From: Marco Felsch <m.felsch@pengutronix.de>
To: Ahmad Fatoum <a.fatoum@pengutronix.de>
Cc: barebox@lists.infradead.org
Subject: Re: [PATCH 2/2] mfd: da9063: fix watchdog ping execution
Date: Mon, 4 Nov 2019 10:44:52 +0100 [thread overview]
Message-ID: <20191104094452.xxpoqvm6d33h4wnh@pengutronix.de> (raw)
In-Reply-To: <2d52df4c-9501-01a2-d67e-3086c1538bdb@pengutronix.de>
Hi Ahmad,
On 19-11-04 09:51, Ahmad Fatoum wrote:
> Hello Marco,
>
> On 11/4/19 9:34 AM, Marco Felsch wrote:
> > On 19-11-04 09:27, Sascha Hauer wrote:
> >> On Wed, Oct 30, 2019 at 06:06:53PM +0100, Marco Felsch wrote:
> >>> The watchdog resets the system if the watchdog gets pinged to fast.
> >>> Between each watchdog ping must be a pause of at least 200ms.
>
> I assume you're using the boot.watchdog_timeout parameter? The time contained
> in this parameter is communicated to the watchdog in boot_entry, which is called
> twice in a normal bootchooser boot, once from the boot command and once more
> from bootchooser.
Thats correct. I saw this bug during testing my boot.default param so
didn't noticed that bootchooser uses boot_entry too.
> This means that your boot time would increase by 200 ms. If this matter to you,
> you might want to change this, so watchdog_set_timeout is called only once.
Increasing the delay isn't a big deal. But after we discussed it again I
will send a v2 which handles the to fast pings by dropping those.
> And if you do so, you could drop this patch. The only other places that feed
> the watchdog are the watchdog poller and the wd command. The watchdog poller
> already waits 500 ms between pings and the command is meant for debugging/testing.
> If someone wants to feed the watchdog that fast while testing, why prevent them?
Becuase if the watchdog gets feeded to fast then the system gets
reseted. So dropping the patch isn't a option.
> (I assume you don't need to wait 200 ms between ping and disabling WDT, if you do,
> one more place is the .priority watchdog device parameter in barebox-next
Sorry, I don't get this. You don't need to wait 200ms between ping and
disabling.
Regards,
Marco
> Cheers
> Ahmad
>
> >>>
> >>> Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
> >>> ---
> >>> drivers/mfd/da9063.c | 11 +++++++++++
> >>> 1 file changed, 11 insertions(+)
> >>>
> >>> diff --git a/drivers/mfd/da9063.c b/drivers/mfd/da9063.c
> >>> index 4d459c7f18..ab57885240 100644
> >>> --- a/drivers/mfd/da9063.c
> >>> +++ b/drivers/mfd/da9063.c
> >>> @@ -14,6 +14,7 @@
> >>> */
> >>>
> >>> #include <common.h>
> >>> +#include <clock.h>
> >>> #include <driver.h>
> >>> #include <gpio.h>
> >>> #include <restart.h>
> >>> @@ -33,6 +34,7 @@ struct da9063 {
> >>> struct i2c_client *client1;
> >>> struct device_d *dev;
> >>> unsigned int timeout;
> >>> + uint64_t last_ping;
> >>> };
> >>>
> >>> /* forbidden/impossible value; timeout will be set to this value initially to
> >>> @@ -237,6 +239,13 @@ static int da9063_watchdog_ping(struct da9063 *priv)
> >>> int ret;
> >>> u8 val;
> >>>
> >>> + /* We need to wait at least 200ms till we can resend a ping */
> >>> + if (!is_timeout_non_interruptible(priv->last_ping, 200 * MSECOND)) {
> >>> + dev_dbg(priv->dev, "active ping delay\n");
> >>> + mdelay(50);
> >>
> >> I would expect to wait the missing time to 200ms here. Maybe doing
> >> nothing in this case would be more appropriate here. I mean, why should
> >> you slow down barebox here when some code triggers the watchdog too
> >> often?
> >>
> >>> + return da9063_watchdog_ping(priv);
> >>
> >> Drop this, just fall through.
> >
> > Just prepared a v2 with a busy wait after discussed it with Lucas.
> > Thanks for your input too :)
> >
> > Regards,
> > Marco
> >
> >> 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 |
> >>
> >
>
> --
> 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 |
>
--
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@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2019-11-04 9:44 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-30 17:06 [PATCH 1/2] mfd: da9063: fix TWDSCALE debug message Marco Felsch
2019-10-30 17:06 ` [PATCH 2/2] mfd: da9063: fix watchdog ping execution Marco Felsch
2019-11-04 8:27 ` Sascha Hauer
2019-11-04 8:34 ` Marco Felsch
2019-11-04 8:51 ` Ahmad Fatoum
2019-11-04 9:44 ` Marco Felsch [this message]
2019-11-04 9:57 ` Ahmad Fatoum
2019-11-04 10:13 ` Marco Felsch
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191104094452.xxpoqvm6d33h4wnh@pengutronix.de \
--to=m.felsch@pengutronix.de \
--cc=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox