On 2013-10-09 11:13, Hidde Verstoep (E2M) wrote: > Both our versions use the linux timers. While trying to find a > solution I found that the current rt_e1000 driver also uses linux > timers.
The key is that the rt_e1000 doesn't call Linux timer services from real-time contexts or while holding real-time locks. > A lot of other rt drivers use busy waits for small timeouts. > However, the timeout here is quite large and a busy wait does not seem > like the right solution. I wanted to try to use the rtdm timer service > (http://www.xenomai.org/documentation/trunk/html/api/group__rtdmtimer.html). > However, these timers are not able to pass context to a handler. So, a > handler wouldn't know which device to check in case of multiple > devices. Could you give a suggestion as how to solve it Jan? container_of(timer...) - the timer object is usually embedded in a struct that describes the required context. But you don't need real-timer timers unless the timeout is on the critical path. That wasn't the case with RTnet drivers so far. Jan
signature.asc
Description: OpenPGP digital signature
------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
_______________________________________________ RTnet-users mailing list RTnet-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rtnet-users
