halide/Halide v21.0.0

on GitHub

Release highlights

We have deliberately skipped version 20.0.0 to align with the LLVM version we are now using. Note that LLVM 21.1.1 or higher is required as LLVM 21.1.0 has a major bug in the NVPTX backend.

Major changes

• The rfactor scheduling directive was rewritten and enhanced. It is now compatible with autoschedulers.
• The Mullapudi2016 autoscheduler now supports experimental GPU scheduling.
• The Python bindings have been substantially improved, with many missing bindings filled in.
• HL_DEBUG_CODEGEN gained a new filtering mode. Debug levels can now be set on a per-file/per-function basis.
• Support was added for AMD Zen5 and the iOS Simulator.
• The strict_float feature has been reimplemented and should be much more reliable.
• Lots of bugfixes, performance improvements, and build system improvements. We spent a lot of time fixing issues with our testing infrastructure and are looking forward to implementing a more stable contribution experience going forward.

Deprecations

• LLVM 19 and below are no longer supported, in keeping with our support policy.
• Halide_BUNDLE_STATIC will be removed in the next release. If you are using it, please migrate to the shared library instead.
• Support for Python 3.8 has been dropped.

Changelog

Scheduling

• The rfactor scheduling directive was rewritten and enhanced.
  • Rewrite the rfactor scheduling directive by @alexreinking in #8490
  • Dequalify names when constructing RVars in rfactor by @alexreinking in #8560
  • Add promise_clamped in rfactor by @alexreinking in #8608
  • Add rfactor patterns for NaN-propagating min/max by @alexreinking in #8587
• The Mullapudi2016 autoscheduler now supports experimental GPU scheduling.
  • GPU autoscheduling with Mullapudi2016: the reference implementation by @antonysigma in #7787
  • Mullapudi2016-GPU: Reorder to avoid for-loops to be sandwiched between gpu_blocks. by @antonysigma in #8647
  • Enable experimental Mullapudi2016 GPU scheduler for test-bench by @antonysigma in #8650
  • Highlight Metal GPU code in stmt_html by @antonysigma in #8659
  • Always ensure gpu_threads count >= warp size of 32 by @antonysigma in #8656
• Fix incorrect natural vector size on Zen4 by @abadams in #8570
• Make it an error to use a device extern stage without target support by @abadams in #8794
• Add support for adding tuple outputs in the configure() method by @abadams in #8649

Python

• Fix argument order in rpow by @alexreinking in #8677
• Drop support for Python 3.8 by @alexreinking in #8678
• Fix segfault in RDom's operator<< by @alexreinking in #8679
• Use ruff to format and lint Python code by @alexreinking in #8684
• Get raw Runtime::Buffer from Buffer in Python rather than use PyBuffer by @alexreinking in #8682
• Bind in-place update operators (e.g. +=) in Python by @alexreinking in #8683
• Clean up Python dependencies; document uv usage by @alexreinking in #8694
• Fix several printing segfaults. by @alexreinking in #8700
• Add Python bindings for serialization by @alexreinking in #8718
• Add all remaining IROperator ops to Python bindings by @alexreinking in #8771
• Fix up memoize; bind to Python by @alexreinking in #8778
• Fix invalid Python type annotation and return types (#8772) by @rtzam in #8773
• Expose Runtime::Buffer::cropped to C++ and Python Buffer by @rtzam in #8787

Debugging

• New feature flag to allow for stack backtrace/unwind by @mcourteaux in #8703
• Add filtering capabilities to HL_DEBUG_CODEGEN by @alexreinking in #8627
• Adding worker_thread_idle() for more informative profiling by @slomp in #8719
• Color IR output in cout and cerr. by @mcourteaux in #8635
• Improve output format for lowering passes timing. by @mcourteaux in #8749
• fix(stmt-html): Fix embedded Buffer processing performance issue. by @mcourteaux in #8748
• Use AArch64 assembly syntax on macOS with LLVM<22 by @alexreinking in #8710

