ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-02-10 08:20:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Enable and clean up compiler warnings in src (#824)

Browse Source 
* WIP: enable and clean up warnings in src

* All warnings handled

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2025-10-11 16:01:13 +03:00
committed by GitHub
parent 4daff01b39
commit 764eefd1bc
6 changed files with 76 additions and 317 deletions
Download Patch File Download Diff File Expand all files Collapse all files

108
src/llama.cpp
View File

@@ -154,26 +154,6 @@ static std::string trim(const std::string & str) {
return str.substr(start, end - start);
}
static bool is_float_close(float a, float b, float abs_tol) {
// Check for non-negative tolerance
if (abs_tol < 0.0) {
throw std::invalid_argument("Tolerance must be non-negative");
}
// Exact equality check
if (a == b) {
return true;
}
// Check for infinities
if (std::isinf(a) || std::isinf(b)) {
return false;
}
// Regular comparison using the provided absolute tolerance
return std::fabs(b - a) <= abs_tol;
}
enum llm_chat_template {
LLM_CHAT_TEMPLATE_CHATML,
LLM_CHAT_TEMPLATE_LLAMA_2,
@@ -320,23 +300,6 @@ std::string gguf_kv_to_str(const gguf_context * ctx_gguf, int i) {
// llama helpers
//
// NOTE: avoid ever using this except for building the token_to_piece caches
static std::string llama_token_to_piece(const struct llama_model * model, llama_token token, bool special) {
std::string piece;
piece.resize(piece.capacity()); // using string internal cache
const int n_chars = llama_token_to_piece(model, token, &piece[0], piece.size(), 0, special);
if (n_chars < 0) {
piece.resize(-n_chars);
int check = llama_token_to_piece(model, token, &piece[0], piece.size(), 0, special);
GGML_ASSERT(check == -n_chars);
}
else {
piece.resize(n_chars);
}
return piece;
}
ggml_backend_buffer_type_t llama_default_buffer_type_cpu(bool host_buffer) {
ggml_backend_buffer_type_t buft = nullptr;
@@ -666,12 +629,9 @@ static bool llama_kv_cache_init(
}
if (needs_v_cache) cache.v_l.reserve(n_layer);
bool warn = true;
int n_mla = 0;
for (int i = 0; i < (int) n_layer; i++) {
const uint32_t n_embd_k_gqa = hparams.n_embd_k_gqa(i) + hparams.n_embd_k_s();
const uint32_t n_embd_v_gqa = hparams.n_embd_v_gqa(i) + hparams.n_embd_v_s();
const uint32_t n_head = hparams.n_head(i);
const uint32_t n_head_kv = hparams.n_head_kv(i);
const uint32_t n_embd_head_k= hparams.n_embd_head_k;
@@ -682,7 +642,6 @@ static bool llama_kv_cache_init(
if (cparams.mla_attn) {
// DeepSeek MLA
const uint32_t n_embd_head_qk_rope = hparams.n_rot;
const uint32_t n_embd_head_qk_nope = hparams.n_embd_head_k - hparams.n_rot;
const uint32_t kv_lora_rank = hparams.n_lora_kv;
//LLAMA_LOG_INFO("%s: layer %d: n_embd_head_qk_rope = %d, kv_lora_rank = %d\n", __func__, i, n_embd_head_qk_rope, kv_lora_rank);
if (cparams.flash_attn) {
@@ -1090,17 +1049,6 @@ static uint32_t llama_kv_cache_get_padding(const struct llama_cparams & cparams)
// load LLaMA models
//
static const char * llama_model_vocab_type_name(enum llama_vocab_type type){
switch (type) {
case LLAMA_VOCAB_TYPE_NONE: return "no vocab";
case LLAMA_VOCAB_TYPE_SPM: return "SPM";
case LLAMA_VOCAB_TYPE_BPE: return "BPE";
case LLAMA_VOCAB_TYPE_WPM: return "WPM";
case LLAMA_VOCAB_TYPE_UGM: return "UGM";
default: return "unknown";
}
}
void llm_load_arch(llama_model_loader & ml, llama_model & model) {
model.arch = ml.get_arch();
if (model.arch == LLM_ARCH_UNKNOWN) {
@@ -1277,7 +1225,6 @@ static void llm_prepare_mla(llama_model & model, int mla) {
// is no MLA + FA for prompt processing, and MLA + FA for token generation, it would be useful
// to change the MLA setting on the fly, depending on context. In that case, having prepared
// the MLA tensors here is the right ting to do^TM.
const uint32_t n_embd_head_qk_rope = hparams.n_rot;
const uint32_t n_embd_head_qk_nope = hparams.n_embd_head_k - hparams.n_rot;
const uint32_t kv_lora_rank = hparams.n_lora_kv;
const int32_t n_embd_head_v = hparams.n_embd_head_v;
@@ -1288,9 +1235,8 @@ static void llm_prepare_mla(llama_model & model, int mla) {
// Somehow the number of heads is being defined as being per layer. Not sure why this is the
// case, but for now we do not support strange models that have different numbers of heads
// in different model layers.
