Re: ASM_IDENTIFY_LANGUAGE advice

[Michael Meissner <meissner at cygnus dot com>]

On Fri, Feb 04, 2000 at 02:49:58PM -0700, Jeffrey A Law wrote:
>  In message <20000123012341.A96411@dragon.nuxi.com>you write:
>   > What is the best handling for "ASM_IDENTIFY_LANGUAGE"?
>   > Looking at the various files these are the two formats used.  With a full
>   > GNU toolchain is there any reason to be worried about confusing gdb?
>   > 
>   > alpha/elf.h
>   > ~~~~~~~~~~~
>   > #define ASM_IDENTIFY_LANGUAGE(FILE)                     \
>   >  fprintf(FILE, "\t%s \"GCC (%s) %s\"\n", IDENT_ASM_OP,  \
>   >          lang_identify(), version_string)
>   > 
>   > 
>   > mips/osfrose.h
>   > ~~~~~~~~~~~~~~
>   > /* Identify the front-end which produced this file.  To keep symbol
>   >    space down, and not confuse kdb, only do this if the language is
>   >    not C.  */
>   > 
>   > #define ASM_IDENTIFY_LANGUAGE(STREAM)                                   \
>   > {                                                                       \
>   >   if (strcmp (lang_identify (), "c") != 0)                              \
>   >     output_lang_identify (STREAM);                                      \
>   > }
> I'd tend to lean towards the ELF stuff -- osfrose was used on relatively
> obscure systems and probably isn't the cleanest code we've got in the
> compiler :-)

Hey I ressemble that remark :-).  Of course the osfrose stuff has not been used
in quite some time.  It had some major problems with it, and came about because
one person at OSF decided we needed our own object format, and that it was time
to reimplement Multics.  I think we should just eliminate the mips/i386/ns32k
osfrose support and the halfpic files.

> Presumably the GNU assembler knows the full syntax of the IDENT directive.
> 
> 
> jeff
> 

-- 
Michael Meissner, Red Hat, Inc.
PMB 198, 174 Littleton Road #3, Westford, Massachusetts 01886
Work:	  meissner@redhat.com		phone: 978-486-9304 fax: 978-692-4482
Non-work: meissner@spectacle-pond.org