Auto-format code changes (#645)

Auto-format code using Clang-Format Co-authored-by: GitHub Actions <actions@github.com>
NCAR · Sep 4, 2024 · 7c5f4bf · 7c5f4bf
1 parent 93dba08
commit 7c5f4bf
Show file tree

Hide file tree

Showing 10 changed files with 404 additions and 107 deletions.
diff --git a/include/micm/cuda/util/cuda_dense_matrix.hpp b/include/micm/cuda/util/cuda_dense_matrix.hpp
@@ -210,7 +210,13 @@ namespace micm
  {
  // the cudaMemset function only works for integer types and is an asynchronous function:
  // https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__MEMORY.html#group__CUDART__MEMORY_1gf7338650f7683c51ee26aadc6973c63a
- CHECK_CUDA_ERROR(cudaMemsetAsync(this->param_.d_data_, val, sizeof(T) * this->param_.number_of_elements_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemset");
+ CHECK_CUDA_ERROR(
+ cudaMemsetAsync(
+ this->param_.d_data_,
+ val,
+ sizeof(T) * this->param_.number_of_elements_,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemset");
  }
  else
  {

diff --git a/include/micm/cuda/util/cuda_util.cuh b/include/micm/cuda/util/cuda_util.cuh
@@ -6,9 +6,9 @@
 #include <cuda_runtime.h>
 
 #include <map>
-#include <string>
 #include <memory>
 #include <mutex>
+#include <string>
 
 #define CHECK_CUDA_ERROR(err, msg) micm::cuda::CheckCudaError(err, __FILE__, __LINE__, msg)
 #define CHECK_CUBLAS_ERROR(err, msg) micm::cuda::CheckCublasError(err, __FILE__, __LINE__, msg)
@@ -54,24 +54,23 @@ namespace micm
  /// @brief Singleton class to manage CUDA streams
  class CudaStreamSingleton
  {
- public:
-
+ public:
  ~CudaStreamSingleton() = default;
- 
+
  CudaStreamSingleton(const CudaStreamSingleton&) = delete;
 
  CudaStreamSingleton& operator=(const CudaStreamSingleton&) = delete;
 
  // Get the only one instance of the singleton class
  static CudaStreamSingleton& GetInstance();
- 
+
  // Get the CUDA stream given a stream ID
  cudaStream_t& GetCudaStream(std::size_t stream_id);
 
  // Empty the map variable to clean up all CUDA streams
  void CleanUp();
- 
- private:
+
+  private:
  // Private constructor to prevent direct instantiation
  CudaStreamSingleton() = default;
 

diff --git a/src/process/process_set.cu b/src/process/process_set.cu
@@ -136,11 +136,32 @@ namespace micm
  ProcessSetParam devstruct;
 
  /// Allocate memory space on the device
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.number_of_reactants_), number_of_reactants_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.reactant_ids_), reactant_ids_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.number_of_products_), number_of_products_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.product_ids_), product_ids_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.yields_), yields_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.number_of_reactants_),
+ number_of_reactants_bytes,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.reactant_ids_),
+ reactant_ids_bytes,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.number_of_products_),
+ number_of_products_bytes,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.product_ids_), product_ids_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.yields_), yields_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
 
  /// Copy the data from host to device
  CHECK_CUDA_ERROR(
@@ -152,7 +173,12 @@ namespace micm
  micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
  "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.reactant_ids_, hoststruct.reactant_ids_, reactant_ids_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ cudaMemcpyAsync(
+ devstruct.reactant_ids_,
+ hoststruct.reactant_ids_,
+ reactant_ids_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
  "cudaMemcpy");
  CHECK_CUDA_ERROR(
  cudaMemcpyAsync(
@@ -163,10 +189,21 @@ namespace micm
  micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
  "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.product_ids_, hoststruct.product_ids_, product_ids_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ cudaMemcpyAsync(
+ devstruct.product_ids_,
+ hoststruct.product_ids_,
+ product_ids_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
  "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.yields_, hoststruct.yields_, yields_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemcpy");
+ cudaMemcpyAsync(
+ devstruct.yields_,
+ hoststruct.yields_,
+ yields_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemcpy");
 
  devstruct.number_of_reactants_size_ = hoststruct.number_of_reactants_size_;
  devstruct.reactant_ids_size_ = hoststruct.reactant_ids_size_;
@@ -186,12 +223,21 @@ namespace micm
  size_t jacobian_flat_ids_bytes = sizeof(size_t) * hoststruct.jacobian_flat_ids_size_;
 
  /// Allocate memory space on the device
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.jacobian_flat_ids_), jacobian_flat_ids_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.jacobian_flat_ids_),
+ jacobian_flat_ids_bytes,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
 
  /// Copy the data from host to device
  CHECK_CUDA_ERROR(
  cudaMemcpyAsync(
- devstruct.jacobian_flat_ids_, hoststruct.jacobian_flat_ids_, jacobian_flat_ids_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ devstruct.jacobian_flat_ids_,
+ hoststruct.jacobian_flat_ids_,
+ jacobian_flat_ids_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
  "cudaMemcpy");
 
  devstruct.jacobian_flat_ids_size_ = hoststruct.jacobian_flat_ids_size_;
@@ -202,18 +248,29 @@ namespace micm
  void FreeConstData(ProcessSetParam& devstruct)
  {
  if (devstruct.number_of_reactants_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.number_of_reactants_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.number_of_reactants_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaFree");
  if (devstruct.reactant_ids_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.reactant_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.reactant_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaFree");
  if (devstruct.number_of_products_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.number_of_products_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.number_of_products_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaFree");
  if (devstruct.product_ids_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.product_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.product_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaFree");
  if (devstruct.yields_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.yields_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.yields_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
  if (devstruct.jacobian_flat_ids_ != nullptr)
  {
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.jacobian_flat_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.jacobian_flat_ids_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaFree");
  }
  }
 
