RE: [avr-gcc-list] possible 4.1.2 bug

[Eric Weddington]

> -----Original Message-----
> From: 
> address@hidden 
> [mailto:address@hidden
> org] On Behalf Of John Regehr
> Sent: Wednesday, March 28, 2007 10:15 PM
> To: address@hidden
> Subject: [avr-gcc-list] possible 4.1.2 bug
> 
> 
> void foo (char, char, char, char, char);
> 
> short bar (void)
> {
>    char a = 0;
>    volatile char b = 0;
>    char c = 0;
>    char d = 0;
>    char e = 0;
>    int in;
>    int out;
> 
>    foo (e, d, c, b, a);
> 
>    __asm__ ("swap %B1 \n\t"
>          "swap %C1 \n\t"
>          "swap %D1 \n\t"
>          "ldi %A1, 0xf0 \n\t"
>          "eor %D1, %C1 \n\t"
>          "and %D1, %A1 \n\t"
>          "eor %D1, %C1 \n\t"
>          "eor %C1, %B1 \n\t"
>          "and %C1, %A1 \n\t"
>          "eor %C1, %B1 \n\t"
>          "movw %A0, %C1 \n\t"
>          : "=&r" (out) : "r" (in));
> 
>    return out;
> }

Besides what Joerg wrote, I'm puzzled about the above inline assembly. This
code uses 4 parts (bytes), A through D, as if "in" was a 32-bit value. But
both variables, out and in, are defined to be int, which on an AVR are 16
bits. Either there's a descrepency here, or I'm just not understanding
what's going on here.

Eric