File: atomics-remarks-gfx90a.cu

package info (click to toggle)
llvm-toolchain-14 1%3A14.0.6-16
  • links: PTS, VCS
  • area: main
  • in suites: sid
  • size: 1,496,368 kB
  • sloc: cpp: 5,593,980; ansic: 986,873; asm: 585,869; python: 184,223; objc: 72,530; lisp: 31,119; f90: 27,793; javascript: 9,780; pascal: 9,762; sh: 9,482; perl: 7,468; ml: 5,432; awk: 3,523; makefile: 2,547; xml: 953; cs: 573; fortran: 567
file content (16 lines) | stat: -rw-r--r-- 617 bytes parent folder | download | duplicates (3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
// RUN: %clang_cc1 %s -triple=amdgcn-amd-amdhsa -fcuda-is-device \
// RUN:   -target-cpu gfx90a -Rpass=atomic-expand -S -o - 2>&1 | \
// RUN:   FileCheck %s --check-prefix=GFX90A-CAS

// REQUIRES: amdgpu-registered-target

#include "Inputs/cuda.h"
#include <stdatomic.h>

// GFX90A-CAS: A compare and swap loop was generated for an atomic fadd operation at system memory scope
// GFX90A-CAS-LABEL: _Z14atomic_add_casPf
// GFX90A-CAS:  flat_atomic_cmpswap v0, v[2:3], v[4:5] glc
// GFX90A-CAS:  s_cbranch_execnz
__device__ float atomic_add_cas(float *p) {
  return __atomic_fetch_add(p, 1.0f, memory_order_relaxed);
}