EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 Projects Releases Wiki Activity

ggml: allow casting between f32 and i32 (#15783)

Browse Source 
* ggml: allow casting between f32 and i32

* fix cuda

* add vulkan

* fix CPU non-cont

* add non-cont test case

* add note

* extend test number range

* correct note

* add cont version for vulkan
This commit is contained in:
Xuan-Son Nguyen
2025-09-08 17:33:01 +07:00
committed by GitHub
parent 5ef22d281d
commit 9fcb29f22f
12 changed files with 247 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-metal/ggml-metal.metal
View File

@@ -5338,6 +5338,8 @@ typedef decltype(kernel_cpy<float, float>) kernel_cpy_t;
template [[host_name("kernel_cpy_f32_f32")]] kernel kernel_cpy_t kernel_cpy<float, float>;
template [[host_name("kernel_cpy_f32_f16")]] kernel kernel_cpy_t kernel_cpy<float, half>;
template [[host_name("kernel_cpy_f32_i32")]] kernel kernel_cpy_t kernel_cpy<float, int32_t>;
template [[host_name("kernel_cpy_i32_f32")]] kernel kernel_cpy_t kernel_cpy<int32_t, float>;
#if defined(GGML_METAL_USE_BF16)
template [[host_name("kernel_cpy_f32_bf16")]] kernel kernel_cpy_t kernel_cpy<float, bfloat>;
#endif