Skip to contentSkip to conversation

Armada 370/XP watchdog

PATCH Superseded Ezequiel Garcia 5 participants 40 messages Feb 6, 2014 – Feb 19, 2014

Subsystems (6)

v6 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

v6 cover Armada 370/XP watchdog

A new round, adding all Tested,Acked-by's and fixing Russell's comment about the leaking clock.

This series depends on Sebastian's irqchip fixes to clear bridge irqs cause, Jason has picked it here:

http://git.infradead.org/linux-mvebu.git mvebu-next/irqchip-fixes

The first patch (for clocksource) is independent of the others.
Daniel: Will you take it?

Wim: Can we take all the watchdog changes through mvebu, with your Ack? This way, we can sort out the irqchip dependency.

I've pushed a branch, in case anybody feels like doing a test:

https://github.com/MISL-EBU-System-SW/mainline-public/tree/wdt_for_v3.14_v6

Ezequiel Garcia (19):
clocksource: orion: Use atomic access for shared registers
watchdog: orion: Add clock error handling
watchdog: orion: Use atomic access for shared registers
watchdog: orion: Remove unused macros
watchdog: orion: Make sure the watchdog is initially stopped
watchdog: orion: Handle the interrupt so it's properly acked
watchdog: orion: Make RSTOUT register a separate resource
watchdog: orion: Remove unneeded BRIDGE_CAUSE clear
watchdog: orion: Introduce an orion_watchdog device structure
watchdog: orion: Introduce per-compatible of_device_id data
watchdog: orion: Add per-compatible clock initialization
watchdog: orion: Add per-compatible watchdog start implementation
watchdog: orion: Add support for Armada 370 and Armada XP SoC
ARM: mvebu: Enable Armada 370/XP watchdog in the devicetree
ARM: kirkwood: Add RSTOUT 'reg' entry to devicetree
ARM: dove: Enable Dove watchdog in the devicetree
watchdog: orion: Enable the build on ARCH_MVEBU
ARM: mvebu: Enable watchdog support in defconfig
ARM: dove: Enable watchdog support in the defconfig

 .../devicetree/bindings/watchdog/marvel.txt        |  11 +-
 arch/arm/boot/dts/armada-370-xp.dtsi               |   4 +
 arch/arm/boot/dts/armada-370.dtsi                  |   5 +
 arch/arm/boot/dts/armada-xp.dtsi                   |   6 +
 arch/arm/boot/dts/dove.dtsi                        |   8 +
 arch/arm/boot/dts/kirkwood.dtsi                    |   2 +-
 arch/arm/configs/dove_defconfig                    |   2 +
 arch/arm/configs/mvebu_defconfig                   |   2 +
 arch/arm/mach-dove/include/mach/bridge-regs.h      |   1 +
 arch/arm/mach-kirkwood/include/mach/bridge-regs.h  |   1 +
 arch/arm/mach-mv78xx0/include/mach/bridge-regs.h   |   1 +
 arch/arm/mach-orion5x/include/mach/bridge-regs.h   |   1 +
 arch/arm/plat-orion/common.c                       |  10 +-
 drivers/clocksource/time-orion.c                   |  28 +-
 drivers/watchdog/Kconfig                           |   2 +-
 drivers/watchdog/orion_wdt.c                       | 367 ++++++++++++++++-----
 16 files changed, 345 insertions(+), 106 deletions(-)
Signature · 1 line
1.8.1.5
#1 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

---
 drivers/clocksource/time-orion.c | 28 ++++++++++------------------
 1 file changed, 10 insertions(+), 18 deletions(-)
drivers/clocksource/time-orion.c +10 -18
index 2006622..0b3ce03 100644
@@ -35,20 +35,6 @@
 #define ORION_ONESHOT_MAX	0xfffffffe
 
 static void __iomem *timer_base;
-static DEFINE_SPINLOCK(timer_ctrl_lock);
-
-/*
- * Thread-safe access to TIMER_CTRL register
- * (shared with watchdog timer)
- */
-void orion_timer_ctrl_clrset(u32 clr, u32 set)
-{
-	spin_lock(&timer_ctrl_lock);
-	writel((readl(timer_base + TIMER_CTRL) & ~clr) | set,
-		timer_base + TIMER_CTRL);
-	spin_unlock(&timer_ctrl_lock);
-}
-EXPORT_SYMBOL(orion_timer_ctrl_clrset);
 
 /*
  * Free-running clocksource handling.
@@ -68,7 +54,8 @@ static int orion_clkevt_next_event(unsigned long delta,
 {
 	/* setup and enable one-shot timer */
 	writel(delta, timer_base + TIMER1_VAL);
-	orion_timer_ctrl_clrset(TIMER1_RELOAD_EN, TIMER1_EN);
+	atomic_io_modify(timer_base + TIMER_CTRL,
+		TIMER1_RELOAD_EN | TIMER1_EN, TIMER1_EN);
 
 	return 0;
 }
@@ -80,10 +67,13 @@ static void orion_clkevt_mode(enum clock_event_mode mode,
 		/* setup and enable periodic timer at 1/HZ intervals */
 		writel(ticks_per_jiffy - 1, timer_base + TIMER1_RELOAD);
 		writel(ticks_per_jiffy - 1, timer_base + TIMER1_VAL);
-		orion_timer_ctrl_clrset(0, TIMER1_RELOAD_EN | TIMER1_EN);
+		atomic_io_modify(timer_base + TIMER_CTRL,
+			TIMER1_RELOAD_EN | TIMER1_EN,
+			TIMER1_RELOAD_EN | TIMER1_EN);
 	} else {
 		/* disable timer */
-		orion_timer_ctrl_clrset(TIMER1_RELOAD_EN | TIMER1_EN, 0);
+		atomic_io_modify(timer_base + TIMER_CTRL,
+			TIMER1_RELOAD_EN | TIMER1_EN, 0);
 	}
 }
 
@@ -131,7 +121,9 @@ static void __init orion_timer_init(struct device_node *np)
 	/* setup timer0 as free-running clocksource */
 	writel(~0, timer_base + TIMER0_VAL);
 	writel(~0, timer_base + TIMER0_RELOAD);
-	orion_timer_ctrl_clrset(0, TIMER0_RELOAD_EN | TIMER0_EN);
+	atomic_io_modify(timer_base + TIMER_CTRL,
+		TIMER0_RELOAD_EN | TIMER0_EN,
+		TIMER0_RELOAD_EN | TIMER0_EN);
 	clocksource_mmio_init(timer_base + TIMER0_VAL, "orion_clocksource",
 			      clk_get_rate(clk), 300, 32,
 			      clocksource_mmio_readl_down);
Signature · 1 line
1.8.1.5
#2 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Quoted 8 lines · Ezequiel Garcia

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

The patch looks good for me.

Signature · 5 lines
  <http://www.linaro.org/> Linaro.org ? Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
#3 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Quoted 10 lines · Daniel Lezcano
Quoted 8 lines · Ezequiel Garcia

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

Quoted 4 lines

The patch looks good for me.

Daniel,

Jason acked this patch for you to take it. Or do you prefer that we merge it with the rest of the watchdog series?

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#4 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Quoted 13 lines · Ezequiel Garcia
Quoted 12 lines · Daniel Lezcano
Quoted 8 lines · Ezequiel Garcia

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

The patch looks good for me.

Quoted 7 lines
Quoted 2 lines

Daniel,

Jason acked this patch for you to take it. Or do you prefer that we merge it with the rest of the watchdog series?

I can pick it, but doesn't the watchdog series depend on it ?

Signature · 5 lines
  <http://www.linaro.org/> Linaro.org ? Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
#5 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Quoted 17 lines · Daniel Lezcano
Quoted 16 lines · Ezequiel Garcia
Quoted 14 lines · Daniel Lezcano
Quoted 8 lines · Ezequiel Garcia

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

The patch looks good for me.

Quoted 7 lines
Quoted 4 lines

Daniel,

Jason acked this patch for you to take it. Or do you prefer that we merge it with the rest of the watchdog series?

I can pick it, but doesn't the watchdog series depend on it ?

Nope. It's indepedent of it.

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#6 lore raw

PATCH v6 1/19 clocksource: orion: Use atomic access for shared registers

Quoted 20 lines · Ezequiel Garcia
Quoted 19 lines · Daniel Lezcano
Quoted 18 lines · Ezequiel Garcia
Quoted 10 lines · Daniel Lezcano
Quoted 8 lines · Ezequiel Garcia

Replace the driver-specific thread-safe shared register API by the recently introduced atomic_io_clear_set().

Quoted 4 lines

The patch looks good for me.

Daniel,

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

Jason acked this patch for you to take it. Or do you prefer that we merge it with the rest of the watchdog series?

I can pick it, but doesn't the watchdog series depend on it ?

Nope. It's indepedent of it.

Daniel,

Ping? This patch is independent of the watchdog series, and there won't be any conflicts with it, so I think it's safe for you to take it.

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#7 lore raw

PATCH v6 2/19 watchdog: orion: Add clock error handling

In reply to Ezequiel Garcia

This commit adds a check for clk_prepare_enable success and introduces an error path to disable the clock properly.

