public inbox for gdb@sourceware.org
 help / color / mirror / Atom feed
From: Claude Robitaille <claude-robitaille@hotmail.com>
To: Simon Marchi <simon.marchi@polymtl.ca>,
	"gdb@sourceware.org" <gdb@sourceware.org>,
	Claude Robitaille <claude-robitaille@hotmail.com>
Subject: Re: GDB is preventing my catch blocks in my C++ program
Date: Thu, 14 Apr 2022 15:06:22 +0000	[thread overview]
Message-ID: <DS7PR07MB768614FA0A0A8C8AD625CBBAF7EF9@DS7PR07MB7686.namprd07.prod.outlook.com> (raw)
In-Reply-To: <DS7PR07MB768627AD9817A6F52AD9B599F7EC9@DS7PR07MB7686.namprd07.prod.outlook.com>

I created a sample program to reproduce the issues.

#include<iostream>
#include <stdexcept>
int main()
{
  std::cout<<"Before the try block" << std::endl;

  try {
    throw std::runtime_error("Test exception");
  } catch (...) {
    std::cout << "Catched the exception" << std::endl;
  }

  std::cout<<"After the try block" << std::endl;
  return 0;
}


I compiled it using this

x86_64-w64-mingw32-g++ -static hello.cpp -static-libstdc++ -static-libgcc -o hello

When running outside gdb, I get this output

Before the try block
Catched the exception
After the try block


But under gdbserver, I get

Listening on port 2345
Remote debugging from host 192.168.70.8, port 44782
Before the try block
terminate called after throwing an instance of 'std::runtime_error'

Child exited with status 3






________________________________
From: Gdb <gdb-bounces+claude=cbcr.me@sourceware.org> on behalf of Claude Robitaille via Gdb <gdb@sourceware.org>
Sent: April 12, 2022 9:39 PM
To: Simon Marchi <simon.marchi@polymtl.ca>; gdb@sourceware.org <gdb@sourceware.org>
Subject: Re: GDB is preventing my catch blocks in my C++ program

To include the list...
________________________________
From: Claude Robitaille <claude-robitaille@hotmail.com>
Sent: April 12, 2022 9:28 PM
To: Simon Marchi <simon.marchi@polymtl.ca>
Subject: Re: GDB is preventing my catch blocks in my C++ program

Thanks Simon for the quick reply.

I can single step the program up to the point where an exception is thrown. And I did set a breakpoint inside the catch statement but no success (earlier breakpoint in the code are working as expected).

Here is what appears on the console (yes, it is Windows)

Listening on port 2345
Remote debugging from host 192.168.70.8, port 34542
Error while doing a GetTokenInformation - Unknown errorLOGGER NOT INITIALIZED:
                *********************************************************************************************
terminate called after throwing an instance of 'sf::context::tGlobal::envVariableMissing'

Child exited with status 3


The string that starts with Error blah blah is my own std::cout earlier in the program. Same with the line with the ******
and the LOGGER xxxx is from the g3log library in my program.

The exception instance is definitely what my program is throwing.

I guess Child exited is from gdb.


________________________________
From: Simon Marchi <simon.marchi@polymtl.ca>
Sent: April 12, 2022 9:17 PM
To: Claude Robitaille <claude-robitaille@hotmail.com>; gdb@sourceware.org <gdb@sourceware.org>
Subject: Re: GDB is preventing my catch blocks in my C++ program



On 2022-04-12 20:27, Claude Robitaille via Gdb wrote:
> Hi all,
>
> I am having an issue with exception in C++. When I run my program outside gdb (gsbserver actually) all is good and my exceptions are properly catched.
>
> However, when running un gdbserver I get:
> "terminate called after throwing an instance of XXXX"
>
> The actuall catch statmeent is catch (...) so it is a catch all.
>
> I obvisouly suspect that, somehow, gdbserver is preventing the normal try/catch/throw machinery to work as it should. The "terminate called..." string is most likely coming form the language library.
>
> Background
> I am crosscompiling with mingw-w64 on Ubuntu. The compiler is 9.3 while both gdb and gdbserver are 9.1 I run gdb under eclipse
>
> I run under WIndows 10
>
> I spend the whole afternoon on Google but all I found are cases where the programmer did not properly coded the catch (or something similar), which, indeed, is not my problem since all is good when running outside of gdb.
>
> Not sure how to tackle this one!!

Just to try to get the obvious out of the way first... are you sure it's
not GDBserver itself crashing?  GDBserver is a C++ program, so it's
possible that an exception is thrown and not caught in it.

Can you paste the logs of a session that shows your problem?

GDBserver and your program are running on Windows IIUC?

Simon

  reply	other threads:[~2022-04-14 15:06 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-13  0:27 Claude Robitaille
2022-04-13  1:17 ` Simon Marchi
     [not found]   ` <DS7PR07MB768661F868A4131D995B3449F7EC9@DS7PR07MB7686.namprd07.prod.outlook.com>
2022-04-13  1:33     ` Simon Marchi
2022-04-13  1:39     ` Claude Robitaille
2022-04-14 15:06       ` Claude Robitaille [this message]
2022-04-15 15:55         ` Claude Robitaille
2022-04-18  0:45           ` Simon Marchi

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=DS7PR07MB768614FA0A0A8C8AD625CBBAF7EF9@DS7PR07MB7686.namprd07.prod.outlook.com \
    --to=claude-robitaille@hotmail.com \
    --cc=gdb@sourceware.org \
    --cc=simon.marchi@polymtl.ca \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).