Files
llama.cpp/ggml
Hongqiang WangandLi He b062ba735e opencl: port fused ssm_scan kernel (Mamba-2, d_state in {128, 256}) to GPU (#26439)
* opencl: port fused ssm_scan kernel (Mamba-2, d_state in {128, 256})

Fold the fused per-token SSM_SCAN recurrent step from opencl/gdn-qwen36-35b
onto the unified base. Previously SSM_SCAN fell back to CPU here; now scalar-A
Mamba-2 with d_state in {128,256}, all-f32, runs on GPU. Other shapes (incl.
Mamba-1 element-wise A) still fall back. test-backend-ops -o SSM_SCAN passes on
Adreno X2-90. opt-out via GGML_OPENCL_DISABLE_SSM_SCAN=1.

* opencl: cleanup

* opencl: require K == 1

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>
2026-08-19 13:35:17 -07:00
..
2024-07-13 18:12:39 +02:00