This file has been truncated, but you can view the full file.
This file contains 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
Args: ./build/bin/torch-mlir-opt --torchscript-module-to-torch-backend-pipeline --torch-backend-to-linalg-on-tensors-backend-pipeline -debug -print-ir-after-all ../../misc/frontend.mlir | |
Load new dialect in Context builtin | |
Load new dialect in Context builtin | |
Load new dialect in Context torch | |
Load new dialect in Context std | |
Load new dialect in Context arith | |
Load new dialect in Context affine | |
Load new dialect in Context linalg | |
Load new dialect in Context math | |
Load new dialect in Context memref |
This file has been truncated, but you can view the full file.
This file contains 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
// -----// IR Dump Before Canonicalizer //----- // | |
func private @_forward(%arg0: tensor<1x7xi64>) -> tensor<?x2xf32> { | |
%c12_i64 = arith.constant 12 : i64 | |
%c7 = arith.constant 7 : index | |
%c30522_i64 = arith.constant 30522 : i64 | |
%c0_i64 = arith.constant 0 : i64 | |
%c1_i64 = arith.constant 1 : i64 | |
%cst = arith.constant 0.000000e+00 : f32 | |
%c384_i64 = arith.constant 384 : i64 | |
%c4_i64 = arith.constant 4 : i64 |
This file contains 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
// -----// IR Dump Before Canonicalizer //----- // | |
func private @_forward(%arg0: tensor<1x512xi64>) -> tensor<?x2xf32> { | |
%c12_i64 = arith.constant 12 : i64 | |
%c512 = arith.constant 512 : index | |
%c30522_i64 = arith.constant 30522 : i64 | |
%c0_i64 = arith.constant 0 : i64 | |
%c1_i64 = arith.constant 1 : i64 | |
%cst = arith.constant 0.000000e+00 : f32 | |
%c384_i64 = arith.constant 384 : i64 | |
%c4_i64 = arith.constant 4 : i64 |
This file contains 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
iree-translate: /home/vivek/work/02_07/iree-npcomp/external/iree/third_party/llvm-project/mlir/include/mlir/IR/UseDefLists.h:170: mlir::IRObjectWithUseList<mlir::OpOperand>::~IRObjectWithUseList() [OperandType = mlir::OpOperand]: Assertion `use_empty() && "Cannot destroy a value that still has uses!"' failed. | |
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace. | |
Stack dump: | |
0. Program arguments: ./build/iree/tools/iree-translate ../../temp.mlir -mlir-elide-elementsattrs-if-larger=16 --iree-hal-target-backends=dylib --iree-mlir-to-vm-bytecode-module -print-ir-after-all -mlir-disable-threading --iree-llvm-debug-symbols=false -o bert4.vmfb | |
#0 0x0000000006efc2f3 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/vivek/work/02_07/iree-npcomp/external/iree/third_party/llvm-project/llvm/lib/Support/Unix/Signals.inc:565:13 | |
#1 0x0000000006efa4b0 llvm::sys::RunSignalHandlers() /home/vivek/work/02_07/iree-npcomp/external/iree/third_party/llvm-project/llvm/li |
This file contains 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
bert_input_ir.mlir:1133:12: error: 'linalg.generic' op unexpected result less than 0 at expression #0 in (d0, d1, d2, d3) -> (d0, d3) | |
%404 = linalg.generic {indexing_maps = [#map2, #map1, #map2], iterator_types = ["parallel", "parallel", "parallel", "parallel"]} ins(%401, %96 : tensor<?x12x?x?xf32>, tensor<?x1x1x?xf32>) outs(%398 : tensor<?x12x?x?xf32>) { | |
^ | |
bert_input_ir.mlir:20:3: note: called from | |
func @forward(%arg0: tensor<1x512xi64>) -> tensor<?x2xf32> { | |
^ | |
bert_input_ir.mlir:1133:12: note: see current operation: %333 = "linalg.generic"(%332, %263, %330) ({ | |
^bb0(%arg1: f32, %arg2: f32, %arg3: f32): | |
%600 = "arith.truncf"(%9) : (f64) -> f32 |
This file contains 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
iree-translate: /home/vivek/work/02_07/vivekkhandelwal1-iree/third_party/llvm-project/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp:1690: static mlir::Type mlir::memref::SubViewOp::inferResultType(mlir::MemRefType, ArrayRef<int64_t>, ArrayRef<int64_t>, ArrayRef<int64_t>): Assertion `staticOffsets.size() == rank && "staticOffsets length mismatch"' failed. | |
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace. | |
Stack dump: | |
