How to make a bit structure ?

[milan.rajik]

How to make a bit structure ?

I have ports like PORTB, PORTD, DDRD, etc... In mikroC PRO AVR Compiler we can write like DDRB.F0 = 1 to set bit 0 of DDRB register. I want to do a similar thing in Atmel Studio projects. How can I define a structure and use the DDRx, PORTx, PINx, or any required registers with bit access like DDRB.F4 ?

 

[Dan Mills]

Use a union of a PORT and a bitfield of the appropriate type

If the type underlying say PORTB is say an unsigned char then write a definition something like this:

union {
 PORTB;
 struct {
  unsigned char B0 : 1;
  unsigned char B1 : 1;
  .
  .
  .
  unsigned char B7 :1;
 };
} PORTBbits;

You can then write PORTBbits.B6 = 1;

As the union overlays the two members in memory this will likely have the desired effect, it is however implementation defined per the C standard IIRC and is dependent on PORTB expanding into a dereferenced pointer to a memory mapped peripheral.

Note that the strucure alias onto a (probably) volatile memory mapped port may need a few things tweaking in the optimiser settings to work correctly.

Regards, Dan.

 

[milan.rajik]

Thanks. I want to use bits like DDRB.F0

Can I define like this ?

---

Code C - [expand]
1
2
3
4
5
6
7
8
9
10
11
union {
 PORTB;
 struct {
  unsigned char B0 : 1;
  unsigned char B1 : 1;
  .
  .
  .
  unsigned char B7 :1;
 };
} PORTB;

---

I tried this but it gives error.

---

Code C - [expand]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
union {
    PORTC;
    struct {
        unsigned char F0 : 1;
        unsigned char F1 : 1;
        unsigned char F2 : 1;
        unsigned char F3 : 1;
        unsigned char F4 : 1;
        unsigned char F5 : 1;
        unsigned char F6 : 1;
        unsigned char F7 : 1;
        
    };
} PORTCbits;
 
union {
    DDRC;
    struct {
        unsigned char F0 : 1;
        unsigned char F1 : 1;
        unsigned char F2 : 1;
        unsigned char F3 : 1;
        unsigned char F4 : 1;
        unsigned char F5 : 1;
        unsigned char F6 : 1;
        unsigned char F7 : 1;
        
    };
} DDRCbits;
 
#define LCD_RS  PORTCbits.F0
#define LCD_EN  PORTCbits.F1
#define LCD_D4  PORTCbits.F2
#define LCD_D5  PORTCbits.F3
#define LCD_D6  PORTCbits.F4
#define LCD_D7  PORTCbits.F5
 
#define LCD_RS_Direction DDRCbits.F0
#define LCD_EN_Direction DDRCbits.F1
#define LCD_D4_Direction DDRCbits.F2
#define LCD_D5_Direction DDRCbits.F3
#define LCD_D6_Direction DDRCbits.F4
#define LCD_D7_Direction DDRCbits.F5

---

 

[Raady Here]

what is the error that you get ?

in program line 38 , instead of #define LCD_RS_Direction DDRCbits.F0
you need to write #define LCD_RS_Direction DDRC.F0

for structured bits you can write as Xbits.x

 

[ep.hobbyiest]

i also have basic doubt of this
if i used char, unsigned int or int instead of unsigned char in bit field of structure then will it affect on any thing?

 

[FvM]

union {
 PORTB;
 struct {
  unsigned char B0 : 1;
  unsigned char B1 : 1;
  .
  .
  .
  unsigned char B7 :1;
 };
} PORTBbits;

The unamed union member specification PORTB doesn't comply with the C standard. May the syntax is accepted by some compilers, but if so, no necessarily by others.

Embedded compilers have different means to allocate a memory object at a specific address so that it can be used e.g. as alias for special function registers. But thease methods aren't portable. The only generic C method is to use an initialized pointer to the SFR, which would require a syntax like
pDDRC->F0
But have to ensure that your syntax is actually mapped to direct SFR bit accesses by the compiler.