claude_code/procedural-3d-engine
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Copy cube map texture faces from host visible buffer instead of linear image (Refs #140)

Browse source
This commit is contained in:
saschawillems 2016-05-14 11:46:56 +02:00
parent ebff829030
commit c7a0d67448
1 changed files with 97 additions and 184 deletions
Download patch file Download diff file Expand all files Collapse all files

271
texturecubemap/texturecubemap.cpp
View file

@ -127,109 +127,89 @@ public:
cubeMap.width = texCube[0].dimensions().x;
cubeMap.height = texCube[0].dimensions().y;
// Get device properites for the requested texture format
VkFormatProperties formatProperties;
vkGetPhysicalDeviceFormatProperties(physicalDevice, format, &formatProperties);
VkImageCreateInfo imageCreateInfo = vkTools::initializers::imageCreateInfo();
imageCreateInfo.imageType = VK_IMAGE_TYPE_2D;
imageCreateInfo.format = format;
imageCreateInfo.extent = { cubeMap.width, cubeMap.height, 1 };
imageCreateInfo.mipLevels = 1;
imageCreateInfo.arrayLayers = 1;
imageCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
imageCreateInfo.tiling = VK_IMAGE_TILING_LINEAR;
imageCreateInfo.usage = VK_IMAGE_USAGE_TRANSFER_SRC_BIT;
imageCreateInfo.sharingMode = VK_SHARING_MODE_EXCLUSIVE;
imageCreateInfo.initialLayout = VK_IMAGE_LAYOUT_PREINITIALIZED;
imageCreateInfo.flags = 0;
VkMemoryAllocateInfo memAllocInfo = vkTools::initializers::memoryAllocateInfo();
VkMemoryRequirements memReqs;
struct {
VkImage image;
VkDeviceMemory memory;
} cubeFace[6];
// Create a host-visible staging buffer that contains the raw image data
VkBuffer stagingBuffer;
VkDeviceMemory stagingMemory;
// Allocate command buffer for image copies and layouts
VkCommandBuffer cmdBuffer;
VkCommandBufferAllocateInfo cmdBufAlllocatInfo =
vkTools::initializers::commandBufferAllocateInfo(
cmdPool,
VK_COMMAND_BUFFER_LEVEL_PRIMARY,
1);
VkResult err = vkAllocateCommandBuffers(device, &cmdBufAlllocatInfo, &cmdBuffer);
assert(!err);
VkBufferCreateInfo bufferCreateInfo = vkTools::initializers::bufferCreateInfo();
bufferCreateInfo.size = texCube.size();
// This buffer is used as a transfer source for the buffer copy
bufferCreateInfo.usage = VK_BUFFER_USAGE_TRANSFER_SRC_BIT;
bufferCreateInfo.sharingMode = VK_SHARING_MODE_EXCLUSIVE;
VkCommandBufferBeginInfo cmdBufInfo =
vkTools::initializers::commandBufferBeginInfo();
vkTools::checkResult(vkCreateBuffer(device, &bufferCreateInfo, nullptr, &stagingBuffer));
err = vkBeginCommandBuffer(cmdBuffer, &cmdBufInfo);
assert(!err);
// Get memory requirements for the staging buffer (alignment, memory type bits)
vkGetBufferMemoryRequirements(device, stagingBuffer, &memReqs);
// Load separate cube map faces into linear tiled textures
for (uint32_t face = 0; face < 6; ++face)
{
err = vkCreateImage(device, &imageCreateInfo, nullptr, &cubeFace[face].image);
assert(!err);
vkGetImageMemoryRequirements(device, cubeFace[face].image, &memReqs);
memAllocInfo.allocationSize = memReqs.size;
getMemoryType(memReqs.memoryTypeBits, VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, &memAllocInfo.memoryTypeIndex);
err = vkAllocateMemory(device, &memAllocInfo, nullptr, &cubeFace[face].memory);
assert(!err);
err = vkBindImageMemory(device, cubeFace[face].image, cubeFace[face].memory, 0);
assert(!err);
// Get memory type index for a host visible buffer
