Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 5, 2024

• [pp] pipelining + dcp unit test (#139633)

  Show description for 75eeefb
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  75eeefb

  Copy full SHA for 75eeefb

• Add Autograd Fallback for MTIA (#139211)

  Show description for 1a70185
  

  egienvalue

  authored and

  pytorchmergebot

  committed
  1a70185

  Copy full SHA for 1a70185

• Migrate inductor-perf-test-nightly.yml to use linux.aws.a100 (#139657)

  Show description for 59b6694
  

  jeanschmidt

  and

  huydhn

  authored
  59b6694

  Copy full SHA for 59b6694

• [hop free symbols] refactor tensor.to_list implementation to call wrap_fx_proxy. (#139663)

  Show description for 6734cb7
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  6734cb7

  Copy full SHA for 6734cb7

• [aarch64] add CUDA 12.6 to docker for sbsa wheel (#138562)

  Show description for ae86939
  

  tinglvv

  authored and

  pytorchmergebot

  committed
  ae86939

  Copy full SHA for ae86939

• [fr][rfc] use a logger to control output for flight recorder analyzer (#139656)

  Show description for d549ddf
  

  c-p-i-o

  authored and

  pytorchmergebot

  committed
  d549ddf

  Copy full SHA for d549ddf

• Add repro instructions to fx_graph_runnable.py (#139481)

  Show description for b9f0563
  

  zou3519

  authored and

  pytorchmergebot

  committed
  b9f0563

  Copy full SHA for b9f0563

• [dynamo][NFC] Remove some dead code paths (#139674)

  Show description for 01bcf37
  

  StrongerXi

  authored and

  pytorchmergebot

  committed
  01bcf37

  Copy full SHA for 01bcf37

• [dynamo] Add `is_mutable()` and `is_immutable()` methods to `VariableTracker` (#139341)

  Show description for 2b3a227
  

  StrongerXi

  authored and

  pytorchmergebot

  committed
  2b3a227

  Copy full SHA for 2b3a227

• [dynamo][NFC] Move `MutationType` classes into `variables/base.py` (#139340)

  Show description for 0ba3962
  

  StrongerXi

  authored and

  pytorchmergebot

  committed
  0ba3962

  Copy full SHA for 0ba3962

• [dynamo][NFC] Rename `mutable_local` and add documentation (#139339)

  Show description for 693a0a1
  

  StrongerXi

  authored and

  pytorchmergebot

  committed
  693a0a1

  Copy full SHA for 693a0a1

• [PGNCCL] Watchdog prints call-time traceback when reporting timeout (#139659)

  Show description for 5f2ed50
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  5f2ed50

  Copy full SHA for 5f2ed50

• [SymmetricMemory] introduce a binding for cuMemset32Async (#138755)

  Show description for ee42a99
  

  Yifu Wang

  authored and

  pytorchmergebot

  committed
  ee42a99

  Copy full SHA for ee42a99

• [AOTAutograd] Handle edge cases for donated buffer & enable in oss (#139669)

  Show description for 87059d4
  

  BoyuanFeng

  authored and

  pytorchmergebot

  committed
  87059d4

  Copy full SHA for 87059d4

• Optimize mutable torch.library.custom_op overhead (#139513)

  Show description for 27ec392
  

  zou3519

  authored and

  pytorchmergebot

  committed
  27ec392

  Copy full SHA for 27ec392

• handle more devices in method_type method of TensorVariable (#138078)

  Show description for 9dc5851
  

  tbohutyn

  authored and

  pytorchmergebot

  committed
  9dc5851

  Copy full SHA for 9dc5851

• [export] Dedup data-dependent errors based on stacktrace (#139540)

  Show description for de509ab
  

  angelayi

  authored and

  pytorchmergebot

  committed
  de509ab

  Copy full SHA for de509ab

• [inductor] Error on unsupported autotuner configs (#139658)

