Catcrafts/Crafter.Graphics
2
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

perf(window): cache per-frame heap-bind structs across frames

Browse source 
Render() rebuilt two VkBindHeapInfoEXT structs every frame and
recomputed reservedRangeOffset via a runtime subtract + bitmask
against Device::descriptorHeapProperties. The heap address/size per
slot are stable for a given heap; only currentBuffer varies.

Precompute numFrames resource/sampler bind structs and cache them,
keyed on the descriptorHeap pointer. The cache rebuilds whenever the
heap is (re)assigned and otherwise just indexes by currentBuffer.
Pointer-keyed invalidation catches any heap reassignment — the same
point a setter hook would fire — and is independent of onResize, as
the heaps never resize today (per the issue's correctness caveat).

Resolves #42

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
catbot 2026-06-16 15:27:10 +00:00
commit 8b95b7e883
2 changed files with 41 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

50
implementations/Crafter.Graphics-Window.cpp
View file

@ -734,27 +734,35 @@ void Window::Render() {
// Pass-side dispatches still run with the same heaps bound — moving // Pass-side dispatches still run with the same heaps bound — moving
// the bind earlier doesn't change anything for them. // the bind earlier doesn't change anything for them.
if (descriptorHeap) { if (descriptorHeap) {
VkBindHeapInfoEXT resourceHeapInfo = { // Rebuild the cached per-frame bind structs only when the heap changes.
.sType = VK_STRUCTURE_TYPE_BIND_HEAP_INFO_EXT, // The address/size of each frame's heap and the reservedRangeOffset
.heapRange = { // bitmask are stable for a given heap, so this runs once per assignment
.address = descriptorHeap->resourceHeap[currentBuffer].address, // rather than every frame.
.size = static_cast<std::uint32_t>(descriptorHeap->resourceHeap[currentBuffer].size) if (descriptorHeap != cachedDescriptorHeap) {
}, for (std::uint8_t i = 0; i < numFrames; ++i) {
.reservedRangeOffset = (descriptorHeap->resourceHeap[currentBuffer].size - Device::descriptorHeapProperties.minResourceHeapReservedRange) & ~(Device::descriptorHeapProperties.imageDescriptorAlignment - 1), resourceHeapInfos[i] = {
.reservedRangeSize = Device::descriptorHeapProperties.minResourceHeapReservedRange .sType = VK_STRUCTURE_TYPE_BIND_HEAP_INFO_EXT,
}; .heapRange = {
Device::vkCmdBindResourceHeapEXT(drawCmdBuffers[currentBuffer], &resourceHeapInfo); .address = descriptorHeap->resourceHeap[i].address,
.size = static_cast<std::uint32_t>(descriptorHeap->resourceHeap[i].size)
VkBindHeapInfoEXT samplerHeapInfo = { },
.sType = VK_STRUCTURE_TYPE_BIND_HEAP_INFO_EXT, .reservedRangeOffset = (descriptorHeap->resourceHeap[i].size - Device::descriptorHeapProperties.minResourceHeapReservedRange) & ~(Device::descriptorHeapProperties.imageDescriptorAlignment - 1),
.heapRange = { .reservedRangeSize = Device::descriptorHeapProperties.minResourceHeapReservedRange
.address = descriptorHeap->samplerHeap[currentBuffer].address, };
.size = static_cast<std::uint32_t>(descriptorHeap->samplerHeap[currentBuffer].size) samplerHeapInfos[i] = {
}, .sType = VK_STRUCTURE_TYPE_BIND_HEAP_INFO_EXT,
.reservedRangeOffset = descriptorHeap->samplerHeap[currentBuffer].size - Device::descriptorHeapProperties.minSamplerHeapReservedRange, .heapRange = {
.reservedRangeSize = Device::descriptorHeapProperties.minSamplerHeapReservedRange .address = descriptorHeap->samplerHeap[i].address,
}; .size = static_cast<std::uint32_t>(descriptorHeap->samplerHeap[i].size)
Device::vkCmdBindSamplerHeapEXT(drawCmdBuffers[currentBuffer], &samplerHeapInfo); },
.reservedRangeOffset = descriptorHeap->samplerHeap[i].size - Device::descriptorHeapProperties.minSamplerHeapReservedRange,
.reservedRangeSize = Device::descriptorHeapProperties.minSamplerHeapReservedRange
};
}
cachedDescriptorHeap = descriptorHeap;
}
Device::vkCmdBindResourceHeapEXT(drawCmdBuffers[currentBuffer], &resourceHeapInfos[currentBuffer]);
Device::vkCmdBindSamplerHeapEXT(drawCmdBuffers[currentBuffer], &samplerHeapInfos[currentBuffer]);
} }
onUpdate.Invoke({startTime, startTime-lastFrameBegin}); onUpdate.Invoke({startTime, startTime-lastFrameBegin});

12
interfaces/Crafter.Graphics-Window.cppm
View file

@ -264,6 +264,18 @@ export namespace Crafter {
VkPipelineStageFlags submitPipelineStages = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT; VkPipelineStageFlags submitPipelineStages = VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT;
std::vector<RenderPass*> passes; std::vector<RenderPass*> passes;
DescriptorHeapVulkan* descriptorHeap = nullptr; DescriptorHeapVulkan* descriptorHeap = nullptr;
// Cached per-frame heap-bind structs. The heap address/size per slot
// are stable once a heap is assigned; only currentBuffer varies frame
// to frame, so the structs (and the reservedRangeOffset arithmetic)
// are computed once per heap rather than every frame in Render().
// Keyed on the heap pointer: whenever descriptorHeap differs from
// cachedDescriptorHeap the cache is rebuilt. This invalidates on any
// heap (re)assignment — heaps never resize today, so a stable pointer
// means stable ranges. Not tied to onResize (the heap is independent
// of swapchain size).
VkBindHeapInfoEXT resourceHeapInfos[numFrames];
VkBindHeapInfoEXT samplerHeapInfos[numFrames];
DescriptorHeapVulkan* cachedDescriptorHeap = nullptr;
std::optional<std::array<float, 4>> clearColor; std::optional<std::array<float, 4>> clearColor;
#else #else
// DOM mode: the page IS the window. WebGPU device and canvas are // DOM mode: the page IS the window. WebGPU device and canvas are