C code clarification for SRAM aliasing

Participant

Hi Carmine,

In the pag.11 of the book there is the next C line to convert SRAM address to alias region:

#define BITBAND_SRAM(a,b) ((ALIAS_SRAM_BASE + ((uint32_t)&(a)-BITBAND_SRAM_BASE)*32 + (b*4)))

Is (uint32_t)&(a) code an error? If not could you explain what it does? For peripheral aliasing you are writing (uint32_t)a.

Thanks,
Alberto

Carmine Noviello

Keymaster

Hi Alberto,
Well, no. It’ depends on your favorite way to use the BITBAND_SRAM() macro
The current macro definition is made so that you can use it in the following way:

uint32_t myvar;
…
unit32_t *bbmyvar = BITBAND_SRAM(myvar, 0);

If you define the macro in that other way:

#define BITBAND_SRAM(a,b) ((ALIAS_SRAM_BASE + ((uint32_t)&(a)-BITBAND_SRAM_BASE)*32 + (b*4)))

Then you have to use it differently:

…
unit32_t *bbmyvar = BITBAND_SRAM(&myvar, 0);

To alias a given variable, which is bound to a given memory address, then you have to know that address.

The difference with BITBAND_PERI() is that the peripheral macros (for example, GPIOA_ODR) are nothing more than an address in the peripheral mapped region.

Hope this clarifies.

This reply was modified 2 years, 4 months ago by Carmine Noviello.
This reply was modified 2 years, 4 months ago by Carmine Noviello.
This reply was modified 2 years, 4 months ago by Carmine Noviello.