Mike Iovine 63c1f2fef9 [Static Runtime] Fold linear prepack ops (#85289) ... Summary: Split `quantized_linear_unpacked_weight_v2` into `linear_prepack` and `quantized_linear` so that the prepacking operation may be eliminated by constant folding. Test Plan: Fixes a huge regression in an internal model: ``` Before 89.6141 ms. 99.0923%. fb::quantized_linear_unpacked_weight_v2 (12 nodes) After 0.806852 ms. 53.5365%. quantized::linear (12 nodes, out variant) (prepacking eliminated) ``` Differential Revision: D39622530 Pull Request resolved: https://github.com/pytorch/pytorch/pull/85289 Approved by: https://github.com/davidberard98		2022-09-22 20:23:07 +00:00
..
interpreter
static	[Static Runtime] Fold linear prepack ops (#85289)	2022-09-22 20:23:07 +00:00
argument_spec.cpp
argument_spec.h
autodiff.cpp
autodiff.h
calculate_necessary_args.h
custom_operator.h	Revert "Fix crash on unload torch cpu dll (#67632)"	2022-08-02 00:56:18 +00:00
decomposition_registry_util.cpp
decomposition_registry_util.h
decomposition_registry.cpp	[reland 2] Call jit decomp in VariableType to improve forward AD coverage (#84976)	2022-09-15 22:46:19 +00:00
decomposition_registry.h	[reland 2] Call jit decomp in VariableType to improve forward AD coverage (#84976)	2022-09-15 22:46:19 +00:00
exception_message.h
graph_executor_impl.h
graph_executor.cpp	Autograd graphtask trim unnecessary edges (#82544)	2022-08-11 18:50:09 +00:00
graph_executor.h
graph_iterator.h
instruction.cpp
instruction.h
interpreter.cpp
interpreter.h
jit_exception.cpp
jit_exception.h
jit_trace.cpp
jit_trace.h
logging.cpp
logging.h
operator_options.h
operator.cpp
operator.h	Add OpOverload.decompose API (#83075)	2022-08-09 18:53:19 +00:00
print_handler.cpp
print_handler.h
profiling_graph_executor_impl.cpp
profiling_graph_executor_impl.h
profiling_record.cpp	Revert "Fix crash on unload torch cpu dll (#67632)"	2022-08-02 00:56:18 +00:00
profiling_record.h	Revert "Fix crash on unload torch cpu dll (#67632)"	2022-08-02 00:56:18 +00:00
register_c10_ops.cpp
register_cuda_ops.cpp	[ROCm] Enable/fix unit tests test_stream_args and test_event_args (#82346)	2022-08-01 22:55:15 +00:00
register_distributed_ops.cpp
register_ops_utils.cpp
register_ops_utils.h
register_prim_ops_fulljit.cpp
register_prim_ops.cpp	Performance optimizations to proxy tensor (#85049)	2022-09-16 00:28:50 +00:00
register_special_ops.cpp
script_profile.cpp
script_profile.h
serialized_shape_function_registry.cpp	Fix for transposed convolution shape functions (#83557)	2022-08-22 19:05:41 +00:00
serialized_shape_function_registry.h
shape_function_registry.h
simple_graph_executor_impl.cpp
simple_graph_executor_impl.h
slice_indices_adjust.cpp
slice_indices_adjust.h
symbolic_script.cpp	Supports symbolic diff for silu (#81724)	2022-08-09 01:18:10 +00:00
symbolic_script.h
symbolic_shape_registry_util.cpp	[NNC] add eltwise OPs: mish and elu (#80586)	2022-09-17 01:44:34 +00:00
symbolic_shape_registry_util.h
symbolic_shape_registry.cpp	Adding additional debug logging and documentation for shape functions (#77115)	2022-08-15 23:39:28 +00:00
symbolic_shape_registry.h
vararg_functions.cpp
vararg_functions.h
variable_tensor_list.h