---
 drivers/watchdog/orion_wdt.c | 29 +++++++++++++++++++----------
 1 file changed, 19 insertions(+), 10 deletions(-)
drivers/watchdog/orion_wdt.c +19 -10
index f7722a4..7f19fa3 100644
@@ -151,17 +151,24 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(clk)) {
 		dev_err(&pdev->dev, "Orion Watchdog missing clock\n");
-		return -ENODEV;
+		return PTR_ERR(clk);
 	}
-	clk_prepare_enable(clk);
+	ret = clk_prepare_enable(clk);
+	if (ret)
+		return ret;
 	wdt_tclk = clk_get_rate(clk);
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res)
-		return -ENODEV;
+	if (!res) {
+		ret = -ENODEV;
+		goto disable_clk;
+	}
+
 	wdt_reg = devm_ioremap(&pdev->dev, res->start, resource_size(res));
-	if (!wdt_reg)
-		return -ENOMEM;
+	if (!wdt_reg) {
+		ret = -ENOMEM;
+		goto disable_clk;
+	}
 
 	wdt_max_duration = WDT_MAX_CYCLE_COUNT / wdt_tclk;
 
@@ -171,14 +178,16 @@ static int orion_wdt_probe(struct platform_device *pdev)
 
 	watchdog_set_nowayout(&orion_wdt, nowayout);
 	ret = watchdog_register_device(&orion_wdt);
-	if (ret) {
-		clk_disable_unprepare(clk);
-		return ret;
-	}
+	if (ret)
+		goto disable_clk;
 
 	pr_info("Initial timeout %d sec%s\n",
 		orion_wdt.timeout, nowayout ? ", nowayout" : "");
 	return 0;
+
+disable_clk:
+	clk_disable_unprepare(clk);
+	return ret;
 }
 
 static int orion_wdt_remove(struct platform_device *pdev)
Signature · 1 line
1.8.1.5
#8 lore raw

PATCH v6 3/19 watchdog: orion: Use atomic access for shared registers

In reply to Ezequiel Garcia

Since the timer control register is shared with the clocksource driver, use the recently introduced atomic_io_clear_set() to access such register. Given the watchdog core already provides serialization for all the watchdog ops, this commit allows to remove the spinlock entirely.

---
 drivers/watchdog/orion_wdt.c | 42 +++++-------------------------------------
 1 file changed, 5 insertions(+), 37 deletions(-)
drivers/watchdog/orion_wdt.c +5 -37
index 7f19fa3..b92a991 100644
@@ -20,7 +20,6 @@
 #include <linux/watchdog.h>
 #include <linux/init.h>
 #include <linux/io.h>
-#include <linux/spinlock.h>
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/of.h>
@@ -46,25 +45,16 @@ static unsigned int wdt_max_duration;	/* (seconds) */
 static struct clk *clk;
 static unsigned int wdt_tclk;
 static void __iomem *wdt_reg;
-static DEFINE_SPINLOCK(wdt_lock);
 
 static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 {
-	spin_lock(&wdt_lock);
-
 	/* Reload watchdog duration */
 	writel(wdt_tclk * wdt_dev->timeout, wdt_reg + WDT_VAL);
-
-	spin_unlock(&wdt_lock);
 	return 0;
 }
 
 static int orion_wdt_start(struct watchdog_device *wdt_dev)
 {
-	u32 reg;
-
-	spin_lock(&wdt_lock);
-
 	/* Set watchdog duration */
 	writel(wdt_tclk * wdt_dev->timeout, wdt_reg + WDT_VAL);
 
@@ -72,48 +62,26 @@ static int orion_wdt_start(struct watchdog_device *wdt_dev)
 	writel(~WDT_INT_REQ, BRIDGE_CAUSE);
 
 	/* Enable watchdog timer */
-	reg = readl(wdt_reg + TIMER_CTRL);
-	reg |= WDT_EN;
-	writel(reg, wdt_reg + TIMER_CTRL);
+	atomic_io_modify(wdt_reg + TIMER_CTRL, WDT_EN, WDT_EN);
 
 	/* Enable reset on watchdog */
-	reg = readl(RSTOUTn_MASK);
-	reg |= WDT_RESET_OUT_EN;
-	writel(reg, RSTOUTn_MASK);
-
-	spin_unlock(&wdt_lock);
+	atomic_io_modify(RSTOUTn_MASK, WDT_RESET_OUT_EN, WDT_RESET_OUT_EN);
 	return 0;
 }
 
 static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 {
-	u32 reg;
-
-	spin_lock(&wdt_lock);
-
 	/* Disable reset on watchdog */
-	reg = readl(RSTOUTn_MASK);
-	reg &= ~WDT_RESET_OUT_EN;
-	writel(reg, RSTOUTn_MASK);
+	atomic_io_modify(RSTOUTn_MASK, WDT_RESET_OUT_EN, 0);
 
 	/* Disable watchdog timer */
-	reg = readl(wdt_reg + TIMER_CTRL);
-	reg &= ~WDT_EN;
-	writel(reg, wdt_reg + TIMER_CTRL);
-
-	spin_unlock(&wdt_lock);
+	atomic_io_modify(wdt_reg + TIMER_CTRL, WDT_EN, 0);
 	return 0;
 }
 
 static unsigned int orion_wdt_get_timeleft(struct watchdog_device *wdt_dev)
 {
-	unsigned int time_left;
-
-	spin_lock(&wdt_lock);
-	time_left = readl(wdt_reg + WDT_VAL) / wdt_tclk;
-	spin_unlock(&wdt_lock);
-
-	return time_left;
+	return readl(wdt_reg + WDT_VAL) / wdt_tclk;
 }
 
 static int orion_wdt_set_timeout(struct watchdog_device *wdt_dev,
Signature · 1 line
1.8.1.5
#9 lore raw

PATCH v6 4/19 watchdog: orion: Remove unused macros

In reply to Ezequiel Garcia

These are not used anywhere so it's safe to remove them.

---
 drivers/watchdog/orion_wdt.c | 2 --
 1 file changed, 2 deletions(-)
drivers/watchdog/orion_wdt.c +0 -2
index b92a991..6746033 100644
@@ -33,8 +33,6 @@
 #define WDT_VAL			0x0024
 
 #define WDT_MAX_CYCLE_COUNT	0xffffffff
-#define WDT_IN_USE		0
-#define WDT_OK_TO_CLOSE		1
 
 #define WDT_RESET_OUT_EN	BIT(1)
 #define WDT_INT_REQ		BIT(3)
Signature · 1 line
1.8.1.5
#10 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

In reply to Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
 drivers/watchdog/orion_wdt.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +3 -0
index 6746033..2dbeee9 100644
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	orion_wdt.max_timeout = wdt_max_duration;
 	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+
 	watchdog_set_nowayout(&orion_wdt, nowayout);
 	ret = watchdog_register_device(&orion_wdt);
 	if (ret)
Signature · 1 line
1.8.1.5
maintainer
#11 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
  drivers/watchdog/orion_wdt.c | 3 +++
  1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
  	orion_wdt.max_timeout = wdt_max_duration;
  	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Thanks,
Guenter

Quoted 4 lines
drivers/watchdog/orion_wdt.c +0 -0
  	watchdog_set_nowayout(&orion_wdt, nowayout);
  	ret = watchdog_register_device(&orion_wdt);
  	if (ret)
#12 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 27 lines · Guenter Roeck
Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
  drivers/watchdog/orion_wdt.c | 3 +++
  1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
  	orion_wdt.max_timeout = wdt_max_duration;
  	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Quoted 4 lines

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

Hm... I see.

Quoted 3 lines

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The watchdog stop was agreed specifically [2].

Ideas?

[1] http://www.spinics.net/lists/arm-kernel/msg302340.html
[2] http://www.spinics.net/lists/arm-kernel/msg302507.html

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
maintainer
#13 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 48 lines · Ezequiel Garcia
Quoted 27 lines · Guenter Roeck
Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
   drivers/watchdog/orion_wdt.c | 3 +++
   1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
   	orion_wdt.max_timeout = wdt_max_duration;
   	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Quoted 4 lines

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

Hm... I see.

Quoted 3 lines

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

Quoted 6 lines

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The watchdog stop was agreed specifically [2].

Ideas?

Other drivers assume that if the watchdog is running, it is supposed to be running. The more common approach in such cases is to ping the watchdog once to give userspace more time to get ready, but leave it enabled. So you could check if the watchdog is enabled, and if it was enabled re-enable it after initialization is complete (and maybe log a message stating that the watchdog is enabled).

If you don't want to do that, and if you are defeating nowayout on purpose to fix a problem with a broken bootloader,
you should at least put in comment describing the problem you are trying to solve, and that you accept breaking nowayout with your fix.

Thanks,
Guenter

#14 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 63 lines · Guenter Roeck
Quoted 48 lines · Ezequiel Garcia
Quoted 27 lines · Guenter Roeck
Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
   drivers/watchdog/orion_wdt.c | 3 +++
   1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
   	orion_wdt.max_timeout = wdt_max_duration;
   	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Quoted 4 lines

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

Hm... I see.

Quoted 3 lines

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

Quoted 6 lines

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The watchdog stop was agreed specifically [2].

Ideas?

Other drivers assume that if the watchdog is running, it is supposed to be running. The more common approach in such cases is to ping the watchdog once to give userspace more time to get ready, but leave it enabled. So you could check if the watchdog is enabled, and if it was enabled re-enable it after initialization is complete (and maybe log a message stating that the watchdog is enabled).

Quoted 5 lines

If you don't want to do that, and if you are defeating nowayout on purpose to fix a problem with a broken bootloader,
you should at least put in comment describing the problem you are trying to solve, and that you accept breaking nowayout with your fix.

