This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Wide characters and GCC
- To: <gcc at gcc dot gnu dot org>
- Subject: Wide characters and GCC
- From: "Joseph S. Myers" <jsm28 at cam dot ac dot uk>
- Date: Wed, 6 Jun 2001 12:17:18 +0100 (BST)
- cc: <m dot hayes at elec dot canterbury dot ac dot nz>, <lars at nocrew dot org>
Do we support target wchar_t wider than HOST_WIDE_INT in GCC? There are
places this doesn't work - but is this something that we should in
principle support, or should be document it to be unsupported and then
give a compile time error if MAX_WCHAR_TYPE_SIZE > HOST_BITS_PER_WIDE_INT?
If we support wchar_t wider than HOST_WIDE_INT, do we support it wider
than HOST_WIDEST_INT?
We seem to support CHAR_TYPE_SIZE different from BITS_PER_UNIT - which at
the C level would mean that the C byte with sizeof returning 1 is not the
same as the natural hardware byte. No machine description actually seems
to use this feature. To what extent is this a genuinely supported useful
feature?
What should the formats of both narrow and wide STRING_CSTs be when the
target byte size (either hardware or C-level) is not the same as the host
byte size? The only vaguely active ports for which this is the case are I
think the C4X and PDP-10 ports; I've CC:ed their maintainers. What
testcases from the GCC testsuite fail on such targets because of problems
with GCC's support for different sizes of char?
--
Joseph S. Myers
jsm28@cam.ac.uk