mac os Big Sur 11.7.4 Linking Error , Undefined symbols

[junlong]

$ make
[ 40%] Built target InferLLM
[ 43%] Linking CXX shared library libInferLLMShared.dylib
Apple clang version 13.0.0 (clang-1300.0.29.30)
Target: x86_64-apple-darwin20.6.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin
"/Library/Developer/CommandLineTools/usr/bin/ld" -demangle -lto_library /Library/Developer/CommandLineTools/usr/lib/libLTO.dylib -dynamic -dylib -arch x86_64 -dylib_install_name @rpath/libInferLLMShared.dylib -platform_version macos 11.7.0 12.1 -syslibroot /Library/Developer/CommandLineTools/SDKs/MacOSX12.1.sdk -o libInferLLMShared.dylib -headerpad_max_install_names CMakeFiles/InferLLMShared.dir/src/core/device.cpp.o CMakeFiles/InferLLMShared.dir/src/core/graph.cpp.o CMakeFiles/InferLLMShared.dir/src/core/kvstotage.cpp.o CMakeFiles/InferLLMShared.dir/src/core/model.cpp.o CMakeFiles/InferLLMShared.dir/src/core/model_imp.cpp.o CMakeFiles/InferLLMShared.dir/src/core/op.cpp.o CMakeFiles/InferLLMShared.dir/src/core/tensor.cpp.o CMakeFiles/InferLLMShared.dir/src/core/thread_pool.cpp.o CMakeFiles/InferLLMShared.dir/src/file.cpp.o CMakeFiles/InferLLMShared.dir/src/graph/chatGLM.cpp.o CMakeFiles/InferLLMShared.dir/src/graph/chatGLM2.cpp.o CMakeFiles/InferLLMShared.dir/src/graph/ggml_llama.cpp.o CMakeFiles/InferLLMShared.dir/src/graph/graph_imp.cpp.o CMakeFiles/InferLLMShared.dir/src/graph/llama_like.cpp.o CMakeFiles/InferLLMShared.dir/src/kern/naive/naive.cpp.o CMakeFiles/InferLLMShared.dir/src/utils.cpp.o CMakeFiles/InferLLMShared.dir/src/kern/optimized/x86/kernel.cpp.o -lc++ -lSystem /Library/Developer/CommandLineTools/usr/lib/clang/13.0.0/lib/darwin/libclang_rt.osx.a
Undefined symbols for architecture x86_64:
"_mm_max_ps(float vector[4], float vector[4])", referenced from:
inferllm::opt::reduce_max(int, float const*) (.avx2) in kernel.cpp.o
"_mm256_add_ps(float vector[8], float vector[8])", referenced from:
inferllm::opt::elemwise_vector_add(int, float const*, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::reduce_square_sum(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
inferllm::opt::compute_src_offset_embd_matmul(float const*, int, float const*, int, float*, int, int, int) (.avx2) in kernel.cpp.o
inferllm::opt::comput_matmul_with_dst_uncontinue(float*, int, float const*, int, float const*, int, int, int) (.avx2) in kernel.cpp.o
"_mm256_div_ps(float vector[8], float vector[8])", referenced from:
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
"_mm256_max_ps(float vector[8], float vector[8])", referenced from:
inferllm::opt::reduce_max(int, float const*) (.avx2) in kernel.cpp.o
"_mm256_sub_ps(float vector[8], float vector[8])", referenced from:
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
"_mm256_hadd_ps(float vector[8], float vector[8])", referenced from:
inferllm::opt::reduce_square_sum(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
inferllm::opt::compute_src_offset_embd_matmul(float const*, int, float const*, int, float*, int, int, int) (.avx2) in kernel.cpp.o
"_mm256_set1_ps(float)", referenced from:
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vec_scale(int, float const*, float, float*) (.avx2) in kernel.cpp.o
inferllm::opt::reduce_max(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
inferllm::opt::comput_matmul_with_dst_uncontinue(float*, int, float const*, int, float const*, int, int, int) (.avx2) in kernel.cpp.o
"_mm256_loadu_ps(float const*)", referenced from:
inferllm::opt::elemwise_vector_add(int, float const*, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_mul(int, float const*, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::reduce_square_sum(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vec_scale(int, float const*, float, float*) (.avx2) in kernel.cpp.o
inferllm::opt::reduce_max(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
...
"_mm256_blendv_ps(float vector[8], float vector[8], float vector[8])", referenced from:
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
"_mm256_storeu_ps(float*, float vector[8])", referenced from:
inferllm::opt::dequantize_row_q4_0(void const*, float*, int) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_add(int, float const*, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_mul(int, float const*, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vec_scale(int, float const*, float, float*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
inferllm::opt::comput_matmul_with_dst_uncontinue(float*, int, float const*, int, float const*, int, int, int) (.avx2) in kernel.cpp.o
...
"_mm256_cvtepi8_epi16(long long vector[2])", referenced from:
inferllm::opt::dequantize_row_q4_0(void const*, float*, int) (.avx2) in kernel.cpp.o
"_mm256_cvtepi16_epi32(long long vector[2])", referenced from:
inferllm::opt::dequantize_row_q4_0(void const*, float*, int) (.avx2) in kernel.cpp.o
"exp(float)", referenced from:
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::elemwise_vector_silu(int, float const*, float*) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) in kernel.cpp.o
"isnan(float)", referenced from:
inferllm::opt::dequantize_row_q4_0(void const*, float*, int) in kernel.cpp.o
"inferllm::opt::exp256_ps(float vector[8])", referenced from:
inferllm::opt::elemwise_vector_silu(int, float const*, float*) (.avx2) in kernel.cpp.o
inferllm::opt::select_sub_max_and_reduce_sum(int, float const*, float*, float) (.avx2) in kernel.cpp.o
"float const& std::__1::max(float const&, float const&)", referenced from:
inferllm::opt::reduce_max(int, float const*) (.avx2) in kernel.cpp.o
inferllm::opt::reduce_max(int, float const*) in kernel.cpp.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [libInferLLMShared.dylib] Error 1
make[1]: *** [CMakeFiles/InferLLMShared.dir/all] Error 2
make: *** [all] Error 2

[junlong]

update mac os to 14.1 and it works