Skip to content

Commit

Permalink
Merge pull request #22 from IQTLabs/cleanup
Browse files Browse the repository at this point in the history 
VkFFT cleanup.
  • Loading branch information
@anarkiwi
anarkiwi authored Jan 15, 2024
2 parents fcd3cd2 + f179299 commit 317821c
Showing 1 changed file with 41 additions and 51 deletions.
92 changes: 41 additions & 51 deletions lib/vkfft.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,22 +15,17 @@ static VkDeviceMemory stagingBufferMemory = {0};
static uint64_t fftBufferSize = 0;
static uint64_t stagingBufferSize = 0;

VkFFTResult _transferDataFromCPU(char *cpu_arr, std::size_t transferSize) {
VkFFTResult _setupCopy(VkCommandBuffer &commandBuffer,
VkBufferCopy &copyRegion) {
VkResult res = VK_SUCCESS;
VkBuffer *buffer = &vkConfiguration.buffer[0];
char *data;
res = vkMapMemory(vkGPU.device, stagingBufferMemory, 0, transferSize, 0,
(void **)&data);
if (res != VK_SUCCESS)
return VKFFT_ERROR_MALLOC_FAILED;
memcpy(data, cpu_arr, transferSize);
vkUnmapMemory(vkGPU.device, stagingBufferMemory);
copyRegion.srcOffset = 0;
copyRegion.dstOffset = 0;
copyRegion.size = stagingBufferSize;
VkCommandBufferAllocateInfo commandBufferAllocateInfo = {
VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO};
commandBufferAllocateInfo.commandPool = vkGPU.commandPool;
commandBufferAllocateInfo.level = VK_COMMAND_BUFFER_LEVEL_PRIMARY;
commandBufferAllocateInfo.commandBufferCount = 1;
VkCommandBuffer commandBuffer = {0};
res = vkAllocateCommandBuffers(vkGPU.device, &commandBufferAllocateInfo,
&commandBuffer);
if (res != VK_SUCCESS)
Expand All @@ -41,12 +36,11 @@ VkFFTResult _transferDataFromCPU(char *cpu_arr, std::size_t transferSize) {
res = vkBeginCommandBuffer(commandBuffer, &commandBufferBeginInfo);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_BEGIN_COMMAND_BUFFER;
VkBufferCopy copyRegion = {0};
copyRegion.srcOffset = 0;
copyRegion.dstOffset = 0;
copyRegion.size = transferSize;
vkCmdCopyBuffer(commandBuffer, stagingBuffer, buffer[0], 1, &copyRegion);
res = vkEndCommandBuffer(commandBuffer);
return VKFFT_SUCCESS;
}

VkFFTResult _runCopy(VkCommandBuffer &commandBuffer) {
VkResult res = vkEndCommandBuffer(commandBuffer);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_END_COMMAND_BUFFER;
VkSubmitInfo submitInfo = {VK_STRUCTURE_TYPE_SUBMIT_INFO};
Expand All @@ -65,48 +59,42 @@ VkFFTResult _transferDataFromCPU(char *cpu_arr, std::size_t transferSize) {
return VKFFT_SUCCESS;
}

VkFFTResult _transferDataFromCPU(char *cpu_arr, std::size_t transferSize) {
VkResult res = VK_SUCCESS;
VkBuffer *buffer = &vkConfiguration.buffer[0];
char *data;
res = vkMapMemory(vkGPU.device, stagingBufferMemory, 0, transferSize, 0,
(void **)&data);
if (res != VK_SUCCESS)
return VKFFT_ERROR_MALLOC_FAILED;
memcpy(data, cpu_arr, transferSize);
vkUnmapMemory(vkGPU.device, stagingBufferMemory);
VkCommandBuffer commandBuffer = {0};
VkBufferCopy copyRegion = {0};
VkFFTResult setup_res = _setupCopy(commandBuffer, copyRegion);
if (setup_res != VKFFT_SUCCESS) {
return setup_res;
}
vkCmdCopyBuffer(commandBuffer, stagingBuffer, buffer[0], 1, &copyRegion);
return _runCopy(commandBuffer);
}

