Re: [avr-libc-dev] [bugs #4101] setjmp/longjmp destroy changes in global registers

[E. Weddington]

Paul Schlie wrote:

> > From: Joerg Wunsch <address@hidden>
> >    
> >
> > > As Paul Schlie wrote:
> > > - ok, I accept that possibility; please correct my misunderstanding.
> > >      
> > We basically implement the ABI you're suggesting.
> >    
>
> - which is good, even if only by accident; but it needs to be documented
>  which subset of caller-saved registers are safe/prefered to use, if
>  desired, and are considered officially not available for asm code use.
>  

If the avr-libc documentation is found lacking in this respect, please 
submit a patch.

>  
>
> > While we do have a lot
> > of assembler files (I'm not particularly happy with the situation
> > either, e.g. fplib is rather hard to maintain), they follow the exact
> > same ABI as the C compiler, so I don't see any difference in that.
> >    
>
> - almost, the compiler can/will avoid using compile-time declared reserved
>  global registers; asm routines can't, therefore must restrict their
>  their use in anticipation of the possibility, regardless if they're
>  used or not. (as in fact the ABI is actually somewhat soft, not hard)
>
>  

In practice, the "reserve global registers" feature is a C language 
extension of GCC.
<http://gcc.gnu.org/onlinedocs/gcc-3.4.3/gcc/Global-Reg-Vars.html#Global-Reg-Vars>

The vast majority of people do not use this feature in AVR GCC. And the 
people who do use this, have been sufficiently warned that they're 
playing with fire.

avr-libc doesn't have to do anything to accomodate this feature. The 
return on investment would approach zero.

Eric