This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Professional support for Vxworks cross-compiler
- To: Mike Stump <mrs at windriver dot com>
- Subject: Re: Professional support for Vxworks cross-compiler
- From: David Edelsohn <dje at watson dot ibm dot com>
- Date: Fri, 07 Jul 2000 18:30:44 -0400
- cc: gcc at gcc dot gnu dot org
>>>>> Mike Stump writes:
>> Also, what is the whole mark_fn_as_defined_in_this_file()
>> machinery doing? Setting SYMBOL_REF_FLAG under more circumstances?
Mike> We basically remember if something is actually defined in the
Mike> translation unit, and we later make a codegen choice based upon that.
Mike> If we don't have information, we guess conservatively. If a sym has
Mike> been defined, then we assume that the whole translation unit fits
Mike> within 32 MB (and hence all defined syms are short and undefined syms
Mike> are long).
alpha, arm, and rs6000 all define current_file_function_operand()
to check SYMBOL_REF_FLAG. That flag already exists without your patch, so
it seems that the patch expands the number of cases where that flag is
set. I am trying to understand why you need to set SYMBOL_REF_FLAG. Were
cases being missed?
AIX linkage already behaves similarly, but without all of the
shortcall/longcall hints.
David