alpaka
alpaka copied to clipboard
Abstraction Library for Parallel Kernel Acceleration :llama:
Move `#include`s from `alpaka::cpu::detail` to the global namespace (see #1775). On Linux, query the OS for free pages instead of reading `/proc/meminfo` (see #1774). Replace the nested namespace with a...
Depending on the architecture, the implementation of `alpaka::cpu::detail::cpuid()` may pull a system header in the `alpaka::cpu::detail` namespace: https://github.com/alpaka-group/alpaka/blob/f27d78c23c22f4eb5044a39cbfa60360ad1cd129/include/alpaka/dev/cpu/SysInfo.hpp#L42-L66 . Would it be possible to move the `#include` statements outside of...
On Linux, `getFreeGlobalMemSizeBytes()` is implemented parsing `/proc/meminfo`: https://github.com/alpaka-group/alpaka/blob/f27d78c23c22f4eb5044a39cbfa60360ad1cd129/include/alpaka/dev/cpu/SysInfo.hpp#L220-L245 I would suggest replacing it with reading the number of free pages from `sysconf()`, for example: ```c++ # if defined(_SC_AVPHYS_PAGES) return static_cast(sysconf(_SC_AVPHYS_PAGES))...
clang CUDA is exposing `atomic*_block()` function signatures even if these can not be used by the selected architecture. This leads to compile issues if clang is used as CUDA compiler....
To know where a test fails the line and file can be very useful. before: ``` 1: ALPAKA_CHECK failed because '!(equals(operandOrig, ret))' 1: ALPAKA_CHECK failed because '!(equals(operandOrig, ret))' ``` after:...
The unit test for `atomicDec` is not testing the flip around based on the values. https://github.com/alpaka-group/alpaka/blob/0b96515b1c8a24d0c3a2904081bc301311503770/include/alpaka/atomic/Op.hpp#L131 Some for `atomicInc` https://github.com/alpaka-group/alpaka/blob/0b96515b1c8a24d0c3a2904081bc301311503770/include/alpaka/atomic/Op.hpp#L115 We simply test if an value is incremented or decremented:...
This is a successor PR to #1713. * Adds clang-14 to CI. * Adds clang-14 as CUDA compiler to GitLab CI (CUDA versions 9.2 - 11.5). * ~~Adds clang-14 as...
After removing old Clang from the CI in the PR #1300 , we should also add the latest versions of Clang/LLVM. Since Clang/LLVM 11, Clang is forward compatible to new...
CI update
Summary: * ~~added gcc-12 to CI~~ - handled in #1765 * ~~added clang-14 to CI~~ - handled in #1766 * ~~added Xcode 13.3.1 to CI -> we now support macOS...
Hi, I'm looking at different flavors of "Abstraction Library for Parallel Kernel Acceleration". How is Alpaka different from Sycl, Kokkos, Raja or OpenCL ? Pros, cons. Thanks