CVS log for pkgsrc/parallel/pocl/Makefile

Up to [cvs.NetBSD.org] / pkgsrc / parallel / pocl

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

*: bump for llvm 19 (shlib major changed)

pocl: add LLVM 19 support

pocl: mask out on i386 due to float16 usage

*: recursive bump for icu 77 and libxml2 2.14

*: recursive bump for icu 76 shlib major version bump

*: revbump for icu downgrade

*: recursive bump for icu 76.1 shlib bump

pocl: needs atomic64

*: replace CMAKE_ARGS with CMAKE_CONFIGURE_ARGS

pocl: updated to 6.0

PoCL 6.0

Highlights

Support for Clang/LLVM 18.0
New driver: cpu-tbb, uses Intel OneTBB library for scheduling
CPU driver new features: OpenMP support, new extensions (cl_pocl_svm_rect...), clGetDeviceAndHostTimer supported
Remote driver new features: coarse-grained SVM, vsock, clCompileProgram/clLinkProgram...
LevelZero driver new features: host synchronization optimization, >4GB buffers (CL_DEVICE_MAX_MEM_ALLOC_SIZE)
CUDA driver new features: CL_DEVICE_MAX_MEM_ALLOC_SIZE increased

revbump after icu and protobuf updates

pocl: updated to 5.0

5.0
http://portablecl.org/docs/html/notes_5_0.html

*: recursive bump for icu 74.1

pocl: Update to 4.0. Unbreaks with latest LLVM.

parallel: Adapt packages (where possible) to USE_(CC|CXX)_FEATURES

revbump after textproc/icu update

massive revision bump after textproc/icu update

pocl: mark as BROKEN

This needs to be updated to 3.0 (available) to support llvm 14.

revbump for textproc/icu update

revbump for icu and libffi

pocl: update to 1.8.

Notable User Facing Changes
---------------------------

- support for LLVM 13
- CMake: Inter-Procedural Optimization is enabled on code of runtime library
  (libpocl.so is compiled with -flto on systems that support it).
- LTTng tracing improved - more command types are traced, and also
  some synchronous API calls (like clCreateBuffer) are traced.
- poclcc, tests and examples can be disabled with CMake options
- Valgrind support improved by making Valgrind aware of pocl's
  reference counting of cl_* objects
- kernels which are called by kernels are now force-inlined
- Support for NetBSD.
- Support for Unix systems without libdl.
- PoCL can now (optionally) respond to SIGUSR2 by printing
  some live debug information.
- improved SPIR support for CUDA devices

Notable Bug Fixes
-----------------

- Fixed a potential crash on Unix systems without sysfs mounted.
- Fixed compilation errors when building on macOS.
  - Fixed POCL_FAST_INIT macro; POCL_INIT_LOCK must be invoked with only one argument.
  - Fix bin/poclcc to not depend on OpenCL 2.0 symbols
- Fixed miscompilation in kernel loops with multiple conditionals with barriers in them.

Other
-----
- Add cmake options PARALLEL_COMPILE_JOBS, PARALLEL_LINK_JOBS to
  use ninja's seperate compile and link job pools.

- Improve memory architecture, buffer migration and allocation.
  Buffers are now allocated on a device when first used
  (previously each buffer was allocated on every device in context).

- the single global LLVMContext was replaced with
  multiple LLVMContexts, one per OpenCL cl_context.
  OpenCL code can now be compiled in parallel
  when using separate cl_contexts. This feature
  is disabled by default since it significantly slowed
  down PyOpenCL. This should be resolved by separating
  LLVM compilation in their own threads in the future.

- a new OpenCL extension was added to PoCL: cl_pocl_content_size.
  The extension allows the user to give optimization hint to PoCL,
  which will be used internally by PoCL to optimize buffer transfers
  between multiple devices.

pocl: work around broken CPU detection on aarch64

pocl: add test target

pocl: Fix a null pointer dererence if sysfs is not present *sigh*

pocl: fix name of vendor directory

add parallel/pocl

OpenCL (Open Computing Language) is an open, royalty-free standard for
cross-platform, parallel programming of diverse accelerators found in
supercomputers, cloud servers, personal computers, mobile devices and embedded
platforms.

PoCL is a portable open source (MIT-licensed) implementation of the OpenCL
standard (1.2 with some 2.0 features supported). In addition to being an easily
portable multi-device (truely heterogeneous) open-source OpenCL implementation,
a major goal of this project is improving interoperability of diversity of
OpenCL-capable devices by integrating them to a single centrally orchestrated
platform.