Hi,
On Tue, May 16, 2023 at 02:07:31PM +0200, Linus Walleij wrote:
> On Mon, May 15, 2023 at 11:02 PM Aaro Koskinen <aaro.koskinen@xxxxxx> wrote:
>
> > There is still a problem that smc_irq is not working. It seems when I
> > tested the previous version, I only quickly checked that the eth0 again
> > probes and the link comes up, without testing any actual traffic - sorry.
> >
> > It seems the irq is stuck hi:
> >
> > gpiochip1: GPIOs 208-223, parent: platform/omap_gpio.1, gpio-0-15:
> > gpio-208 ( |smc_irq ) in hi IRQ
> >
> > To fix it I had to add:
> >
> > irq_set_irq_type(gpiod_to_irq(d), IRQ_TYPE_EDGE_RISING);
>
> I just added that onliner, it gives a nice symmetry to the CF card
> IRQ.
It seems you forgot to add braces as they are now needed for if .. else:
+ d = gpiod_get(NULL, "smc_irq", GPIOD_IN);
+ if (IS_ERR(d))
+ pr_err("Unable to get SMC IRQ GPIO descriptor\n");
+ else
+ irq_set_irq_type(gpiod_to_irq(d), IRQ_TYPE_EDGE_RISING);
+ osk5912_smc91x_resources[1] = DEFINE_RES_IRQ(gpiod_to_irq(d));
+
> > But I'm not sure why this is now needed?
>
> Me neither. I hope some people will test the other boards as well,
> or I will fix them as they report breakage, as is custom.
I will try to test all the OMAP1 boards.
A.
