| .. |
|
backends
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
polyfills
|
[Set] Update set.union and set.update to support *args (#152989)
|
2025-05-16 14:28:32 +00:00 |
|
repro
|
Revert "[BE]: Enable RUFF TRY400 rule - log.exception (#153473)"
|
2025-05-16 08:29:26 +00:00 |
|
variables
|
[user triton] dynamo support for new host-side TMA API (#155662)
|
2025-06-12 12:56:23 +00:00 |
|
__init__.py
|
[dynamo, nested graph breaks] add skip_frame debugging function (#153773)
|
2025-05-28 23:29:37 +00:00 |
|
_trace_wrapped_higher_order_op.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
bytecode_analysis.py
|
Remove code for Python < 3.9 (#147097)
|
2025-02-14 03:22:49 +00:00 |
|
bytecode_transformation.py
|
[dynamo, nested graph breaks] small fixes to resume function generation (#151056)
|
2025-05-28 23:29:37 +00:00 |
|
cache_size.py
|
|
|
|
callback.py
|
Redo D75092426: [internal] Expose additional metadata to compilation callbacks (#155063)
|
2025-06-05 23:40:31 +00:00 |
|
code_context.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
codegen.py
|
[dynamo] Avoid unncessary caching source codegen (#155376)
|
2025-06-10 19:38:15 +00:00 |
|
compiled_autograd.py
|
[ca] better error message for subclasses not supported by FakeTensor (#155481)
|
2025-06-11 19:09:29 +00:00 |
|
comptime.py
|
[BE][PYFMT] migrate PYFMT for torch._dynamo to ruff format (#144549)
|
2025-02-28 03:03:53 +00:00 |
|
config.py
|
[reland][dynamo] Record the pre-graph bytecode using fast record function event (#154974)
|
2025-06-06 13:11:03 +00:00 |
|
convert_frame.py
|
Replace frame_traced_fn hook with get_traced_code() util (#155249)
|
2025-06-10 22:40:58 +00:00 |
|
create_parameter_op.py
|
|
|
|
current_scope_id.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
debug_utils.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
decorators.py
|
[precompile] Ensure @disable()-ed function won't trigger recompile from precompile bytecode. (#155363)
|
2025-06-10 16:13:38 +00:00 |
|
device_interface.py
|
Use torch.types.Device in device_interface.py (#152935)
|
2025-05-07 13:20:10 +00:00 |
|
distributed.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
eval_frame.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
exc.py
|
[test] use JK to force graph break on slow aliasing/mutation/dynamic_shape behavior (#155257)
|
2025-06-09 16:21:59 +00:00 |
|
external_utils.py
|
[ca] mark scalar int sizes as dynamic via tensor wrapping (#151731)
|
2025-05-08 15:12:08 +00:00 |
|
funcname_cache.py
|
gracefully handle tokenize.TokenError in funcname parser. Adds support for non-Python source (#148737)
|
2025-03-11 19:49:28 +00:00 |
|
graph_break_hints.py
|
[dynamo] unimplemented -> unimplemented_v2 in variables/builder.py (#151044)
|
2025-04-11 09:07:01 +00:00 |
|
graph_deduplication.py
|
[Dynamo] Fix typing in graph_deduplication.py (#152572)
|
2025-05-13 12:17:59 +00:00 |
|
graph_region_tracker.py
|
[Hierarchical-Compile] Hash int args in addition to input shapes (#155655)
|
2025-06-12 06:35:12 +00:00 |
|
graph_utils.py
|
[Hierarchical Compile] Take into account mutation deps in cycle detection (#152506)
|
2025-05-13 12:17:59 +00:00 |
|
guards.py
|
[dynamo] Use BINARY_SUBSCR for pre-graph bytecode for regular dict accesses (#155727)
|
2025-06-12 04:02:29 +00:00 |
|
hooks.py
|
Replace frame_traced_fn hook with get_traced_code() util (#155249)
|
2025-06-10 22:40:58 +00:00 |
|
logging.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
metrics_context.py
|
[logging] Fix duration logging for dynamo_compile (#151749)
|
2025-04-22 03:29:13 +00:00 |
|
mutation_guard.py
|
Add functionality for installing free variables (#151134)
|
2025-04-24 17:57:54 +00:00 |
|
output_graph.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
pgo.py
|
Combine sticky pgo key with job id (#154863)
|
2025-06-03 07:58:38 +00:00 |
|
profiler.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
replay_record.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
resume_execution.py
|
[dynamo, nested graph breaks] refactor codegen to minimize NULL codegen'ing (#153510)
|
2025-05-28 23:29:37 +00:00 |
|
side_effects.py
|
Add docblock to torch/_dynamo/side_effects.py (#155403)
|
2025-06-08 07:02:30 +00:00 |
|
source.py
|
[dynamo] Use BINARY_SUBSCR for pre-graph bytecode for regular dict accesses (#155727)
|
2025-06-12 04:02:29 +00:00 |
|
symbolic_convert.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
|
tensor_version_op.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
test_case.py
|
Fix path matching in CPythonTestCase/setUpClass (#153070)
|
2025-05-13 16:26:25 +00:00 |
|
test_dont_skip_tracing_functions.py
|
[dynamo] context manager/decorator for dynamo config patching during tracing (#150586)
|
2025-04-23 09:12:13 +00:00 |
|
test_minifier_common.py
|
Document dynamo (#146736)
|
2025-02-13 00:02:21 +00:00 |
|
testing.py
|
[inductor][refactor] Refactor the fetching of subgraph names (#152770)
|
2025-05-06 02:55:34 +00:00 |
|
trace_rules.py
|
Support tuning of _grouped_mm (#153953)
|
2025-06-12 15:39:35 +00:00 |
|
types.py
|
[dynamo] Use sentinel value for guard filter. (#151131)
|
2025-04-12 15:29:57 +00:00 |
|
utils.py
|
Migrate from lru_cache to cache (#155613)
|
2025-06-11 19:44:18 +00:00 |
62fa3f5aeb