Bug 19436 - C++11 ABI tag does not work
Summary: C++11 ABI tag does not work
Status: NEW
Alias: None
Product: gdb
Classification: Unclassified
Component: c++ (show other bugs)
Version: HEAD
: P2 normal
Target Milestone: ---
Assignee: Not yet assigned to anyone
URL:
Keywords:
: 18601 (view as bug list)
Depends on:
Blocks:
 
Reported: 2016-01-07 18:11 UTC by Jan Kratochvil
Modified: 2024-06-29 09:52 UTC (History)
8 users (show)

See Also:
Host:
Target:
Build:
Last reconfirmed:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Jan Kratochvil 2016-01-07 18:11:21 UTC
It affects default configuration of latest stable release of GCC.



echo -e '#include <string>\nstd::string foovar;'|g++ -c -o 20.o -Wall -g -x c++ -;nm -C 20.o|grep -w foovar
0000000000000000 B foovar[abi:cxx11]
gdb -batch ./20.o -ex "p foovar" -ex "p 'foovar[abi:cxx11]'"

Actual results:
No symbol "foovar" in current context.
$1 = {static npos = <optimized out>, _M_dataplus = {<std::allocator<char>> = {<__gnu_cxx::new_allocator<char>> = {<No data fields>}, <No data fields>}, _M_p = 0x0 <__static_initialization_and_destruction_0(int, int)>}, _M_string_length = 0, {_M_local_buf = '\000' <repeats 15 times>, _M_allocated_capacity = 0}}

Expected results:
$1 = ""
(I do not care what it should for the second print.)

One can use gcc -D_GLIBCXX_USE_CXX11_ABI=0 to get the old behavior.
One could also configure gcc for the old behavior with:
  --with-default-libstdcxx-abi=gcc4-compatible

GDB: b900245c3b92fc460a3f7fa17d14eb08f9ab4c76 7.10.50.20160107-git
FAIL: g++ (GCC) 5.3.1 20160107
FAIL: g++ (GCC) 6.0.0 20150530 (experimental)
I read it should affect already gcc-5.1 but I haven't verified that.

The pretty printing bug is probably in libstdc++ pretty printers, I do not make that a subject of this PR.
Comment 1 Yao Qi 2016-08-12 13:07:30 UTC
A fail in gdb.cp/no-dmgl-verbose.exp is due to this PR.

break 'f(std::string)'^M
Function "f(std::string)" not defined.^M
(gdb) FAIL: gdb.cp/no-dmgl-verbose.exp: setting breakpoint at 'f(std::string)'

GDB can insert breakpoint if I add __cxx11 after std::, like this,

(gdb) b 'f(std::__cxx11::string)'
Breakpoint 1 at 0x8: file /home/yaoqi01/SourceCode/gnu/build-gdb/gdb/testsuite/../../../binutils-gdb/gdb/testsuite/gdb.cp/no-dmgl-verbose.cc, line 23.

Also, if I compile test case with -D_GLIBCXX_USE_CXX11_ABI=0, gdb works again,
Comment 2 ghjghj530-bubu 2017-07-27 13:39:09 UTC
Still a problem in GCC 6.3, GDB 7.10

-> With -D_GLIBCXX_USE_CXX11_ABI=1

(gdb) p fooVar
p fooVar
No symbol "fooVar" in current context.
(gdb) info variables
...
std::__cxx11::string fooVar[abi:cxx11];
....
(gdb) p 'fooVar[abi:cxx11]'
p 'fooVar[abi:cxx11]'
$2 = "hello world"


-> With -D_GLIBCXX_USE_CXX11_ABI=0

(gdb) p fooVar
p fooVar
$2 = "hello world"


Any chance somebody can have a look at this? thanks!
Comment 3 Keith Seitz 2017-07-28 16:44:51 UTC
There is a patch series that has been submitted by Pedro which should address (or at least mitigate) this issue. See https://sourceware.org/ml/gdb-patches/2017-06/msg00012.html .
Comment 4 Keith Seitz 2017-08-09 19:34:08 UTC
*** Bug 18601 has been marked as a duplicate of this bug. ***
Comment 5 Sourceware Commits 2017-11-29 19:49:24 UTC
The master branch has been updated by Pedro Alves <palves@sourceware.org>:

https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=bd69330db86b2367aac8aac5915f1686451c9d5d

commit bd69330db86b2367aac8aac5915f1686451c9d5d
Author: Pedro Alves <palves@redhat.com>
Date:   Wed Nov 29 19:33:24 2017 +0000

    Breakpoints in symbols with ABI tags (PR c++/19436)
    
    Trying to set a breakpoint in a function with an ABI tag does not work
    currently.  E.g., debugging gdb itself, we see this with the
    "string_printf" function:
    
     (top-gdb) b string_print                               [TAB]
     (top-gdb) b string_printf[abi:cxx11](char const*, ...) [RET]
     No source file named string_printf[abi.
     Make breakpoint pending on future shared library load? (y or [n])
    
    Quoting doesn't help:
     (top-gdb) b 'string_printf[abi:cxx11]'(char const*, ...)
     malformed linespec error: unexpected string, "(char const*, ...)"
     (top-gdb) b 'string_printf[abi:cxx11](char const*, ...)'
     No source file named string_printf[abi.
     Make breakpoint pending on future shared library load? (y or [n]) n
    
    This patch fixes this, and takes it a bit further.
    
