rocPRIM 2.13.1 for ROCm 5.7.0

Changed

• Deprecated configuration radix_sort_config for device-level radix sort as it no longer matches the algorithm's parameters. New configuration radix_sort_config_v2 is preferred instead.
• Removed erroneous implementation of device-level inclusive_scan and exclusive_scan. The prior default implementation using lookback-scan now is the only available implementation.
• The benchmark metric indicating the bytes processed for exclusive_scan_by_key and inclusive_scan_by_key has been changed to incorporate the key type. Furthermore, the benchmark log has been changed such that these algorithms are reported as scan and scan_by_key instead of scan_exclusive and scan_inclusive.
• Deprecated configurations scan_config and scan_by_key_config for device-level scans, as they no longer match the algorithm's parameters. New configurations scan_config_v2 and scan_by_key_config_v2 are preferred instead.

Fixed

• Fixed build issue caused by missing header in thread/thread_search.hpp.

rocPRIM 2.13.0 for ROCm 5.6.1

rocPRIM code for ROCm 5.6.1 did not change. The library was rebuilt for the updated ROCm 5.6.1 stack.

rocPRIM 2.13.0 for ROCm 5.6.0

rocPRIM code for ROCm 5.6.0 did not change. The library was rebuilt for the updated ROCm 5.6.0 stack.

rocPRIM 2.12.1 for ROCm 5.4.4

Fixed

• Fixed compilation and execution issues for benchmarks with HIP on Windows

rocPRIM 2.13.0 for ROCm 5.5.1

rocPRIM code for ROCm 5.5.1 did not change. The library was rebuilt for the updated ROCm 5.5.1 stack.

rocPRIM 2.13.0 for ROCm 5.5.0

Added

• New block level radix_rank primitive.
• New block level radix_rank_match primitive.

Changed

• Improved the performance of block_radix_sort and device_radix_sort.

Known Issues

• Disabled GPU error messages relating to incorrect warp operation usage with Navi GPUs on Windows, due to GPU printf performance issues on Windows.

Fixed

• Fixed benchmark build on Windows

rocPRIM 2.12.0 for ROCm 5.4.3

rocPRIM code for ROCm 5.4.3 did not change. The library was rebuilt for the updated ROCm 5.4.3 stack.

rocPRIM 2.12.0 for ROCm 5.4.2

rocPRIM code for ROCm 5.4.2 did not change. The library was rebuilt for the updated ROCm 5.4.2 stack.

rocPRIM 2.12.0 for ROCm 5.4.1

rocPRIM code for ROCm 5.4.1 did not change. The library was rebuilt for the updated ROCm 5.4.1 stack.

rocPRIM 2.12.0 for ROCm 5.4.0

Changed

• device_partition, device_unique, and device_reduce_by_key now support problem
  sizes larger than 2^32 items.

Removed

• block_sort::sort() overload for keys and values with a dynamic size. This overload was documented but the
  implementation is missing. To avoid further confusion the documentation is removed until a decision is made on
  implementing the function.

Fixed

• Fixed the compilation failure in device_merge if the two key iterators don't match.