Changes Of Note since Halide 17

• Ring-buffering now supported in schedules (Func::ring_buffer()). This is distinct from fold_storage in that it folds across time (the loop variables) rather than folding across space (the pure vars of the Func).
• Fixed a longstanding bug in lossless_cast()
• Lots of fixes for Vulkan backend
• OpenGLCompute is no longer supported
• Added support for ARM SVE2
• Added (basic) support for Intel APX and AVX10
• Added support for Hexagon HVX v68
• Added support for numpy's .npy format to .debug_to_file() and the code in halide_image_io.h
• Python bindings now support bfloat and int64 properly
• Hacky code that auto-named Funcs, Vars etc via DWARF introspection was removed
• The profiler was revamped to behave better when multiple Halide pipelines are in flight at the same time.
• Numerous lowering passes were sped up, resulting in faster compilation for large pipelines. However, time spent in LLVM is still the long pole for most pipelines.
• Fixed-point instruction selection has been improved via tracking constant integer bounds of expressions.
• Adds feature detection for ARM CPUs to the runtime library and to the host target feature computation. Supports Windows, macOS, Linux, iOS, and Android.

Deprecations / Removals

• tuple_select() has been removed in favor of overloads to select().
• Various fixed-point operators have been removed from the Halide::Internal namespace and are now in the public Halide namespace.

What's Changed

• Detect ARM CPU features for host target and in runtime (#8298)
• Scheduling directive to support ring buffering by @vksnk in https://github.com/halide/Halide/pull/7967
• Don't add ring_buffer semaphores if the function is not scheduled as async by @vksnk in https://github.com/halide/Halide/pull/8015
• Quick fix for crash that is occurring in SVE2 tests. by @zvookin in https://github.com/halide/Halide/pull/8020
• Don't use variable-length arrays by @steven-johnson in https://github.com/halide/Halide/pull/8021
• Set warnings on tests as well as src by @steven-johnson in https://github.com/halide/Halide/pull/8022
• Stronger chain detection in LoopCarry pass by @vksnk in https://github.com/halide/Halide/pull/8016
• adds mappings for f16 variants of halide float math by @mikewoodworth in https://github.com/halide/Halide/pull/8029
• Require LLVM >= 16.0 by @steven-johnson in https://github.com/halide/Halide/pull/8003
• Add test for #8029 by @steven-johnson in https://github.com/halide/Halide/pull/8032
• Tweak the Printer code in runtime for smaller code by @steven-johnson in https://github.com/halide/Halide/pull/8023
• Fix bounds_of_nested_lanes by @abadams in https://github.com/halide/Halide/pull/8039
• Track whether or not let expressions failed to solve in solver by @abadams in https://github.com/halide/Halide/pull/7982
• Fix type error in VectorizeLoops by @abadams in https://github.com/halide/Halide/pull/8055
• Update makefile to use test/common/terminate_handler.cpp by @abadams in https://github.com/halide/Halide/pull/8066
• add unsafe_promise_clamped by @wraith1995 in https://github.com/halide/Halide/pull/8071
• Don't require Halide_WebGPU when using wasm (#8063) by @steven-johnson in https://github.com/halide/Halide/pull/8065
• Outsmart the LLVM optimizer by @steven-johnson in https://github.com/halide/Halide/pull/8073
• Add hexagon_benchmarks app for CMake builds by @prasmish in https://github.com/halide/Halide/pull/8069
• Fix bool conversion bug in Vulkan code generator by @derek-gerstmann in https://github.com/halide/Halide/pull/8067
• Better validation of gpu schedules by @abadams in https://github.com/halide/Halide/pull/8068
• Add an easy way to print vectors in debug output. by @zvookin in https://github.com/halide/Halide/pull/8072
• [WebGPU] Update to latest native headers by @jrprice in https://github.com/halide/Halide/pull/8081
• Remove OpenGLCompute by @steven-johnson in https://github.com/halide/Halide/pull/8077
• Add checks to prevent people from using negative split factors by @abadams in https://github.com/halide/Halide/pull/8076
• Fix rfactor adding too many pure loops by @abadams in https://github.com/halide/Halide/pull/8086

New Contributors

• @tylerhou made their first contribution in https://github.com/halide/Halide/pull/8013
• @wraith1995 made their first contribution in https://github.com/halide/Halide/pull/8071
• @prasmish made their first contribution in https://github.com/halide/Halide/pull/8069
• @2022tgoel made their first contribution in https://github.com/halide/Halide/pull/8111
• @FabianSchuetze made their first contribution in https://github.com/halide/Halide/pull/8182
• @FindHao made their first contribution in https://github.com/halide/Halide/pull/8322

Full Changelog: https://github.com/halide/Halide/compare/v17.0.2...v18.0.0