Re: [RFA]: Add IN_EPILOGUE() predicate

[Andrew Cagney <ac131313 at cygnus dot com>]

> On Mon, 5 Nov 2001, Andrew Cagney wrote:
> 
> 
>> > I actually think that gdbarch-related functions should be described
>> > in their own chapter/section, perhaps with subsections devoted to
>> > different aspects of gdbarch.  So could you please start such a
>> > section somewhere in the chapter "Target Architecture Definition",
>> > and put the text you wanted to add there?  Please use the general
>> > idea of the section "Target Conditionals" for the layout and the text
>> > that preceeds the list of the functions/macros.
> 
>> 
>> Eli, I don't know that there is much to gain from this.  The old macro's 
>> and new functions are really the same thing (the architecture vector) 
>> and belong in the same section.
> 
> 
> I thought about starting a new section for reasons that are partly 
> pedagogical: we want to break away with the Old Ways and start our Brave 
> New World!  [;-)] 

:-) Hmm, an x86 maintainer, be very afraid :-)

> In addition, I suspect it would be confusing to have in the same section 
> description of two essentially parallel, but subtly different (e.g., 
> functions vs macros) systems.  It seemed to me that if we say clearly in 
> each section/chapter what system it describes, the reader will be less 
> confused.

Hopefully the upper case vs lower case will help people differentiate 
between the two.

> However, if you feel we should document them all together, I won't 
> object; the above are not strong feelings.

I'm in the same boat.  I suspect that it will be easier to track if 
someone submitting a patch for a macro->function conversion does the 
change in-place rather than has to also move the definition from 
chanpter A to chapter B.

Andrew