[PATCH] x86: reject architecture settings that are invalid to be set from the command line (v2)

[H.J. Lu]

On Thu, Jun 10, 2010 at 9:06 AM, Jan Beulich <JBeulich@novell.com> wrote:
>>>> On 10.06.10 at 17:55, "H.J. Lu" <hjl.tools@gmail.com> wrote:
>> On Thu, Jun 10, 2010 at 8:13 AM, Jan Beulich <JBeulich@novell.com> wrote:
>>> So far, options like -march=i8086 were accepted despite the assembler
>>> subsequently choking on other consistency checks, leading to reasonably
>>> cryptic error messages. This patch makes it so that impossible
>>> architecure settings are neither accepted nor displayed (i.e. it is now
>>> made sure that those settings can only be used via directives).
>>>
>>> gas/
>>> 2010-06-10  Jan Beulich  <jbeulich@novell.com>
>>>
>>>        * config/tc-i386.c (md_parse_option): Ignore impossible processor
>>>        types.
>>>        (show_arch): New parameter 'check'.
>>>        (md_show_usage): Adjust calls to show_arch().
>>>
>>> --- a/gas/config/tc-i386.c
>>> +++ b/gas/config/tc-i386.c
>>> @@ -8166,6 +8166,11 @@ md_parse_option (int c, char *arg)
>>>              if (strcmp (arch, cpu_arch [j].name) == 0)
>>>                {
>>>                  /* Processor.  */
>>> +                 if (! (strcmp (default_arch, "i386")
>>> +                        ? cpu_arch[j].flags.bitfield.cpulm
>>> +                        : cpu_arch[j].flags.bitfield.cpui386))
>>> +                   continue;
>>
>>
>> This is still wrong:
>>
>> [hjl@gnu-6 gas]$ ./as-new -march=i386 --32 x.s -o x.o
>> Assembler messages:
>> Fatal error: Invalid -march= option: `i386'
>> [hjl@gnu-6 gas]$
>>
>> Please change it to
>>
>> if (!cpu_arch[j].flags.bitfield.cpui386)
>> continue;
>
> Didn't your original response mean only the other instance? I think
> it is perfectly valid to require the options to be passed in sane
> order (i.e. --32 first if you want to use -march=i386). I'm precisely
> trying to reduce the number of cases where that (much less
> meaningful imo) "xx-bit mode not supported" message appears.
>

I don't think there is such a requirement before and I don't
believe we should start it now.

-- 
H.J.