    The actual symbol name as demangled by libiberty's demangler is really
    
     string_printf[abi:cxx11](char const*, ...)
    
    however, this patch makes it possible to set the breakpoint with
    
     string_printf(char const*, ...)
    
    too.  I.e., ignoring the ABI tag.
    
    And to match, it teaches the completer to complete the symbol name
    without the ABI tag, i.e.,
    
      "string_pri<TAB>"  -> "string_printf(char const*, ...)"
    
    If however, you really want to break on a symbol with the tag, then
    you simply start writing the tag, and GDB will preserve it, like:
    
      "string_printf[a<TAB>"  -> "string_printf[abi:cxx11](char const*, ...)"
    
    Grows the gdb.linespec/ tests like this:
    
      -# of expected passes           8977
      +# of expected passes           9176
    
    gdb/ChangeLog:
    2017-11-29  Pedro Alves  <palves@redhat.com>
    
    	PR c++/19436
    	* NEWS: Mention setting breakpoints on functions with C++ ABI
    	tags.
    	* completer.h (completion_match_for_lcd) <match,
    	mark_ignored_range>: New methods.
    	<finish>: Consider ignored ranges.
    	<clear>: Clear ignored ranges.
    	<m_ignored_ranges, m_finished_storage>: New fields.
    	* cp-support.c (cp_search_name_hash): Ignore ABI tags.
    	(cp_symbol_name_matches_1, cp_fq_symbol_name_matches): Pass the
    	completion_match_for_lcd pointer to strncmp_iw_with_mode.
    	(test_cp_symbol_name_cmp): Add [abi:...] tags unit tests.
    	* language.c (default_symbol_name_matcher): Pass the
    	completion_match_for_lcd pointer to strncmp_iw_with_mode.
    	* linespec.c (linespec_lexer_lex_string): Don't tokenize ABI tags.
    	* utils.c (skip_abi_tag): New function.
    	(strncmp_iw_with_mode): Add completion_match_for_lcd parameter.
    	Handle ABI tags.
    	* utils.h (strncmp_iw_with_mode): Add completion_match_for_lcd
    	parameter.
    
    gdb/testsuite/ChangeLog:
    2017-11-29  Pedro Alves  <palves@redhat.com>
    
    	PR c++/19436
    	* gdb.linespec/cpls-abi-tag.cc: New file.
    	* gdb.linespec/cpls-abi-tag.exp: New file.
    
    gdb/doc/ChangeLog:
    2017-11-29  Pedro Alves  <palves@redhat.com>
    
    	PR c++/19436
    	* gdb.texinfo (Debugging C Plus Plus): Document setting
    	breakpoints in functions with ABI tags.
Comment 6 Pedro Alves 2017-11-29 20:18:33 UTC
"print" (In reply to Jan Kratochvil from comment #0)
> It affects default configuration of latest stable release of GCC.
> 
> 
> 
> echo -e '#include <string>\nstd::string foovar;'|g++ -c -o 20.o -Wall -g -x
> c++ -;nm -C 20.o|grep -w foovar
> 0000000000000000 B foovar[abi:cxx11]
> gdb -batch ./20.o -ex "p foovar" -ex "p 'foovar[abi:cxx11]'"

This now works in master:

gdb -batch ./20.o -ex "p foovar" -ex "p 'foovar[abi:cxx11]'"
$1 = ""
$2 = ""

which turns out to be a happy consequence of the commit listed above.
I'm leaving the bug open because while bd69330db86b added comprehensive tests for breakpoints-on-symbols-with-abi-tags, it added none for print/expressions.
Comment 7 Tom de Vries 2019-10-13 09:48:01 UTC
(In reply to Yao Qi from comment #1)
> A fail in gdb.cp/no-dmgl-verbose.exp is due to this PR.

FTR, this FAIL still exists, as mentioned here: https://sourceware.org/ml/gdb-patches/2019-10/msg00308.html.
Comment 8 Hannes Domani 2024-06-29 09:52:55 UTC
(In reply to Tom de Vries from comment #7)
> (In reply to Yao Qi from comment #1)
> > A fail in gdb.cp/no-dmgl-verbose.exp is due to this PR.
> 
> FTR, this FAIL still exists, as mentioned here:
> https://sourceware.org/ml/gdb-patches/2019-10/msg00308.html.

gdb.cp/no-dmgl-verbose.exp was removed by this commit:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=cb2cd8cba82a0a5480a147d95b16098ad74d33c6

Can this be closed now?