marfrit/rocky-vulkan-llama
1
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

Shader iter6: BK_STEP=1 regressed (barrier crossover); BK_STEP=2 is sweet spot, reverted to iter5 champion

Browse Source
This commit is contained in:
Claude (noether)
2026-07-11 18:36:21 +02:00
parent 220a3b1667
commit 6ee32814b2
1 changed files with 9 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/shader-tuning.md
+9
View File
@@ -71,3 +71,12 @@ tuples carry the 12th element (=2), so L gets BK_STEP=2 while M/S inherit defaul
| 128 (L) | 28.1 | **28.08** | held (L win kept) | | 128 (L) | 28.1 | **28.08** | held (L win kept) |
| 256 (L) | 28.5 | **28.46** | held | | 256 (L) | 28.5 | **28.46** | held |
L-path 28 t/s WITHOUT the M regression. Shippable champion. Progression (L): 5.8→23.4→26.2→28.1. L-path 28 t/s WITHOUT the M regression. Shippable champion. Progression (L): 5.8→23.4→26.2→28.1.
## Iteration 6 — BK_STEP=1 (full 2048 occupancy attempt) — REGRESSED (barrier crossover)
| ubatch | iter5 | iter6 | |
|--------|-------|-------|--|
| 128 (L) | 28.1 | **23.1** | ↓ 4× K-loop barriers outran the occupancy gain |
| 256 (L) | 28.5 | **23.3** | ↓ |
BK_STEP=2 is the occupancy sweet spot; reverted to iter5 champion (28 t/s). Shared-based occupancy
exhausted. Next lever: **f16 accumulators** (register-based occupancy, no extra barriers) to break
the 33–64 reg band toward 2048 resident threads.