Skip to content

Commit 9c70676

Browse files
vivekkhandelwal1vivekkhandelwal1
authored
build: manually update PyTorch version (#3727)
Set PyTorch and TorchVision version to nightly release 2024-10-15. Tracker issue for the failing tests added to xfail_set in this PR. Issue: #3796 This commit disables the failing sparse tensor tests since they are not maintained on day-to-day basis and blocks the roll PyTorch update for now. Signed-Off By: Vivek Khandelwal <[email protected]>
1 parent dc7a1ff commit 9c70676

File tree

11 files changed

+232
-191
lines changed

11 files changed

+232
-191
lines changed

‎include/torch-mlir/Dialect/Torch/IR/GeneratedTorchOps.td‎

Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6319,6 +6319,30 @@ def Torch_AtenDotOp : Torch_Op<"aten.dot", [
63196319
let hasCanonicalizer = 1;
63206320
}
63216321

6322+
def Torch_AtenOuterOp : Torch_Op<"aten.outer", [
6323+
AllowsTypeRefinement,
6324+
HasValueSemantics,
6325+
ReadOnly
6326+
]> {
6327+
let summary = "Generated op for `aten::outer : (Tensor, Tensor) -> (Tensor)`";
6328+
let arguments = (ins
6329+
AnyTorchTensorType:$self,
6330+
AnyTorchTensorType:$vec2
6331+
);
6332+
let results = (outs
6333+
AnyTorchOptionalTensorType:$result
6334+
);
6335+
let hasCustomAssemblyFormat = 1;
6336+
let extraClassDefinition = [{
6337+
ParseResult AtenOuterOp::parse(OpAsmParser &parser, OperationState &result) {
6338+
return parseDefaultTorchOp(parser, result, 2, 1);
6339+
}
6340+
void AtenOuterOp::print(OpAsmPrinter &printer) {
6341+
printDefaultTorchOp(printer, *this, 2, 1);
6342+
}
6343+
}];
6344+
}
6345+
63226346
def Torch_AtenCosineSimilarityOp : Torch_Op<"aten.cosine_similarity", [
63236347
AllowsTypeRefinement,
63246348
HasValueSemantics,

‎lib/Dialect/Torch/Transforms/AbstractInterpLibrary.cpp‎

Lines changed: 24 additions & 42 deletions
Original file line numberDiff line numberDiff line change
@@ -7601,6 +7601,13 @@ StringRef mlir::torch::Torch::getAbstractInterpLibrary() {
76017601
" } : (!torch.int, !torch.bool) -> ()\n"
76027602
" return %0 : !torch.list<int>\n"
76037603
" }\n"
7604+
" func.func @\"__torch_mlir_shape_fn.aten.outer\"(%arg0: !torch.list<int>, %arg1: !torch.list<int>) -> !torch.list<int> {\n"
7605+
" %int0 = torch.constant.int 0\n"
7606+
" %0 = torch.aten.__getitem__.t %arg0, %int0 : !torch.list<int>, !torch.int -> !torch.int\n"
7607+
" %1 = torch.aten.__getitem__.t %arg1, %int0 : !torch.list<int>, !torch.int -> !torch.int\n"
7608+
" %2 = torch.prim.ListConstruct %0, %1 : (!torch.int, !torch.int) -> !torch.list<int>\n"
7609+
" return %2 : !torch.list<int>\n"
7610+
" }\n"
76047611
" func.func @\"__torch_mlir_shape_fn.aten.dot\"(%arg0: !torch.list<int>, %arg1: !torch.list<int>) -> !torch.list<int> {\n"
76057612
" %0 = torch.prim.ListConstruct : () -> !torch.list<int>\n"
76067613
" return %0 : !torch.list<int>\n"
@@ -13403,6 +13410,14 @@ StringRef mlir::torch::Torch::getAbstractInterpLibrary() {
1340313410
" %4 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%2, %3) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
1340413411
" return %4 : !torch.int\n"
1340513412
" }\n"
13413+
" func.func @\"__torch_mlir_dtype_fn.aten.outer\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.tuple<int, int>) -> !torch.int {\n"
13414+
" %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
13415+
" %1:2 = torch.prim.TupleUnpack %arg1 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
13416+
" %2 = torch.prim.ListConstruct %0#0, %1#0 : (!torch.int, !torch.int) -> !torch.list<optional<int>>\n"
13417+
" %3 = torch.prim.ListConstruct %0#1, %1#1 : (!torch.int, !torch.int) -> !torch.list<int>\n"
13418+
" %4 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%2, %3) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13419+
" return %4 : !torch.int\n"
13420+
" }\n"
1340613421
" func.func @\"__torch_mlir_dtype_fn.aten.mm\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.tuple<int, int>) -> !torch.int {\n"
1340713422
" %false = torch.constant.bool false\n"
1340813423
" %int5 = torch.constant.int 5\n"
@@ -13813,63 +13828,30 @@ StringRef mlir::torch::Torch::getAbstractInterpLibrary() {
1381313828
" return %5 : !torch.int\n"
1381413829
" }\n"
1381513830
" func.func @\"__torch_mlir_dtype_fn.aten.lerp.Scalar\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.tuple<int, int>, %arg2: !torch.number) -> !torch.int {\n"
13816-
" %none = torch.constant.none\n"
1381713831
" %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1381813832
" %1:2 = torch.prim.TupleUnpack %arg1 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
13819-
" %2 = torch.prim.ListConstruct %0#0, %1#0, %none : (!torch.int, !torch.int, !torch.none) -> !torch.list<optional<int>>\n"
13820-
" %3 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.get_dtype_of_scalar(%arg2) : (!torch.number) -> !torch.int\n"
13821-
" %4 = torch.prim.ListConstruct %0#1, %1#1, %3 : (!torch.int, !torch.int, !torch.int) -> !torch.list<int>\n"
13822-
" %5 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%2, %4) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13823-
" return %5 : !torch.int\n"
13833+
" %2 = torch.prim.ListConstruct %0#0, %1#0 : (!torch.int, !torch.int) -> !torch.list<optional<int>>\n"
13834+
" %3 = torch.prim.ListConstruct %0#1, %1#1 : (!torch.int, !torch.int) -> !torch.list<int>\n"
13835+
" %4 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%2, %3) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13836+
" return %4 : !torch.int\n"
1382413837
" }\n"
1382513838
" func.func @\"__torch_mlir_dtype_fn.aten.addcmul\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.tuple<int, int>, %arg2: !torch.tuple<int, int>, %arg3: !torch.number) -> !torch.int {\n"
13826-
" %none = torch.constant.none\n"
13827-
" %str = torch.constant.str \"AssertionError: \"\n"
13828-
" %int11 = torch.constant.int 11\n"
1382913839
" %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1383013840
" %1:2 = torch.prim.TupleUnpack %arg1 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1383113841
" %2:2 = torch.prim.TupleUnpack %arg2 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
13832-
" %3 = torch.aten.ne.int %0#1, %int11 : !torch.int, !torch.int -> !torch.bool\n"
13833-
" torch.prim.If %3 -> () {\n"
13834-
" torch.prim.If.yield\n"
13835-
" } else {\n"
13836-
" torch.prim.RaiseException %str, %none : !torch.str, !torch.none\n"
13837-
" torch.prim.If.yield\n"
13838-
" }\n"
13839-
" %4 = torch.aten.ne.int %1#1, %int11 : !torch.int, !torch.int -> !torch.bool\n"
13840-
" torch.prim.If %4 -> () {\n"
13841-
" torch.prim.If.yield\n"
13842-
" } else {\n"
13843-
" torch.prim.RaiseException %str, %none : !torch.str, !torch.none\n"
13844-
" torch.prim.If.yield\n"
13845-
" }\n"
13846-
" %5 = torch.aten.ne.int %2#1, %int11 : !torch.int, !torch.int -> !torch.bool\n"
13847-
" torch.prim.If %5 -> () {\n"
13848-
" torch.prim.If.yield\n"
13849-
" } else {\n"
13850-
" torch.prim.RaiseException %str, %none : !torch.str, !torch.none\n"
13851-
" torch.prim.If.yield\n"
13852-
" }\n"
13853-
" %6 = torch.prim.ListConstruct %0#0, %1#0, %2#0 : (!torch.int, !torch.int, !torch.int) -> !torch.list<optional<int>>\n"
13854-
" %7 = torch.prim.ListConstruct %0#1, %1#1, %2#1 : (!torch.int, !torch.int, !torch.int) -> !torch.list<int>\n"
13855-
" %8 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%6, %7) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13856-
" return %8 : !torch.int\n"
13842+
" %3 = torch.prim.ListConstruct %0#0, %1#0, %2#0 : (!torch.int, !torch.int, !torch.int) -> !torch.list<optional<int>>\n"
13843+
" %4 = torch.prim.ListConstruct %0#1, %1#1, %2#1 : (!torch.int, !torch.int, !torch.int) -> !torch.list<int>\n"
13844+
" %5 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%3, %4) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13845+
" return %5 : !torch.int\n"
1385713846
" }\n"
1385813847
" func.func @\"__torch_mlir_dtype_fn.aten.addcdiv\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.tuple<int, int>, %arg2: !torch.tuple<int, int>, %arg3: !torch.number) -> !torch.int {\n"
13859-
" %int6 = torch.constant.int 6\n"
1386013848
" %0:2 = torch.prim.TupleUnpack %arg0 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1386113849
" %1:2 = torch.prim.TupleUnpack %arg1 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1386213850
" %2:2 = torch.prim.TupleUnpack %arg2 : !torch.tuple<int, int> -> !torch.int, !torch.int\n"
1386313851
" %3 = torch.prim.ListConstruct %0#0, %1#0, %2#0 : (!torch.int, !torch.int, !torch.int) -> !torch.list<optional<int>>\n"
1386413852
" %4 = torch.prim.ListConstruct %0#1, %1#1, %2#1 : (!torch.int, !torch.int, !torch.int) -> !torch.list<int>\n"
1386513853
" %5 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.promote_dtypes(%3, %4) : (!torch.list<optional<int>>, !torch.list<int>) -> !torch.int\n"
13866-
" %6 = call @__torch__.torch_mlir.jit_ir_importer.build_tools.library_generator.is_integer_dtype(%5) : (!torch.int) -> !torch.bool\n"
13867-
" %7 = torch.prim.If %6 -> (!torch.int) {\n"
13868-
" torch.prim.If.yield %int6 : !torch.int\n"
13869-
" } else {\n"
13870-
" torch.prim.If.yield %5 : !torch.int\n"
13871-
" }\n"
13872-
" return %7 : !torch.int\n"
13854+
" return %5 : !torch.int\n"
1387313855
" }\n"
1387413856
" func.func @\"__torch_mlir_dtype_fn.aten.add.Scalar\"(%arg0: !torch.tuple<int, int>, %arg1: !torch.number, %arg2: !torch.number) -> !torch.int {\n"
1387513857
" %none = torch.constant.none\n"

