[User] Insert summary of your issue or enhancement..

[adityachallapally]

Prerequisites

Please answer the following questions for yourself before submitting an issue.

• [Y ] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
• [ Y] I carefully followed the README.md.
• [ Y] I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• [Y ] I reviewed the Discussions, and have a new bug or useful enhancement to share.

Expected Behavior

I expect it to run but it keeps on not running after I followed the instructions.

Current Behavior

Keeps on providing errors to ggml functions:

I llama.cpp build info:
I UNAME_S: Linux
I UNAME_P: unknown
I UNAME_M: x86_64
I CFLAGS: -I. -O3 -std=c11 -fPIC -DNDEBUG -Wall -Wextra -Wpedantic -Wcast-qual -Wdouble-promotion -Wshadow -Wstrict-prototypes -Wpointer-arith -pthread -march=native -mtune=native
I CXXFLAGS: -I. -I./examples -O3 -std=c++11 -fPIC -DNDEBUG -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wno-multichar -pthread -march=native -mtune=native
I LDFLAGS:
I CC: cc (GCC) 13.1.1 20230429
I CXX: g++ (GCC) 13.1.1 20230429

g++ -I. -I./examples -O3 -std=c++11 -fPIC -DNDEBUG -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wno-multichar -pthread -march=native -mtune=native examples/main/main.cpp ggml.o llama.o common.o -o main
/usr/bin/ld: llama.o: in function llama_free': llama.cpp:(.text+0x28e1): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0x2924): undefined reference to ggml_cuda_host_free' /usr/bin/ld: llama.cpp:(.text+0x2aac): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0x2adb): undefined reference to ggml_cuda_host_free' /usr/bin/ld: llama.o: in function llama_model_load_internal(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, llama_context&, int, int, ggml_type, bool, bool, bool, void ()(float, void), void*)':
llama.cpp:(.text+0x9020): undefined reference to ggml_cuda_host_free' /usr/bin/ld: llama.cpp:(.text+0x9033): undefined reference to ggml_cuda_host_malloc'
/usr/bin/ld: llama.cpp:(.text+0xa6ff): undefined reference to ggml_cuda_transform_tensor' /usr/bin/ld: llama.cpp:(.text+0xa714): undefined reference to ggml_cuda_transform_tensor'
/usr/bin/ld: llama.cpp:(.text+0xa729): undefined reference to ggml_cuda_transform_tensor' /usr/bin/ld: llama.cpp:(.text+0xa73e): undefined reference to ggml_cuda_transform_tensor'
/usr/bin/ld: llama.cpp:(.text+0xa756): undefined reference to ggml_cuda_transform_tensor' /usr/bin/ld: llama.o:llama.cpp:(.text+0xa76b): more undefined references to ggml_cuda_transform_tensor' follow
/usr/bin/ld: llama.o: in function llama_init_from_file': llama.cpp:(.text+0xb968): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0xb97b): undefined reference to ggml_cuda_host_malloc' /usr/bin/ld: llama.cpp:(.text+0xbb96): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0xbba9): undefined reference to ggml_cuda_host_malloc' /usr/bin/ld: llama.cpp:(.text+0xbc4f): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0xbc62): undefined reference to ggml_cuda_host_malloc' /usr/bin/ld: llama.cpp:(.text+0xbd0b): undefined reference to ggml_cuda_host_free'
/usr/bin/ld: llama.cpp:(.text+0xbd1e): undefined reference to ggml_cuda_host_malloc' /usr/bin/ld: ggml.o: in function ggml_compute_forward_mul_mat_q_f32':
ggml.c:(.text+0x2817): undefined reference to ggml_cuda_can_mul_mat' /usr/bin/ld: ggml.o: in function ggml_compute_forward_mul_mat_f16_f32':
ggml.c:(.text+0x602d): undefined reference to ggml_cuda_can_mul_mat' /usr/bin/ld: ggml.o: in function ggml_init':
ggml.c:(.text+0x12068): undefined reference to ggml_init_cublas' /usr/bin/ld: ggml.o: in function ggml_compute_forward':
ggml.c:(.text+0x1452b): undefined reference to ggml_cuda_can_mul_mat' /usr/bin/ld: ggml.o: in function ggml_graph_compute':
ggml.c:(.text+0x1f317): undefined reference to ggml_cuda_can_mul_mat' /usr/bin/ld: ggml.c:(.text+0x1f947): undefined reference to ggml_cuda_mul_mat_get_wsize'
/usr/bin/ld: ggml.o: in function ggml_compute_forward_mul_mat_q_f32': ggml.c:(.text+0x2ba5): undefined reference to ggml_cuda_mul_mat'
/usr/bin/ld: ggml.o: in function ggml_compute_forward_mul_mat_f16_f32': ggml.c:(.text+0x6469): undefined reference to ggml_cuda_mul_mat'
/usr/bin/ld: ggml.o: in function ggml_compute_forward': ggml.c:(.text+0x16564): undefined reference to ggml_cuda_mul_mat'

Environment and Context

Linux on 8G GPU

[SlyEcho]

That means you compiled ggml.o with LLAMA_CUBLAS=1 and main without it.

Please run make clean before building different configurations, or use CMake.

[BlessMario]

make clean was the remedy...thanks

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.