pytorch

mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00

History

lezcano 9a5b4d2403 Do not forward parent's value range to CSE variable for variables created within codegen. (#123099 ) Consider we are generating code for `ops.gt`, and within it we call `ops.to_dtype`. Before, we would forward the bounds from `gt` to the to the result of `to_dtype`, which is wrong. Pull Request resolved: https://github.com/pytorch/pytorch/pull/123099 Approved by: https://github.com/jgong5, https://github.com/peterbell10		2024-04-23 06:26:39 +00:00
..
codegen	Do not forward parent's value range to CSE variable for variables created within codegen. (#123099 )	2024-04-23 06:26:39 +00:00
fx_passes	[inductor] Refactor runtime files into torch._inductor.runtime (part 3) (#124557 )	2024-04-22 18:46:24 +00:00
kernel	Add support for capturing tensors with score_mod (#124444 )	2024-04-23 06:20:13 +00:00
runtime	[inductor] Use compile time config values in runtime (#124561 )	2024-04-22 18:46:40 +00:00
__init__.py	[aoti] Change aot_compile callsites (#122225 )	2024-03-29 21:34:20 +00:00
autotune_process.py	[inductor] Refactor runtime files into torch._inductor.runtime (part 3) (#124557 )	2024-04-22 18:46:24 +00:00
bounds.py	Some minor type stub improvements (#118529 )	2024-02-04 00:19:00 +00:00
codecache.py	Add AOTInductor generated cpp code to TORCH_TRACE (#124617 )	2024-04-22 19:25:20 +00:00
comm_analysis.py	[BE]: TRY002 - Ban raising vanilla exceptions (#124570 )	2024-04-21 22:26:40 +00:00
comms.py	[BE]: TRY002 - Ban raising vanilla exceptions (#124570 )	2024-04-21 22:26:40 +00:00
compile_fx.py	rename sl to strobelight (#124455 )	2024-04-19 22:50:13 +00:00
config.py	[inductor] Use compile time config values in runtime (#124561 )	2024-04-22 18:46:40 +00:00
constant_folding.py	[inductor] Handle meta tensor ops in graph (#123786 )	2024-04-12 19:03:13 +00:00
cudagraph_trees.py	[CUDAGraphTree] Support mutated inputs from prior cudagraph pool (#123231 )	2024-04-19 10:32:12 +00:00
cudagraph_utils.py	[CUDAGraphTree] Support mutated inputs from prior cudagraph pool (#123231 )	2024-04-19 10:32:12 +00:00
debug.py	[inductor] Add FileLock around V.debug.copy (#122665 )	2024-03-28 03:17:33 +00:00
decomposition.py	[AOTI] Fixes ScatterFallback codegen (#124580 )	2024-04-22 20:47:26 +00:00
dependencies.py	Make torch._inductor.dependencies.Dep a proper class (#124407 )	2024-04-22 17:09:34 +00:00
exc.py	[AOTInductor] Improve validation for C++ wrapper codegen (#111102 )	2023-10-13 08:46:17 +00:00
freezing.py	Use graph.find_nodes in inductor (#122256 )	2024-04-07 18:51:14 +00:00
fx_utils.py	Fix FakeTensorUpdater logic for updating fake tensors (#116168 )	2024-03-15 18:22:24 +00:00
graph.py	Add support for capturing tensors with score_mod (#124444 )	2024-04-23 06:20:13 +00:00
hooks.py	Enable local_partial_types (#118467 )	2024-01-28 13:38:22 +00:00
index_propagation.py	[inductor] Use python constants in IndexPropagation (#122031 )	2024-03-21 21:53:22 +00:00
inductor_prims.py	[inductor] Add explicit ops.fma and use it in softmax_backward (#122518 )	2024-04-06 02:15:16 +00:00
ir.py	[AOTI] Fixes ScatterFallback codegen (#124580 )	2024-04-22 20:47:26 +00:00
lowering.py	[AOTI] Fixes ScatterFallback codegen (#124580 )	2024-04-22 20:47:26 +00:00
metrics.py	[inductor] comprehensive padding (#120758 )	2024-04-15 19:05:51 +00:00
mkldnn_lowerings.py	[inductor] easy: move mkldnn lowerings to its own file (#123556 )	2024-04-09 03:44:27 +00:00
ops_handler.py	remove log2 decomposition; add log2 lowering (#123112 )	2024-04-02 16:16:26 +00:00
optimize_indexing.py	[BE]: TRY002 - Ban raising vanilla exceptions (#124570 )	2024-04-21 22:26:40 +00:00
pattern_matcher.py	[BE]: TRY002 - Ban raising vanilla exceptions (#124570 )	2024-04-21 22:26:40 +00:00
quantized_lowerings.py	[AOTInductor] Support quantized linear on CPU with fbgemm (#123069 )	2024-04-01 09:15:05 +00:00
scheduler.py	[inductor] Refactor runtime files into torch._inductor.runtime (part 3) (#124557 )	2024-04-22 18:46:24 +00:00
script.ld	Place .lrodata later in the binary (#117575 )	2024-01-18 17:58:18 +00:00
select_algorithm.py	Add support for capturing tensors with score_mod (#124444 )	2024-04-23 06:20:13 +00:00
sizevars.py	[inductor] simplify expr when looking up size hint (#123140 )	2024-04-04 04:59:59 +00:00
test_case.py	[inductor] Fix fresh_inductor_cache() (#122661 )	2024-04-15 20:28:54 +00:00
test_operators.py	Grandfather in some more pytorch ops to be pt2_compliant (#113050 )	2023-11-09 02:35:33 +00:00
utils.py	Don't clean up fresh inductor cache on error (#124620 )	2024-04-23 02:13:05 +00:00
virtualized.py	[EASY] Move OpsHandler and MockHandler to their own file (#119851 )	2024-02-15 18:54:41 +00:00
wrapper_benchmark.py	[inductor] Refactor runtime files into torch._inductor.runtime (part 3) (#124557 )	2024-04-22 18:46:24 +00:00