| Issue |
180310
|
| Summary |
mlir-opt crash: Unknown primary type in sparse_tensor.new during --sparse-tensor-conversion
|
| Labels |
new issue
|
| Assignees |
|
| Reporter |
xvv4402
|
**Bug Op:** sparse_tensor.new
**Bug Pass:** --sparse-tensor-conversion
**Detailed Description:** The mlir-opt tool hits a llvm_unreachable ("Unknown primary type") during the sparse tensor conversion pass. The crash occurs when sparse_tensor.new attempts to lower a tensor with an index element type. The utility function primaryTypeEncoding in the SparseTensor dialect does not recognize index as a valid primary type for encoding, leading to the crash during reader generation (genReader).
```mlir
module {
func.func @main() {
%0 = gpu.num_subgroups : index
%1 = smt.constant true
%2 = sparse_tensor.new %0 : index to tensor<?xindex, #sparse_tensor.encoding<{ map = (d0) -> (d0 : compressed) }>>
return
}
}
```
```
mlir-opt --sparse-tensor-conversion temp.mlir
```
```
Unknown primary type
UNREACHABLE executed at llvm-project-llvmorg-22.1.0-rc1/mlir/lib/Dialect/SparseTensor/Transforms/Utils/CodegenUtils.cpp:126!
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace and instructions to reproduce the bug.
Stack dump:
0. Program arguments: llvmorg-22.1.0-install/bin/mlir-opt --sparse-tensor-conversion temp.mlir
#0 0x0000605b9c185e32 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) (llvmorg-22.1.0-install/bin/mlir-opt+0x252fe32)
#1 0x0000605b9c1825af llvm::sys::RunSignalHandlers() (llvmorg-22.1.0-install/bin/mlir-opt+0x252c5af)
#2 0x0000605b9c1826fc SignalHandler(int, siginfo_t*, void*) Signals.cpp:0:0
#3 0x00007a3cbe845330 (/lib/x86_64-linux-gnu/libc.so.6+0x45330)
#4 0x00007a3cbe89eb2c __pthread_kill_implementation ./nptl/pthread_kill.c:44:76
#5 0x00007a3cbe89eb2c __pthread_kill_internal ./nptl/pthread_kill.c:78:10
#6 0x00007a3cbe89eb2c pthread_kill ./nptl/pthread_kill.c:89:10
#7 0x00007a3cbe84527e raise ./signal/../sysdeps/posix/raise.c:27:6
#8 0x00007a3cbe8288ff abort ./stdlib/abort.c:81:7
#9 0x0000605b9c16512e (llvmorg-22.1.0-install/bin/mlir-opt+0x250f12e)
#10 0x0000605b9ce10d96 mlir::sparse_tensor::primaryTypeEncoding(mlir::Type) (llvmorg-22.1.0-install/bin/mlir-opt+0x31bad96)
#11 0x0000605b9ce16e30 mlir::sparse_tensor::genReader(mlir::OpBuilder&, mlir::Location, mlir::sparse_tensor::SparseTensorType, mlir::Value, llvm::SmallVectorImpl<mlir::Value>&, mlir::Value&) (llvmorg-22.1.0-install/bin/mlir-opt+0x31c0e30)
#12 0x0000605b9ce8eb3c (anonymous namespace)::SparseTensorNewConverter::matchAndRewrite(mlir::sparse_tensor::NewOp, mlir::sparse_tensor::NewOpAdaptor, mlir::ConversionPatternRewriter&) const SparseTensorConversion.cpp:0:0
#13 0x0000605b9ce83352 llvm::LogicalResult mlir::ConversionPattern::dispatchTo1To1<mlir::OpConversionPattern<mlir::sparse_tensor::NewOp>, mlir::sparse_tensor::NewOp>(mlir::OpConversionPattern<mlir::sparse_tensor::NewOp> const&, mlir::sparse_tensor::NewOp, mlir::sparse_tensor::NewOp::GenericAdaptor<llvm::ArrayRef<mlir::ValueRange>>, mlir::ConversionPatternRewriter&) (llvmorg-22.1.0-install/bin/mlir-opt+0x322d352)
#14 0x0000605b9ce83474 mlir::OpConversionPattern<mlir::sparse_tensor::NewOp>::matchAndRewrite(mlir::Operation*, llvm::ArrayRef<mlir::ValueRange>, mlir::ConversionPatternRewriter&) const (llvmorg-22.1.0-install/bin/mlir-opt+0x322d474)
#15 0x0000605ba396332e mlir::ConversionPattern::matchAndRewrite(mlir::Operation*, mlir::PatternRewriter&) const (llvmorg-22.1.0-install/bin/mlir-opt+0x9d0d32e)
```
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
