llama.cpp/ggml/src/ggml-cann/kernels
leo-pony c18610b4ee
CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216)
* CANN Support Ascend310P to accelerate F32 and F16 Model

* Add compile option soc type macro ASCEND_310P to ggml-cann lib

* Remove unused code

* Remove the ascend soc_type hard code compile option in CMakelist.txt
2024-11-22 14:07:20 +08:00
..
ascendc_kernels.h cann: support q4_0 model (#8822) 2024-08-05 12:22:30 +08:00
CMakeLists.txt CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216) 2024-11-22 14:07:20 +08:00
dup.cpp CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216) 2024-11-22 14:07:20 +08:00
get_row_f16.cpp CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216) 2024-11-22 14:07:20 +08:00
get_row_f32.cpp CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216) 2024-11-22 14:07:20 +08:00
get_row_q4_0.cpp CANN: Support Ascend310P to accelerate F32 and F16 Model (#10216) 2024-11-22 14:07:20 +08:00
get_row_q8_0.cpp [CANN] Add Ascend NPU backend (#6035) 2024-07-17 14:23:50 +03:00
quantize_f16_q8_0.cpp [CANN] Add Ascend NPU backend (#6035) 2024-07-17 14:23:50 +03:00
quantize_f32_q8_0.cpp [CANN] Add Ascend NPU backend (#6035) 2024-07-17 14:23:50 +03:00
quantize_float_to_q4_0.cpp cann: fix buffer_num and runtime speed slowly error (#8865) 2024-08-05 21:10:37 +08:00