This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.


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

Re: FYI: -march/-mcpu patch for mips



> 
> Here it is.
> 

I like it mostly.

You should probably put in a check to make sure that both -march and
-mcpu are both not given on the command line (and different).

What I would do, since both can be on the gcc command line, is to
produce a warning if both are given and use the value from -march.

Otherwise it looks good to me.  We'll need Nick to approve it though.

-eric

>  
> +    case OPTION_MARCH:
>      case OPTION_MCPU:
>        {
> +     int cpu = CPU_UNKNOWN;
> +
>       /* Identify the processor type.  */
> -     if (strcasecmp (arg, "default") == 0)
> -       mips_cpu = CPU_UNKNOWN;
> -     else
> +     if (strcasecmp (arg, "default") != 0)
>         {
>           const struct mips_cpu_info *ci;
>  
>           ci = mips_cpu_info_from_name (arg);
>           if (ci == NULL || ci->is_isa)
> -           as_bad (_("invalid architecture -mcpu=%s"), arg);
> +           as_fatal (_("invalid architecture -march/mcpu=%s"), arg);
>           else
> -           mips_cpu = ci->cpu;
> +           cpu = ci->cpu;
> +       }
> +
> +     switch (c)
> +       {
> +       case OPTION_MARCH:
> +         mips_arch = cpu;
> +         break;
> +       case OPTION_MCPU:
> +         mips_cpu = cpu;
> +         break;
>         }



-- 
I will not grease the monkey bars


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