While comparing the behavior of GCC mainline and Studio 12.1 cc wrt. UCN
identifiers, I noticed that while gcc passes them unmodified to the
assembler, cc uses some encoding:
$ cat ucn.c
int \u00C0;
$ c99 -S ucn.c
$ cat ucn.s
[...]
.section ".rodata",#alloc
Drodata.rodata:
.skip 0
.type Drodata.rodata,#object
.size Drodata.rodata,0
.section ".bss",#alloc,#write
.common __2cBQhK_,4,4
[...]
$ gcc -std=c99 -fextended-identifiers -S ucn.c
$ cat ucn.s
.file "ucn.c"
.common ?,4,4
.ident "GCC: (GNU) 4.4.2"
It would be really helpful if cc and gcc could be made compatible here,
so I'd like to request a specification of the encoding used by cc to
implement this in gcc.
Thanks.
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
