CCS C Software and Maintenance Offers
FAQFAQ   FAQForum Help   FAQOfficial CCS Support   SearchSearch  RegisterRegister 

ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

CCS does not monitor this forum on a regular basis.

Please do not post bug reports on this forum. Send them to support@ccsinfo.com
write_eeprom and PIC16F684

 
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion
View previous topic :: View next topic  
Author Message
Fabri



Joined: 22 Aug 2005
Posts: 275

View user's profile Send private message

write_eeprom and PIC16F684
PostPosted: Sat Aug 26, 2006 12:43 am     Reply with quote
Hi Everybody,

I have a problem with write_eeprom and PIC16F684. The same routine work fine with other pic, for example PIC16F877 and PIC16F690.

this is the part of routine:


Quote:

#define adr_zero_l 10
#define adr_zero_m 11
#define adr_zero_h 12

unsigned int32 zero;

zero=0x00023280;
write_eeprom(adr_zero_h,make8(zero,2));
write_eeprom(adr_zero_m,make8(zero,1));
write_eeprom(adr_zero_l,make8(zero,0));


When I read, in second step, the 3 location I found the last value I tryed to write.

In adr_zero_h I found 0x80;
In adr_zero_m I found 0x80;
In adr_zero_l I found 0x80;

I tryed to save the 24bit of zero in more step and writing any byte and testing all but the result doesn't change.

I belive there's a problem in writing routine.
My compiler il PCWH V 3.249.

Somebody help me about ?

Regards,













[/quote]
Ttelmah
Guest
PostPosted: Sat Aug 26, 2006 8:48 am     Reply with quote
I have just tested it with the following code, and 3.249:
Code:

#define adr_zero_l 10
#define adr_zero_m 11
#define adr_zero_h 12


void main()
{
   unsigned int32 zero;
   int8 temp;
   zero=0x00023280;
   write_eeprom(adr_zero_h,make8(zero,2));
   write_eeprom(adr_zero_m,make8(zero,1));
   write_eeprom(adr_zero_l,make8(zero,0));
   temp=read_eeprom(adr_zero_h);
   printf("high : %u/n",temp);
   temp=read_eeprom(adr_zero_m);
   printf("mid : %u/n",temp);
   temp=read_eeprom(adr_zero_l);
   printf("low : %u/n",temp);

   while(1) ;
}

and it worked fine, giving 2, 50, 128. Also, reading the chip in the programmer after running, showed the bytes happily sitting in the EEPROM.
Two possibilities. The first is that you have a chip with a faulty EEPROM (or have 'worn it out', by doing some program loop with rapid writes - it is suprising how quickly the EEPROM like can be used up like this. The second is that something else is wrong in the code.

Best Wishes
Fabri



Joined: 22 Aug 2005
Posts: 275

View user's profile Send private message

PostPosted: Sat Aug 26, 2006 1:43 pm     Reply with quote
Hi Ttelmah,

As I told you, the same program work fine with PIC16F690. I changed also chip without difference. I developed the program for PIC16F690 with ICD2 debugger and after I traslate it to PIC16F684 and I controlled data printing out as you done.

I don't realize were's the problem....

By the way, shall I install more then one version of compiler in the same PC ?

Thanks for test compiler versione,

Fabri
Ttelmah
Guest
PostPosted: Sat Aug 26, 2006 3:24 pm     Reply with quote
I tested it on a 684...
It runs fine.

Best Wishes
Fabri



Joined: 22 Aug 2005
Posts: 275

View user's profile Send private message

PostPosted: Mon Aug 28, 2006 1:59 am     Reply with quote
Hi,

The problem was in printing out 32 bit of zero. Actually the eeprom location was right.

I used:
Quote:

printf("zero= %LX",zero);


So I print out the 32 bit in three time.

Which the right way to print out 32 in one time ?

Thanks for help,[/quote]
Guest
PostPosted: Mon Aug 28, 2006 4:18 am     Reply with quote
"%LX", prints out the whole 32bit number.
What you show is 'right'.
If you want the output to be constant length (as your input value, with leading zeroes), then use "%08LX".
To print the number as input, use "0x%08LX".
As an 'aside', though make8, and make32 exist, and work, they are 'non portable', if you ever want to move your code to another processor. Personally, I prefer to do the transfers using a union, which is standard C.
So:
Code:


int8 ctr;
union {
   int8 b[4];
   int32 zero;
} join;

join.zero=0x00023280;
for (ctr=0;ctr<3;ctr++) write_eeprom(addr_zero_l+ctr,join.b[ctr+1]);
//Because we only transfer the low three bytes, there is an offset needed.


//Transfers all three bytes to the eeprom

join.zero=0;
for (ctr=0;ctr<3>0;) printf("Byte %d = %x/n",ctr,join.b[(ctr--)+1]);


Best Wishes
Fabri



Joined: 22 Aug 2005
Posts: 275

View user's profile Send private message

PostPosted: Mon Aug 28, 2006 5:40 am     Reply with quote
Hi Everibody,

I'm sorry for confusion but I'm quite nervous about this software.

Printing out work fine with option "%LX" but the real problem, in my software, is convertion with make32.

I use make32 to load the 3 eeprom location in one 32 bit location so:
Quote:

zero=make32(0,read_eeprom(adr_zero_h),read_eeprom(adr_zero_m)
,read_eeprom(adr_zero_l));


For example:
Location adr_zero_h contain 0x20;
Location adr_zero_m contain 0x30;
Location adr_zero_l contain 0x40;

I'm waiting for zero=0x203040 but make32 return me zero=0x404040.

otherwise,
Quote:

p_h=read_eeprom(adr_zero_h);
p_m=read_eeprom(adr_zero_m);
p_l=read_eeprom(adr_zero_l);

zero=make32(0,p_h,p_m,p_l);


work fine and in zero I found the correct value,

Thanks for any opinion about,
Ttelmah
Guest
PostPosted: Mon Aug 28, 2006 7:37 am     Reply with quote
The problem here is that the function expects to receive numeric values, _not_ the output from other functions. There is only a single 'scratch' area used for return values, so when the third function is executed the return values from the earlier functions have been destroyed, giving you the '404040' result. You need to transfer the return values into variables, and call make32 with these. Using a union, is safer here, since the values are transferred to the required bytes of the union one at a time, getting rid of this problem, without needing three extra scratch variables...
Note that the manual, specifically lists the values needed for make32, as variables, for this reason.

Best Wishes
Fabri



Joined: 22 Aug 2005
Posts: 275

View user's profile Send private message

PostPosted: Mon Aug 28, 2006 9:27 am     Reply with quote
Hi,

I changed return function with numeric value and now work fine.

I understand now but I don't realize why the software work with PIC16F690 and the return value of make32 is correct as I espect.


Regards,
Fabri
Display posts from previous:   
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group