On Sat, Dec 31, 2005 at 01:25:12PM -0500, Mulyadi Santosa wrote: > Hello Mr McKinney... > > > set_task_state() ensures that no later memory references will be reordered, > > either by the CPU or by the compiler, to occur before the assignment > > to task_struct->state. > > > > A simple assignment does -not- make this guarantee. > > In this case, AFAIK HT is also considered as SMP....if it is, then > set_task_state() is the correct way to go? HT is indeed SMP as far as memory barriers are concerned. However, this does not necessarily mean that each and every assignment to task_struct->state is a bug. You need to look at the context -- what locks are held, whether the task structure in question is accessible to other CPUs (e.g., during initialization), and so on. Of course, if you have a test case that forces a failure, and changing to use set_task_state() prevents that failure, please submit the patch to LKML. Thanx, Paul -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