  Show description for cc25b6d
  

  SamGinzburg

  authored and

  pytorchmergebot

  committed
  cc25b6d

  Copy full SHA for cc25b6d

• [logging][ez] Add timer logging for pickling and unpickle for object based collective (#139757)

  Show description for 41e4d88
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  41e4d88

  Copy full SHA for 41e4d88

• Use Manylinux2_28 for wheel builds (#138732)

  Show description for 5860c8e
  

  atalman

  authored and

  pytorchmergebot

  committed
  5860c8e

  Copy full SHA for 5860c8e

• End TritonBundle on non-cache write codepaths (#139698)

  Show description for c0d21b6
  

  oulgen

  authored and

  pytorchmergebot

  committed
  c0d21b6

  Copy full SHA for c0d21b6

• Revert "[dynamo][guards] Consider tensors as immutable for dict tag matches (#139560)"

  Show description for 4d5cc1b
  pytorchmergebot

  committed
  4d5cc1b

  Copy full SHA for 4d5cc1b

• Use clang-tidy 17 (#139678)

  Show description for a2bc2e3
  

  cyyever

  authored and

  pytorchmergebot

  committed
  a2bc2e3

  Copy full SHA for a2bc2e3

• HACK: use FB proxy for testowners (#139473)

  Show description for e0156f9
  

  ezyang

  authored and

  pytorchmergebot

  committed
  e0156f9

  Copy full SHA for e0156f9

• [Torch Elastic] Fix the bug caused by wrong host address in creating TCPStore server inside dynamic rendezvous (#139702)

  Show description for 13eb3b3
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  13eb3b3

  Copy full SHA for 13eb3b3

• [CUDA][CI][cusparselt] Only CUDA 11.8 ships the libcusparseLt.so.0, CUDA 12 would use PYPI libcusparselt (#138547)

  Show description for 53f164c
  

  nWEIdia

  authored and

  pytorchmergebot

  committed
  53f164c

  Copy full SHA for 53f164c

• Fix compiler collective TORCH_TRACE and improve code state printing (#139716)

  Show description for 349cd49
  

  ezyang

  authored and

  pytorchmergebot

  committed
  349cd49

  Copy full SHA for 349cd49

• [7/N] Don't skip ASAN on some tests (#139675)

  Show description for 546318e
  

  cyyever

  authored and

  pytorchmergebot

  committed
  546318e

  Copy full SHA for 546318e

• Fix for gcc10 torch.compile compiler error when march=aarch64+sve (#137795)

  Show description for f551d90
  

  robert-hardwick

  authored and

  pytorchmergebot

  committed
  f551d90

  Copy full SHA for f551d90

• Deprecate `torch._utils.is_compiling()` and `torch._dynamo.external_utils.is_compiling()` (#127690)

  Show description for e84d112
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  e84d112

  Copy full SHA for e84d112

• Fix torch.histc not checking min > max on cuda for int8 tensors (#139372)

  Show description for ffb7a08
  

  zeshengzong

  authored and

  pytorchmergebot

  committed
  ffb7a08

  Copy full SHA for ffb7a08

• [Intel GPU] Avoid target_link_libraries twice for torch_xpu_ops which will potentially cause multiple definition symbol linker error. (#139024)

  Show description for 356fc41
  

  etaf

  authored and

  pytorchmergebot

  committed
  356fc41

  Copy full SHA for 356fc41

• use torch.sym_sum instead of incremental sum in _cat_meta (#139653)

  Show description for 6ad52db
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  6ad52db

  Copy full SHA for 6ad52db

• Fix existing lint issues in ir.py (#139237)

  Show description for 51a3d6d
  

  aorenste

  authored and

  pytorchmergebot

  committed
  51a3d6d

  Copy full SHA for 51a3d6d

• RMSNorms docs - remove biases initialization (#139620)

  Show description for b2f5a53
  

  elisim

  authored and

  pytorchmergebot

  committed
  b2f5a53

  Copy full SHA for b2f5a53

Pagination

Previous

Next