Re: [PATCH 1/2] new memory-changed MI notification.

[Eli Zaretskii <eliz at gnu dot org>]

> Date: Fri, 28 Sep 2012 15:57:59 +0800
> From: Yao Qi <yao@codesourcery.com>
> CC: <gdb-patches@sourceware.org>
> 
> On 09/28/2012 03:36 PM, Eli Zaretskii wrote:
> > If @var{type} can only be "code", then I suggest to say
> >
> >    ...[,type=code]
> >
> > explicitly.
> >
> 
> @var{type} can only "code" nowadays, but I am wondering we may have 
> other types, such as "data", in the future.  I don't want to give the 
> consumer of this notification an impression that "type=code" is 
> hard-coded into notification.  If we don't have to worry about it at 
> all, "[,type=code]" is fine to me.

We don't have to worry about this at this time.  The explanatory text
says that type can only be "code" anyway, and will have to be revised
when other types can be emitted.  So not saying 'code" saves us
nothing.

> > Btw, why "code"?  If this is the name of the section, it should be
> > ".text", not code.
> 
>  From the consumer of this notification's point of view, 'type' is more 
> useful than 'name', because the consumer may don't know what ".text" 
> section is, or on some ports, text section is not named as ".text", such 
> as section ".text_vle" for VLE.

In that case, saying that this identifies a section is inaccurate.

But if we are going to put "code" explicitly in the notification line,
then this is perhaps a moot point.  The explanatory text should say
something like

  The optional @code{type="code"} part is reported if the memory
  written to holds executable code.