This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
RE: Enable EDB7xxx interrupts
- To: "'Lewin A.R.W. Edwards'" <larwe at larwe dot com>,"'eCos discussion'" <ecos-discuss at sources dot redhat dot com>
- Subject: RE: [ECOS] Enable EDB7xxx interrupts
- From: "Trenton D. Adams" <tadams at extremeeng dot com>
- Date: Tue, 26 Jun 2001 14:42:06 -0600
- Organization: Extreme Engineering
> >I did test the EINT1 interrupt on the CPU as you suggested with the
> >multi tester by pulling it low, and fired just fine.
>
> OK, that means your software is good, but we still haven't
> narrowed the
> problem down beyond these possibilities:
>
> * PS6700 not configured correctly to generate interrupts
> * PS6700 configured correctly and generating interrupts, but those
> interrupts not [electrically] making it to the CPU
> * PS6700 fried
Have you ever seen a device that shows that it's generating an interrupt
by looking at it's internal registers, but the actual interrupt is not
making it to the CPU?
> >Yep, already did that, a jumper is required for the FIQ, but not the
> >IRQ line. The FIQ jumper has been on the entire time however.
>
> If the line is tied to another outputs from a different
> device, or has an
> overly aggressive pullup/pulldown then the 6700 might not be
> able to pull
> the CPU's input high/low enough to assert an interrupt
> request if the other
> device's output is not tristated. When you poked it with your
> wire, you
> forced the issue.
>
> If possible it could be instructive to isolate the int out
> pin(s) from the
> 6700 and see if they show activity when electrically isolated
> from the rest
> of the circuit.
How would you do this when they are wired into the board?