Merge remote-tracking branch 'upstream/main' into merge-main

.github/workflows/RollPyTorch.yml

@@ -79,6 +79,7 @@ jobs:
         TORCH_MLIR_SRC_PYTORCH_BRANCH="${{ env.PT_HASH }}" \
         TORCH_MLIR_SRC_PYTORCH_RELEASE="${{ env.PT_RELEASE }}" \
         TM_UPDATE_ODS_AND_ABSTRACT_INTERP_LIB="ON" \
+        TM_PYTHON_VERSIONS="cp311-cp311" \
         ./build_tools/python_deploy/build_linux_packages.sh
 
     - name: Post issue comment on build failure

.github/workflows/buildAndTest.yml

@@ -63,6 +63,7 @@ jobs:
       uses: actions/checkout@v3
       with:
         submodules: 'true'
+        fetch-depth: 0
 
     - name: Fetch PyTorch commit hash
       if: ${{ matrix.os-arch != 'windows-x86_64' }}

.github/workflows/buildRelease.yml

@@ -36,6 +36,8 @@ jobs:
       uses: actions/checkout@v3
       with:
         submodules: 'true'
+        fetch-depth: 0
+
     - uses: ./.github/actions/setup-build
       with:
         cache-suffix: 'release'

README.md

@@ -12,6 +12,7 @@ PyTorch is an open source machine learning framework that facilitates the seamle
 
 [MLIR](https://mlir.llvm.org)
 The MLIR project offers a novel approach for building extensible and reusable compiler architectures, which address the issue of software fragmentation, reduce the cost of developing domain-specific compilers, improve compilation for heterogeneous hardware, and promote compatibility between existing compilers.
+
 [Torch-MLIR](https://github.com/llvm/torch-mlir)
 Several vendors have adopted MLIR as the middle layer in their systems, enabling them to map frameworks such as PyTorch, JAX, and TensorFlow into MLIR and subsequently lower them to their target hardware. We have observed half a dozen custom lowerings from PyTorch to MLIR, making it easier for hardware vendors to focus on their unique value, rather than needing to implement yet another PyTorch frontend for MLIR. The ultimate aim is to be similar to the current hardware vendors adding LLVM target support, rather than each one implementing Clang or a C++ frontend.
 

docs/long_term_roadmap.md

@@ -246,3 +246,19 @@ for current LTC-based toolchains onto TorchDynamo. This migration will improve
 the end-user experience since TorchDynamo is more seamless, but it is a
 end-user-impacting migration nonetheless and we will want to phase it
 appropriately with the community.
+
+### End-to-end (E2E) testing
+
+Torch-MLIR currently maintains its own test suite with
+[hundreds of end-to-end tests](https://github.com/llvm/torch-mlir/tree/main/python/torch_mlir_e2e_test/test_suite)
+that verify the correctness and completeness of our op lowerings.
+These tests are tedious to write, and also sometimes hit corners
+of PyTorch's API that aren't usually reachable by user code.
+PyTorch already has an [end-to-end op test suite](https://github.com/pytorch/pytorch/blob/ead51864622467acd6835b6da86a166c1a32aa55/torch/testing/_internal/common_methods_invocations.py#L1)
+and we should just plug into it. Here is [an example](https://github.com/pytorch/pytorch/blob/ead51864622467acd6835b6da86a166c1a32aa55/test/test_proxy_tensor.py#L1573) of doing so.
+Even better, it would be great if TorchDynamo/PyTorch 2.0
+directly provided a way to plug into this.
+
+Additionally, we can leverage the [`pytorch-jit-paritybench`](https://github.com/jansel/pytorch-jit-paritybench)
+to verify our end-to-end correctness on real models.
+

e2e_testing/main.py

@@ -27,7 +27,14 @@
 from torch_mlir_e2e_test.stablehlo_backends.linalg_on_tensors import LinalgOnTensorsStablehloBackend
 from torch_mlir_e2e_test.tosa_backends.linalg_on_tensors import LinalgOnTensorsTosaBackend
 
-from .xfail_sets import LINALG_XFAIL_SET, STABLEHLO_PASS_SET, TOSA_PASS_SET, LTC_XFAIL_SET, TORCHDYNAMO_XFAIL_SET
+from .xfail_sets import (
+    LINALG_XFAIL_SET,
+    STABLEHLO_PASS_SET,
+    TOSA_PASS_SET,
+    LTC_XFAIL_SET,
+    TORCHDYNAMO_XFAIL_SET,
+    TORCHDYNAMO_CRASHING_SET
+)
 
 # Import tests to register them in the global registry.
 from torch_mlir_e2e_test.test_suite import register_all_tests
@@ -77,26 +84,33 @@ def main():
     if args.config == "linalg":
         config = LinalgOnTensorsBackendTestConfig(RefBackendLinalgOnTensorsBackend())
         xfail_set = LINALG_XFAIL_SET
-    if args.config == "tosa":
+        crashing_set = set()
+    elif args.config == "tosa":
         config = TosaBackendTestConfig(LinalgOnTensorsTosaBackend())
         xfail_set = all_test_unique_names - TOSA_PASS_SET
-    if args.config == "stablehlo":
+        crashing_set = set()
+    elif args.config == "stablehlo":
         config = StablehloBackendTestConfig(LinalgOnTensorsStablehloBackend())
         xfail_set = all_test_unique_names - STABLEHLO_PASS_SET
+        crashing_set = set()
     elif args.config == "native_torch":
         config = NativeTorchTestConfig()
-        xfail_set = {}
+        xfail_set = set()
+        crashing_set = set()
     elif args.config == "torchscript":
         config = TorchScriptTestConfig()
-        xfail_set = {}
+        xfail_set = set()
+        crashing_set = set()
     elif args.config == "lazy_tensor_core":
         config = LazyTensorCoreTestConfig()
         xfail_set = LTC_XFAIL_SET
+        crashing_set = set()
     elif args.config == "torchdynamo":
         config = TorchDynamoTestConfig()
         xfail_set = TORCHDYNAMO_XFAIL_SET
+        crashing_set = TORCHDYNAMO_CRASHING_SET
 
-    do_not_attempt = set(args.crashing_tests_to_not_attempt_to_run_and_a_bug_is_filed or [])
+    do_not_attempt = set(args.crashing_tests_to_not_attempt_to_run_and_a_bug_is_filed or []).union(crashing_set)
     available_tests = [test for test in GLOBAL_TEST_REGISTRY if test.unique_name not in do_not_attempt]
     if args.crashing_tests_to_not_attempt_to_run_and_a_bug_is_filed is not None:
         for arg in args.crashing_tests_to_not_attempt_to_run_and_a_bug_is_filed: