* [PATCH 0/2] watchdog: rti: adjust initial ping for attach @ 2020-06-24 11:45 Tero Kristo 2020-06-24 11:45 ` [PATCH 1/2] watchdog: use __watchdog_ping in startup Tero Kristo 2020-06-24 11:45 ` [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window Tero Kristo 0 siblings, 2 replies; 13+ messages in thread From: Tero Kristo @ 2020-06-24 11:45 UTC (permalink / raw) To: wim, linux, linux-watchdog; +Cc: linux-kernel, jan.kiszka Hi, This series fixes attaching the RTI watchdog driver to an already running timer; it can be started by boot loader for example. In this case, we must read the current remaining timeout, and adjust the min_hw_heartbeat based on that so that we don't attempt to either pet the watchdog too early, or too late. The reason for all this is because the K3 RTI watchdog runs in windowed mode only, petting it either too early, or too late causes a watchdog reset. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply [flat|nested] 13+ messages in thread
* [PATCH 1/2] watchdog: use __watchdog_ping in startup 2020-06-24 11:45 [PATCH 0/2] watchdog: rti: adjust initial ping for attach Tero Kristo @ 2020-06-24 11:45 ` Tero Kristo 2020-06-30 20:36 ` Guenter Roeck 2020-06-24 11:45 ` [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window Tero Kristo 1 sibling, 1 reply; 13+ messages in thread From: Tero Kristo @ 2020-06-24 11:45 UTC (permalink / raw) To: wim, linux, linux-watchdog; +Cc: linux-kernel, jan.kiszka Current watchdog startup functionality does not respect the minimum hw heartbeat setup and the last watchdog ping timeframe when watchdog is already running and userspace process attaches to it. Fix this by using the __watchdog_ping from the startup also. For this code path, we can also let the __watchdog_ping handle the bookkeeping for the worker and last keepalive times. Signed-off-by: Tero Kristo <t-kristo@ti.com> --- drivers/watchdog/watchdog_dev.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c index 7e4cd34a8c20..bc1cfa288553 100644 --- a/drivers/watchdog/watchdog_dev.c +++ b/drivers/watchdog/watchdog_dev.c @@ -275,15 +275,18 @@ static int watchdog_start(struct watchdog_device *wdd) set_bit(_WDOG_KEEPALIVE, &wd_data->status); started_at = ktime_get(); - if (watchdog_hw_running(wdd) && wdd->ops->ping) - err = wdd->ops->ping(wdd); - else + if (watchdog_hw_running(wdd) && wdd->ops->ping) { + err = __watchdog_ping(wdd); + if (err == 0) + set_bit(WDOG_ACTIVE, &wdd->status); + } else { err = wdd->ops->start(wdd); - if (err == 0) { - set_bit(WDOG_ACTIVE, &wdd->status); - wd_data->last_keepalive = started_at; - wd_data->last_hw_keepalive = started_at; - watchdog_update_worker(wdd); + if (err == 0) { + set_bit(WDOG_ACTIVE, &wdd->status); + wd_data->last_keepalive = started_at; + wd_data->last_hw_keepalive = started_at; + watchdog_update_worker(wdd); + } } return err; -- 2.17.1 -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply related [flat|nested] 13+ messages in thread
* Re: [PATCH 1/2] watchdog: use __watchdog_ping in startup 2020-06-24 11:45 ` [PATCH 1/2] watchdog: use __watchdog_ping in startup Tero Kristo @ 2020-06-30 20:36 ` Guenter Roeck 0 siblings, 0 replies; 13+ messages in thread From: Guenter Roeck @ 2020-06-30 20:36 UTC (permalink / raw) To: Tero Kristo; +Cc: wim, linux-watchdog, linux-kernel, jan.kiszka On Wed, Jun 24, 2020 at 02:45:33PM +0300, Tero Kristo wrote: > Current watchdog startup functionality does not respect the minimum hw > heartbeat setup and the last watchdog ping timeframe when watchdog is > already running and userspace process attaches to it. Fix this by using > the __watchdog_ping from the startup also. For this code path, we can > also let the __watchdog_ping handle the bookkeeping for the worker and > last keepalive times. > > Signed-off-by: Tero Kristo <t-kristo@ti.com> Reviewed-by: Guenter Roeck <linux@roeck-us.net> > --- > drivers/watchdog/watchdog_dev.c | 19 +++++++++++-------- > 1 file changed, 11 insertions(+), 8 deletions(-) > > diff --git a/drivers/watchdog/watchdog_dev.c b/drivers/watchdog/watchdog_dev.c > index 7e4cd34a8c20..bc1cfa288553 100644 > --- a/drivers/watchdog/watchdog_dev.c > +++ b/drivers/watchdog/watchdog_dev.c > @@ -275,15 +275,18 @@ static int watchdog_start(struct watchdog_device *wdd) > set_bit(_WDOG_KEEPALIVE, &wd_data->status); > > started_at = ktime_get(); > - if (watchdog_hw_running(wdd) && wdd->ops->ping) > - err = wdd->ops->ping(wdd); > - else > + if (watchdog_hw_running(wdd) && wdd->ops->ping) { > + err = __watchdog_ping(wdd); > + if (err == 0) > + set_bit(WDOG_ACTIVE, &wdd->status); > + } else { > err = wdd->ops->start(wdd); > - if (err == 0) { > - set_bit(WDOG_ACTIVE, &wdd->status); > - wd_data->last_keepalive = started_at; > - wd_data->last_hw_keepalive = started_at; > - watchdog_update_worker(wdd); > + if (err == 0) { > + set_bit(WDOG_ACTIVE, &wdd->status); > + wd_data->last_keepalive = started_at; > + wd_data->last_hw_keepalive = started_at; > + watchdog_update_worker(wdd); > + } > } > > return err; ^ permalink raw reply [flat|nested] 13+ messages in thread
