diff mbox

[v2] gpio: mvebu: Add clk support to prevent lockup

Message ID 1359887666-17552-1-git-send-email-andrew@lunn.ch (mailing list archive)
State New, archived
Headers show

Commit Message

Andrew Lunn Feb. 3, 2013, 10:34 a.m. UTC
The kirkwood SoC GPIO cores use the runit clock. Add code to
clk_prepare_enable() runit, otherwise there is a danger of locking up
the SoC by accessing the GPIO registers when runit clock is not
ticking.

Reported-by: Simon Baatz <gmbnomis@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---
Since v1:
ChangeLog: tclk->runit clock

 arch/arm/boot/dts/kirkwood.dtsi |    2 ++
 drivers/gpio/gpio-mvebu.c       |    7 +++++++
 2 files changed, 9 insertions(+)

Comments

Simon Baatz Feb. 3, 2013, 7:48 p.m. UTC | #1
On Sun, Feb 03, 2013 at 11:34:26AM +0100, Andrew Lunn wrote:
> The kirkwood SoC GPIO cores use the runit clock. Add code to
> clk_prepare_enable() runit, otherwise there is a danger of locking up
> the SoC by accessing the GPIO registers when runit clock is not
> ticking.
> 
> Reported-by: Simon Baatz <gmbnomis@gmail.com>
> Signed-off-by: Andrew Lunn <andrew@lunn.ch>

Tested-by: Simon Baatz <gmbnomis@gmail.com>

After boot (GPIO is the only driver enabling runit at this point in
time):

# cat /sys/kernel/debug/clk/tclk/runit/clk_enable_count
2


- Simon
Jason Cooper Feb. 4, 2013, 1:44 a.m. UTC | #2
On Sun, Feb 03, 2013 at 11:34:26AM +0100, Andrew Lunn wrote:
> The kirkwood SoC GPIO cores use the runit clock. Add code to
> clk_prepare_enable() runit, otherwise there is a danger of locking up
> the SoC by accessing the GPIO registers when runit clock is not
> ticking.
> 
> Reported-by: Simon Baatz <gmbnomis@gmail.com>
> Signed-off-by: Andrew Lunn <andrew@lunn.ch>
> ---
> Since v1:
> ChangeLog: tclk->runit clock
> 
>  arch/arm/boot/dts/kirkwood.dtsi |    2 ++
>  drivers/gpio/gpio-mvebu.c       |    7 +++++++
>  2 files changed, 9 insertions(+)

Linus,

I merge-tested this against your gpio/fixes branch since you had a patch
to the same file:

11c0cee gpio: mvebu: Don't free chip label memory

There were no conflicts.  If you don't mind, I'd like to take this
through mvebu/fixes -> arm-soc since this is one of four patches to fix
one reported issue.  I'd like to keep them in the same branch.

If this causes a headache for you, please let me know and you can take
it.

Otherwise, applied to mvebu/fixes with Simon's Tested-by:

thx,

Jason.
Linus Walleij Feb. 4, 2013, 8:45 p.m. UTC | #3
On Mon, Feb 4, 2013 at 2:44 AM, Jason Cooper <jason@lakedaemon.net> wrote:
> On Sun, Feb 03, 2013 at 11:34:26AM +0100, Andrew Lunn wrote:
>> The kirkwood SoC GPIO cores use the runit clock. Add code to
>> clk_prepare_enable() runit, otherwise there is a danger of locking up
>> the SoC by accessing the GPIO registers when runit clock is not
>> ticking.
>>
>> Reported-by: Simon Baatz <gmbnomis@gmail.com>
>> Signed-off-by: Andrew Lunn <andrew@lunn.ch>
>> ---
>> Since v1:
>> ChangeLog: tclk->runit clock
>>
>>  arch/arm/boot/dts/kirkwood.dtsi |    2 ++
>>  drivers/gpio/gpio-mvebu.c       |    7 +++++++
>>  2 files changed, 9 insertions(+)
>
> Linus,
>
> I merge-tested this against your gpio/fixes branch since you had a patch
> to the same file:
>
> 11c0cee gpio: mvebu: Don't free chip label memory
>
> There were no conflicts.  If you don't mind, I'd like to take this
> through mvebu/fixes -> arm-soc since this is one of four patches to fix
> one reported issue.  I'd like to keep them in the same branch.

No problem, go ahead.
Acked-by: Linus Walleij <linus.walleij@linaro.org>

Yours,
Linus Walleij
diff mbox

Patch

diff --git a/arch/arm/boot/dts/kirkwood.dtsi b/arch/arm/boot/dts/kirkwood.dtsi
index d6ab442..6c5d75f 100644
--- a/arch/arm/boot/dts/kirkwood.dtsi
+++ b/arch/arm/boot/dts/kirkwood.dtsi
@@ -38,6 +38,7 @@ 
 			interrupt-controller;
 			#interrupt-cells = <2>;
 			interrupts = <35>, <36>, <37>, <38>;
+			clocks = <&gate_clk 7>;
 		};
 
 		gpio1: gpio@10140 {
@@ -49,6 +50,7 @@ 
 			interrupt-controller;
 			#interrupt-cells = <2>;
 			interrupts = <39>, <40>, <41>;
+			clocks = <&gate_clk 7>;
 		};
 
 		serial@12000 {
diff --git a/drivers/gpio/gpio-mvebu.c b/drivers/gpio/gpio-mvebu.c
index 6819d63..456663c 100644
--- a/drivers/gpio/gpio-mvebu.c
+++ b/drivers/gpio/gpio-mvebu.c
@@ -41,6 +41,7 @@ 
 #include <linux/io.h>
 #include <linux/of_irq.h>
 #include <linux/of_device.h>
+#include <linux/clk.h>
 #include <linux/pinctrl/consumer.h>
 
 /*
@@ -495,6 +496,7 @@  static int mvebu_gpio_probe(struct platform_device *pdev)
 	struct resource *res;
 	struct irq_chip_generic *gc;
 	struct irq_chip_type *ct;
+	struct clk *clk;
 	unsigned int ngpios;
 	int soc_variant;
 	int i, cpu, id;
@@ -528,6 +530,11 @@  static int mvebu_gpio_probe(struct platform_device *pdev)
 		return id;
 	}
 
+	clk = devm_clk_get(&pdev->dev, NULL);
+	/* Not all SoCs require a clock.*/
+	if (!IS_ERR(clk))
+		clk_prepare_enable(clk);
+
 	mvchip->soc_variant = soc_variant;
 	mvchip->chip.label = dev_name(&pdev->dev);
 	mvchip->chip.dev = &pdev->dev;