[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20110216024826.GL28005@angua.secretlab.ca>
Date: Tue, 15 Feb 2011 19:48:26 -0700
From: Grant Likely <grant.likely@...retlab.ca>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: LKML <linux-kernel@...r.kernel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Gregory Bean <gbean@...eaurora.org>,
Jean Delvare <khali@...ux-fr.org>,
Lennert Buytenhek <buytenh@...retlab.ca>
Subject: Re: [patch] drivers: gpio: sx150x: Do not access I2C from
mask/unmask functions
On Mon, Feb 14, 2011 at 06:08:51PM +0100, Thomas Gleixner wrote:
> On Mon, 14 Feb 2011, Thomas Gleixner wrote:
>
> Ooops. quilt refresh should be mandatory for quilt mail. One day I'll fix that.
>
> ------->
> Subject: drivers: gpio: sx150x: Do not access I2C from mask/unmask functions
> From: Thomas Gleixner <tglx@...utronix.de>
> Date: Mon, 14 Feb 2011 16:58:11 +0100
>
> irq_chip->irq_mask/unmask are called with interrupts disabled and
> irq_desc->lock held. So we cannot access i2c from this context. That's
> what irq_bus_sync_unlock() is for.
>
> Store the masked information in the chip data structure and update the
> i2c bus from the irq_bus_sync_unlock() callback.
>
> This does not need a while(pending) loop because the update to this is
> always serialized via the bus lock, so we never have more than one pin
> update pending.
>
> Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
> Cc: Andrew Morton <akpm@...ux-foundation.org>
> Cc: Gregory Bean <gbean@...eaurora.org>
> Cc: Jean Delvare <khali@...ux-fr.org>
> Cc: Lennert Buytenhek <buytenh@...retlab.ca>
Applied, thanks.
g.
> ---
> drivers/gpio/sx150x.c | 54 ++++++++++++++++++++++++++++++++++----------------
> 1 file changed, 37 insertions(+), 17 deletions(-)
>
> Index: linux-2.6-tip/drivers/gpio/sx150x.c
> ===================================================================
> --- linux-2.6-tip.orig/drivers/gpio/sx150x.c
> +++ linux-2.6-tip/drivers/gpio/sx150x.c
> @@ -25,6 +25,8 @@
> #include <linux/workqueue.h>
> #include <linux/i2c/sx150x.h>
>
> +#define NO_UPDATE_PENDING -1
> +
> struct sx150x_device_data {
> u8 reg_pullup;
> u8 reg_pulldn;
> @@ -47,8 +49,11 @@ struct sx150x_chip {
> const struct sx150x_device_data *dev_cfg;
> int irq_summary;
> int irq_base;
> + int irq_update;
> u32 irq_sense;
> - unsigned long irq_set_type_pending;
> + u32 irq_masked;
> + u32 dev_sense;
> + u32 dev_masked;
> struct irq_chip irq_chip;
> struct mutex lock;
> };
> @@ -312,9 +317,8 @@ static void sx150x_irq_mask(struct irq_d
>
> chip = container_of(ic, struct sx150x_chip, irq_chip);
> n = d->irq - chip->irq_base;
> -
> - sx150x_write_cfg(chip, n, 1, chip->dev_cfg->reg_irq_mask, 1);
> - sx150x_write_cfg(chip, n, 2, chip->dev_cfg->reg_sense, 0);
> + chip->irq_masked |= (1 << n);
> + chip->irq_update = n;
> }
>
> static void sx150x_irq_unmask(struct irq_data *d)
> @@ -326,9 +330,8 @@ static void sx150x_irq_unmask(struct irq
> chip = container_of(ic, struct sx150x_chip, irq_chip);
> n = d->irq - chip->irq_base;
>
> - sx150x_write_cfg(chip, n, 1, chip->dev_cfg->reg_irq_mask, 0);
> - sx150x_write_cfg(chip, n, 2, chip->dev_cfg->reg_sense,
> - chip->irq_sense >> (n * 2));
> + chip->irq_masked &= ~(1 << n);
> + chip->irq_update = n;
> }
>
> static int sx150x_irq_set_type(struct irq_data *d, unsigned int flow_type)
> @@ -350,7 +353,7 @@ static int sx150x_irq_set_type(struct ir
>
> chip->irq_sense &= ~(3UL << (n * 2));
> chip->irq_sense |= val << (n * 2);
> - chip->irq_set_type_pending |= BIT(n);
> + chip->irq_update = n;
> return 0;
> }
>
> @@ -404,15 +407,29 @@ static void sx150x_irq_bus_sync_unlock(s
>
> chip = container_of(ic, struct sx150x_chip, irq_chip);
>
> - while (chip->irq_set_type_pending) {
> - n = __ffs(chip->irq_set_type_pending);
> - chip->irq_set_type_pending &= ~BIT(n);
> - if (!(irq_to_desc(n + chip->irq_base)->status & IRQ_MASKED))
> - sx150x_write_cfg(chip, n, 2,
> - chip->dev_cfg->reg_sense,
> - chip->irq_sense >> (n * 2));
> - }
> + if (chip->irq_update == NO_UPDATE_PENDING)
> + goto out;
>
> + n = chip->irq_update;
> + chip->irq_update = NO_UPDATE_PENDING;
> +
> + /* Avoid updates if nothing changed */
> + if (chip->dev_sense == chip->irq_sense &&
> + chip->dev_sense == chip->irq_masked)
> + goto out;
> +
> + chip->dev_sense = chip->irq_sense;
> + chip->dev_masked = chip->irq_masked;
> +
> + if (chip->irq_masked & (1 << n)) {
> + sx150x_write_cfg(chip, n, 1, chip->dev_cfg->reg_irq_mask, 1);
> + sx150x_write_cfg(chip, n, 2, chip->dev_cfg->reg_sense, 0);
> + } else {
> + sx150x_write_cfg(chip, n, 1, chip->dev_cfg->reg_irq_mask, 0);
> + sx150x_write_cfg(chip, n, 2, chip->dev_cfg->reg_sense,
> + chip->irq_sense >> (n * 2));
> + }
> +out:
> mutex_unlock(&chip->lock);
> }
>
> @@ -445,8 +462,11 @@ static void sx150x_init_chip(struct sx15
> chip->irq_chip.irq_bus_sync_unlock = sx150x_irq_bus_sync_unlock;
> chip->irq_summary = -1;
> chip->irq_base = -1;
> + chip->irq_masked = ~0;
> chip->irq_sense = 0;
> - chip->irq_set_type_pending = 0;
> + chip->dev_masked = ~0;
> + chip->dev_sense = 0;
> + chip->irq_update = NO_UPDATE_PENDING;
> }
>
> static int sx150x_init_io(struct sx150x_chip *chip, u8 base, u16 cfg)
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists