MIPS: octeon: Fix GPIO number in IRQ chip private data
authorAlexander Sverdlin <alexander.sverdlin.ext@nsn.com>
Thu, 11 Apr 2013 15:29:39 +0000 (17:29 +0200)
committerRalf Baechle <ralf@linux-mips.org>
Tue, 7 May 2013 23:19:07 +0000 (01:19 +0200)
Current GPIO chip implementation in octeon-irq is still broken, even after upstream
commit 87161ccdc61862c8b49e75c21209d7f79dc758e9 (MIPS: Octeon: Fix broken interrupt
controller code). It works for GPIO IRQs that have reset-default configuration, but
not for edge-triggered ones.

The problem is in octeon_irq_gpio_map_common(), which passes modified "hw" variable
(which has range of possible values 16..31) as "gpio_line" parameter to
octeon_irq_set_ciu_mapping(), which saves it in private data of the IRQ chip. Later,
neither octeon_irq_gpio_setup() is able to re-configure GPIOs (cvmx_write_csr() is
writing to non-existent CVMX_GPIO_BIT_CFGX), nor octeon_irq_ciu_gpio_ack() is able
to acknowledge such IRQ, because "mask" is incorrect.

Fix is trivial and has been tested on Cavium Octeon II -based board, including
both level-triggered and edge-triggered GPIO IRQs.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin.ext@nsn.com>
Cc: David Daney <david.daney@cavium.com>
Acked-by: David Daney <david.daney@cavium.com>
Patchwork: http://patchwork.linux-mips.org/patch/4980/
Acked-by: John Crispin <blogic@openwrt.org>
arch/mips/cavium-octeon/octeon-irq.c

index 156aa6143e1117646ab7b81ccf57ca623a258513..a22f06a6f7cac31e3506619371615b8f4e1c8d4b 100644 (file)
@@ -1032,9 +1032,8 @@ static int octeon_irq_gpio_map_common(struct irq_domain *d,
        if (!octeon_irq_virq_in_range(virq))
                return -EINVAL;
 
-       hw += gpiod->base_hwirq;
-       line = hw >> 6;
-       bit = hw & 63;
+       line = (hw + gpiod->base_hwirq) >> 6;
+       bit = (hw + gpiod->base_hwirq) & 63;
        if (line > line_limit || octeon_irq_ciu_to_irq[line][bit] != 0)
                return -EINVAL;