Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)

* Extend llama_kv_cache_seq_rm to allow matichng any sequence

* Replace llama_kv_cache_tokens_rm with llama_kv_cache_clear

Use llama_kv_cache_clear for cache clearing

Change calls to llama_kv_cache_tokens_rm that want to delete by position to use llama_kv_cache_seq_rm functionality
This commit is contained in:
Kerfuffle authored and GitHub committed 2023-10-29 11:31:40 -06:00
1 parent 2046eb4345
commit 6e08281e58
8 files changed
+30 -32

No files matched your search

+1 -1
View File
@@ -857,7 +857,7 @@ struct llama_server_context
void kv_cache_clear() {
// clear the entire KV cache
llama_kv_cache_tokens_rm(ctx, -1, -1);
llama_kv_cache_clear(ctx);
clean_kv_cache = false;
}