diff --git a/src/llama-kv-cache-hybrid-recurrent.cpp b/src/llama-kv-cache-hybrid-recurrent.cpp index d269a6b50..8871dbf63 100644 --- a/src/llama-kv-cache-hybrid-recurrent.cpp +++ b/src/llama-kv-cache-hybrid-recurrent.cpp @@ -102,9 +102,10 @@ bool llama_kv_cache_hybrid_recurrent::get_can_shift() const { // Shifting is trivially supported for recurrent return kv_attn->get_can_shift(); } -void llama_kv_cache_hybrid_recurrent::clear() { - kv_attn ->clear(); - kv_recurrent->clear(); + +void llama_kv_cache_hybrid_recurrent::clear(bool data) { + kv_attn ->clear(data); + kv_recurrent->clear(data); } bool llama_kv_cache_hybrid_recurrent::seq_rm(llama_seq_id seq_id, llama_pos p0, llama_pos p1) { diff --git a/src/llama-kv-cache-hybrid-recurrent.h b/src/llama-kv-cache-hybrid-recurrent.h index e504631e4..8728fd733 100644 --- a/src/llama-kv-cache-hybrid-recurrent.h +++ b/src/llama-kv-cache-hybrid-recurrent.h @@ -55,7 +55,7 @@ public: bool get_can_shift() const override; - void clear() override; + void clear(bool data) override; bool seq_rm (llama_seq_id seq_id, llama_pos p0, llama_pos p1) override; void seq_cp (llama_seq_id seq_id_src, llama_seq_id seq_id_dst, llama_pos p0, llama_pos p1) override;