shalinib-ibm ce2c003f17 ggml-cpu: Enable tiled matmul on AIX (llama/25199) ...

The matmul_tiled path uses large local stack buffers for A_pack and B_pack. On AIX this can trigger a segmentation fault, so reduce the buffer footprint there to keep the tiled path usable.

 Performance Impact:
    ~ 2x gains in PP_Speed for FP32, Q4_0 and Q8_0 models tested with llama-bench, llama-batched-bench and llama-cli.
    Models used: Llama3.2 3b Instruct F32, qwen 2.5 3b Q4_0 and Q8_0

2026-07-10 13:06:42 +03:00

..

cmake

ggml : Parallelize quant LUT init (llama/23595)

2026-05-25 12:26:07 +03:00

include

sycl : support --split-mode tensor (llama/24152)

2026-06-26 16:03:57 +03:00

src

ggml-cpu: Enable tiled matmul on AIX (llama/25199)

2026-07-10 13:06:42 +03:00

.gitignore

…

CMakeLists.txt

ggml : bump version to 0.15.3 (ggml/1550)

2026-06-26 16:03:57 +03:00