Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Hannes Domani <ssbssa@yahoo.de>
To: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>,
	 Simon Marchi <simark@simark.ca>
Subject: Re: [PATCH] Fix ctrl-c when debugging WOW64 processes
Date: Thu, 17 Sep 2020 20:23:45 +0000 (UTC)	[thread overview]
Message-ID: <1811381142.1345868.1600374225078@mail.yahoo.com> (raw)
In-Reply-To: <0c8dc56c-22cd-be59-3fdb-bae6878ea331@simark.ca>

 Am Donnerstag, 17. September 2020, 22:00:29 MESZ hat Simon Marchi <simark@simark.ca> Folgendes geschrieben:

> On 2020-09-17 2:03 p.m., Hannes Domani via Gdb-patches wrote:
> > DebugBreakProcess starts a new thread in the target process with the
> > entry point DbgUiRemoteBreakin, where an int3 triggers a breakpoint
> > exception for gdb.
> >
> > But this uses DbgUiRemoteBreakin of the 64bit ntdll.dll even for
> > WOW64 processes.
> > It stops in 64bit code, Wow64GetThreadContext reports a wrong pc without
> > the int3, and gdb lets the target process continue.
> >
> > So this uses DbgUiRemoteBreakin of the 32bit ntdll.dll as the thread
> > entry point for WOW64 processes instead.
>
> Interesting.  Is there any reference somewhere that explains that things
> must be done this way when a 64 bit process is debugging a 32 bit
> process?

Sadly, I could find no real reference for this.
It took me quite a few hours to debug/understand this problem.


> In any case, the patch is OK, with the formatting nits below fixed.
>
> > @@ -1522,9 +1524,36 @@ ctrl_c_handler (DWORD event_type)
> >    if (!new_console && !attach_flag)
> >      return TRUE;
> >
> > -  if (!DebugBreakProcess (current_process_handle))
> > -    warning (_("Could not interrupt program.  "
> > -          "Press Ctrl-c in the program console."));
> > +#ifdef __x86_64__
> > +  if (wow64_process)
> > +    {
> > +      /* Call DbgUiRemoteBreakin of the 32bit ntdll.dll in the target process.
> > +    DebugBreakProcess would call the one of the 64bit ntdll.dll, which
> > +    can't be correctly handled by gdb.  */
> > +      if (!wow64_dbgbreak)
>
> wow64_dbgbreak == nullptr
>
>
> > +    {
> > +      CORE_ADDR addr;
> > +      if (!find_minimal_symbol_address ("ntdll!DbgUiRemoteBreakin",
> > +                        &addr, 0))
> > +        wow64_dbgbreak = (void *) addr;
> > +    }
> > +
> > +      if (wow64_dbgbreak)
>
>
> wow64_dbgbreak != nullptr

Pushed with these changes, thanks.


Hannes


  reply	other threads:[~2020-09-17 20:23 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20200917180337.1984-1-ssbssa.ref@yahoo.de>
2020-09-17 18:03 ` Hannes Domani
2020-09-17 20:00   ` Simon Marchi
2020-09-17 20:23     ` Hannes Domani [this message]
2020-09-17 20:17   ` Tom Tromey
2020-09-17 20:56     ` Hannes Domani
2020-09-18 14:27       ` Hannes Domani
2020-09-18 15:48         ` Simon Marchi
2020-09-18 16:09           ` Hannes Domani

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=1811381142.1345868.1600374225078@mail.yahoo.com \
    --to=ssbssa@yahoo.de \
    --cc=gdb-patches@sourceware.org \
    --cc=simark@simark.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