This is an archive of the discontinued LLVM Phabricator instance.

Differential D137981

[mlir][sparse] avoid nop rewriting on runtime lib path in pipeline
ClosedPublic

Authored by aartbik on Nov 14 2022, 1:54 PM.

Details

Reviewers
nicolasvasilache
Peiming
bixia
anlunx
wrengr
Commits
rG910d6270dba7: [mlir][sparse] avoid nop rewriting on runtime lib path in pipeline

Diff Detail

Event Timeline

aartbik created this revision.Nov 14 2022, 1:54 PM
Herald added a project: Restricted Project. · View Herald TranscriptNov 14 2022, 1:54 PM
Herald added subscribers: Moerafaat, anlunx, bzcheeseman and 20 others. · View Herald Transcript
aartbik requested review of this revision.Nov 14 2022, 1:54 PM
Herald added a reviewer: nicolasvasilache. · View Herald TranscriptNov 14 2022, 1:54 PM
Herald added a project: Restricted Project. · View Herald Transcript
Herald added subscribers: stephenneuendorffer, nicolasvasilache. · View Herald Transcript
aartbik added reviewers: Peiming, bixia, anlunx, wrengr.Nov 14 2022, 1:55 PM
Peiming accepted this revision.Nov 14 2022, 1:55 PM
This revision is now accepted and ready to land.Nov 14 2022, 1:55 PM
Harbormaster completed remote builds in B197615: Diff 475267.Nov 14 2022, 2:19 PM
This revision was landed with ongoing or failed builds.Nov 14 2022, 4:39 PM
Closed by commit rG910d6270dba7: [mlir][sparse] avoid nop rewriting on runtime lib path in pipeline (authored by aartbik). · Explain Why
This revision was automatically updated to reflect the committed changes.
aartbik added a commit: rG910d6270dba7: [mlir][sparse] avoid nop rewriting on runtime lib path in pipeline.

Revision Contents

PathSize
mlir/
lib/
Dialect/
SparseTensor/
Pipelines/
8 lines

Diff 475308

mlir/lib/Dialect/SparseTensor/Pipelines/SparseTensorPipelines.cpp

Show First 20 Lines • Show All 53 Lines • ▼ Show 20 Linesvoid mlir::sparse_tensor::buildSparseCompiler(
pm.addNestedPass<func::FuncOp>(createLinalgGeneralizationPass()); pm.addNestedPass<func::FuncOp>(createLinalgGeneralizationPass());
pm.addPass( pm.addPass(
bufferization::createTensorCopyInsertionPass(getBufferizationOptions( bufferization::createTensorCopyInsertionPass(getBufferizationOptions(
/*analysisOnly=*/options.testBufferizationAnalysisOnly))); /*analysisOnly=*/options.testBufferizationAnalysisOnly)));
if (options.testBufferizationAnalysisOnly) if (options.testBufferizationAnalysisOnly)
return; return;
pm.addPass(createSparseTensorRewritePass(options.enableRuntimeLibrary)); pm.addPass(createSparseTensorRewritePass(options.enableRuntimeLibrary));
pm.addPass(createSparsificationPass(options.sparsificationOptions())); pm.addPass(createSparsificationPass(options.sparsificationOptions()));
if (options.enableRuntimeLibrary) if (options.enableRuntimeLibrary) {
pm.addPass(createSparseTensorConversionPass( pm.addPass(createSparseTensorConversionPass(
options.sparseTensorConversionOptions())); options.sparseTensorConversionOptions()));
else } else {
pm.addPass( pm.addPass(
createSparseTensorCodegenPass(options.enableBufferInitialization)); createSparseTensorCodegenPass(options.enableBufferInitialization));
pm.addPass(createSparseBufferRewritePass(options.enableBufferInitialization)); pm.addPass(
createSparseBufferRewritePass(options.enableBufferInitialization));
}
pm.addPass(createDenseBufferizationPass( pm.addPass(createDenseBufferizationPass(
getBufferizationOptions(/*analysisOnly=*/false))); getBufferizationOptions(/*analysisOnly=*/false)));
pm.addNestedPass<func::FuncOp>(createCanonicalizerPass()); pm.addNestedPass<func::FuncOp>(createCanonicalizerPass());
pm.addNestedPass<func::FuncOp>( pm.addNestedPass<func::FuncOp>(
mlir::bufferization::createFinalizingBufferizePass()); mlir::bufferization::createFinalizingBufferizePass());
// TODO(springerm): Add sparse support to the BufferDeallocation pass and add // TODO(springerm): Add sparse support to the BufferDeallocation pass and add
// it to this pipeline. // it to this pipeline.
pm.addNestedPass<func::FuncOp>(createConvertLinalgToLoopsPass()); pm.addNestedPass<func::FuncOp>(createConvertLinalgToLoopsPass());
Show All 27 Lines