Commits

Branch selector

 gh/laithsakka/254/head

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 12, 2025

• Update

  Show description for 3f8a222
  laithsakka

  committed
  3f8a222

  Copy full SHA for 3f8a222

• Update

  Show description for 0a38407
  laithsakka

  committed
  0a38407

  Copy full SHA for 0a38407

• Update

  Show description for bb0c12f
  laithsakka

  committed
  bb0c12f

  Copy full SHA for bb0c12f

Commits on Aug 9, 2025

• Update

  Show description for ed37641
  laithsakka

  committed
  ed37641

  Copy full SHA for ed37641

• Update

  Show description for 55997bb
  laithsakka

  committed
  55997bb

  Copy full SHA for 55997bb

• Update

  Show description for c174a70
  laithsakka

  committed
  c174a70

  Copy full SHA for c174a70

• Update

  Show description for d1e4f75
  laithsakka

  committed
  d1e4f75

  Copy full SHA for d1e4f75

• Update (base update)

  Show description for 98a579b
  laithsakka

  committed
  98a579b

  Copy full SHA for 98a579b

Commits on Aug 5, 2025

• Allow controlling PG backend and options via init_device_mesh (#159371)

  Show description for aeb5321
  

  lw

  authored and

  pytorchmergebot

  committed
  aeb5321

  Copy full SHA for aeb5321

• [inductor] consolidate common GEMM triton param retrieval (#159383)

  Show description for 625108e
  

  coconutruben

  authored and

  pytorchmergebot

  committed
  625108e

  Copy full SHA for 625108e

• Improve graph output alias with subclass error message (#159619)

  Show description for 09e5a93
  

  ezyang

  authored and

  pytorchmergebot

  committed
  09e5a93

  Copy full SHA for 09e5a93

• Generalize torch._C._set_allocator_settings to be generic (#156175)

  Show description for 908c5cc
  

  guangyey

  authored and

  pytorchmergebot

  committed
  908c5cc

  Copy full SHA for 908c5cc

• Deprecate overleap functions in CUDAAllocatorConfig, use AcceleratorAllocatorConfig instead (#156165)

  Show description for c114585
  

  guangyey

  authored and

  pytorchmergebot

  committed
  c114585

  Copy full SHA for c114585

• Refactor CUDAAllocatorConfig to reuse AcceleratorAllocatorConfig (#150312)

  Show description for ae1a706
  

  guangyey

  authored and

  pytorchmergebot

  committed
  ae1a706

  Copy full SHA for ae1a706

• Fix AllocatorConfig potential SIO issue (#159629)

  Show description for 56d19a5
  

  guangyey

  authored and

  pytorchmergebot

  committed
  56d19a5

  Copy full SHA for 56d19a5

• [Dynamo][Better Engineering] Type annotation for `torch/_dynamo/output_graph.py` (#159602)

  Show description for b6c5338
  

  Lucaskabela

  authored and

  pytorchmergebot

  committed
  b6c5338

  Copy full SHA for b6c5338

• skip XPU for dataloader CPU only unit test (#159811)

  Show description for 4fd5fab
  

  divyanshk

  authored and

  pytorchmergebot

  committed
  4fd5fab

  Copy full SHA for 4fd5fab

• [Inductor][Triton] Support TMA before strict 3.4 cutoff (#159777)

  Show description for bbc0df1
  

  njriasan

  authored and

  pytorchmergebot

  committed
  bbc0df1

  Copy full SHA for bbc0df1

• Remove pin on libuv from instructions (#159504)

  Show description for 33ec6e3
  

  hmaarrfk

  authored and

  pytorchmergebot

  committed
  33ec6e3

  Copy full SHA for 33ec6e3

• Add cascade sum support for Inductor CPP backend (#156296)

  Show description for efc4b46
  

  CaoE

  authored and

  pytorchmergebot

  committed
  efc4b46

  Copy full SHA for efc4b46

• [BE][MPS] Remove unused size12 variable (#159832)

  Show description for 1ca8388
  

  malfet

  authored and

  pytorchmergebot

  committed
  1ca8388

  Copy full SHA for 1ca8388

• [nativert] force resize to zero. (#159683)

  Show description for b694973
  

  dolpm

  authored and

  pytorchmergebot

  committed
  b694973

  Copy full SHA for b694973

• [C10D] fix slow init due to repeated dns resolution failure (#159596)

  Show description for 482f069
  

  wconstab

  authored and

  pytorchmergebot

  committed
  482f069

  Copy full SHA for 482f069

• Use uppercase OR when checking for system XNNPACK (#159527)

  Show description for 85d931f
  

  benjaminhottell

  authored and

  pytorchmergebot

  committed
  85d931f

  Copy full SHA for 85d931f

• Recursively sync fbgemm submodules before build (#159477)

  Show description for 8a2f53c
  

  jataylo

  authored and

  pytorchmergebot

  committed
  8a2f53c

  Copy full SHA for 8a2f53c

• Add `avg_pool3d` backward pass for MPS (#159089)

  Show description for b59b61a
  

  kurtamohler

  authored and

  pytorchmergebot

  committed
  b59b61a

  Copy full SHA for b59b61a

• Update torch-xpu-ops commit pin (#159621)

  Show description for 57ab39f
  

  CuiYifeng

  authored and

  pytorchmergebot

  committed
  57ab39f

  Copy full SHA for 57ab39f

• [Dynamo] Enable torch function dispatch on HOPs (#159708)

  Show description for 182975e
  

  mlazos

  authored and

  pytorchmergebot

  committed
  182975e

  Copy full SHA for 182975e

• [Dynamo] Fix arg ordering in tf modes (#159707)

  Show description for 9f8cfe7
  

  mlazos

  authored and

  pytorchmergebot

  committed
  9f8cfe7

  Copy full SHA for 9f8cfe7

• Fix failing test (#159800)

  Show description for e273ff0
  

  oulgen

  authored and

  pytorchmergebot

  committed
  e273ff0

  Copy full SHA for e273ff0

• [AOTI] don't allow int32 indices if {non-inf, > int32_max} upper bound is provided (#159433)

  Show description for 5e0fc2c
  

  davidberard98

  authored and

  pytorchmergebot

  committed
  5e0fc2c

  Copy full SHA for 5e0fc2c

Commits on Aug 4, 2025

• DeviceCopy should have the same layout as input (#159615)

  Show description for bc4b04e
  

  yushangdi

  authored and

  pytorchmergebot

  committed
  bc4b04e

  Copy full SHA for bc4b04e

• Revert "[inductor] add lowering for repeat_interleave.Tensor with output size specified (#147160) (#158462)" (#159798)

  Show description for 6b414f5
  

  davidberard98

  authored and

  pytorchmergebot

  committed
  6b414f5

  Copy full SHA for 6b414f5

• Revert "[mps] Turn on inductor dynamic shapes tests (#159456)"

  Show description for fb8f32e
  pytorchmergebot

  committed
  fb8f32e

  Copy full SHA for fb8f32e

• [Cutlass] Fix wrapper code generation breakage (#159760)

  Show description for 7ba996b
  

  mlazos

  authored and

  pytorchmergebot

  committed
  7ba996b

  Copy full SHA for 7ba996b

Pagination

Previous

Next