trunk/09232df64e161efc316a66bdbc3755991a1f594a
[Testcase Refactoring] Add hw_classification in test/test_fx.py (#192…
[Testcase Refactoring] Add hw_classification in test/test_fx.py (#192…
Fixes #196631 Summary TorchInductor already optimizes index_add when the index is a sliced permutation: index = torch.randperm(x.shape[0], device=x.device)[: y.shape[0]] result = torch.index_add(x, dim=0, source=y, index=index) Because the indices are unique and in bounds, the accumulating update can be rewritten using unsafe indexing and a non-accumulating index_put. The full-permutation case: in…
On MPS, torch.linalg.lstsq raised IndexError: max(): Expected reduction dim 0 to have non-zero size whenever min(m, n) == 0. Root cause: the MPS kernel solves the system through an SVD and computes the rank cutoff from the largest singular value, S.max(-1). An empty matrix has no singular values, and max over an empty dimension has no identity value, so it raises (the same as on CPU). Fix: when mi…
Python 3.10 reaches EOL this month and was dropped from the nightly binary build matrix in #198233, but 13 of the 31 images in .ci/docker/build.sh were still pinned to it. Most did not say so: only two carry the version in their tag, so images like pytorch-linux-jammy-cuda13.2-cudnn9-py3-gcc11 -- 28 workflow references -- were quietly 3.10. Bumps the eleven whose tags do not encode a version, so n…
Summary: test_ordered_set.py imports CPython's test.support for two helpers, check_free_after_iterating and gc_collect. Not every Python distribution ships the test package, and where it is missing the whole module fails to import, so no test in it runs: ImportError: cannot import name 'support' from 'test' (unknown location) Copy check_free_after_iterating into the test file and call gc.collect()…
Safari rendering broke at some point with the trace plot overlapping. Fixed with some new divs. AI slop says: Safari resolves height: 100% on the timeline SVG grid items against the entire grid container instead of their assigned rows. This makes the main timeline and minimap overflow their tracks and overlap the stack trace panel. The stretched SVG also distorts the y-axis labels. Place each time…
Summary Fix the remaining base types.DynamicClassAttribute failure from the CPython 3.13 port. DynamicClassAttribute is a pure-Python descriptor. When one is created inside a compiled class body and later read raw through C.__dict__["attr"], Dynamo sends that ephemeral descriptor through SourcelessBuilder. The builder had no case for the exact DynamicClassAttribute type, so it graph-broke with "Un…
[Inductor] Skip ExternKernelSchedulerNode in create_foreach_nodes (#1…
[Inductor] Skip ExternKernelSchedulerNode in create_foreach_nodes (#1…
This PR is auto-generated nightly by this action. Update the pinned torchtitan hash. Pull Request resolved: #199668 Approved by: https://github.com/pytorchbot
[py3.14] Fix partial class attribute in inductor dynamic shapes tes…
[py3.14] Fix partial class attribute in inductor dynamic shapes tes…
[inductor][test] Require SM90 for the AOTI CUTLASS profiler test (#19…
[inductor] Wake the AsyncCompile pool for Triton-backed devices, not …
Summary: Pull Request resolved: #199318 Test PR to trigger CI Test Plan: CI Differential Revision: D120640754
simplify tests added
[aoti] Use the proxy executor for _quantized ops without a C shim (#1…
[aoti] Use the proxy executor for _quantized ops without a C shim (#1…
[aoti] Clone outputs that alias a constant through a fallback view op…
[aoti] Share extern kernel nodes between the constant-folding and mai…
[inductor][cpu] Cast vectorized bool in CppVecOverrides.masked (#1986…
update audio commit hash
update vllm commit hash
update torchtitan commit hash
update torchtitan commit hash
Summary: Pull Request resolved: #199318 Test PR to trigger CI Test Plan: CI Differential Revision: D120640754
Fixes #185684 Root cause any_user_may_alias in _is_fake_tensor_same references torch._inductor.fx_passes.reinplace._generalized_scatter by full attribute path, but fx_utils.py never imports that submodule. Its only fx_passes import is from torch._inductor.fx_passes.control_dependencies import control_deps. That import binds torch._inductor.fx_passes, but fx_passes/__init__.py is empty, so reinplac…
Fix shape inference symbol key mismatch with ShapeEnv global symbol n…
Fixes #150476 Adds focused unit test coverage for previously under-tested functionality in torch/_inductor/scheduler.py. The added tests cover: FusionResult immediate and deferred fusion decisions. PendingFusion state and fusion-node bookkeeping. NodeUser identity, equality, hashing, and conservative merging behavior. Nested reduction candidate selection and fusion legality checks. Classification…
Fixes #150476 Adds focused unit test coverage for previously under-tested functionality in torch/_inductor/scheduler.py. The added tests cover: FusionResult immediate and deferred fusion decisions. PendingFusion state and fusion-node bookkeeping. NodeUser identity, equality, hashing, and conservative merging behavior. Nested reduction candidate selection and fusion legality checks. Classification…