CLOGS is a library for higher-level operations on top of the OpenCL C++ API. It is designed to integrate with other OpenCL code, including synchronization using OpenCL events. Currently only two operations are supported: radix sorting and exclusive scan. Radix sort supports all the unsigned integral types as keys, and all the built-in scalar and vector types suitable for storage in buffers as values. Scan supports all the integral types. It also supports vector types, which allows limited multi-scan capabilities.
|Tags||OpenCL sort scan prefix sum|
|Operating Systems||Posix compliant Windows|
Release Notes: This release adds support for out-of-place prefix sums (although with the same type for source and destination). It also saves program binaries in the cache for faster startup on some platforms.
Release Notes: This release fixes a bug that could cause incorrect sorting results, depending on the results of autotuning.
Release Notes: This bugfix release fixes a race condition in the radix sort, introduced in 1.2.1. It also works around a driver bug in the AMD APP SDK for CPU devices.
Release Notes: This release has no API changes, but significantly improves performance. Recent AMD GPUs get more than 3 times faster in some cases, but NVIDIA GPUs also see a nice performance boost.
Release Notes: This release autotunes kernel parameters on the target system for improved performance, in some cases a 2x improvement. Refer to the user manual for instructions on performing the autotuning. There are also some minor bugfixes.