Message ID | 20190610171103.30903-8-grygorii.strashko@ti.com |
---|---|
State | New |
Headers | show |
Series | gpio: gpio-omap: set of fixes and big clean-up | expand |
On Mon, Jun 10, 2019 at 7:12 PM Grygorii Strashko <grygorii.strashko@ti.com> wrote: > From: Russell King <rmk+kernel@armlinux.org.uk> > > Architectures are single-copy atomic, which means that simply reading > a register is an inherently atomic operation. There is no need to > take a spinlock here. > > Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> > Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com> Patch applied. This makes me wonder how many more instances we have of this kind of mistake in the kernel :/ I guess it is mostly harmless but it sure makes for massive code complexity. I would sure like e.g. drivers/gpio/gpio-mmio.c to avoid taking locks on architectures where this is not a problem given it is used on a plethora of architectures. Yours, Linus Walleij
diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c index cc320d09d074..44a4287cce9e 100644 --- a/drivers/gpio/gpio-omap.c +++ b/drivers/gpio/gpio-omap.c @@ -953,17 +953,10 @@ static void omap_gpio_free(struct gpio_chip *chip, unsigned offset) static int omap_gpio_get_direction(struct gpio_chip *chip, unsigned offset) { - struct gpio_bank *bank; - unsigned long flags; - void __iomem *reg; - int dir; + struct gpio_bank *bank = gpiochip_get_data(chip); - bank = gpiochip_get_data(chip); - reg = bank->base + bank->regs->direction; - raw_spin_lock_irqsave(&bank->lock, flags); - dir = !!(readl_relaxed(reg) & BIT(offset)); - raw_spin_unlock_irqrestore(&bank->lock, flags); - return dir; + return !!(readl_relaxed(bank->base + bank->regs->direction) & + BIT(offset)); } static int omap_gpio_input(struct gpio_chip *chip, unsigned offset)