Version 4.0.0

Version 4.0.0

Feature Additions:

• RISC-V support for host target (either cross compiled or native)
• Features can be enabled for host targets using the cmake
  option CA_HOST_TARGET_<ARCH>_FEATURES or the environment variable
  CA_HOST_TARGET_FEATURES. This uses a format similar to -mattr on tools like
  opt e.g. "+v;-zfencei".
• The compiler-passes directory has been created to just allow important
  compiler passes to be made available externally such as the vectorizer vecz.
• Support for SPIR 1.2 programs (cl_khr_spir) has been dropped.

Upgrade guidance:

• LLVM 17 and 18 only are supported in this release.
• Added support for a Remote HAL over tcp/ip, primarily for demoing new RISC-V
  targets quickly
• The cmake variable CA_HOST_TARGET_CPU has been split into capitalized
  architecture variants of the form CA_HOST_TARGET_<ARCH>_CPU e.g
  CA_HOST_TARGET_X86_64_CPU, CA_HOST_TARGET_AARCH64_CPU and
  CA_HOST_TARGET_RISCV64_CPU. The environment variable CA_HOST_TARGET_CPU
  remains the same name. Note that CA_HOST_TARGET_CPU_NATIVE is no longer
  supported but can be achieved by using native as a value for the variants.
• The mux spec has been bumped:
  • 0.77.0: to loosen the requirements on the mux event type used by
    DMA builtins.
  • 0.78.0: to introduce mux builtins for sub-group, work-group, and
    vector-group operations.
  • 0.79.0: to introduce mux builtins for sub-group shuffle operations.
  • 0.80.0: to introduce support for 64-bit atomic operations.
• The compiler::ImageArgumentSubstitutionPass now replaces sampler typed
  parameters in kernel functions with i32 parameters via a wrapper function.
• The host target as a consequence now passes samplers to kernels as 32-bit
  integer arguments, not as integer arguments disguised as pointer values.
• The compiler::utils::ReplaceBarriersPass has been replaced with the
  compiler::utils::LowerToMuxBuiltinsPass.
• The compiler::utils::HandleBarriersPass has been renamed to the
  compiler::utils::WorkItemLoopsPass.
• The compiler::utils::createLoop API has moved its list of IVs parameter
  into its compiler::utils::CreateLoopOpts parameter. It can now also set the
  IV names via a second CreateLoopOpts field.
• Support for LLVM versions is now limited to LLVM 17 and LLVM 18. Support for
  earlier LLVM versions has been removed.
• Support for FMA (fused multiply-add) is required for the device. For the host
  device for x86-64, this means only x86-64-v3 and newer are supported. This
  roughly translates to 2015 or newer, both for Intel and for AMD.
• Although hardware support for FMA is available on all platforms we currently
  test, if you are using OCK on a platform we do not test and encounter
  issues, please let us know by opening an issue!
• compiler-utils library has been split into compiler-pipeline and
  compiler-binary-metadata to allow use of compiler pipeline utilities without
  the binary metadata requirements. Both will be needed for mux targets.
• The utility function addParamToAllFunctions has been moved to
  ReplaceLocalModuleScopeVariablesPass and renamed as it is only used there.
• OpenCL-Headers now fetches from the github repo with tag v2024.05.08.This can
  be overridden using -DFETCHCONTENT_SOURCE_DIR_OPENCL_HEADERS to point to a
  different repo.

OCK Demo Package 2024-03-06-f211f42

ock-demo build 2024-03-06-f211f42

OCK daily 2024-03-05-a390b0f

Daily build 2024-03-05-a390b0f

OCK Demo Package 2024-02-29-f0588da

ock-demo build 2024-02-29-f0588da

OCK daily 2024-02-06

Daily build 2024-02-06

OCK daily 2024-01-27

Daily build 2024-01-27

OCK daily 2024-01-20

Daily build 2024-01-20

OCK daily 2024-01-15-691b289

Daily build 2024-01-15-691b289

OCK daily 2024-01-08-ec45b4f

Daily build 2024-01-08-ec45b4f

Version 3.0.0

Version 3.0.0

Upgrade guidance:

• The mux spec has been bumped:
  • 0.74.0: to account for changes to BaseModule.
  • 0.75.0: to remove code supporting cl_codeplay_program_snapshot.
  • 0.76.0: to add code supporting cl_intel_required_subgroup_size.
• The cl_codeplay_program_snapshot extension has been removed.
• cl::binary::ProgramInfo and cl::binary::KernelInfo have been removed and
  replaced with equivalent structures in the compiler namespace.
• compiler::Module::finalize no longer takes a callback to populate program
  info - it now takes compiler::ProgramInfo by pointer and populates it
  itself if passed a non-null address.

Feature additions:

• muxc now supports OpenCL C inputs, before running a (now optional) pass
  pipeline. A device is required for this to work.The file type is worked out
  from the file extension, and LLVM IR is assumed otherwise. Options can be
  provided to this step with -cl-options. This also provides the ability to
  select a device based on its index from 0, using --device-idx <idx>.
• The cl_intel_required_subgroup_size OpenCL extension is now supported. Note
  that no in-tree targets currently report any supported sub-group sizes so no
  kernels compiled with a required sub-group size will successfully compile.
• The SPIR-V SubgroupSize execution mode is now supported via the
  SubgroupDispatch capability. Note that other aspects of this capability are
  not supported.
• The SPIR-V OptNoneINTEL capability is now supported.
• compiler::BaseModulePassMachinery can now be given compiler options to
  guide its decisions.
• A new method - compiler::BaseModulePassMachinery::handlePipelineElement -
  has been added to allow more customizable target-specific pipeline component

Future feature dropping for next major release (4.0.0):

• SPIR 1.2 is deprecated and will be removed.
• LLVM 15 support will switch to LLVM 17.