tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-01 13:05:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

memory : rename interface to llama_memory_context_i (#14296)

Browse Source 
* memory : rename interface to llama_memory_context_i

ggml-ci

* cont : fix comments

* cont : use "mctx" for referencing a memory context

ggml-ci
This commit is contained in:
Georgi Gerganov
2025-06-21 08:03:46 +03:00
committed by GitHub
parent b23fa0b3f4
commit 692e3cdd0a
14 changed files with 339 additions and 341 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
src/llama-memory.h
View File

@ -3,7 +3,6 @@
#include "llama.h"
#include <memory>
#include <vector>
struct llama_ubatch;
@ -28,23 +27,21 @@ enum llama_memory_status {
LLAMA_MEMORY_STATUS_FAILED_COMPUTE,
};
// helper function for combining the status of two memory states
// helper function for combining the status of two memory contexts
// useful for implementing hybrid memory types (e.g. iSWA)
llama_memory_status llama_memory_status_combine(llama_memory_status s0, llama_memory_status s1);
// the interface for managing the memory state during batch processing
// the interface for managing the memory context during batch processing
// this interface is implemented per memory type. see:
// - llama_kv_cache_unified_state
// - llama_kv_cache_unified_iswa_state
// - llama_kv_cache_unified_context
// - llama_kv_cache_unified_iswa_context
// ...
//
// the only method that can mutate the memory and the memory state is llama_memory_i::apply()
//
// TODO: rename to llama_memory_context_i ?
struct llama_memory_state_i {
virtual ~llama_memory_state_i() = default;
// the only method that should mutate the memory and the memory context is llama_memory_i::apply()
struct llama_memory_context_i {
virtual ~llama_memory_context_i() = default;
// consume the current ubatch from the state and proceed to the next one
// consume the current ubatch from the context and proceed to the next one
// return false if we are done
virtual bool next() = 0;
@ -55,11 +52,11 @@ struct llama_memory_state_i {
// get the current ubatch
virtual const llama_ubatch & get_ubatch() const = 0;
// get the status of the memory state - used for error handling and checking if any updates would be applied
// get the status of the memory context - used for error handling and checking if any updates would be applied
virtual llama_memory_status get_status() const = 0;
};
using llama_memory_state_ptr = std::unique_ptr<llama_memory_state_i>;
using llama_memory_context_ptr = std::unique_ptr<llama_memory_context_i>;
// general concept of LLM memory
// the KV cache is a type of LLM memory, but there can be other types
@ -67,19 +64,19 @@ struct llama_memory_i {
virtual ~llama_memory_i() = default;
// split the input batch into a set of ubatches and verify that they can fit into the cache
// return a state object containing the ubatches and KV cache state required to process them
// check the llama_memory_state_i::get_status() for the result
virtual llama_memory_state_ptr init_batch(
// return a context object containing the ubatches and memory state required to process them
// check the llama_memory_context_i::get_status() for the result
virtual llama_memory_context_ptr init_batch(
llama_batch_allocr & balloc,
uint32_t n_ubatch,
bool embd_all) = 0;
// simulate full cache, used for allocating worst-case compute buffers
virtual llama_memory_state_ptr init_full() = 0;
virtual llama_memory_context_ptr init_full() = 0;
// prepare for any pending memory updates, such as shifts, defrags, etc.
// status == LLAMA_MEMORY_STATUS_NO_UPDATE if there is nothing to update
virtual llama_memory_state_ptr init_update(llama_context * lctx, bool optimize) = 0;
virtual llama_memory_context_ptr init_update(llama_context * lctx, bool optimize) = 0;
// getters
virtual bool get_can_shift() const = 0;