* [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-24 11:45 [PATCH 0/2] watchdog: rti: adjust initial ping for attach Tero Kristo 2020-06-24 11:45 ` [PATCH 1/2] watchdog: use __watchdog_ping in startup Tero Kristo @ 2020-06-24 11:45 ` Tero Kristo 2020-06-24 15:24 ` Jan Kiszka 1 sibling, 1 reply; 13+ messages in thread From: Tero Kristo @ 2020-06-24 11:45 UTC (permalink / raw) To: wim, linux, linux-watchdog; +Cc: linux-kernel, jan.kiszka If the RTI watchdog has been started by someone (like bootloader) when the driver probes, we must adjust the initial ping timeout to match the currently running watchdog window to avoid generating watchdog reset. Signed-off-by: Tero Kristo <t-kristo@ti.com> --- drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c index d456dd72d99a..02ea2b2435f5 100644 --- a/drivers/watchdog/rti_wdt.c +++ b/drivers/watchdog/rti_wdt.c @@ -55,11 +55,13 @@ static int heartbeat; * @base - base io address of WD device * @freq - source clock frequency of WDT * @wdd - hold watchdog device as is in WDT core + * @min_hw_heartbeat_save - save of the min hw heartbeat value */ struct rti_wdt_device { void __iomem *base; unsigned long freq; struct watchdog_device wdd; + unsigned int min_hw_heartbeat_save; }; static int rti_wdt_start(struct watchdog_device *wdd) @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) /* put watchdog in active state */ writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); + if (wdt->min_hw_heartbeat_save) { + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; + wdt->min_hw_heartbeat_save = 0; + } + return 0; } @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) goto err_iomap; } + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { + u32 time_left; + u32 heartbeat; + + set_bit(WDOG_HW_RUNNING, &wdd->status); + time_left = rti_wdt_get_timeleft(wdd); + heartbeat = readl(wdt->base + RTIDWDPRLD); + heartbeat <<= WDT_PRELOAD_SHIFT; + heartbeat /= wdt->freq; + if (time_left < heartbeat / 2) + wdd->min_hw_heartbeat_ms = 0; + else + wdd->min_hw_heartbeat_ms = + (time_left - heartbeat / 2 + 1) * 1000; + + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; + } + ret = watchdog_register_device(wdd); if (ret) { dev_err(dev, "cannot register watchdog device\n"); -- 2.17.1 -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply related [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-24 11:45 ` [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window Tero Kristo @ 2020-06-24 15:24 ` Jan Kiszka 2020-06-25 8:32 ` Tero Kristo 0 siblings, 1 reply; 13+ messages in thread From: Jan Kiszka @ 2020-06-24 15:24 UTC (permalink / raw) To: Tero Kristo, wim, linux, linux-watchdog; +Cc: linux-kernel On 24.06.20 13:45, Tero Kristo wrote: > If the RTI watchdog has been started by someone (like bootloader) when > the driver probes, we must adjust the initial ping timeout to match the > currently running watchdog window to avoid generating watchdog reset. > > Signed-off-by: Tero Kristo <t-kristo@ti.com> > --- > drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ > 1 file changed, 25 insertions(+) > > diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c > index d456dd72d99a..02ea2b2435f5 100644 > --- a/drivers/watchdog/rti_wdt.c > +++ b/drivers/watchdog/rti_wdt.c > @@ -55,11 +55,13 @@ static int heartbeat; > * @base - base io address of WD device > * @freq - source clock frequency of WDT > * @wdd - hold watchdog device as is in WDT core > + * @min_hw_heartbeat_save - save of the min hw heartbeat value > */ > struct rti_wdt_device { > void __iomem *base; > unsigned long freq; > struct watchdog_device wdd; > + unsigned int min_hw_heartbeat_save; > }; > > static int rti_wdt_start(struct watchdog_device *wdd) > @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) > /* put watchdog in active state */ > writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); > > + if (wdt->min_hw_heartbeat_save) { > + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; > + wdt->min_hw_heartbeat_save = 0; > + } > + > return 0; > } > > @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) > goto err_iomap; > } > > + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { > + u32 time_left; > + u32 heartbeat; > + > + set_bit(WDOG_HW_RUNNING, &wdd->status); > + time_left = rti_wdt_get_timeleft(wdd); > + heartbeat = readl(wdt->base + RTIDWDPRLD); > + heartbeat <<= WDT_PRELOAD_SHIFT; > + heartbeat /= wdt->freq; > + if (time_left < heartbeat / 2) > + wdd->min_hw_heartbeat_ms = 0; > + else > + wdd->min_hw_heartbeat_ms = > + (time_left - heartbeat / 2 + 1) * 1000; > + > + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; > + } > + > ret = watchdog_register_device(wdd); > if (ret) { > dev_err(dev, "cannot register watchdog device\n"); > This assumes that the bootloader also programmed a 50% window, right? The pending U-Boot patch will do that, but what if that may chance or someone uses a different setup? Jan -- Siemens AG, Corporate Technology, CT RDA IOT SES-DE Corporate Competence Center Embedded Linux ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-24 15:24 ` Jan Kiszka @ 2020-06-25 8:32 ` Tero Kristo 2020-06-25 10:00 ` Jan Kiszka 2020-06-25 13:35 ` Guenter Roeck 0 siblings, 2 replies; 13+ messages in thread From: Tero Kristo @ 2020-06-25 8:32 UTC (permalink / raw) To: Jan Kiszka, wim, linux, linux-watchdog; +Cc: linux-kernel On 24/06/2020 18:24, Jan Kiszka wrote: > On 24.06.20 13:45, Tero Kristo wrote: >> If the RTI watchdog has been started by someone (like bootloader) when >> the driver probes, we must adjust the initial ping timeout to match the >> currently running watchdog window to avoid generating watchdog reset. >> >> Signed-off-by: Tero Kristo <t-kristo@ti.com> >> --- >> drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ >> 1 file changed, 25 insertions(+) >> >> diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c >> index d456dd72d99a..02ea2b2435f5 100644 >> --- a/drivers/watchdog/rti_wdt.c >> +++ b/drivers/watchdog/rti_wdt.c >> @@ -55,11 +55,13 @@ static int heartbeat; >> * @base - base io address of WD device >> * @freq - source clock frequency of WDT >> * @wdd - hold watchdog device as is in WDT core >> + * @min_hw_heartbeat_save - save of the min hw heartbeat value >> */ >> struct rti_wdt_device { >> void __iomem *base; >> unsigned long freq; >> struct watchdog_device wdd; >> + unsigned int min_hw_heartbeat_save; >> }; >> static int rti_wdt_start(struct watchdog_device *wdd) >> @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) >> /* put watchdog in active state */ >> writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); >> + if (wdt->min_hw_heartbeat_save) { >> + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; >> + wdt->min_hw_heartbeat_save = 0; >> + } >> + >> return 0; >> } >> @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device >> *pdev) >> goto err_iomap; >> } >> + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { >> + u32 time_left; >> + u32 heartbeat; >> + >> + set_bit(WDOG_HW_RUNNING, &wdd->status); >> + time_left = rti_wdt_get_timeleft(wdd); >> + heartbeat = readl(wdt->base + RTIDWDPRLD); >> + heartbeat <<= WDT_PRELOAD_SHIFT; >> + heartbeat /= wdt->freq; >> + if (time_left < heartbeat / 2) >> + wdd->min_hw_heartbeat_ms = 0; >> + else >> + wdd->min_hw_heartbeat_ms = >> + (time_left - heartbeat / 2 + 1) * 1000; >> + >> + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; >> + } >> + >> ret = watchdog_register_device(wdd); >> if (ret) { >> dev_err(dev, "cannot register watchdog device\n"); >> > > This assumes that the bootloader also programmed a 50% window, right? > The pending U-Boot patch will do that, but what if that may chance or > someone uses a different setup? Yes, we assume 50%. I think based on the hw design, 50% is the only sane value to be used, otherwise you just shrink the open window too much and for no apparent reason. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-25 8:32 ` Tero Kristo @ 2020-06-25 10:00 ` Jan Kiszka 2020-06-25 13:35 ` Guenter Roeck 1 sibling, 0 replies; 13+ messages in thread From: Jan Kiszka @ 2020-06-25 10:00 UTC (permalink / raw) To: Tero Kristo, wim, linux, linux-watchdog; +Cc: linux-kernel On 25.06.20 10:32, Tero Kristo wrote: > On 24/06/2020 18:24, Jan Kiszka wrote: >> On 24.06.20 13:45, Tero Kristo wrote: >>> If the RTI watchdog has been started by someone (like bootloader) when >>> the driver probes, we must adjust the initial ping timeout to match the >>> currently running watchdog window to avoid generating watchdog reset. >>> >>> Signed-off-by: Tero Kristo <t-kristo@ti.com> >>> --- >>> drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ >>> 1 file changed, 25 insertions(+) >>> >>> diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c >>> index d456dd72d99a..02ea2b2435f5 100644 >>> --- a/drivers/watchdog/rti_wdt.c >>> +++ b/drivers/watchdog/rti_wdt.c >>> @@ -55,11 +55,13 @@ static int heartbeat; >>> * @base - base io address of WD device >>> * @freq - source clock frequency of WDT >>> * @wdd - hold watchdog device as is in WDT core >>> + * @min_hw_heartbeat_save - save of the min hw heartbeat value >>> */ >>> struct rti_wdt_device { >>> void __iomem *base; >>> unsigned long freq; >>> struct watchdog_device wdd; >>> + unsigned int min_hw_heartbeat_save; >>> }; >>> static int rti_wdt_start(struct watchdog_device *wdd) >>> @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device >>> *wdd) >>> /* put watchdog in active state */ >>> writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); >>> + if (wdt->min_hw_heartbeat_save) { >>> + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; >>> + wdt->min_hw_heartbeat_save = 0; >>> + } >>> + >>> return 0; >>> } >>> @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device >>> *pdev) >>> goto err_iomap; >>> } >>> + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { >>> + u32 time_left; >>> + u32 heartbeat; >>> + >>> + set_bit(WDOG_HW_RUNNING, &wdd->status); >>> + time_left = rti_wdt_get_timeleft(wdd); >>> + heartbeat = readl(wdt->base + RTIDWDPRLD); >>> + heartbeat <<= WDT_PRELOAD_SHIFT; >>> + heartbeat /= wdt->freq; >>> + if (time_left < heartbeat / 2) >>> + wdd->min_hw_heartbeat_ms = 0; >>> + else >>> + wdd->min_hw_heartbeat_ms = >>> + (time_left - heartbeat / 2 + 1) * 1000; >>> + >>> + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; >>> + } >>> + >>> ret = watchdog_register_device(wdd); >>> if (ret) { >>> dev_err(dev, "cannot register watchdog device\n"); >>> >> >> This assumes that the bootloader also programmed a 50% window, right? >> The pending U-Boot patch will do that, but what if that may chance or >> someone uses a different setup? > > Yes, we assume 50%. I think based on the hw design, 50% is the only sane > value to be used, otherwise you just shrink the open window too much and > for no apparent reason. Fine with me, but should we check that assumption when adopting the watchdog? Jan -- Siemens AG, Corporate Technology, CT RDA IOT SES-DE Corporate Competence Center Embedded Linux ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-25 8:32 ` Tero Kristo 2020-06-25 10:00 ` Jan Kiszka @ 2020-06-25 13:35 ` Guenter Roeck 2020-06-25 17:04 ` Tero Kristo 1 sibling, 1 reply; 13+ messages in thread From: Guenter Roeck @ 2020-06-25 13:35 UTC (permalink / raw) To: Tero Kristo, Jan Kiszka, wim, linux-watchdog; +Cc: linux-kernel On 6/25/20 1:32 AM, Tero Kristo wrote: > On 24/06/2020 18:24, Jan Kiszka wrote: >> On 24.06.20 13:45, Tero Kristo wrote: >>> If the RTI watchdog has been started by someone (like bootloader) when >>> the driver probes, we must adjust the initial ping timeout to match the >>> currently running watchdog window to avoid generating watchdog reset. >>> >>> Signed-off-by: Tero Kristo <t-kristo@ti.com> >>> --- >>> drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ >>> 1 file changed, 25 insertions(+) >>> >>> diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c >>> index d456dd72d99a..02ea2b2435f5 100644 >>> --- a/drivers/watchdog/rti_wdt.c >>> +++ b/drivers/watchdog/rti_wdt.c >>> @@ -55,11 +55,13 @@ static int heartbeat; >>> * @base - base io address of WD device >>> * @freq - source clock frequency of WDT >>> * @wdd - hold watchdog device as is in WDT core >>> + * @min_hw_heartbeat_save - save of the min hw heartbeat value >>> */ >>> struct rti_wdt_device { >>> void __iomem *base; >>> unsigned long freq; >>> struct watchdog_device wdd; >>> + unsigned int min_hw_heartbeat_save; >>> }; >>> static int rti_wdt_start(struct watchdog_device *wdd) >>> @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) >>> /* put watchdog in active state */ >>> writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); >>> + if (wdt->min_hw_heartbeat_save) { >>> + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; >>> + wdt->min_hw_heartbeat_save = 0; >>> + } >>> + >>> return 0; >>> } >>> @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) >>> goto err_iomap; >>> } >>> + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { >>> + u32 time_left; >>> + u32 heartbeat; >>> + >>> + set_bit(WDOG_HW_RUNNING, &wdd->status); >>> + time_left = rti_wdt_get_timeleft(wdd); >>> + heartbeat = readl(wdt->base + RTIDWDPRLD); >>> + heartbeat <<= WDT_PRELOAD_SHIFT; >>> + heartbeat /= wdt->freq; >>> + if (time_left < heartbeat / 2) >>> + wdd->min_hw_heartbeat_ms = 0; >>> + else >>> + wdd->min_hw_heartbeat_ms = >>> + (time_left - heartbeat / 2 + 1) * 1000; >>> + >>> + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; >>> + } >>> + >>> ret = watchdog_register_device(wdd); >>> if (ret) { >>> dev_err(dev, "cannot register watchdog device\n"); >>> >> >> This assumes that the bootloader also programmed a 50% window, right? The pending U-Boot patch will do that, but what if that may chance or someone uses a different setup? > > Yes, we assume 50%. I think based on the hw design, 50% is the only sane value to be used, otherwise you just shrink the open window too much and for no apparent reason. > Not sure if that is a valid assumption. Someone