Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 19, 2025

• Nccl update to 2.25.1 for cuda 12.4-12.8 (#146073)

  Show description for 4ece056
  

  atalman

  authored and

  pytorchmergebot

  committed
  4ece056

  Copy full SHA for 4ece056

• fix pt2e block wise quantization unit test (#147406)

  Show description for bd370c1
  

  cccclai

  authored and

  pytorchmergebot

  committed
  bd370c1

  Copy full SHA for bd370c1

• [cutlass backend] forward fix of standalone runner for fbcode (#147158)

  Show description for 5006932
  

  henrylhtsang

  authored and

  pytorchmergebot

  committed
  5006932

  Copy full SHA for 5006932

• [OSS] Update FileSystem methods to properly handle a string argument (#145751)

  Show description for f16d301
  

  ankitageorge

  authored and

  pytorchmergebot

  committed
  f16d301

  Copy full SHA for f16d301

• [ONNX] Pick up missing types in dynamic shapes renaming (#147407)

  Show description for 953f783
  

  titaiwangms

  authored and

  pytorchmergebot

  committed
  953f783

  Copy full SHA for 953f783

• [CD] Increase timeout for windows binary builds (#147390)

  Show description for 757d7f2
  

  atalman

  authored and

  pytorchmergebot

  committed
  757d7f2

  Copy full SHA for 757d7f2

• [ONNX] Move and improve error reproduction logic in test (#147391)

  Show description for 959d79f
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  959d79f

  Copy full SHA for 959d79f

Commits on Feb 18, 2025

• Revert "Add torch._scaled_mm for CPU (#139975)"

  Show description for babb2dc
  pytorchmergebot

  committed
  babb2dc

  Copy full SHA for babb2dc

• add unbacked strict mode (#147333)

  Show description for 525ca80
  

  bobrenjc93

  authored and

  pytorchmergebot

  committed
  525ca80

  Copy full SHA for 525ca80

• Add no_data_dependent_graph_break mode (#147342)

  Show description for 5d547d8
  

  bobrenjc93

  authored and

  pytorchmergebot

  committed
  5d547d8

  Copy full SHA for 5d547d8

• Update addr doc (#146482)

  Show description for bae049b
  

  albanD

  authored and

  pytorchmergebot

  committed
  bae049b

  Copy full SHA for bae049b

• [Sigmoid] Fix issues with constant folding and fba_ops (#146948)

  Show description for ca397d8
  

  trieuat

  authored and

  pytorchmergebot

  committed
  ca397d8

  Copy full SHA for ca397d8

• [FSDP2] Simplify shard_placement_fn in test (#146847)

  Show description for c9a15d9
  

  tsunghsienlee

  authored and

  pytorchmergebot

  committed
  c9a15d9

  Copy full SHA for c9a15d9

• [BE] correct docs for clock_rate to MHz, fixes #147098 (#147393)

  Show description for c8433c2
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  c8433c2

  Copy full SHA for c8433c2

• Add fqn_modifier at loading_state_dict and unit test (#146557)

  Show description for a21a123
  

  mori360

  authored and

  pytorchmergebot

  committed
  a21a123

  Copy full SHA for a21a123

• Revert "Nccl update to 2.25.1 for cuda 12.4-12.8 (#146073)"

  Show description for 7622e29
  pytorchmergebot

  committed
  7622e29

  Copy full SHA for 7622e29

• More precise check for shared storage check in inductor/reinplace pass (#147050)

  Show description for 3f35664
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  3f35664

  Copy full SHA for 3f35664

• [dynamo] replace hardcoded eval frame control flags skip_code_recursive_flag/cache_limit_hit_flag (#146355)

  Show description for 63e8ad4
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  63e8ad4

  Copy full SHA for 63e8ad4

• [dynamo] refactor dynamo__custom_eval_frame to C++, refactor SKIP_CODE[_RECURSIVE] (#145603)

  Show description for 75db0fd
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  75db0fd

  Copy full SHA for 75db0fd

• [codegen] enable SORT and TUPLE_REDUCTION for AMD Triton (#147340)

  Show description for eb892cd
  

  Yang Chen

  authored and

  pytorchmergebot

  committed
  eb892cd

  Copy full SHA for eb892cd

• Add link to non_blocking/pinmem tutorial in `Tensor.to` docstrings (#145651)

  Show description for 1b047d5
  

  Vincent Moens

  authored and

  pytorchmergebot

  committed
  1b047d5

  Copy full SHA for 1b047d5

• dynamo: Don't crash when encountering a object with no __name__ (#147246)

  Show description for 166419b
  

  c00w

  authored and

  pytorchmergebot

  committed
  166419b

  Copy full SHA for 166419b

• Fix typo (#147330)

  Show description for 74682e8
  

  12v

  authored and

  pytorchmergebot

  committed
  74682e8

  Copy full SHA for 74682e8

• Fix linter warnings (#147386)

  Show description for d9b3d76
  

  ahmadsharif1

  authored and

  pytorchmergebot

  committed
  d9b3d76

  Copy full SHA for d9b3d76

• Revert "Fix non-bitwise type annotations for Tensor operators (see #145838) (#146845)"

  Show description for 302f56a
  pytorchmergebot

  committed
  302f56a

  Copy full SHA for 302f56a

• [symbolic shapes] Add replacement for backed symints (#147240)

  Show description for 57060be
  

  angelayi

  authored and

  pytorchmergebot

  committed
  57060be

  Copy full SHA for 57060be

• [export] Log evaluate_expr (#146939)

  Show description for 84abeaa
  

  angelayi

  authored and

  pytorchmergebot

  committed
  84abeaa

  Copy full SHA for 84abeaa

• Deprecate `skip_code_recursive_on_cache_limit_hit` config flag (#136970)

  Show description for c6b331f
  

  zeshengzong

  authored and

  pytorchmergebot

  committed
  c6b331f

  Copy full SHA for c6b331f

• Add torch._scaled_mm for CPU (#139975)

  Show description for 6f7e67c
  

  yanbing-j

  authored and

  pytorchmergebot

  committed
  6f7e67c

  Copy full SHA for 6f7e67c

• Fix the AOTI compile failure with ARM CPU for Meta internal (#147204)

  Show description for dd2a943
  

  hl475

  authored and

  pytorchmergebot

  committed
  dd2a943

  Copy full SHA for dd2a943

• Update ck (#144799)

  Show description for 5d675de
  

  alugorey

  authored and

  pytorchmergebot

  committed
  5d675de

  Copy full SHA for 5d675de

• s390x: add cleanup for cancelled docker image builds (#147110)

  Show description for a00d2b5
  

  AlekseiNikiforovIBM

  authored and

  pytorchmergebot

  committed
  a00d2b5

  Copy full SHA for a00d2b5

• Update torch-xpu-ops commit pin (#147358)

  Show description for 6edc419
  

  toyxu

  authored and

  pytorchmergebot

  committed
  6edc419

  Copy full SHA for 6edc419

• [export] Loosen symint input serialization (#147237)

  Show description for 0c8028e
  

  angelayi

  authored and

  pytorchmergebot

  committed
  0c8028e

  Copy full SHA for 0c8028e

• Unify all sympy versions to avoid conflicts within PyTorch (#147197)

  Show description for b10ba0a
  

  fffrog

  authored and

  pytorchmergebot

  committed
  b10ba0a

  Copy full SHA for b10ba0a

Pagination

Previous

Next