CodeGen

• Mark our PTX kernels as kernels, to stop them from being stripped by @abadams in #8571
• Math functions renaming table for GPU backends to support vectorized evaluation of math functions. by @mcourteaux in #8595
• Apply version constraints to iOS objects by @alexreinking in #8546
• Redirect bitwise ops to logical ops in case the arguments are bool. by @mcourteaux in #8597
• scalarize select condition for LLVM where possible by @abadams in #8575
• Add missing addition simplifier rules by @abadams in #8630
• Bounds and alignment analysis through bitwise ops by @abadams in #8574
• Make the vld2 pattern more obviously profitable by @abadams in #8765
• Fix vector shuffle for Vulkan CodeGen by @derek-gerstmann in #8621
• Suppress warning on Windows for duplicate constant symbols. by @mcourteaux in #8555
• Use lossless_cast for saturating casts from unsigned to signed on x86 by @abadams in #8527
• AMD Zen5 support by @changhoon-sung in #8612

Compiler

• Rework strict_float to use individual op intrinsics instead by @abadams in #8641
• Don't cache mutations of Exprs that have only one reference to them by @abadams in #8518
• Only use the nodes-visited set for nodes with multiple refs by @abadams in #8547
• In graph_equal(), call the correct implementation for comparing equalities between statements and expressions by @BachiLi in #8611

Runtime

• Support copying the overlapping region from one buffer to another. by @mcourteaux in #8463
• Add (iOS) simulator target feature. by @alexreinking in #8623
• Opt out of JIT exceptions by @abadams in #8615
• Experimental: support removing unused runtime functions via
  HL_RUNTIME_DROP_FUNCS environment variable.
  • PoC feature: drop functions from the runtime by @mcourteaux in #8653

Apps

• The onnx app now builds with CMake:
  • Add CMake for onnx app by @vawale in #8707
  • Fix halide_as_onnx_backend_test by @alexreinking in #8784

Documentation

• Add note about relative paths to readme by @abadams in #8613

Bugfixes

• Fix #8534 [Buffer serialization does not match deserialization] by @abadams in #8535
• Fix CUDA HTML code printing bug. by @mcourteaux in #8558
• Fix halide_get_cpu_features() linkage to avoid name mangling issues by @derek-gerstmann in #8573
• Fix for #8578 by @mcourteaux in #8579
• Fix shuffle bug in CodeGen C. by @mcourteaux in #8567
• Check if expression is defined before trying to compute its constant_integer_bounds by @vksnk in #8599
• Drop invalid "in-bounds" GEP for constant offsets by @alexreinking in #8768
• Record trace_loads directly on ImageParam. by @alexreinking in #8803
• RewriteLoadsAs32Bit should use the mutated index by @rootjalex in #8581
• Set any_strict_float for wrapper module if target has strict_flag feature by @vksnk in #8681
• Fix wrong type of the bound by @vksnk in #8781
• Fix UB-introducing rewrite in FindIntrinsics by @abadams in #8539
• Fix rewrite that doesn't preserve type by @abadams in #8674
• Fix nested select handling in remove_undef by @abadams in #8669
• Add an underlying type to the halide_buffer_flags to prevent UB in C++ by @mcourteaux in #8690

Testing / CI

• Limit depth more strictly in CSE fuzz test by @abadams in #8512
• Skip fast exp/log/pow/sin/cosine tests without sse 4.1 by @abadams in #8541
• Hopefully fix flaky mullapudi reorder test by @abadams in #8542
• Skip test when code could be using x87 by @abadams in #8537
• Fix stale GPU lifetime management tests for Vulkan. by @derek-gerstmann in #8601
• Upgrade runner for cmake_cmake_file_lists job by @alexreinking in #8609
• Buildbot fixes by @alexreinking in #8706
• Fix the pip packaging workflow by @alexreinking in #8708
• Fix complexity of bounds of nested pure intrinsics by @abadams in #8689
• Skip two sub-tests on llvm 21.1 by @abadams in #8782
• Speed up simd_op_check_wasm by @abadams in #8780
• Reduce the beam size in the adams2019 apps test to avoid timeouts by @abadams in #8786
• Workaround llvm slow compile time bug in Mullapudi overlap test by @abadams in #8793
• Restore concurrent behavior to gpu_allocation_cache test by @abadams in #8792
• Revert "Skip two sub-tests on llvm 21.1" by @abadams in #8806
• Fix WASM splat op check test. by @mcourteaux in #8705

