Hi, I am a new user of libevent - I ran into the following valgrind error on my machine:
Version: libevent-2.0.8-rc OS: Ubuntu 10.04 LTS amd64 Of interest is (bufferevent_sock.c:321) inside: struct bufferevent *bufferevent_socket_new() struct bufferevent_private *bufev_p; // ... if ((bufev_p = mm_calloc(1, sizeof(struct bufferevent_private)))== NULL) // ... // ---------------------------------------- bufev_p doesn't seem to get freed with mm_free() unless there is an error. My question is, what to do about this valgrind error? 1. There is no bug - suppress the valgrind message? 2. There is a flaw in the way I am calling bufferevent_socket_new(). In other words, I am not calling bufferevent_free() at the right time? 3. This code needs to be patched? 4. Something else? Please advise. Thank you. - Ken ==2933== 2,728 (496 direct, 2,232 indirect) bytes in 1 blocks are definitely lost in loss record 9 of 9 ==2933== at 0x4C278FC: calloc (vg_replace_malloc.c:467) ==2933== by 0x5781B55: bufferevent_socket_new (bufferevent_sock.c:321) ==2933== by 0x4CE3A5: accept_conn_cb(evconnlistener*, int, sockaddr*, int, void*) (Socket.cpp:71) ==2933== by 0x5783804: listener_read_cb (listener.c:377) ==2933== by 0x57796BB: event_base_loop (event.c:1287) ==2933== by 0x4CF773: robolab::Socket::startup() (Socket.cpp:226) ==2933== by 0x48ADF1: robolab::ClientServer::startup() (ClientServer.cpp:117) ==2933== by 0x473A76: serverThreadCB(void*) (Test_EchoClientServer.cpp:47) ==2933== by 0x66519C9: start_thread (pthread_create.c:300) ==2933== by 0x75526FC: clone (clone.S:112) *********************************************************************** To unsubscribe, send an e-mail to majord...@freehaven.net with unsubscribe libevent-users in the body.
