On Tuesday 13 January 2009, Tom Wheeler wrote:
> I am curious if anyone has successfully used the DaVinci gpio's as
> interrupts in the DaVinci git repository.
Sure. Several dm355evm IRQs use the GPIOs:
dm355evm:~# cat /proc/int*
CPU0
12: 43 AINTC musb_hdrc
16: 0 AINTC edma
17: 0 AINTC edma_error
23: 0 AINTC mmc0
26: 4239 AINTC mmc0
27: 8681 AINTC mmc1
31: 0 AINTC mmc1
32: 21057 AINTC clockevent
33: 4 AINTC free-run counter
39: 23929 AINTC i2c_davinci
40: 493 AINTC serial
65: 965 GPIO eth0
69: 26 GPIO dm355evm_keys
Err: 0
dm355evm:~#
> The gpio_to_irq code in
> arch/arm/mach-davinci/include/mach/gpio.h appears to incorrectly
> calculate the irq number. For instance, the irq returned for GPIO0 is
> 64 when it should be 48.
There are two ways to use certain GPIOs as IRQs: through the
GPIO controller, and directly. For GPIO-0 that's the difference
between those numbers ... 64 goes through the controller.
I've *never* seen the "direct" route work, but going through
the GPIO controller has always worked just fine. The issue may
be that if the "bank 0" IRQ is enabled, the "direct" versions
of those low numbered IRQs won't work. So far as I can tell,
that's no loss.
- Dave
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
