Commits

Branch selector

 codex/add-helper-function-to-sizevars.py

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 5, 2025

• is_size_one -> is_size_one_or_false

  Show description for b6c4296
  ColinPeppler

  authored
  b6c4296

  Copy full SHA for b6c4296

Commits on Sep 4, 2025

• Update torch/_inductor/lowering.py

  ColinPeppler

  authored
  a4f9132

  Copy full SHA for a4f9132

• Update torch/_inductor/lowering.py

  ColinPeppler

  authored
  f80a169

  Copy full SHA for f80a169

• Update torch/_inductor/lowering.py

  

  ColinPeppler

  authored and

  pytorchmergebot

  committed
  f657d11

  Copy full SHA for f657d11

• inductor: tighten is_size_one docs

  

  ColinPeppler

  authored and

  pytorchmergebot

  committed
  ff2bceb

  Copy full SHA for ff2bceb

• [SymmMem] Feed tensor.data_ptr instead of handle.buffer_ptr into kernels (#162193)

  Show description for 89d41d3
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  89d41d3

  Copy full SHA for 89d41d3

• [SymmMem] Add root argument to broadcast op (#161090)

  Show description for 9bdcee0
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  9bdcee0

  Copy full SHA for 9bdcee0

• [ROCm] Enabling several UTs (#161715)

  Show description for b9ba612
  

  pragupta

  authored and

  pytorchmergebot

  committed
  b9ba612

  Copy full SHA for b9ba612

• Revert "[SymmMem] Add root argument to broadcast op (#161090)"

  Show description for d5b3841
  pytorchmergebot

  committed
  d5b3841

  Copy full SHA for d5b3841

• Revert "Fix usage of forwarding references (#161094)"

  Show description for 48bedd7
  pytorchmergebot

  committed
  48bedd7

  Copy full SHA for 48bedd7

• Apply Triton tensor descriptor for flex-decoding for performance (#161643)

  Show description for a3d72b0
  

  EikanWang

  authored and

  pytorchmergebot

  committed
  a3d72b0

  Copy full SHA for a3d72b0

• Make distributed modules importable even when backend not built (#159889)

  Show description for ef3be67
  

  ezyang

  authored and

  pytorchmergebot

  committed
  ef3be67

  Copy full SHA for ef3be67

• Revert "[nativert] triton runtime implementation (#161798)"

  Show description for 95ee0bf
  pytorchmergebot

  committed
  95ee0bf

  Copy full SHA for 95ee0bf

• Fix Arm64 OSS pytorch build with FBGEMM (#161527)

  Show description for dbec087
  

  mcfi

  authored and

  pytorchmergebot

  committed
  dbec087

  Copy full SHA for dbec087

• Revert "[BLAS] Avoid downcasts for fp16fp16->fp32 BLAS (#161999)"

  Show description for c3d54de
  pytorchmergebot

  committed
  c3d54de

  Copy full SHA for c3d54de

• Revert "[BE] Cleanup stale comments/copy from `gemm` (#162001)"

  Show description for afa6e56
  pytorchmergebot

  committed
  afa6e56

  Copy full SHA for afa6e56

• Revert "[MPS] enable cat op for sparse (#162007)"

  Show description for 9e5247f
  pytorchmergebot

  committed
  9e5247f

  Copy full SHA for 9e5247f

• Always build USE_DISTRIBUTED. (#160449)

  Show description for c371032
  

  ezyang

  authored and

  pytorchmergebot

  committed
  c371032

  Copy full SHA for c371032

• [nativert] triton runtime implementation (#161798)

  Show description for 3dde5d7
  

  dolpm

  authored and

  pytorchmergebot

  committed
  3dde5d7

  Copy full SHA for 3dde5d7

• [BE]: Update cpp-httplib submodule to 0.26.0 (#162181)

  Show description for 1f51056
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  1f51056

  Copy full SHA for 1f51056

• [dynamo][hops] Remove const outputs from the speculated subgraph (#161355)

  Show description for 6b1900c
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  6b1900c

  Copy full SHA for 6b1900c

• Modified the docs to add example for torch.is_floating_point and torc… (#161951)

  Show description for 9480cdc
  

  mansiag05

  authored and

  pytorchmergebot

  committed
  9480cdc

  Copy full SHA for 9480cdc

• [cuDNN][SDPA] Enable cuDNN SDPA by default for SM 9.0, SM 10.0 (#162073)

  Show description for 6f7608d
  

  eqy

  authored and

  pytorchmergebot

  committed
  6f7608d

  Copy full SHA for 6f7608d

• export: add explicit decomposition for aten.expand_copy and unit test (#161688)

  Show description for d1a15ab
  

  albertw7711

  authored and

  pytorchmergebot

  committed
  d1a15ab

  Copy full SHA for d1a15ab

• [dynamo] Make the MRO walk more narrow (#162105)

  Show description for 3302859
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  3302859

  Copy full SHA for 3302859

• enable float32 and float16 in `torch._grouped_mm` fallback (#162059)

  Show description for 9eadb37
  

  vkuzo

  authored and

  pytorchmergebot

  committed
  9eadb37

  Copy full SHA for 9eadb37

• move `_grouped_mm` fallback to composite explicit autograd (#161717)

  Show description for 61fb632
  

  vkuzo

  authored and

  pytorchmergebot

  committed
  61fb632

  Copy full SHA for 61fb632

• create torch._grouped_mm fallback path with for loops / bmm (#161407)

  Show description for 8a736fa
  

  vkuzo

  authored and

  pytorchmergebot

  committed
  8a736fa

  Copy full SHA for 8a736fa

• [SymmMem] Increase signal pad size for NVL72 (#162026)

  Show description for 8bb213b
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  8bb213b

  Copy full SHA for 8bb213b

• [SymmMem] Add a helper API to distinguish intra- and inter- node (#161984)

  Show description for 869cbcc
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  869cbcc

  Copy full SHA for 869cbcc

• [CUDA] Reuse blocks with record_stream during CUDA Graph capture in the CUDACachingAllocator (#158352)

  Show description for 0c0e056
  

  2 peopleauthored and

  pytorchmergebot

  committed
  0c0e056

  Copy full SHA for 0c0e056

• [dynamo] change error_on_graph_break/fullgraph semantics (#161747)

  Show description for f36f285
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  f36f285

  Copy full SHA for f36f285

• Update torch-xpu-ops commit pin (#162062)

  Show description for ba7f546
  

  CuiYifeng

  authored and

  pytorchmergebot

  committed
  ba7f546

  Copy full SHA for ba7f546

• Add dependency-groups.dev to pyproject.toml (#161216)

  Show description for 43b7c86
  

  lakshayg

  authored and

  pytorchmergebot

  committed
  43b7c86

  Copy full SHA for 43b7c86

• [ROCm] [CK] Composable Kernel integration for inductor backend (#158747)

  Show description for 019fed3
  

  5 peopleauthored and

  pytorchmergebot

  committed
  019fed3

  Copy full SHA for 019fed3

Pagination

Previous

Next