llama.cpp/ggml
Gong-Mi 56b8f27fab vulkan: optimize ARM Mali tile sizes and re-enable FP16
- Tunes l_warptile to match m_warptile (64x64) for ARM GPUs, fixing low occupancy on medium-sized matrices.

- Re-enables FP16/BF16 support for ARM as the tiling fix resolves the performance regression.

- Adds comments clarifying the UMA memory allocation fallback strategy.
2026-01-02 12:04:32 +08:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) 2025-08-07 13:45:41 +02:00
include vulkan: extend topk_moe to handle sigmoid w/exp_probs_b for nemotron (#18295) 2026-01-01 08:58:27 +01:00
src vulkan: optimize ARM Mali tile sizes and re-enable FP16 2026-01-02 12:04:32 +08:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt ggml : bump version to 0.9.5 (ggml/1410) 2025-12-31 18:54:43 +02:00