Repository navigation
Vulkan Performance
This section documents Vulkan renderer performance by code path and ownership boundary.
flowchart LR
A[Guest/NV2A state] --> B[Shader state + binding]
B --> C[Stage modules]
C --> D[Graphics pipeline]
D --> E[Draw recording]
E --> F[Vertex/data synchronization]
F --> G[Command buffer / render pass]
G --> H[GPU execution]
A slowdown can appear at any of these layers. Similar symptoms do not imply the same cause.
Vulkan Shader Binding & Compilation explains how shader state becomes generated GLSL, SPIR-V, VkShaderModule objects, reflection state, shader bindings, and ultimately a graphics pipeline. It also documents why shader-binding misses, stage-module misses, and pipeline misses must be counted separately.
Vulkan Ubershader & Hybrid Specialization documents the fallback/interpreter route used to avoid making every admitted first-seen fragment specialization a synchronous draw dependency. It covers the control ABI, fail-closed admission, route policy, background compiler, required compiler lane, publication boundary, and descriptor-layout consequences.
Vulkan Vertex Uploads & Render-Pass Lifetime documents the Vulkan-side mirror of guest vertex RAM and the proof required before replacing an ordered staged upload with a mapped CPU write during an active batch.
flowchart TD
A[Measured bad frame] --> B[Classify shader/pipeline cost]
B --> C[PR #68 diagnostic evidence]
C --> D{First-seen specialization still blocks draw?}
D -->|Yes| E[Fallback ubershader + background specialization]
E --> F[PR #71]
B --> G[Inspect other synchronization / pass disruption]
G --> H[Vertex mirror lifetime proof]
H --> I[PR #87]
PRs are implementation history and research references. The architecture pages are the durable source of explanation.
DSP tooling guide and navigation updated November 2, 2026 for experimental PR #384, source 8f51d237. Capture/replay usage and limits.
The earlier wiki-wide review was published November 1, 2026. Emulator source: b7140848. Reviewed wiki content: 1d2b671. Current status · Release availability · Publication process.
- Home
- Current status
- Stable versus main
- Implemented mechanisms
- Controls and defaults
- Measured results and limits
- Complete change register
- System
- CPU and vCPU
- Memory and DMA
- Graphics / NV2A
- APU and audio
- Timing and timers
- Platform and I/O
- Hardware observations
- Fidelity boundaries
- Methodology
- PGR2 routes
- Debugging lag / hot paths
- Tools and commands
- DSP capture, replay and fuzzing
- HLE readiness and public XISO proof
- Concurrent DSP shadow and OpenAL proof
- DSP HLE operation inventory
- DSP HLE fix examples and testing
- Development
- Upstream shortlist
- Wiki maintenance
- Research / retired work
- Optimization catalog
- Hot-path shortcuts
- Researched experiments
- Unproven proposal directory
- PR lessons
- September26 snapshot
Repository · PRs · Issues