who designs a watchdog with such a narrow ping window might as well also use it. The question is if you want to rely on that assumption, or check and change it if needed. Also, I wonder if we should add an API function such as "set_last_hw_keepalive()" to avoid all that complexity. Thanks, Guenter ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-25 13:35 ` Guenter Roeck @ 2020-06-25 17:04 ` Tero Kristo 2020-06-30 20:23 ` Guenter Roeck 0 siblings, 1 reply; 13+ messages in thread From: Tero Kristo @ 2020-06-25 17:04 UTC (permalink / raw) To: Guenter Roeck, Jan Kiszka, wim, linux-watchdog; +Cc: linux-kernel On 25/06/2020 16:35, Guenter Roeck wrote: > On 6/25/20 1:32 AM, Tero Kristo wrote: >> On 24/06/2020 18:24, Jan Kiszka wrote: >>> On 24.06.20 13:45, Tero Kristo wrote: >>>> If the RTI watchdog has been started by someone (like bootloader) when >>>> the driver probes, we must adjust the initial ping timeout to match the >>>> currently running watchdog window to avoid generating watchdog reset. >>>> >>>> Signed-off-by: Tero Kristo <t-kristo@ti.com> >>>> --- >>>> drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ >>>> 1 file changed, 25 insertions(+) >>>> >>>> diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c >>>> index d456dd72d99a..02ea2b2435f5 100644 >>>> --- a/drivers/watchdog/rti_wdt.c >>>> +++ b/drivers/watchdog/rti_wdt.c >>>> @@ -55,11 +55,13 @@ static int heartbeat; >>>> * @base - base io address of WD device >>>> * @freq - source clock frequency of WDT >>>> * @wdd - hold watchdog device as is in WDT core >>>> + * @min_hw_heartbeat_save - save of the min hw heartbeat value >>>> */ >>>> struct rti_wdt_device { >>>> void __iomem *base; >>>> unsigned long freq; >>>> struct watchdog_device wdd; >>>> + unsigned int min_hw_heartbeat_save; >>>> }; >>>> static int rti_wdt_start(struct watchdog_device *wdd) >>>> @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) >>>> /* put watchdog in active state */ >>>> writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); >>>> + if (wdt->min_hw_heartbeat_save) { >>>> + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; >>>> + wdt->min_hw_heartbeat_save = 0; >>>> + } >>>> + >>>> return 0; >>>> } >>>> @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) >>>> goto err_iomap; >>>> } >>>> + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { >>>> + u32 time_left; >>>> + u32 heartbeat; >>>> + >>>> + set_bit(WDOG_HW_RUNNING, &wdd->status); >>>> + time_left = rti_wdt_get_timeleft(wdd); >>>> + heartbeat = readl(wdt->base + RTIDWDPRLD); >>>> + heartbeat <<= WDT_PRELOAD_SHIFT; >>>> + heartbeat /= wdt->freq; >>>> + if (time_left < heartbeat / 2) >>>> + wdd->min_hw_heartbeat_ms = 0; >>>> + else >>>> + wdd->min_hw_heartbeat_ms = >>>> + (time_left - heartbeat / 2 + 1) * 1000; >>>> + >>>> + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; >>>> + } >>>> + >>>> ret = watchdog_register_device(wdd); >>>> if (ret) { >>>> dev_err(dev, "cannot register watchdog device\n"); >>>> >>> >>> This assumes that the bootloader also programmed a 50% window, right? The pending U-Boot patch will do that, but what if that may chance or someone uses a different setup? >> >> Yes, we assume 50%. I think based on the hw design, 50% is the only sane value to be used, otherwise you just shrink the open window too much and for no apparent reason. >> > > Not sure if that is a valid assumption. Someone who designs a watchdog > with such a narrow ping window might as well also use it. The question > is if you want to rely on that assumption, or check and change it if needed. Right, if that is a blocker, I can modify the code. Should be maybe couple of lines addition. > Also, I wonder if we should add an API function such as > "set_last_hw_keepalive()" to avoid all that complexity. I can try adding that also if it is desirable. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-25 17:04 ` Tero Kristo @ 2020-06-30 20:23 ` Guenter Roeck 2020-07-01 5:50 ` Tero Kristo 0 siblings, 1 reply; 13+ messages in thread From: Guenter Roeck @ 2020-06-30 20:23 UTC (permalink / raw) To: Tero Kristo; +Cc: Jan Kiszka, wim, linux-watchdog, linux-kernel On Thu, Jun 25, 2020 at 08:04:50PM +0300, Tero Kristo wrote: > On 25/06/2020 16:35, Guenter Roeck wrote: > > On 6/25/20 1:32 AM, Tero Kristo wrote: > > > On 24/06/2020 18:24, Jan Kiszka wrote: > > > > On 24.06.20 13:45, Tero Kristo wrote: > > > > > If the RTI watchdog has been started by someone (like bootloader) when > > > > > the driver probes, we must adjust the initial ping timeout to match the > > > > > currently running watchdog window to avoid generating watchdog reset. > > > > > > > > > > Signed-off-by: Tero Kristo <t-kristo@ti.com> > > > > > --- > > > > > drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ > > > > > 