TVM Monthly - October 2023

Note: This monthly report contains main branch only.

As discussed by the TVM PMC, our goal is to provide a monthly summary of the project so users and developers can get a better understanding of the goings on of the TVM community.

Feedback and suggestions are welcomed so that we can further improve these updates.

Community

None.

We continue to improve CI, Frontend and other Runtimes. Note again: This monthly report contains main branch only.

Pull Requests

The below is high-level summary of the PRs closed in the last month grouped by area.

Adreno & ArmComputeLibrary

  • #15897 - [Adreno][CI]Few updates to Adreno docker setup
  • #15990 - [ACL] Update Compute Library to v23.08

Arith

  • #15881 - Simplify the result of non-divisible floordiv

CMSIS-NN

  • #15747 - Move CMSIS_5 from SHA to release based upgrade

Frontend

  • #15502 - [relay]aten::copy_ support for pytorch
  • #15821 - [TFLite]Support quantized ELU
  • #15844 - [TFLite]Fix test failures caused by div-by-zero

LLVM

  • #15921 - Fix for llvm CodeGenOpt API change

Relay

  • #15928 - Expose qnn ops directly from relay.qnn module
  • #15889 - fix the conflicted documentation description

TIR & TVMScript

  • #15913 - [TIR] Fix offset_factor in cuda tensor core intrins
  • #15906 - [TIR] Fix the error example in the documentation for pad_einsum
  • #15953 - [TIR] Pretty print TIR LLVM function name
  • #15972 - delete print extra info at parsing

micoNPU

  • #15929 - [microNPU][ETHOSU] Fix rounding mode in requantize operation
  • #15780 - [microNPU][ETHOSU] MatMul legalization support

CI & Docker & BugFix

  • CI
    • #16018 - [Tests] Fix work_dir location used by test_micro_tuning_with_meta_schedule
    • #16019 - [Tests] Check int8+int32 testcases in test_estimate_peak_flops_cpu
    • #16017 - [Tests] Fix str vs. int comparison in test_num_threads
    • #15903 - [Target]Add LLVM functions for current system info
    • #15836 - Update ci-gpu image
  • Docker
    • #15799 - Add LLVM 17 to the LLVM install script
    • #15862 - Upgrade oneflow to v0.8.0
  • BugFix
    • #16000 - [Fix][TIR]fix symbolic strides lower
    • #15970 - [Hotfix] Mark python-FFI handling with TVM_DLL
    • #15965 - [CI]Better to pass the build folder
    • #15891 - [Relay]fix axis parsing of repeat converter in the MXNet frontend
    • #15873 - [Fix] Remove duplicated words from comments, NFC
    • #15868 - [Relay]Fix conv transpose with default strides in ONNX frontend

Misc

  • #16015 - [Pylint] fix pylint issues for cblas
  • #15955 - [FFI][Python] Handle error propagation when line number is missing
  • #15982 - Bump werkzeug from 2.2.3 to 3.0.1 in /apps/microtvm
  • #15966 - [CMake] Fix order of GNUInstallDirs module
  • #15952 - Update ci_arm Docker tag
  • #15940 - [Minor] Fix compilation warnings for clang
  • #15947 - Bump urllib3 from 1.26.9 to 1.26.18 in /docker/python
  • #15835 - [CodeGenC][Redo] Handle GlobalVar callee as internal function call
  • #15945 - Bump urllib3 from 1.26.15 to 1.26.18 in /apps/microtvm
  • #15934 - [Release] [Dont Squash] Modify version number to 0.14.0 and 0.15.0.dev on main branch
  • #15847 - [release] Update version to 0.14.0 and 0.15.0.dev on main branch
  • #15867 - Bump pillow from 9.3.0 to 10.0.1 in /apps/microtvm/ethosu
  • #15866 - Bump pillow from 9.3.0 to 10.0.1 in /apps/microtvm/cmsisnn
  • #15865 - Bump pillow from 9.2.0 to 10.0.1 in /apps/microtvm
  • #15833 - [VM] Memory Manager moved up to runtime
  • #15859 - [Script] Fix miscs of make_notes.py