PyTorch MergeBot 60fa9d8ad0 Revert "[Inductor] Fix assertion failure in efficient_conv_bn_eval for default args (#169126)" ...

This reverts commit 27163a02b6.

Reverted https://github.com/pytorch/pytorch/pull/169126 on behalf of https://github.com/pytorch-auto-revert due to Reverted automatically by pytorch's autorevert, to avoid this behaviour add the tag autorevert: disable ([comment](https://github.com/pytorch/pytorch/pull/169126#issuecomment-3684907851))

2025-12-23 03:10:47 +00:00

..

analysis

[XPU][Fix] Register convolution_overrideable for flops count (#166839)

2025-11-04 05:56:29 +00:00

autoheuristic

Remove useless super() delegation (#168235)

2025-11-21 09:08:45 +00:00

codegen

Fix use_compute_types parameter name in CPP backend to_dtype (#168125)

2025-12-22 04:43:29 +00:00

compile_worker

[4/N] Use context managers (#169169)

2025-12-03 00:52:10 +00:00

fx_passes

Revert "[Inductor] Fix assertion failure in efficient_conv_bn_eval for default args (#169126)"

2025-12-23 03:10:47 +00:00

kernel

[Inductor][Triton][FP8] Support (BlockWise128x128, BlockWise1x128) scaling in Inductor (#170748)

2025-12-22 17:21:21 +00:00

lookup_table

…

package

…

runtime

[Inductor] Support appending TLX-only CUDA options in triton template (#171117)

2025-12-23 00:51:09 +00:00

template_heuristics

Add Inductor mm template in TLX (#170410)

2025-12-20 20:48:17 +00:00

__autotune_main__.py

…

__init__.py

Fix aot_compile typing. (#168320)

2025-11-21 23:18:37 +00:00

analyze_preserves_zero_mask.py

…

aoti_eager.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

async_compile.py

[RFC] Add experimental Pallas TorchInductor backend (#166822)

2025-11-05 00:52:41 +00:00

augmented_graph_helper.py

[1/N] Use key in dict for existence checks (#167035)

2025-11-06 02:25:10 +00:00

autotune_process.py

Avoid closing random file handles in Inductor (#169065)

2025-12-19 23:31:11 +00:00

await_utils.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

bounds.py

[1/N] Use key in dict for existence checks (#167035)

2025-11-06 02:25:10 +00:00

cache.py

…

choices.py

Upgrade pyrefly version (#169633)

2025-12-12 08:33:22 +00:00

codecache.py

Revert "[Inductor XPU GEMM] Step 1/N: Refactor cutlass configuration. (#160174)"

2025-12-12 00:13:49 +00:00

comm_analysis.py

[Inductor] Add Blackwell specs to comm analysis (#169817)

2025-12-12 07:37:07 +00:00

comm_lowering.py

Add GroupName NewType (#167552)

2025-12-05 16:33:35 +00:00

comms_debug.py

…

comms.py

Upgrade pyrefly version (#169633)

2025-12-12 08:33:22 +00:00

compile_fx_async.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

compile_fx_ext.py

[inductor, 3.14] catch pickle.PicklingError exceptions (#167383)

2025-11-13 21:47:14 +00:00

compile_fx_subproc.py

…

compile_fx.py

[inductor] Fix cudagraph skip for index_put_ with boolean indices, graph partitioning logic (#170103)

2025-12-18 22:43:04 +00:00

compiler_bisector.py

inductor compiler bisector run mode (#170717)

2025-12-18 20:22:08 +00:00

config_comms.py

[inductor][1/N] optimize reorder/sink passes compile time (#169370)

2025-12-10 16:24:03 +00:00

config.py

Add Inductor mm template in TLX (#170410)

2025-12-20 20:48:17 +00:00

constant_folding.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

cpp_builder.py

[Inductor][2/2] Decouple flags for optimization and debug symbols (#167575)

2025-11-13 00:59:15 +00:00

cpu_vec_isa.py

Upgrade pyrefly version (#169633)

2025-12-12 08:33:22 +00:00

cudagraph_trees.py

Cudagraph trees add helpful code for debuggin (#170208)

2025-12-12 18:42:35 +00:00

cudagraph_utils.py

Cudagraph trees add helpful code for debuggin (#170208)

2025-12-12 18:42:35 +00:00

custom_graph_pass.py

Fix incorrect homogeneous tuple types (#169463)

