Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 26, 2024

• [Inductor XPU] Fix wrong device check before skip concat linear. (pytorch#140916)

  Show description for 9dd3b85
  

  etaf

  authored and

  pytorchmergebot

  committed
  9dd3b85

  Copy full SHA for 9dd3b85

• [AOTI XPU] Enable Cpp wraper for Intel GPU. (pytorch#135318)

  Show description for 4742080
  

  etaf

  authored and

  pytorchmergebot

  committed
  4742080

  Copy full SHA for 4742080

• [Intel GPU] XPUInductorQuantizer for XPU int8 recipe customization (pytorch#139578)

  Show description for c418a9a
  

  ZhiweiYan-96

  authored and

  pytorchmergebot

  committed
  c418a9a

  Copy full SHA for c418a9a

• Revert "[sparse] add extra options to _cslt_spare_mm (pytorch#137427)"

  Show description for 5318bf8
  pytorchmergebot

  committed
  5318bf8

  Copy full SHA for 5318bf8

• Remove linking of private cuda targets (pytorch#141463)

  Show description for 6d4cd3e
  

  cyyever

  authored and

  pytorchmergebot

  committed
  6d4cd3e

  Copy full SHA for 6d4cd3e

• [Intel GPU] qconv at XPU backend (pytorch#133080)

  Show description for 648f5d9
  

  ZhiweiYan-96

  authored and

  pytorchmergebot

  committed
  648f5d9

  Copy full SHA for 648f5d9

• [BE] Use torch.special.expm1 (pytorch#141518)

  Show description for f2d388e
  

  malfet

  authored and

  pytorchmergebot

  committed
  f2d388e

  Copy full SHA for f2d388e

• [Dynamo][autograd.Function] Use fake tensor prop to infer fwd output (pytorch#136184)

  Show description for dcd16bd
  

  yanboliang

  authored and

  pytorchmergebot

  committed
  dcd16bd

  Copy full SHA for dcd16bd

• Fix some typos in cuda.cmake (pytorch#141462)

  Show description for 6b60f4b
  

  cyyever

  authored and

  pytorchmergebot

  committed
  6b60f4b

  Copy full SHA for 6b60f4b

• [IntraNodeComm] fix a recent breakage (pytorch#141200)

  Show description for 6a22cae
  

  yifuwang

  authored and

  pytorchmergebot

  committed
  6a22cae

  Copy full SHA for 6a22cae

• [dynamo] Fix and simplify hanlding of Set.update method (pytorch#141286)

  Show description for 583484b
  

  StrongerXi

  authored and

  pytorchmergebot

  committed
  583484b

  Copy full SHA for 583484b

• fix non termination in unflatten + state (pytorch#141494)

  Show description for 5d7c370
  

  avikchaudhuri

  authored and

  pytorchmergebot

  committed
  5d7c370

  Copy full SHA for 5d7c370

• Upgrade ROCm wheels to manylinux2_28 - 1 of 2 (docker images) (pytorch#140681)

  Show description for 9ccbd84
  

  jithunnair-amd

  authored and

  pytorchmergebot

  committed
  9ccbd84

  Copy full SHA for 9ccbd84

• [Build] Add COMMIT_SHA to caffe2::GetBuildOptions (pytorch#141313)

  Show description for 8f5ce86
  

  malfet

  authored and

  pytorchmergebot

  committed
  8f5ce86

  Copy full SHA for 8f5ce86

• Revert "Always unspecialize float in OSS (pytorch#138922)"

  Show description for ad37afd
  pytorchmergebot

  committed
  ad37afd

  Copy full SHA for ad37afd

• Revert "Remove THC from OSS build (pytorch#134969)"

  Show description for 964655b
  pytorchmergebot

  committed
  964655b

  Copy full SHA for 964655b

Commits on Nov 25, 2024

• [sparse] add extra options to _cslt_spare_mm (pytorch#137427)

