[m-rev.] for review: ho_inst_info clarification

[Mark Brown]

On Thu, Nov 12, 2015 at 9:53 AM, Julien Fischer <jfischer at opturion.com> wrote:
>
> Hi Mark,
>
> On Sun, 8 Nov 2015, Mark Brown wrote:
>
>> I've gone through all of the uses of ho_inst_info 'none' in the
>> compiler and identified places I suspect may be a problem, and marked
>> them with an XXX comment. I haven't tried to fix anything in this
>> change; I'll look at fixing inst_match.m, inst_util.m and
>> modecheck_unify.m in separate changes. The problems in inst_match.m
>> look to be responsible for mantis bug 264.
>>
>> The XXX comments in compiler/float_regs.m are for review by Peter.
>>
>> Suggestions welcome for an alternative to the new name
>> none_or_standard_func.
>
>
> ...
>
>> Change 'none' to 'none_or_standard_func'
>
>
> I suggest 'none_or_default_func' since the term "default function mode"
> is used pretty much throughout the system and introducing a second term
> to describe the same concept is potentially confusing.

Okay. The code already uses the name "standard" in some places, so
I'll make all of them consistent.

>
> The diff looks fine.  Others may have comments regarding the individual
> XXXs,
> but that shouldn't prevent you committing.

Okay. It's mainly float_regs.m that I wasn't sure about, since I don't
understand the code as well as Peter. The issues flagged there may not
actually be bugs at all.

Thanks,
Mark.