Commits

Branch selector

 main

User selector

nirvedhmeshram

Datepicker

All time

Commit History

Commits on Apr 16, 2024

• Add boolean option to do RELU in mlp plugin (iree-org#17058)

  Show description for ace6397
  nirvedhmeshram

  authored
  ace6397

  Copy full SHA for ace6397

Commits on Apr 15, 2024

• Add ability to call the same custom dispatch multiple times when using pdl patterns. (iree-org#16967)

  Show description for e36844f
  nirvedhmeshram

  authored
  e36844f

  Copy full SHA for e36844f

Commits on Apr 3, 2024

• Revert "Revert "Add PDL pre-processing pass to compiler" (iree-org#16968)" (iree-org#16969)

  Show description for d33f216
  nirvedhmeshram

  authored
  d33f216

  Copy full SHA for d33f216

• Add PDL pre-processing pass to compiler (iree-org#16945)

  Show description for 180e458
  nirvedhmeshram

  authored
  180e458

  Copy full SHA for 180e458

Commits on Feb 20, 2024

• Integrate llvm 20240219 (iree-org#16483)

  nirvedhmeshram

  authored
  cd18fdf

  Copy full SHA for cd18fdf

Commits on Feb 2, 2024

• [NFC] Update typo in iree-e2e-matmul-test.cc (iree-org#16293)

  nirvedhmeshram

  authored
  cddd084

  Copy full SHA for cddd084

• Add a reference matmul to e2e matmul tests (iree-org#16280)

  Show description for a957fde
  nirvedhmeshram

  authored
  a957fde

  Copy full SHA for a957fde

Commits on Oct 31, 2023

• [ROCM] add device path and use it to setup device (iree-org#15234)

  Show description for 2706526
  

  nirvedhmeshram

  and

  PhaneeshB

  authored
  2706526

  Copy full SHA for 2706526

Commits on Sep 29, 2023

• [LLVMGPU] Convert maximumf/minimumf to max/min for ROCM (iree-org#15069)

  Show description for 113f9d2
  nirvedhmeshram

  authored
  113f9d2

  Copy full SHA for 113f9d2

Commits on Oct 31, 2022

• Extend subspan byte offset folding to GPU MMA load/store (iree-org#10958)

  Show description for 2b2db6e
  nirvedhmeshram

  authored
  2b2db6e

  Copy full SHA for 2b2db6e

Commits on Oct 30, 2022

• Update cooperative matrix pipeline to use upstream flow (iree-org#10931)

  Show description for c557c16
  

  nirvedhmeshram

  and

  antiagainst

  authored
  c557c16

  Copy full SHA for c557c16

Commits on Jun 29, 2022

• Disable warp reduction for LLVMGPU backends that are not CUDA (iree-org#9631)

  nirvedhmeshram

  authored
  0a48b3b

  Copy full SHA for 0a48b3b

Commits on Jun 23, 2022

• Drop hc.bc from the list of device libs pulled in by ROCM backend (iree-org#9600)

  nirvedhmeshram

  authored
  b7f3a68

  Copy full SHA for b7f3a68

• Fix rocm fill buffer function (iree-org#9595)

  Show description for 6332652
  nirvedhmeshram

  authored
  6332652

  Copy full SHA for 6332652

Commits on Jun 21, 2022

• Fix offset error in ROCM command buffer (iree-org#9562)

  nirvedhmeshram

  authored
  2cad0ed

  Copy full SHA for 2cad0ed

Commits on Jun 14, 2022

• Some updates to build external rocm (iree-org#9502)

  nirvedhmeshram

  authored
  bba260d

  Copy full SHA for bba260d

Commits on May 17, 2022

• [ROCM] Add iree-hal-rocm-dump-isa flag (iree-org#9141)

  nirvedhmeshram

  authored
  6cb2475

  Copy full SHA for 6cb2475

Commits on May 16, 2022

• Add software pipeline depth to lowering config (iree-org#9114)

  nirvedhmeshram

  authored
  5a00d42

  Copy full SHA for 5a00d42

Commits on May 13, 2022

• Remove comma between fields in LoweringConfig assemblyFormat (iree-org#9116)

  Show description for c8a68f1
  nirvedhmeshram

  authored
  c8a68f1

  Copy full SHA for c8a68f1

Commits on May 9, 2022

• [CUDA] Allow async copy for fp16 and copy chunks of 8 elements for vector load (iree-org#8949)

  nirvedhmeshram

  authored
  fe1f493

  Copy full SHA for fe1f493

Commits on Apr 19, 2022

• Simplify builtin fill/splat dispatches (iree-org#8919)

  nirvedhmeshram

  authored
  b970332

  Copy full SHA for b970332

Commits on Apr 12, 2022

• Fix tracing macro for IREE_HAL_CUDA_ALLOCATOR_ID (iree-org#8852)

  nirvedhmeshram

  authored
  b788b1b

  Copy full SHA for b788b1b

• Add tracing macro for IREE_HAL_CUDA_ALLOCATOR_ID (iree-org#8845)

  nirvedhmeshram

  authored
  fed9cca

  Copy full SHA for fed9cca

Commits on Apr 8, 2022

• Add workgroup swizzling for better cache reuse (iree-org#8789)

  Show description for 793b9cf
  nirvedhmeshram

  authored
  793b9cf

  Copy full SHA for 793b9cf

Commits on Mar 26, 2022

• [CUDA] Add pattern to collapse vector.transfer_read to 2D when possible (iree-org#8636)

  Show description for 5420350
  nirvedhmeshram

  authored
  5420350

  Copy full SHA for 5420350

Commits on Mar 4, 2022

• [CUDA] Expose cuda default device selection as a flag (iree-org#8455)

  nirvedhmeshram

  authored
  a9dc253

  Copy full SHA for a9dc253

Commits on Feb 26, 2022

• Adding dynamic shared memory for cuda (iree-org#8405)

  nirvedhmeshram

  authored
  878952e

  Copy full SHA for 878952e

Commits on Feb 14, 2022

• Adding BatchMatmulOp to use tensorcore when possible (iree-org#8282)

  nirvedhmeshram

  authored
  b959830

  Copy full SHA for b959830