mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2025-06-26 19:55:04 +00:00
@ -1705,7 +1705,7 @@ size_t llama_context::state_write_data(llama_io_write_i & io) {
|
||||
}
|
||||
|
||||
llama_kv_cache * kv_self = static_cast<llama_kv_cache *>(memory.get());
|
||||
|
||||
|
||||
if (kv_self != nullptr) {
|
||||
LLAMA_LOG_DEBUG("%s: - writing KV self\n", __func__);
|
||||
kv_self->state_write(io);
|
||||
|
Reference in New Issue
Block a user