[Tosa] Add tosa-to-linalg-pipeline for testing #69997
Merged
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
|
@llvm/pr-subscribers-mlir-linalg @llvm/pr-subscribers-mlir Author: Tai Ly (Tai78641) ChangesThis patch changes TosaValidation pass so that it works as either a pass on FuncOp or a pass on ModuleOp Tosa Variable checks are only enabled on ModuleOp because variable declarations may be outside of functions. Also added a pass on ModuleOp, --tosa-to-linalg-pipeline and a test, tosa-to-linalg-pipeline.mlir Full diff: https://github.com/llvm/llvm-project/pull/69997.diff 7 Files Affected:
diff --git a/mlir/include/mlir/Conversion/Passes.td b/mlir/include/mlir/Conversion/Passes.td
index 274784fe4a7b29c..f1df226d7058955 100644
--- a/mlir/include/mlir/Conversion/Passes.td
+++ b/mlir/include/mlir/Conversion/Passes.td
@@ -388,8 +388,8 @@ def ConvertFuncToLLVMPass : Pass<"convert-func-to-llvm", "ModuleOp"> {
already present in the IR will be kept as is.
An LLVM datalayout string can be attached as an attribute to the module on
- which the pass anchors. Such an attribute is attached by calling the
- set-module-datalayout pass. If present, an llvm::DataLayout object is
+ which the pass anchors. Such an attribute is attached by calling the
+ set-module-datalayout pass. If present, an llvm::DataLayout object is
created from this attribute and used in the conversion to LLVM.
#### Output IR
@@ -816,12 +816,12 @@ def ConvertMemRefToSPIRV : Pass<"convert-memref-to-spirv"> {
def ConvertNVVMToLLVMPass : Pass<"convert-nvvm-to-llvm"> {
let summary = "Convert NVVM to PTX with Inline Assembly in LLVM dialect";
let description = [{
- This pass generates PTX instructions using inline assembly for NVVM
+ This pass generates PTX instructions using inline assembly for NVVM
operations implements `BasicPtxBuilderInterface`.
}];
let dependentDialects = [
"NVVM::NVVMDialect",
- ];
+ ];
}
//===----------------------------------------------------------------------===//
@@ -1129,6 +1129,22 @@ def TosaToLinalgNamed
let constructor = "tosa::createTosaToLinalgNamed()";
}
+//===----------------------------------------------------------------------===//
+// TosaToLinalgPipeline
+//===----------------------------------------------------------------------===//
+
+def TosaToLinalgPipeline
+ : Pass<"tosa-to-linalg-pipeline", "ModuleOp"> {
+ let summary = "Lower TOSA to LinAlg on tensors and named operations with validation";
+ let description = [{
+ Pass that converts TOSA operations to the equivalent operations using the
+ tensor operations in LinAlg as well as LinAlg named operations.
+ This invokes addTosaToLinalgPasses pipeline to allow testing.
+ }];
+
+ let constructor = "tosa::createTosaToLinalgPipeline()";
+}
+
//===----------------------------------------------------------------------===//
// TosaToSCF
//===----------------------------------------------------------------------===//
diff --git a/mlir/include/mlir/Conversion/TosaToLinalg/TosaToLinalg.h b/mlir/include/mlir/Conversion/TosaToLinalg/TosaToLinalg.h
index c411010603ac61f..19906461892501f 100644
--- a/mlir/include/mlir/Conversion/TosaToLinalg/TosaToLinalg.h
+++ b/mlir/include/mlir/Conversion/TosaToLinalg/TosaToLinalg.h
@@ -27,6 +27,7 @@ namespace tosa {
std::unique_ptr<Pass> createTosaToLinalg();
std::unique_ptr<Pass> createTosaToLinalgNamed();
+std::unique_ptr<Pass> createTosaToLinalgPipeline();
/// Populates passes to convert from TOSA to Linalg on buffers. At the end of
/// the pass, the function will only contain linalg ops or standard ops if the
diff --git a/mlir/include/mlir/Dialect/Tosa/Transforms/Passes.td b/mlir/include/mlir/Dialect/Tosa/Transforms/Passes.td
index a0f670de20150fb..81932ba8b8dd38a 100644
--- a/mlir/include/mlir/Dialect/Tosa/Transforms/Passes.td
+++ b/mlir/include/mlir/Dialect/Tosa/Transforms/Passes.td
@@ -89,7 +89,7 @@ def TosaLevelType : I32EnumAttr<"TosaLevelEnum", "Tosa level",
let cppNamespace = "mlir::tosa";
}
-def TosaValidation : Pass<"tosa-validate", "mlir::ModuleOp"> {
+def TosaValidation : Pass<"tosa-validate"> {
let summary = "Validates TOSA dialect";
let description = [{
This pass validates if input TOSA operations match the specification for given
diff --git a/mlir/lib/Conversion/TosaToLinalg/CMakeLists.txt b/mlir/lib/Conversion/TosaToLinalg/CMakeLists.txt
index 4b79bf82810c58d..f35cbc9e8dcd1fd 100644
--- a/mlir/lib/Conversion/TosaToLinalg/CMakeLists.txt
+++ b/mlir/lib/Conversion/TosaToLinalg/CMakeLists.txt
@@ -3,6 +3,7 @@ add_mlir_conversion_library(MLIRTosaToLinalg
TosaToLinalgNamed.cpp
TosaToLinalgNamedPass.cpp
TosaToLinalgPass.cpp
+ TosaToLinalgPipeline.cpp
ADDITIONAL_HEADER_DIRS
${MLIR_MAIN_INCLUDE_DIR}/mlir/Dialect/Tosa
diff --git a/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgPipeline.cpp b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgPipeline.cpp
new file mode 100644
index 000000000000000..514011fc92accc0
--- /dev/null
+++ b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgPipeline.cpp
@@ -0,0 +1,65 @@
+//===- TosaToLinalgPipeline.cpp - Lowering Tosa to Linalg Dialect ---------===//
+//
+// 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
+//
+//===----------------------------------------------------------------------===//
+//
+// This transformation pass legalizes Tosa operations to the Linalg dialect.
