Update base for Update on "[slimtensor] Add SlimTensor class with basic properties and CPU copy operation"

**Key components:**

1. **`c10/core/Contiguity.h`** - Contiguity checking utility:
   - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order)

2. **`core/SlimTensor.h`** - Main SlimTensor class with:
   - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset
   - **Property accessors**:
     - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support
     - `strides()`, `stride(dim)` - get tensor strides with negative indexing support
     - `dtype()`, `device()`, `device_type()`, `device_index()`
     - `numel()`, `dim()`, `nbytes()`, `itemsize()`
     - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset)
     - `storage_offset()`, `storage()`
   - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()`
   - **Copy operation**: `copy_(other)` - copies data from another tensor
     - Fast path: uses memcpy for both-contiguous tensors
     - Slow path: element-wise copy respecting strides for non-contiguous tensors
   - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()`

**Curretnt constraints:**
- Only CPU device supported
- Only Float32 dtype tested
- copy_() only supports CPU-to-CPU copy

Those contraints will be further improved in the following diffs

Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/)

[ghstack-poisoned]

Author: Gasoonjia

.github/workflows/add-unanswered-to-project.yml

@@ -2,14 +2,20 @@ name: Add Open External Contributor PRs and Issues to PyTorch Org Project 136
 
 on:
   workflow_dispatch:
-  pull_request: 
-   paths: 
-     .github/workflows/add-unanswered-to-project.yml
+  schedule:
+    # GitHub Actions cron uses UTC. These run at:
+    #  - 14:00 UTC  -> 08:00 CST (UTC-6)
+    #  - 19:00 UTC  -> 13:00 CST (UTC-6)
+    - cron: "0 14 * * *"
+    - cron: "0 19 * * *"
+  pull_request:
+    paths:
+      - .github/workflows/add-unanswered-to-project.yml
 jobs:
   add_to_project:
     runs-on: ubuntu-latest
     steps:
-      - name: Add open issues and open, non-draft PRs to org project (excluding certain authors)
+      - name: Add open issues and open, non-draft PRs to org project (excluding certain authors and bots)
         uses: actions/github-script@v7
         with:
           github-token: ${{ secrets.ET_EXT_CONTRIB }}
@@ -41,13 +47,26 @@ jobs:
               "app/dependabot", "Erik-Lundell", "zingo", "AdrianLundell", "oscarandersson8218", "per", "Sebastian-Larsson", "SaoirseARM", 
               "robell", "mansnils", "martinlsm", "freddan80", "YufengShi-dudu", "tom-arm", "perheld", "Jerry-Ge", "gggekov", "fumchin", 
               "wwwind", "benkli01", "Tessil", "maddun01", "Michiel-Olieslagers", "armwaheed", "agrima1304", "emmakujala", "annietllnd", 
-              "MatthiasHertel80", "AlexTawseArm", "jmahbs", "morgolock", "Christoffer-JL", "ArmRyan", "xingguo01", "haowhsu-quic", 
-              "shewu-quic", "winskuo-quic", "chunit-quic", "DannyYuyang-quic", "chuntl", "thchenqti", "jethroqti", "chenweng-quic", 
-              "cymbalrush", "DenisVieriu97", "billmguo", "StrycekSimon", "jirioc", "robert-kalmar", "skywall", "MartinPavella", 
-              "roman-janik-nxp", "novak-vaclav ", "neuropilot-captain", "dijopaul", "cad-rlc", "cad-audio", "ynimmaga", "daniil-lyakhov", 
-              "emmanuel-ferdman", "cavusmustafa", "anzr299", "Jiseong-oh", "alexdean08"
+              "MatthiasHertel80", "AlexTawseArm", "jmahbs", "morgolock", "Christoffer-JL", "ArmRyan", "xingguo01", "tgonzalezorlandoarm", 
+              "haowhsu-quic", "shewu-quic", "winskuo-quic", "chunit-quic", "DannyYuyang-quic", "chuntl", "thchenqti", "jethroqti", 
+              "chenweng-quic", "cymbalrush", "DenisVieriu97", "billmguo", "StrycekSimon", "jirioc", "robert-kalmar", "skywall", 
+              "MartinPavella", "roman-janik-nxp", "novak-vaclav", "neuropilot-captain", "dijopaul", "cad-rlc", "cad-audio", "ynimmaga", 
+              "daniil-lyakhov", "emmanuel-ferdman", "cavusmustafa", "anzr299", "Jiseong-oh", "alexdean08",
+              // explicitly include the dependabot bot login seen in PRs
+              "dependabot[bot]"
             ]);
 
