[m-rev.] for review: ho_inst_info clarification

Julien Fischer jfischer at opturion.com
Thu Nov 12 09:53:26 AEDT 2015

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.

> in ho_inst_info. This value
> is used for first-order values as well as for functions that have the
> standard inst, and the new name better reflects that.
> Places using this value have been checked for correctness. Issues that
> need to be looked at have been marked with XXX, although have not been
> addressed in this change.
> compiler/prog_data.m:
> 	Update the type.
> compiler/inst_match.m:
> 	Add XXX comments. This module needs to check for non-standard
> 	function insts in a bunch of places.
> compiler/inst_util.m:
> 	Add XXX comments. We lose information about non-standard function
> 	insts when merging bound and any. The information needs to be
> 	either preserved or disallowed entirely.
> compiler/float_regs.m:
> 	Add XXX comments. This module may miss cases involving standard
> 	function insts.
> compiler/modecheck_unify.m:
> 	Add XXX comment. We should exclude pred and non-standard func
> 	lambda non-locals from becoming locked, an addition to standard
> 	functions.
> compiler/*.m:
> 	No special handling is required for other modules.

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


More information about the reviews mailing list