I'm not fond of not having "nowayout" option on our driver, given I'm sure it's a watchdog feature for a good reason.

On the other side, I can't see how can we distinguish a previously and explicitly enabled watchdog, from a spurious enable by broken bootloader.

Jason, what do you say?

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#15 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 30 lines · Ezequiel Garcia
Quoted 29 lines · Guenter Roeck
Quoted 28 lines · Ezequiel Garcia
Quoted 27 lines · Guenter Roeck
Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
   drivers/watchdog/orion_wdt.c | 3 +++
   1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
   	orion_wdt.max_timeout = wdt_max_duration;
   	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

How often would a user legitimately want to unload/load the watchdog module?

Quoted 30 lines
Quoted 30 lines
Quoted 16 lines

Hm... I see.

Quoted 3 lines

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

Quoted 6 lines

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The watchdog stop was agreed specifically [2].

Ideas?

Other drivers assume that if the watchdog is running, it is supposed to be running. The more common approach in such cases is to ping the watchdog once to give userspace more time to get ready, but leave it enabled. So you could check if the watchdog is enabled, and if it was enabled re-enable it after initialization is complete (and maybe log a message stating that the watchdog is enabled).

Quoted 4 lines
Quoted 4 lines

If you don't want to do that, and if you are defeating nowayout on purpose to fix a problem with a broken bootloader,
you should at least put in comment describing the problem you are trying to solve, and that you accept breaking nowayout with your fix.

Yes, this should be commented.

Quoted 5 lines

I'm not fond of not having "nowayout" option on our driver, given I'm sure it's a watchdog feature for a good reason.

On the other side, I can't see how can we distinguish a previously and explicitly enabled watchdog, from a spurious enable by broken bootloader.

How about we just don't define module_exit() and leave a comment as such? It's not unprecedented, a couple of the atm drivers are explicitly setup like this (uPD98402.c, zatm.c, eni.c).

thx,

Jason.

maintainer
#16 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 31 lines · Jason Cooper
Quoted 30 lines · Ezequiel Garcia
Quoted 29 lines · Guenter Roeck
Quoted 28 lines · Ezequiel Garcia
Quoted 27 lines · Guenter Roeck
Quoted 17 lines · Ezequiel Garcia

Having the watchdog initially fully stopped is important to avoid any spurious watchdog triggers, in case the registers are not in its reset state.

---
   drivers/watchdog/orion_wdt.c | 3 +++
   1 file changed, 3 insertions(+)
drivers/watchdog/orion_wdt.c +0 -0
index 6746033..2dbeee9 100644

drivers/watchdog/orion_wdt.c static int orion_wdt_probe(struct platform_device *pdev) · line 147

Quoted 6 lines
drivers/watchdog/orion_wdt.c +3 -0
@@ -142,6 +142,9 @@ static int orion_wdt_probe(struct platform_device *pdev)
   	orion_wdt.max_timeout = wdt_max_duration;
   	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
 
+	/* Let's make sure the watchdog is fully stopped */
+	orion_wdt_stop(&orion_wdt);
+

Actually we just had that in another driver, and I stumbled over it there.

Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

Problem with stopping the watchdog in probe unconditionally is that you can use it to defeat nowayout: unload the module, then load it again, and the watchdog is stopped even if nowayout is true.

How often would a user legitimately want to unload/load the watchdog module?

Not sure what you are saying here. If you don't think the nowayout option should be supported, drop it. Don't claim it is supported when it isn't.

Quoted 40 lines
Quoted 30 lines
Quoted 30 lines
Quoted 16 lines

Hm... I see.

Quoted 3 lines

Is this really what you want ? Or, in other words, what is the problem you are trying to solve ?

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

Quoted 6 lines

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The watchdog stop was agreed specifically [2].

Ideas?

Other drivers assume that if the watchdog is running, it is supposed to be running. The more common approach in such cases is to ping the watchdog once to give userspace more time to get ready, but leave it enabled. So you could check if the watchdog is enabled, and if it was enabled re-enable it after initialization is complete (and maybe log a message stating that the watchdog is enabled).

Quoted 4 lines
Quoted 4 lines

If you don't want to do that, and if you are defeating nowayout on purpose to fix a problem with a broken bootloader,
you should at least put in comment describing the problem you are trying to solve, and that you accept breaking nowayout with your fix.

Yes, this should be commented.

Quoted 3 lines

I'm not fond of not having "nowayout" option on our driver, given I'm sure it's a watchdog feature for a good reason.

Quoted 7 lines
Quoted 2 lines

On the other side, I can't see how can we distinguish a previously and explicitly enabled watchdog, from a spurious enable by broken bootloader.

How about we just don't define module_exit() and leave a comment as such? It's not unprecedented, a couple of the atm drivers are explicitly setup like this (uPD98402.c, zatm.c, eni.c).

Or don't support tristate in the first place. There was some argument from others earlier that a watchdog should never be optional. Or drop the nowayout option from the driver.

There is one problem, though, if you don't support a pre-enabled watchdog:
If the system dies before the watchdog application starts running, it will hang. This is the reason why the watchdog is enabled on purpose by some boot loaders.

Guenter

#17 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

In reply to Ezequiel Garcia
Quoted 10 lines · Ezequiel Garcia

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The issue here is the driver configures two 'machine kill' elements:
the PANIC IRQ and the RstOut setup.

Before configuring either of those the driver needs to ensure that any old watchdog events are cleared out of the HW. We must not get a spurious event.

I agree not disabling an already functional and properly configured counter from the bootloader is desirable.

So lets break it down a bit..

1) The IRQ:
It looks like the cause bit latches high on watchdog timer expiration but has no side effect unless it is unmasked.

The new IRQ flow code ensures the bit is cleared during request_irq so no old events can trigger the IRQ. Thus it is solved now.

2) The RstOut:
It is a bit unspecific, but I think this signal latches high when any unmasked rst event occurs and stays high until the RstIn pin is asserted.

So we don't need to worry about clearing old events from here

Alternatively the WDRstOutEn routes the Cause bit into the RstOut pin.

3) The timer itself:
The WDT is just a general timer with an optional hookup to the rst control. If it is harmlessly counting but not resetting we need to stop that before enabling rst out.

So, how about this for psuedo-code in probe:

if (readl(RSTOUTn) & WDRstOutEn)

{
    /* Watchdog is configured and may be down counting,
       don't touch it */
    request_irq(..);
}
else
{
    /* Watchdog is not configured, fully disable the timer
       and configure for watchdog operation. */
    disable_watchdog();
    request_irq();
    writel(RSTOUTn), .. WDRstOutEn);
}

Jason

#18 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 23 lines · Jason Gunthorpe
Quoted 10 lines · Ezequiel Garcia

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The issue here is the driver configures two 'machine kill' elements:
the PANIC IRQ and the RstOut setup.

Before configuring either of those the driver needs to ensure that any old watchdog events are cleared out of the HW. We must not get a spurious event.

I agree not disabling an already functional and properly configured counter from the bootloader is desirable.

Quoted 9 lines

So lets break it down a bit..

1) The IRQ:
It looks like the cause bit latches high on watchdog timer expiration but has no side effect unless it is unmasked.

The new IRQ flow code ensures the bit is cleared during request_irq so no old events can trigger the IRQ. Thus it is solved now.

Agreed.

Quoted 5 lines

3) The timer itself:
The WDT is just a general timer with an optional hookup to the rst control. If it is harmlessly counting but not resetting we need to stop that before enabling rst out.

Actually, the current flow is to:

1. Disable rst out and then disable the counter, in probe().

2. Enable the counter, and then enable rst out, in start().

Quoted 17 lines

So, how about this for psuedo-code in probe:

if (readl(RSTOUTn) & WDRstOutEn)

{
    /* Watchdog is configured and may be down counting,
       don't touch it */
    request_irq(..);
}
else
{
    /* Watchdog is not configured, fully disable the timer
       and configure for watchdog operation. */
    disable_watchdog();
    request_irq();
    writel(RSTOUTn), .. WDRstOutEn);
}

Sounds good, although it seems to me it's actually simpler:

  /* Let's make sure the watchdog is fully stopped, unless
   * it's explicitly enabled and running
   */
  if ( !(wdt_rst_out_en && wdt_timer_enabled) ) {
    watchdog_stop();
  }
Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#19 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoting Ezequiel Garcia <[email protected]>:

Quoted 66 lines · Jason Gunthorpe
Quoted 23 lines · Ezequiel Garcia
Quoted 11 lines

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The issue here is the driver configures two 'machine kill' elements:
the PANIC IRQ and the RstOut setup.

Before configuring either of those the driver needs to ensure that any old watchdog events are cleared out of the HW. We must not get a spurious event.

I agree not disabling an already functional and properly configured counter from the bootloader is desirable.

Quoted 9 lines

So lets break it down a bit..

1) The IRQ:
It looks like the cause bit latches high on watchdog timer expiration but has no side effect unless it is unmasked.

The new IRQ flow code ensures the bit is cleared during request_irq so no old events can trigger the IRQ. Thus it is solved now.

Agreed.

Quoted 5 lines

3) The timer itself:
The WDT is just a general timer with an optional hookup to the rst control. If it is harmlessly counting but not resetting we need to stop that before enabling rst out.