VkFFTResult _transferDataToCPU(char *cpu_arr, std::size_t transferSize) {
// a function that transfers data from the GPU to the CPU using staging
// buffer, because the GPU memory is not host-coherent
VkResult res = VK_SUCCESS;
VkBuffer *buffer = &vkConfiguration.buffer[0];
VkCommandBufferAllocateInfo commandBufferAllocateInfo = {
VK_STRUCTURE_TYPE_COMMAND_BUFFER_ALLOCATE_INFO};
commandBufferAllocateInfo.commandPool = vkGPU.commandPool;
commandBufferAllocateInfo.level = VK_COMMAND_BUFFER_LEVEL_PRIMARY;
commandBufferAllocateInfo.commandBufferCount = 1;
VkCommandBuffer commandBuffer = {0};
res = vkAllocateCommandBuffers(vkGPU.device, &commandBufferAllocateInfo,
&commandBuffer);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_ALLOCATE_COMMAND_BUFFERS;
VkCommandBufferBeginInfo commandBufferBeginInfo = {
VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO};
commandBufferBeginInfo.flags = VK_COMMAND_BUFFER_USAGE_ONE_TIME_SUBMIT_BIT;
res = vkBeginCommandBuffer(commandBuffer, &commandBufferBeginInfo);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_BEGIN_COMMAND_BUFFER;
VkBufferCopy copyRegion = {0};
copyRegion.srcOffset = 0;
copyRegion.dstOffset = 0;
copyRegion.size = transferSize;
VkFFTResult setup_res = _setupCopy(commandBuffer, copyRegion);
if (setup_res != VKFFT_SUCCESS) {
return setup_res;
}
vkCmdCopyBuffer(commandBuffer, buffer[0], stagingBuffer, 1, &copyRegion);
res = vkEndCommandBuffer(commandBuffer);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_END_COMMAND_BUFFER;
VkSubmitInfo submitInfo = {VK_STRUCTURE_TYPE_SUBMIT_INFO};
submitInfo.commandBufferCount = 1;
submitInfo.pCommandBuffers = &commandBuffer;
res = vkQueueSubmit(vkGPU.queue, 1, &submitInfo, vkGPU.fence);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_SUBMIT_QUEUE;
res = vkWaitForFences(vkGPU.device, 1, &vkGPU.fence, VK_TRUE, 100000000000);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_WAIT_FOR_FENCES;
res = vkResetFences(vkGPU.device, 1, &vkGPU.fence);
if (res != VK_SUCCESS)
return VKFFT_ERROR_FAILED_TO_RESET_FENCES;
vkFreeCommandBuffers(vkGPU.device, vkGPU.commandPool, 1, &commandBuffer);
VkFFTResult copy_res = _runCopy(commandBuffer);
if (copy_res != VKFFT_SUCCESS) {
return copy_res;
}
char *data;
res = vkMapMemory(vkGPU.device, stagingBufferMemory, 0, transferSize, 0,
(void **)&data);
Expand Down Expand Up @@ -220,7 +208,9 @@ int64_t init_vkfft(std::size_t batches, std::size_t nfft,
return resFFT;

resFFT = allocateBuffer(&vkGPU, &stagingBuffer, &stagingBufferMemory,
VK_BUFFER_USAGE_TRANSFER_SRC_BIT,
VK_BUFFER_USAGE_STORAGE_BUFFER_BIT |
VK_BUFFER_USAGE_TRANSFER_SRC_BIT |
VK_BUFFER_USAGE_TRANSFER_DST_BIT,
VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT |
VK_MEMORY_PROPERTY_HOST_COHERENT_BIT,
stagingBufferSize);
Expand Down

0 comments on commit 317821c

Please sign in to comment.