Function llama_load_session_file_internal

llama.cpp:10500–10552

}

static bool llama_load_session_file_internal(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out) {
    llama_file file(path_session, "rb");

    // sanity checks
    {
        const uint32_t magic   = file.read_u32();
        const uint32_t version = file.read_u32();

        if (magic != LLAMA_SESSION_MAGIC || version != LLAMA_SESSION_VERSION) {
            LLAMA_LOG_ERROR("%s : unknown (magic, version) for session file: %08x, %08x\n", __func__, magic, version);
            return false;
        }

        llama_hparams session_hparams;
        file.read_raw(&session_hparams, sizeof(llama_hparams));

        if (session_hparams != ctx->model.hparams) {
            LLAMA_LOG_INFO("%s : model hparams didn't match from session file!\n", __func__);
            return false;
        }
    }

    // load the prompt
    {
        const uint32_t n_token_count = file.read_u32();

        if (n_token_count > n_token_capacity) {
            LLAMA_LOG_ERROR("%s : token count in session file exceeded capacity! %u > %zu\n", __func__, n_token_count, n_token_capacity);
            return false;
        }

        file.read_raw(tokens_out, sizeof(llama_token) * n_token_count);
        *n_token_count_out = n_token_count;
    }

    // restore the context state
    {
        const size_t n_state_size_cur = file.size - file.tell();
        const size_t n_state_size_max = llama_get_state_size(ctx);

        if (n_state_size_cur > n_state_size_max) {
            LLAMA_LOG_ERROR("%s : the state size in session file is too big! max %zu, got %zu\n", __func__, n_state_size_max, n_state_size_cur);
            return false;
        }

        std::vector<uint8_t> state_data(n_state_size_max);
        file.read_raw(state_data.data(), n_state_size_cur);

        llama_set_state_data(ctx, state_data.data());
    }

    return true;
}

bool llama_load_session_file(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out) {
    try {