• State Not Answered
  • Locked Locked
  • Replies 9 replies
  • Subscribers 21 subscribers
  • Views 7403 views
  • Users 0 members are here
2025 survey

We are running a survey to help us improve the experience for all of our members. If you see the survey appear, please take the time to tell us about your experience if you can.

Options
Related
How was your experience today?
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Creating a not-so-portable Delay library

James Corbin

Hello,

I'm rather new to Keil C and just getting geared up to use it on my next project. ATM, I'm overwhelmed with the number of options and extensions that are available. I intend to write a C "library" that consists of 3 functions:

Delay()
DelayMS()
DelayUS()

These will be used as follows (examples):

Delay(2);        // creates a 2 second delay
DelayMS(500);    // creates 500ms delay
DelayUS(10);     // creates a 10us delay

I realize a timer could be used, but I'd like to have a few generic approximated delay functions available that do not occupy a timer module.

Is this even possible? I realize exact accuracy would not be possible. But a 502ms delay (or even as much as 10ms off) would be sufficient for my needs.

I'd guess I will need to write a delay loop in assembly and call it from C correct? And of course if a controller or clock got changed, i'd have to update my library..

John

Parents
  • 0 in reply to Andy Neil

    Just for the sake of variety:

    If you have a timer running, you can read the timer value to implement your delay loop. You don't have to set a timer value and code up an interrupt handler, or interfere with the existing use of the timer for another purpose. It's sufficient to read the timer, calculate your desired end value, and just spin while polling the timer value until the timer gets to the end.

    As with anything time-related, this is easiest to code if you know your delay time will be less than half the maximum range of the timer value.

Reply
  • 0 in reply to Andy Neil

    Just for the sake of variety:

    If you have a timer running, you can read the timer value to implement your delay loop. You don't have to set a timer value and code up an interrupt handler, or interfere with the existing use of the timer for another purpose. It's sufficient to read the timer, calculate your desired end value, and just spin while polling the timer value until the timer gets to the end.

    As with anything time-related, this is easiest to code if you know your delay time will be less than half the maximum range of the timer value.

Children
  • 0 in reply to Drew Davis

    Some time ago Jon Ward posted this macro here: http://www.keil.com/forum/docs/thread1940.asp
    You may find it useful for short delays:

    #define NOPS(x)	\ 
	(((x) & 1) ? _nop_() : 0),\ 
	(((x) & 2) ? _nop_(),_nop_() : 0),\ 
	(((x) & 4) ? _nop_(),_nop_(),_nop_(),_nop_() : 0),\ 
	(((x) & 8) ? _nop_(),_nop_(),_nop_(),_nop_(),_nop_(),_nop_(),_nop_(),_nop_() : 0)
    What would be really useful is to be able to make the source code sensitive to the target options such as crystal speed (by means of a predefined macro constant) so that the same code could be used again and again for different projects.

    Of course, the compiler would also have to know how many clock cycles are required for a NOP - since the simulator needs to know this for timeing, presumably this information is available somewhere within Keil.

  • 0 in reply to Graham Cole

    Oh dear! My post did not do that on preview!