pytorch

autoheuristic

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

codegen

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

compile_worker

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

fx_passes

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

kernel

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

package

[aoti package] seek 0 after loading buffer (#142204 )

2024-12-09 21:53:28 +00:00

runtime

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

__init__.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

aoti_eager.py

filelock: Make waitcounter variant to use (#139816 )

2024-12-12 01:18:34 +00:00

async_compile.py

[logging] Add triton_compile_time_us column to dynamo_compile (#142068 )

2024-12-06 16:11:57 +00:00

autotune_process.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

bounds.py

remove allow-untyped-defs from _inductor/bounds.py (#141440 )

2024-11-24 16:23:31 +00:00

choices.py

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

codecache.py

filelock: Make waitcounter variant to use (#139816 )

2024-12-12 01:18:34 +00:00

comm_analysis.py

[BE][Easy][16/19] enforce style for empty lines in import segments in torch/_i*/ (#129768 )

2024-07-20 16:20:58 +00:00

comm_lowering.py

[inductor] Refactor ir.Layout into ir.OutputSpec (#140910 )

2024-11-21 20:01:57 +00:00

comms.py

add torchrec collectives to enforce global ordering (#141970 )

2024-12-11 02:45:24 +00:00

compile_fx.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

compiler_bisector.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

config.py

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

constant_folding.py

[Quant][Inductor][X86] add fusion pass for linear_dynamic_fp16 (#141549 )

2024-12-07 03:08:08 +00:00

cpp_builder.py

filelock: Make waitcounter variant to use (#139816 )

2024-12-12 01:18:34 +00:00

cpu_vec_isa.py

filelock: Make waitcounter variant to use (#139816 )

2024-12-12 01:18:34 +00:00

cudagraph_trees.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

cudagraph_utils.py

[CUDAGraph] Warn once if too many distinct sizes (#132832 )

2024-08-07 19:48:06 +00:00

custom_graph_pass.py

[inductor] Add API to make post_grad_custom passes cache-able (#137298 )

2024-10-07 21:11:54 +00:00

debug.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

decomposition.py

Add support for bfloat16 atomic adds in fbcode (#141857 )

2024-12-10 11:40:15 +00:00

dependencies.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

dtype_propagation.py

Revert "Infer whether prologues can be computed without upcasting to fp32 without changing numerics (#142402 )"

2024-12-11 17:32:25 +00:00

exc.py

Type _inductor/exc.py (#137595 )

2024-10-09 23:15:06 +00:00

extern_node_serializer.py

[BE][Easy][14/19] enforce style for empty lines in import segments in torch/_[a-c]*/ and torch/_[e-h]*/ and torch/_[j-z]*/ (#129765 )

2024-07-31 10:42:50 +00:00

freezing.py

type annotations for meta_utils (#140203 )

2024-11-13 20:07:47 +00:00

fx_utils.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

graph.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

hooks.py

[BE][Easy][16/19] enforce style for empty lines in import segments in torch/_i*/ (#129768 )

2024-07-20 16:20:58 +00:00

index_propagation.py

[inductor] sympy.Integer([01]) -> sympy.S.(Zero|One) (#139523 )

2024-11-04 04:28:40 +00:00

inductor_prims.py

inductor: dont use default_dtype during rng functionalization (#136041 )

2024-09-17 03:40:54 +00:00

ir.py

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

jagged_lowerings.py

[inductor] Refactor MutableBox to make IRNode typing easier (#140895 )

2024-11-20 19:50:46 +00:00

loop_body.py

[inductor] Fix 3d tiling (#141709 )

2024-12-01 19:47:41 +00:00

lowering.py

[foreach_map] Initial foreach map HOP impl for inference (#142098 )

2024-12-11 21:32:11 +00:00

memory.py

[inductor] Refactor ir.Layout into ir.OutputSpec (#140910 )

2024-11-21 20:01:57 +00:00

metrics.py

[inductor] Multi-kernel + cooperative reductions (#138893 )

2024-10-29 15:45:17 +00:00

mkldnn_ir.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

mkldnn_lowerings.py

[inductor][cpp] Add BMM kernel template for autotuning (#129772 )

2024-12-06 04:54:00 +00:00

mock_cache.py

Profile guided optimization for automatic_dynamic (#139001 )

2024-11-03 06:29:57 +00:00

ops_handler.py

Revert "[Easy] factor out inductor ophandler decompositions (#142400 )"

2024-12-11 17:32:26 +00:00

optimize_indexing.py

type _inductor/optimize_indexing.py (#137599 )

2024-10-09 23:29:47 +00:00

output_code.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

pattern_matcher.py

Save indexing for getitem nodes when do custom replacements (#140193 )

2024-11-27 20:19:13 +00:00

quantized_lowerings.py

[inductor][cpp] Add BMM kernel template for autotuning (#129772 )

2024-12-06 04:54:00 +00:00

remote_cache.py

Make cache keys consistent between OSS and internal (#142147 )

2024-12-05 22:29:07 +00:00

scheduler.py

Revert "Prologue Fusion (#134532 )"

2024-12-11 17:32:26 +00:00

script.ld

Place .lrodata later in the binary (#117575 )

2024-01-18 17:58:18 +00:00

select_algorithm.py

filelock: Make waitcounter variant to use (#139816 )

2024-12-12 01:18:34 +00:00

sizevars.py

Fix undesired specialization on slice after split. (#142372 )

2024-12-11 18:52:17 +00:00

subgraph_lowering.py

[FlexAttention] add support for learnable biases in Inductor (#137452 )

2024-11-25 19:08:34 +00:00

test_case.py

Enable autograd cache on inductor tests (#140890 )

2024-11-27 20:41:43 +00:00

test_operators.py

[BE][Easy][16/19] enforce style for empty lines in import segments in torch/_i*/ (#129768 )

2024-07-20 16:20:58 +00:00

triton_bundler.py

pytorch/features: Make a feature logger and record triton bundling (#141056 )

2024-11-22 01:31:08 +00:00

utils.py

Revert "Remove unused Python variables in torch/[_-a]* (#133492 )"

2024-12-11 17:29:12 +00:00

virtualized.py

[inductor] Refactor reduction type choices into V.choices (#139585 )

2024-11-17 16:10:37 +00:00

wrapper_benchmark.py

Add ncu profile to final output_code.py (#142259 )

2024-12-07 07:54:43 +00:00