Actually, the current flow is to:

1. Disable rst out and then disable the counter, in probe().

2. Enable the counter, and then enable rst out, in start().

Quoted 17 lines

So, how about this for psuedo-code in probe:

if (readl(RSTOUTn) & WDRstOutEn)

{
    /* Watchdog is configured and may be down counting,
       don't touch it */
    request_irq(..);
}
else
{
    /* Watchdog is not configured, fully disable the timer
       and configure for watchdog operation. */
    disable_watchdog();
    request_irq();
    writel(RSTOUTn), .. WDRstOutEn);
}
Quoted 9 lines

Sounds good, although it seems to me it's actually simpler:

  /* Let's make sure the watchdog is fully stopped, unless
   * it's explicitly enabled and running
   */
  if ( !(wdt_rst_out_en && wdt_timer_enabled) ) {
    watchdog_stop();
  }
     if (!wdt_rst_out_en || !wdt_timer_enabled)
         watchdog_stop();

seems to be a bit easier to understand.

Also watch out for coding style issues. Above code snippet has multiple violations ;-).

Guenter

#20 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoted 78 lines · linux at roeck-us.net

Quoting Ezequiel Garcia <[email protected]>:

Quoted 66 lines · Jason Gunthorpe
Quoted 23 lines · Ezequiel Garcia
Quoted 11 lines

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog stop for (1).

The issue here is the driver configures two 'machine kill' elements:
the PANIC IRQ and the RstOut setup.

Before configuring either of those the driver needs to ensure that any old watchdog events are cleared out of the HW. We must not get a spurious event.

I agree not disabling an already functional and properly configured counter from the bootloader is desirable.

Quoted 9 lines

So lets break it down a bit..

1) The IRQ:
It looks like the cause bit latches high on watchdog timer expiration but has no side effect unless it is unmasked.

The new IRQ flow code ensures the bit is cleared during request_irq so no old events can trigger the IRQ. Thus it is solved now.

Agreed.

Quoted 5 lines

3) The timer itself:
The WDT is just a general timer with an optional hookup to the rst control. If it is harmlessly counting but not resetting we need to stop that before enabling rst out.

Actually, the current flow is to:

1. Disable rst out and then disable the counter, in probe().

2. Enable the counter, and then enable rst out, in start().

Quoted 17 lines

So, how about this for psuedo-code in probe:

if (readl(RSTOUTn) & WDRstOutEn)

{
    /* Watchdog is configured and may be down counting,
       don't touch it */
    request_irq(..);
}
else
{
    /* Watchdog is not configured, fully disable the timer
       and configure for watchdog operation. */
    disable_watchdog();
    request_irq();
    writel(RSTOUTn), .. WDRstOutEn);
}
Quoted 9 lines

Sounds good, although it seems to me it's actually simpler:

  /* Let's make sure the watchdog is fully stopped, unless
   * it's explicitly enabled and running
   */
  if ( !(wdt_rst_out_en && wdt_timer_enabled) ) {
    watchdog_stop();
  }
Quoted 5 lines
     if (!wdt_rst_out_en || !wdt_timer_enabled)
         watchdog_stop();

seems to be a bit easier to understand.

Yeah, I was actually planning to have a orion_wdt_enabled() function to get the running and enabled status. Looks cleaner I think.

	if (!orion_wdt_enabled())
		watchdog_stop();

So the idea is OK?

I'll push the new series in a short while. Unfortunately, this change means I have to rebase almost all the series, because I introduce the orion watchdog struct :-(

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#21 lore raw

PATCH v6 5/19 watchdog: orion: Make sure the watchdog is initially stopped

Quoting Ezequiel Garcia <[email protected]>:

Quoted 86 lines · linux at roeck-us.net
Quoted 79 lines

Quoting Ezequiel Garcia <[email protected]>:

Quoted 6 lines · Jason Gunthorpe
Quoted 5 lines · Ezequiel Garcia
Quoted 4 lines

Well, this is related to the discussion about the bootloader not reseting the watchdog properly, provoking spurious watchdog triggering.

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

Jason Gunthorpe explained [1] that we needed a particular sequence:

1. Disable WDT
2. Clear bridge
3. Enable WDT

We added the irq handling to satisfy (2), and the watchdog

stop for (1).

Quoted 53 lines
Quoted 11 lines

The issue here is the driver configures two 'machine kill' elements:
the PANIC IRQ and the RstOut setup.

Before configuring either of those the driver needs to ensure that any old watchdog events are cleared out of the HW. We must not get a spurious event.

I agree not disabling an already functional and properly configured counter from the bootloader is desirable.

Quoted 9 lines

So lets break it down a bit..

1) The IRQ:
It looks like the cause bit latches high on watchdog timer expiration but has no side effect unless it is unmasked.

The new IRQ flow code ensures the bit is cleared during request_irq so no old events can trigger the IRQ. Thus it is solved now.

Agreed.

Quoted 5 lines

3) The timer itself:
The WDT is just a general timer with an optional hookup to the rst control. If it is harmlessly counting but not resetting we need to stop that before enabling rst out.

Actually, the current flow is to:

1. Disable rst out and then disable the counter, in probe().

2. Enable the counter, and then enable rst out, in start().

Quoted 17 lines

So, how about this for psuedo-code in probe:

if (readl(RSTOUTn) & WDRstOutEn)

{
    /* Watchdog is configured and may be down counting,
       don't touch it */
    request_irq(..);
}
else
{
    /* Watchdog is not configured, fully disable the timer
       and configure for watchdog operation. */
    disable_watchdog();
    request_irq();
    writel(RSTOUTn), .. WDRstOutEn);
}
Quoted 9 lines

Sounds good, although it seems to me it's actually simpler:

  /* Let's make sure the watchdog is fully stopped, unless
   * it's explicitly enabled and running
   */
  if ( !(wdt_rst_out_en && wdt_timer_enabled) ) {
    watchdog_stop();
  }
Quoted 5 lines
     if (!wdt_rst_out_en || !wdt_timer_enabled)
         watchdog_stop();

seems to be a bit easier to understand.

Quoted 8 lines

Yeah, I was actually planning to have a orion_wdt_enabled() function to get the running and enabled status. Looks cleaner I think.

	if (!orion_wdt_enabled())
		watchdog_stop();

So the idea is OK?

Sounds good to me.

Quoted 3 lines

I'll push the new series in a short while. Unfortunately, this change means I have to rebase almost all the series, because I introduce the orion watchdog struct :-(

Just trying to keep you busy :-)

Guenter

#22 lore raw

PATCH v6 6/19 watchdog: orion: Handle the interrupt so it's properly acked

In reply to Ezequiel Garcia

DT-enabled plaforms, where the irqchip driver for the brigde interrupt controller is available, can handle the watchdog IRQ properly. Therefore, request the interrupt and add a dummy handler that merely calls panic().

This is done in order to have an initial 'ack' of the interruption, which clears the watchdog state.

Furthermore, since some platforms don't have such IRQ, this commit makes the interrupt specification optional.

---
 .../devicetree/bindings/watchdog/marvel.txt        |  2 ++
 drivers/watchdog/orion_wdt.c                       | 24 +++++++++++++++++++++-
 2 files changed, 25 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/watchdog/marvel.txt +2 -0
index 5dc8d30..0731fbd 100644
@@ -7,6 +7,7 @@ Required Properties:
 
 Optional properties:
 
+- interrupts	: Contains the IRQ for watchdog expiration
 - timeout-sec	: Contains the watchdog timeout in seconds
 
 Example:
@@ -14,6 +15,7 @@ Example:
 	wdt at 20300 {
 		compatible = "marvell,orion-wdt";
 		reg = <0x20300 0x28>;
+		interrupts = <3>;
 		timeout-sec = <10>;
 		status = "okay";
 	};
drivers/watchdog/orion_wdt.c +23 -1
index 2dbeee9..5b9b80c 100644
@@ -19,6 +19,7 @@
 #include <linux/platform_device.h>
 #include <linux/watchdog.h>
 #include <linux/init.h>
+#include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/clk.h>
 #include <linux/err.h>
@@ -109,10 +110,16 @@ static struct watchdog_device orion_wdt = {
 	.min_timeout = 1,
 };
 
+static irqreturn_t orion_wdt_irq(int irq, void *devid)
+{
+	panic("Watchdog Timeout");
+	return IRQ_HANDLED;
+}
+
 static int orion_wdt_probe(struct platform_device *pdev)
 {
 	struct resource *res;
-	int ret;
+	int ret, irq;
 
 	clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(clk)) {
@@ -145,6 +152,21 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	/* Let's make sure the watchdog is fully stopped */
 	orion_wdt_stop(&orion_wdt);
 
+	/* Request the IRQ only after the watchdog is disabled */
+	irq = platform_get_irq(pdev, 0);
+	if (irq > 0) {
+		/*
+		 * Not all supported platforms specify an interrupt for the
+		 * watchdog, so let's make it optional.
+		 */
+		ret = devm_request_irq(&pdev->dev, irq, orion_wdt_irq, 0,
+				       pdev->name, &orion_wdt);
+		if (ret < 0) {
+			dev_err(&pdev->dev, "failed to request IRQ\n");
+			goto disable_clk;
+		}
+	}
+
 	watchdog_set_nowayout(&orion_wdt, nowayout);
 	ret = watchdog_register_device(&orion_wdt);
 	if (ret)
Signature · 1 line
1.8.1.5
#23 lore raw

PATCH v6 7/19 watchdog: orion: Make RSTOUT register a separate resource

In reply to Ezequiel Garcia

In order to support other SoC, it's required to distinguish the 'control' timer register, from the 'rstout' register that enables system reset on watchdog expiration.

