CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-03 09:22:01 +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

36
src/llama-kv-cache-unified-iswa.h
View File

@@ -31,14 +31,14 @@ public:
// llama_memory_i
//
llama_memory_state_ptr init_batch(
llama_memory_context_ptr init_batch(
llama_batch_allocr & balloc,
uint32_t n_ubatch,
bool embd_all) override;
llama_memory_state_ptr init_full() override;
llama_memory_context_ptr init_full() override;
llama_memory_state_ptr init_update(llama_context * lctx, bool optimize) override;
llama_memory_context_ptr init_update(llama_context * lctx, bool optimize) override;
bool get_can_shift() const override;
@@ -72,32 +72,32 @@ private:
std::unique_ptr<llama_kv_cache_unified> kv_swa;
};
class llama_kv_cache_unified_iswa_state : public llama_memory_state_i {
class llama_kv_cache_unified_iswa_context : public llama_memory_context_i {
public:
// used for errors
llama_kv_cache_unified_iswa_state(llama_memory_status status);
llama_kv_cache_unified_iswa_context(llama_memory_status status);
// used to create a full-cache state
llama_kv_cache_unified_iswa_state(
// used to create a full-cache context
llama_kv_cache_unified_iswa_context(
llama_kv_cache_unified_iswa * kv);
// used to create an update state
llama_kv_cache_unified_iswa_state(
// used to create an update context
llama_kv_cache_unified_iswa_context(
llama_kv_cache_unified_iswa * kv,
llama_context * lctx,
bool optimize);
// used to create a state from a batch
llama_kv_cache_unified_iswa_state(
// used to create a batch processing context from a batch
llama_kv_cache_unified_iswa_context(
llama_kv_cache_unified_iswa * kv,
std::vector<uint32_t> heads_base,
std::vector<uint32_t> heads_swa,
std::vector<llama_ubatch> ubatches);
virtual ~llama_kv_cache_unified_iswa_state();
virtual ~llama_kv_cache_unified_iswa_context();
//
// llama_memory_state_i
// llama_memory_context_i
//
bool next() override;
@@ -107,11 +107,11 @@ public:
const llama_ubatch & get_ubatch() const override;
//
// llama_kv_cache_unified_iswa_state specific API
// llama_kv_cache_unified_iswa_context specific API
//
const llama_kv_cache_unified_state * get_base() const;
const llama_kv_cache_unified_state * get_swa() const;
const llama_kv_cache_unified_context * get_base() const;
const llama_kv_cache_unified_context * get_swa() const;
private:
//llama_kv_cache_unified_iswa * kv;
@@ -121,8 +121,8 @@ private:
std::vector<llama_ubatch> ubatches;
const llama_memory_state_ptr state_base;
const llama_memory_state_ptr state_swa;
const llama_memory_context_ptr ctx_base;
const llama_memory_context_ptr ctx_swa;
const llama_memory_status status;
};