RE: [avr-libc-dev] eeprom_read_word causes problems for the compiler

[Wouter van Gulik]

> I agree, this whole setup seems strange. Perhaps if you look in the CVS
> log
> (you can do it online at the avr-libc project), perhaps there may be some
> comments as to why it was originally done this way. Note that the EEPROM
> API
> was originally written (copyrighted) by Marek Michalkiewicz. Perhaps Marek
> (CCed) could explain more why it was done this way?
> 

AFAIK the EEPROM routines are rewritten from scratch by Bjoern Haase (CCed).
The eeprom register addresses changed for a lot devices, there for the old
library was not ok anymore. Bjorn did a good job on rewriting it, his post
to the list describes all routines in details except the read word.

In the CVS log's I see his patch changed made the read word return to
r30:r31. But I can't figure out why.

> I like the idea of your proposed approach, to change the EEPROM routines
> into C functions.

Well we can only do it for the eeprom_read_word. Changing the read_byte
function would cause a growth in size. As I said, Bjoern did a good job :D
I think the advantage of the inline-C-function-with-inline-assembler-call is
the reduced register used. In stead of a call the compiler sees a block
assembler which uses the X register and the R30:R31 (or R25:R24 if we change
it). But it still does not solve the mystery on using r31:r30.

> Is there any way that you would be willing to work up an
> implementation?
>

Yes I will try to put together a patch with the minimal changes. I only need
to know one thing; can I assume that using this construction:

/** \ingroup avr_eeprom
    Read one 16-bit word (little endian) from EEPROM address \c addr. */
uint16_t
eeprom_read_word (const uint16_t *addr)
{
  register uint16_t result asm("r24"); //I can specify r24 only

  __asm__ __volatile__ (
        XCALL " __eeprom_read_word_" _REG_LOCATION_SUFFIX CR_TAB
       : "+x" (addr),
         "=w" (result) //Use w restrict to adiw capable registers
       : );
  return result;
}

that the compiler will not use r25 for other purposes? Note that this goes
inline!
More generally is this a sane thing to do? Is there any objection for doing
it this way?

Wouter