ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)

Browse source 
* Add SVE support for q4_0_q8_0 q8_0_q8_0

* remove ifdef
This commit is contained in:
Masaya, Kato 2024-05-25 17:42:31 +09:00 • committed by GitHub
parent 9791f40258
commit faa0e6979a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 85 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

1
llama.cpp
View file

@ -18337,6 +18337,7 @@ const char * llama_print_system_info(void) {
s += "AVX512_BF16 = " + std::to_string(ggml_cpu_has_avx512_bf16()) + " | ";
s += "FMA = " + std::to_string(ggml_cpu_has_fma()) + " | ";
s += "NEON = " + std::to_string(ggml_cpu_has_neon()) + " | ";
s += "SVE = " + std::to_string(ggml_cpu_has_sve()) + " | ";
s += "ARM_FMA = " + std::to_string(ggml_cpu_has_arm_fma()) + " | ";
s += "F16C = " + std::to_string(ggml_cpu_has_f16c()) + " | ";
s += "FP16_VA = " + std::to_string(ggml_cpu_has_fp16_va()) + " | ";