David Brownell wrote: > On Friday 08 May 2009, Narnakaje, Snehaprabha wrote: >> I had tried Tom's patch to use the direct interrupts and it >> worked fine (no netdev transmit errors) until recently. >> But it stopped working with the latest on linux-davinci git >> (has GPIO driver changes from Dave). Basically the link >> itself is not getting detected. > > This is explained by Kevin's observation that for GPIO IRQs, > the edge control registers always need to be used ... even > for when banked IRQs are not used. That it worked before > was a function of a minor bug. > > Those non-banked IRQs will need a set_type() method that > works just like gpio_irq_type(). Their enable() methods > would call gpio_irq_enable() and davinci_unmask_irq(), more > or less. Their disable() methods could use the standard > AINTC code. > > - Dave
Noone has complained about the other GPIO hardware bug. That is, the edge registers (falling/rising) both do the same thing. By that I mean, if you enable rising edge interrupts, then disable falling edge interrupts, all interrupts will be off. If you disable falling edge interrupts and then enable rising edge interrupts, both edge interrupts will be on. Please feel free to verify this as it sounds absurd. Troy _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
