Dear all, after reading chapter 6 of the Linux Device Driver Book 2nd Edition, I still do not know the best way to achieve a timed execution of kernel code. For this reason, I contact this list to hope for a good hint. The Problem: In kernel space, I run some byte counter of which the code is loaded as a module. This works perfectly. I would like to have a function that gets triggered every second to read the actual state of the counters, this function is called timed_func() later. The function should be triggered irrespective of the status of the other functionality /* I know about the race conditions... and the fact that the linux kernel is not interruptible per node */ Ideas: In my mind the following ideas on came up how this problem could be solved in principle: a) implement timed_func() as a "normal" function in the same module -- this would be the easiest. b) implement timed_func() as a separate module-function. There, I assume there is no difference executionwise. c) implement timed_func() as an own kernel_thread such that it gets a schedulable entity. My Questions: - Is it possible to have timed_func() being launched every second in cases a) & b) or is the whole kernel module blocked until the function gets executed? If it would be possible to implement a) or b), how could I achieve this? - Is there another solution/idea that has not come to my mind so far? Thanks in advance for any hint! wbr, Lukas -- Lukas Ruf | Wanna know anything about raw | <http://www.lpr.ch> | IP? <http://www.rawip.org> | -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
