On Fri, Jul 17, 2015 at 04:57:48PM +0200, Jan Lübbe wrote: > On Fr, 2015-07-17 at 15:39 +0200, Daniel Schultz wrote: > > The dmtimer0 is too inaccurate to be used for measurements. > > We switch to the more accurate dmtimer2. > > Ah, OK, after looking at the TRM again, the real reason for the > inaccuracy seems to be that the 32KiHz for dmtimer0 is *not* derived > from the RTC crystal, but from an internal RC oscillator. > > The kernel uses dmtimer1 for the clocksource by switching it to the > M_OSC in the PRCM. > > The dmtimer2 is used for clockevents and runs from M_OSC by default. > While we could set the dmtimer1 source to M_OSC in the clock setup, it > seems simpler to just use dmtimer2. > > The commit message should explain the reasoning behind the switch. Changed the commit message while applying to: The 32KHz from dmtimer0 is derived from a SoC internal RC oscillator which is quite inaccurate. Switch to dmtimer2 which is driven from the high frequency oscillator clock. Sascha -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox
