rocPRIM-2.10.12 for ROCm 5.0.2

rocPRIM code for ROCm 5.0.2 is unchanged from rocPRIM for ROCm 5.0.1. The library was rebuilt for the updated ROCm 5.0.2 stack.

rocPRIM-2.10.12 for ROCm 5.0.1

rocPRIM code for ROCm 5.0.1 is unchanged from rocPRIM for ROCm 5.0.0. The library was rebuilt for the updated ROCm 5.0.1 stack.

rocPRIM-2.10.12 for ROCm 5.0.0

Fixed

• Enable bfloat16 tests and reduce threshold for bfloat16
• Fix device scan limit_size feature
• Non-optimized builds no longer trigger local memory limit errors

Added

• Added scan size limit feature
• Added reduce size limit feature
• Added transform size limit feature
• Add block_load_striped and block_store_striped
• Add gather_to_blocked to gather values from other threads into a blocked arrangement
• The block sizes for device merge sorts initial block sort and its merge steps are now separate in its kernel config
  • the block sort step supports multiple items per thread

Changed

• size_limit for scan, reduce and transform can now be set in the config struct instead of a parameter
• Device_scan and device_segmented_scan: inclusive_scan now uses the input-type as accumulator-type, exclusive_scan uses initial-value-type.
  • This particularly changes behaviour of small-size input types with large-size output types (e.g. short input, int output).
  • And low-res input with high-res output (e.g. float input, double output)
• Revert old Fiji workaround, because they solved the issue at compiler side
• Update README cmake minimum version number
• Block sort support multiple items per thread
  • currently only powers of two block sizes, and items per threads are supported and only for full blocks
• Bumped the minimum required version of CMake to 3.16

Known issues

• Unit tests may soft hang on MI200 when running in hipMallocManaged mode.
• device_segmented_radix_sort, device_scan unit tests failing for HIP on Windows
• ReduceEmptyInput cause random faulire with bfloat16

rocPRIM-2.10.11 for ROCm 4.5.2

rocPRIM code for ROCm 4.5.2 is unchanged from rocPRIM for ROCm 4.5.0. The library was rebuilt for the updated ROCm 4.5.2 stack.

rocPRIM-2.10.11 for ROCm 4.5.0

Addded

• Code coverage tools build option
• Address sanitizer build option
• gfx1030 support added.
• Experimental HIP-CPU support; build using GCC/Clang/MSVC on Win/Linux. It is work in progress, many algorithms still known to fail.
• Initial HIP on Windows support. See README for instructions on how to build and install.
• bfloat16 support added.

Optimizations

• Added single tile radix sort for smaller sizes.
• Improved performance for radix sort for larger element sizes.

Changed

• Package renamed to rocprim-dev for .deb, and to rocprim-devel for .rpm. As rocPRIM is a header-only library, there is no associated runtime package, so for compatibility this development package provides the package rocprim. The provides feature in packaging is introduced as a deprecated feature and will be removed in a future rocm release.

Deprecated

• The warp_size() function is now deprecated; please switch to host_warp_size() and device_warp_size() for host and device references respectively.

rocPRIM-2.10.10 for ROCm 4.3.1

No changes made for ROCm 4.3.1.

rocPRIM-2.10.10 for ROCm 4.3.0

Fixed

• Bugfix & minor performance improvement for merge_sort when input and output storage are the same.

Added

• gfx90a support added.

Deprecated

• The warp_size() function is now deprecated; please switch to host_warp_size() and device_warp_size() for host and device references respectively.

rocPRIM-2.10.9 for ROCm 4.2.0

Fixed

• Size zero inputs are now properly handled with newer ROCm builds that no longer allow zero-size kernel grid/block dimensions

Changed

• Minimum cmake version required is now 3.10.2

Known issues

• Device scan unit test currently failing due to LLVM bug.

rocPRIM-2.10.8 for ROCm 4.1.0

Fixed

• Texture cache iteration support has been re-enabled.
• Benchmark builds have been re-enabled.
• Unique operator no longer called on invalid elements.

Known issues

• Device scan unit test currently failing due to LLVM bug.

Known Issues

• None

rocPRIM-2.10.6 for ROCm 4.0.0

New Features

• No new features

Known Issues

• None