happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

rm redundant includes

This commit is contained in:
Xuan Son Nguyen 2025-11-18 15:01:17 +01:00
parent 3b7946034c
commit ca993bad51
6 changed files with 59 additions and 62 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
tools/server/server-common.h
View File

@ -42,11 +42,6 @@ using json = nlohmann::ordered_json;
#define SRV_ERR(fmt, ...) LOG_ERR("srv %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define SRV_DBG(fmt, ...) LOG_DBG("srv %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_INF(fmt, ...) LOG_INF("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_WRN(fmt, ...) LOG_WRN("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_ERR(fmt, ...) LOG_ERR("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_DBG(fmt, ...) LOG_DBG("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
using raw_buffer = std::vector<uint8_t>;
template <typename T>
@ -71,41 +66,6 @@ enum stop_type {
STOP_TYPE_LIMIT,
};
// state diagram: https://github.com/ggml-org/llama.cpp/pull/9283
enum slot_state {
SLOT_STATE_IDLE,
SLOT_STATE_STARTED, // TODO: this state is only used for setting up the initial prompt processing; maybe merge it with launch_slot_with_task in the future
SLOT_STATE_PROCESSING_PROMPT,
SLOT_STATE_DONE_PROMPT,
SLOT_STATE_GENERATING,
};
enum server_state {
SERVER_STATE_LOADING_MODEL, // Server is starting up, model not fully loaded yet
SERVER_STATE_READY, // Server is ready and model is loaded
};
enum server_task_type {
SERVER_TASK_TYPE_COMPLETION,
SERVER_TASK_TYPE_EMBEDDING,
SERVER_TASK_TYPE_RERANK,
SERVER_TASK_TYPE_INFILL,
SERVER_TASK_TYPE_CANCEL,
SERVER_TASK_TYPE_NEXT_RESPONSE,
SERVER_TASK_TYPE_METRICS,
SERVER_TASK_TYPE_SLOT_SAVE,
SERVER_TASK_TYPE_SLOT_RESTORE,
SERVER_TASK_TYPE_SLOT_ERASE,
SERVER_TASK_TYPE_SET_LORA,
};
enum oaicompat_type {
OAICOMPAT_TYPE_NONE,
OAICOMPAT_TYPE_CHAT,
OAICOMPAT_TYPE_COMPLETION,
OAICOMPAT_TYPE_EMBEDDING,
};
// https://community.openai.com/t/openai-chat-list-of-error-codes-and-types/357791/11
enum error_type {
ERROR_TYPE_INVALID_REQUEST,

33
tools/server/server-queue.cpp
View File

@ -1,12 +1,17 @@
#include "server-common.h"
#include "server-task.h"
#include "server-queue.h"
#include <chrono>
#include <condition_variable>
#include <deque>
#include <mutex>
#include <unordered_set>
#include "log.h"
#define QUE_INF(fmt, ...) LOG_INF("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_WRN(fmt, ...) LOG_WRN("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_ERR(fmt, ...) LOG_ERR("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define QUE_DBG(fmt, ...) LOG_DBG("que %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define RES_INF(fmt, ...) LOG_INF("res %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define RES_WRN(fmt, ...) LOG_WRN("res %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define RES_ERR(fmt, ...) LOG_ERR("res %12.*s: " fmt, 12, __func__, __VA_ARGS__)
#define RES_DBG(fmt, ...) LOG_DBG("res %12.*s: " fmt, 12, __func__, __VA_ARGS__)
//
// server_queue
@ -150,7 +155,7 @@ void server_queue::cleanup_pending_task(int id_target) {
//
void server_response::add_waiting_task_id(int id_task) {
SRV_DBG("add task %d to waiting list. current waiting = %d (before add)\n", id_task, (int) waiting_task_ids.size());
RES_DBG("add task %d to waiting list. current waiting = %d (before add)\n", id_task, (int) waiting_task_ids.size());
std::unique_lock<std::mutex> lock(mutex_results);
waiting_task_ids.insert(id_task);
@ -160,13 +165,13 @@ void server_response::add_waiting_tasks(const std::vector<server_task> & tasks)
std::unique_lock<std::mutex> lock(mutex_results);
for (const auto & task : tasks) {
SRV_DBG("add task %d to waiting list. current waiting = %d (before add)\n", task.id, (int) waiting_task_ids.size());
RES_DBG("add task %d to waiting list. current waiting = %d (before add)\n", task.id, (int) waiting_task_ids.size());
waiting_task_ids.insert(task.id);
}
}
void server_response::remove_waiting_task_id(int id_task) {
SRV_DBG("remove task %d from waiting list. current waiting = %d (before remove)\n", id_task, (int) waiting_task_ids.size());
RES_DBG("remove task %d from waiting list. current waiting = %d (before remove)\n", id_task, (int) waiting_task_ids.size());
std::unique_lock<std::mutex> lock(mutex_results);
waiting_task_ids.erase(id_task);
@ -182,7 +187,7 @@ void server_response::remove_waiting_task_ids(const std::unordered_set<int> & id
std::unique_lock<std::mutex> lock(mutex_results);
for (const auto & id_task : id_tasks) {
SRV_DBG("remove task %d from waiting list. current waiting = %d (before remove)\n", id_task, (int) waiting_task_ids.size());
RES_DBG("remove task %d from waiting list. current waiting = %d (before remove)\n", id_task, (int) waiting_task_ids.size());
waiting_task_ids.erase(id_task);
}
}
@ -192,7 +197,7 @@ server_task_result_ptr server_response::recv(const std::unordered_set<int> & id_
std::unique_lock<std::mutex> lock(mutex_results);
condition_results.wait(lock, [&]{
if (!running) {
SRV_DBG("%s : queue result stop\n", __func__);
RES_DBG("%s : queue result stop\n", __func__);
std::terminate(); // we cannot return here since the caller is HTTP code
}
return !queue_results.empty();
@ -224,7 +229,7 @@ server_task_result_ptr server_response::recv_with_timeout(const std::unordered_s
std::cv_status cr_res = condition_results.wait_for(lock, std::chrono::seconds(timeout));
if (!running) {
SRV_DBG("%s : queue result stop\n", __func__);
RES_DBG("%s : queue result stop\n", __func__);
std::terminate(); // we cannot return here since the caller is HTTP code
}
if (cr_res == std::cv_status::timeout) {
@ -241,12 +246,12 @@ server_task_result_ptr server_response::recv(int id_task) {
}
void server_response::send(server_task_result_ptr && result) {
SRV_DBG("sending result for task id = %d\n", result->id);
RES_DBG("sending result for task id = %d\n", result->id);
std::unique_lock<std::mutex> lock(mutex_results);
for (const auto & id_task : waiting_task_ids) {
if (result->id == id_task) {
SRV_DBG("task id = %d pushed to result queue\n", result->id);
RES_DBG("task id = %d pushed to result queue\n", result->id);
queue_results.emplace_back(std::move(result));
condition_results.notify_all();