1 file changed, 25 insertions(+) > > > > > > > > > > diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c > > > > > index d456dd72d99a..02ea2b2435f5 100644 > > > > > --- a/drivers/watchdog/rti_wdt.c > > > > > +++ b/drivers/watchdog/rti_wdt.c > > > > > @@ -55,11 +55,13 @@ static int heartbeat; > > > > > * @base - base io address of WD device > > > > > * @freq - source clock frequency of WDT > > > > > * @wdd - hold watchdog device as is in WDT core > > > > > + * @min_hw_heartbeat_save - save of the min hw heartbeat value > > > > > */ > > > > > struct rti_wdt_device { > > > > > void __iomem *base; > > > > > unsigned long freq; > > > > > struct watchdog_device wdd; > > > > > + unsigned int min_hw_heartbeat_save; > > > > > }; > > > > > static int rti_wdt_start(struct watchdog_device *wdd) > > > > > @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) > > > > > /* put watchdog in active state */ > > > > > writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); > > > > > + if (wdt->min_hw_heartbeat_save) { > > > > > + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; > > > > > + wdt->min_hw_heartbeat_save = 0; > > > > > + } > > > > > + > > > > > return 0; > > > > > } > > > > > @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) > > > > > goto err_iomap; > > > > > } > > > > > + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { > > > > > + u32 time_left; > > > > > + u32 heartbeat; > > > > > + > > > > > + set_bit(WDOG_HW_RUNNING, &wdd->status); > > > > > + time_left = rti_wdt_get_timeleft(wdd); > > > > > + heartbeat = readl(wdt->base + RTIDWDPRLD); > > > > > + heartbeat <<= WDT_PRELOAD_SHIFT; > > > > > + heartbeat /= wdt->freq; > > > > > + if (time_left < heartbeat / 2) > > > > > + wdd->min_hw_heartbeat_ms = 0; > > > > > + else > > > > > + wdd->min_hw_heartbeat_ms = > > > > > + (time_left - heartbeat / 2 + 1) * 1000; > > > > > + > > > > > + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; > > > > > + } > > > > > + > > > > > ret = watchdog_register_device(wdd); > > > > > if (ret) { > > > > > dev_err(dev, "cannot register watchdog device\n"); > > > > > > > > > > > > > This assumes that the bootloader also programmed a 50% window, right? The pending U-Boot patch will do that, but what if that may chance or someone uses a different setup? > > > > > > Yes, we assume 50%. I think based on the hw design, 50% is the only sane value to be used, otherwise you just shrink the open window too much and for no apparent reason. > > > > > > > Not sure if that is a valid assumption. Someone who designs a watchdog > > with such a narrow ping window might as well also use it. The question > > is if you want to rely on that assumption, or check and change it if needed. > > Right, if that is a blocker, I can modify the code. Should be maybe couple > of lines addition. > > > Also, I wonder if we should add an API function such as > > "set_last_hw_keepalive()" to avoid all that complexity. > > I can try adding that also if it is desirable. > But wait, the code doesn't really match what the description of this patch claims, or at least the description is misleading. Per the description, this is to prevent an early timeout. However, the problem here is that the watchdog core does not generate a ping, even if requested, because it believes that it just generated one right before the watchdog timer was registered, and that it can not generate another one because min_hw_heartbeat_ms has not elapsed. With that in mind, the problem is a bit more complex. First, the driver doesn't really update the current timeout to the value that is currently configured and enabled. Instead, it just uses/assumes the default (DEFAULT_HEARTBEAT or whatever the heartbeat module parameter is set to). This means that it is still possible for an early timeout to occur if there is a mismatch between the bootloader timeout and the timeout assumed by the driver. Worse, the timeout is only updated in the start function - and the start function isn't called if the watchdog is already running. Actually, the driver does not support updating the timeout at all. This means that a mismatch between the bootloader timeout and the timeout assumed by the driver is not handled well. To solve this, the driver would have to update the actual timeout to whatever is programmed into the chip and ignore any module parameter and default settings if the watchdog is already running. Alternatively, it would have to support updating the timeout (if the hardware supports that) after the watchdog was started. Second, handling min_hw_heartbeat_ms properly should really be implemented in the watchdog core. Instead of assuming that the most recent keepalive happened "just before now", as it currently does, it should call the timeleft function (if available and if the watchdog is running) and calculate the most recent keepalive (and thus the earliest acceptable next keepalive) from its return value. Thanks, Guenter ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-06-30 20:23 ` Guenter Roeck @ 2020-07-01 5:50 ` Tero Kristo 2020-07-01 13:34 ` Guenter Roeck 0 siblings, 1 reply; 13+ messages in thread From: Tero Kristo @ 2020-07-01 5:50 UTC (permalink / raw) To: Guenter Roeck; +Cc: Jan Kiszka, wim, linux-watchdog, linux-kernel On 30/06/2020 23:23, Guenter Roeck wrote: > On Thu, Jun 25, 2020 at 08:04:50PM +0300, Tero Kristo wrote: >> On 25/06/2020 16:35, Guenter Roeck wrote: >>> On 6/25/20 1:32 AM, Tero Kristo wrote: >>>> On 24/06/2020 18:24, Jan Kiszka wrote: >>>>> On 24.06.20 13:45, Tero Kristo wrote: >>>>>> If the RTI watchdog has been started by someone (like bootloader) when >>>>>> the driver probes, we must adjust the initial ping timeout to match the >>>>>> currently running watchdog window to avoid generating watchdog reset. >>>>>> >>>>>> Signed-off-by: Tero Kristo <t-kristo@ti.com> >>>>>> --- >>>>>> drivers/watchdog/rti_wdt.c | 25 +++++++++++++++++++++++++ >>>>>> 1 file changed, 25 insertions(+) >>>>>> >>>>>> diff --git a/drivers/watchdog/rti_wdt.c b/drivers/watchdog/rti_wdt.c >>>>>> index d456dd72d99a..02ea2b2435f5 100644 >>>>>> --- a/drivers/watchdog/rti_wdt.c >>>>>> +++ b/drivers/watchdog/rti_wdt.c >>>>>> @@ -55,11 +55,13 @@ static int heartbeat; >>>>>> * @base - base io address of WD device >>>>>> * @freq - source clock frequency of WDT >>>>>> * @wdd - hold watchdog device as is in WDT core >>>>>> + * @min_hw_heartbeat_save - save of the min hw heartbeat value >>>>>> */ >>>>>> struct rti_wdt_device { >>>>>> void __iomem *base; >>>>>> unsigned long freq; >>>>>> struct watchdog_device wdd; >>>>>> + unsigned int min_hw_heartbeat_save; >>>>>> }; >>>>>> static int rti_wdt_start(struct watchdog_device *wdd) >>>>>> @@ -107,6 +109,11 @@ static int rti_wdt_ping(struct watchdog_device *wdd) >>>>>> /* put watchdog in active state */ >>>>>> writel_relaxed(WDKEY_SEQ1, wdt->base + RTIWDKEY); >>>>>> + if (wdt->min_hw_heartbeat_save) { >>>>>> + wdd->min_hw_heartbeat_ms = wdt->min_hw_heartbeat_save; >>>>>> + wdt->min_hw_heartbeat_save = 0; >>>>>> + } >>>>>> + >>>>>> return 0; >>>>>> } >>>>>> @@ -201,6 +208,24 @@ static int rti_wdt_probe(struct platform_device *pdev) >>>>>> goto err_iomap; >>>>>> } >>>>>> + if (readl(wdt->base + RTIDWDCTRL) == WDENABLE_KEY) { >>>>>> + u32 time_left; >>>>>> + u32 heartbeat; >>>>>> + >>>>>> + set_bit(WDOG_HW_RUNNING, &wdd->status); >>>>>> + time_left = rti_wdt_get_timeleft(wdd); >>>>>> + heartbeat = readl(wdt->base + RTIDWDPRLD); >>>>>> + heartbeat <<= WDT_PRELOAD_SHIFT; >>>>>> + heartbeat /= wdt->freq; >>>>>> + if (time_left < heartbeat / 2) >>>>>> + wdd->min_hw_heartbeat_ms = 0; >>>>>> + else >>>>>> + wdd->min_hw_heartbeat_ms = >>>>>> + (time_left - heartbeat / 2 + 1) * 1000; >>>>>> + >>>>>> + wdt->min_hw_heartbeat_save = 11 * heartbeat * 1000 / 20; >>>>>> + } >>>>>> + >>>>>> ret = watchdog_register_device(wdd); >>>>>> if (ret) { >>>>>> dev_err(dev, "cannot register watchdog device\n"); >>>>>> >>>>> >>>>> This assumes that the bootloader also programmed a 50% window, right? The pending U-Boot patch will do that, but what if that may chance or someone uses a different setup? >>>> >>>> Yes, we assume 50%. I think based on the hw design, 50% is the only sane value to be used, otherwise you just shrink the open window too much and for no apparent reason. >>>> >>> >>> Not sure if that is a valid assumption. Someone who designs a watchdog >>> with such a narrow ping window might as well also use it. The question >>> is if you want to rely on that assumption, or check and change it if needed. >> >> Right, if that is a blocker, I can modify the code. Should be maybe couple >> of lines addition. >> >>> Also, I wonder if we should add an API function such as >>> "set_last_hw_keepalive()" to avoid all that complexity. >> >> I can try adding that also if it is desirable. >> > > But wait, the code doesn't really match what the description of this > patch claims, or at least the description is misleading. Per the > description, this is to prevent an early timeout. However, the problem > here is that the watchdog core does not generate a ping, even if > requested, because it believes that it just generated one right before > the watchdog timer was registered, and that it can not generate another > one because min_hw_heartbeat_ms has not elapsed. You are right. Maybe the patch description could use some more beef into it. > > With that in mind, the problem is a bit more complex. > > First, the driver doesn't really update the current timeout to the > value that is currently configured and enabled. Instead, it