[PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Mika Westerberg]

Guenter Roeck reported an interrupt storm on a prototype system which is
based on Cyan Chromebook. The root cause turned out to be a incorrectly
configured pin that triggers spurious interrupts. This will be fixed in
coreboot but currently we need to prevent the interrupt storm from
happening by masking all interrupts (but not GPEs) on those systems.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
Reported-by: Guenter Roeck <linux@roeck-us.net>
Reported-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
---
 drivers/pinctrl/intel/pinctrl-cherryview.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/pinctrl/intel/pinctrl-cherryview.c b/drivers/pinctrl/intel/pinctrl-cherryview.c
index aa6c9f569c2b..083ffd397154 100644
--- a/drivers/pinctrl/intel/pinctrl-cherryview.c
+++ b/drivers/pinctrl/intel/pinctrl-cherryview.c
@@ -1601,6 +1601,22 @@ static int chv_gpio_probe(struct chv_pinctrl *pctrl, int irq)
 			clear_bit(i, chip->irq.valid_mask);
 	}
 
+	/*
+	 * The same set of machines in chv_no_valid_mask[] have incorrectly
+	 * configured GPIOs that generate spurious interrupts so we use
+	 * this same list to apply another quirk for them.
+	 *
+	 * See also https://bugzilla.kernel.org/show_bug.cgi?id=197953.
+	 */
+	if (!need_valid_mask) {
+		/*
+		 * Mask all interrupts the community is able to generate
+		 * but leave the ones that can only generate GPEs unmasked.
+		 */
+		chv_writel(~(BIT(pctrl->community->nirqs) - 1),
+			   pctrl->regs + CHV_INTMASK);
+	}
+
 	/* Clear all interrupts */
 	chv_writel(0xffff, pctrl->regs + CHV_INTSTAT);
 
-- 
2.15.0

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Andy Shevchenko]

On Fri, 2017-12-01 at 13:25 +0300, Mika Westerberg wrote:
> Guenter Roeck reported an interrupt storm on a prototype system which
> is
> based on Cyan Chromebook. The root cause turned out to be a
> incorrectly
> configured pin that triggers spurious interrupts. This will be fixed
> in
> coreboot but currently we need to prevent the interrupt storm from
> happening by masking all interrupts (but not GPEs) on those systems.
> 
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945

Shouldn't be this URL to the bug mentioned in the comments below?

> +	/*
> +	 * The same set of machines in chv_no_valid_mask[] have
> incorrectly
> +	 * configured GPIOs that generate spurious interrupts so we
> use
> +	 * this same list to apply another quirk for them.
> +	 *
> +	 * See also https://bugzilla.kernel.org/show_bug.cgi?id=19795
> 3.
> +	 */

> +		chv_writel(~(BIT(pctrl->community->nirqs) - 1),
> +			   pctrl->regs + CHV_INTMASK);

It might be considered to use

GENMASK(32, pctrl->community->nirqs)

instead of

~(BIT(pctrl->community->nirqs) - 1)

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Mika Westerberg]

On Fri, Dec 01, 2017 at 02:35:32PM +0200, Andy Shevchenko wrote:
> On Fri, 2017-12-01 at 13:25 +0300, Mika Westerberg wrote:
> > Guenter Roeck reported an interrupt storm on a prototype system which
> > is
> > based on Cyan Chromebook. The root cause turned out to be a
> > incorrectly
> > configured pin that triggers spurious interrupts. This will be fixed
> > in
> > coreboot but currently we need to prevent the interrupt storm from
> > happening by masking all interrupts (but not GPEs) on those systems.
> > 
> > Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
> 
> Shouldn't be this URL to the bug mentioned in the comments below?

Doh! Yes, it should. I'll fix it up.

> > +	/*
> > +	 * The same set of machines in chv_no_valid_mask[] have
> > incorrectly
> > +	 * configured GPIOs that generate spurious interrupts so we
> > use
> > +	 * this same list to apply another quirk for them.
> > +	 *
> > +	 * See also https://bugzilla.kernel.org/show_bug.cgi?id=19795
> > 3.
> > +	 */
> 
> > +		chv_writel(~(BIT(pctrl->community->nirqs) - 1),
> > +			   pctrl->regs + CHV_INTMASK);
> 
> It might be considered to use
> 
> GENMASK(32, pctrl->community->nirqs)
> 
> instead of
> 
> ~(BIT(pctrl->community->nirqs) - 1)

Indeed it looks better with GENMASK() so I'll change it in the next
version.

