Commits

Branch selector

 main

User selector

All users

All time

Commit History

Commits on Jan 31, 2026

• [LinalgExt] Add OuterReduction tiling strategy for ArgCompareOp (#23102)

  Show description for 0d71d25
  bangtianliu

  authored
  0d71d25

  Copy full SHA for 0d71d25

Commits on Jan 30, 2026

• [DispatchCreation] Support split reduction on weight backward CNHW layout (#23343)

  Show description for 41cae6d
  yzhang93

  authored
  41cae6d

  Copy full SHA for 41cae6d

• [CI] Update iree-test-suite to run convolution tests (#23312)

  Show description for 8659263
  amd-eochoalo

  authored
  8659263

  Copy full SHA for 8659263

• [LLVMCPU] Allow generic conv ops too in KernelDispatch (#23295)

  Show description for fce95cd
  Abhishek-Varma

  authored
  fce95cd

  Copy full SHA for fce95cd

• [TensorExt] Propagate and fold iree_tensor_ext.bitcast (#23182)

  Show description for 18fe092
  Yu-Zhewen

  authored
  18fe092

  Copy full SHA for 18fe092

• [StableHLO] Rework zero-extent canonicalizer (#23227)

  Show description for c0885ee
  sommerlukas

  authored
  c0885ee

  Copy full SHA for c0885ee

• iree-bazel-* improvements for handling multiple targets + options. (#23330)

  Show description for 3824de7
  

  benvanik

  and

  claude

  authored
  3824de7

  Copy full SHA for 3824de7

• [CI][Torch models] Update golden dispatch counts (#23336)

  efric

  authored
  c6ac63e

  Copy full SHA for c6ac63e

Commits on Jan 29, 2026

• Integrate llvm/llvm-project@db6b186e (#23328)

  Show description for ab71aab
  IanWood1

  authored
  ab71aab

  Copy full SHA for ab71aab

• Switching iree_time_now() to CLOCK_MONOTONIC + cleanup. (#23325)

  Show description for 4f92211
  

  benvanik

  and

  claude

  authored
  4f92211

  Copy full SHA for 4f92211

• [Dispatch Creation] Restrict broadcasting consumer fusion (#23232)

  Show description for 8c014e5
  IanWood1

  authored
  8c014e5

  Copy full SHA for 8c014e5

• [Integrate] Prioritize IREE's PackOp/UnPackOp bufferization implementation. (#23326)

  Show description for 79166ed
  hanhanW

  authored
  79166ed

  Copy full SHA for 79166ed

• [plugins][Torch] Add a flag to externalize transients during torch-to-IREE conversion (#23224)

  Show description for 24af126
  zjgarvey

  authored
  24af126

  Copy full SHA for 24af126

• [GPU] Don't swap expand with slice in the same block (#23267)

  Show description for ea72d00
  Max191

  authored
  ea72d00

  Copy full SHA for ea72d00

• [Stream] Verify `flow.executable` before lowering (#23262)

  Show description for 34d2446
  sommerlukas

  authored
  34d2446

  Copy full SHA for 34d2446

• [CI] Try reenabling onnx ops tests for rdna4 (#23324)

  Show description for 8f8b29c
  kuhar

  authored
  8f8b29c

  Copy full SHA for 8f8b29c

• [Codegen][GPU] Add subgroup and lane execution scopes (#23097)

  Show description for 602aa40
  qedawkins

  authored
  602aa40

  Copy full SHA for 602aa40

• [GPU] Support some collapse_shape ops in GPUReduceBankConflicts (#23301)

  Show description for 96f3974
  Max191

  authored
  96f3974

  Copy full SHA for 96f3974

• [Stream] Restrict import of dynamically shaped tensors (#23214)

  Show description for d88f560
  sommerlukas

  authored
  d88f560

  Copy full SHA for d88f560

• [Input] Do not attempt to convert LLVM dialect functions (#23299)

  Show description for a7328f7
  sommerlukas

  authored
  a7328f7

  Copy full SHA for a7328f7

• [LinalgExt] Add pattern to canonicalize identity map_gather into a linalg.copy (#23240)

  Show description for 709ab4f
  Abhishek-Varma

  authored
  709ab4f

  Copy full SHA for 709ab4f

• [GPU] MmaSchedule configuration crashes when lacking PerfTflops (#23303)

  Show description for e968799
  rsuderman

  authored
  e968799

  Copy full SHA for e968799

• [AMDGPU][LDS] Support linearized DMA for small innermost dimensions (#23056)

  Show description for ea01b8a
  lialan

  authored
  ea01b8a

  Copy full SHA for ea01b8a

Commits on Jan 28, 2026

• [GPU] Add workgroupMemoryBankCount parameter to TargetWgpAttr (#23273)

  Show description for fc06a5d
  Muzammiluddin-Syed-ECE

  authored
  fc06a5d

  Copy full SHA for fc06a5d

• [compiler][NFC] Follow camelCase naming convention. (#23316)

  Show description for 315cac2
  hanhanW

  authored
  315cac2

  Copy full SHA for 315cac2

• Integrate llvm/llvm-project@3446ff1 (#23302)

  Show description for 31f794a
  IanWood1

  authored
  31f794a

  Copy full SHA for 31f794a

• Revert "[CPU] Support dynamic attention by tiling K1 when needed." (#23313)

  Show description for 36ac5f4
  hanhanW

  authored
  36ac5f4

  Copy full SHA for 36ac5f4

• [CI] Update golden time for gfx1201 (#23310)

  Show description for d9aec69
  amd-eochoalo

  authored
  d9aec69

  Copy full SHA for d9aec69

• [CPU] Support dynamic attention by tiling K1 when needed. (#23304)

  hanhanW

  authored
  761bd9c

  Copy full SHA for 761bd9c

• [CI] disable failing cts tests (#23300)

  Show description for e3dfd29
  amd-eochoalo

  authored
  e3dfd29

  Copy full SHA for e3dfd29

• [GlobalOptimization] Fix output_shape handling in SinkTransposeThroughExpandShape (#23308)

  Show description for 81b508d
  

  qedawkins

  and

  claude

  authored
  81b508d

  Copy full SHA for 81b508d

• [Encoding] Add verifier for encoding_dims on (Un)SetEncodingOp (#23245)

  Show description for fbc4499
  jtuyls

  authored
  fbc4499

  Copy full SHA for fbc4499

• Implement missing stablehlo.fft operations (#22829)

  Show description for 839085a
  pstarkcdpr

  authored
  839085a

  Copy full SHA for 839085a

Commits on Jan 27, 2026

• [Codegen] Use safer hoisting in OptimizeTensorInsertExtractSlices (#23280)

  Show description for 789859e
  Max191

  authored
  789859e

  Copy full SHA for 789859e

• [AMDGPU][LDS] Adding 1k, 2k, 4k, 8k static shape e2e tests for coalesced gather DMA op (#22884)

  Show description for af093a8
  lialan

  authored
  af093a8

  Copy full SHA for af093a8