Mark Wendt wrote: > > I was thinking more of hardware resources, such as interrupts. What > happens when both instances require the same interrupt at the same time? > Unlike shared memory, interrupts are a bit harder to share, or have > concurrent use. Can the separate instances somehow share the IO bus? Or > is there some kind of real time scheduler that looks ahead for all possible > conflicts, and works it out? > This has always been a big problem on the PC platform, ie. the extreme shortage of interrupt vectors. This is mostly due to an interrupt architecture that assigns specific bus wires for a single interrupt. Some other platforms, such as the ARM, have interrupt requests that provide a bit field to specify the vector address (like the PDP-11) where hundreds of vectors can be available. Just another annoyance on the PC platform. At least on RTAI, there were some problems in having regularly scheduled threads running from interrupt lines instead of the timer. I hope Xenomai or prempt can be more flexible. On my interface boards, they can sample encoder position and then request an interrupt through the parallel port, which would reduce the effect of task latency on position sampling to zero, as long as the thread completed before the next interrupt.
Jon ------------------------------------------------------------------------------ Try New Relic Now & We'll Send You this Cool Shirt New Relic is the only SaaS-based application performance monitoring service that delivers powerful full stack analytics. Optimize and monitor your browser, app, & servers with just a few lines of code. Try New Relic and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_apr _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
