pytorch/test/dynamo
Tugsbayasgalan Manlaibaatar 20c5add133 [export] Refactor constrain_as_value and constrain_as_size (#106591)
Some notable changes:
1. `constrain_as_size` allows min value to be less than 2 as it will unconditionally assume min >= 2 for compiler purposes. Instead, we add additional check to make sure max value is always greater than 2.
2. Previously, we used to runtime assert on the unbacked symint's val range which would be always between [2, max]. I modified this logic to assert on [0, max] unless user explicitly specifies the min range.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/106591
Approved by: https://github.com/gmagogsfm, https://github.com/ezyang
2023-08-15 05:41:43 +00:00
..
mock_modules
__init__.py
test_activation_checkpointing.py [partitioners][ac][dynamic] Fix output signature of fwd with symints (#105771) 2023-07-22 03:04:11 +00:00
test_after_aot.py
test_aot_autograd.py Add sequence_nr to aot_autograd to map forward ops to their corresponding backward ops (#103129) 2023-08-02 00:52:52 +00:00
test_autograd_function.py [BE] Enable ruff's UP rules and autoformat dynamo / functorch and refs (#105432) 2023-07-19 13:48:44 +00:00
test_backends.py Remove dynamo+nvfuser (#105789) 2023-08-08 22:29:32 +00:00
test_compile.py [BE] Enable ruff's UP rules and autoformat dynamo / functorch and refs (#105432) 2023-07-19 13:48:44 +00:00
test_comptime.py [dynamo] Simulate torch function enablement state (#105091) 2023-07-13 17:42:20 +00:00
test_config.py
test_ctx_manager.py Change _dynamo.export to be export(f)(*args, **kwargs) (#106109) 2023-07-27 21:41:13 +00:00
test_cudagraphs.py [ROCm] enable cudagraph inductor UTs on ROCm (#105662) 2023-08-01 20:55:27 +00:00
test_decorators.py Refactor stack handling to always use TracingContext to populate real stack on exception (#106277) 2023-08-02 00:09:16 +00:00
test_dynamic_shapes.py [dynamo][easy] Only xfail test_dynamic_shapes_float_guard_dynamic_shapes if z3 is available (#107137) 2023-08-14 19:47:21 +00:00
test_exc.py Run translation validation on tracing error. (#106645) 2023-08-14 13:43:34 +00:00
test_export.py [export] Refactor constrain_as_value and constrain_as_size (#106591) 2023-08-15 05:41:43 +00:00
test_export_mutations.py Change _dynamo.export to be export(f)(*args, **kwargs) (#106109) 2023-07-27 21:41:13 +00:00
test_functions.py NumPy support in torch.compile (#106211) 2023-08-11 00:39:32 +00:00
test_global.py
test_higher_order_ops.py [pt2] support vmap (#101707) 2023-08-09 03:39:33 +00:00
test_input_attr_tracking.py
test_interop.py
test_logging.py [dynamo, logging] add default pt2 logging group (#106417) 2023-08-04 20:34:42 +00:00
test_minifier.py
test_misc.py [Dynamo x FSDP][1/x] Builder support for deque, appendleft (#106884) 2023-08-11 03:26:12 +00:00
test_model_output.py
test_modules.py Back out "Reland "Make adding buffers more like adding parameters (#104069)" (#106224)" (#106743) 2023-08-08 15:27:34 +00:00
test_nops.py
test_optimizers.py Enable more e2e foreach optimizer compilation tests (#105438) 2023-07-20 02:41:19 +00:00
test_pre_dispatch.py
test_profiler.py [profiler] Fix profiling shapes with PT2 + lists of dynamic shapes (#105893) 2023-07-26 13:41:07 +00:00
test_python_autograd.py
test_recompile_ux.py Switch automatic_dynamic_shapes to True by default in fbcode (#104883) 2023-07-13 17:37:57 +00:00
test_recompiles.py Enable dynamic shapes of torch.nn.Parameter (#105855) 2023-08-08 05:40:01 +00:00
test_replay_record.py
test_repros.py Back out "Reland "Make adding buffers more like adding parameters (#104069)" (#106224)" (#106743) 2023-08-08 15:27:34 +00:00
test_skip_non_tensor.py
test_sources.py
test_subclasses.py [dynamo] Simulate torch function enablement state (#105091) 2023-07-13 17:42:20 +00:00
test_subgraphs.py Dynamo tensor aliasing guards, dedup graphargs (#104921) 2023-07-13 22:18:08 +00:00
test_unspec.py NumPy support in torch.compile (#106211) 2023-08-11 00:39:32 +00:00
test_verify_correctness.py
utils.py Grab bag of DTensor enablement stuff (Enable whole graph capture for DTensor) (#105787) 2023-07-30 00:17:45 +00:00