+            function isBotOrExcluded(user) {
+              if (!user) return false;
+              // GitHub sometimes marks bots with user.type === "Bot"
+              if (user.type && user.type.toLowerCase() === "bot") return true;
+              // Some bots use logins that end with [bot], e.g. dependabot[bot]
+              if (user.login && user.login.endsWith("[bot]")) return true;
+              // Explicit excluded list
+              if (excludedAuthors.has(user.login)) return true;
+              return false;
+            }
+
             async function addItem(contentId, type, number) {
               try {
                 await github.graphql(`
@@ -69,7 +88,7 @@ jobs:
             }
 
             try {
-              // Add open issues (not PRs) and exclude by author
+              // Add open issues (not PRs) and exclude by author/bots
               const issues = await github.paginate(
                 github.rest.issues.listForRepo,
                 {
@@ -80,12 +99,14 @@ jobs:
                 }
               );
               for (const issue of issues) {
-                if (!issue.pull_request && !excludedAuthors.has(issue.user.login)) {
+                if (!issue.pull_request && !isBotOrExcluded(issue.user)) {
                   await addItem(issue.node_id, 'issue', issue.number);
+                } else {
+                  console.log(`Skipping issue #${issue.number} by ${issue.user && issue.user.login}`);
                 }
               }
 
-              // Add open, non-draft PRs (regardless of review state), exclude by author
+              // Add open, non-draft PRs (regardless of review state), exclude by author/bots
               const prs = await github.paginate(
                 github.rest.pulls.list,
                 {
@@ -95,8 +116,10 @@ jobs:
                 }
               );
               for (const pr of prs) {
-                if (!pr.draft && !excludedAuthors.has(pr.user.login)) {
+                if (!pr.draft && !isBotOrExcluded(pr.user)) {
                   await addItem(pr.node_id, 'pr', pr.number);
+                } else {
+                  console.log(`Skipping PR #${pr.number} by ${pr.user && pr.user.login}`);
                 }
               }
             } catch (error) {

.github/workflows/cuda.yml

@@ -87,8 +87,8 @@ jobs:
         export LD_LIBRARY_PATH=/opt/conda/lib:$LD_LIBRARY_PATH
         PYTHON_EXECUTABLE=python source .ci/scripts/test_model.sh "${{ matrix.model }}" cmake cuda
 
-  test-cuda-shims:
-    name: test-cuda-shims
+  unittest-cuda:
+    name: unittest-cuda
     uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@main
     permissions:
       id-token: write
@@ -103,17 +103,20 @@ jobs:
       ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
       script: |
         set -eux
-        # Install requirements
-        bash ./install_requirements.sh
+        # Install executorch in editable mode so custom op libs land in-tree
+        bash ./install_executorch.sh
 
         # Build ExecuTorch with CUDA support
         cmake --workflow --preset llm-release-cuda
 
-        # Build and run CUDA shim tests
+        # Build and run CUDA shim tests (C++)
         pushd backends/cuda/runtime/shims/tests
         cmake --workflow --preset default
         popd
 
+        # Run CUDA backend Python tests, overrides addopts so that we don't run all tests in pytest.ini
+        python -m pytest backends/cuda/tests backends/cuda/passes/tests -v -o "addopts="
+
   export-model-cuda-artifact:
     name: export-model-cuda-artifact
     # Skip this job if the pull request is from a fork (HuggingFace secrets are not available)

backends/aoti/aoti_backend.py

@@ -156,7 +156,10 @@ def preprocess(
         # Apply custom backend-specific passes
         custom_passes = cls.get_custom_passes(compile_specs)
         for custom_pass in custom_passes:
-            custom_pass(device_edge_program.graph_module)
+            if getattr(custom_pass, "requires_exported_program", False):
+                custom_pass(device_edge_program)
+            else:
+                custom_pass(device_edge_program.graph_module)
 
         # Run decompositions if any
         if decomposition_table:

backends/aoti/slim/core/test/targets.bzl

@@ -4,9 +4,9 @@ def define_common_targets():
     """Define test targets for SlimTensor core module."""
 