Build

• Fix workflow for next release by @alexreinking in #8514
• Fix Debian packaging by @alexreinking in #8524
• Remove llvm version check from Makefile by @abadams in #8533
• Drop deprecated / unsupported setups for Halide 20 by @alexreinking in #8508
• Fix check for Windows never having aligned_alloc available. by @mcourteaux in #8551
• Don't include CMAKE_INSTALL_PREFIX when LIBDIR is absolute by @alexreinking in #8552
• Add target-nvptx to target-all in vcpkg.json by @alexreinking in #8562
• Fix top of LLVM, and remove upper limit of LLVM version from CMakeLists. by @mcourteaux in #8568
• build_halide_h asserts that every header it slurps in is one of the args by @abadams in #8559
• Upgrade pybind11 to 2.11.1 by @alexreinking in #8616
• Drop check for LLVM_LIBCXX in FindHalide_LLVM.cmake by @alexreinking in #8617
• Fix finding LLD on Homebrew when multiple versions are installed. by @alexreinking in #8619
• Fix build on GCC 15 (Comes with Fedora 42). by @mcourteaux in #8626
• Constrain Clang and LLD searches to LLVM version by @alexreinking in #8634
• Disallow empty CMAKE_BUILD_TYPE on single-config generators by @alexreinking in #8651
• Add missing outputs to add_halide_library; fix advice in Lesson 21. by @alexreinking in #8660
• Bump the LLVM version in the pip package to 20.1.8 by @alexreinking in #8698
• Prefer to build against libjpeg-turbo and document this. by @alexreinking in #8775
• Add C++17 requirement to RunGenMain CMake target by @alexreinking in #8795
• Allow llvm-ar in BundleStatic.cmake by @alexreinking in #8799
• Fix dubious find_package logic in test/generator by @alexreinking in #8804
• Warning when extra-output is requested w/o filename by @FabianSchuetze in #8671
• Makefile linker flag fixes and cleanups by @abadams in #8764

Ongoing maintenance

• Fix clang-tidy-19 errors by @steven-johnson in #8509
• Remove unused function in HexagonOptimize by @steven-johnson in #8511
• Fix two non-idiomatic uses of node_type by @abadams in #8520
• Handle some misc TODOs by @abadams in #8528
• Use a consistent idiom for visit_let by @abadams in #8540
• Upgrade to clang-format 19 by @alexreinking in #8543
• Suppress clang-tidy warning for make_with_shape_of() by @steven-johnson in #8545
• Remove debugging print left in by @abadams in #8572
• Fixes for llvm trunk by @abadams in #8590
• Another fix for llvm trunk by @abadams in #8591
• Our internal error macros were redesigned:
  • Accurately annotate Error system with [[noreturn]] by @alexreinking in #8564
  • Teach compilers that internal_error does not return. by @alexreinking in #8807
  • Use a new macro trick to avoid throwing in destructors. by @alexreinking in #8774
• Move to opaque llvm pointers by @abadams in #8614
• Avoid throwing from a destructor in PartitionLoops.cpp by @alexreinking in #8767
• Bump version to 21.0.0 by @alexreinking in #8810
• remove superfluous overload that causes compile errors by @ongjunjie in #8654
• Remove obsolete WasmExecutor specific debug macro. by @zvookin in #8670
• Add missing header by @vksnk in #8680
• Attempted fix for LLVM change by @abadams in #8642
• Fix top LLVM: renamed NVPTX barrier intrinsics. by @mcourteaux in #8631

New Contributors

• @changhoon-sung made their first contribution in #8612
• @vawale made their first contribution in #8707
• @rtzam made their first contribution in #8773

Full Changelog: v19.0.0...v21.0.0