app->getLoc(),
app->getAffineMap(), operands);
unsigned idx = 0;
for (auto *v : app->getResults()) {
v->replaceAllUsesWith(newApp->getResult(idx++));
}
}
Uniformize composition of AffineApplyOp by construction This CL is the 5th on the path to simplifying AffineMap composition. This removes the distinction between normalized single-result AffineMap and more general composed multi-result map. One nice byproduct of making the implementation driven by single-result is that the multi-result extension is a trivial change: the implementation is still single-result and we just use: ``` unsigned idx = getIndexOf(...); map.getResult(idx); ``` This CL also fixes an AffineNormalizer implementation issue related to symbols. Namely it stops performing substitutions on symbols in AffineNormalizer and instead concatenates them all to be consistent with the call to `AffineMap::compose(AffineMap)`. This latter call to `compose` cannot perform simplifications of symbols coming from different maps based on positions only: i.e. dims are applied and renumbered but symbols must be concatenated. The only way to determine whether symbols from different AffineApply are the same is to look at the concrete values. The canonicalizeMapAndOperands is thus extended with behavior to support replacing operands that appear multiple times. Lastly, this CL demonstrates that the implementation is correct by rewriting ComposeAffineMaps using only `makeComposedAffineApply`. The implementation uses a matcher because AffineApplyOp are introduced as composed operations on the fly instead of iteratively forwardSubstituting. For this purpose, a walker would revisit freshly introduced AffineApplyOp. Regardless, ComposeAffineMaps is scheduled to disappear, this CL replaces the implementation based on iterative `forwardSubstitute` by a composed-by-construction `makeComposedAffineApply`. Remaining calls to `forwardSubstitute` will be removed in the next CL. PiperOrigin-RevId: 228830443
2019-01-10 21:54:34 -08:00
{
auto pattern = Op(affineApplyOp);
auto apps = pattern.match(f);
std::reverse(apps.begin(), apps.end());
for (auto m : apps) {
auto app = cast<OperationInst>(m.first)->cast<AffineApplyOp>();
bool hasNonEmptyUse = llvm::any_of(
app->getResults(), [](Value *r) { return !r->use_empty(); });
if (!hasNonEmptyUse) {
m.first->erase();
}
}
}
return success();
}
static PassRegistration<ComposeAffineMaps> pass("compose-affine-maps",
"Compose affine maps");