neovide/neovide · error

Failed to create query readback buffer.

Error message

Failed to create query readback buffer.

What it means

In `create_d3d_gpu_context` (src/profiling/d3d.rs:197), the Tracy GPU profiling context creates a readback buffer in a default heap to pull query results back to the CPU. `ID3D12Device::CreateCommittedResource` returned a failure HRESULT, so the code panics via `.expect()`. Typical causes are out of memory, an invalid device, or unsupported heap/resource parameters.

Source

Thrown at src/profiling/d3d.rs:197

        let readback_heap_props = D3D12_HEAP_PROPERTIES {
            Type: D3D12_HEAP_TYPE_READBACK,
            CPUPageProperty: D3D12_CPU_PAGE_PROPERTY_UNKNOWN,
            MemoryPoolPreference: D3D12_MEMORY_POOL_UNKNOWN,
            CreationNodeMask: 0,
            VisibleNodeMask: 0, // TODO: Support multiple adapters.
        };

        let mut readback_buffer: Option<ID3D12Resource> = None;
        device
            .CreateCommittedResource(
                &readback_heap_props,
                D3D12_HEAP_FLAG_NONE,
                &readback_buffer_desc,
                D3D12_RESOURCE_STATE_COPY_DEST,
                None,
                &mut readback_buffer,
            )
            .expect("Failed to create query readback buffer.");

        let payload_fence: ID3D12Fence =
            device.CreateFence(0, D3D12_FENCE_FLAG_NONE).expect("Failed to create payload fence.");

        let command_allocator: ID3D12CommandAllocator = device
            .CreateCommandAllocator(D3D12_COMMAND_LIST_TYPE_DIRECT)
            .expect("Failed to create command allocator");

        let command_list: ID3D12GraphicsCommandList = device
            .CreateCommandList(
                0,
                D3D12_COMMAND_LIST_TYPE_DIRECT,
                &command_allocator,
                &ID3D12PipelineState::from_raw(null_mut()),
            )
            .expect("Failed to create command list");

        (

View on GitHub (pinned to ade2d9cda7)

Solutions

1. Update GPU drivers and verify the D3D12 device is healthy (no D3D12_ERROR_DEVICE_REMOVED / DXGI_ERROR_DEVICE_REMOVED).
2. Check free memory; the readback buffer allocation can fail under VRAM/system memory pressure.
3. Disable the gpu_profiling feature if Tracy GPU profiling is not needed.
4. Enable the D3D12 debug layer to get the precise HRESULT reason from the debug output.

// before enabling GPU profiling
if !device.is_valid() || device_removed { /* skip tracy_create_gpu_context */ }

// keep profiling optional
let result = std::panic::catch_unwind(|| tracy_create_gpu_context(...));
if result.is_err() { eprintln!("GPU profiling unavailable"); }

When it happens

Trigger: Calling `tracy_create_gpu_context` (which invokes `create_d3d_gpu_context`) when `CreateCommittedResource` for the readback buffer fails — e.g. device removed, E_OUTOFMEMORY, or the D3D12 device was created with debug-layer-detected invalid state.

Common situations: GPU driver crash / device-removed state; running out of VRAM; enabling gpu_profiling on systems with broken or outdated D3D12 drivers; creating the profiling context after the device was already destroyed.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

• Failed to create payload fence.
• Failed to create command allocator
• Failed to create command list
• Profiling context not initialized for current thread
• Failed to create Direct3D debug controller

AI-assisted analysis of neovide/neovide@ade2d9cda7 (2026-09-06). Data as JSON: /api/errors/b5add45343c6492b. Report an issue: GitHub.