TMS320F28379D: Problem defining an 8-bit variable (uint8_t)

Sinan,

The C28x core is based on 16-bit addressable word length and a byte on this architecture is defined as 16-bits. 

We do provide some ability to force manipulation at the 8-bit level, please see this wiki article: https://software-dl.ti.com/ccs/esd/documents/c2000_byte-accesses-with-the-c28x-cpu.html

Best,

Matthew

Hi, Matthew:

MatthewPate said:

We do provide some ability to force manipulation at the 8-bit level, please see this wiki article: https://software-dl.ti.com/ccs/esd/documents/c2000_byte-accesses-with-the-c28x-cpu.html

I've looked through the examples here but the examples didn't work. Can you show me a working 8 bit example?

I need to convert 16 bits to 8 bits. 

I need 8 bit variables (uint8_t) and arrays like in my application below. In my application below, my function takes 8 bit parameters. And my function returns 8 bit variable.

uint8_t ReadApp(uint8_t p1, uint8_t p2)
{
    uint8_t Data[2];
    ...


    return Data[1];
}

Sinan,

Thanks and best regards.

Sinan,

The _byte intrinsic allows for access/manipulation at the 8-bit level of an existing 16-bit memory address. 

The C2000 cannot instantiate variables at a 8-bit word boundary since it is a 16-bit machine and the lowest level of addressing is at the 16-bit level.

In order to manipulate at the 8-bit level you'll have to declare your vars as Uint16's and use the byte to pack/unpack at the 8-bit level.

Best,

Matthew

Hi, Matthew:

I'm again having problem with defining 8 bit variable.

MatthewPate said:

In order to manipulate at the 8-bit level you'll have to declare your vars as Uint16's and use the byte to pack/unpack at the 8-bit level.

I did what you told me. However, that didn't work.

The code statement below is not working. I want to get the 0xCD variable.

uint16_t data = 0xABCD;

__mov_byte(data, 0);

Sinan,

Thanks.