OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
b71e813bce
pytorch/torch/_inductor
History
Bin Bao 8c17830dea [AOTI] Unify how weights are stored as data section (#139471) 
Summary: https://github.com/pytorch/pytorch/pull/118076 introduced a cleaner way to link weights as a data section for macos. Unify the code by adopting that approach for Linux as well.

Test Plan: CI

Differential Revision: D65302273

Pull Request resolved: https://github.com/pytorch/pytorch/pull/139471
Approved by: https://github.com/chenyang78
2024-11-02 00:23:24 +00:00
..
autoheuristic AutoHeuristic: mm ranking heuristic h100 (#133608) 2024-08-16 16:20:38 +00:00
codegen [inductor] Simplify remove_kernel_local_buffers (#139452) 2024-11-01 20:36:39 +00:00
compile_worker typing subproc_pool.py (#138032) 2024-10-18 15:31:05 +00:00
fx_passes [Easy] Add joint graph passes, fallback_random to bisector (#139295) 2024-11-01 23:21:53 +00:00
kernel [Flex Attention] Paged Attention (#137164) 2024-10-29 17:05:22 +00:00
package [aoti] Accept constant inputs (#137197) 2024-10-03 20:59:33 +00:00
runtime [inductor] set sanitize_overflow=False for triton kernels (#139502) 2024-11-01 23:10:21 +00:00
__init__.py [pt2] improve error logs for torch.cond and aoti package (#138647) 2024-10-24 23:38:07 +00:00
aoti_eager.py [4/N] Non-Tensor: Support layout, device and dtype for aten operations (#125897) 2024-07-23 17:50:17 +00:00
async_compile.py [easy] Log subproc pool creation (#138642) 2024-10-23 02:41:42 +00:00
autotune_process.py Add workspace to TritonTemplates (#138050) 2024-10-29 18:17:54 +00:00
bounds.py Existing mypy issues (#136236) 2024-09-24 01:02:07 +00:00
codecache.py [AOTI] Unify how weights are stored as data section (#139471) 2024-11-02 00:23:24 +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 Preliminary registered-buffer collective support via Inductor (#138029) 2024-10-30 18:11:09 +00:00
comms.py [Traceable FSDP2] Use .copy_ instead of .set_ for unsharded_param inplace update; Replace unsharded_param graph input usage with graph intermediate; Support FSDP2+LoRA (#133730) 2024-09-11 23:01:05 +00:00
compile_fx.py [fx graph cache] Support freezing with FX graph caching (#136505) 2024-11-01 18:29:29 +00:00
compiler_bisector.py Class rename (#139490) 2024-11-02 00:10:17 +00:00
config.py [AOTI] Unify how weights are stored as data section (#139471) 2024-11-02 00:23:24 +00:00
constant_folding.py Constant folding for lifted graph (#135060) 2024-10-28 06:28:31 +00:00
cpp_builder.py Enable Windows Arm64 (#133088) 2024-10-24 16:10:44 +00:00
cpu_vec_isa.py [PyTorch] Use 128-bit vectors for ARM64 (#137426) 2024-10-26 00:20:35 +00:00
cudagraph_trees.py typing compile_fx.py (#138033) 2024-10-21 18:14:59 +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 Refactor FxGraphDrawer to use HTML-like labels (#137726) 2024-11-01 23:19:50 +00:00
decomposition.py Revert "Fix unbind_copy and add its decomposition (#134319)" 2024-10-29 04:54:37 +00:00
dependencies.py [inductor] Fix reduction_hint sum to single element (#137754) 2024-10-13 01:08:23 +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 [aotd] Cleanup on subclasses in inductor freezing (#136549) 2024-09-27 11:20:03 +00:00
fx_utils.py Revert "Dont decompose aten.baddmm in inductor (#137904)" 2024-10-29 04:08:11 +00:00
graph.py Class rename (#139490) 2024-11-02 00:10:17 +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 Add try except for _maybe_evaluate_static call in IndexPropagation (#132128) 2024-08-05 01:02:51 +00:00
inductor_prims.py inductor: dont use default_dtype during rng functionalization (#136041) 2024-09-17 03:40:54 +00:00
ir.py [inductor] make requires_stride_order more unbacked-symint-aware (#137063) 2024-10-31 13:11:02 +00:00
jagged_lowerings.py Add lowering for aten.searchsorted (#135701) 2024-10-04 19:26:05 +00:00
loop_body.py [inductor] Add LoopBody.op_counts (#137945) 2024-10-16 06:35:10 +00:00
lowering.py [inductor] make requires_stride_order more unbacked-symint-aware (#137063) 2024-10-31 13:11:02 +00:00
memory.py add condition for freeable input buffer (#139480) 2024-11-01 21:15:40 +00:00
metrics.py [inductor] Multi-kernel + cooperative reductions (#138893) 2024-10-29 15:45:17 +00:00
mkldnn_ir.py [AOTI] add C shim for QConvPointWise (#138540) 2024-10-31 02:03:01 +00:00
mkldnn_lowerings.py [AOTI] add C shim for QConvPointWise (#138540) 2024-10-31 02:03:01 +00:00
ops_handler.py Add lowering for aten.searchsorted (#135701) 2024-10-04 19:26:05 +00:00
optimize_indexing.py type _inductor/optimize_indexing.py (#137599) 2024-10-09 23:29:47 +00:00
pattern_matcher.py [Easy] Refactor post grad application of passes (#139293) 2024-10-31 17:05:27 +00:00
quantized_lowerings.py type _inductor/quantized_lowerings.py (#137598) 2024-10-09 23:29:26 +00:00
remote_cache.py Profile guided optimization for automatic_dynamic (#139001) 2024-11-01 21:43:25 +00:00
scheduler.py [inductor] Simplify remove_kernel_local_buffers (#139452) 2024-11-01 20:36:39 +00:00
script.ld
select_algorithm.py [inductor] Refactor kernel args into SIMDKernelFeatures (#139327) 2024-11-01 00:30:14 +00:00
sizevars.py In Inductor, be willing to generate deferred runtime asserts when unbacked (#138804) 2024-10-28 02:19:55 +00:00
subgraph_lowering.py [BE] typing for decorators - fx/_compatibility (part 1) (#134202) 2024-08-22 17:07:33 +00:00
test_case.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
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 Consolidate Triton cache into Inductor cache (#138239) 2024-10-31 01:37:16 +00:00
utils.py [inductor] Remove SIMDKernel.last_usage (#139364) 2024-11-01 16:28:15 +00:00
virtualized.py [inductor] Move LoopBody to its own file (#135257) 2024-09-07 16:29:15 +00:00
wrapper_benchmark.py [fx graph cache] Support freezing with FX graph caching (#136505) 2024-11-01 18:29:29 +00:00