0. Program arguments: ./build/iree/tools/iree-translate bert_input_ir.mlir -mlir-elide-elementsattrs-if-larger=16 --iree-hal-target-backends=dylib --iree-mlir-to-vm-bytecode-module -print-ir-after-all -mlir-disable-threading --iree-llvm-debug-symbols=false -o bert4.vmfb | |
#0 0x0000000006ddcef3 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/vivek/work/02_07/vivekkhandelwal1-iree/third_party/llvm-project/llvm/lib/Support/Unix/Signals.inc:565:13 | |
#1 0x0000000006ddb0b0 llvm::sys::RunSignalHandlers() /home/vivek/work/02_07/vivekkhandelwal1-i |
This file contains 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
./build/bin/torch-mlir-opt --torchscript-module-to-torch-backend-pipeline --torch-backend-to-linalg-on-tensors-backend-pipeline /tmp/BincountModule.mlir | |
torch-mlir-opt: /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/external/llvm-project/llvm/include/llvm/ADT/SmallVector.h:273: llvm::SmallVectorTemplateCommon::reference llvm::SmallVectorTemplateCommon<llvm::SmallVector<long, 2>, void>::operator[](llvm::SmallVectorTemplateCommon::size_type) [T = llvm::SmallVector<long, 2>]: Assertion `idx < size()' failed. | |
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace. | |
Stack dump: | |
0. Program arguments: ./build/bin/torch-mlir-opt --torchscript-module-to-torch-backend-pipeline --torch-backend-to-linalg-on-tensors-backend-pipeline /tmp/BincountModule.mlir | |
#0 0x000000000345665a llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/external/llvm-project/llvm/lib/Support/Unix/Signals.inc:565:11 | |
#1 0x000000000345682b P |
This file contains 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
torch-mlir-opt: /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/external/llvm-project/llvm/include/llvm/ADT/ArrayRef.h:194: ArrayRef<T> llvm::ArrayRef<long>::slice(size_t, size_t) const [T = long]: Assertion `N+M <= size() && "Invalid specifier"' failed. | |
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace. | |
Stack dump: | |
0. Program arguments: ./build/bin/torch-mlir-opt --torchscript-module-to-torch-backend-pipeline --torch-backend-to-linalg-on-tensors-backend-pipeline /tmp/BincountModule.mlir | |
#0 0x000000000345669a llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/external/llvm-project/llvm/lib/Support/Unix/Signals.inc:565:11 | |
#1 0x000000000345686b PrintStackTraceSignalHandler(void*) /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/external/llvm-project/llvm/lib/Support/Unix/Signals.inc:632:1 | |
#2 0x0000000003454e45 llvm::sys::RunSignalHandlers() /home/vivek/work/02_07/vivekkhandelwal1-torch-mlir/exter |
This file contains 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
// -----// IR Dump After VerifyInvariantsBeforeBackendLowering //----- // | |
module attributes {torch.debug_module_name = "BincountModule"} { | |
func @forward(%arg0: !torch.vtensor<[?],si64>) -> !torch.vtensor<[?],si64> { | |
%none = torch.constant.none | |
%int0 = torch.constant.int 0 | |
%0 = torch.aten.bincount %arg0, %none, %int0 : !torch.vtensor<[?],si64>, !torch.none, !torch.int -> !torch.vtensor<[?],si64> | |
return %0 : !torch.vtensor<[?],si64> | |
} | |
} |
This file contains 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
Args: ./build/bin/torch-mlir-opt -pass-pipeline=builtin.func(scf-bufferize),builtin.func(linalg-bufferize),builtin.func(refback-munge-memref-copy),func-bufferize,arith-bufferize,builtin.func(tensor-bufferize),builtin.func(finalizing-bufferize),refback-munge-calling-conventions,refback-insert-rng-globals,builtin.func(convert-linalg-to-loops),builtin.func(lower-affine),convert-scf-to-cf,builtin.func(refback-expand-ops-for-llvm),builtin.func(arith-expand),builtin.func(convert-math-to-llvm),convert-linalg-to-llvm,convert-memref-to-llvm,builtin.func(convert-arith-to-llvm),convert-std-to-llvm,convert-cf-to-llvm,reconcile-unrealized-casts /tmp/BincountModule.mlir --print-ir-after-all -mlir-disable-threading --debug | |
Load new dialect in Context builtin | |
Load new dialect in Context builtin | |
Load new dialect in Context torch | |
Load new dialect in Context arith | |
Load new dialect in Context linalg | |
Load new dialect in Context affine | |
Load new dialect in Context math | |
Load new dialect in Context memref | |
Load new dialect in Context |
OlderNewer