[PATCH v5 8/8] x86: further re-work insn/suffix recognition to also cover MOVSX

[H.J. Lu]

On Wed, Oct 26, 2022 at 2:07 AM Jan Beulich <jbeulich@suse.com> wrote:
>
> On 25.10.2022 19:10, H.J. Lu wrote:
> > On Tue, Oct 25, 2022 at 12:29 AM Jan Beulich <jbeulich@suse.com> wrote:
> >>
> >> PR gas/29524
> >>
> >> Having templates with a suffix explicitly present has always been
> >> quirky. After prior adjustment all that's left to also eliminate the
> >
> > I prefer keeping the current MOVS behavior.  Please submit a new patch
> > set without MOVS changes.
>
> You're kidding? If you _still_ think this patch isn't worth it despite
> its merits, and despite it now only reducing the set of templates and
> adjusting/extending the testsuite, without any changes to tc-i386.c you

Without this change, the Pass2 change can be dropped.

> can simply give your okay to the first 7 patches. Why would I spam the
> list with an identical re-submission of those first 7 patches?
>
> Independent of that I'd like to understand the reasons for your
> preference, in particular when considering the benefits as well as the
> reported bug which is being fixed (and which you marked as WONTFIX for
> a subjective reason). So far your objection was that the change
> complicates the code. Now it actually simplifies it slightly by
> halving the number of involved templates.
>
> Jan



-- 
H.J.