@@ -224,7 +281,11 @@ namespace micm
  const ProcessSetParam& devstruct)
  {
  size_t number_of_blocks = (rate_constants_param.number_of_grid_cells_ + BLOCK_SIZE - 1) / BLOCK_SIZE;
- SubtractJacobianTermsKernel<<<number_of_blocks, BLOCK_SIZE, 0, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(
+ SubtractJacobianTermsKernel<<<
+ number_of_blocks,
+ BLOCK_SIZE,
+ 0,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(
  rate_constants_param, state_variables_param, jacobian_param, devstruct);
  } // end of SubtractJacobianTermsKernelDriver
 
@@ -235,7 +296,11 @@ namespace micm
  const ProcessSetParam& devstruct)
  {
  size_t number_of_blocks = (rate_constants_param.number_of_grid_cells_ + BLOCK_SIZE - 1) / BLOCK_SIZE;
- AddForcingTermsKernel<<<number_of_blocks, BLOCK_SIZE, 0, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(
+ AddForcingTermsKernel<<<
+ number_of_blocks,
+ BLOCK_SIZE,
+ 0,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(
  rate_constants_param, state_variables_param, forcing_param, devstruct);
  } // end of AddForcingTermsKernelDriver
  } // namespace cuda

diff --git a/src/solver/linear_solver.cu b/src/solver/linear_solver.cu
@@ -89,20 +89,56 @@ namespace micm
 
  /// Create a struct whose members contain the addresses in the device memory.
  LinearSolverParam devstruct;
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.nLij_Lii_), nLij_Lii_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.Lij_yj_), Lij_yj_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.nUij_Uii_), nUij_Uii_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
- CHECK_CUDA_ERROR(cudaMallocAsync(&(devstruct.Uij_xj_), Uij_xj_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.nLij_Lii_), nLij_Lii_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.Lij_yj_), Lij_yj_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.nUij_Uii_), nUij_Uii_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
+ CHECK_CUDA_ERROR(
+ cudaMallocAsync(
+ &(devstruct.Uij_xj_), Uij_xj_bytes, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMalloc");
 
  /// Copy the data from host to device
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.nLij_Lii_, hoststruct.nLij_Lii_, nLij_Lii_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemcpy");
+ cudaMemcpyAsync(
+ devstruct.nLij_Lii_,
+ hoststruct.nLij_Lii_,
+ nLij_Lii_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.Lij_yj_, hoststruct.Lij_yj_, Lij_yj_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemcpy");
+ cudaMemcpyAsync(
+ devstruct.Lij_yj_,
+ hoststruct.Lij_yj_,
+ Lij_yj_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.nUij_Uii_, hoststruct.nUij_Uii_, nUij_Uii_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemcpy");
+ cudaMemcpyAsync(
+ devstruct.nUij_Uii_,
+ hoststruct.nUij_Uii_,
+ nUij_Uii_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemcpy");
  CHECK_CUDA_ERROR(
- cudaMemcpyAsync(devstruct.Uij_xj_, hoststruct.Uij_xj_, Uij_xj_bytes, cudaMemcpyHostToDevice, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaMemcpy");
+ cudaMemcpyAsync(
+ devstruct.Uij_xj_,
+ hoststruct.Uij_xj_,
+ Uij_xj_bytes,
+ cudaMemcpyHostToDevice,
+ micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)),
+ "cudaMemcpy");
 
  devstruct.nLij_Lii_size_ = hoststruct.nLij_Lii_size_;
  devstruct.Lij_yj_size_ = hoststruct.Lij_yj_size_;
@@ -117,13 +153,17 @@ namespace micm
  void FreeConstData(LinearSolverParam& devstruct)
  {
  if (devstruct.nLij_Lii_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.nLij_Lii_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.nLij_Lii_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
  if (devstruct.Lij_yj_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.Lij_yj_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.Lij_yj_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
  if (devstruct.nUij_Uii_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.nUij_Uii_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.nUij_Uii_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
  if (devstruct.Uij_xj_ != nullptr)
- CHECK_CUDA_ERROR(cudaFreeAsync(devstruct.Uij_xj_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
+ CHECK_CUDA_ERROR(
+ cudaFreeAsync(devstruct.Uij_xj_, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)), "cudaFree");
  }
 
  void SolveKernelDriver(
@@ -133,7 +173,8 @@ namespace micm
  const LinearSolverParam& devstruct)
  {
  size_t number_of_blocks = (x_param.number_of_grid_cells_ + BLOCK_SIZE - 1) / BLOCK_SIZE;
- SolveKernel<<<number_of_blocks, BLOCK_SIZE, 0, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(x_param, L_param, U_param, devstruct);
+ SolveKernel<<<number_of_blocks, BLOCK_SIZE, 0, micm::cuda::CudaStreamSingleton::GetInstance().GetCudaStream(0)>>>(
+ x_param, L_param, U_param, devstruct);
  }
  } // namespace cuda
 } // namespace micm