Commits

Breadcrumbs

History for

onpull-request

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 7, 2024

• [Intel GPU] Add XPU memory-related APIs (#129919)

  Show description for b53d97c
  

  guangyey

  authored and

  pytorchmergebot

  committed
  b53d97c

  Copy full SHA for b53d97c

Commits on Sep 6, 2024

• [dynamo] recursively skip frames when Dynamo cache limit is hit (#135144)

  Show description for 95e976a
  

  williamwen42

  authored and

  pytorchmergebot

  committed
  95e976a

  Copy full SHA for 95e976a

Commits on Sep 5, 2024

• Revise CPU vectorization ISA support API (#135075)

  Show description for 43dcb4b
  

  sanchitintel

  authored and

  pytorchmergebot

  committed
  43dcb4b

  Copy full SHA for 43dcb4b

Commits on Aug 30, 2024

• Revert "[c10d] Remove Option for ProcessGroup and Expose backend Options to reflect the correct code structure (#132931)"

  Show description for 351ba3e
  pytorchmergebot

  committed
  351ba3e

  Copy full SHA for 351ba3e

Commits on Aug 29, 2024

• [c10d] Remove Option for ProcessGroup and Expose backend Options to reflect the correct code structure (#132931)

  Show description for 65864d0
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  65864d0

  Copy full SHA for 65864d0

• Uses MemPoolContext to route allocations from CUDACachingAllocator (#134685)

  Show description for 4655eb3
  

  syed-ahmed

  authored and

  pytorchmergebot

  committed
  4655eb3

  Copy full SHA for 4655eb3

Commits on Aug 27, 2024

• Add compile time instruction count metric (#133834)

  Show description for d6091c8
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  d6091c8

  Copy full SHA for d6091c8

• [PT2/Profiler] Add Context Info to Torch-Compiled Regions (#132765)

  Show description for 0b81f70
  

  sraikund16

  authored and

  pytorchmergebot

  committed
  0b81f70

  Copy full SHA for 0b81f70

Commits on Aug 21, 2024

• add relevant function (#133946)

  Show description for a5ef04a
  

  FFFrog

  authored and

  pytorchmergebot

  committed
  a5ef04a

  Copy full SHA for a5ef04a

• [sparse] Add cuSPARSELt as a backend (#128534)

  Show description for 255cd75
  

  jcaip

  authored and

  pytorchmergebot

  committed
  255cd75

  Copy full SHA for 255cd75

Commits on Aug 20, 2024

• Add C API to return all torch function disablement status (#133136)

  Show description for 48ee098
  

  mlazos

  authored and

  pytorchmergebot

  committed
  48ee098

  Copy full SHA for 48ee098

• [Dynamo] Add torch function mode stack guard to dynamo (#133130)

  Show description for 09e366c
  

  mlazos

  authored and

  pytorchmergebot

  committed
  09e366c

  Copy full SHA for 09e366c

Commits on Aug 15, 2024

• [sigmoid] Support custom obj deserialization. (#133463)

  Show description for 59b3f59
  

  zhxchen17

  authored and

  pytorchmergebot

  committed
  59b3f59

  Copy full SHA for 59b3f59

• [Reland] Add wrappers for synchronous GPUDirect Storage APIs (#133489)

  Show description for 018e48c
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  018e48c

  Copy full SHA for 018e48c

Commits on Aug 9, 2024

• [Profiler] Add API for Dynamic Activity Toggling [2/n] (#133035)

  Show description for d2ecdcb
  

  sraikund16

  authored and

  pytorchmergebot

  committed
  d2ecdcb

  Copy full SHA for d2ecdcb

Commits on Aug 8, 2024

• Populate submodules of torch._C to sys.modules recursively (#132216)

  Show description for 24dee99
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  24dee99

  Copy full SHA for 24dee99

• Remove ProcessGroupRoundRobin (#132888)

  Show description for 0f90ffe
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  0f90ffe

  Copy full SHA for 0f90ffe

Commits on Aug 7, 2024

• Revert "Populate submodules of torch._C to sys.modules recursively (#132216)"

