Re: Antworten: Re: [avr-gcc-list] Problem with the calling convention

[Dmitry K.]

> As E. Weddington wrote:
> > Do we know if this is a regression? If we can discover that this is
> > a regression then we're more likely to get a fix in say 3.4.2
> > instead of having to wait until 3.5.
>
> Right you are!
>
> I tried it on a GCC 3.3.1 I've still got around on one of my machines,
> and it properly moves arg `d' away into r14..r17, and uses these
> registers in the following statements.
>
> So yes, it is a regression.

3.3.2 and 3.3.4 produce true code (`d' saved in r14..r17 before calls).