Re: assert creates unusable core dump on current stable Cygwin release

[Brian Inglis <Brian.Inglis@SystematicSw.ab.ca>]

On 2019-10-09 11:10, Jon Turney wrote:
> On 09/10/2019 16:31, Brian Inglis wrote:
>> On 2019-10-09 01:14, Biswapriyo Nath wrote:
>>> * mintty version: mintty 3.0.6 (x86_64-pc-cygwin) * cygwin version:
>>> 3.1.0-0.6 * code:
>>> 
>>> #include <assert.h>
>>> 
>>> int main() { int x = 1; assert(x == 0); }
>>> 
>>> * Expected result: The terminal should show this message which appers in
>>> latest stable cygwin version. assertion "x == 0" failed: file "test.c",
>>> line 6, function: main
>>> 
>>> * Actual result: terminal only shows "Aborted (core dumped)".
>> 
>> I also get the core dump, which is un-gdb-able below, as is it's core dump,
>> on *current stable* Cygwin 64 releases *AND* see the message!
> 
> As far as I'm aware, the core-dump tool 'dumper' has never been fixed for 
> x86_64, so I wouldn't expect this to work. See [1].
> 
> [1] https://cygwin.com/ml/cygwin/2016-03/msg00464.html

Given that it's used elsewhere, is there more to it than just:

int
dumper::init_core_dump ()
{
  bfd_init ();

- core_bfd = bfd_openw (file_name, "elf32-i386");
+ core_bfd = bfd_openw (file_name, "elf64-x86");
  if (core_bfd == NULL)
    {
      bfd_perror ("opening bfd");
      goto failed;
    }

  if (!bfd_set_format (core_bfd, bfd_core))
    {
      bfd_perror ("setting bfd format");
      goto failed;
    }

- if (!bfd_set_arch_mach (core_bfd, bfd_arch_i386, 0))
+ if (!bfd_set_arch_mach (core_bfd, bfd_arch_i386, bfd_mach_x86_64))
    {
      bfd_perror ("setting bfd architecture");
      goto failed;
    }

  return 1;

failed:
  dumper_abort ();
  return 0;

}

-- 
Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada

This email may be disturbing to some readers as it contains
too much technical detail. Reader discretion is advised.

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple