On Mon, Apr 27, 2009 at 01:11:39PM +0200, Robert P. J. Day wrote:
> On Mon, 27 Apr 2009, Mikael Starvik wrote:
> > should that not be "spin_unlock_irqrestore()"?
>
> um ... the code is correct, or my observation is correct? in any
> event, i'll leave this with you.
Hi, I've just added the following patch to the CRIS-tree:
=============
Subject: [PATCH] CRISv32: Fix typo compile error in ARTPEC-3 gpio driver.
arch/cris/arch-v32/drivers/mach-a3/gpio.c:
+spin_lock_irqrestore(&gpio_lock, flags);
arch/cris/arch-v32/drivers/mach-a3/gpio.c:
+spin_lock_irqrestore(&gpio_lock, flags);
should that not be "spin_unlock_irqrestore()"?
The code in question was inside an (most often) undefined ifdef.
Reported-by: "Robert P. J. Day" <[email protected]>
Signed-off-by: Jesper Nilsson <[email protected]>
---
arch/cris/arch-v32/drivers/mach-a3/gpio.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/cris/arch-v32/drivers/mach-a3/gpio.c b/arch/cris/arch-v32/drivers/mach-a3/gpio.c
index 7a87bc0..97357cf 100644
--- a/arch/cris/arch-v32/drivers/mach-a3/gpio.c
+++ b/arch/cris/arch-v32/drivers/mach-a3/gpio.c
@@ -681,7 +681,7 @@ static int virtual_gpio_ioctl(struct file *file, unsigned int cmd,
shadow |= ~readl(dir_oe[priv->minor]) |
(arg & changeable_bits[priv->minor]);
i2c_write(VIRT_I2C_ADDR, (void *)&shadow, sizeof(shadow));
- spin_lock_irqrestore(&gpio_lock, flags);
+ spin_unlock_irqrestore(&gpio_lock, flags);
break;
case IO_CLRBITS:
spin_lock_irqsave(&gpio_lock, flags);
@@ -690,7 +690,7 @@ static int virtual_gpio_ioctl(struct file *file, unsigned int cmd,
shadow |= ~readl(dir_oe[priv->minor]) &
~(arg & changeable_bits[priv->minor]);
i2c_write(VIRT_I2C_ADDR, (void *)&shadow, sizeof(shadow));
- spin_lock_irqrestore(&gpio_lock, flags);
+ spin_unlock_irqrestore(&gpio_lock, flags);
break;
case IO_HIGHALARM:
/* Set alarm when bits with 1 in arg go high. */
--
1.6.1
/^JN - Jesper Nilsson
--
Jesper Nilsson -- [email protected]