  Show description for ff81ca8
  pytorchmergebot

  committed
  ff81ca8

  Copy full SHA for ff81ca8

• Add padding_side to pad_sequence with "left" and "right" options ("right" as default) (#131884)

  Show description for 258f47f
  

  ringohoffman

  authored and

  pytorchmergebot

  committed
  258f47f

  Copy full SHA for 258f47f

• Grouped Query Attention (#132689)

  Show description for 8bc5ef5
  

  jainapurva

  authored and

  pytorchmergebot

  committed
  8bc5ef5

  Copy full SHA for 8bc5ef5

• add L2 cache size to device properties (#132819)

  Show description for 527f104
  

  nmacchioni

  authored and

  pytorchmergebot

  committed
  527f104

  Copy full SHA for 527f104

Commits on Aug 6, 2024

• torch.autograd.graph.increment_version: accept List[Tensor], use in AOTDispatcher (#132652)

  Show description for e3394e5
  

  bdhirsh

  authored and

  pytorchmergebot

  committed
  e3394e5

  Copy full SHA for e3394e5

• [c10d] Add a new API for adding ephemeral timeout for one local rank and the timeout will reset when the first collective finishes (#130905)

  Show description for 4e61092
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  4e61092

  Copy full SHA for 4e61092

Commits on Aug 4, 2024

• [BE] Format uncategorized Python files with ruff format (#132576)

  Show description for 4226ed1
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  4226ed1

  Copy full SHA for 4226ed1

Commits on Aug 2, 2024

• Revert "Grouped Query Attention (#128898)"

  Show description for bcb4f7c
  pytorchmergebot

  committed
  bcb4f7c

  Copy full SHA for bcb4f7c

Commits on Aug 1, 2024

• Populate submodules of torch._C to sys.modules recursively (#132216)

  Show description for 672ce46
  

  XuehaiPan

  authored and

  pytorchmergebot

  committed
  672ce46

  Copy full SHA for 672ce46

• Implements torch.cuda.MemPool() API (#131152)

  Show description for 7c89ec0
  

  syed-ahmed

  authored and

  pytorchmergebot

  committed
  7c89ec0

  Copy full SHA for 7c89ec0

Commits on Jul 31, 2024

• Grouped Query Attention (#128898)

  Show description for 6b28af1
  

  jainapurva

  authored and

  pytorchmergebot

  committed
  6b28af1

  Copy full SHA for 6b28af1

• Expose function to probe whether PyTorch was built with FlashAttention (#131894)

  Show description for f4f7aba
  

  lw

  authored and

  pytorchmergebot

  committed
  f4f7aba

  Copy full SHA for f4f7aba

• Merge torch.nn.utils.rnn type stubs (#131872)

  Show description for deb788f
  

  ringohoffman

  authored and

  pytorchmergebot

  committed
  deb788f

  Copy full SHA for deb788f

Commits on Jul 30, 2024

• Fix pyi annotation for ProcessGroupGloo.Options (#132080)

  Show description for 12b67bd
  

  ishon19

  authored and

  pytorchmergebot

  committed
  12b67bd

  Copy full SHA for 12b67bd

• Revert "Grouped Query Attention (#128898)"

  Show description for 499ead9
  pytorchmergebot

  committed
  499ead9

  Copy full SHA for 499ead9

Commits on Jul 29, 2024

• Grouped Query Attention (#128898)

  Show description for d039b14
  

  jainapurva

  authored and

  pytorchmergebot

  committed
  d039b14

  Copy full SHA for d039b14

• [Land Internally] MTIA equivalent of torch.cuda.memory_stats (#132007)

  Show description for dcb0310
  

  nautsimon

  authored and

  pytorchmergebot

  committed
  dcb0310

  Copy full SHA for dcb0310

• Fix pyi annotation for ProcessGroupNCCL.Options (#130957)

  Show description for 6c6fbb4
  

  ezyang

  authored and

  pytorchmergebot

  committed
  6c6fbb4

  Copy full SHA for 6c6fbb4

Pagination

Previous

Next