To prevent a compatibility break, this commit adds a fallback to a hardcoded RSTOUT address.

---
 .../devicetree/bindings/watchdog/marvel.txt        |  6 ++-
 arch/arm/mach-dove/include/mach/bridge-regs.h      |  1 +
 arch/arm/mach-kirkwood/include/mach/bridge-regs.h  |  1 +
 arch/arm/mach-mv78xx0/include/mach/bridge-regs.h   |  1 +
 arch/arm/mach-orion5x/include/mach/bridge-regs.h   |  1 +
 arch/arm/plat-orion/common.c                       | 10 +++--
 drivers/watchdog/orion_wdt.c                       | 45 +++++++++++++++++++++-
 7 files changed, 57 insertions(+), 8 deletions(-)
Documentation/devicetree/bindings/watchdog/marvel.txt +4 -2
index 0731fbd..1544fe9 100644
@@ -3,7 +3,9 @@
 Required Properties:
 
 - Compatibility : "marvell,orion-wdt"
-- reg		: Address of the timer registers
+- reg		: Should contain two entries: first one with the
+		  timer control address, second one with the
+		  rstout enable address.
 
 Optional properties:
 
@@ -14,7 +16,7 @@ Example:
 
 	wdt at 20300 {
 		compatible = "marvell,orion-wdt";
-		reg = <0x20300 0x28>;
+		reg = <0x20300 0x28>, <0x20108 0x4>;
 		interrupts = <3>;
 		timeout-sec = <10>;
 		status = "okay";
arch/arm/mach-dove/include/mach/bridge-regs.h +1 -0
index 5362df3..f4a5b34 100644
@@ -21,6 +21,7 @@
 #define  CPU_CTRL_PCIE1_LINK	0x00000008
 
 #define RSTOUTn_MASK		(BRIDGE_VIRT_BASE + 0x0108)
+#define RSTOUTn_MASK_PHYS	(BRIDGE_PHYS_BASE + 0x0108)
 #define  SOFT_RESET_OUT_EN	0x00000004
 
 #define SYSTEM_SOFT_RESET	(BRIDGE_VIRT_BASE + 0x010c)
arch/arm/mach-kirkwood/include/mach/bridge-regs.h +1 -0
index 8b9d1c9..60f6421 100644
@@ -21,6 +21,7 @@
 #define CPU_RESET		0x00000002
 
 #define RSTOUTn_MASK		(BRIDGE_VIRT_BASE + 0x0108)
+#define RSTOUTn_MASK_PHYS	(BRIDGE_PHYS_BASE + 0x0108)
 #define SOFT_RESET_OUT_EN	0x00000004
 
 #define SYSTEM_SOFT_RESET	(BRIDGE_VIRT_BASE + 0x010c)
arch/arm/mach-mv78xx0/include/mach/bridge-regs.h +1 -0
index 5f03484..e20d6da 100644
@@ -15,6 +15,7 @@
 #define L2_WRITETHROUGH		0x00020000
 
 #define RSTOUTn_MASK		(BRIDGE_VIRT_BASE + 0x0108)
+#define RSTOUTn_MASK_PHYS	(BRIDGE_PHYS_BASE + 0x0108)
 #define SOFT_RESET_OUT_EN	0x00000004
 
 #define SYSTEM_SOFT_RESET	(BRIDGE_VIRT_BASE + 0x010c)
arch/arm/mach-orion5x/include/mach/bridge-regs.h +1 -0
index f727d03..5766e3f 100644
@@ -18,6 +18,7 @@
 #define CPU_CTRL		(ORION5X_BRIDGE_VIRT_BASE + 0x104)
 
 #define RSTOUTn_MASK		(ORION5X_BRIDGE_VIRT_BASE + 0x108)
+#define RSTOUTn_MASK_PHYS	(ORION5X_BRIDGE_PHYS_BASE + 0x108)
 
 #define CPU_SOFT_RESET		(ORION5X_BRIDGE_VIRT_BASE + 0x10c)
 
arch/arm/plat-orion/common.c +6 -4
index 830ff07..3ec6e8e 100644
@@ -595,14 +595,16 @@ void __init orion_spi_1_init(unsigned long mapbase)
 /*****************************************************************************
  * Watchdog
  ****************************************************************************/
-static struct resource orion_wdt_resource =
-		DEFINE_RES_MEM(TIMER_PHYS_BASE, 0x28);
+static struct resource orion_wdt_resource[] = {
+		DEFINE_RES_MEM(TIMER_PHYS_BASE, 0x04),
+		DEFINE_RES_MEM(RSTOUTn_MASK_PHYS, 0x04),
+};
 
 static struct platform_device orion_wdt_device = {
 	.name		= "orion_wdt",
 	.id		= -1,
-	.num_resources	= 1,
-	.resource	= &orion_wdt_resource,
+	.num_resources	= ARRAY_SIZE(orion_wdt_resource),
+	.resource	= orion_wdt_resource,
 };
 
 void __init orion_wdt_init(void)
drivers/watchdog/orion_wdt.c +43 -2

81 lines not on this page · raw message

Signature · 1 line
1.8.1.5
#24 lore raw

PATCH v6 8/19 watchdog: orion: Remove unneeded BRIDGE_CAUSE clear

In reply to Ezequiel Garcia

After adding the IRQ request, the BRIDGE_CAUSE bit should be cleared by the bridge interrupt controller. There's no longer a need to do it in the watchdog driver, so we can simply remove it.

---
 drivers/watchdog/orion_wdt.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/watchdog/orion_wdt.c +0 -5
index ad141d5..d280b17 100644
@@ -24,7 +24,6 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/of.h>
-#include <mach/bridge-regs.h>
 
 /* RSTOUT mask register physical address for Orion5x, Kirkwood and Dove */
 #define ORION_RSTOUT_MASK_OFFSET	0x20108
@@ -42,7 +41,6 @@
 #define WDT_MAX_CYCLE_COUNT	0xffffffff
 
 #define WDT_RESET_OUT_EN	BIT(1)
-#define WDT_INT_REQ		BIT(3)
 
 static bool nowayout = WATCHDOG_NOWAYOUT;
 static int heartbeat = -1;		/* module parameter (seconds) */
@@ -64,9 +62,6 @@ static int orion_wdt_start(struct watchdog_device *wdt_dev)
 	/* Set watchdog duration */
 	writel(wdt_tclk * wdt_dev->timeout, wdt_reg + WDT_VAL);
 
-	/* Clear watchdog timer interrupt */
-	writel(~WDT_INT_REQ, BRIDGE_CAUSE);
-
 	/* Enable watchdog timer */
 	atomic_io_modify(wdt_reg + TIMER_CTRL, WDT_EN, WDT_EN);
 
Signature · 1 line
1.8.1.5
#25 lore raw

PATCH v6 9/19 watchdog: orion: Introduce an orion_watchdog device structure

In reply to Ezequiel Garcia

In order to prepare to support multiple compatible-strings, this commit adds a device structure to hold the driver's state.

---
 drivers/watchdog/orion_wdt.c | 106 ++++++++++++++++++++++++++-----------------
 1 file changed, 65 insertions(+), 41 deletions(-)
drivers/watchdog/orion_wdt.c +65 -41
index d280b17..30ce504 100644
@@ -44,45 +44,56 @@
 
 static bool nowayout = WATCHDOG_NOWAYOUT;
 static int heartbeat = -1;		/* module parameter (seconds) */
-static unsigned int wdt_max_duration;	/* (seconds) */
-static struct clk *clk;
-static unsigned int wdt_tclk;
-static void __iomem *wdt_reg;
-static void __iomem *wdt_rstout;
+
+struct orion_watchdog {
+	struct watchdog_device wdt;
+	void __iomem *reg;
+	void __iomem *rstout;
+	unsigned long clk_rate;
+	struct clk *clk;
+};
 
 static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 {
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 	/* Reload watchdog duration */
-	writel(wdt_tclk * wdt_dev->timeout, wdt_reg + WDT_VAL);
+	writel(dev->clk_rate * wdt_dev->timeout, dev->reg + WDT_VAL);
 	return 0;
 }
 
 static int orion_wdt_start(struct watchdog_device *wdt_dev)
 {
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
 	/* Set watchdog duration */
-	writel(wdt_tclk * wdt_dev->timeout, wdt_reg + WDT_VAL);
+	writel(dev->clk_rate * wdt_dev->timeout, dev->reg + WDT_VAL);
 
 	/* Enable watchdog timer */
-	atomic_io_modify(wdt_reg + TIMER_CTRL, WDT_EN, WDT_EN);
+	atomic_io_modify(dev->reg + TIMER_CTRL, WDT_EN, WDT_EN);
 
 	/* Enable reset on watchdog */
-	atomic_io_modify(wdt_rstout, WDT_RESET_OUT_EN, WDT_RESET_OUT_EN);
+	atomic_io_modify(dev->rstout, WDT_RESET_OUT_EN, WDT_RESET_OUT_EN);
+
 	return 0;
 }
 
 static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 {
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
 	/* Disable reset on watchdog */
-	atomic_io_modify(wdt_rstout, WDT_RESET_OUT_EN, 0);
+	atomic_io_modify(dev->rstout, WDT_RESET_OUT_EN, 0);
 
