[libc-commits] [clang] [compiler-rt] [libc] [llvm] [PGO][AMDGPU] Add offload profiling with uniformity-aware optimization (PR #177665)

[Yaxun Liu via libc-commits]

================
@@ -0,0 +1,68 @@
+/*===- InstrProfilingGPU.c - GPU profile counter functions ----------------===*\
+|*
+|* 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
+|*
+\*===----------------------------------------------------------------------===*/
+
+#if defined(__AMDGPU__) || defined(__NVPTX__)
+
+#include <gpuintrin.h>
+#include <stdint.h>
+
+#define ATOMIC_ADD(ptr, val)                                                   \
+  __scoped_atomic_fetch_add(ptr, val, __ATOMIC_RELAXED, __MEMORY_SCOPE_DEVICE)
----------------
yxsamliu wrote:

Resolved — macro removed, using __scoped_atomic_fetch_add directly in __llvm_profile_instrument_gpu.

https://github.com/llvm/llvm-project/pull/177665