rpc: fix segfault on invalid endpoint format (#18387)

* rpc: fix segfault on invalid endpoint format

* rpc: add error log for failed endpoint connection

diff --git a/common/arg.cpp b/common/arg.cpp
index 189470182abe92501400fd6f49ee84a65f733848..62d31393c4390be0eb58145c547675e4140a091e 100644 (file)
--- a/common/arg.cpp
+++ b/common/arg.cpp
@@ -2017,7 +2017,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
     if (llama_supports_rpc()) {
         add_opt(common_arg(
             {"--rpc"}, "SERVERS",
-            "comma separated list of RPC servers",
+            "comma separated list of RPC servers (host:port)",
             [](common_params & params, const std::string & value) {
                 add_rpc_devices(value);
                 GGML_UNUSED(params);

diff --git a/ggml/src/ggml-rpc/ggml-rpc.cpp b/ggml/src/ggml-rpc/ggml-rpc.cpp
index e7890a5ee935448dfb56f9b3e8f7d7fefa06b21d..164b39d01e2d89c9e71fd3693a5edb48ad6dfd09 100644 (file)
--- a/ggml/src/ggml-rpc/ggml-rpc.cpp
+++ b/ggml/src/ggml-rpc/ggml-rpc.cpp
@@ -524,6 +524,7 @@ static std::shared_ptr<socket_t> get_socket(const std::string & endpoint) {
     std::string host;
     int port;
     if (!parse_endpoint(endpoint, host, port)) {
+        GGML_LOG_ERROR("Failed to parse endpoint: %s\n", endpoint.c_str());
         return nullptr;
     }
 #ifdef _WIN32
@@ -2053,6 +2054,10 @@ ggml_backend_reg_t ggml_backend_rpc_reg(void) {
 
 static uint32_t ggml_backend_rpc_get_device_count(const char * endpoint) {
     auto sock = get_socket(endpoint);
+    if (sock == nullptr) {
+        GGML_LOG_ERROR("Failed to connect to %s\n", endpoint);
+        return 0;
+    }
     rpc_msg_device_count_rsp response;
     bool status = send_rpc_cmd(sock, RPC_CMD_DEVICE_COUNT, nullptr, 0, &response, sizeof(response));
     RPC_STATUS_ASSERT(status);