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

address feedbacks to avoid using exceptions

Browse source
This commit is contained in:
randxie 2023-06-29 07:49:31 +08:00
parent 67be3fc743
commit fcb0a77b13
1 changed files with 10 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

17
llama.cpp
View file

@ -3336,7 +3336,7 @@ size_t llama_set_state_data(struct llama_context * ctx, uint8_t * src) {
return nread;
}
static void llama_load_session_file_internal(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out) {
static bool llama_load_session_file_internal(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out) {
llama_file file(path_session, "rb");
// sanity checks
@ -3345,14 +3345,16 @@ static void llama_load_session_file_internal(struct llama_context * ctx, const c
const uint32_t version = file.read_u32();
if (magic != LLAMA_SESSION_MAGIC || version != LLAMA_SESSION_VERSION) {
throw std::runtime_error(format("%s : unknown (magic, version) for session file: %08x, %08x\n", __func__, magic, version));
fprintf(stderr, "%s : unknown (magic, version) for session file: %08x, %08x\n", __func__, magic, version);
return false;
}
llama_hparams session_hparams;
file.read_raw(&session_hparams, sizeof(llama_hparams));
if (session_hparams != ctx->model.hparams) {
throw std::runtime_error(format("%s : model hparams didn't match from session file!\n", __func__));
fprintf(stderr, "%s : model hparams didn't match from session file!\n", __func__);
return false;
}
}
@ -3361,7 +3363,8 @@ static void llama_load_session_file_internal(struct llama_context * ctx, const c
const uint32_t n_token_count = file.read_u32();
if (n_token_count > n_token_capacity) {
throw std::runtime_error(format("%s : token count in session file exceeded capacity! %u > %zu\n", __func__, n_token_count, n_token_capacity));
fprintf(stderr, "%s : token count in session file exceeded capacity! %u > %zu\n", __func__, n_token_count, n_token_capacity);
return false;
}
file.read_raw(tokens_out, sizeof(llama_token) * n_token_count);
@ -3374,7 +3377,8 @@ static void llama_load_session_file_internal(struct llama_context * ctx, const c
const size_t n_state_size_max = llama_get_state_size(ctx);
if (n_state_size_cur > n_state_size_max) {
throw std::runtime_error(format("%s : the state size in session file is too big! max %zu, got %zu\n", __func__, n_state_size_max, n_state_size_cur));
fprintf(stderr, "%s : the state size in session file is too big! max %zu, got %zu\n", __func__, n_state_size_max, n_state_size_cur);
return false;
}
std::vector<uint8_t> state_data(n_state_size_max);
@ -3386,8 +3390,7 @@ static void llama_load_session_file_internal(struct llama_context * ctx, const c
bool llama_load_session_file(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out) {
try {
llama_load_session_file_internal(ctx, path_session, tokens_out, n_token_capacity, n_token_count_out);
return true;
return llama_load_session_file_internal(ctx, path_session, tokens_out, n_token_capacity, n_token_count_out);
} catch (const std::exception & err) {
fprintf(stderr, "error loading session file: %s\n", err.what());
return false;