Open code, practice, and notes →
Download source with English comments · for reading
WebGPU · 15 A First Compute Array Transform
Series contents · Technology map
Stage: Compute and Data Movement · Editable example · Storage ID: webgpu-step-15
Goals and learning outcomes
- Implement independent-element compute.
- Handle dispatch boundaries.
Prerequisites
Continue using the output and notes from unit 14. Preserve the previous inputs and reference outputs before starting "A First Compute Array Transform".
Key concepts and explanations
Parallel Indices
global_invocation_id can identify the element to process. Dispatch size depends on both element count and workgroup size.
Bounds Checks
The final workgroup may contain more invocations than actual data elements. In independent-element kernels, check the length before accessing data.
Practice setup
First read the execution-scope guidance on the example screen. The following preparation is for extending the assignment in a separate project. This is a document-guided exercise to implement yourself in a separate localhost or HTTPS project. Check the secure context, navigator.gpu, and adapter request result, and provide an unsupported-environment message. Preserve device.lost handling and resource release.
The linked example is a starting point. It does not implement every task below. Complete work beyond its supported scope in a separate project.
Practice tasks
- In compute.wgsl, write a kernel that doubles each input array element.
- Create two storage buffers and a compute pipeline, and dispatch ceil(N/64) workgroups.
- Use N=65 and copy the output to a staging buffer to check through the last element.
Variables to observe
| Variable | Baseline | Change | Expected change / hypothesis |
|---|---|---|---|
| Array length | 64 | 65 | Only one element in the second workgroup is expected to perform a valid calculation. |
Predictions in this table are not measured results. Change one condition at a time and record why the actual result differs.
Completion criteria
- All 65 results match the CPU doubling results.
- Exclude invocations beyond the length before buffer access.
Self-check
- Does dispatchWorkgroups take an element count?
- Does adding a barrier require reviewing early-return conditions?
Official references
Linked example and execution scope
Example source · Runtime environments and limits
Runs real WebGPU/WGSL. Requires HTTPS/localhost and a supported adapter; unsupported environments display a reason, and CPU results are not substituted for GPU results.
Source code and records are not overwritten automatically. Resetting an existing example also preserves your notes.
Next learning
Previous: Buffers and Render Passes · Next: Storage and Ping-Pong