‎projects/pt1/e2e_testing/xfail_sets.py‎

Lines changed: 74 additions & 27 deletions
Original file line numberDiff line numberDiff line change
@@ -442,10 +442,6 @@
442442
"ElementwiseDequantizePerTensorModule_basic",
443443
"ElementwiseQuantizePerTensorModule_basic",
444444
"ElementwiseQuantizePerTensorUIntModule_basic",
445-
"ElementwiseRreluEvalModule_basic",
446-
"ElementwiseRreluEvalStaticModule_basic",
447-
"ElementwiseRreluTrainModule_basic",
448-
"ElementwiseRreluTrainStaticModule_basic",
449445
"ElementwiseToDtypeI64ToUI8Module_basic",
450446
"EqIntModule_basic",
451447
"FloatImplicitModule_basic",
@@ -487,9 +483,6 @@
487483
"ReduceMinAlongDimUnsignedInt_basic",
488484
"RsubInt0d_NumToTensor_Module_basic",
489485
"ScalarImplicitFloatModule_basic",
490-
"SignAndLogarithmOfDeterminantModule_F32",
491-
"SignAndLogarithmOfDeterminantBatchedModule_F32",
492-
"SignAndLogarithmOfDeterminantDynamicModule_F32",
493486
"SortIntListReverse_basic",
494487
"SortIntList_basic",
495488
"SplitDimDynamicModule_basic",
@@ -519,13 +512,42 @@
519512
"SplitTensorNegativeDimModule_basic",
520513
"SplitWithSizesListUnpackModule_basic",
521514
"SplitWithSizes_Module_basic",
515+
"AdaptiveAvgPool1dGeneralDynamic_basic",
516+
"AdaptiveAvgPool1dStaticEvenMultiple_basic",
517+
"AdaptiveAvgPool1dStaticLargerOutput_basic",
518+
"AdaptiveAvgPool2dDynamicNoBatch_basic",
519+
"AdaptiveAvgPool2dDynamic_basic",
520+
"AdaptiveMaxPool1dDynamicNoBatch_basic",
521+
"AdaptiveMaxPool1dDynamic_basic",
522+
"AdaptiveMaxPool1dStatic_basic",
523+
"CrossEntropyLossModule_basic",
524+
"CrossEntropyLossNoReductionModule_basic",
525+
"ElementwiseExpm1IntModule_basic",
526+
"ElementwiseExpm1Module_basic",
527+
"IndexPutImpl1DFloatAccumulateModule_basic",
528+
"IndexPutImpl1DFloatNonAccumulateModule_basic",
529+
"IndexPutImpl1DIntAccumulateModule_basic",
530+
"IndexPutImpl1DIntNonAccumulateModule_basic",
531+
"IndexPutImpl2DFloatNonAccumulateModule_basic",
532+
"IndexPutImpl2DImplicitModule_basic",
533+
"IndexPutImpl2DIndexModule_basic",
534+
"IndexPutImpl2DNoneIndexStaticModule_basic",
535+
"IndexPutImpl3DFloatNonAccumulateModule_basic",
536+
"IndexPutImplIndexWithNoneModule_basic",
537+
"InterpolateDynamicModule_sizes_nearest",
538+
"IouOfModule_basic",
539+
"MeshgridIndexingIJ_basic",
540+
"MeshgridIndexingXY_basic",
541+
"Meshgrid_basic",
542+
"OneHotModule_basic",
522543
}
523544