memAllocInfo.memoryTypeIndex = getMemoryType(memReqs.memoryTypeBits, VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT);
VkImageSubresource subRes = {};
subRes.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
vkTools::checkResult(vkAllocateMemory(device, &memAllocInfo, nullptr, &stagingMemory));
vkTools::checkResult(vkBindBufferMemory(device, stagingBuffer, stagingMemory, 0));
VkSubresourceLayout subResLayout;
void *data;
// Copy texture data into staging buffer
uint8_t *data;
vkTools::checkResult(vkMapMemory(device, stagingMemory, 0, memReqs.size, 0, (void **)&data));
memcpy(data, texCube.data(), texCube.size());
vkUnmapMemory(device, stagingMemory);
vkGetImageSubresourceLayout(device, cubeFace[face].image, &subRes, &subResLayout);
assert(!err);
err = vkMapMemory(device, cubeFace[face].memory, 0, memReqs.size, 0, &data);
assert(!err);
memcpy(data, texCube[face][subRes.mipLevel].data(), texCube[face][subRes.mipLevel].size());
vkUnmapMemory(device, cubeFace[face].memory);
// Setup buffer copy regions for all cube faces
std::vector<VkBufferImageCopy> bufferCopyRegions;
uint32_t offset = 0;
// Image barrier for linear image (base)
// Linear image will be used as a source for the copy
vkTools::setImageLayout(
cmdBuffer,
cubeFace[face].image,
VK_IMAGE_ASPECT_COLOR_BIT,
VK_IMAGE_LAYOUT_PREINITIALIZED,
VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL);
for (uint32_t face = 0; face < 6; face++)
{
VkBufferImageCopy bufferCopyRegion = {};
bufferCopyRegion.imageSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
bufferCopyRegion.imageSubresource.mipLevel = 0;
bufferCopyRegion.imageSubresource.baseArrayLayer = face;
bufferCopyRegion.imageSubresource.layerCount = 1;
bufferCopyRegion.imageExtent.width = texCube[face].dimensions().x;
bufferCopyRegion.imageExtent.height = texCube[face].dimensions().y;
bufferCopyRegion.imageExtent.depth = 1;
bufferCopyRegion.bufferOffset = offset;
bufferCopyRegions.push_back(bufferCopyRegion);
offset += texCube[face].size();
}
// Transfer cube map faces to optimal tiling
// Setup texture as blit target with optimal tiling
// Create optimal tiled target image
VkImageCreateInfo imageCreateInfo = vkTools::initializers::imageCreateInfo();
imageCreateInfo.imageType = VK_IMAGE_TYPE_2D;
imageCreateInfo.format = format;
imageCreateInfo.mipLevels = 1;
imageCreateInfo.samples = VK_SAMPLE_COUNT_1_BIT;
imageCreateInfo.tiling = VK_IMAGE_TILING_OPTIMAL;
imageCreateInfo.usage = VK_IMAGE_USAGE_SAMPLED_BIT;
imageCreateInfo.sharingMode = VK_SHARING_MODE_EXCLUSIVE;
imageCreateInfo.initialLayout = VK_IMAGE_LAYOUT_PREINITIALIZED;
imageCreateInfo.extent = { cubeMap.width, cubeMap.height, 1 };
imageCreateInfo.usage = VK_IMAGE_USAGE_TRANSFER_DST_BIT | VK_IMAGE_USAGE_SAMPLED_BIT;
imageCreateInfo.flags = VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT;
// Cube faces count as array layers in Vulkan
imageCreateInfo.arrayLayers = 6;
// This flag is required for cube map images
imageCreateInfo.flags = VK_IMAGE_CREATE_CUBE_COMPATIBLE_BIT;
err = vkCreateImage(device, &imageCreateInfo, nullptr, &cubeMap.image);
assert(!err);
VK_CHECK_RESULT(vkCreateImage(device, &imageCreateInfo, nullptr, &cubeMap.image));
vkGetImageMemoryRequirements(device, cubeMap.image, &memReqs);
memAllocInfo.allocationSize = memReqs.size;
memAllocInfo.memoryTypeIndex = getMemoryType(memReqs.memoryTypeBits, VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT);
