CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-11 10:36:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

context : remove batch_manager

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-02-14 16:10:55 +02:00
parent 131743ff4f
commit d5e8e1a2ba
4 changed files with 242 additions and 291 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/llama-kv-cache.h
View File

@@ -150,7 +150,9 @@ struct llama_kv_slot_restorer {
bool do_restore = false;
explicit llama_kv_slot_restorer(const struct llama_kv_cache & cache) {
llama_kv_cache & cache;
explicit llama_kv_slot_restorer(llama_kv_cache & cache) : cache(cache) {
old_state.head = cache.head;
old_state.n = cache.n;
}
@@ -167,7 +169,7 @@ struct llama_kv_slot_restorer {
// must be explicitly called to restore the kv_cache state
// and rollback changes from all llama_kv_cache_find_slot calls
void restore(struct llama_kv_cache & cache) {
void restore() {
if (do_restore) {
cache.head = old_state.head;
cache.n = old_state.n;