Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Jul 30, 2025

• [Dynamo][Better Engineering] Add typing annotations to guard and source (#158397) (#159491)

  Show description for 2b1ae29
  

  Lucaskabela

  authored and

  pytorchmergebot

  committed
  2b1ae29View commit details

  Copy full SHA for 2b1ae29

  Browse repository at this point

• [MPS] Add `simd_[arg][max|min]` (#158990)

  Show description for 1293405
  

  malfet

  authored and

  pytorchmergebot

  committed
  1293405View commit details

  Copy full SHA for 1293405

  Browse repository at this point

• [dynamo, easy] add comment on skipping sys.monitoring frames (#159493)

  Show description for 3a65ff8
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  3a65ff8View commit details

  Copy full SHA for 3a65ff8

  Browse repository at this point

• Fix a bug of distributed 'gather' with uncontiguous tensors on the Gloo backend (#158903)

  Show description for acf13a9
  

  tiandeyu-cs

  authored and

  pytorchmergebot

  committed
  acf13a9View commit details

  Copy full SHA for acf13a9

  Browse repository at this point

• fix strategy hashing arg mismatch (#159506)

  Show description for 3a55676
  

  zpcore

  authored and

  pytorchmergebot

  committed
  3a55676View commit details

  Copy full SHA for 3a55676

  Browse repository at this point

• Don't use torch.backends.cuda.matmul.allow_tf32 in inductor cache key (#159480)

  Show description for af39144
  

  masnesral

  authored and

  pytorchmergebot

  committed
  af39144View commit details

  Copy full SHA for af39144

  Browse repository at this point

• [ATen][CUDA][cuFFT] Guard against deprecated error codes (#159466)

  Show description for 25343b3
  

  Aidyn-A

  authored and

  pytorchmergebot

  committed
  25343b3View commit details

  Copy full SHA for 25343b3

  Browse repository at this point

• [ContextParallel][FlexAttention] Prototype of supporting FlexAttention in Context Parallel (#158692)

  Show description for 07fad04
  

  XilunWu

  authored and

  pytorchmergebot

  committed
  07fad04View commit details

  Copy full SHA for 07fad04

  Browse repository at this point

• Revert "Fix rand_like decomposition to preserve strides (#159294)"

  Show description for 7ac70ac
  pytorchmergebot

  committed
  7ac70acView commit details

  Copy full SHA for 7ac70ac

  Browse repository at this point

• [Code Motion]Restructure flex attention kernel into flex subdirectory (#159437)

  Show description for e221a1c
  

  drisspg

  authored and

  pytorchmergebot

  committed
  e221a1cView commit details

  Copy full SHA for e221a1c

  Browse repository at this point

• [c10d] Fix setGroupName and setGroupDesc in `group_split` and `merge_remote_group` (#159429)

  Show description for 4defea1
  

  Junjie Wang (PyTorch)

  authored and

  pytorchmergebot

  committed
  4defea1View commit details

  Copy full SHA for 4defea1

  Browse repository at this point

• [ROCm CI] Migrate to MI325 Capacity. (#159059)

  Show description for 53d68b9
  

  2 peopleauthored and

  pytorchmergebot

  committed
  53d68b9View commit details

  Copy full SHA for 53d68b9

  Browse repository at this point

• [PP] Fix zero bubble schedules for eval() (#159475)

  Show description for f74842d
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  f74842dView commit details

  Copy full SHA for f74842d

  Browse repository at this point

• Fix TestAutogradFallback flaky tests under Dynamo: migrate to lib._destroy() (#159443)

  Show description for 644fee2
  

  xmfan

  authored and

  pytorchmergebot

  committed
  644fee2View commit details

  Copy full SHA for 644fee2

  Browse repository at this point

• [BE] Clarify comment to not revert when command has been edited (#159495)

  Show description for 7821fbc
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  7821fbcView commit details

  Copy full SHA for 7821fbc

  Browse repository at this point

• [ONNX] RMS Norm (#159377)

  Show description for 73ee323
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  73ee323View commit details

  Copy full SHA for 73ee323

  Browse repository at this point

• Update CODEOWNERS for ONNX (#159390)

  Show description for 176c644
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  176c644View commit details

