Commits

Branch selector

 bugfix_cuda_abs_jit

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 29, 2025

• [cuda][complex] Use scaling to compute the absolute value of complex numbers, since hypot cannot handle overflow properly.

  thenumberouscode

  committed
  4a1ec62

  Copy full SHA for 4a1ec62

Commits on Jul 17, 2025

• Add unified memory APIs for torch.accelerator (#152932)

  Show description for 2ad5c25
  

  guangyey

  authored and

  pytorchmergebot

  committed
  2ad5c25

  Copy full SHA for 2ad5c25

• Add DeviceAllocator as the base device allocator (#138222)

  Show description for 1179e33
  

  guangyey

  authored and

  pytorchmergebot

  committed
  1179e33

  Copy full SHA for 1179e33

• Always disable ShardingPropagation cache if compiling (#156868)

  Show description for f6d1388
  

  azahed98

  authored and

  pytorchmergebot

  committed
  f6d1388

  Copy full SHA for f6d1388

• [Device] Add support for PrivateUse1 device type in parse_type function (#157609)

  Show description for c09eba8
  

  trajepl

  authored and

  pytorchmergebot

  committed
  c09eba8

  Copy full SHA for c09eba8

• [easy][guards] Add developer comment for posterity (#158471)

  Show description for 2179afd
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  2179afd

  Copy full SHA for 2179afd

• [dynamo] Constant fold torch.autograd._profiler_enabled (#158482)

  Show description for d7e1b8b
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  d7e1b8b

  Copy full SHA for d7e1b8b

• [AOT_inductor] model_base.h add Windows include files. (#158477)

  Show description for b6454a9
  

  xuhancn

  authored and

  pytorchmergebot

  committed
  b6454a9

  Copy full SHA for b6454a9

• ci: Add reusable workflow to get changed files in PRs (#158517)

  Show description for e9367a7
  

  seemethere

  authored and

  pytorchmergebot

  committed
  e9367a7

  Copy full SHA for e9367a7

• inductor: Fix crash in split_cat when tensors is a Node (#157155)

  Show description for e78f2ac
  

  c00w

  authored and

  pytorchmergebot

  committed
  e78f2ac

  Copy full SHA for e78f2ac

• Refactor Provenance Tracking (#158399)

  Show description for 82a1ee1
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  82a1ee1

  Copy full SHA for 82a1ee1

• update expeced results (#158497)

  Show description for 306dd19
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  306dd19

  Copy full SHA for 306dd19

Commits on Jul 16, 2025

• [PP] Add eval() API to schedule (#157795)

  Show description for 1d58476
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  1d58476

  Copy full SHA for 1d58476

• [Dynamo][Better Engineering] Add enhanced typing support to `_dynamo/eval_frame.py` (#158276)

  Show description for a4d7532
  

  2 peopleauthored and

  pytorchmergebot

  committed
  a4d7532

  Copy full SHA for a4d7532

• [CUDA] Use runtime driver API for cuStreamWriteValue32 (#158295)

  Show description for a9f902a
  

  2 peopleauthored and

  pytorchmergebot

  committed
  a9f902a

  Copy full SHA for a9f902a

• Add transpose to torch/csrc/stable (#158160)

  Show description for e311886
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  e311886

  Copy full SHA for e311886

• [aoti][mps] Enable test_aot_inductor.py tests (#155598)

  Show description for 3cb1187
  

  angelayi

  authored and

  pytorchmergebot

  committed
  3cb1187

  Copy full SHA for 3cb1187

• [Dynamo][Better Engineering] Support typing in codegen.py (#158386)

  Show description for 5951fcd
  

  Lucaskabela

  authored and

  pytorchmergebot

  committed
  5951fcd

  Copy full SHA for 5951fcd

• [Dynamo][Better Engineering] Add typing to bytecode analysis and transform (#158293)

  Show description for ada44e5
  

  Lucaskabela

  authored and

  pytorchmergebot

  committed
  ada44e5

  Copy full SHA for ada44e5

• [BE][testing] Disable test_static_cuda_launcher:test_floats internally (#158296)

  Show description for 9df0176
  

  masnesral

  authored and

  pytorchmergebot

  committed
  9df0176

  Copy full SHA for 9df0176

• [DTensor][BE] add document to ShardingPropagator.register_op_strategy (#158362)

  Show description for 94c746b
  

  XilunWu

  authored and

  pytorchmergebot

  committed
  94c746b

  Copy full SHA for 94c746b

• [ez][lint] Add pr_time_benchmarks to merge conflictless csv linter (#158353)

  Show description for 473208c
  

  clee2000

  authored and

  pytorchmergebot

  committed
  473208c

  Copy full SHA for 473208c

• Add warning about removed sm50 and sm60 arches (#158301)

  Show description for fb731fe
  

  atalman

  authored and

  pytorchmergebot

  committed
  fb731fe

  Copy full SHA for fb731fe

• [4/n] Remove references to TorchScript in PyTorch docs (#158317)

  Show description for a9ee425
  

  yiming0416

  authored and

  pytorchmergebot

  committed
  a9ee425

  Copy full SHA for a9ee425

• Revert "recovering node source from dict (#158373)"

  Show description for 14ecc03
  pytorchmergebot

  committed
  14ecc03

  Copy full SHA for 14ecc03

• [export] Update docs (#157750)

  Show description for 1cc62c2
  

  angelayi

  authored and

  pytorchmergebot

  committed
  1cc62c2

  Copy full SHA for 1cc62c2

• [c10d]Prototype of remote_group_merge (#158287)

  Show description for f58a680
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  f58a680

  Copy full SHA for f58a680

• Revert "[cuda][cupy] Improve cupy device placement when device is provided (#158320)"

  Show description for 944a140
  pytorchmergebot

  committed
  944a140

  Copy full SHA for 944a140

• Fix invalid formatting (#158436)

  Show description for 79ab84e
  

  cyyever

  authored and

  pytorchmergebot

  committed
  79ab84e

  Copy full SHA for 79ab84e

• Move c10/macros/Macros.h to headeronly (#158365)

  Show description for 2b0f9b1
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  2b0f9b1

  Copy full SHA for 2b0f9b1

• Move the rest of c10/macros/Export.h (#158358)

  Show description for b40f48d
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  b40f48d

  Copy full SHA for b40f48d

• recovering node source from dict (#158373)

  Show description for 4d05598
  

  Gasoonjia

  authored and

  pytorchmergebot

  committed
  4d05598

  Copy full SHA for 4d05598

• Fix indexing with multi-dimensional boolean mask (#158369)

  Show description for bc9091a
  

  manuelcandales

  authored and

  pytorchmergebot

  committed
  bc9091a

  Copy full SHA for bc9091a

• Don't need to handle PyTrace_EXCEPTION in pyProfileFn (#154392)

  Show description for a26bf38
  

  D-D-H

  authored and

  pytorchmergebot

  committed
  a26bf38

  Copy full SHA for a26bf38

• [cond] add _FlopCounterMode support for cond (#158067)

  Show description for da05b7f
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  da05b7f

  Copy full SHA for da05b7f

Pagination

Previous

Next