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Conversion/TosaToLinalg/TosaToLinalg.h"
+
+#include "mlir/Dialect/Arith/IR/Arith.h"
+#include "mlir/Dialect/Func/IR/FuncOps.h"
+#include "mlir/Dialect/Linalg/IR/Linalg.h"
+#include "mlir/Dialect/Math/IR/Math.h"
+#include "mlir/Dialect/SCF/IR/SCF.h"
+#include "mlir/Dialect/Tensor/IR/Tensor.h"
+#include "mlir/Dialect/Tosa/IR/TosaOps.h"
+#include "mlir/Dialect/Tosa/Transforms/Passes.h"
+#include "mlir/Dialect/Tosa/Utils/QuantUtils.h"
+#include "mlir/IR/PatternMatch.h"
+#include "mlir/Pass/PassManager.h"
+#include "mlir/Transforms/DialectConversion.h"
+#include "mlir/Transforms/GreedyPatternRewriteDriver.h"
+#include "mlir/Transforms/Passes.h"
+
+namespace mlir {
+#define GEN_PASS_DEF_TOSATOLINALGPIPELINE
+#include "mlir/Conversion/Passes.h.inc"
+} // namespace mlir
+
+using namespace mlir;
+
+namespace {
+struct TosaToLinalgPipeline
+ : public impl::TosaToLinalgPipelineBase<TosaToLinalgPipeline> {
+public:
+ void getDependentDialects(DialectRegistry ®istry) const override {
+ registry
+ .insert<arith::ArithDialect, linalg::LinalgDialect, math::MathDialect,
+ tensor::TensorDialect, scf::SCFDialect>();
+ }
+
+ void runOnOperation() override {
+ OpPassManager pm("builtin.module");
+
+ tosa::addTosaToLinalgPasses(pm,
+ /* disableTosaDecompositions = */ false,
+ /* validationOptions = */
+ {tosa::TosaProfileEnum::BaseInference,
+ /* StrictOperationSpecAlignment = */ true,
+ tosa::TosaLevelEnum::EightK});
+
+ if (failed(runPipeline(pm, getOperation())))
+ signalPassFailure();
+ }
+};
+} // namespace
+
+std::unique_ptr<Pass> mlir::tosa::createTosaToLinalgPipeline() {
+ return std::make_unique<TosaToLinalgPipeline>();
+}
diff --git a/mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp b/mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp
index 424a31175d61707..88bf02205c689f4 100644
--- a/mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp
+++ b/mlir/lib/Dialect/Tosa/Transforms/TosaValidation.cpp
@@ -15,7 +15,6 @@
#include "mlir/Dialect/Tosa/Transforms/PassesEnums.cpp.inc"
#include <string>
-#include <unordered_map>
#include "mlir/Dialect/Func/IR/FuncOps.h"
#include "mlir/Dialect/Tosa/IR/TosaOps.h"
@@ -506,7 +505,9 @@ LogicalResult TosaValidation::applyVariableCheck(Operation *op) {
void TosaValidation::runOnOperation() {
configLevelAndProfile();
- getOperation().walk([&](Operation *op) {
+ Operation *topOp = getOperation();
+ const bool isModule = isa<ModuleOp>(topOp);
+ topOp->walk([&](Operation *op) {
for (Value operand : op->getOperands()) {
if ((profile == TosaProfileEnum::BaseInference) &&
isa<FloatType>(getElementTypeOrSelf(operand))) {
@@ -526,8 +527,8 @@ void TosaValidation::runOnOperation() {
if (failed(applyLevelCheck(op)))
signalPassFailure();
- // do variable type checks
- if (failed(applyVariableCheck(op)))
+ // do variable type checks iff topOp is a ModuleOp
+ if (isModule && failed(applyVariableCheck(op)))
signalPassFailure();
});
}
diff --git a/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-pipeline.mlir b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-pipeline.mlir
new file mode 100644
index 000000000000000..ff932af18926464
--- /dev/null
+++ b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-pipeline.mlir
@@ -0,0 +1,31 @@
+// RUN: mlir-opt %s --split-input-file --tosa-to-linalg-pipeline -verify-diagnostics