just > uses/assumes the default (DEFAULT_HEARTBEAT or whatever the heartbeat > module parameter is set to). This means that it is still possible for > an early timeout to occur if there is a mismatch between the bootloader > timeout and the timeout assumed by the driver. Worse, the timeout > is only updated in the start function - and the start function isn't > called if the watchdog is already running. Actually, the driver does > not support updating the timeout at all. This means that a mismatch > between the bootloader timeout and the timeout assumed by the driver > is not handled well. > > To solve this, the driver would have to update the actual timeout to > whatever is programmed into the chip and ignore any module parameter > and default settings if the watchdog is already running. Alternatively, > it would have to support updating the timeout (if the hardware supports > that) after the watchdog was started. Hardware supports changing the timeout value, however it only updates this during the next window (preload values are picked once user pings the watchdog.) > Second, handling min_hw_heartbeat_ms properly should really be implemented > in the watchdog core. Instead of assuming that the most recent keepalive > happened "just before now", as it currently does, it should call the > timeleft function (if available and if the watchdog is running) and > calculate the most recent keepalive (and thus the earliest acceptable > next keepalive) from its return value. Yes, it all becomes bit complex if we let the bootloader configure the values freely. Current bootloader implementation does not do this, as it is mostly a copy of the kernel driver. However, I can modify the kernel driver to take all this into account, even if the code becomes a bit more complex due to this. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-07-01 5:50 ` Tero Kristo @ 2020-07-01 13:34 ` Guenter Roeck 2020-07-01 14:36 ` Tero Kristo 0 siblings, 1 reply; 13+ messages in thread From: Guenter Roeck @ 2020-07-01 13:34 UTC (permalink / raw) To: Tero Kristo; +Cc: Jan Kiszka, wim, linux-watchdog, linux-kernel On 6/30/20 10:50 PM, Tero Kristo wrote: [ ... ] > Hardware supports changing the timeout value, however it only updates this during the next window (preload values are picked once user pings the watchdog.) > The current driver doesn't support or use that, though, since the start function is only called once to start the watchdog, and not at all if the watchdog is already running. So, if the bootloader sets the timeout to X, and the user sets a timeout of, say, X * 4, userspace will never ping the watchdog often enough. The driver will have to address that to support already-running watchdogs. Thanks, Guenter ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window 2020-07-01 13:34 ` Guenter Roeck @ 2020-07-01 14:36 ` Tero Kristo 0 siblings, 0 replies; 13+ messages in thread From: Tero Kristo @ 2020-07-01 14:36 UTC (permalink / raw) To: Guenter Roeck; +Cc: Jan Kiszka, wim, linux-watchdog, linux-kernel On 01/07/2020 16:34, Guenter Roeck wrote: > On 6/30/20 10:50 PM, Tero Kristo wrote: > [ ... ] > >> Hardware supports changing the timeout value, however it only updates this during the next window (preload values are picked once user pings the watchdog.) >> > The current driver doesn't support or use that, though, since the start > function is only called once to start the watchdog, and not at all if > the watchdog is already running. So, if the bootloader sets the timeout > to X, and the user sets a timeout of, say, X * 4, userspace will never > ping the watchdog often enough. The driver will have to address that > to support already-running watchdogs. Yeah, I will modify that. I think I will just prevent changing the timeout if watchdog has been enabled from boot, it is probably cleanest approach. Unless I happen to come up with some sane way to change it on fly. -Tero -- Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki ^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2020-07-01 14:48 UTC | newest] Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-06-24 11:45 [PATCH 0/2] watchdog: rti: adjust initial ping for attach Tero Kristo 2020-06-24 11:45 ` [PATCH 1/2] watchdog: use __watchdog_ping in startup Tero Kristo 2020-06-30 20:36 ` Guenter Roeck 2020-06-24 11:45 ` [PATCH 2/2] watchdog: rti: tweak min_hw_heartbeat_ms to match initial allowed window Tero Kristo 2020-06-24 15:24 ` Jan Kiszka 2020-06-25 8:32 ` Tero Kristo 2020-06-25 10:00 ` Jan Kiszka 2020-06-25 13:35 ` Guenter Roeck 2020-06-25 17:04 ` Tero Kristo 2020-06-30 20:23 ` Guenter Roeck 2020-07-01 5:50 ` Tero Kristo 2020-07-01 13:34 ` Guenter Roeck 2020-07-01 14:36 ` Tero Kristo
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).