Re: [PATCH] Symbol Separation - Take 2

[Daniel Jacobowitz <drow at mvista dot com>]

On Sat, Jul 19, 2003 at 04:39:31PM -0700, Devang Patel wrote:
> 
> On Saturday, July 19, 2003, at 8:40 AM, Neil Booth wrote:
> 
> >Devang Patel wrote:-
> >
> >>Well, symbol separation can not be invalid. It is a method to create
> >>and use
> >>separate debugging symbol repositories. If user reads documentation
> >>(and developer reads first paragraph in cppss.c) then it is obvious
> >>what is
> >>'sr' and hence -Winvalid-sr.
> >
> >PCH is a standard abbreviation that most compiler users know, SR is
> >not.  So please expand sr.
> 
> Most users do not know what is symbol repository, so even if I expand
> it to -Winvalid-symbol-repository I'm afraid it will not be obvious to 
> them.
> They will have to read documentation. And once they read it they will
> start using SR as abbreviation of Symbol Repository.

You can search (info browser, or google) for Symbol Repository.  Not
for SR.

> >>>OK, I'll need to look at these in more detail.
> >>
> >>I'm still waiting.
> >
> >Sorry, I've only had a bit of time in the evenings and wanted to spend
> >that on finishing my --help changes.
> >
> >I've had a look at your patch.  From what I can see, cppss.c doesn't
> >belong in cpplib at all.  Maybe you can put it in c-pch.c, or in a
> >new c-symsep.c?
> 
>  I think it belongs to cpplib. Can you explain me which function does
> not belong to cppss.c ?

Cpplib is essentially a preprocessor and token-stream library.  Soon it
will be built completely independently of GCC, and probably used in
other projects.  In what way does a sophisticated mechanism for
managing debug information belong to it?

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer