In a nutshell, this moves our libomptarget code to populate the offload subproject. With this commit, users need to enable the new LLVM/Offload subproject as a runtime in their cmake configuration. No further changes are expected for downstream code. Tests and other components still depend on OpenMP and have also not been renamed. The results below are for a build in which OpenMP and Offload are enabled runtimes. In addition to the pure `git mv`, we needed to adjust some CMake files. Nothing is intended to change semantics. ``` ninja check-offload ``` Works with the X86 and AMDGPU offload tests ``` ninja check-openmp ``` Still works but doesn't build offload tests anymore. ``` ls install/lib ``` Shows all expected libraries, incl. - `libomptarget.devicertl.a` - `libomptarget-nvptx-sm_90.bc` - `libomptarget.rtl.amdgpu.so` -> `libomptarget.rtl.amdgpu.so.18git` - `libomptarget.so` -> `libomptarget.so.18git` Fixes: https://github.com/llvm/llvm-project/issues/75124 --------- Co-authored-by: Saiyedul Islam <Saiyedul.Islam@amd.com>
Summary: Recent patches have added some missing intrinsic functions NVPTX. This patch gets rid of all the remaining uses of inline assembly. The one change that wasn't directly replaced with a built-in was the `pack` and `unpack` implementations. However, using the generic C implementation is equivalent to the output SASS when run through PTXAS.
Summary: A recent patch allowed us to easily replace GNU atomics with scoped variants that make use of the backend's handling for more permissive scopes. The default is full "system" scope, that means the atomic operation must be consistent with operations that may happen on the host's memory. This is generally only required for processes that are communicating with something via global fine-grained memory. This patch uses these atomics to make everything device scoped, as nothing in the OpenMP runtime should depend on the host. This is only provided as a very new clang extension but the DeviceRTL is only compiled with clang so it is always available.
Summary: While this is technically a no-op for AMDGPU hardware, in cases where the user would see fit to add an explicit wavefront sync on Nvidia hardware, we should also inform the LLVM optimizer that this control flow is convergent so we do not reorder blocks.
- [Libomptarget] Make the references to 'malloc' and 'free' weak. - [Libomptarget][NFC] Use C++ style attributes instead
This feature was supposed to allow you to trace execution inside of Libomptarget. However, this never really worked properly. The printing was always reoganized, only worked for single threads, and pretty much only told you a handful of things about a runtime library that's an implementation detail to all users. Despite this, it contributed about 40% of the total filesize of the deviceRTL. This patch simply removes this functionalit which I think was past due. Reviewed By: tianshilei1992 Differential Revision: https://reviews.llvm.org/D157001
Differential Revision: https://reviews.llvm.org/D156729
This change makes the naming more consistent, I hope.
We ended up with `llvm.assume(icmp ne ptr as(4) null, as(4) @str)` because the string in address space 4 was not known to be non-null. There is no need to create these assumes.
With https://reviews.llvm.org/D137524, memory scope and ordering attributes are being used to generate the required instructions for atomic inc/dec on AMDGPU. This patch adds the memory scope attribute to the atomic::inc API and uses the device scope in reduction. Without the device scope in atomic_inc, the default system scope leads to unnecessary L2 write-backs/invalidates. Reviewed By: arsenm Differential Revision: https://reviews.llvm.org/D154172