Re: Deletion of removed targets (AMD 29K, ELXSI, etc)

["Joseph S. Myers" <jsm28 at cam dot ac dot uk>]

On Thu, 5 Dec 2002, Larin Hennessy wrote:

> boehm-gc
> 2002-12-05  Larin Hennessy  <larin@oregonstate.edu>
> 
> 	* config.guess: Remove clipper and Convex entries
> 	* config.sub: Remove MIL-STD-1750a, AMD 29K, Clipper, Convex,
> 	 Elxsi, Intel i860, Sun PicoJava, and Western Electric 32000 entries.

Not appropriate.  These are common GNU files and some GNU software may
support these systems.

> 	* expr.c: Remove comment about Convex

Generic comments like this one about variation between different machines 
may still be perfectly relevant even though the particular machine 
mentioned as an example isn't supported any longer.  Is there some 
currently supported machine that would be a better example in

-  /* Try indexing by frame ptr and try by stack ptr.
-     It is known that on the Convex the stack ptr isn't a valid index.
-     With luck, one or the other is valid on any machine.  */
+  /* Try indexing by frame ptr and try by stack ptr. */

?

> 	* doc/trouble.texi: Remove WE32K entry

I think there's a lot more in trouble.texi relating only to removed 
targets (possibly removed targets on CPUs for which other targets are 
still supported).

> gcc/po
> 2002-12-05  Larin Hennessy <larin@oregonstate.edu>
> 
> 	* da.po: Remove MIL-STD-1750a, AMD 29K, Clipper, Convex,
> 	 Elxsi, Intel i860, Sun PicoJava, and Western Electric 32000 entries.
> 	* el.po (ditto)
> 	* es.po (ditto)
> 	* fr.po (ditto)
> 	* ja.po (ditto)
> 	* nl.po (ditto)
> 	* sv.po (ditto)
> 	* tr.po (ditto)

Changes to message catalogs need to go via the translation project.  In 
this case, what's needed is for gcc.pot to be regenerated and sent to the 
translation project, and the translators should then remove obsolete 
messages.

Could the i18n maintainer deal with regenerating gcc.pot, mainline and 3.3
branch, once 3.3 has branched, and submit a 3.3 branch snapshot to the
robot, to get this in motion?

-- 
Joseph S. Myers
jsm28@cam.ac.uk