 	/* Disable watchdog timer */
-	atomic_io_modify(wdt_reg + TIMER_CTRL, WDT_EN, 0);
+	atomic_io_modify(dev->reg + TIMER_CTRL, WDT_EN, 0);
+
 	return 0;
 }
 
 static unsigned int orion_wdt_get_timeleft(struct watchdog_device *wdt_dev)
 {
-	return readl(wdt_reg + WDT_VAL) / wdt_tclk;
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+	return readl(dev->reg + WDT_VAL) / dev->clk_rate;
 }
 
 static int orion_wdt_set_timeout(struct watchdog_device *wdt_dev,
@@ -106,12 +117,6 @@ static const struct watchdog_ops orion_wdt_ops = {
 	.get_timeleft = orion_wdt_get_timeleft,
 };
 
-static struct watchdog_device orion_wdt = {
-	.info = &orion_wdt_info,
-	.ops = &orion_wdt_ops,
-	.min_timeout = 1,
-};
-
 static irqreturn_t orion_wdt_irq(int irq, void *devid)
 {
 	panic("Watchdog Timeout");
@@ -147,18 +152,29 @@ static void __iomem *orion_wdt_ioremap_rstout(struct platform_device *pdev,
 
 static int orion_wdt_probe(struct platform_device *pdev)
 {
+	struct orion_watchdog *dev;
+	unsigned int wdt_max_duration;	/* (seconds) */
 	struct resource *res;
 	int ret, irq;
 
-	clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(clk)) {
+	dev = devm_kzalloc(&pdev->dev, sizeof(struct orion_watchdog),
+			   GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+
+	dev->wdt.info = &orion_wdt_info;
+	dev->wdt.ops = &orion_wdt_ops;
+	dev->wdt.min_timeout = 1;
+
+	dev->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(dev->clk)) {
 		dev_err(&pdev->dev, "Orion Watchdog missing clock\n");
-		return PTR_ERR(clk);
+		return PTR_ERR(dev->clk);
 	}
-	ret = clk_prepare_enable(clk);
+	ret = clk_prepare_enable(dev->clk);
 	if (ret)
 		return ret;
-	wdt_tclk = clk_get_rate(clk);
+	dev->clk_rate = clk_get_rate(dev->clk);
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!res) {
@@ -166,27 +182,31 @@ static int orion_wdt_probe(struct platform_device *pdev)
 		goto disable_clk;
 	}
 
-	wdt_reg = devm_ioremap(&pdev->dev, res->start, resource_size(res));
-	if (!wdt_reg) {
+	dev->reg = devm_ioremap(&pdev->dev, res->start,
+			       resource_size(res));
+	if (!dev->reg) {
 		ret = -ENOMEM;
 		goto disable_clk;
 	}
 
-	wdt_rstout = orion_wdt_ioremap_rstout(pdev, res->start &
-						    INTERNAL_REGS_MASK);
-	if (!wdt_rstout) {
+	dev->rstout = orion_wdt_ioremap_rstout(pdev, res->start &
+						     INTERNAL_REGS_MASK);
+	if (!dev->rstout) {
 		ret = -ENODEV;
 		goto disable_clk;
 	}
 
-	wdt_max_duration = WDT_MAX_CYCLE_COUNT / wdt_tclk;
+	wdt_max_duration = WDT_MAX_CYCLE_COUNT / dev->clk_rate;
+
+	dev->wdt.timeout = wdt_max_duration;
+	dev->wdt.max_timeout = wdt_max_duration;
+	watchdog_init_timeout(&dev->wdt, heartbeat, &pdev->dev);
 
-	orion_wdt.timeout = wdt_max_duration;
-	orion_wdt.max_timeout = wdt_max_duration;
-	watchdog_init_timeout(&orion_wdt, heartbeat, &pdev->dev);
+	platform_set_drvdata(pdev, &dev->wdt);
+	watchdog_set_drvdata(&dev->wdt, dev);
 
 	/* Let's make sure the watchdog is fully stopped */
-	orion_wdt_stop(&orion_wdt);
+	orion_wdt_stop(&dev->wdt);
 
 	/* Request the IRQ only after the watchdog is disabled */
 	irq = platform_get_irq(pdev, 0);
@@ -196,37 +216,41 @@ static int orion_wdt_probe(struct platform_device *pdev)
 		 * watchdog, so let's make it optional.
 		 */
 		ret = devm_request_irq(&pdev->dev, irq, orion_wdt_irq, 0,
-				       pdev->name, &orion_wdt);
+				       pdev->name, dev);
 		if (ret < 0) {
 			dev_err(&pdev->dev, "failed to request IRQ\n");
 			goto disable_clk;
 		}
 	}
 
-	watchdog_set_nowayout(&orion_wdt, nowayout);
-	ret = watchdog_register_device(&orion_wdt);
+	watchdog_set_nowayout(&dev->wdt, nowayout);
+	ret = watchdog_register_device(&dev->wdt);
 	if (ret)
 		goto disable_clk;
 
 	pr_info("Initial timeout %d sec%s\n",
-		orion_wdt.timeout, nowayout ? ", nowayout" : "");
+		dev->wdt.timeout, nowayout ? ", nowayout" : "");
 	return 0;
 
 disable_clk:
-	clk_disable_unprepare(clk);
+	clk_disable_unprepare(dev->clk);
 	return ret;
 }
 
 static int orion_wdt_remove(struct platform_device *pdev)
 {
-	watchdog_unregister_device(&orion_wdt);
-	clk_disable_unprepare(clk);
+	struct watchdog_device *wdt_dev = platform_get_drvdata(pdev);
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
+	watchdog_unregister_device(wdt_dev);
+	clk_disable_unprepare(dev->clk);
 	return 0;
 }
 
 static void orion_wdt_shutdown(struct platform_device *pdev)
 {
-	orion_wdt_stop(&orion_wdt);
+	struct watchdog_device *wdt_dev = platform_get_drvdata(pdev);
+	orion_wdt_stop(wdt_dev);
 }
 
 static const struct of_device_id orion_wdt_of_match_table[] = {
Signature · 1 line
1.8.1.5
#26 lore raw

PATCH v6 10/19 watchdog: orion: Introduce per-compatible of_device_id data

In reply to Ezequiel Garcia

This commit adds an orion_watchdog_data structure to hold compatible-data information. This allows to remove the driver-wide definition and to be able to add support for multiple compatible-strings in the future.

---
 drivers/watchdog/orion_wdt.c | 58 +++++++++++++++++++++++++++++++-------------
 1 file changed, 41 insertions(+), 17 deletions(-)
drivers/watchdog/orion_wdt.c +41 -17
index 30ce504..12fedf4 100644
@@ -24,6 +24,7 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 
 /* RSTOUT mask register physical address for Orion5x, Kirkwood and Dove */
 #define ORION_RSTOUT_MASK_OFFSET	0x20108
@@ -35,29 +36,33 @@
  * Watchdog timer block registers.
  */
 #define TIMER_CTRL		0x0000
-#define WDT_EN			0x0010
-#define WDT_VAL			0x0024
 
 #define WDT_MAX_CYCLE_COUNT	0xffffffff
 
-#define WDT_RESET_OUT_EN	BIT(1)
-
 static bool nowayout = WATCHDOG_NOWAYOUT;
 static int heartbeat = -1;		/* module parameter (seconds) */
 
+struct orion_watchdog_data {
+	int wdt_counter_offset;
+	int wdt_enable_bit;
+	int rstout_enable_bit;
+};
+
 struct orion_watchdog {
 	struct watchdog_device wdt;
 	void __iomem *reg;
 	void __iomem *rstout;
 	unsigned long clk_rate;
 	struct clk *clk;
+	const struct orion_watchdog_data *data;
 };
 
 static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 	/* Reload watchdog duration */
-	writel(dev->clk_rate * wdt_dev->timeout, dev->reg + WDT_VAL);
+	writel(dev->clk_rate * wdt_dev->timeout,
+	       dev->reg + dev->data->wdt_counter_offset);
 	return 0;
 }
 
@@ -66,13 +71,16 @@ static int orion_wdt_start(struct watchdog_device *wdt_dev)
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 
 	/* Set watchdog duration */
-	writel(dev->clk_rate * wdt_dev->timeout, dev->reg + WDT_VAL);
+	writel(dev->clk_rate * wdt_dev->timeout,
+	       dev->reg + dev->data->wdt_counter_offset);
 
 	/* Enable watchdog timer */
-	atomic_io_modify(dev->reg + TIMER_CTRL, WDT_EN, WDT_EN);
+	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit,
+						dev->data->wdt_enable_bit);
 
 	/* Enable reset on watchdog */
-	atomic_io_modify(dev->rstout, WDT_RESET_OUT_EN, WDT_RESET_OUT_EN);
+	atomic_io_modify(dev->rstout, dev->data->rstout_enable_bit,
+				      dev->data->rstout_enable_bit);
 
 	return 0;
 }
@@ -82,10 +90,10 @@ static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 
 	/* Disable reset on watchdog */
-	atomic_io_modify(dev->rstout, WDT_RESET_OUT_EN, 0);
+	atomic_io_modify(dev->rstout, dev->data->rstout_enable_bit, 0);
 
