From 8437f1485eefdb8163fb449f99abbf7178b83c2f Mon Sep 17 00:00:00 2001
From: Wiebe van Breukelen <breukelen@astron.nl>
Date: Tue, 5 Nov 2024 09:00:31 +0100
Subject: [PATCH] Renamed functions and removed getTotalConstMem()

---
 CHANGELOG.md                   |  2 +-
 include/cudawrappers/cu.hpp    |  7 +------
 include/cudawrappers/cufft.hpp | 30 +++++++++++++++---------------
 tests/test_cu.cpp              | 11 ++---------
 tests/test_cufft.cpp           |  4 ++--
 5 files changed, 21 insertions(+), 33 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 9f2d1cd..d379acf 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -17,7 +17,7 @@ project adheres to [Semantic Versioning](http://semver.org/).
 - Added `cu::Stream::memcpyHtoD2DAsync()`, `cu::Stream::memcpyDtoHD2Async()`,
   and `cu::Stream::memcpyDtoD2DAsync()`
 - Added `cu::DeviceMemory::memset2D()` and `cu::Stream::memset2DAsync()`
-- Added `cufft::FFT1D_R2C` and `cufft::FFT1D_C2R`
+- Added `cufft::FFT1DR2C` and `cufft::FFT1DC2R`
 - Added `cu::Device::getOrdinal()`
 
 ### Changed
diff --git a/include/cudawrappers/cu.hpp b/include/cudawrappers/cu.hpp
index 202ebbd..bacec83 100644
--- a/include/cudawrappers/cu.hpp
+++ b/include/cudawrappers/cu.hpp
@@ -182,17 +182,12 @@ class Device : public Wrapper<CUdevice> {
 #endif
   }
 
-  size_t getTotalMem() const {
+  size_t totalMem() const {
     size_t size{};
     checkCudaCall(cuDeviceTotalMem(&size, _obj));
     return size;
   }
 
-  size_t getTotalConstMem() const {
-    return static_cast<size_t>(
-        getAttribute(CU_DEVICE_ATTRIBUTE_TOTAL_CONSTANT_MEMORY));
-  }
-
   int getOrdinal() const { return _ordinal; }
 
   // Primary Context Management
diff --git a/include/cudawrappers/cufft.hpp b/include/cudawrappers/cufft.hpp
index fb96132..57a08df 100644
--- a/include/cudawrappers/cufft.hpp
+++ b/include/cudawrappers/cufft.hpp
@@ -231,30 +231,30 @@ FFT2D<CUDA_C_16F>::FFT2D(const int nx, const int ny)
     : FFT2D(nx, ny, 1, nx * ny, 1) {}
 
 /*
- * FFT1D_R2C
+ * FFT1DR2C
  */
 template <cudaDataType_t T>
-class FFT1D_R2C : public FFT {
+class FFT1DR2C : public FFT {
  public:
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_R2C(const int nx) = delete;
+  FFT1DR2C(const int nx) = delete;
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_R2C(const int nx, const int batch) = delete;
+  FFT1DR2C(const int nx, const int batch) = delete;
 
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_R2C(const int nx, const int batch, long long inembed,
-            long long ouembed) = delete;
+  FFT1DR2C(const int nx, const int batch, long long inembed,
+           long long ouembed) = delete;
 };
 
 template <>
-FFT1D_R2C<CUDA_R_32F>::FFT1D_R2C(const int nx, const int batch,
-                                 long long inembed, long long ouembed) {
+FFT1DR2C<CUDA_R_32F>::FFT1DR2C(const int nx, const int batch, long long inembed,
+                               long long ouembed) {
   checkCuFFTCall(cufftCreate(plan()));
   const int rank = 1;
   size_t ws = 0;
@@ -273,26 +273,26 @@ FFT1D_R2C<CUDA_R_32F>::FFT1D_R2C(const int nx, const int batch,
  * FFT1D_C2R
  */
 template <cudaDataType_t T>
-class FFT1D_C2R : public FFT {
+class FFT1DC2R : public FFT {
  public:
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_C2R(const int nx) = delete;
+  FFT1DC2R(const int nx) = delete;
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_C2R(const int nx, const int batch) = delete;
+  FFT1DC2R(const int nx, const int batch) = delete;
 #if defined(__HIP__)
   __host__
 #endif
-  FFT1D_C2R(const int nx, const int batch, long long inembed,
-            long long ouembed) = delete;
+  FFT1DC2R(const int nx, const int batch, long long inembed,
+           long long ouembed) = delete;
 };
 
 template <>
-FFT1D_C2R<CUDA_C_32F>::FFT1D_C2R(const int nx, const int batch,
-                                 long long inembed, long long ouembed) {
+FFT1DC2R<CUDA_C_32F>::FFT1DC2R(const int nx, const int batch, long long inembed,
+                               long long ouembed) {
   checkCuFFTCall(cufftCreate(plan()));
   const int rank = 1;
   size_t ws = 0;
diff --git a/tests/test_cu.cpp b/tests/test_cu.cpp
index cf5bc0a..2986498 100644
--- a/tests/test_cu.cpp
+++ b/tests/test_cu.cpp
@@ -24,20 +24,13 @@ TEST_CASE("Test cu::Device", "[device]") {
     CHECK(arch.size() > 0);
   }
 
-  SECTION("Test device.getTotalMem", "[device]") {
-    const size_t total_mem = device.getTotalMem();
+  SECTION("Test device.totalMem", "[device]") {
+    const size_t total_mem = device.totalMem();
     std::cout << "Device total memory: " << (total_mem / (1024 * 1024))
               << " bytes" << std::endl;
     CHECK(total_mem > 0);
   }
 
-  SECTION("Test Device.getTotalConstMem", "[device]") {
-    const size_t const_mem = device.getTotalConstMem();
-    std::cout << "Device constant memory: " << const_mem << " bytes"
-              << std::endl;
-    CHECK(const_mem > 0);
-  }
-
   SECTION("Test Device.getOrdinal", "[device]") {
     const int dev_ordinal = device.getOrdinal();
     CHECK(dev_ordinal >= 0);
diff --git a/tests/test_cufft.cpp b/tests/test_cufft.cpp
index 96194aa..afc5354 100644
--- a/tests/test_cufft.cpp
+++ b/tests/test_cufft.cpp
@@ -127,8 +127,8 @@ TEST_CASE("Test 1D FFT", "[FFT1D]") {
     generateSignal(static_cast<cufftComplex *>(h_in), size, patchSize, {1, 1});
     stream.memcpyHtoDAsync(d_in, h_in, arraySize);
 
-    cufft::FFT1D_R2C<CUDA_R_32F> fft_r2c(size, 1, 1, 1);
-    cufft::FFT1D_C2R<CUDA_C_32F> fft_c2r(size, 1, 1, 1);
+    cufft::FFT1DR2C<CUDA_R_32F> fft_r2c(size, 1, 1, 1);
+    cufft::FFT1DC2R<CUDA_C_32F> fft_c2r(size, 1, 1, 1);
     fft_r2c.setStream(stream);
     fft_c2r.setStream(stream);