TheRock/rocm-systems/projects/hip-tests/catch/unit/stream/hipMultiStream.cc

/*
 * Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
 *
 * SPDX-License-Identifier: MIT
 */

#include <hip_test_common.hh>
#include <iostream>
#include <vector>
constexpr int NN = 1 << 21;
__global__ void kernel_do_nothing(__attribute__((unused)) int a) {
  // empty kernel
}
__global__ void kernel(float* x, float* y, int n) {
  size_t tid{threadIdx.x};
  if (tid < 1) {
    for (int i = 0; i < n; i++) {
      x[i] = sqrt(powf(3.14159, i));
    }
    y[tid] = y[tid] + 1.0f;
  }
}
__global__ void nKernel(float* y) {
  size_t tid{threadIdx.x};
  y[tid] = y[tid] + 1.0f;
}
HIP_TEST_CASE(Unit_hipMultiStream_sameDevice) {
  constexpr int num_streams{8};
  hipStream_t streams[num_streams];
  float *data[num_streams], *yd, *xd;
  float y{1.0f}, x{1.0f};
  const int n = isQuickLevel() ? (1 << 12) : NN;
  HIP_CHECK(hipMalloc((void**)&yd, sizeof(float)));
  HIP_CHECK(hipMalloc((void**)&xd, sizeof(float)));
  HIP_CHECK(hipMemcpy(yd, &y, sizeof(float), hipMemcpyHostToDevice));
  HIP_CHECK(hipMemcpy(xd, &x, sizeof(float), hipMemcpyHostToDevice));
  for (int i = 0; i < num_streams; i++) {
    HIP_CHECK(hipStreamCreate(&streams[i]));
    HIP_CHECK(hipMalloc(&data[i], n * sizeof(float)));
    hipLaunchKernelGGL(kernel, dim3(1), dim3(1), 0, streams[i], data[i], xd, n);
    HIP_CHECK(hipGetLastError());
    hipLaunchKernelGGL(HIP_KERNEL_NAME(nKernel), dim3(1), dim3(1), 0, 0, yd);
    HIP_CHECK(hipGetLastError());
    HIP_CHECK(hipFree(data[i]));
    HIP_CHECK(hipStreamDestroy(streams[i]));
  }
  HIP_CHECK(hipMemcpy(&x, xd, sizeof(float), hipMemcpyDeviceToHost));
  HIP_CHECK(hipMemcpy(&y, yd, sizeof(float), hipMemcpyDeviceToHost));
  HIP_CHECK(hipFree(xd));
  HIP_CHECK(hipFree(yd));
  REQUIRE(x == Catch::Approx(y));
}

HIP_TEST_CASE(Unit_hipMultiStream_multimeDevice) {
  const int nLoops = isQuickLevel() ? 500 : 50000;
  constexpr int nStreams = 2;
  std::vector<hipStream_t> streams(nStreams);
  int nGpu = 0;
  HIP_CHECK(hipGetDeviceCount(&nGpu));
  if (nGpu < 1) {
    HIP_SKIP_TEST(HipTest::SkipReason::kNoGpuDevice);
  }
  static int device = 0;
  HIP_CHECK(hipSetDevice(device));
  hipDeviceProp_t props;
  HIP_CHECK(hipGetDeviceProperties(&props, device));
  INFO("Running on Bus: " << props.pciBusID << " " << props.name);
  for (int i = 0; i < nStreams; i++) {
    HIP_CHECK(hipStreamCreate(&streams[i]));
  }
  for (int k = 0; k <= nLoops; ++k) {
    HIP_CHECK(hipDeviceSynchronize());
    // Launch kernel with default stream
    hipLaunchKernelGGL(kernel_do_nothing, dim3(1), dim3(1), 0, 0, 1);
    HIP_CHECK(hipGetLastError());
    // Launch kernel on all streams
    for (int i = 0; i < nStreams; i++) {
      hipLaunchKernelGGL(kernel_do_nothing, dim3(1), dim3(1), 0, streams[i], 1);
      HIP_CHECK(hipGetLastError());
    }
    // Sync stream 1
    HIP_CHECK(hipStreamSynchronize(streams[0]));
    if (k % 10000 == 0 || k == nLoops) {
      INFO("Iter: " << k);
    }
  }
  HIP_CHECK(hipDeviceSynchronize());
  // Clean up
  for (int i = 0; i < nStreams; i++) {
    HIP_CHECK(hipStreamDestroy(streams[i]));
  }
}