Hi Trent, On 16/03/2018 11:12:02-0700, Trent Piepho wrote: > In order to read correctly from asynchronously updated RTC registers, it's > necessary to read repeatedly until their values do not change from read to > read. There is no timeout in this code and it could possibly loop > forever. > > It's also necessary to wait for three RTC clock ticks for certain > operations to take effect and the driver might wait forever for this to > happen. > > To avoid kernel hangs, put in timeouts. > > These hangs will happen when running under qemu, which doesn't emulate the > SNVS RTC. It could also happen if the RTC block where somehow placed into > reset or the slow speed clock that drives the RTC counter (but not the > CPU) were to stop. > > The symptoms are a work queue hang on rtc_timer_do_work(), which > eventually blocks a systemd fsnotify operation that triggers a work queue > flush, causing systemd to hang and thus causing all services that should > be started by systemd, like a console getty, to fail to start or stop. > > Also optimize the wait code to wait less. It only needs to wait for the > clock to advance three ticks, not to see it change three times. > > Signed-off-by: Trent Piepho <tpiepho@xxxxxxxxxx> > Cc: Alessandro Zummo <a.zummo@xxxxxxxxxxxx> > Cc: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxxxxxxxxx> > Cc: linux-rtc@xxxxxxxxxxxxxxx > Cc: Shawn Guo <shawnguo@xxxxxxxxxx> > Cc: Sascha Hauer <kernel@xxxxxxxxxxxxxx> > Cc: Fabio Estevam <fabio.estevam@xxxxxxx> > Can you rebase that patch on top of rtc-next if you feel this is still necessary? Also, it would be great to have Shawn or Fabio's ack. Thanks! -- Alexandre Belloni, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering https://bootlin.com
