Commits

Branch selector

 patch-32

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 7, 2025

• Update upload-test-stats.yml

  jithunnair-amd

  authored
  2d1081e

  Copy full SHA for 2d1081e

• Update pytorch-probot.yml

  jithunnair-amd

  authored
  7abc0e4

  Copy full SHA for 7abc0e4

• Add sync-tag

  jithunnair-amd

  authored
  817c1ea

  Copy full SHA for 817c1ea

Commits on Nov 6, 2025

• Change name and add tag

  amdfaa

  authored
  8df4f3f

  Copy full SHA for 8df4f3f

• Create slow-rocm-mi200.yml

  amdfaa

  authored
  f103928

  Copy full SHA for f103928

• Remove rocm from slow workflow

  amdfaa

  authored
  afceb60

  Copy full SHA for afceb60

• Revert "[Inductor] addmm with bias -> unfuse bias if there is a pointwise/reduction consumer (#166165)"

  Show description for 9b4ac45
  pytorchmergebot

  committed
  9b4ac45

  Copy full SHA for 9b4ac45

• Fix boxcox to return same result for same input in one batch (#166986)

  Show description for a45a17f
  

  yunwu52

  authored and

  pytorchmergebot

  committed
  a45a17f

  Copy full SHA for a45a17f

• Fix flaky memory profiler test (#167168)

  Show description for c5593e7
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  c5593e7

  Copy full SHA for c5593e7

• Update pythoncapi_compat.h (#167138)

  Show description for c90a976
  

  colesbury

  authored and

  pytorchmergebot

  committed
  c90a976

  Copy full SHA for c90a976

• Move enrich_profiler_metadata config import out of gm.recompile() (#167114)

  Show description for d144382
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  d144382

  Copy full SHA for d144382

• Distributed Autotuning (#163369)

  Show description for 78827c5
  

  aorenste

  authored and

  pytorchmergebot

  committed
  78827c5

  Copy full SHA for 78827c5

• [ez] avoid log spam when random data is generated (#166919)

  Show description for ab1e734
  

  shunting314

  authored and

  pytorchmergebot

  committed
  ab1e734

  Copy full SHA for ab1e734

• Prevent torch._check causing graph breaks (#164676)

  Show description for 888958a
  

  2 peopleauthored and

  pytorchmergebot

  committed
  888958a

  Copy full SHA for 888958a

• [BE][Ez]: Update fmtlib submodule to 12.1.0 (#166983)

  Show description for d19f36b
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  d19f36b

  Copy full SHA for d19f36b

• [CUDA][cuBLASLt] addmm -- enable 2D bias in the Lt path when followed by an activation (#165548)

  Show description for 096c935
  

  nikitaved

  authored and

  pytorchmergebot

  committed
  096c935

  Copy full SHA for 096c935

• Add guidance on how to migrate kernels to the libtorch stable ABI (#167112)

  Show description for 03dea56
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  03dea56

  Copy full SHA for 03dea56

• [pp] Add reduce_grad Action (#166449)

  Show description for 2e83ae2
  

  IvanKobzarev

  authored and

  pytorchmergebot

  committed
  2e83ae2

  Copy full SHA for 2e83ae2

• [Inductor][Grouped Gemm] Add Blackwell CuTeDSL Kernel (#167182)

  Show description for 77b7097
  

  NikhilAPatel

  authored and

  pytorchmergebot

  committed
  77b7097

  Copy full SHA for 77b7097

• [Sigmoid][Delta Update][2/N] update delta update api to load original value first before casting to target dtype (#167039)

  Show description for c9b2db7
  

  jijunyan

  authored and

  pytorchmergebot

  committed
  c9b2db7

  Copy full SHA for c9b2db7

• [ROCm] Enable StaticCudaLauncher for ROCm (#166492)

  Show description for ba2e6b0
  

  chinmaydk99

  authored and

  pytorchmergebot

  committed
  ba2e6b0

  Copy full SHA for ba2e6b0

• Fix python -m build: error: unrecognized arguments: --no-build-isolation (#166848)

  Show description for 8523a64
  

  taras-janea

  authored and

  pytorchmergebot

  committed
  8523a64

  Copy full SHA for 8523a64

• [ROCm] Enable multi-arch compilation and unit tests for AOT Inductor (#166357)

  Show description for 9fef18e
  

  chinmaydk99

  authored and

  pytorchmergebot

  committed
  9fef18e

  Copy full SHA for 9fef18e

• [annotate][export] Add annotation to assertion nodes in export (#167171)

  Show description for aaea391
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  aaea391

  Copy full SHA for aaea391

• Update torch.var documentation to use modern API (#167209)

  Show description for 7206668
  

  mohsinm-dev

  authored and

  pytorchmergebot

  committed
  7206668

  Copy full SHA for 7206668

• Build libgomp (gcc-13) from src on AArch64 (#166549)

  Show description for 7729de0
  

  fadara01

  authored and

  pytorchmergebot

  committed
  7729de0

  Copy full SHA for 7729de0

• Add missing super().setUp() (#167163)

  Show description for 73078f3
  

  huydhn

  authored and

  pytorchmergebot

  committed
  73078f3

  Copy full SHA for 73078f3

• fix static_input_indices subclass remapping under training (#167127)

  Show description for ea7add4
  

  bdhirsh

  authored and

  pytorchmergebot

  committed
  ea7add4

  Copy full SHA for ea7add4

• [ROCm][CI] Run rocm.yml and inductor-rocm.yml every 3rd hour (#167220)

  Show description for 0ed4119
  

  amdfaa

  authored and

  pytorchmergebot

  committed
  0ed4119

  Copy full SHA for 0ed4119

• [Flight Recorder] Reverted to include stack traces for dump pipe triggered FR dump (#167023)

  Show description for 03fd2b7
  VieEeEw

  authored
  03fd2b7

  Copy full SHA for 03fd2b7

• [Inductor] Fix unbacked float symbol handling in kernel codegen (#166890)

  Show description for fd7bf9c
  

  karthickai

  authored and

  pytorchmergebot

  committed
  fd7bf9c

  Copy full SHA for fd7bf9c

• Update Sphinx dependencies (#164901)

  Show description for 41c9eee
  

  svekars

  authored and

  pytorchmergebot

  committed
  41c9eee

  Copy full SHA for 41c9eee

• `nn.Linear`: nD contiguous input + bias -- dispatch to addmm also when weight is sparse (#166071)

  Show description for bfc0ba4
  

  nikitaved

  authored and

  pytorchmergebot

  committed
  bfc0ba4

  Copy full SHA for bfc0ba4

• Make CUDA preload logic more straightforward (#167046)

  Show description for 3fdc5db
  

  malfet

  authored and

  pytorchmergebot

  committed
  3fdc5db

  Copy full SHA for 3fdc5db

• [inductor] Use runtime estimations in iterative sink waits pass (#167081)

  Show description for cc477f6
  

  IvanKobzarev

  authored and

  pytorchmergebot

  committed
  cc477f6

  Copy full SHA for cc477f6

Pagination

Previous

Next