 	/* Disable watchdog timer */
-	atomic_io_modify(dev->reg + TIMER_CTRL, WDT_EN, 0);
+	atomic_io_modify(dev->reg + TIMER_CTRL, dev->data->wdt_enable_bit, 0);
 
 	return 0;
 }
@@ -93,7 +101,7 @@ static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 static unsigned int orion_wdt_get_timeleft(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
-	return readl(dev->reg + WDT_VAL) / dev->clk_rate;
+	return readl(dev->reg + dev->data->wdt_counter_offset) / dev->clk_rate;
 }
 
 static int orion_wdt_set_timeout(struct watchdog_device *wdt_dev,
@@ -150,9 +158,25 @@ static void __iomem *orion_wdt_ioremap_rstout(struct platform_device *pdev,
 	return devm_ioremap(&pdev->dev, rstout, 0x4);
 }
 
+static const struct orion_watchdog_data orion_data = {
+	.rstout_enable_bit = BIT(1),
+	.wdt_enable_bit = BIT(4),
+	.wdt_counter_offset = 0x24,
+};
+
+static const struct of_device_id orion_wdt_of_match_table[] = {
+	{
+		.compatible = "marvell,orion-wdt",
+		.data = &orion_data,
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, orion_wdt_of_match_table);
+
 static int orion_wdt_probe(struct platform_device *pdev)
 {
 	struct orion_watchdog *dev;
+	const struct of_device_id *match;
 	unsigned int wdt_max_duration;	/* (seconds) */
 	struct resource *res;
 	int ret, irq;
@@ -162,9 +186,15 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	if (!dev)
 		return -ENOMEM;
 
+	match = of_match_device(orion_wdt_of_match_table, &pdev->dev);
+	if (!match)
+		/* Default legacy match */
+		match = &orion_wdt_of_match_table[0];
+
 	dev->wdt.info = &orion_wdt_info;
 	dev->wdt.ops = &orion_wdt_ops;
 	dev->wdt.min_timeout = 1;
+	dev->data = match->data;
 
 	dev->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(dev->clk)) {
@@ -253,12 +283,6 @@ static void orion_wdt_shutdown(struct platform_device *pdev)
 	orion_wdt_stop(wdt_dev);
 }
 
-static const struct of_device_id orion_wdt_of_match_table[] = {
-	{ .compatible = "marvell,orion-wdt", },
-	{},
-};
-MODULE_DEVICE_TABLE(of, orion_wdt_of_match_table);
-
 static struct platform_driver orion_wdt_driver = {
 	.probe		= orion_wdt_probe,
 	.remove		= orion_wdt_remove,
Signature · 1 line
1.8.1.5
#27 lore raw

PATCH v6 11/19 watchdog: orion: Add per-compatible clock initialization

In reply to Ezequiel Garcia

Following the introduction of the compatible-data field, it's now possible to further abstract the clock initialization. This will allow to support SoC with a different clock setup.

---
 drivers/watchdog/orion_wdt.c | 53 ++++++++++++++++++++++++++------------------
 1 file changed, 32 insertions(+), 21 deletions(-)
drivers/watchdog/orion_wdt.c +32 -21
index 12fedf4..de7a10b 100644
@@ -42,10 +42,14 @@
 static bool nowayout = WATCHDOG_NOWAYOUT;
 static int heartbeat = -1;		/* module parameter (seconds) */
 
+struct orion_watchdog;
+
 struct orion_watchdog_data {
 	int wdt_counter_offset;
 	int wdt_enable_bit;
 	int rstout_enable_bit;
+	int (*clock_init)(struct platform_device *,
+			  struct orion_watchdog *);
 };
 
 struct orion_watchdog {
@@ -57,6 +61,22 @@ struct orion_watchdog {
 	const struct orion_watchdog_data *data;
 };
 
+static int orion_wdt_clock_init(struct platform_device *pdev,
+				struct orion_watchdog *dev)
+{
+	int ret;
+
+	dev->clk = devm_clk_get(&pdev->dev, NULL);
+	if (IS_ERR(dev->clk))
+		return PTR_ERR(dev->clk);
+	ret = clk_prepare_enable(dev->clk);
+	if (ret)
+		return ret;
+
+	dev->clk_rate = clk_get_rate(dev->clk);
+	return 0;
+}
+
 static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
@@ -162,6 +182,7 @@ static const struct orion_watchdog_data orion_data = {
 	.rstout_enable_bit = BIT(1),
 	.wdt_enable_bit = BIT(4),
 	.wdt_counter_offset = 0x24,
+	.clock_init = orion_wdt_clock_init,
 };
 
 static const struct of_device_id orion_wdt_of_match_table[] = {
@@ -196,34 +217,24 @@ static int orion_wdt_probe(struct platform_device *pdev)
 	dev->wdt.min_timeout = 1;
 	dev->data = match->data;
 
-	dev->clk = devm_clk_get(&pdev->dev, NULL);
-	if (IS_ERR(dev->clk)) {
-		dev_err(&pdev->dev, "Orion Watchdog missing clock\n");
-		return PTR_ERR(dev->clk);
-	}
-	ret = clk_prepare_enable(dev->clk);
-	if (ret)
-		return ret;
-	dev->clk_rate = clk_get_rate(dev->clk);
-
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		ret = -ENODEV;
-		goto disable_clk;
-	}
+	if (!res)
+		return -ENODEV;
 
 	dev->reg = devm_ioremap(&pdev->dev, res->start,
 			       resource_size(res));
-	if (!dev->reg) {
-		ret = -ENOMEM;
-		goto disable_clk;
-	}
+	if (!dev->reg)
+		return -ENOMEM;
 
 	dev->rstout = orion_wdt_ioremap_rstout(pdev, res->start &
 						     INTERNAL_REGS_MASK);
-	if (!dev->rstout) {
-		ret = -ENODEV;
-		goto disable_clk;
+	if (!dev->rstout)
+		return -ENODEV;
+
+	ret = dev->data->clock_init(pdev, dev);
+	if (ret) {
+		dev_err(&pdev->dev, "cannot initialize clock\n");
+		return ret;
 	}
 
 	wdt_max_duration = WDT_MAX_CYCLE_COUNT / dev->clk_rate;
Signature · 1 line
1.8.1.5
#28 lore raw

PATCH v6 12/19 watchdog: orion: Add per-compatible watchdog start implementation

In reply to Ezequiel Garcia

To handle differences between SoCs this commit adds per-compatible string start() function for the watchdog kick-off. This is preparation work and makes no functionality changes to the current driver.

---
 drivers/watchdog/orion_wdt.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/watchdog/orion_wdt.c +11 -1
index de7a10b..ce0c365 100644
@@ -50,6 +50,7 @@ struct orion_watchdog_data {
 	int rstout_enable_bit;
 	int (*clock_init)(struct platform_device *,
 			  struct orion_watchdog *);
+	int (*start)(struct watchdog_device *);
 };
 
 struct orion_watchdog {
@@ -86,7 +87,7 @@ static int orion_wdt_ping(struct watchdog_device *wdt_dev)
 	return 0;
 }
 
-static int orion_wdt_start(struct watchdog_device *wdt_dev)
+static int orion_start(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
 
@@ -105,6 +106,14 @@ static int orion_wdt_start(struct watchdog_device *wdt_dev)
 	return 0;
 }
 
+static int orion_wdt_start(struct watchdog_device *wdt_dev)
+{
+	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
+
+	/* There are some per-SoC quirks to handle */
+	return dev->data->start(wdt_dev);
+}
+
 static int orion_wdt_stop(struct watchdog_device *wdt_dev)
 {
 	struct orion_watchdog *dev = watchdog_get_drvdata(wdt_dev);
@@ -183,6 +192,7 @@ static const struct orion_watchdog_data orion_data = {
 	.wdt_enable_bit = BIT(4),
 	.wdt_counter_offset = 0x24,
 	.clock_init = orion_wdt_clock_init,
+	.start = orion_start,
 };
 
 static const struct of_device_id orion_wdt_of_match_table[] = {
Signature · 1 line
1.8.1.5
#29 lore raw

PATCH v6 13/19 watchdog: orion: Add support for Armada 370 and Armada XP SoC

In reply to Ezequiel Garcia

Using the added infrastructure for handling SoC differences, this commit adds support for the watchdog controller available in Armada 370 and Armada XP SoCs.

Also, and because the AXP clock initialization uses of_clk_get_by_name, this commit changes the orion clock initialization to use clk_get() and adds a proper clk_put() on the common exit/error paths.

---
 .../devicetree/bindings/watchdog/marvel.txt        |   3 +
 drivers/watchdog/orion_wdt.c                       | 106 ++++++++++++++++++++-
 2 files changed, 107 insertions(+), 2 deletions(-)
Documentation/devicetree/bindings/watchdog/marvel.txt +3 -0
index 1544fe9..de11eb4 100644
@@ -3,6 +3,9 @@
 Required Properties:
 
 - Compatibility : "marvell,orion-wdt"
+		  "marvell,armada-370-wdt"
+		  "marvell,armada-xp-wdt"
+
 - reg		: Should contain two entries: first one with the
 		  timer control address, second one with the
 		  rstout enable address.
drivers/watchdog/orion_wdt.c +104 -2

154 lines not on this page · raw message

Signature · 1 line
1.8.1.5
#30 lore raw

PATCH v6 14/19 ARM: mvebu: Enable Armada 370/XP watchdog in the devicetree

In reply to Ezequiel Garcia

Add the DT nodes to enable watchdog support available in Armada 370 and Armada XP SoCs.