getMemoryType(memReqs.memoryTypeBits, VK_MEMORY_PROPERTY_DEVICE_LOCAL_BIT, &memAllocInfo.memoryTypeIndex);
err = vkAllocateMemory(device, &memAllocInfo, nullptr, &cubeMap.deviceMemory);
assert(!err);
err = vkBindImageMemory(device, cubeMap.image, cubeMap.deviceMemory, 0);
assert(!err);
VK_CHECK_RESULT(vkAllocateMemory(device, &memAllocInfo, nullptr, &cubeMap.deviceMemory));
VK_CHECK_RESULT(vkBindImageMemory(device, cubeMap.image, cubeMap.deviceMemory, 0));
VkCommandBuffer copyCmd = VulkanExampleBase::createCommandBuffer(VK_COMMAND_BUFFER_LEVEL_PRIMARY, true);
// Image barrier for optimal image (target)
// Optimal image will be used as destination for the copy
// Set initial layout for all array layers of the optimal (target) tiled texture
// Set initial layout for all array layers (faces) of the optimal (target) tiled texture
VkImageSubresourceRange subresourceRange = {};
subresourceRange.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
subresourceRange.baseMipLevel = 0;
@ -237,70 +217,34 @@ public:
subresourceRange.layerCount = 6;
vkTools::setImageLayout(
cmdBuffer,
copyCmd,
cubeMap.image,
VK_IMAGE_ASPECT_COLOR_BIT,
VK_IMAGE_LAYOUT_PREINITIALIZED,
VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
subresourceRange);
// Copy cube map faces one by one
for (uint32_t face = 0; face < 6; ++face)
{
// Copy region for image blit
VkImageCopy copyRegion = {};
copyRegion.srcSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
copyRegion.srcSubresource.baseArrayLayer = 0;
copyRegion.srcSubresource.mipLevel = 0;
copyRegion.srcSubresource.layerCount = 1;
copyRegion.srcOffset = { 0, 0, 0 };
copyRegion.dstSubresource.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT;
copyRegion.dstSubresource.baseArrayLayer = face;
copyRegion.dstSubresource.mipLevel = 0;
copyRegion.dstSubresource.layerCount = 1;
copyRegion.dstOffset = { 0, 0, 0 };
copyRegion.extent.width = cubeMap.width;
copyRegion.extent.height = cubeMap.height;
copyRegion.extent.depth = 1;
// Put image copy into command buffer
vkCmdCopyImage(
cmdBuffer,
cubeFace[face].image,
VK_IMAGE_LAYOUT_TRANSFER_SRC_OPTIMAL,
// Copy the cube map faces from the staging buffer to the optimal tiled image
vkCmdCopyBufferToImage(
copyCmd,
stagingBuffer,
cubeMap.image,
VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
1, &copyRegion);
}
bufferCopyRegions.size(),
bufferCopyRegions.data()
);
// Change texture image layout to shader read after all faces have been copied
cubeMap.imageLayout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL;
vkTools::setImageLayout(
cmdBuffer,
copyCmd,
cubeMap.image,
VK_IMAGE_ASPECT_COLOR_BIT,
VK_IMAGE_LAYOUT_TRANSFER_DST_OPTIMAL,
cubeMap.imageLayout,
subresourceRange);
err = vkEndCommandBuffer(cmdBuffer);
assert(!err);
VkFence nullFence = { VK_NULL_HANDLE };
// Submit command buffer to graphis queue
VkSubmitInfo submitInfo = vkTools::initializers::submitInfo();
submitInfo.commandBufferCount = 1;
submitInfo.pCommandBuffers = &cmdBuffer;
err = vkQueueSubmit(queue, 1, &submitInfo, nullFence);
assert(!err);
err = vkQueueWaitIdle(queue);
assert(!err);
VulkanExampleBase::flushCommandBuffer(copyCmd, queue, true);
// Create sampler
VkSamplerCreateInfo sampler = vkTools::initializers::samplerCreateInfo();
@ -316,27 +260,23 @@ public:
sampler.minLod = 0.0f;
sampler.maxLod = 0.0f;
sampler.borderColor = VK_BORDER_COLOR_FLOAT_OPAQUE_WHITE;
