Loading...
Searching...
No Matches
interop Directory Reference
Files | |
| AmdPinnedBuffers.hpp | |
| AMD GL pinned-memory and bus-addressable extension wrapper. | |
| AVHWFrameToQRhi.hpp | |
| Small inline helpers for the AVFrame opaque → QRhiTexture pattern. | |
| BorrowedHostImportCapture.hpp | |
| Zero-copy capture from buffers the producer already owns, via VK_EXT_external_memory_host. | |
| CaptureCorrelator.hpp | |
| Assembles one CaptureFrameSet out of several independent capture threads, for a rig whose sensors are separate V4L2 devices. | |
| CaptureStrategyCommon.hpp | |
| Backend-neutral plumbing shared by GPU-direct video CAPTURE strategies (no graphics-API headers). | |
| CaptureSyncGroup.hpp | |
| Frame-accurate correlation across several capture streams. | |
| ComputeRingDispatcher.hpp | |
| Bundles a ImportedGpuBufferRing + per-slot ComputeEncoder + InteropFence into the per-frame "encode + signal" loop every vendor GPU-direct path runs. | |
| CpuStagedCapture.hpp | |
| Generic CPU-staging (host-staged) video capture strategy. | |
| CpuStagedVideoOutput.hpp | |
| Vendor-neutral host-staged video OUTPUT helper. | |
| CudaFunctions.hpp | |
| Shared dlopen'd CUDA driver-API table for score-plugin-gfx. | |
| CudaInterop.cpp | |
| Driver-API-only impl of the CudaInterop C API. | |
| CudaVmmAllocator.hpp | |
| BAR1-mappable GPU memory allocator using the CUDA VMM API. | |
| D3D12HostImportUpload.hpp | |
| Zero-copy CPU->GPU upload on D3D12, the counterpart of VkHostImportUpload. | |
| DirectShowPixelFormat.hpp | |
| DirectShow / Video-for-Windows fourcc <-> VideoPixelFormat. | |
| DirectShowSubtype.hpp | |
| The DirectShow enumeration decision, expressed without windows.h. | |
| DMABufImport.hpp | |
| DmaBufImportCapture.hpp | |
| Zero-copy capture strategy: import a producer's DMA-BUF ring straight into the renderer's sampled texture. | |
| DmaLockPolicy.hpp | |
| DMA-lock policy concept for the shared NVIDIA-DVP shims. | |
| DrmFourcc.hpp | |
| DrmFunctions.hpp | |
| dlopen'd libdrm entry points, so KMS output adds no link-time dep. | |
| DrmKmsDevice.hpp | |
| Direct DRM/KMS scanout — output with no compositor in the path. | |
| DrmPixelFormat.hpp | |
| DRM fourcc <-> VideoPixelFormat, and the PipeWire SPA formats. | |
| DrmPrimeWrap.hpp | |
| DvpCaptureD3D11.hpp | |
| DvpCaptureGl.hpp | |
| DvpOutputD3D11.hpp | |
| DvpOutputGl.hpp | |
| EglDmaBufExport.hpp | |
| EglDmaBufImport.hpp | |
| GLCaptureUpload.hpp | |
| OpenGL-specific upload helpers for GPU-direct video CAPTURE strategies that land a captured frame in a QRhi GL texture. | |
| GpuCapabilities.hpp | |
| Runtime probe of which GPU interop mechanisms are available on this system. | |
| GStreamerPixelFormat.hpp | |
| GStreamer video-format name <-> VideoPixelFormat. | |
| HostFramePool.hpp | |
| App-owned host frame pool behind a FrameMemoryProvider. | |
| HostPinnedRing.hpp | |
| Vendor-neutral N-slot sysmem ring for DVP / AMD-pinned / host-copy transfers. | |
| ImportedGpuBufferRing.hpp | |
| Backend-uniform N-slot exportable GPU buffer ring for vendor P2P. | |
| InteropFence.hpp | |
| Backend-uniform fence between QRhi command-buffer + CUDA stream. | |
| PacedFramePump.hpp | |
| Vendor-neutral VBI/clock-paced frame submission. | |
| RdmaGpuBuffer.hpp | |
| Vendor-neutral allocator of RDMA-capable GPU VRAM. | |
| RdmaPlayoutProbe.hpp | |
| Content-verified one-time probe of the GPU->peer (playout) P2P path. | |
| RdmaRingDepth.hpp | |
| Shared BAR1-aware slot-count policy for the RDMA rings. | |
| RdmaVideoOutput.hpp | |
| Vendor-neutral GPU-direct output strategy. | |
| StageProfiler.hpp | |
| Env-gated per-stage wall-time accumulator for the GPU-direct video pipelines. | |
| TextureShare.hpp | |
| TripleBufferIndex.hpp | |
| Lock-free slot rotation behind TextureShare's producer/consumer pair. | |
| V4L2Loader.hpp | |
| dlopen'd libv4l2 entry points, shared by every V4L2 consumer. | |
| V4L2PixelFormat.hpp | |
| The one V4L2 fourcc <-> VideoPixelFormat table. | |
| VendorDmaRegistrar.hpp | |
| Vendor-supplied pin / unpin callbacks shared by the GPU-direct and host-staged output helpers. | |
| VideoCaptureStrategy.hpp | |
| Vendor-neutral interface for GPU-direct video CAPTURE strategies. | |
| VideoOutputStrategy.hpp | |
| Vendor-neutral interface for GPU-direct video OUTPUT strategies. | |
| VideoOutputStrategySelect.hpp | |
| Generic "try GPU-direct output strategies in order" picker. | |
| VideoPixelFormat.hpp | |
| Vendor-neutral pixel format vocabulary + descriptive metadata. | |
| VideoPixelFormatAV.hpp | |
Bridge between the wire-format enum VideoPixelFormat and libav's AVPixelFormat. | |
| VideoPixelFormatQRhi.hpp | |
| Bridge between a CPU/DMA buffer layout and a GPU texture format. | |
| VkCudaSemaphore.hpp | |
| Exportable Vulkan timeline semaphore + CUDA-side import pair. | |
| VkExternalMemoryHelpers.hpp | |
| Vulkan external-memory image/buffer create + export + import helpers. | |
| VkHostImportUpload.hpp | |
| Zero-copy CPU->GPU upload on Vulkan via VK_EXT_external_memory_host. | |
| VulkanCudaBounce.hpp | |
| CUDA-VMM bounce buffers imported into Vulkan — the bridge between QRhi-owned resources and a vendor DMA engine. | |
| VulkanRhiContext.hpp | |
| Acquire the raw Vulkan handles (instance/physDev/device/queue/family) behind a Vulkan QRhi, for the RDMA GPU-direct interop paths. | |