This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: PING: A Far Less Ambitious AltiVec patch


Ziemowit Laski <zlaski@ziemas.net> writes:

> Geoff,
> 
> This is not the latest patch. :-)  Look here instead:
> 
> http://gcc.gnu.org/ml/gcc-patches/2004-02/msg01101.html

I'll look at that.

> >
> > --- 8960,8966 ----
> >           || REGNO (XEXP (x, 0)) >= 32)
> >         output_operand_lossage ("invalid %%P value");
> >         else
> > !       fprintf (file, "%s", reg_names[REGNO (XEXP (x, 0))]);
> >         return;
> >
> >       case 'q':
> >
> > You should update the comment immediately above this code.
> 
> The latest version looks like this:
> 
> --- 9019,9031 ----
>            || REGNO (XEXP (x, 0)) >= 32)
>          output_operand_lossage ("invalid %%P value");
>          else
> + #if TARGET_MACHO
> +       /* The Darwin assembler requires a register name rather than
> +          merely a number.  */
> +       fprintf (file, "%s", reg_names[REGNO (XEXP (x, 0))]);
> + #else
>          fprintf (file, "%d", REGNO (XEXP (x, 0)));
> + #endif
>          return;
> 

That's worse.  Use reg_names always.

> I think I've standardized on just using powerpc*-*-* in all the test
> cases in the latest patch.  Is this OK?

That's good.

-- 
- Geoffrey Keating <geoffk@geoffk.org>


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]