Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on May 7, 2024

• Separate arm64 and amd64 docker builds (pytorch#125617)

  Show description for b29d77b
  

  atalman

  authored and

  pytorchmergebot

  committed
  b29d77b

  Copy full SHA for b29d77b

• Fix & optimize open device registration test. (pytorch#125572)

  Show description for 5dee462
  

  FFFrog

  authored and

  pytorchmergebot

  committed
  5dee462

  Copy full SHA for 5dee462

• Enable dynamo-traced optimizer peak memory tests (pytorch#124543)

  Show description for f0c6d61
  

  mlazos

  authored and

  pytorchmergebot

  committed
  f0c6d61

  Copy full SHA for f0c6d61

• Disable fb_memcache for MTIA (pytorch#125658)

  Show description for 5033d3b
  

  oulgen

  authored and

  pytorchmergebot

  committed
  5033d3b

  Copy full SHA for 5033d3b

• [PT2D] Fix the circular import issue (pytorch#125618)

  Show description for e72936c
  

  fegin

  authored and

  pytorchmergebot

  committed
  e72936c

  Copy full SHA for e72936c

• Rename TorchDynamo -> Dyanamo in the dynamo tutorial doc (pytorch#123431)

  Show description for acafaba
  

  lezcano

  authored and

  pytorchmergebot

  committed
  acafaba

  Copy full SHA for acafaba

• [inductor][cpp] support int64 vertical vec reduction (fix pytorch#124821) (pytorch#125563)

  Show description for 058e281
  

  Jiong Gong

  authored and

  pytorchmergebot

  committed
  058e281

  Copy full SHA for 058e281

• refactor: extract get_lr warning (pytorch#125545)

  Show description for a60fa96
  

  david20571015

  authored and

  pytorchmergebot

  committed
  a60fa96

  Copy full SHA for a60fa96

• [dynamo] support torchbind object input (pytorch#124978)

  Show description for 461ffaa
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  461ffaa

  Copy full SHA for 461ffaa

• Enable UFMT on test_decomp.py, test_expanded_weights.py and some files (pytorch#125117)

  Show description for c165a8e
  

  shink

  authored and

  pytorchmergebot

  committed
  c165a8e

  Copy full SHA for c165a8e

• [Inductor] log fusion failure due to index mismatch (pytorch#124986)

  Show description for 48b6c8d
  

  shunting314

  authored and

  pytorchmergebot

  committed
  48b6c8d

  Copy full SHA for 48b6c8d

• add uuid in cudaDeviceProperties (pytorch#125083)

  Show description for f35fe4e
  

  jeffdaily

  authored and

  pytorchmergebot

  committed
  f35fe4e

  Copy full SHA for f35fe4e

• [export] Allow constant attr mutation (pytorch#125424)

  Show description for 4332fc4
  

  angelayi

  authored and

  pytorchmergebot

  committed
  4332fc4

  Copy full SHA for 4332fc4

• Updated docs to add the error case for torch.multinomial Issue#125388 (pytorch#125495)

  Show description for c0c2f61
  

  jainapurva

  authored and

  pytorchmergebot

  committed
  c0c2f61

  Copy full SHA for c0c2f61

• DTensor Fused ADAM (pytorch#125369)

  Show description for 3407899
  

  msaroufim

  authored and

  pytorchmergebot

  committed
  3407899

  Copy full SHA for 3407899

• [BE] Delete unused AT_FORALL_SCALAR_TYPES_AND[456] (pytorch#125607)

  Show description for 65fc3c3
  

  malfet

  authored and

  pytorchmergebot

  committed
  65fc3c3

  Copy full SHA for 65fc3c3

Commits on May 6, 2024

• fix loading optimizer options from archive (pytorch#125215)

  Show description for 3411d54
  

  daizhirui

  authored and

  pytorchmergebot

  committed
  3411d54

  Copy full SHA for 3411d54

• [CUDNN] Remove defunct cuDNN V8 API build flag (pytorch#120006)

