Stefani Seibold wrote: > > Hi folks, > > there is some troubles with delayed non-realtime interrupts in > rtlinux. > > The problem is in the rtl_schedule() function, which did not call > the rtl_process_pending() function (schedulers/rtl_sched.c) if > rtlinux goes into the idle mode state. > > This will delay the pending non-realtime interrupts until a > rtl_soft_sti() (main/rtl_core.c) will be called and this can be > a very long time (for example: on a NSC Geode more than 4ms).
Apart from the specific problem you cite have you checked Geode intrinsic interrupt latency? On a Geode 266 based PC-104 board I found that the interrupt latency can be up to 300/400 us anyhow. You could check it under RTL by running the RTL scheduler in periodic mode and measuring the ticking period with an rdtsc directly in the RTL scheduler interrupt handler, or toggling a bit on the parport as well. Not that I checked the problem under a doing nothing DOS by trapping irq0, i.e. the standard 8254 timer interrupt, thus with the barest OS interaction. I'm curious to know if it is intrinsically related to the CPU/Chipset or to my specific PC-104 board. Paolo. -- [rtl] --- To unsubscribe: echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED] -- For more information on Real-Time Linux see: http://www.rtlinux.org/
