Menu

Unable to generate Interrupt Vector Table

kent_twt4
2020-05-14
2020-05-16
  • kent_twt4

    kent_twt4 - 2020-05-14

    Back to playing with the ATTiny 0, 1-series chips. In particular the ATTiny416- XNano. I am trying to put together a Boost regulator for a friend. Taking one thing at a time, and that is going OK until now where I seem to be stuck on Interrupts. They are defined in my experimental chip data file, which is both auto generated, and then hand crafted when I get in a pickle.

    GCB does not seem to be picking up my Interrupt Vector definitions from the chipdata file. When using sub Interrupt no vector gerneration, and when I try a the On Interrupt also no luck. Tried aliasing the Timer1Overflow to the TCA0_OVF, and that did not work either. When using the On Interrupt, I get an invalid interrupt event error. So it could be the chip? or the interrupt definition TCA0_OVF are not being picked up by GCB?

    Currently using GCB 98.04

     

    Last edit: kent_twt4 2020-05-14
  • Anobium

    Anobium - 2020-05-14

    0.98.04? :-(

    Is this the hand crafted .dat file? The auto-generation was never completed. So, it is 1/2 done?

     
  • kent_twt4

    kent_twt4 - 2020-05-14

    Yes 98.04

    Combo of mostly auto generated then hand crafted when lacking or inconvienant. Say for instance, the attiny416 naming convention for PortB is PortB_Out, so would alias PortB register location as PortB,1060 which is the same as PortB_Out,1060. Same for the SREG and CPU_SREG and others.

    I honestly forget just how far along I got. But was feeling a bit overwhelmed when generating all of the tiny or mega files and then trying to reconcile against port mux tables or previous GCB chipdata files. Think the Register/Bits got done, but I can't remember if I actually got a compile that generated all the files including package/pins without error? Remember lastly trying to stamp out the multitude of package/pins combos to the point of ignoring some packages.

    Have never ported to current GCB tools, so there is that. Not really ready to dive back in at this time.

     
  • Anobium

    Anobium - 2020-05-14

    I am working flat out to get the next release out.

    i had a quick look and this is very broken. There would be much time required to resolve. And, I would only fix for latest version.

    When I corrected the event data the compler fell over with a serious error. It will take time to resolve and work through.

    Can you workaround by polling?

     
  • kent_twt4

    kent_twt4 - 2020-05-14

    Yes I'll try polling in a tight loop and should be able to schedule other functions accordingly. Why couldn't I think of that? the old gray matter is deteriorating.

    Thanks for the peek, and good luck with the new release. Your efforts are greatly appreciated.

     
    • Anobium

      Anobium - 2020-05-14

      Cheers,

      You need to get off that old release. We need your help on the newer releases. :-)

       
  • kent_twt4

    kent_twt4 - 2020-05-16

    Demo for polled TCA0 Overflow Interrupt. Had to adjust calculated period registers. Tweeked datfile to accomodate 16 bit period register, in lieu of some sort of macro.

     

    Last edit: kent_twt4 2020-05-16

Log in to post a comment.

Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.