current CVS: C++ ICE in bootstrap with --enable-checking

Zack Weinberg zack@rabi.columbia.edu
Thu Dec 23 21:55:00 GMT 1999


While compiling libstdc++.

/home/zack/src/egcs/build/gcc/xgcc -B/home/zack/src/egcs/build/gcc/
    -B/home/zack/egcsinst/i586-pc-linux-gnu/bin/ -c -g -O2 -fvtable-thunks
    -D_GNU_SOURCE -fno-implicit-templates -I../../../libstdc++
    -I../../../libstdc++/stl -I../libio -I../../../libstdc++/../libio
    -nostdinc++ -D_IO_MTSAFE_IO ../../../libstdc++/stdexcepti.cc
In file included from ../../../libstdc++/stdexcepti.cc:8:
../../../libstdc++/stl/stl_alloc.h: In function `size_t
  __default_alloc_template<threads, inst>::_S_freelist_index (unsigned int)
  [with bool threads = true, int inst = 0]':
../../../libstdc++/stl/stl_alloc.h:436:   instantiated from
  `__default_alloc_template<threads, inst>::deallocate (void *, unsigned int)
  [with bool threads = true, int inst = 0]'
../../../libstdc++/std/bastring.cc:42:   instantiated from
  `basic_string<charT, traits, Allocator>::Rep::operatordelete (void *) [with
  charT = char, traits = string_char_traits<char>, Allocator =
  __default_alloc_template<true, 0>]'
../../../libstdc++/std/bastring.h:114:   instantiated from
  `basic_string<charT, traits, Allocator>::Rep::release () [with charT = char,
  traits = string_char_traits<char>, Allocator =
  __default_alloc_template<true, 0>]'
../../../libstdc++/std/bastring.h:195:   instantiated from
  `basic_string<charT, traits, Allocator>::basic_string (int) [with charT =
  char, traits = string_char_traits<char>, Allocator =
  __default_alloc_template<true, 0>]'
../../../libstdc++/stdexcept:45:   instantiated from here
../../../libstdc++/stl/stl_alloc.h:347: Tree check: expected function_decl,
  have template_decl
../../../libstdc++/stl/stl_alloc.h:347: Internal compiler error in
  `lookup_arg_dependent', at ../gcc/cp/decl2.c:4832
Please submit a full bug report.
See <URL: http://www.gnu.org/software/gcc/faq.html#bugreport > for instructions.
make[2]: *** [stdexcepti.o] Error 1

Incidentally, "basic_string<...>::Rep::operatordelete" looks really weird.

zw


More information about the Gcc-bugs mailing list