Hi, > Hi Masatake, > > On Wed, Jul 11, 2012 at 01:58:31PM +0900, Masatake YAMATO wrote: >> (The patch sets are rebased to bluetooth-next. Unnecessary white >> spaces are trimmed.) >> >> lsof command can tell the type of socket processes are using. >> Internal lsof uses inode numbers on socket fs to resolve the type of >> sockets. Files under /proc/net/, such as tcp, udp, unix, etc provides >> such inode information. >> >> Unfortunately bluetooth related protocols don't provide such inode >> information. This patch series introduces /proc/net files for the protocols. >> >> This patch against af_bluetooth.c provides facility to the implementation >> of protocols. This patch extends bt_sock_list and introduces two exported >> function bt_procfs_init, bt_procfs_cleanup. >> >> The type bt_sock_list is already used in some of implementation of >> protocols. bt_procfs_init prepare seq_operations which converts >> protocol own bt_sock_list data to protocol own proc entry when the >> entry is accessed. >> >> What I, lsof user, need is just inode number of bluetooth >> socket. However, people may want more information. The bt_procfs_init >> takes a function pointer for customizing the show handler of >> seq_operations. > > I've tested the patch and it creates entries in /proc/net Thanks. > BTW: Shall lsof be able to identify l2cap protocol? > > for l2test it prints: "can't identify protocol" No. I'll work on lsof side after the kernel side patches are merged to the main line. Masatake YAMATO > Best regards > Andrei Emeltchenko > -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
