From mboxrd@z Thu Jan 1 00:00:00 1970 From: Laurent Pinchart Date: Tue, 03 Feb 2015 08:52:04 +0000 Subject: Re: [RFC 5/5] watchdog: sh_mobile: add driver Message-Id: <1803320.jWEVcQJxer@avalon> List-Id: References: <1422802074-1921-6-git-send-email-wsa@the-dreams.de> In-Reply-To: <1422802074-1921-6-git-send-email-wsa@the-dreams.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: linux-sh@vger.kernel.org Hi Wolfram, On Monday 02 February 2015 14:24:44 Wolfram Sang wrote: > >> + sh_wdt_start(&priv->wdev); > >> + sh_wdt_write(priv, 0xfff0, RWTCNT); > >=20 > > Is there a reason to set the counter to 0xfff0 instead of 0xffff ? >=20 > Giving serial console some time to push out FIFOs. But yeah, it is an > arbitrary value. 16 cycles as 32.768 kHz would be 488=B5s. Assuming a 115200 serial baud rat= e=20 that would allow transmission of a bit less than 6 characters. I wonder if = it's worth it. Shouldn't the restart core code instead flush the serial FIF= Os=20 explicitly before calling the restart handler if that behaviour is desired ? > >> + priv->restart_handler.notifier_call =3D sh_mobile_wdt_restart_handle= r; > >> + priv->restart_handler.priority =3D 192; > >=20 > > Just for my information, how did you choose 192 for the priority ? >=20 > Assuming DA9063 is 128, then I wanted to give this priority since it > doesn't need I2C access which might be stalled. --=20 Regards, Laurent Pinchart