Re: Not emitting .ident directives

[Richard Henderson <rth at cygnus dot com>]

On Sun, Jan 31, 1999 at 03:20:56PM -0500, Zack Weinberg wrote:
> Ok.  Is version_string declared in an existing header file?

Apparently not.  I'd put it in output.h.

> I did it this way for consistency.  It would be easy for the various tm.h
> files to get out of sync - in fact some other ports do ASM_FILE_END almost
> but not quite the same as this.  (I left those alone in this patch).  Would
> you rather I kept to the status quo, or do you have a better idea?

I guess I'd rather stay with the status quo, barring a major cleanup
of the target headers.

> >Nothing that needs solving here, but a point for future cleanup is
> >to allow translate_options to do some transmogrification on options
> >with a single leading - as well.
> 
> Sorry, I don't know what you mean by this.

translate_options only handles `--' options.  If it handled `-'
options as well, you could have translated -Qn to -fno-ident.

Or does -Qn mean something to some assembler?


r~