• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 31 subscribers
  • Views 578 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TM4C123GH6PM Datasheet - small error on SSI chapter

Luis Afonso
Guru 20670 points
Other Parts Discussed in Thread: TM4C123GH6PM

Hi everyone,

A bit of small matter.

In the datasheet of the TM4C123GH6PM: DS-TM4C123GH6PM-15741.2722 SPMS376D


There seems to be a small error. On 15.3.3 Interrupts it's written "All of the interrupt events are ORed together before being sent to the interrupt controller"
Is "ORed" something I am not aware of or is it just a little error?
I belive it should be "Read" but just wanted small confirmation (preparing some information for my colleagues and bumped into this)

  • 0
    TI__Guru**** 244400 points
    Hello Luis

    There are multiple interrupt sources. However the NVIC has only one interrupt line for the SSI Module. So all the bits have to be OR-ed together so that any interrupt source when set can cause the CPU to service the interrupt

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Guru 20670 points
    Ah! Thank you Amit. So that's what it's referring too
    Kinda weird for me that, ORed. I can't help but read it O-Red.
  • 0 in reply to Luis Afonso
    Guru 47900 points

    OR'ed has been around & used long before MCUs walked this land.   And - like Sasquatch - has a family: NOR'ed, And'ed, even XOR'ed.

    In dayze of olde we soldered wires or made pcb routings to create "wired-OR connections" - and it became easier to describe these as, "OR'ed."

    As one (modern) specific example - which our small firm finds useful (hopefully others, too) - when connecting to matrix keypads - instead of "continually scanning" those keys (potentially generating EMI while raising MCU current drain) we tie all keypad rows (or columns) to an 8 input Nand gate - and then drive all keypad columns (or rows) to ground.  (strange that - eh?)   Yet - the moment a key-press occurs - that 8 input Nand gate toggles it's output - interrupting our MCU - and only then do we initiate keypad scan.  Costs an IC - but reduces the load upon the MCU & reduces EMI.

    There are cases where continual keypad scan is desirable.  Employing that same scan to drive a multiplexed (usually 7 segment display) is one example.  The efficiency of making that keypad-scan perform multiple, simultaneous tasks is compelling.  Final application dictates what's best and even the solution may be OR'ed!

  • 0 in reply to cb1
    Guru 20670 points
    I guess it's because English isn't my native language that I never knew about that term.

    Basically you are doing the same that those SSI interrupts do. Any can trigger an interrupt and then in the ISR it's checked which one it was.
    That seems like a really useful feature having a interrupt when any key is pressed.
  • 0 in reply to Luis Afonso
    Guru 47900 points

    Luis Afonso said:
    Any can trigger an interrupt and then in the ISR it's checked

    Mais certainement - mon ami - indeed that's the point.

    In some cases - we may need to scan multiple sensors or communication channels - and these may exceed 8 in number.  Thus a single Port proves insufficient as an "Interrupt Source Detector."  (i.e. hasn't enough pins!)   Now - our alternative method - via multi-input Nand Gate - (and with those multiple inputs Nand"ed) a single gate output causes the interrupt - and enables (far greater than 8 independent signal sources) to be examined for source determination...

    MCU by itself - does not always prove "best/brightest."  Same holds true for less adaptable hardware.  Yet - somewhere in-between - (may) lurk the best solution...