llama : refactor src/llama.cpp (#10902)

* llama : scatter llama.cpp into multiple modules (wip)

* llama : control-vector -> adapter

* llama : arch

* llama : mmap

ggml-ci

* ci : remove BUILD_SHARED_LIBS=OFF

ggml-ci

* llama : arch (cont)

ggml-ci

* llama : chat

ggml-ci

* llama : model

ggml-ci

* llama : hparams

ggml-ci

* llama : adapter

ggml-ci

* examples : fix

ggml-ci

* rebase

ggml-ci

* minor

* llama : kv cache

ggml-ci

* llama : impl

ggml-ci

* llama : batch

ggml-ci

* cont

ggml-ci

* llama : context

ggml-ci

* minor

* llama : context (cont)

ggml-ci

* llama : model loader

ggml-ci

* common : update lora

ggml-ci

* llama : quant

ggml-ci

* llama : quant (cont)

ggml-ci

* minor [no ci]
This commit is contained in:
Georgi Gerganov 2025-01-03 10:18:53 +02:00 committed by GitHub
parent 2f0ee84b9b
commit f66f582927
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
61 changed files with 12193 additions and 11649 deletions

67
src/llama-mmap.h Normal file
View file

@ -0,0 +1,67 @@
#pragma once
#include <memory>
#include <vector>
struct llama_file;
struct llama_mmap;
struct llama_mlock;
using llama_files = std::vector<std::unique_ptr<llama_file>>;
using llama_mmaps = std::vector<std::unique_ptr<llama_mmap>>;
using llama_mlocks = std::vector<std::unique_ptr<llama_mlock>>;
struct llama_file {
llama_file(const char * fname, const char * mode);
~llama_file();
size_t tell() const;
size_t size() const;
int fileno() const;
void seek(size_t offset, int whence) const;
void read_raw(void * ptr, size_t len) const;
uint32_t read_u32() const;
void write_raw(const void * ptr, size_t len) const;
void write_u32(uint32_t val) const;
private:
struct impl;
std::unique_ptr<impl> pimpl;
};
struct llama_mmap {
llama_mmap(const llama_mmap &) = delete;
llama_mmap(struct llama_file * file, size_t prefetch = (size_t) -1, bool numa = false);
~llama_mmap();
size_t size() const;
void * addr() const;
void unmap_fragment(size_t first, size_t last);
static const bool SUPPORTED;
private:
struct impl;
std::unique_ptr<impl> pimpl;
};
struct llama_mlock {
llama_mlock();
~llama_mlock();
void init(void * ptr);
void grow_to(size_t target_size);
static const bool SUPPORTED;
private:
struct impl;
std::unique_ptr<impl> pimpl;
};
size_t llama_path_max();