Commits

Branch selector

 patch-27

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 3, 2025

• Update rocm.yml

  amdfaa

  authored
  2fb4875

  Copy full SHA for 2fb4875

• Change inductor-rocm.yml cron schedule to run every hour

  amdfaa

  authored
  ef6ae30

  Copy full SHA for ef6ae30

• [Inductor] addmm with bias -> unfuse bias if there is a pointwise/reduction consumer (#166165)

  Show description for 94f2657
  

  nikitaved

  authored and

  pytorchmergebot

  committed
  94f2657

  Copy full SHA for 94f2657

• Remove tools from BC linter (#166858)

  Show description for 3f6538f
  

  ezyang

  authored and

  pytorchmergebot

  committed
  3f6538f

  Copy full SHA for 3f6538f

• Switch to pyrefly as only type checker (#166197)

  Show description for f33abae
  

  maggiemoss

  authored and

  pytorchmergebot

  committed
  f33abae

  Copy full SHA for f33abae

• [MPS] Error out when BatchNorm is called for Complex (#166215)

  Show description for 73da7a4
  

  malfet

  authored and

  pytorchmergebot

  committed
  73da7a4

  Copy full SHA for 73da7a4

• Avoid std::copy_n in CopyKernel and IndexKernel (#143544)

  Show description for 335b5c7
  

  cyyever

  authored and

  pytorchmergebot

  committed
  335b5c7

  Copy full SHA for 335b5c7

• Revert "Back out "Do not decompose in functionalization/proxy tensor if autograd wouldn't have decomposed (#164939)" (#165910)" (#166812)

  Show description for 76bb27e
  

  ezyang

  authored and

  pytorchmergebot

  committed
  76bb27e

  Copy full SHA for 76bb27e

• Remove nightly pth check from pyrefly (#166857)

  Show description for a2da693
  

  ezyang

  authored and

  pytorchmergebot

  committed
  a2da693

  Copy full SHA for a2da693

• [Code Clean] Clean asserts in torch/ao/quantization (root, quantizer, backend_config) (#165433)

  Show description for d177900
  

  zhudada0120

  authored and

  pytorchmergebot

  committed
  d177900

  Copy full SHA for d177900

• Revert "Fixes torch.compile(nn.ModuleList()) changes bool() behavior (#159208)"

  Show description for 61bcc8d
  pytorchmergebot

  committed
  61bcc8d

  Copy full SHA for 61bcc8d

• Revert "[MPS] Fix `smooth_l1_loss` backward for fp16 (#166687)"

  Show description for 1656b25
  pytorchmergebot

  committed
  1656b25

  Copy full SHA for 1656b25

• Revert "Give full Dynamo stack traces in CI (#160417)"

  Show description for 5d62307
  pytorchmergebot

  committed
  5d62307

  Copy full SHA for 5d62307

• Revert "[MPS] Error out when BatchNorm is called for Complex (#166215)"

  Show description for a4077b5
  pytorchmergebot

  committed
  a4077b5

  Copy full SHA for a4077b5

• [inductor] Collectives estimations: option to use nccl estimator for fx node (#166521)

  Show description for ae038f8
  

  IvanKobzarev

  authored and

  pytorchmergebot

  committed
  ae038f8

  Copy full SHA for ae038f8

• [xla hash update] update the pinned xla hash (#166845)

  Show description for defac66
  

  pytorchupdatebot

  authored and

  pytorchmergebot

  committed
  defac66

  Copy full SHA for defac66

• Reapply "Back out "Do not decompose in functionalization/proxy tensor if autograd wouldn't have decomposed (#164939)" (#165910)" (#166812)

  Show description for 061fa73
  pytorchmergebot

  committed
  061fa73

  Copy full SHA for 061fa73

• [caffe2] Ignore -Wswitch-enum warnings (#166760)

  Show description for 9501405
  

  NSProgrammer

  authored and

  pytorchmergebot

  committed
  9501405

  Copy full SHA for 9501405

• Give full Dynamo stack traces in CI (#160417)

  Show description for e0791fc
  

  ezyang

  authored and

  pytorchmergebot

  committed
  e0791fc

  Copy full SHA for e0791fc

• [2/N] Change C-style casts to static_cast or reinterpret_cast (#165891)

  Show description for e1d011d
  

  2 peopleauthored and

  pytorchmergebot

  committed
  e1d011d

  Copy full SHA for e1d011d

• [3/N] Add clang-tidy readability checks (#164692)

  Show description for 3f54010
  

  cyyever

  authored and

  pytorchmergebot

  committed
  3f54010

  Copy full SHA for 3f54010

• Revert "Back out "Do not decompose in functionalization/proxy tensor if autograd wouldn't have decomposed (#164939)" (#165910)" (#166812)

  Show description for 5a3930a
  

  ezyang

  authored and

  pytorchmergebot

  committed
  5a3930a

  Copy full SHA for 5a3930a

• torch.cond supports autograd now (#165908)

  Show description for a5f0007
  

  ezyang

  authored and

  pytorchmergebot

  committed
  a5f0007

  Copy full SHA for a5f0007

• Fix: type promotion in FakeTensor (#166522)

  Show description for 69fb3eb
  

  krastogi-in

  authored and

  pytorchmergebot

  committed
  69fb3eb

  Copy full SHA for 69fb3eb

• [2/N] Correctly use test parameters (#166783)

  Show description for 1c4ced2
  

  cyyever

  authored and

  pytorchmergebot

  committed
  1c4ced2

  Copy full SHA for 1c4ced2

• [6/N] Remove unused loop variables in tests (#166785)

  Show description for 392acee
  

  cyyever

  authored and

  pytorchmergebot

  committed
  392acee

  Copy full SHA for 392acee

• [DebugMode] add stack traces (#166440)

  Show description for fee1ac9
  

  pianpwk

  authored and

  pytorchmergebot

  committed
  fee1ac9

  Copy full SHA for fee1ac9

• [dynamo] fix pos-only names should can be collected in `**kwargs` (#166798)

  Show description for 4a7fefd
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  4a7fefd

  Copy full SHA for 4a7fefd

• [export] Fix static_input_indices for aot_export_joint (#166761)

  Show description for 3b43159
  

  angelayi

  authored and

  pytorchmergebot

  committed
  3b43159

  Copy full SHA for 3b43159

Commits on Nov 2, 2025

• Revert "Add min/max support for barebones uint types (#166813)"

  Show description for 3eddf04
  pytorchmergebot

  committed
  3eddf04

  Copy full SHA for 3eddf04

• [BE] Using std::move to reduce copy constructor calls by one. (#163599)

  Show description for 7c203b8
  

  thenumberouscode

  authored and

  pytorchmergebot

  committed
  7c203b8

  Copy full SHA for 7c203b8

• Add claude skills for uint support and AT_DISPATCH_V2 (#166814)

  Show description for 3ca216a
  

  ezyang

  authored and

  pytorchmergebot

  committed
  3ca216a

  Copy full SHA for 3ca216a

• Add min/max support for barebones uint types (#166813)

  Show description for 9c22bbb
  

  ezyang

  authored and

  pytorchmergebot

  committed
  9c22bbb

  Copy full SHA for 9c22bbb

• [MPS] Refactor `torch.cat` and add fast path for contiguous inputs (#166556)

  Show description for 6268883
  

  kurtamohler

  authored and

  pytorchmergebot

  committed
  6268883

  Copy full SHA for 6268883

• [Sparse] support for exp op (#166801)

  Show description for 16212f0
  

  Isalia20

  authored and

  pytorchmergebot

  committed
  16212f0

  Copy full SHA for 16212f0

Pagination

Previous

Next