Change register in background C program

[tweekzilla]

I'm running a toy model to try and switch Gate3[1].Chan[0].AtanEna in a background C program. (the AtanEna is bit 18 of inCtrl). I can read the bit without any problems and it recognizes when I flip the bit using the terminal, however, when I try and set the bit in a background C program it does not work. Am I able to do this? Code below

 

int main(void) 
{
volatile GateArray3 *tga;
int inCtrl;

InitLibrary();
tga = GetGate3MemPtr(1);

inCtrl = tga->Chan[0].InCtrl;
printf("before %i\n",inCtrl); //AtanEna = 1
tga->Chan[0].InCtrl &= ~(1<<18); //AtanEna should = 0
inCtrl = tga->Chan[0].InCtrl;
printf("flipped %i\n",inCtrl); //Still shows 1
CloseLibrary();	
return 0;
}

 

There is no change in the value of inCtrl upon setting the bit to zero.

[Omron Forums Support]

I could be wrong, but I think you may need to change the line where you are actually writing the bit low to the following:

 

tga->Chan[0].InCtrl &= (tga->Chan[0].InCtrl)^(1<<18); //AtanEna should = 0

 

Does that work?

[tweekzilla]

Hi Charles

 

Unfortunately that doesn't work. I know from doing the following the the bit-operation works:

 

inCtrl = tga->Chan[0].InCtrl
inCtrl &= ~(1<<18) // When printed this sets bit 18 to zero

[JeffLowe]

I could be wrong, but I think you may need to change the line where you are actually writing the bit low to the following:

 

tga->Chan[0].InCtrl &= (tga->Chan[0].InCtrl)^(1<<18); //AtanEna should = 0

 

Does that work?

 

Does Gate3.WpKey have to be $AAAAAAAA for this?

[Omron Forums Support]

I tested this and you're right, it doesn't set the 18th bit low. Also, setting Sys.WpKey=$AAAAAAAA has no effect.

 

My PPMAC is malfunctioning right now; otherwise, I would test further. For now, can you just use SetPmacVar()?

 

For example, SetPmacVar("Gate3[1].Chan[0].AtanEna",0.0)

 

This will set AtanEna to 0. Using SetPmacVar is slower than writing the bit directly but it will work in a pinch.

[tweekzilla]

Yes the SetPmacVar does work so long as you make sure that WpKey is set. It's strange though that I can't set it by simply addressing the pointer in C. It might be a write protect issue but I though that there was no write protect when accessing variables from C?

[shansen]

Are you sure that the printf isn't misleading you? The reason I ask is that I ran into a similar problem where I would set a register, and then immediately printf the register to read its value.

 

The problem is that whenever you modify a register in C you are probably doing a read-modify-write operation. If your program hasn't completed the "write" operation before your printf, you'll get the wrong value printed to the terminal.

 

Try inserting a usleep between the write and the printf, see if that changes anything.

 

This would also explain why SetPmacVar works, but the direct write appears not to work.

[tweekzilla]

I put a sleep in (1 second) and I still see the same effect. I also check using the terminal to see if the bit has been set to zero.

 

Another think I tried was just writing the value to the register i.e.

 

tga->Chan[0].InCtrl = 4194375;

 

And that did not work either.

[shansen]

Are you sure you are setting tga->WpKey=$AAAAAAAA before the write?

[tweekzilla]

tga->WpKey=0xAAAAAAAA Did the trick! Thanks everyone and as always it was in the software reference manual although I was being distracted by:

 

"Furthermore, in the Script environment, many elements are write-protected so user code cannot change their values, but these elements can be read at any time. In the C environment, any element for which you are provided access (that is, whose name is provided in the RtGpShm.h header file) can be written to. "

 

Which I wrongly assumed meant that the registers were not write-protected.