aleleba/lightcraft
1
0
Fork 0
mirror of https://github.com/storytold/lightcraft.git synced 2026-11-01 05:57:09 -06:00
Code Issues Packages Projects Releases Wiki Activity

Issue #2936: gpu-pipeline.md says what is measured about the remaining kernel time

Browse Source 
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018QWU3BMJMENFqFqxDHzV3b
This commit is contained in:
Ingmar Lübbe
2026-10-25 16:39:14 +01:00
co-authored by Claude Opus 5.5
parent fa4bccff64
commit 1ada7c2af0
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/gpu-pipeline.md
+1 -1
View File
@@ -306,7 +306,7 @@ On Windows / DX12 the kernels cost more: FXC compiles each pipeline single-threa
`create_compute_pipeline` (issue #471). Built one after another, the 27 took ~2.2 s on an Intel Core
Ultra 7 165U iGPU (`app.gpu` 2.4 s median, up to 7 s on the first run after an install), so
`ctx::build_kernels` shares them out to worker threads (issue #2936): ~1.05 s there (`app.gpu` 1.2 s),
now bounded by the largest kernel (`finish.wgsl`). wgpu's error scopes are per thread, so every worker
most likely the largest kernel (`finish.wgsl`) alone (not timed per kernel). wgpu's error scopes are per thread, so every worker
opens its own; a refused kernel still means no GPU. The `gpu:` info log line gives the kernel count
and build time.