     runtime.cxx_test(
-        name = "test_storage_cpu",
+        name = "test_storage",
         srcs = [
-            "test_storage_cpu.cpp",
+            "test_storage.cpp",
         ],
         deps = [
             "//executorch/backends/aoti/slim/core:storage",

…aoti/slim/core/test/test_storage_cpu.cpp …nds/aoti/slim/core/test/test_storage.cppbackends/aoti/slim/core/test/test_storage_cpu.cpp renamed to backends/aoti/slim/core/test/test_storage.cpp backends/aoti/slim/core/test/test_storage_cpu.cpp renamed to backends/aoti/slim/core/test/test_storage.cpp

@@ -113,6 +113,9 @@
 from .replace_scalar_with_tensor_pass import (  # noqa
     ReplaceScalarWithTensorByProfilePass,
 )
+from .rewrite_bool_bitwise_not_to_logical_not_pass import (  # noqa
+    RewriteBoolBitwiseNotToLogicalNotPass,
+)
 from .rewrite_bool_to_fp32_cast_via_int8_pass import (  # noqa
     RewriteBoolToFp32CastViaInt8Pass,
 )

backends/arm/_passes/__init__.py

@@ -103,6 +103,7 @@
     RemoveNoopPass,
     ReplaceInfAndLimitValuesPass,
     ReplaceScalarWithTensorByProfilePass,
+    RewriteBoolBitwiseNotToLogicalNotPass,
     RewriteBoolToFp32CastViaInt8Pass,
     RewriteConvPass,
     RewriteMatmulPass,
@@ -222,6 +223,7 @@ def _tosa_pipeline(
         self.add_passes(
             [
                 FuseQuantizedActivationPass(),
+                RewriteBoolBitwiseNotToLogicalNotPass(),
                 RewriteBoolToFp32CastViaInt8Pass(),
                 ConvertToClampPass(),
                 DecomposeTOSAUnsupportedClampPass(),

backends/arm/_passes/arm_pass_manager.py

@@ -0,0 +1,43 @@
+# Copyright 2025 Arm Limited and/or its affiliates.
+#
+# This source code is licensed under the BSD-style license found in the
+# LICENSE file in the root directory of this source tree.
+
+
+from typing import Set, Type
+
+import torch
+from executorch.backends.arm._passes import ArmPass
+from executorch.exir.dialects._ops import ops as exir_ops
+from executorch.exir.pass_base import ExportPass
+
+
+class RewriteBoolBitwiseNotToLogicalNotPass(ArmPass):
+    """
+    Rewrites ``aten.bitwise_not`` on boolean tensors to ``aten.logical_not``.
+
+    TOSA ``bitwise_not`` does not support boolean inputs. On boolean tensors,
+    ``bitwise_not`` is equivalent to ``logical_not``, so this rewrite preserves
+    semantics while enabling lowering.
+    """
+
+    _passes_required_after: Set[Type[ExportPass]] = set()
+
+    _TARGET_OPS = {
+        exir_ops.edge.aten.bitwise_not.default,
+    }
+
+    def call_operator(self, op, args, kwargs, meta):
+        if op not in self._TARGET_OPS:
+            return super().call_operator(op, args, kwargs, meta)
+
+        if meta["val"].dtype == torch.bool:
+            x = args[0]
+            return super().call_operator(
+                exir_ops.edge.aten.logical_not.default,
+                (x,),
+                {},
+                meta,
+            )
+
+        return super().call_operator(op, args, kwargs, meta)

backends/arm/_passes/rewrite_bool_bitwise_not_to_logical_not_pass.py

@@ -77,6 +77,8 @@ def __init__(
         if "u55" in target_lower:
             # Add the Ethos-U55 extension marker
             base_tosa_version += "+u55"
+        if "u85" in self.target:
+            base_tosa_version += "+cf"
         tosa_spec = TosaSpecification.create_from_string(base_tosa_version)
 
         self._set_compile_specs(tosa_spec, compiler_flags)

backends/arm/ethosu/compile_spec.py

@@ -3,5 +3,5 @@
 # This source code is licensed under the BSD-style license found in the
 # LICENSE file in the root directory of this source tree.
 
-ethos-u-vela == 4.4.1
+ethos-u-vela == 4.5.0
 pte-adapter-model-explorer == 0.0.2