Re: [PATCH 3/7] x86: move / quiesce pre-386 non-16-bit warning

["H.J. Lu" <hjl.tools@gmail.com>]

On Thu, Aug 18, 2022 at 11:13 PM Jan Beulich <jbeulich@suse.com> wrote:
>
> On 18.08.2022 17:30, H.J. Lu wrote:
> > On Thu, Aug 18, 2022 at 12:21 AM Jan Beulich <jbeulich@suse.com> wrote:
> >>
> >> On 17.08.2022 21:21, H.J. Lu wrote:
> >>> On Tue, Aug 16, 2022 at 12:31 AM Jan Beulich <jbeulich@suse.com> wrote:
> >>>>
> >>>> Emitting this warning for every insn, including ones having actual
> >>>> errors, is annoying. Introduce a boolean variable to emit the warning
> >>>> just once on the first insn after .arch may have changed the things, and
> >>>> move the warning to output_insn(). (I didn't want to go as far as
> >>>> checking whether the .arch actually turned off the i386 bit, but doing
> >>>> so would be an option.)
> >>>> ---
> >>>> Otoh I wonder whether switching to a pre-386 architecture shouldn't
> >>>> automatically move to CODE_16BIT: Us emitting operand- or address-size
> >>>> prefixes violates the architecture specification. Alternatively we
> >>>> could outright reject such .arch directives when not already in 16-bit
> >>>> mode.
> >>
> >> Mind me asking - no opinion here?
> >
> > We shouldn't change the current behavior to avoid any surprises.
>
> And continue to emit non-working code. Recall that the warning message
> talks of only "addressing mode", which even I initially took to mean
> what it say, and hence being entirely bogus to emit for insns without
> memory operand (or anything else susceptible to address size setting).

If it is a real error, we should issue an error.

> I did actually try to find better wording for the warning, but
> couldn't come up with anything halfway sensible and not overly long.
>
> Jan



-- 
H.J.