  Show description for ee4cafa
  

  eqy

  authored and

  pytorchmergebot

  committed
  ee4cafa

  Copy full SHA for ee4cafa

• Better repro command: include test class + fix paths for py3.8 (pytorch#125498)

  Show description for b98c689
  

  jbschlosser

  authored and

  pytorchmergebot

  committed
  b98c689

  Copy full SHA for b98c689

• Initial implementation of Inductor FX Graph Remote Cache (pytorch#124669)

  Show description for 22bcfc2
  

  oulgen

  authored and

  pytorchmergebot

  committed
  22bcfc2

  Copy full SHA for 22bcfc2

• Nested Tensor + AOTI test (pytorch#125513)

  Show description for 05bd7fe
  

  davidberard98

  authored and

  pytorchmergebot

  committed
  05bd7fe

  Copy full SHA for 05bd7fe

• [TD] Enable TD on AVX related configs (pytorch#125482)

  Show description for 1b3fd83
  

  clee2000

  authored and

  pytorchmergebot

  committed
  1b3fd83

  Copy full SHA for 1b3fd83

• Reduce the number of layers for mixtral moe model to adapt CI memory limitation (pytorch#125608)

  Show description for 8c74162
  

  yanboliang

  authored and

  pytorchmergebot

  committed
  8c74162

  Copy full SHA for 8c74162

• xfail codegen dynamic if the test is xfailed (pytorch#125573)

  Show description for 7ddf57e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  7ddf57e

  Copy full SHA for 7ddf57e

• [dynamo] better file open method in funcname_cache (pytorch#125435)

  Show description for 373a00d
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  373a00d

  Copy full SHA for 373a00d

• [pipelining] Add tests for tracing frontend (pytorch#125449)

  Show description for cbb3791
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  cbb3791

  Copy full SHA for cbb3791

• [dynamo] fix potentially missing _torchdynamo_inline from ScriptFunction (pytorch#125447)

  Show description for bdaa7bb
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  bdaa7bb

  Copy full SHA for bdaa7bb

• Move autocast op list to autocast_mode.h to make sure other backends can reuse it. (pytorch#125114)

  Show description for ad9a27f
  

  PHLens

  authored and

  pytorchmergebot

  committed
  ad9a27f

  Copy full SHA for ad9a27f

• Revert "Compute bounds for the variables created during codegen (pytorch#123100)"

  Show description for 2a42c40
  pytorchmergebot

  committed
  2a42c40

  Copy full SHA for 2a42c40

• [FSDP] mark pre_backward_hook unserializable (pytorch#125464)

  Show description for 9cd4bcb
  

  ppwwyyxx

  authored and

  pytorchmergebot

  committed
  9cd4bcb

  Copy full SHA for 9cd4bcb

• Allow building for sm90a (pytorch#125523)

  Show description for 7d10b06
  

  drisspg

  authored and

  pytorchmergebot

  committed
  7d10b06

  Copy full SHA for 7d10b06

• Revert "Upgrade submodule oneDNN to v3.4 (pytorch#122472)"

  Show description for ee0c473
  pytorchmergebot

  committed
  ee0c473

  Copy full SHA for ee0c473

• Remove some pre-c++17 cruft (pytorch#125590)

  Show description for af14413
  

  r-barnes

  authored and

  pytorchmergebot

  committed
  af14413

  Copy full SHA for af14413

• try to fix the warning in distribute_tensor (pytorch#125476)

  Show description for daf1eb4
  

  wanchaol

  authored and

  pytorchmergebot

  committed
  daf1eb4

  Copy full SHA for daf1eb4

• Revert "[FX] Update type hints in torch.fx._compatibility.py (pytorch#125469)"

  Show description for 7ffa555
  pytorchmergebot

  committed
  7ffa555

  Copy full SHA for 7ffa555

Pagination

Previous

Next