[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <4be661fc-033d-7be6-cda6-8e90510f1738@roeck-us.net>
Date: Fri, 9 Mar 2018 20:04:12 -0800
From: Guenter Roeck <linux@...ck-us.net>
To: Brian Norris <briannorris@...omium.org>,
Wim Van Sebroeck <wim@...ux-watchdog.org>
Cc: linux-kernel@...r.kernel.org, linux-watchdog@...r.kernel.org,
Doug Anderson <dianders@...omium.org>,
Matthias Kaehlcke <mka@...omium.org>,
linux-rockchip@...ts.infradead.org
Subject: Re: [PATCH v2 1/2] watchdog: dw: RMW the control register
On 03/09/2018 07:46 PM, Brian Norris wrote:
> RK3399 has rst_pulse_length in CONTROL_REG[4:2], determining the length
> of pulse to issue for system reset. We shouldn't clobber this value,
> because that might make the system reset ineffective. On RK3399, we're
> seeing that a value of 000b (meaning 2 cycles) yields an unreliable
> (partial?) reset, and so we only fully reset after the watchdog fires a
> second time. If we retain the system default (010b, or 8 clock cycles),
> then the watchdog reset is much more reliable.
>
> Read-modify-write retains the system value and improves reset
> reliability.
>
> It seems we were intentionally clobbering the response mode previously,
> to ensure we performed a system reset (we don't support an interrupt
> notification), so retain that explicitly.
>
> Signed-off-by: Brian Norris <briannorris@...omium.org>
Reviewed-by: Guenter Roeck <linux@...ck-us.net>
> ---
> v2:
> * factor out helper
> * handle both start() and restart() cases
> * note the RESP_MODE handling in commit message
> ---
> drivers/watchdog/dw_wdt.c | 23 +++++++++++++++--------
> 1 file changed, 15 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/watchdog/dw_wdt.c b/drivers/watchdog/dw_wdt.c
> index c2f4ff516230..918357bccf5e 100644
> --- a/drivers/watchdog/dw_wdt.c
> +++ b/drivers/watchdog/dw_wdt.c
> @@ -34,6 +34,7 @@
>
> #define WDOG_CONTROL_REG_OFFSET 0x00
> #define WDOG_CONTROL_REG_WDT_EN_MASK 0x01
> +#define WDOG_CONTROL_REG_RESP_MODE_MASK 0x02
> #define WDOG_TIMEOUT_RANGE_REG_OFFSET 0x04
> #define WDOG_TIMEOUT_RANGE_TOPINIT_SHIFT 4
> #define WDOG_CURRENT_COUNT_REG_OFFSET 0x08
> @@ -121,14 +122,23 @@ static int dw_wdt_set_timeout(struct watchdog_device *wdd, unsigned int top_s)
> return 0;
> }
>
> +static void dw_wdt_arm_system_reset(struct dw_wdt *dw_wdt)
> +{
> + u32 val = readl(dw_wdt->regs + WDOG_CONTROL_REG_OFFSET);
> +
> + /* Disable interrupt mode; always perform system reset. */
> + val &= ~WDOG_CONTROL_REG_RESP_MODE_MASK;
> + /* Enable watchdog. */
> + val |= WDOG_CONTROL_REG_WDT_EN_MASK;
> + writel(val, dw_wdt->regs + WDOG_CONTROL_REG_OFFSET);
> +}
> +
> static int dw_wdt_start(struct watchdog_device *wdd)
> {
> struct dw_wdt *dw_wdt = to_dw_wdt(wdd);
>
> dw_wdt_set_timeout(wdd, wdd->timeout);
> -
> - writel(WDOG_CONTROL_REG_WDT_EN_MASK,
> - dw_wdt->regs + WDOG_CONTROL_REG_OFFSET);
> + dw_wdt_arm_system_reset(dw_wdt);
>
> return 0;
> }
> @@ -152,16 +162,13 @@ static int dw_wdt_restart(struct watchdog_device *wdd,
> unsigned long action, void *data)
> {
> struct dw_wdt *dw_wdt = to_dw_wdt(wdd);
> - u32 val;
>
> writel(0, dw_wdt->regs + WDOG_TIMEOUT_RANGE_REG_OFFSET);
> - val = readl(dw_wdt->regs + WDOG_CONTROL_REG_OFFSET);
> - if (val & WDOG_CONTROL_REG_WDT_EN_MASK)
> + if (dw_wdt_is_enabled(dw_wdt))
> writel(WDOG_COUNTER_RESTART_KICK_VALUE,
> dw_wdt->regs + WDOG_COUNTER_RESTART_REG_OFFSET);
> else
> - writel(WDOG_CONTROL_REG_WDT_EN_MASK,
> - dw_wdt->regs + WDOG_CONTROL_REG_OFFSET);
> + dw_wdt_arm_system_reset(dw_wdt);
>
> /* wait for reset to assert... */
> mdelay(500);
>
Powered by blists - more mailing lists