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

remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (#10797)

Browse source 
other windows build fixes
This commit is contained in:
Diego Devesa 2024-12-12 19:02:49 +01:00 • committed by GitHub
parent 4064c0e3b6
commit cb13ef85a4
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
14 changed files with 76 additions and 65 deletions
Download patch file Download diff file Expand all files Collapse all files

12
examples/CMakeLists.txt
View file

@ -20,7 +20,12 @@ else()
add_subdirectory(batched)
add_subdirectory(embedding)
add_subdirectory(eval-callback)
add_subdirectory(gbnf-validator)
if (NOT WIN32)
# disabled on Windows because it uses internal functions not exported with LLAMA_API
add_subdirectory(gbnf-validator)
endif()
add_subdirectory(gguf-hash)
add_subdirectory(gguf-split)
add_subdirectory(gguf)
@ -51,7 +56,10 @@ else()
add_subdirectory(convert-llama2c-to-ggml)
add_subdirectory(cvector-generator)
add_subdirectory(export-lora)
add_subdirectory(quantize-stats)
if (NOT WIN32)
# disabled on Windows because it uses internal functions not exported with LLAMA_API
add_subdirectory(quantize-stats)
endif()
add_subdirectory(llava)
if (GGML_RPC)
add_subdirectory(rpc)

4
examples/gguf-split/gguf-split.cpp
View file

@ -287,7 +287,7 @@ struct split_strategy {
}
void print_info() {
printf("n_split: %ld\n", ctx_outs.size());
printf("n_split: %zu\n", ctx_outs.size());
int i_split = 0;
for (auto & ctx_out : ctx_outs) {
// re-calculate the real gguf size for each split (= metadata size + total size of all tensors)
@ -297,7 +297,7 @@ struct split_strategy {
total_size += ggml_nbytes(t);
}
total_size = total_size / 1000 / 1000; // convert to megabytes
printf("split %05d: n_tensors = %d, total_size = %ldM\n", i_split + 1, gguf_get_n_tensors(ctx_out), total_size);
printf("split %05d: n_tensors = %d, total_size = %zuM\n", i_split + 1, gguf_get_n_tensors(ctx_out), total_size);
i_split++;
}
}

10
examples/llama-bench/llama-bench.cpp
View file

@ -1521,7 +1521,7 @@ int main(int argc, char ** argv) {
for (const auto & inst : params_instances) {
params_idx++;
if (params.progress) {
fprintf(stderr, "llama-bench: benchmark %d/%ld: starting\n", params_idx, params_count);
fprintf(stderr, "llama-bench: benchmark %d/%zu: starting\n", params_idx, params_count);
}
// keep the same model between tests when possible
if (!lmodel || !prev_inst || !inst.equal_mparams(*prev_inst)) {
@ -1573,14 +1573,14 @@ int main(int argc, char ** argv) {
// warmup run
if (t.n_prompt > 0) {
if (params.progress) {
fprintf(stderr, "llama-bench: benchmark %d/%ld: warmup prompt run\n", params_idx, params_count);
fprintf(stderr, "llama-bench: benchmark %d/%zu: warmup prompt run\n", params_idx, params_count);
}
//test_prompt(ctx, std::min(t.n_batch, std::min(t.n_prompt, 32)), 0, t.n_batch, t.n_threads);
test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
}
if (t.n_gen > 0) {
if (params.progress) {
fprintf(stderr, "llama-bench: benchmark %d/%ld: warmup generation run\n", params_idx, params_count);
fprintf(stderr, "llama-bench: benchmark %d/%zu: warmup generation run\n", params_idx, params_count);
}
test_gen(ctx, 1, t.n_threads);
}
@ -1592,14 +1592,14 @@ int main(int argc, char ** argv) {
if (t.n_prompt > 0) {
if (params.progress) {
fprintf(stderr, "llama-bench: benchmark %d/%ld: prompt run %d/%d\n", params_idx, params_count,
fprintf(stderr, "llama-bench: benchmark %d/%zu: prompt run %d/%d\n", params_idx, params_count,
i + 1, params.reps);
}
test_prompt(ctx, t.n_prompt, t.n_batch, t.n_threads);
}
if (t.n_gen > 0) {
if (params.progress) {
fprintf(stderr, "llama-bench: benchmark %d/%ld: generation run %d/%d\n", params_idx, params_count,
fprintf(stderr, "llama-bench: benchmark %d/%zu: generation run %d/%d\n", params_idx, params_count,
i + 1, params.reps);
}
test_gen(ctx, t.n_gen, t.n_threads);

2
examples/retrieval/retrieval.cpp
View file

@ -143,7 +143,7 @@ int main(int argc, char ** argv) {
std::vector<chunk> file_chunk = chunk_file(context_file, params.chunk_size, params.chunk_separator);
chunks.insert(chunks.end(), file_chunk.begin(), file_chunk.end());
}
LOG_INF("Number of chunks: %ld\n", chunks.size());
LOG_INF("Number of chunks: %zu\n", chunks.size());
llama_backend_init();
llama_numa_init(params.numa);

2
examples/tokenize/tokenize.cpp
View file

@ -394,7 +394,7 @@ int main(int raw_argc, char ** raw_argv) {
}
if (show_token_count) {
printf("Total number of tokens: %ld\n", tokens.size());
printf("Total number of tokens: %zu\n", tokens.size());
}
// silence valgrind
llama_free(ctx);