Markus Fritsche 636fa430cd Champion shader ported to July build (29x over stock GPU) BUT ties CPU prefill / loses decode -> no faster-than-CPU win on Mali (VA squeeze). Honest correction.

2026-07-12 07:05:03 +02:00

..

benchmarks.md

panvk WLS fix v1 + confirmed mechanism (partial: reuse halves but doesn't eliminate)

2026-07-11 14:06:47 +02:00

fable-review.md

panvk WLS fix v1 + confirmed mechanism (partial: reuse halves but doesn't eliminate)

2026-07-11 14:06:47 +02:00

fix-plan.md

Add fix plan: cleanup_frequency=10 is the accumulation trigger; consumer-side bounding is the primary lever

2026-07-11 14:58:50 +02:00

gemma-gpu-kernel-bringup-2026-07-12.md

Champion shader ported to July build (29x over stock GPU) BUT ties CPU prefill / loses decode -> no faster-than-CPU win on Mali (VA squeeze). Honest correction.

2026-07-12 07:05:03 +02:00

RESULTS.md

FIX: panvk priv-heap exhaustion solved (shared TLS + WLS reuse); crash gone ub64->1024

2026-07-11 16:30:55 +02:00

root-cause-final.md

Phase 1: consumer-side levers falsified; accumulation is in-flight cmdbufs x 304MB TLS

2026-07-11 16:00:09 +02:00

root-cause.md

rocky-vulkan-llama: skeleton — deps, root-cause diagnosis, llama.cpp submit patch

2026-07-11 13:33:00 +02:00

shader-tuning.md

Shader CHAMPION converged: 28 t/s (4.8x), correctness-gated; f16acc refused; ~15% left is quantize-dispatch (profile, not tile)

2026-07-11 19:11:46 +02:00