2025-12-06 09:18:11 +00:00

debug.py

[dynamo] Fix more local test failures (#169228)

2025-12-02 16:31:33 +00:00

decomposition.py

Revert "Ldexp out of range (#168357)"

2025-12-19 23:24:42 +00:00

dependencies.py

[inductor] do mix order reduction fusion earlier and allow fusing more nodes into it (#168209)

2025-12-05 23:45:35 +00:00

distributed_autotune.py

[inductor] fix typing for TensorBox.create (#169992)

2025-12-16 23:30:10 +00:00

dtype_propagation.py

Upgrade pyrefly version (#169633)

2025-12-12 08:33:22 +00:00

exc.py

…

extern_node_serializer.py

…

freezing_utils.py

…

freezing.py

[2/N] Use 'is' in callable comparisons (#166685)

2025-10-31 08:08:07 +00:00

fuzzer.py

Revert "[Inductor XPU GEMM] Step 1/N: Refactor cutlass configuration. (#160174)"

2025-12-12 00:13:49 +00:00

fx_utils.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

graph.py

[inductor] fix typing for TensorBox.create (#169992)

2025-12-16 23:30:10 +00:00

hooks.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

index_propagation.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

inductor_prims.py

…

invert_expr_analysis.py

Invert unary read and write for fusion (#161404)

2025-11-05 16:10:52 +00:00

ir.py

Fix syntax for suppression comments. (#167088)

2025-12-22 02:44:50 +00:00

jagged_lowerings.py

[inductor] fix typing for TensorBox.create (#169992)

2025-12-16 23:30:10 +00:00

kernel_inputs.py

[BugFix] Fix error param (#170952)

2025-12-20 21:34:31 +00:00

kernel_template_choice.py

…

loop_body.py

Invert unary read and write for fusion (#161404)

2025-11-05 16:10:52 +00:00

lowering.py

[inductor] Fix cudagraph skip for index_put_ with boolean indices, graph partitioning logic (#170103)

2025-12-18 22:43:04 +00:00

memory.py

Filter control deps while calculating lifetime (#169496)

2025-12-08 23:26:44 +00:00

metrics.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

mkldnn_ir.py

[inductor] fix typing for TensorBox.create (#169992)

2025-12-16 23:30:10 +00:00

mkldnn_lowerings.py

[inductor] fix typing for TensorBox.create (#169992)

2025-12-16 23:30:10 +00:00

mock_cache.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

ops_handler.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

optimize_indexing.py

…

output_code.py

Upgrade pyrefly version (#169633)

2025-12-12 08:33:22 +00:00

pattern_matcher.py

[easy][inductor] already matched node do not need to match again in R… (#170283)

2025-12-19 02:05:37 +00:00

quantized_lowerings.py

…

remote_cache.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

remote_gemm_autotune_cache.py

…

rocm_multiarch_utils.py

[ROCm] Enable multi-arch compilation and unit tests for AOT Inductor (#166357)

2025-11-06 19:08:15 +00:00

scheduler.py

[Inductor] Allow statically analyzing epilogues (#170001)

2025-12-18 22:50:25 +00:00

script.ld

…

select_algorithm.py

[Inductor] Support appending TLX-only CUDA options in triton template (#171117)

2025-12-23 00:51:09 +00:00

shape_propagation.py

Revert "[inductor] Remove implicit float64 upcast in triton (#170393)"

2025-12-19 00:26:35 +00:00

sizevars.py

Revert "no need to bail out for unbacked in statically_known_multiple_of (#170058)"

2025-12-18 18:37:19 +00:00

standalone_compile.py

[11/N] Apply ruff UP035 rule (#166225)

2025-11-04 04:53:40 +00:00

subgraph_lowering.py

[Flex] Fix symbolic shapes lowering (#168383)

2025-11-22 07:59:15 +00:00

test_case.py

[inductor][ez] skip cache for unit test via envvar (#167237)

2025-11-13 22:28:16 +00:00

test_operators.py

…

tiling_utils.py

Fix coalesce analysis to work with symints (#161819)

2025-12-15 23:30:18 +00:00

triton_bundler.py

…

utils.py

[Inductor] Support appending TLX-only CUDA options in triton template (#171117)

2025-12-23 00:51:09 +00:00

virtualized.py

Distributed Autotuning (#163369)

2025-11-06 21:10:21 +00:00

wrapper_benchmark.py

[inductor] Expand use of generic benchmark function (#164938)

2025-11-03 20:15:25 +00:00