reload problem in current CVS

[Bernd Schmidt]

> > Reload 0: reload_in (SF) = (reg/v:SF 25)
> >         reload_out (SF) = (reg:SF 0 eax)
> >         FLOAT_INT_REGS, RELOAD_OTHER (opnum = 0)
> >         reload_in_reg: (reg/v:SF 25)
> >         reload_out_reg: (reg:SF 0 eax)
> >         reload_reg_rtx: (reg:SF 0 eax)
> > 
> > This is OK, we don't need secondary memory if we use register eax as reload
> > register.  However, choose_reload_regs decides that it's better to inherit
> > the value of reg 25, which is still lying around in reg 8 (a stack reg).
> 
> For my own edification, when does the matching constraint get enforced?

I've already forgotten some of the details.  I believe we are still using
eax as reload register, but copying from reg 8 (rather than reg 25's memory
location).  That copy needs secondary memory.


Bernd