if (hparams.n_head(il) != n_head) throw std::runtime_error("Unsupported configuration");
if ((int)hparams.n_head(il) != n_head) throw std::runtime_error("Unsupported configuration");
}
auto total_size_wkb = 0;
size_t max_wkv_size = 0;
size_t max_wk_size = 0;
for (auto& l : model.layers) {
@@ -1414,10 +1360,6 @@ static void llm_prepare_mla(llama_model & model, int mla) {
// preparation of wkv_b impossible. It also has the benefit that wkv_b will get automatically
// run-time repacked if the rtr option is set.
//
const uint32_t n_embd_head_qk_rope = hparams.n_rot;
const uint32_t n_embd_head_qk_nope = hparams.n_embd_head_k - hparams.n_rot;
const uint32_t kv_lora_rank = hparams.n_lora_kv;
const int32_t n_embd_head_v = hparams.n_embd_head_v;
const int32_t n_head = hparams.n_head(0);
std::vector<uint8_t> work_data;
LLAMA_LOG_INFO("============ %s: need to compute %d wkv_b tensors\n", __func__, n_to_compute);
@@ -1425,7 +1367,7 @@ static void llm_prepare_mla(llama_model & model, int mla) {
// Somehow the number of heads is being defined as being per layer. Not sure why this is the
// case, but for now we do not support strange models that have different numbers of heads
// in different model layers.
if (hparams.n_head(il) != n_head) throw std::runtime_error("Unsupported configuration");
if ((int)hparams.n_head(il) != n_head) throw std::runtime_error("Unsupported configuration");
}
size_t context_size = ggml_tensor_overhead()*16*n_layer;
@@ -1434,12 +1376,8 @@ static void llm_prepare_mla(llama_model & model, int mla) {
auto ctx = ggml_init(params);
auto graph = ggml_new_graph_custom(ctx, 8, false);
//layer.wk_b = create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_K_B, "weight", i), {n_embd_head_qk_nope, kv_lora_rank, n_head}, 0);
//layer.wv_b = create_tensor(ctx_split, tn(LLM_TENSOR_ATTN_V_B, "weight", i), {kv_lora_rank, n_embd_head_v, n_head}, 0);
std::vector<char> wk_buffer, wv_buffer;
std::vector<char> tmp_buffer;
//std::vector<uint8_t> tensor_data(2*n_embd_head_qk_nope*kv_lora_rank*n_head*sizeof(float) + max_wk_size);
for (int il = 0; il < n_layer; ++il) {
auto& l = model.layers[il];
if (l.wkv_b || !l.wk_b || !l.wv_b) continue;
@@ -3514,6 +3452,7 @@ struct llama_model_params llama_model_default_params() {
/*.check_tensors =*/ false,
/*.repack_tensors =*/ false,
/*.use_thp =*/ false,
/*.validate_quants =*/ false,
};
#ifdef GGML_USE_METAL
@@ -4298,10 +4237,6 @@ int32_t llama_n_layer(const struct llama_model * model) {
return model->hparams.n_layer;
}
int32_t llama_n_head(const struct llama_model * model) {
return model->hparams.n_head();
}
float llama_rope_freq_scale_train(const struct llama_model * model) {
return model->hparams.rope_freq_scale_train;
}
@@ -5903,11 +5838,6 @@ llama_token llama_token_eos(const struct llama_model * model) {
return model->vocab.token_eos();
}
// What is cls?
//llama_token llama_token_cls(const struct llama_model * model) {
// return llama_token_cls_impl(model->vocab);
//}
llama_token llama_token_sep(const struct llama_model * model) {
return model->vocab.token_sep();
}
@@ -5944,36 +5874,6 @@ llama_token llama_token_eot(const struct llama_model * model) {
return model->vocab.token_eot();
}
// deprecated
llama_token llama_token_fim_pre(const struct llama_model * model) {
return model->vocab.token_fim_pre();
}
// deprecated
llama_token llama_token_fim_suf(const struct llama_model * model) {
return model->vocab.token_fim_suf();
}
// deprecated
llama_token llama_token_fim_mid(const struct llama_model * model) {
return model->vocab.token_fim_mid();
}
// deprecated
llama_token llama_token_fim_pad(const struct llama_model * model) {
return model->vocab.token_fim_pad();
}
// deprecated
llama_token llama_token_fim_rep(const struct llama_model * model) {
return model->vocab.token_fim_rep();
}
// deprecated
llama_token llama_token_fim_sep(const struct llama_model * model) {
return model->vocab.token_fim_sep();
}
//
// tokenization
//
@@ -6781,7 +6681,7 @@ void llama_sample_top_n_sigma(struct llama_context * ctx, llama_token_data_array
}
void llama_sample_dry(struct llama_context* ctx, struct llama_sampler_dry* smpl, llama_token_data_array* candidates_p) {
void llama_sample_dry([[maybe_unused]] struct llama_context* ctx, struct llama_sampler_dry* smpl, llama_token_data_array* candidates_p) {
llama_sampler_dry_apply(smpl, candidates_p);
}