  Show description for f145116
  

  jcaip

  authored and

  pytorchmergebot

  committed
  f145116

  Copy full SHA for f145116

• Populate nn.module.stack in _fuse_conv_bn_qat (pytorch#141400)

  Show description for 02990fe
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  02990fe

  Copy full SHA for 02990fe

• [ROCm] Remove gfx906 from CI docker build (pytorch#141523)

  Show description for 851edf2
  

  jithunnair-amd

  authored and

  pytorchmergebot

  committed
  851edf2

  Copy full SHA for 851edf2

• [PGNCCL] Record device index for GPU guarding during NCCLComm method calls (pytorch#141270)

  Show description for 9156253
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  9156253

  Copy full SHA for 9156253

• [c10d][CI] Use new store for PG restart tests (pytorch#141374)

  Show description for af4522b
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  af4522b

  Copy full SHA for af4522b

• [dynamo] support list.sort sort non-constant iterable with constant keys (pytorch#141485)

  Show description for b18bbc9
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  b18bbc9

  Copy full SHA for b18bbc9

• cpp_wrapper tests: Fix tests assuming non-cpp_wrapper code (pytorch#141175)

  Show description for efec302
  

  benjaminglass1

  authored and

  pytorchmergebot

  committed
  efec302

  Copy full SHA for efec302

• Update triton wheel install script with new versioning (pytorch#141497)

  Show description for 78491d6
  

  benjaminglass1

  authored and

  pytorchmergebot

  committed
  78491d6

  Copy full SHA for 78491d6

• [FlexAttention] add support for learnable biases in Inductor (pytorch#137452)

  Show description for 91f7c54
  

  drisspg

  authored and

  pytorchmergebot

  committed
  91f7c54

  Copy full SHA for 91f7c54

• [MPS] Make MetalShaderLibrary usable from C++ (pytorch#141477)

  Show description for de6d69e
  

  malfet

  authored and

  pytorchmergebot

  committed
  de6d69e

  Copy full SHA for de6d69e

• [MPS][BE] Add virtual destructor (pytorch#141476)

  Show description for 953e5f9
  

  malfet

  authored and

  pytorchmergebot

  committed
  953e5f9

  Copy full SHA for 953e5f9

• [MPS] Move MetalShaderLibrary to its own header (pytorch#141475)

  Show description for b532a84
  

  malfet

  authored and

  pytorchmergebot

  committed
  b532a84

  Copy full SHA for b532a84

• [MPS][BE] Remove unused definitions (pytorch#141474)

  Show description for 1bca122
  

  malfet

  authored and

  pytorchmergebot

  committed
  1bca122

  Copy full SHA for 1bca122

• [inductor] Refactor dependencies.extract_loop_body_with_args (pytorch#141404)

  Show description for 9a09011
  

  jansel

  authored and

  pytorchmergebot

  committed
  9a09011

  Copy full SHA for 9a09011

• [CUTLASS] Lift shape & stride information as kernel args (pytorch#138611)

  Show description for 8f5edcb
  

  ColinPeppler

  authored and

  pytorchmergebot

  committed
  8f5edcb

  Copy full SHA for 8f5edcb

• Revert "export AOTI_TORCH_EXPORT on Windows. (pytorch#140030)"

  Show description for 2325749
  pytorchmergebot

  committed
  2325749

  Copy full SHA for 2325749

• [AMD] Add ncclRemoteError back (pytorch#141461)

  Show description for 4a378d7
  

  xw285cornell

  authored and

  pytorchmergebot

  committed
  4a378d7

  Copy full SHA for 4a378d7

• [ROCm] Select gpu targets according to PYTORCH_ROCM_ARCH when building AOTriton from source (pytorch#139432)

  Show description for 5ececd4
  

  vickytsang

  authored and

  pytorchmergebot

  committed
  5ececd4

  Copy full SHA for 5ececd4

• [ONNX] Use the torchlib opset number and fix opset import logic (pytorch#141413)

  Show description for 419b566
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  419b566

  Copy full SHA for 419b566

Pagination

Previous

Next