1
tools/server/server-queue.h
View File

@ -1,6 +1,5 @@
#pragma once
#include "server-common.h"
#include "server-task.h"
#include <chrono>

10
tools/server/server-task.cpp
View File

@ -1,16 +1,12 @@
#include "server-common.h"
#include "server-task.h"
#include "common.h"
#include "llama.h"
#include "chat.h"
#include "sampling.h"
#include "json-schema-to-grammar.h"
#include <string>
#include <unordered_set>
#include <list>
#include "server-common.h"
#include "server-task.h"
using json = nlohmann::ordered_json;
//

23
tools/server/server-task.h
View File

@ -10,10 +10,33 @@
#include <unordered_set>
#include <list>
// TODO: prevent including the whole server-common.h as we only use server_tokens
#include "server-common.h"
using json = nlohmann::ordered_json;
enum server_task_type {
SERVER_TASK_TYPE_COMPLETION,
SERVER_TASK_TYPE_EMBEDDING,
SERVER_TASK_TYPE_RERANK,
SERVER_TASK_TYPE_INFILL,
SERVER_TASK_TYPE_CANCEL,
SERVER_TASK_TYPE_NEXT_RESPONSE,
SERVER_TASK_TYPE_METRICS,
SERVER_TASK_TYPE_SLOT_SAVE,
SERVER_TASK_TYPE_SLOT_RESTORE,
SERVER_TASK_TYPE_SLOT_ERASE,
SERVER_TASK_TYPE_SET_LORA,
};
// TODO: change this to more generic "response_format" to replace the "format_response_*" in server-common
enum oaicompat_type {
OAICOMPAT_TYPE_NONE,
OAICOMPAT_TYPE_CHAT,
OAICOMPAT_TYPE_COMPLETION,
OAICOMPAT_TYPE_EMBEDDING,
};
struct task_params {
bool stream = true;
bool include_usage = false;

14
tools/server/server.cpp
View File

@ -23,6 +23,20 @@ using json = nlohmann::ordered_json;
constexpr int HTTP_POLLING_SECONDS = 1;
// state diagram: https://github.com/ggml-org/llama.cpp/pull/9283
enum slot_state {
SLOT_STATE_IDLE,
SLOT_STATE_STARTED, // TODO: this state is only used for setting up the initial prompt processing; maybe merge it with launch_slot_with_task in the future
SLOT_STATE_PROCESSING_PROMPT,
SLOT_STATE_DONE_PROMPT,
SLOT_STATE_GENERATING,
};
enum server_state {
SERVER_STATE_LOADING_MODEL, // Server is starting up, model not fully loaded yet
SERVER_STATE_READY, // Server is ready and model is loaded
};
static bool server_task_type_need_embd(server_task_type task_type) {
switch (task_type) {
case SERVER_TASK_TYPE_EMBEDDING: