Heartbeat LED during init script
Marcel Hellwig
barebox+mailing at cookiesoft.de
Thu Feb 18 03:00:42 EST 2021
Hey Sasha,
I have a slight problem here:
This is what I think:
Because the memtest gets called in the init routine, the `ctrl` is completly ignored, at least I can't interrupt it nor the LED blinks.
But if I insert `poller_call`, the led blinks, but the memtest is painfully slow (like 10x slower at least).
I don't think that this will work for me, but I have to accept, that the LED doesn't blink while doing a memtest.
But an interruptable memtest would be nice at least for the user, so my patch I sent yesterday still remains ;)
Greetings,
Marcel
> Sascha Hauer <sha at pengutronix.de> hat am 17.02.2021 14:17 geschrieben:
>
>
> On Wed, Feb 17, 2021 at 01:44:59PM +0100, barebox+mailing at cookiesoft.de wrote:
> > Hey Sasha,
> >
> > thanks for the answer.
> >
> > > barebox doesn't support interrupts. LED blinking is done in pollers
> >
> > That explains the obversation.
> >
> > > add a ctrlc() call to your memtester code inside some loop.
> > > That will cause the poller to run.
> >
> > I'm not entirely sure, if I understand you correctly or you didn't understand me.
> >
> > I use the memtest command[0] from barebox so there is nothing custom written or any loop that I can control.
>
> Ok, somehow I thought you added some memtest code yourself. I was
> mistaken.
>
> > However, if I understand you correctly one could add the `ctrlc`
> > command in all of these[1] three loops inside the
> > `mem_test_moving_inversions` function. Is that correct?
>
> Yes, correct.
>
> > Or is there an
> > other way to not listen on ctrl+c but let the poller itself run?
>
> You could add poller_call() directly. The idea of adding ctrlc() is just
> that this way you could also make the memtest interruptible for the
> user.
>
> Sascha
>
> --
> Pengutronix e.K. | |
> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
More information about the barebox
mailing list