All pull requests

Pull requests

• [executorch][cuda] Run transposed convolutions on the implicit-GEMM template

  CLA Signedmeta-exported
  #25091·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Test 3-D convolutions on the implicit-GEMM template

  CLA Signedmeta-exported
  #25090·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Run 1-D convolutions on the implicit-GEMM template

  CLA Signedmeta-exported
  #25089·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Converted convolutions follow the output layout the graph recorded

  CLA Signedmeta-exported
  #25088·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Route 3-input-channel window and kernel-equals-stride convolutions to the implicit-GEMM template

  CLA Signedmeta-exported
  #25087·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add an implicit-GEMM template for dense triton::convolution

  CLA Signedmeta-exported
  #25086·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Route 2-D depthwise and few-channel grouped convolutions to a Triton direct-stencil kernel

  CLA Signedmeta-exported
  #25085·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add triton::convolution and the conv routing framework

  CLA Signedmeta-exported
  #25084·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add planned replacement to the Triton replacement pass

  CLA Signedmeta-exported
  #25083·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Deprecate triton_kernel_mode=OFF

  CLA Signedmeta-exported
  #25082·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• Do not elide views of control flow submodule inputs

  CLA Signedcommunity: contribution
  #25081·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• [XNNPACK] Do not delegate bilinear upsampling with a fractional scale factor

  CLA Signedcommunity: contribution
  #25080·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• [XNNPACK] Define a statically quantized bias once per consumer

  CLA Signedcommunity: contribution
  #25079·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• [XNNPACK] Run softmax in NCHW

  CLA Signedcommunity: contribution
  #25078·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• portable: return early from amax/amin on an empty input

  CLA Signedcommunity: contribution
  #25077·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• Use Python's sign convention in executorch_prim::mod

  CLA Signedcommunity: contribution
  #25076·AxelNoun opened Nov 14, 2026Contributor·

  ·
  1

  2

• Write ExecuTorch test owners to HUD ClickHouse

  CLA Signedmeta-exported
  #25069·psiddh opened Nov 14, 2026Contributor·

  ·
  4

• Report the Vulkan, WebGPU, QNN, OpenVINO and NXP backend suites to HUD

  CLA Signedmeta-exportedmodule: vulkan
  #25068·psiddh opened Nov 14, 2026Contributor·

  ·
  4

• [executorch][cuda] Add triton::convolution and the conv routing framework

  CLA Signed
  #25066·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add planned replacement to the Triton replacement pass

  CLA Signed
  #25065·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Deprecate triton_kernel_mode=OFF

  CLA Signed
  #25064·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add triton::convolution and the conv routing framework

  CLA Signed
  #25063·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Add planned replacement to the Triton replacement pass

  CLA Signed
  #25062·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• [executorch][cuda] Deprecate triton_kernel_mode=OFF

  CLA Signed
  #25061·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  2

• Read the CUDA cache counters once per autotune capture

  CLA Signed
  #25060·Gasoonjia opened Nov 14, 2026Contributor·

  ·
  7