err = vkCreateSampler(device, &sampler, nullptr, &cubeMap.sampler);
assert(!err);
VK_CHECK_RESULT(vkCreateSampler(device, &sampler, nullptr, &cubeMap.sampler));
// Create image view
VkImageViewCreateInfo view = vkTools::initializers::imageViewCreateInfo();
view.image = VK_NULL_HANDLE;
// Cube map view type
view.viewType = VK_IMAGE_VIEW_TYPE_CUBE;
view.format = format;
view.components = { VK_COMPONENT_SWIZZLE_R, VK_COMPONENT_SWIZZLE_G, VK_COMPONENT_SWIZZLE_B, VK_COMPONENT_SWIZZLE_A };
view.subresourceRange = { VK_IMAGE_ASPECT_COLOR_BIT, 0, 1, 0, 1 };
// 6 array layers (faces)
view.subresourceRange.layerCount = 6;
view.image = cubeMap.image;
err = vkCreateImageView(device, &view, nullptr, &cubeMap.view);
assert(!err);
VK_CHECK_RESULT(vkCreateImageView(device, &view, nullptr, &cubeMap.view));
// Cleanup
for (auto& face : cubeFace)
{
vkDestroyImage(device, face.image, nullptr);
vkFreeMemory(device, face.memory, nullptr);
}
// Clean up staging resources
vkFreeMemory(device, stagingMemory, nullptr);
vkDestroyBuffer(device, stagingBuffer, nullptr);
}
void buildCommandBuffers()
@ -356,30 +296,19 @@ public:
renderPassBeginInfo.clearValueCount = 2;
renderPassBeginInfo.pClearValues = clearValues;
VkResult err;
for (int32_t i = 0; i < drawCmdBuffers.size(); ++i)
{
// Set target frame buffer
renderPassBeginInfo.framebuffer = frameBuffers[i];
err = vkBeginCommandBuffer(drawCmdBuffers[i], &cmdBufInfo);
assert(!err);
VK_CHECK_RESULT(vkBeginCommandBuffer(drawCmdBuffers[i], &cmdBufInfo));
vkCmdBeginRenderPass(drawCmdBuffers[i], &renderPassBeginInfo, VK_SUBPASS_CONTENTS_INLINE);
VkViewport viewport = vkTools::initializers::viewport(
(float)width,
(float)height,
0.0f,
1.0f);
VkViewport viewport = vkTools::initializers::viewport((float)width, (float)height, 0.0f, 1.0f);
vkCmdSetViewport(drawCmdBuffers[i], 0, 1, &viewport);
VkRect2D scissor = vkTools::initializers::rect2D(
width,
height,
0,
0);
VkRect2D scissor = vkTools::initializers::rect2D(width, height, 0, 0);
vkCmdSetScissor(drawCmdBuffers[i], 0, 1, &scissor);
VkDeviceSize offsets[1] = { 0 };
@ -400,18 +329,14 @@ public:
vkCmdEndRenderPass(drawCmdBuffers[i]);
err = vkEndCommandBuffer(drawCmdBuffers[i]);
assert(!err);
VK_CHECK_RESULT(vkEndCommandBuffer(drawCmdBuffers[i]));
}
}
void draw()
{
VkResult err;
// Get next image in the swap chain (back/front buffer)
err = swapChain.acquireNextImage(semaphores.presentComplete, &currentBuffer);
assert(!err);
VK_CHECK_RESULT(swapChain.acquireNextImage(semaphores.presentComplete, &currentBuffer));
submitPostPresentBarrier(swapChain.buffers[currentBuffer].image);
@ -420,16 +345,13 @@ public:
submitInfo.pCommandBuffers = &drawCmdBuffers[currentBuffer];
// Submit to queue
err = vkQueueSubmit(queue, 1, &submitInfo, VK_NULL_HANDLE);
assert(!err);
VK_CHECK_RESULT(vkQueueSubmit(queue, 1, &submitInfo, VK_NULL_HANDLE));
submitPrePresentBarrier(swapChain.buffers[currentBuffer].image);
err = swapChain.queuePresent(queue, currentBuffer, semaphores.renderComplete);
assert(!err);
VK_CHECK_RESULT(swapChain.queuePresent(queue, currentBuffer, semaphores.renderComplete));
err = vkQueueWaitIdle(queue);
assert(!err);
VK_CHECK_RESULT(vkQueueWaitIdle(queue));
}
void loadMeshes()
@ -494,8 +416,7 @@ public:
poolSizes.data(),
2);
VkResult vkRes = vkCreateDescriptorPool(device, &descriptorPoolInfo, nullptr, &descriptorPool);
assert(!vkRes);