+
+
+// -----
+
+// check that -tosa-validate of stateful ops do not kick in
+func.func @test_variable_write_shape(%arg0: tensor<1x4x8xi32>) -> () {
+ // expected-error@+1 {{failed to legalize operation 'tosa.variable'}}
+ tosa.variable @stored_var = dense<-1> : tensor<2x4x8xi32>
+ tosa.variable.write @stored_var, %arg0 : tensor<1x4x8xi32>
+ return
+}
+
+// -----
+
+// check that --tosa-to-linalg kick in
+func.func @tensor_with_unknown_rank(%arg0: tensor<*xi8>) -> tensor<*xi8> {
+ // expected-error@+1 {{failed to legalize operation 'tosa.abs'}}
+ %0 = "tosa.abs"(%arg0) : (tensor<*xi8>) -> tensor<*xi8>
+ return %0 : tensor<*xi8>
+}
+
+// -----
+
+// check that --tosa-validate=strict-op-spec-alignment kick in
+func.func @test_avg_pool2d_zero_dim_input(%arg0: tensor<1x0x?x9xf32>) -> tensor<1x7x7x9xf32> {
+ // expected-error@+1 {{'tosa.avg_pool2d' op tensor has a dimension with size zero. Each dimension of a tensor must have size >= 1}}
+ %0 = "tosa.avg_pool2d"(%arg0) {acc_type = f32, kernel = array<i64: 2, 2>, pad = array<i64: 0, 1, 0, 1>, stride = array<i64: 1, 1>}
+ : (tensor<1x0x?x9xf32>) -> tensor<1x7x7x9xf32>
+ return %0 : tensor<1x7x7x9xf32>
+}
|
Tai78641
force-pushed
the
pr_tosa_linalg_pipeline
branch
from
0aa89c0 to
ea174e5
Compare
joker-eph
reviewed
Oct 24, 2023
| // do variable type checks | ||
| if (failed(applyVariableCheck(op))) | ||
| // do variable type checks iff topOp is a ModuleOp | ||
| if (isModule && failed(applyVariableCheck(op))) |
There was a problem hiding this comment.
That makes the validation behavior different depending on how it is scheduled: seems like something easy to misconfigure? What is the motivation to allow this "partial validation"?
There was a problem hiding this comment.
good point. reverted this change.
|
|
||
| if (failed(runPipeline(pm, getOperation()))) | ||
| signalPassFailure(); | ||
| } |
There was a problem hiding this comment.
Why isn't this just a registered pipeline instead of a pass? https://mlir.llvm.org/docs/PassManagement/#pass-pipeline-registration
There was a problem hiding this comment.
revised as a registered pipeline now
| @@ -90,7 +90,6 @@ void mlir::tosa::addTosaToLinalgPasses( | |||
| pm.addNestedPass<func::FuncOp>(tosa::createTosaLayerwiseConstantFoldPass( | |||
| {options.aggressiveReduceConstant})); | |||
| pm.addNestedPass<func::FuncOp>(tosa::createTosaMakeBroadcastablePass()); | |||
| pm.addNestedPass<mlir::ModuleOp>( | |||
| tosa::createTosaValidation(validationOptions)); | |||
| pm.addNestedPass<func::FuncOp>(tosa::createTosaValidation(validationOptions)); | |||
There was a problem hiding this comment.
Why not simply pm.addPass(tosa::createTosaValidation(validationOptions)); ?
Add tosa-to-linalg-pipeline that calls the function addTosaToLinalgPasses, so it gets tested in core also added tests in tosa-to-linalg-pipeline.mlir Signed-off-by: Tai Ly <[email protected]> Change-Id: Ie0fb6a09c7dd8d4bd5304e283810a5f65f55e912
Tai78641
force-pushed
the
pr_tosa_linalg_pipeline
branch
from
ea174e5 to
df7fda7
Compare
Tai78641
changed the title
joker-eph
approved these changes
Oct 25, 2023
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.
Add tosa-to-linalg-pipeline that calls the function
addTosaToLinalgPasses, so it gets tested in core
also added tests in tosa-to-linalg-pipeline.mlir