mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 32 Packages Projects Releases Wiki Activity

kleidiai: add and integrate SVE 256-bit vector-length kernel (#18458)

Browse Source 
* kleidiai: add and integrate SVE 256-bit vector-length kernel

* updated for review comments
This commit is contained in:
Charles Xu
2025-12-30 13:04:53 +01:00
committed by GitHub
parent d77d7c5c06
commit 2d6c00a9b8
4 changed files with 115 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/CMakeLists.txt
+2 -2
View File
@@ -401,8 +401,8 @@ if (GGML_CPU_ALL_VARIANTS)
ggml_add_cpu_backend_variant(android_armv8.2_2 DOTPROD FP16_VECTOR_ARITHMETIC)
ggml_add_cpu_backend_variant(android_armv8.6_1 DOTPROD FP16_VECTOR_ARITHMETIC MATMUL_INT8)
ggml_add_cpu_backend_variant(android_armv9.0_1 DOTPROD MATMUL_INT8 FP16_VECTOR_ARITHMETIC SVE2)
ggml_add_cpu_backend_variant(android_armv9.2_1 DOTPROD MATMUL_INT8 FP16_VECTOR_ARITHMETIC SME)
ggml_add_cpu_backend_variant(android_armv9.2_2 DOTPROD MATMUL_INT8 FP16_VECTOR_ARITHMETIC SVE SME)
ggml_add_cpu_backend_variant(android_armv9.2_1 DOTPROD MATMUL_INT8 FP16_VECTOR_ARITHMETIC SVE SME)
ggml_add_cpu_backend_variant(android_armv9.2_2 DOTPROD MATMUL_INT8 FP16_VECTOR_ARITHMETIC SVE SVE2 SME)
elseif (APPLE)
ggml_add_cpu_backend_variant(apple_m1 DOTPROD)
ggml_add_cpu_backend_variant(apple_m2_m3 DOTPROD MATMUL_INT8)