524545
FX_IMPORTER_CRASHING_SET = LINALG_CRASHING_SET | {
525546
"HBC_basic",
526547
# Runtime op verification: out-of-bounds access
527548
"_SoftmaxModule_basic",
528549
"UpSampleNearest2dDynamicFactor_basic",
550+
"AdaptiveAvgPool1dGeneralDynamicNoBatches_basic",
529551
}
530552

531553
FX_IMPORTER_STABLEHLO_XFAIL_SET = {
@@ -554,10 +576,6 @@
554576
"ElementwiseRemainderTensorModule_Int_Float_NegativeDivisor_basic",
555577
"ElementwiseRemainderTensorModule_Int_NegativeDividend_basic",
556578
"ElementwiseRemainderTensorModule_Int_NegativeDivisor_basic",
557-
"ElementwiseRreluEvalModule_basic",
558-
"ElementwiseRreluEvalStaticModule_basic",
559-
"ElementwiseRreluTrainModule_basic",
560-
"ElementwiseRreluTrainStaticModule_basic",
561579
"MaxPool1dCeilModeTrueModule_basic",
562580
"MaxPool1dStaticCeilModeTrueModule_basic",
563581
"MaxUnpool3dModulePad0_basic",
@@ -591,7 +609,6 @@
591609
"AdaptiveAvgPool3dDynamic_basic",
592610
"AdaptiveMaxPool1dDynamicNoBatch_basic",
593611
"AdaptiveMaxPool1dDynamic_basic",
594-
"AdaptiveMaxPool1dDimOneStatic_basic",
595612
"AdaptiveMaxPool1dStatic_basic",
596613
"AdaptiveMaxPool2dDynamicNoBatch_basic",
597614
"AdaptiveMaxPool2dDynamicWithIndices_basic",
@@ -758,12 +775,7 @@
758775
"MaxPool2dWithIndicesBackwardStatic3DModule_basic",
759776
"MaxPool2dWithIndicesBackwardStatic4DModule_basic",
760777
"MaxPool3dCeilModeTrueModule_basic",
761-
"MaxPool3dEmptyStrideStaticModule_basic",
762-
"MaxPool3dLargeDatadModule_basic",
763-
"MaxPool3dModuleRandomSimple_basic",
764-
"MaxPool3dModule_basic",
765778
"MaxPool3dStaticCeilModeTrueModule_basic",
766-
"MaxPool3dStaticModule_basic",
767779
"MaxPool3dWithIndicesAllNegativeValuesModule_basic",
768780
"MaxPool3dWithIndicesAllOnesModule_basic",
769781
"MaxPool3dWithIndicesCeilModeTrueModule_basic",
@@ -921,6 +933,51 @@
921933
"Unfold_Module_Rank_Zero_basic",
922934
"Unfold_Module_Rank_Zero_Size_Zero_basic",
923935
"Unfold_Module_Dynamic_basic",
936+
"AdaptiveAvgPool1dGeneralDynamic_basic",
937+
"AdaptiveAvgPool1dGeneralDynamicNoBatches_basic",
938+
"AdaptiveAvgPool1dStaticEvenMultiple_basic",
939+
"AdaptiveAvgPool1dStaticLargerOutput_basic",
940+
"AdaptiveAvgPool2dDynamicNoBatch_basic",
941+
"AdaptiveAvgPool2dDynamic_basic",
942+
"AddIntModule_basic",
943+
"AtenIntTensorByteDtypeModule_basic",
944+
"AtenIntTensorCharDtypeModule_basic",
945+
"AtenItemIntOpModule_basic",
946+
"CrossEntropyLossModule_basic",
947+
"CrossEntropyLossNoReductionModule_basic",
948+
"EinsumStaticContractRhsModule_basic",
949+
"EinsumStaticFourDimensionModule_basic",
950+
"EinsumStaticModule_basic",
951+
"EinsumStaticWithEllipsisSlicingAndBroadcastModule_basic",
952+
"EinsumStaticWithEllipsisSlicingModule_basic",
953+
"ElementwiseExpm1IntModule_basic",
954+
"ElementwiseExpm1Module_basic",
955+
"InterpolateDynamicModule_sizes_nearest",
956+
"IouOfModule_basic",
957+
"IscloseStaticModuleTrue_basic",
958+
"IscloseStaticModule_basic",
959+
"MeshgridIndexingIJ_basic",
960+
"MeshgridIndexingXY_basic",
961+
"Meshgrid_basic",
962+
"MulIntModule_basic",
963+
"OneHotModule_basic",
964+
"ReduceFrobeniusNormComplexModule_basic",
965+
"ScalarImplicitIntModule_basic",
966+
"ScaledDotProductAttentionBoolMaskModule_basic",
967+
"ScaledDotProductAttentionDifferentCausalModule_basic",
968+
"ScaledDotProductAttentionDifferentDynamicCausalModule_basic",
969+
"ScaledDotProductAttentionDifferentModule_basic",
970+
"ScaledDotProductAttentionMaskModule_basic",
971+
"ScaledDotProductAttentionSameCausalModule_basic",
972+
"ScaledDotProductAttentionSameDynamicModule_basic",
973+
"ScaledDotProductAttentionSameModule_basic",
974+
"SubIntModule_basic",
975+
"TensorToIntZeroRank_basic",
976+
"UpSampleNearest2dDynamicFactor_basic",
977+
"UpSampleNearest2dDynamicSize_basic",
978+
"UpSampleNearest2dStaticFactor_basic",
979+
"UpSampleNearest2dStaticSize_basic",
980+
"UpSampleNearest2d_basic",
924981
}
925982

