gnupic: Thread: stimulus/timer problem is back

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

Hi !

Next time stimulus/timer problem:

I put into the simulator the following signal:
>stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
>node test_node
>attach test_node portb0   
and I get a allways high signal on portb0

and the TMR0 Problem is back in 0.19.2:

>TMR0: bug TMR0 is larger than 255...
>cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
>calculated value = 20000086   


Martin ####@####.####

On Fri, 2 Jun 2000, Martin Berentsen wrote:

> Hi !

Martin,

FYI, I got about 10 copies of this message!

> 
> Next time stimulus/timer problem:
> 
> I put into the simulator the following signal:
> >stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end
> >node test_node
> >attach test_node portb0   
> and I get a allways high signal on portb0
> 
> and the TMR0 Problem is back in 0.19.2:
> 
> >TMR0: bug TMR0 is larger than 255...
> >cpu->cycles.value = 44f  last_cycle = ffffffff0000001a  prescale = 8 
> >calculated value = 20000086   

I'll be going out of town in the next few hours, so I'm not going to be able to
fix these problems within at least 2 weeks. Sorry.


I can help with the stimulus. First I made a mistake in a post a few days
ago. To do the square wave stimulus I left out one key component:

gpsim> stimulus sqw period 10000 high_time 1300 start 30000 initial_state 0 end

When you do this, gpsim will create a name for the stimulus. To see the name,
enter the stimulus command without any arguements:

gpsim> stimulus 
Stimulus List
stimulus porta0
stimulus porta1
stimulus porta2
stimulus porta3
stimulus porta4
stimulus porta4
stimulus portb0
stimulus portb1
stimulus portb2
stimulus portb3
stimulus portb4
stimulus portb5
stimulus portb6
stimulus portb7
stimulus sa_square_wave
returning from dump

(The last line is a debug message). The last stimulus in the list is the one
that gpsim has assigned for the sqw.

Create the node:

gpsim> node test_node

Here's where I messed up in the previous post. Attach the stimulus and an I/O
pin to the node:

gpsim> attach test_node portb0 sa_square_wave

Now, that will properly attach the stimulus to portb bit 0. Unfortunately, I
just tried this myself and discovered that the stimulus is always high. To which
I can only respond with dammit. There's definitely a bug here.

If I stand back and take a broader look at the stimulus stuff, I can reiterate
what Byron politely noted: gpsim's stimulus mechanism is complicated. Since I'm
going to need a robust and easy to use (understand) stimulus interface to
support modules, expect some major enhancemants in this area over the next few
months.

Meanwhile, Martin, I suggest that you hold off on using the stimuli until at
least gpsim-0.19.3 is out. Sorry.

Regards,
Scott