Claude (noether) 05f240ef6c Shader iter3 CHAMPION: BLOCK 256 same-registers -> ub128 25.8 t/s (+10%), spill-free ...

Shared-occupancy is the lever. Progression 5.8->23.4->26.2 t/s (~30% roofline). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

2026-07-11 17:58:10 +02:00

..

llama.cpp-0001-vk-tunable-submit-thresholds.patch

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

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

llama.cpp-0002-vk-cmd-cleanup-freq-env.patch

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

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

llama.cpp-0003-mali-warptile-iter1.patch

Shader iter1: Mali L-warptile (WM=WN=16) -> ub128 4x (5.8->23.4 t/s), spill 5328->96

2026-07-11 17:29:55 +02:00

llama.cpp-0003-mali-warptile-iter3-champion.patch

Shader iter3 CHAMPION: BLOCK 256 same-registers -> ub128 25.8 t/s (+10%), spill-free

2026-07-11 17:58:10 +02:00

mesa-panvk-0010-fix-priv-heap-exhaustion-shared-tls-wls.patch

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

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