Hi,

I've updated my patches for 12266, and I *think* I have finally stamped 
out most of the bugs, nearly doubling the number of tests. Of course, it 
is still possible that I've neglected something along the line.

I've attached the "whole" patch again (this is what will actually be 
committed -- it is definitive). I have /not/ included separate patches 
for the "printname" and "cp_demangle_name_parse_free" patches from last 
go around, since those have not changed.

I am attaching, though, the cp_canonicalize_string_no_typedef and tests 
patches, since those have changed non-trivially.

This patchset has been tested on the buildbot.

Keith

PS. Link to previous submission:
http://sourceware.org/ml/gdb-patches/2011-04/msg00405.html
http://sourceware.org/ml/gdb-patches/2011-04/msg00408.html
http://sourceware.org/ml/gdb-patches/2011-04/msg00407.html
http://sourceware.org/ml/gdb-patches/2011-04/msg00406.html
http://sourceware.org/ml/gdb-patches/2011-04/msg00409.html

PPS. ChangeLogs are nearly identical -- one addition:

	* cp-support.c (replace_typedefs_qualified_name): New function.