hi all , i have one more doubt regarding mutual exclusion wihout hardware support and sorry if it is too basic to be asked on the list. i looked at the DEKKER's algo refered by u people but as far as i understood it works only for sequential memory model . what for processors that use store buffer and partial store ordering . if i do'nt have any hardware support for atomic operation (some test-and-set operation , or lock prefix or what ever..) how can i ensure the atomicity of any operation ( such as gaining of spin lock). Thanx in advance . Rgds, pankaj chauhan ________________________________________________________________________ Yahoo! India Mobile: Download the latest polyphonic ringtones. Go to http://in.mobile.yahoo.com -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/