VK_CHECK_RESULT(vkCreateDescriptorPool(device, &descriptorPoolInfo, nullptr, &descriptorPool));
}
void setupDescriptorSetLayout()
@ -519,16 +440,14 @@ public:
setLayoutBindings.data(),
setLayoutBindings.size());
VkResult err = vkCreateDescriptorSetLayout(device, &descriptorLayout, nullptr, &descriptorSetLayout);
assert(!err);
VK_CHECK_RESULT(vkCreateDescriptorSetLayout(device, &descriptorLayout, nullptr, &descriptorSetLayout));
VkPipelineLayoutCreateInfo pPipelineLayoutCreateInfo =
vkTools::initializers::pipelineLayoutCreateInfo(
&descriptorSetLayout,
1);
err = vkCreatePipelineLayout(device, &pPipelineLayoutCreateInfo, nullptr, &pipelineLayout);
assert(!err);
VK_CHECK_RESULT(vkCreatePipelineLayout(device, &pPipelineLayoutCreateInfo, nullptr, &pipelineLayout));
}
void setupDescriptorSets()
@ -547,8 +466,7 @@ public:
1);
// 3D object descriptor set
VkResult vkRes = vkAllocateDescriptorSets(device, &allocInfo, &descriptorSets.object);
assert(!vkRes);
VK_CHECK_RESULT(vkAllocateDescriptorSets(device, &allocInfo, &descriptorSets.object));
std::vector<VkWriteDescriptorSet> writeDescriptorSets =
{
@ -568,8 +486,7 @@ public:
vkUpdateDescriptorSets(device, writeDescriptorSets.size(), writeDescriptorSets.data(), 0, NULL);
// Sky box descriptor set
vkRes = vkAllocateDescriptorSets(device, &allocInfo, &descriptorSets.skybox);
assert(!vkRes);
VK_CHECK_RESULT(vkAllocateDescriptorSets(device, &allocInfo, &descriptorSets.skybox));
writeDescriptorSets =
{
@ -661,15 +578,13 @@ public:
pipelineCreateInfo.stageCount = shaderStages.size();
pipelineCreateInfo.pStages = shaderStages.data();
VkResult err = vkCreateGraphicsPipelines(device, pipelineCache, 1, &pipelineCreateInfo, nullptr, &pipelines.skybox);
assert(!err);
VK_CHECK_RESULT(vkCreateGraphicsPipelines(device, pipelineCache, 1, &pipelineCreateInfo, nullptr, &pipelines.skybox));
// Cube map reflect pipeline
shaderStages[0] = loadShader(getAssetPath() + "shaders/cubemap/reflect.vert.spv", VK_SHADER_STAGE_VERTEX_BIT);
shaderStages[1] = loadShader(getAssetPath() + "shaders/cubemap/reflect.frag.spv", VK_SHADER_STAGE_FRAGMENT_BIT);
depthStencilState.depthWriteEnable = VK_TRUE;
err = vkCreateGraphicsPipelines(device, pipelineCache, 1, &pipelineCreateInfo, nullptr, &pipelines.reflect);
assert(!err);
VK_CHECK_RESULT(vkCreateGraphicsPipelines(device, pipelineCache, 1, &pipelineCreateInfo, nullptr, &pipelines.reflect));
}
// Prepare and initialize uniform buffer containing shader uniforms
@ -708,8 +623,7 @@ public:
uboVS.model = glm::rotate(uboVS.model, glm::radians(rotation.z), glm::vec3(0.0f, 0.0f, 1.0f));
uint8_t *pData;
VkResult err = vkMapMemory(device, uniformData.objectVS.memory, 0, sizeof(uboVS), 0, (void **)&pData);
assert(!err);
VK_CHECK_RESULT(vkMapMemory(device, uniformData.objectVS.memory, 0, sizeof(uboVS), 0, (void **)&pData));
memcpy(pData, &uboVS, sizeof(uboVS));
vkUnmapMemory(device, uniformData.objectVS.memory);
@ -723,8 +637,7 @@ public:
uboVS.model = glm::rotate(uboVS.model, glm::radians(rotation.y), glm::vec3(0.0f, 1.0f, 0.0f));
uboVS.model = glm::rotate(uboVS.model, glm::radians(rotation.z), glm::vec3(0.0f, 0.0f, 1.0f));
err = vkMapMemory(device, uniformData.skyboxVS.memory, 0, sizeof(uboVS), 0, (void **)&pData);
assert(!err);
VK_CHECK_RESULT(vkMapMemory(device, uniformData.skyboxVS.memory, 0, sizeof(uboVS), 0, (void **)&pData));
memcpy(pData, &uboVS, sizeof(uboVS));
vkUnmapMemory(device, uniformData.skyboxVS.memory);
}