---
 arch/arm/boot/dts/armada-370-xp.dtsi | 4 ++++
 arch/arm/boot/dts/armada-370.dtsi    | 5 +++++
 arch/arm/boot/dts/armada-xp.dtsi     | 6 ++++++
 3 files changed, 15 insertions(+)
arch/arm/boot/dts/armada-370-xp.dtsi +4 -0
index 74b5964..7bbc4ac 100644
@@ -199,6 +199,10 @@
 				interrupts = <37>, <38>, <39>, <40>, <5>, <6>;
 			};
 
+			watchdog at 20300 {
+				reg = <0x20300 0x34>, <0x20704 0x4>;
+			};
+
 			usb at 50000 {
 				compatible = "marvell,orion-ehci";
 				reg = <0x50000 0x500>;
arch/arm/boot/dts/armada-370.dtsi +5 -0
index 0d8530c..b7759b2 100644
@@ -196,6 +196,11 @@
 				clocks = <&coreclk 2>;
 			};
 
+			watchdog at 20300 {
+				compatible = "marvell,armada-370-wdt";
+				clocks = <&coreclk 2>;
+			};
+
 			usb at 50000 {
 				clocks = <&coreclk 0>;
 			};
arch/arm/boot/dts/armada-xp.dtsi +6 -0
index b8b84a2..abb9f9d 100644
@@ -111,6 +111,12 @@
 				clock-names = "nbclk", "fixed";
 			};
 
+			watchdog at 20300 {
+				compatible = "marvell,armada-xp-wdt";
+				clocks = <&coreclk 2>, <&refclk>;
+				clock-names = "nbclk", "fixed";
+			};
+
 			armada-370-xp-pmsu at 22000 {
 				compatible = "marvell,armada-370-xp-pmsu";
 				reg = <0x22100 0x400>, <0x20800 0x20>;
Signature · 1 line
1.8.1.5
#31 lore raw

PATCH v6 15/19 ARM: kirkwood: Add RSTOUT 'reg' entry to devicetree

In reply to Ezequiel Garcia

In order to support multiplatform builds the watchdog devicetree binding was modified and now the 'reg' property is specified to need two entries. This commit adds the second entry as-per the new specification.

---
 arch/arm/boot/dts/kirkwood.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/kirkwood.dtsi +1 -1
index 6abf44d..85542c2 100644
@@ -178,7 +178,7 @@
 
 		wdt: watchdog-timer at 20300 {
 			compatible = "marvell,orion-wdt";
-			reg = <0x20300 0x28>;
+			reg = <0x20300 0x28>, <0x20108 0x4>;
 			interrupt-parent = <&bridge_intc>;
 			interrupts = <3>;
 			clocks = <&gate_clk 7>;
Signature · 1 line
1.8.1.5
#32 lore raw

PATCH v6 16/19 ARM: dove: Enable Dove watchdog in the devicetree

In reply to Ezequiel Garcia

Add the devicetree node to enable watchdog support available in Dove SoCs.

---
 arch/arm/boot/dts/dove.dtsi | 8 ++++++++
 1 file changed, 8 insertions(+)
arch/arm/boot/dts/dove.dtsi +8 -0
index 2b76524..a88ed73 100644
@@ -210,6 +210,14 @@
 				clocks = <&core_clk 0>;
 			};
 
+			watchdog at 20300 {
+				compatible = "marvell,orion-wdt";
+				reg = <0x20300 0x28>, <0x20108 0x4>;
+				interrupt-parent = <&bridge_intc>;
+				interrupts = <3>;
+				clocks = <&core_clk 0>;
+			};
+
 			crypto: crypto-engine at 30000 {
 				compatible = "marvell,orion-crypto";
 				reg = <0x30000 0x10000>,
Signature · 1 line
1.8.1.5
#33 lore raw

PATCH v6 17/19 watchdog: orion: Enable the build on ARCH_MVEBU

In reply to Ezequiel Garcia

After adding support for Armada 370/XP SoC let's enable the build on these platforms.

---
 drivers/watchdog/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/watchdog/Kconfig +1 -1
index 4c4c566..2dda6c0 100644
@@ -291,7 +291,7 @@ config DAVINCI_WATCHDOG
 
 config ORION_WATCHDOG
 	tristate "Orion watchdog"
-	depends on ARCH_ORION5X || ARCH_KIRKWOOD || ARCH_DOVE
+	depends on ARCH_ORION5X || ARCH_KIRKWOOD || ARCH_DOVE || ARCH_MVEBU
 	select WATCHDOG_CORE
 	help
 	  Say Y here if to include support for the watchdog timer
Signature · 1 line
1.8.1.5
maintainer
#34 lore raw

PATCH v6 17/19 watchdog: orion: Enable the build on ARCH_MVEBU

Quoted 5 lines · Ezequiel Garcia

After adding support for Armada 370/XP SoC let's enable the build on these platforms.

Quoted 18 lines
---
  drivers/watchdog/Kconfig | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
drivers/watchdog/Kconfig +1 -1
index 4c4c566..2dda6c0 100644
@@ -291,7 +291,7 @@ config DAVINCI_WATCHDOG
 
  config ORION_WATCHDOG
  	tristate "Orion watchdog"
-	depends on ARCH_ORION5X || ARCH_KIRKWOOD || ARCH_DOVE
+	depends on ARCH_ORION5X || ARCH_KIRKWOOD || ARCH_DOVE || ARCH_MVEBU
  	select WATCHDOG_CORE
  	help
  	  Say Y here if to include support for the watchdog timer
#35 lore raw

PATCH v6 18/19 ARM: mvebu: Enable watchdog support in defconfig

In reply to Ezequiel Garcia

Now that we have proper support for Armada 370/XP watchdog let's enable it in the defconfig.

---
 arch/arm/configs/mvebu_defconfig | 2 ++
 1 file changed, 2 insertions(+)
arch/arm/configs/mvebu_defconfig +2 -0
index 0f4511d..31bdc6f 100644
@@ -62,6 +62,8 @@ CONFIG_GPIOLIB=y
 CONFIG_GPIO_SYSFS=y
 CONFIG_THERMAL=y
 CONFIG_ARMADA_THERMAL=y
+CONFIG_WATCHDOG=y
+CONFIG_ORION_WATCHDOG=y
 CONFIG_USB_SUPPORT=y
 CONFIG_USB=y
 CONFIG_USB_EHCI_HCD=y
Signature · 1 line
1.8.1.5
#36 lore raw

PATCH v6 18/19 ARM: mvebu: Enable watchdog support in defconfig

Quoted 7 lines · Ezequiel Garcia

Now that we have proper support for Armada 370/XP watchdog let's enable it in the defconfig.

---
 arch/arm/configs/mvebu_defconfig | 2 ++
 1 file changed, 2 insertions(+)

If this ends up being the last version for this series, I'll add a patch for multi_v7_defconfig. Otherwise, please add for the next round.

thx,

Jason.

#37 lore raw

PATCH v6 18/19 ARM: mvebu: Enable watchdog support in defconfig

Quoted 4 lines · Jason Cooper
Quoted 3 lines · Ezequiel Garcia

Now that we have proper support for Armada 370/XP watchdog let's enable it in the defconfig.

Quoted 8 lines
Quoted 4 lines
---
 arch/arm/configs/mvebu_defconfig | 2 ++
 1 file changed, 2 insertions(+)

If this ends up being the last version for this series, I'll add a patch for multi_v7_defconfig. Otherwise, please add for the next round.

Oh, yes, I forgot about your request on this. Let's hope I don't forget next time.

Signature · 3 lines
Ezequiel Garc?a, Free Electrons
Embedded Linux, Kernel and Android Engineering
http://free-electrons.com
#38 lore raw

PATCH v6 18/19 ARM: mvebu: Enable watchdog support in defconfig

Quoted 10 lines · Ezequiel Garcia
Quoted 9 lines · Jason Cooper
Quoted 7 lines · Ezequiel Garcia

Now that we have proper support for Armada 370/XP watchdog let's enable it in the defconfig.

---
 arch/arm/configs/mvebu_defconfig | 2 ++
 1 file changed, 2 insertions(+)
Quoted 6 lines
Quoted 3 lines

If this ends up being the last version for this series, I'll add a patch for multi_v7_defconfig. Otherwise, please add for the next round.

Oh, yes, I forgot about your request on this. Let's hope I don't forget next time.

Yes, we found the i2c hang because of multi_v7, so it's definitely worth the effort.

thx,

Jason.

#39 lore raw

PATCH v6 19/19 ARM: dove: Enable watchdog support in the defconfig

In reply to Ezequiel Garcia

Now that we have watchdog support, let's add it to the defconfig.

---
 arch/arm/configs/dove_defconfig | 2 ++
 1 file changed, 2 insertions(+)
arch/arm/configs/dove_defconfig +2 -0
index 1101054..a330690 100644
@@ -80,6 +80,8 @@ CONFIG_SPI_ORION=y
 # CONFIG_HWMON is not set
 CONFIG_THERMAL=y
 CONFIG_DOVE_THERMAL=y
+CONFIG_WATCHDOG=y
+CONFIG_ORION_WATCHDOG=y
 CONFIG_USB=y
 CONFIG_USB_XHCI_HCD=y
 CONFIG_USB_EHCI_HCD=y
Signature · 1 line
1.8.1.5