On Fri, 2015-12-18 at 16:18 -0300, Marcos Díaz wrote: > Hi, i'm using c++ chrono to measure a steady signal periods (With > GPIO > interrupts), And I see that periodically I have errors of +1ms and > then of > -1ms. (I discarded GPIOP interrupts problem) I tried using rtems > API's > rtems_clock_get_uptime_nanoseconds > and then I saw the same errors. > I'm suspecting of some problem related with interrupts and the > updating of > bintime uptime in kern_tc.c related to the simple downcounter ticker > this > bsp has. > Has anybody seen something like this in any other bsp?
Yep, this sounds familiar... See https://devel.rtems.org/ticket/2230 (pre-timecounter changes) and https://devel.rtems.org/ticket/2356 (post-). -- Nick Withers Embedded Systems Programmer Department of Nuclear Physics, Research School of Physics and Engineering The Australian National University (CRICOS: 00120C) _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