  Copy full SHA for 176c644

  Browse repository at this point

• Fix large_tensor_test skipping cpu (#158617)

  Show description for debc059
  

  drisspg

  authored and

  pytorchmergebot

  committed
  debc059View commit details

  Copy full SHA for debc059

  Browse repository at this point

• Remove `/d2implyavx512upperregs-` flag (#159431)

  Show description for 0df78f0
  

  malfet

  authored and

  pytorchmergebot

  committed
  0df78f0View commit details

  Copy full SHA for 0df78f0

  Browse repository at this point

• Add CPython test for heapq (#159370)

  Show description for d0e8a0e
  

  guilhermeleobas

  authored and

  pytorchmergebot

  committed
  d0e8a0eView commit details

  Copy full SHA for d0e8a0e

  Browse repository at this point

• [BE]: Update CUTLASS submodule to 4.1.0 (#158854)

  Show description for 2249284
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  2249284View commit details

  Copy full SHA for 2249284

  Browse repository at this point

• fixed typo error (#159451)

  Show description for 5c14315
  

  Rohit Manav

  authored and

  pytorchmergebot

  committed
  5c14315View commit details

  Copy full SHA for 5c14315

  Browse repository at this point

• [BE] Make PyObjectSlot use a global PyInterpreter and remove (#158427)

  Show description for 1b99c18
  

  PaliC

  authored and

  pytorchmergebot

  committed
  1b99c18View commit details

  Copy full SHA for 1b99c18

  Browse repository at this point

• [Graph Partition] add graph partition doc (#159450)

  Show description for 435edbc
  

  BoyuanFeng

  authored and

  pytorchmergebot

  committed
  435edbcView commit details

  Copy full SHA for 435edbc

  Browse repository at this point

• Revert "Fix max_width computation in _tensor_str._Formatter (#126859)"

  Show description for 6c6e11c
  pytorchmergebot

  committed
  6c6e11cView commit details

  Copy full SHA for 6c6e11c

  Browse repository at this point

• [Profiler] Fix lost C call events problem in Python 3.12.0-3.12.4 (#155446)

  Show description for a775c8e
  

  D-D-H

  authored and

  pytorchmergebot

  committed
  a775c8eView commit details

  Copy full SHA for a775c8e

  Browse repository at this point

• [inductor] Fix mm decomposition evaluating symints (#158998)

  Show description for 24d07b3
  

  azahed98

  authored and

  pytorchmergebot

  committed
  24d07b3View commit details

  Copy full SHA for 24d07b3

  Browse repository at this point

• Various bugfixes for running NanoGPT training (#159166)

  Show description for 90fd06b
  

  jamesjwu

  authored and

  pytorchmergebot

  committed
  90fd06bView commit details

  Copy full SHA for 90fd06b

  Browse repository at this point

• [DCP] Improve error handling for process based async checkpointing (#159374)

  Show description for 002f188
  

  meetv18

  authored and

  pytorchmergebot

  committed
  002f188View commit details

  Copy full SHA for 002f188

  Browse repository at this point

• Move Half to headeronly (#159172)

  Show description for 259e79e
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  259e79eView commit details

  Copy full SHA for 259e79e

  Browse repository at this point

• [RPC][TensorPipe] Fix import torch if compiled without TensorPipe (#159461)

  Show description for ee343ce
  

  Aidyn-A

  authored and

  pytorchmergebot

  committed
  ee343ceView commit details

  Copy full SHA for ee343ce

  Browse repository at this point

• unflatten closure (#159418)

  Show description for ea53691
  

  avikchaudhuri

  authored and

  pytorchmergebot

  committed
  ea53691View commit details

  Copy full SHA for ea53691

  Browse repository at this point

• Move Float4 to headeronly (#159414)

  Show description for b268f22
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  b268f22View commit details

  Copy full SHA for b268f22

  Browse repository at this point

• [dynamo][guards] Skip no tensor aliasing guard on inbuilt nn module buffers (#159453)

  Show description for 52a52d1
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  52a52d1View commit details

  Copy full SHA for 52a52d1

  Browse repository at this point

• Revert "Move Half to headeronly (#159172)"

  Show description for eaadd12
  pytorchmergebot

  committed
  eaadd12View commit details

  Copy full SHA for eaadd12

  Browse repository at this point