RE: [PATCH] x86: add missing APX logic to cpu_flags_match()

["Cui, Lili" <lili.cui@intel.com>]

> On 08.01.2024 11:28, Cui, Lili wrote:
> >> On 08.01.2024 09:30, Cui, Lili wrote:
> >>>>>> --- a/gas/config/tc-i386.c
> >>>>>> +++ b/gas/config/tc-i386.c
> >>>>>> @@ -1940,6 +1940,30 @@ cpu_flags_match (const insn_template *t)
> >>>>>>  	      any.bitfield.cpuavx512vl = 0;
> >>>>>>  	    }
> >>>>>>  	}
> >>>>>> +
> >>>>>> +      /* Dual non-APX/APX templates need massaging from what
> >>>>>> + APX_F() in
> >>>> the
> >>>>>> +         opcode table has produced.  While the direct transformation of
> the
> >>>>>> +         incoming cpuid&(cpuid|APX_F) would be to cpuid&(cpuid)
> >>>>>> + /
> >>>>>> cpuid&(APX_F)
> >>>>>> +         respectively, it's cheaper to move to just cpuid / cpuid&APX_F
> >>>>>> +         instead.  */
> >>>>>> +      if (any.bitfield.cpuapx_f
> >>>>>> +	  && (any.bitfield.cpubmi || any.bitfield.cpubmi2
> >>>>>> +	      || any.bitfield.cpuavx512f || any.bitfield.cpuavx512bw
> >>>>>> +	      || any.bitfield.cpuavx512dq || any.bitfield.cpuamx_tile
> >>>>>> +	      || any.bitfield.cpucmpccxadd))
> >>>>>> +	{
> >>>>>> +	  /* These checks (verifying that APX_F() was properly used in
> the
> >>>>>> +	     opcode table entry) make sure there's no need for an "else"
> to
> >>>>>> +	     the "if()" below.  */
> >>>>>> +	  gas_assert (!cpu_flags_all_zero (&all));
> >>>>>> +	  cpu = cpu_flags_and (all, any);
> >>>>>> +	  gas_assert (cpu_flags_equal (&cpu, &all));
> >>>>>> +
> >>>>>> +	  if (need_evex_encoding (t))
> >>>>>> +	    all = any;
> >>>>>> +
> >>>>>
> >>>>>> +	  memset (&any, 0, sizeof (any));
> >>>>>
> >>>>> Wouldn't it make sense to put it in the else branch and clean out
> >>>>> APX-F
> >>>> specifically? Just like you did before.
> >>>>>
> >>>>>   if (need_evex_encoding (t))
> >>>>>     all = any;
> >>>>> else
> >>>>>    any.bitfield.cpuapx_f = 0;
> >>>>
> >>>> That was an alternative I did consider, yes, but the way I've done
> >>>> it is overall more self-consistent imo, at the expense of being
> >>>> less consistent with the
> >>>> AVX/AVX512 logic (the moving of "any" to "all" isn't consistent
> >>>> with that anyway).
> >>>>
> >>>
> >>>   memset (&any, 0, sizeof (any));
> >>>
> >>> I'd say this would make "any" not match the actual value, which
> >>> might be
> >> used later, but it's been cleared here.
> >>
> >> I'm afraid I don't get what you're trying to tell me.
> >>
> >
> > What I mean is that memset will clear the variable "any", there is no problem
> in handling it this way. But I think the following way is more reasonable.
> >
> > For evex it should be:
> > any.bitfield.cpubmi = 1
> > any.bitfield.cpuapx_f = 1
> >
> > For vex it should be:
> > any.bitfield.cpubmi = 1
> > any.bitfield.cpuapx_f = 0
> >
> > Instead of clearing all values in "any".
> 
> But why would you want to have the same in "any" that you already have in
> "all"? That would incur extra checks later in the function for no gain.
> 

Same value but completely different meaning, if you think the current one is better, that's fine with me. By the way, I have a question about " t->cpu_any", is it specific to dual VEX/EVEX templates?

Thanks,
Lili.