search

ianlancetaylor / demangle

C++ symbol name demangler written in Go
BSD 3-Clause "New" or "Revised" License
166 stars 24 forks source link

Mangled name which gets demangled with Go demangler but not with original c++filt ! #9

Closed simonhf closed 4 years ago

simonhf commented 4 years ago

The mangled name is legit because it came from a legit Linux C++ project which compiles and runs without error.

Any ideas why this fails with c++filt, or where I would submit a bug for c++filt?

$ git clone https://github.com/ianlancetaylor/demangle.git
$ cd demangle/
$ mkdir -p src/github.com/ianlancetaylor/demangle/
$ cp *.go src/github.com/ianlancetaylor/demangle/
$ GOPATH=`pwd` go run c++filt.go _ZSt16forward_as_tupleIJZNK6broker5store7requestINS0_4dataEJRKN3caf13atom_constantILNS4_10atom_valueE4115129EEES3_S3_EEENS4_8expectedIT_EEDpOT0_EUlRS3_E_ZNKS2_IS3_JS9_S3_S3_EEESC_SF_EUlRNS4_5errorEE0_EESt5tupleIJDpOT_EESO_
std::tuple<broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data&&, broker::data&&) const::{lambda(broker::data&)#1}&&, broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&) const::{lambda(caf::error&)#2}&&> std::forward_as_tuple<broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data&&, broker::data&&) const::{lambda(broker::data&)#1}, broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&) const::{lambda(caf::error&)#2}>(broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data&&, broker::data&&) const::{lambda(broker::data&)#1}, broker::store::request<broker::data, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data>(caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&, caf::atom_constant<(caf::atom_value)4115129> const&, broker::data, broker::data&&) const::{lambda(caf::error&)#2})

$ c++filt --version | egrep Binutils
GNU c++filt (GNU Binutils for Ubuntu) 2.33
$ c++filt --no-recursion-limit _ZSt16forward_as_tupleIJZNK6broker5store7requestINS0_4dataEJRKN3caf13atom_constantILNS4_10atom_valueE4115129EEES3_S3_EEENS4_8expectedIT_EEDpOT0_EUlRS3_E_ZNKS2_IS3_JS9_S3_S3_EEESC_SF_EUlRNS4_5errorEE0_EESt5tupleIJDpOT_EESO_
_ZSt16forward_as_tupleIJZNK6broker5store7requestINS0_4dataEJRKN3caf13atom_constantILNS4_10atom_valueE4115129EEES3_S3_EEENS4_8expectedIT_EEDpOT0_EUlRS3_E_ZNKS2_IS3_JS9_S3_S3_EEESC_SF_EUlRNS4_5errorEE0_EESt5tupleIJDpOT_EESO_

$ gcc --version | egrep gcc
gcc (Ubuntu 9.2.1-9ubuntu2) 9.2.1 20191008

$ cat --number /usr/include/c++/9/tuple | head -1483 | tail -7 
  1477    // _GLIBCXX_RESOLVE_LIB_DEFECTS
  1478    // 2275. Why is forward_as_tuple not constexpr?
  1479    /// std::forward_as_tuple
  1480    template<typename... _Elements>
  1481      constexpr tuple<_Elements&&...>
  1482      forward_as_tuple(_Elements&&... __args) noexcept
  1483      { return tuple<_Elements&&...>(std::forward<_Elements>(__args)...); }
ianlancetaylor commented 4 years ago

If this package demangles it correctly and the c++filt program does not, then the bug should be filed at https://gcc.gnu.org/bugzilla. (Of course it may already have been fixed there.)

simonhf commented 4 years ago

Thanks! Created the bug here [1].

[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93035