926983
FX_IMPORTER_STABLEHLO_CRASHING_SET = {
@@ -3297,7 +3354,6 @@
32973354
"SplitWithSizesListUnpackModule_basic",
32983355
"SplitWithSizes_Module_basic",
32993356
"ElementwiseCreateComplexModule_basic",
3300-
"AdaptiveMaxPool1dDimOneStatic_basic",
33013357
"AtenPolarDoubleModule_basic",
33023358
"AtenPolarFloatModule_basic",
33033359
"HstackBasicComplexModule_basic",
@@ -3318,10 +3374,6 @@
33183374
"Conv_Transpose3dStaticModule_basic",
33193375
"ElementwiseFloatTensorGtIntTensorModule_basic",
33203376
"ElementwiseIntTensorLtFloatTensorModule_basic",
3321-
"ElementwiseRreluEvalModule_basic",
3322-
"ElementwiseRreluEvalStaticModule_basic",
3323-
"ElementwiseRreluTrainModule_basic",
3324-
"ElementwiseRreluTrainStaticModule_basic",
33253377
"IndexPutWithNoneAndBroadcastModule_basic",
33263378
"MaskedScatterStaticBasic_basic",
33273379
"MaxUnpool3dModulePad0_basic",
@@ -3628,12 +3680,7 @@
36283680
"MaxPool2dWithIndicesNonDefaultStrideModule_basic",
36293681
"MaxPool2dWithIndicesStaticModule_basic",
36303682
"MaxPool3dCeilModeTrueModule_basic",
3631-
"MaxPool3dEmptyStrideStaticModule_basic",
3632-
"MaxPool3dLargeDatadModule_basic",
3633-
"MaxPool3dModuleRandomSimple_basic",
3634-
"MaxPool3dModule_basic",
36353683
"MaxPool3dStaticCeilModeTrueModule_basic",
3636-
"MaxPool3dStaticModule_basic",
36373684
"MaxPool3dWithIndicesAllNegativeValuesModule_basic",
36383685
"MaxPool3dWithIndicesAllOnesModule_basic",
36393686
"MaxPool3dWithIndicesCeilModeTrueModule_basic",

0 commit comments

Comments
 (0)