Objective Development Forums

Re: HW interrupt problem

2010-12-07T21:53:51+02:00

maxi wrote: you must call usbPoll() at least once every 50 ms.

How silly of me. Thanks for pointing out the obvious
Back to the drawing board for my FW design.

Statistics: Posted by Guest — Tue Dec 07, 2010 9:53 pm

Re: HW interrupt problem

2010-12-07T21:47:36+02:00

Anonymous2 wrote: Am I doing something wrong, or is this as specified?

Yes, the 250 ms delay in an interrupt service routine is a BIG no no, regardless of the fact that the you must call usbPoll() at least once every 50 ms. Better increment a (volatile)global counter and toggleGreen every so many iterations.

Hope it helps

Statistics: Posted by maxi — Tue Dec 07, 2010 9:47 pm

HW interrupt problem

2010-12-07T00:03:59+02:00

Hello all,

I try to use v-usb in a project where I need HW interrupts for the ATMega. I use a modified 'custom-class' project. The USB is connected to INT0 and it works nicely.

In INT1 I have a peripheral system that can cause an interrupt. As an interrupt service I have (it is dummy for now)

Code:

ISR( INT1_vect, ISR_NOBLOCK )
{
        toggleGreen();
        _delay_ms(250);
}

And yes, I can toggle the green led with the external interrupt.

Now, if I cause INT1-interrupts then occasionally my PC's host program won't find the USB device.

I run the 'set-led' program in a tight loop in the PC.

So it seems that the INT0 (USB) cannot override the INT1(external). Despite the ATMega docs saying it should. AVR-libc docs say passing ISR_NOBLOCK to the ISR function should not disable interrupts. Am I doing something wrong, or is this as specified?

thanks

Statistics: Posted by Guest — Tue Dec 07, 2010 12:03 am