From: Pedro Alves <pedro@palves.net>
To: gdb-patches@sourceware.org
Subject: [PATCH 0/3] Fix "b func(std::string)", DMGL_VERBOSE, gdb.cp/no-dmgl-verbose.exp
Date: Thu, 5 May 2022 19:50:17 +0100 [thread overview]
Message-ID: <20220505185020.3648774-1-pedro@palves.net> (raw)
This series evolved from the discussion starting at:
[PATCH] Fix gdb.cp/no-dmgl-verbose.exp test
https://sourceware.org/pipermail/gdb-patches/2022-April/188542.html
Turns out gdb.cp/no-dmgl-verbose.exp is right as long as GDB doesn't
use DMGL_VERBOSE (which it doesn't today), _and_ on libstdc++, it is
build for pre-C++11 ABI.
See "GCC5 and the C++11 ABI", at
<https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi>.
And "Dual ABI" at
<https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html>.
On systems with compilers that default to modern C++11 ABI
(practically everywhere these days), gdb.cp/no-dmgl-verbose.exp
exposes that "b func(std::string)" doesn't work any more.
This series fixes this, ensuring that the scenario works on both
pre-C++11, and C++11 ABIs. It does this by switching GDB to use
DMGL_VERBOSE, and removing some related code from gdb/cp-support.c.
This is done in patch #1.
Patch #2 is preparatory work for patch #3.
Patch #3 is about making it so that "b f(std::string)" also works if
the current language is C. I noticed that it doesn't work, because
the testcase debugs a .o file, and in such case, GDB doesn't figure
out that the program's language is C++, and so it assumes the language
is C. Without doing "set language c++", "b f(std::string)" would
still fail. But I think that should have not been necessary, I think
GDB should have been able to set the breakpoint even with "set
language c".
Tested on x86_64 GNU/Linux with no regressions.
Pedro Alves (3):
Fix "b f(std::string)", always use DMGL_VERBOSE
Always pass an explicit language down to c_type_print
Fix "b f(std::string)" when current language is C
gdb/ada-typeprint.c | 2 +-
gdb/c-exp.y | 1 +
gdb/c-lang.c | 8 +-
gdb/c-lang.h | 17 +--
gdb/c-typeprint.c | 25 +---
gdb/cp-name-parser.y | 5 +-
gdb/cp-support.c | 54 ++++++---
gdb/cp-support.h | 12 ++
gdb/d-lang.c | 2 +-
gdb/gnu-v3-abi.c | 3 +-
gdb/go-typeprint.c | 2 +-
gdb/objc-lang.c | 2 +-
gdb/opencl-lang.c | 2 +-
gdb/rust-lang.c | 5 +-
...-dmgl-verbose.cc => break-f-std-string.cc} | 0
gdb/testsuite/gdb.cp/break-f-std-string.exp | 112 ++++++++++++++++++
gdb/testsuite/gdb.cp/no-dmgl-verbose.exp | 35 ------
17 files changed, 192 insertions(+), 95 deletions(-)
rename gdb/testsuite/gdb.cp/{no-dmgl-verbose.cc => break-f-std-string.cc} (100%)
create mode 100644 gdb/testsuite/gdb.cp/break-f-std-string.exp
delete mode 100644 gdb/testsuite/gdb.cp/no-dmgl-verbose.exp
base-commit: 8e1ada9e0bd4f57597b835b2d09b100d24c604d8
--
2.36.0
next reply other threads:[~2022-05-05 18:50 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-05 18:50 Pedro Alves [this message]
2022-05-05 18:50 ` [PATCH 1/3] Fix "b f(std::string)", always use DMGL_VERBOSE Pedro Alves
2022-05-05 20:31 ` Keith Seitz via Gdb-patches
2022-05-05 21:49 ` Carl Love via Gdb-patches
2022-05-06 13:21 ` Pedro Alves
2022-05-06 8:10 ` Lancelot SIX via Gdb-patches
2022-05-06 13:09 ` Pedro Alves
2022-05-05 18:50 ` [PATCH 2/3] Always pass an explicit language down to c_type_print Pedro Alves
2022-05-05 20:47 ` Keith Seitz via Gdb-patches
2022-05-06 12:33 ` Pedro Alves
2022-05-05 18:50 ` [PATCH 3/3] Fix "b f(std::string)" when current language is C Pedro Alves
2022-05-05 20:50 ` Keith Seitz via Gdb-patches
2022-05-06 8:34 ` Lancelot SIX via Gdb-patches
2022-05-06 13:09 ` Pedro Alves
2022-05-09 16:54 ` [PATCH 0/3] Fix "b func(std::string)", DMGL_VERBOSE, gdb.cp/no-dmgl-verbose.exp Tom Tromey
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220505185020.3648774-1-pedro@palves.net \
--to=pedro@palves.net \
--cc=gdb-patches@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox