344 lines
16 KiB
XML
344 lines
16 KiB
XML
<?xml version='1.0' encoding='UTF-8' standalone='no'?>
|
|
<doxygen xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="compound.xsd" version="1.9.1" xml:lang="en-US">
|
|
<compounddef id="release-3-2-0" kind="page">
|
|
<compoundname>release-3-2-0</compoundname>
|
|
<title>Release 3.2.0 (2021/07/29)</title>
|
|
<tableofcontents>
|
|
<tocsect>
|
|
<name>Download</name>
|
|
<reference>release-3-2-0_1release-3-2-0_download</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>System Requirements</name>
|
|
<reference>release-3-2-0_1release-3-2-0_system_requirements</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Working Items</name>
|
|
<reference>release-3-2-0_1release-3-2-0_working_items</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>New Features</name>
|
|
<reference>release-3-2-0_1release-3-2-0_new_features</reference>
|
|
<tableofcontents>
|
|
<tocsect>
|
|
<name>Taskflow Core</name>
|
|
<reference>release-3-2-0_1release-3-2-0_taskflow_core</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>cudaFlow</name>
|
|
<reference>release-3-2-0_1release-3-2-0_cudaflow</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>syclFlow</name>
|
|
<reference>release-3-2-0_1release-3-2-0_syclflow</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>CUDA Standard Parallel Algorithms</name>
|
|
<reference>release-3-2-0_1release-3-2-0_cuda_std_algorithms</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Utilities</name>
|
|
<reference>release-3-2-0_1release-3-2-0_utilities</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Taskflow Profiler (TFProf)</name>
|
|
<reference>release-3-2-0_1release-3-2-0_profiler</reference>
|
|
</tocsect>
|
|
</tableofcontents>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Bug Fixes</name>
|
|
<reference>release-3-2-0_1release-3-2-0_bug_fixes</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Breaking Changes</name>
|
|
<reference>release-3-2-0_1release-3-2-0_breaking_changes</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Deprecated and Removed Items</name>
|
|
<reference>release-3-2-0_1release-3-2-0_deprecated_items</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Documentation</name>
|
|
<reference>release-3-2-0_1release-3-2-0_documentation</reference>
|
|
</tocsect>
|
|
<tocsect>
|
|
<name>Miscellaneous Items</name>
|
|
<reference>release-3-2-0_1release-3-2-0_miscellaneous_items</reference>
|
|
</tocsect>
|
|
</tableofcontents>
|
|
<briefdescription>
|
|
</briefdescription>
|
|
<detaileddescription>
|
|
<para>Taskflow 3.2.0 is the 3rd release in the 3.x line! This release includes several new changes such as CPU-GPU tasking, algorithm collection, enhanced web-based profiler, documentation, and unit tests.</para>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_download">
|
|
<title>Download</title>
|
|
<para>Taskflow 3.2.0 can be downloaded from <ulink url="https://github.com/taskflow/taskflow/releases/tag/v3.2.0">here</ulink>.</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_system_requirements">
|
|
<title>System Requirements</title>
|
|
<para>To use Taskflow v3.2.0, you need a compiler that supports C++17:</para>
|
|
<para><itemizedlist>
|
|
<listitem><para>GNU C++ Compiler at least v8.4 with -std=c++17 </para>
|
|
</listitem>
|
|
<listitem><para>Clang C++ Compiler at least v6.0 with -std=c++17 </para>
|
|
</listitem>
|
|
<listitem><para>Microsoft Visual Studio at least v19.27 with /std:c++17 </para>
|
|
</listitem>
|
|
<listitem><para>AppleClang Xcode Version at least v12.0 with -std=c++17 </para>
|
|
</listitem>
|
|
<listitem><para>Nvidia CUDA Toolkit and Compiler (nvcc) at least v11.1 with -std=c++17 </para>
|
|
</listitem>
|
|
<listitem><para>Intel C++ Compiler at least v19.0.1 with -std=c++17 </para>
|
|
</listitem>
|
|
<listitem><para>Intel DPC++ Clang Compiler at least v13.0.0 with -std=c++17 and SYCL20</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
Taskflow works on Linux, Windows, and Mac OS X.</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_working_items">
|
|
<title>Working Items</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>enhancing support for SYCL with Intel DPC++ </para>
|
|
</listitem>
|
|
<listitem><para>enhancing parallel CPU and GPU algorithms </para>
|
|
</listitem>
|
|
<listitem><para>designing pipeline interface and its scheduling algorithms</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_new_features">
|
|
<title>New Features</title>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_taskflow_core">
|
|
<title>Taskflow Core</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>added <ref refid="classtf_1_1SmallVector" kindref="compound">tf::SmallVector</ref> optimization for optimizing the dependency storage in a graph </para>
|
|
</listitem>
|
|
<listitem><para>added move constructor and move assignment operator for <ref refid="classtf_1_1Taskflow" kindref="compound">tf::Taskflow</ref><itemizedlist>
|
|
<listitem><para><ref refid="classtf_1_1Taskflow_1afd790de6db6d16ddf4729967c1edebb5" kindref="member">tf::Taskflow::Taskflow(Taskflow&&)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Taskflow_1aa4957a41e63e1d1a6f77c540d70d04fa" kindref="member">tf::Taskflow::operator=(Taskflow&&)</ref> </para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
<listitem><para>added moved run in <ref refid="classtf_1_1Executor" kindref="compound">tf::Executor</ref> for automatically managing taskflow's lifetimes<itemizedlist>
|
|
<listitem><para><ref refid="classtf_1_1Executor_1a4b97752c00b788918cb8396fbb1b3ad0" kindref="member">tf::Executor::run(Taskflow&&)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Executor_1a2d0dc32d81100202168e42ac1d3dd8f9" kindref="member">tf::Executor::run(Taskflow&&, C&&)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Executor_1a04aa012598622e807dba3272aa464d7e" kindref="member">tf::Executor::run_n(Taskflow&&, size_t)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Executor_1ae5c5c6a1c1f6d4740414c86daa6a3901" kindref="member">tf::Executor::run_n(Taskflow&&, size_t, C&&)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Executor_1a724d188df2952f64c1f2c60555a0a611" kindref="member">tf::Executor::run_until(Taskflow&&, P&&)</ref></para>
|
|
</listitem><listitem><para><ref refid="classtf_1_1Executor_1ad7aa69f06309b9226a4046f8c400144e" kindref="member">tf::Executor::run_until(Taskflow&&, P&&, C&&)</ref></para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect2>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_cudaflow">
|
|
<title>cudaFlow</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>improved the execution flow of <ref refid="classtf_1_1cudaFlowCapturer" kindref="compound">tf::cudaFlowCapturer</ref> when updates involve</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
New algorithms in <ref refid="classtf_1_1cudaFlow" kindref="compound">tf::cudaFlow</ref> and <ref refid="classtf_1_1cudaFlowCapturer" kindref="compound">tf::cudaFlowCapturer</ref>:</para>
|
|
<para><itemizedlist>
|
|
<listitem><para>added tf::cudaFlow::reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::transform_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::uninitialized_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::transform_uninitialized_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::inclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::exclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::transform_inclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::transform_exclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::merge </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::merge_by_key </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::sort </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::sort_by_key </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::find_if </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::min_element </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlow::max_element </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::transform_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::uninitialized_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::transform_uninitialized_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::inclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::exclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::transform_inclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::transform_exclusive_scan </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::merge </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::merge_by_key </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::sort </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::sort_by_key </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::find_if </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::min_element </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaFlowCapturer::max_element </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cudaLinearCapturing</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect2>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_syclflow">
|
|
<title>syclFlow</title>
|
|
</sect2>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_cuda_std_algorithms">
|
|
<title>CUDA Standard Parallel Algorithms</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>added <ref refid="namespacetf_1a7c449cec0b93503b8280d05add35e9f4" kindref="member">tf::cuda_for_each</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a01ad7ce62fa6f42f2f2fbff3659b7884" kindref="member">tf::cuda_for_each_index</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a3ed764530620a419e3400e1f9ab6c956" kindref="member">tf::cuda_transform</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a8a872d2a0ac73a676713cb5be5aa688c" kindref="member">tf::cuda_reduce</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a492e8410db032a0273a99dd905486161" kindref="member">tf::cuda_uninitialized_reduce</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a4463d06240d608bc31d8b3546a851e4e" kindref="member">tf::cuda_transform_reduce</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added tf::cuda_transform_uninitialized_reduce </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a2e1b44c84a09e0a8495a611cb9a7ea40" kindref="member">tf::cuda_inclusive_scan</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1aeb391c40120844318fd715b8c3a716bb" kindref="member">tf::cuda_exclusive_scan</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1afa4aa760ddb6efbda1b9bab505ad5baf" kindref="member">tf::cuda_transform_inclusive_scan</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a2e739895c1c73538967af060ca714366" kindref="member">tf::cuda_transform_exclusive_scan</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a37ec481149c2f01669353033d75ed72a" kindref="member">tf::cuda_merge</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1aa84d4c68d2cbe9f6efc4a1eb1a115458" kindref="member">tf::cuda_merge_by_key</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a06804cb1598e965febc7bd35fc0fbbb0" kindref="member">tf::cuda_sort</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a3461b9179221dd7230ce2a0e45156c7f" kindref="member">tf::cuda_sort_by_key</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a5f9dabd7c5d0fa5166cf76d9fa5a038e" kindref="member">tf::cuda_find_if</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a572c13198191c46765264f8afabe2e9f" kindref="member">tf::cuda_min_element</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="namespacetf_1a3fc577fd0a8f127770bcf68bc56c073e" kindref="member">tf::cuda_max_element</ref></para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect2>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_utilities">
|
|
<title>Utilities</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>added CUDA meta programming </para>
|
|
</listitem>
|
|
<listitem><para>added SYCL meta programming</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect2>
|
|
<sect2 id="release-3-2-0_1release-3-2-0_profiler">
|
|
<title>Taskflow Profiler (TFProf)</title>
|
|
</sect2>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_bug_fixes">
|
|
<title>Bug Fixes</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>fixed compilation errors in constructing tf::cudaRoundRobinCapturing </para>
|
|
</listitem>
|
|
<listitem><para>fixed compilation errors of TLS worker pointer in <ref refid="classtf_1_1Executor" kindref="compound">tf::Executor</ref> </para>
|
|
</listitem>
|
|
<listitem><para>fixed compilation errors of nvcc v11.3 in auto template deduction<itemizedlist>
|
|
<listitem><para>std::scoped_lock</para>
|
|
</listitem><listitem><para>tf::Serializer and tf::Deserializer </para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
<listitem><para>fixed memory leak when moving a <ref refid="classtf_1_1Taskflow" kindref="compound">tf::Taskflow</ref></para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_breaking_changes">
|
|
<title>Breaking Changes</title>
|
|
<para>There are no breaking changes in this release.</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_deprecated_items">
|
|
<title>Deprecated and Removed Items</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>removed tf::cudaFlow::kernel_on method </para>
|
|
</listitem>
|
|
<listitem><para>removed explicit partitions in parallel iterations and reductions </para>
|
|
</listitem>
|
|
<listitem><para>removed tf::cudaFlowCapturerBase </para>
|
|
</listitem>
|
|
<listitem><para>removed tf::cublasFlowCapturer </para>
|
|
</listitem>
|
|
<listitem><para>renamed update and rebind methods in <ref refid="classtf_1_1cudaFlow" kindref="compound">tf::cudaFlow</ref> and <ref refid="classtf_1_1cudaFlowCapturer" kindref="compound">tf::cudaFlowCapturer</ref> to overloads</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_documentation">
|
|
<title>Documentation</title>
|
|
<para><itemizedlist>
|
|
<listitem><para>revised <ref refid="StaticTasking" kindref="compound">Static Tasking</ref><itemizedlist>
|
|
<listitem><para><ref refid="StaticTasking_1MoveATaskflow" kindref="member">Move a Taskflow</ref> </para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
<listitem><para>revised <ref refid="ExecuteTaskflow" kindref="compound">Executor</ref><itemizedlist>
|
|
<listitem><para><ref refid="ExecuteTaskflow_1ExecuteATaskflowWithTransferredOwnership" kindref="member">Execute a Taskflow with Transferred Ownership</ref> </para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="cudaFlowAlgorithms" kindref="compound">cudaFlow Algorithms</ref> </para>
|
|
</listitem>
|
|
<listitem><para>added <ref refid="cudaStandardAlgorithms" kindref="compound">CUDA Standard Algorithms</ref><itemizedlist>
|
|
<listitem><para><ref refid="CUDASTDExecutionPolicy" kindref="compound">Execution Policy</ref></para>
|
|
</listitem><listitem><para><ref refid="CUDASTDReduce" kindref="compound">Parallel Reduction</ref></para>
|
|
</listitem><listitem><para><ref refid="CUDASTDScan" kindref="compound">Parallel Scan</ref></para>
|
|
</listitem><listitem><para><ref refid="CUDASTDMerge" kindref="compound">Parallel Merge</ref></para>
|
|
</listitem><listitem><para><ref refid="CUDASTDFind" kindref="compound">Parallel Find</ref></para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</listitem>
|
|
</itemizedlist>
|
|
</para>
|
|
</sect1>
|
|
<sect1 id="release-3-2-0_1release-3-2-0_miscellaneous_items">
|
|
<title>Miscellaneous Items</title>
|
|
<para>We have published <ref refid="classtf_1_1cudaFlow" kindref="compound">tf::cudaFlow</ref> in the following conference:<itemizedlist>
|
|
<listitem><para>Dian-Lun Lin and Tsung-Wei Huang, "Efficient GPU Computation using Task <ref refid="classtf_1_1Graph" kindref="compound">Graph</ref> Parallelism," <emphasis>European Conference on Parallel and Distributed Computing (EuroPar)</emphasis>, 2021 </para>
|
|
</listitem></itemizedlist>
|
|
</para>
|
|
</sect1>
|
|
</detaileddescription>
|
|
<location file="doxygen/releases/release-3.2.0.dox"/>
|
|
</compounddef>
|
|
</doxygen>
|