From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-return-29394-listarch-gdb=sources.redhat.com@sourceware.org>
Received: (qmail 19909 invoked by alias); 22 Aug 2007 11:10:42 -0000
Received: (qmail 19806 invoked by uid 22791); 22 Aug 2007 11:10:41 -0000
X-Spam-Check-By: sourceware.org
Received: from Unknown (HELO mail2.br-automation.com) (213.33.116.61)     by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 22 Aug 2007 11:10:39 +0000
X-AuditID: c0a80110-aa710bb0000009d6-8a-46cc19a18d78
Received: from brsmtp01.br-automation.com (unknown [192.168.1.60]) 	by mail2.br-automation.com (Symantec Mail Security) with ESMTP id C543A912 	for <gdb@sourceware.org>; Wed, 22 Aug 2007 13:10:25 +0200 (CEST)
In-Reply-To: <20070821025607.GA14389@ednor.casa.cgf.cx>
To: gdb@sourceware.org
Subject: Re: gdb-6.6 mingw port hangs after Ctrl-C
MIME-Version: 1.0
X-Mailer: Lotus Notes Release 7.0.2 September 26, 2006
Message-ID: <OFE68D095B.3ACA11DC-ONC125733F.00375C8A-C125733F.003D5E58@br-automation.com>
From: Roland Puntaier <Roland.Puntaier@br-automation.com>
Date: Wed, 22 Aug 2007 11:10:00 -0000
Content-Type: text/plain; charset="US-ASCII"
X-IsSubscribed: yes
Mailing-List: contact gdb-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <gdb.sourceware.org>
List-Subscribe: <mailto:gdb-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/gdb/>
List-Post: <mailto:gdb@sourceware.org>
List-Help: <mailto:gdb-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: gdb-owner@sourceware.org
X-SW-Source: 2007-08/txt/msg00163.txt.bz2

> Christopher Faylor:
> You can stop another thread with a "SuspendThread()" call and you can
> cause the thread to resume in another location with a "ResumeThread()"
> by changing the EIP that is passed to it.  However, if you do that while
> you are in the middle of a low-level Windows function which, say, holds
> a mutex, then you are going to eventually run into problems.

I wanted to try this with the following changes in remoted.c,
but I could'nt find a way to get the EIP of the new address, i.e. 
"continueHere:". 
How could that be done?

#include <windows.h>//Added
static HANDLE h_main_thread = 0;//Added

static void
interrupt_query (void)
{
  target_terminal_ours ();

  if (query ("Interrupted while waiting for the program.\n\
Give up (and stop debugging it)? "))
    {
      target_mourn_inferior ();
      if (h_main_thread){//Added
        CONTEXT threadContext;//Added
        SuspendThread(h_main_thread);//Added
        GetThreadContext(h_main_thread,&threadContext);//Added
        threadContext.Eip = &continueHere;//Added and wished it would work
        SetThreadContext(h_main_thread,&threadContext);//Added
        ResumeThread(h_main_thread);//Added
        ExitThread(0);//Added
      }//Added
continueHere://Added and wished it would work
      deprecated_throw_reason (RETURN_QUIT);
    }

  target_terminal_inferior ();
}

...

static ptid_t
remote_wait (ptid_t ptid, struct target_waitstatus *status)
{
...
      h_main_thread=GetCurrentThread();//Added
      ofunc = signal (SIGINT, remote_interrupt);
      getpkt (&rs->buf, &rs->buf_size, 1);
      signal (SIGINT, ofunc);
      h_main_thread=0;//Added
....


> Christopher Faylor:
> You'd need to use overlapped I/O for those situations.  If you do that
> then you'd have to decide whether you want gdb to work on Windows 9x/Me
> or not.
This seems less brute force: 
 WSARecv(), WSAWaitForMultipleEvents() instead of recv() 
 and let WSAWaitForMultipleEvents() return through an event signaled in 
interrupt_query()
 and if that event happened, continue calling deprecated_throw_reason 
(RETURN_QUIT); this time from the main thread.