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

Pass pointer to params in llama_init_from_file

Browse source 
Especially with golang bindings, calling by value has the side-effect of
values not being copied correctly. This has been observed with the
bindings in https://github.com/go-skynet/go-llama.cpp/pull/105.
This commit is contained in:
mudler 2023-06-16 23:43:36 +02:00
parent d411968e99
commit 200892a3a5
7 changed files with 9 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/common.cpp
View file

@ -555,7 +555,7 @@ struct llama_context * llama_init_from_gpt_params(const gpt_params & params) {
lparams.logits_all = params.perplexity;
lparams.embedding = params.embedding;
llama_context * lctx = llama_init_from_file(params.model.c_str(), lparams);
llama_context * lctx = llama_init_from_file(params.model.c_str(), &lparams);
if (lctx == NULL) {
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());

2
examples/quantize-stats/quantize-stats.cpp
View file

@ -330,7 +330,7 @@ int main(int argc, char ** argv) {
lparams.f16_kv = false;
lparams.use_mlock = false;
ctx = llama_init_from_file(params.model.c_str(), lparams);
ctx = llama_init_from_file(params.model.c_str(), &lparams);
if (ctx == NULL) {
fprintf(stderr, "%s: error: failed to load model '%s'\n", __func__, params.model.c_str());

4
examples/save-load-state/save-load-state.cpp
View file

@ -35,7 +35,7 @@ int main(int argc, char ** argv) {
auto last_n_tokens_data = std::vector<llama_token>(params.repeat_last_n, 0);
// init
auto ctx = llama_init_from_file(params.model.c_str(), lparams);
auto ctx = llama_init_from_file(params.model.c_str(), &lparams);
auto tokens = std::vector<llama_token>(params.n_ctx);
auto n_prompt_tokens = llama_tokenize(ctx, params.prompt.c_str(), tokens.data(), int(tokens.size()), true);
@ -95,7 +95,7 @@ int main(int argc, char ** argv) {
llama_free(ctx);
// load new model
auto ctx2 = llama_init_from_file(params.model.c_str(), lparams);
auto ctx2 = llama_init_from_file(params.model.c_str(), &lparams);
// Load state (rng, logits, embedding and kv_cache) from file
{

2
examples/train-text-from-scratch/train-text-from-scratch.cpp
View file

@ -3054,7 +3054,7 @@ int main(int argc, char ** argv) {
struct llama_context_params llama_params = llama_context_default_params();
llama_params.vocab_only = true;
struct llama_context * lctx = llama_init_from_file(params.fn_vocab_model, llama_params);
struct llama_context * lctx = llama_init_from_file(params.fn_vocab_model, &llama_params);
struct llama_vocab vocab;
{

3
llama.cpp
View file

@ -2618,8 +2618,9 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
struct llama_context * llama_init_from_file(
const char * path_model,
struct llama_context_params params) {
const struct llama_context_params * params_ptr) {
ggml_time_init();
struct llama_context_params params = *params_ptr;
llama_context * ctx = new llama_context;

2
llama.h
View file

@ -142,7 +142,7 @@ extern "C" {
// Return NULL on failure
LLAMA_API struct llama_context * llama_init_from_file(
const char * path_model,
struct llama_context_params params);
const struct llama_context_params * params);
// Frees all allocated memory
LLAMA_API void llama_free(struct llama_context * ctx);

2
tests/test-tokenizer-0.cpp
View file

@ -36,7 +36,7 @@ int main(int argc, char **argv) {
lparams.vocab_only = true;
ctx = llama_init_from_file(fname.c_str(), lparams);
ctx = llama_init_from_file(fname.c_str(), &lparams);
if (ctx == NULL) {
fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());