330d8983d2
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>
82 lines
2.5 KiB
C++
82 lines
2.5 KiB
C++
//===------ State.cpp - OpenMP State & ICV interface ------------- C++ -*-===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include "Shared/Environment.h"
|
|
|
|
#include "Allocator.h"
|
|
#include "Configuration.h"
|
|
#include "Mapping.h"
|
|
#include "Synchronization.h"
|
|
#include "Types.h"
|
|
#include "Utils.h"
|
|
|
|
using namespace ompx;
|
|
|
|
#pragma omp begin declare target device_type(nohost)
|
|
|
|
[[gnu::used, gnu::retain, gnu::weak,
|
|
gnu::visibility(
|
|
"protected")]] DeviceMemoryPoolTy __omp_rtl_device_memory_pool;
|
|
[[gnu::used, gnu::retain, gnu::weak,
|
|
gnu::visibility("protected")]] DeviceMemoryPoolTrackingTy
|
|
__omp_rtl_device_memory_pool_tracker;
|
|
|
|
/// Stateless bump allocator that uses the __omp_rtl_device_memory_pool
|
|
/// directly.
|
|
struct BumpAllocatorTy final {
|
|
|
|
void *alloc(uint64_t Size) {
|
|
Size = utils::roundUp(Size, uint64_t(allocator::ALIGNMENT));
|
|
|
|
if (config::isDebugMode(DeviceDebugKind::AllocationTracker)) {
|
|
atomic::add(&__omp_rtl_device_memory_pool_tracker.NumAllocations, 1,
|
|
atomic::seq_cst);
|
|
atomic::add(&__omp_rtl_device_memory_pool_tracker.AllocationTotal, Size,
|
|
atomic::seq_cst);
|
|
atomic::min(&__omp_rtl_device_memory_pool_tracker.AllocationMin, Size,
|
|
atomic::seq_cst);
|
|
atomic::max(&__omp_rtl_device_memory_pool_tracker.AllocationMax, Size,
|
|
atomic::seq_cst);
|
|
}
|
|
|
|
uint64_t *Data =
|
|
reinterpret_cast<uint64_t *>(&__omp_rtl_device_memory_pool.Ptr);
|
|
uint64_t End =
|
|
reinterpret_cast<uint64_t>(Data) + __omp_rtl_device_memory_pool.Size;
|
|
|
|
uint64_t OldData = atomic::add(Data, Size, atomic::seq_cst);
|
|
if (OldData + Size > End)
|
|
__builtin_trap();
|
|
|
|
return reinterpret_cast<void *>(OldData);
|
|
}
|
|
|
|
void free(void *) {}
|
|
};
|
|
|
|
BumpAllocatorTy BumpAllocator;
|
|
|
|
/// allocator namespace implementation
|
|
///
|
|
///{
|
|
|
|
void allocator::init(bool IsSPMD, KernelEnvironmentTy &KernelEnvironment) {
|
|
// TODO: Check KernelEnvironment for an allocator choice as soon as we have
|
|
// more than one.
|
|
}
|
|
|
|
void *allocator::alloc(uint64_t Size) { return BumpAllocator.alloc(Size); }
|
|
|
|
void allocator::free(void *Ptr) { BumpAllocator.free(Ptr); }
|
|
|
|
///}
|
|
|
|
#pragma omp end declare target
|