en Copyright 2015 Java http://172.16.0.5:8080/history/llvm-project-15.0.7/openmp/libomptarget/DeviceRTL/include/generated_microtask_cases.gen#67ab875f [OpenMP] Prototype opt-in new GPU device RTLThe "old" OpenMP GPU device runtime (D14254) has served us well for manyyears but modernizing it has caused some pain recently. This patchintroduces an alternative which is mostly written from scratch embracingOpenMP 5.X, C++, LLVM coding style (where applicable), and conceptualinterfaces. This new runtime is opt-in through a clang flag (D106793).The new runtime is currently only build for nvptx and has "-new" in itsname.The design is tailored towards middle-end optimizations rather thanfront-end code generation choices, a trend we already started in the oldruntime a while back. In contrast to the old one, state is organized ina simple manner rather than a "smart" one. While this can induce costsit helps optimizations. Our expectation is that the majority of codescan be optimized and a "simple" design is therefore preferable. The newruntime does also avoid users to pay for things they do not use,especially wrt. memory. The unlikely case of nested parallelism issupported but costly to make the more likely case use less resources.The worksharing and reduction implementation have been taken from theold runtime and will be rewritten in the future if necessary.Documentation and debug features are still mostly missing and will beadded over time.All external symbols start with `__kmpc` for legacy reasons but shouldbe renamed once we switch over to a single runtime. All internal symbolsare placed in appropriate namespaces (anonymous or `_OMP`) to avoid nameclashes with user symbols.Differential Revision: https://reviews.llvm.org/D106803 List of files: /llvm-project-15.0.7/openmp/libomptarget/DeviceRTL/include/generated_microtask_cases.gen Sun, 25 Jul 2021 18:26:44 +0000 Johannes Doerfert <johannes@jdoerfert.de>