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Guenter Roeck]

On Fri, Dec 01, 2017 at 01:25:04PM +0300, Mika Westerberg wrote:
> Guenter Roeck reported an interrupt storm on a prototype system which is
> based on Cyan Chromebook. The root cause turned out to be a incorrectly
> configured pin that triggers spurious interrupts. This will be fixed in
> coreboot but currently we need to prevent the interrupt storm from
> happening by masking all interrupts (but not GPEs) on those systems.
> 
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
> Reported-by: Guenter Roeck <linux@roeck-us.net>
> Reported-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Tested-by: Guenter Roeck <linux@roeck-us.net>

Guenter

> ---
>  drivers/pinctrl/intel/pinctrl-cherryview.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/drivers/pinctrl/intel/pinctrl-cherryview.c b/drivers/pinctrl/intel/pinctrl-cherryview.c
> index aa6c9f569c2b..083ffd397154 100644
> --- a/drivers/pinctrl/intel/pinctrl-cherryview.c
> +++ b/drivers/pinctrl/intel/pinctrl-cherryview.c
> @@ -1601,6 +1601,22 @@ static int chv_gpio_probe(struct chv_pinctrl *pctrl, int irq)
>  			clear_bit(i, chip->irq.valid_mask);
>  	}
>  
> +	/*
> +	 * The same set of machines in chv_no_valid_mask[] have incorrectly
> +	 * configured GPIOs that generate spurious interrupts so we use
> +	 * this same list to apply another quirk for them.
> +	 *
> +	 * See also https://bugzilla.kernel.org/show_bug.cgi?id=197953.
> +	 */
> +	if (!need_valid_mask) {
> +		/*
> +		 * Mask all interrupts the community is able to generate
> +		 * but leave the ones that can only generate GPEs unmasked.
> +		 */
> +		chv_writel(~(BIT(pctrl->community->nirqs) - 1),
> +			   pctrl->regs + CHV_INTMASK);
> +	}
> +
>  	/* Clear all interrupts */
>  	chv_writel(0xffff, pctrl->regs + CHV_INTSTAT);
>  
> -- 
> 2.15.0
> 

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Linus Walleij]

On Fri, Dec 1, 2017 at 11:25 AM, Mika Westerberg
<mika.westerberg@linux.intel.com> wrote:

> Guenter Roeck reported an interrupt storm on a prototype system which is
> based on Cyan Chromebook. The root cause turned out to be a incorrectly
> configured pin that triggers spurious interrupts. This will be fixed in
> coreboot but currently we need to prevent the interrupt storm from
> happening by masking all interrupts (but not GPEs) on those systems.
>
> Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
> Reported-by: Guenter Roeck <linux@roeck-us.net>
> Reported-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Fixes:?
Cc: stable@vger.kernel.org?

Yours,
Linus Walleij

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Guenter Roeck]

On 12/02/2017 07:51 AM, Linus Walleij wrote:
> On Fri, Dec 1, 2017 at 11:25 AM, Mika Westerberg
> <mika.westerberg@linux.intel.com> wrote:
> 
>> Guenter Roeck reported an interrupt storm on a prototype system which is
>> based on Cyan Chromebook. The root cause turned out to be a incorrectly
>> configured pin that triggers spurious interrupts. This will be fixed in
>> coreboot but currently we need to prevent the interrupt storm from
>> happening by masking all interrupts (but not GPEs) on those systems.
>>
>> Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
>> Reported-by: Guenter Roeck <linux@roeck-us.net>
>> Reported-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
>> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> 
> Fixes:?
> Cc: stable@vger.kernel.org?

Fixes: bcb48cca23ec ("pinctrl: cherryview: Do not mask all interrupts in probe")

Mika, any chance to send an update soon ? I would like to apply the patch to
chromeos-4.14, and I would prefer to use the final version.

On a side note, I confirmed that the problem is also seen on release systems
(Acer Chromebook R11) as well as on other systems using the same chipset,
such as Samsung Chromebook 3.

Thanks,
Guenter

Re: [PATCH] pinctrl: cherryview: Mask all interrupts on Intel_Strago based systems

[Mika Westerberg]

On Sat, Dec 02, 2017 at 09:25:13AM -0800, Guenter Roeck wrote:
> On 12/02/2017 07:51 AM, Linus Walleij wrote:
> > On Fri, Dec 1, 2017 at 11:25 AM, Mika Westerberg
> > <mika.westerberg@linux.intel.com> wrote:
> > 
> > > Guenter Roeck reported an interrupt storm on a prototype system which is
> > > based on Cyan Chromebook. The root cause turned out to be a incorrectly
> > > configured pin that triggers spurious interrupts. This will be fixed in
> > > coreboot but currently we need to prevent the interrupt storm from
> > > happening by masking all interrupts (but not GPEs) on those systems.
> > > 
> > > Link: https://bugzilla.kernel.org/show_bug.cgi?id=194945
> > > Reported-by: Guenter Roeck <linux@roeck-us.net>
> > > Reported-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> > > Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
> > 
> > Fixes:?
> > Cc: stable@vger.kernel.org?
> 
> Fixes: bcb48cca23ec ("pinctrl: cherryview: Do not mask all interrupts in probe")

Right, thanks. I'll add them to v2.

> Mika, any chance to send an update soon ? I would like to apply the patch to
> chromeos-4.14, and I would prefer to use the final version.

I'll send an updated version tomorrow.