[RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

  Using Ctrl-C to interrupt a running application
does not always work nicely because the application
can decide to handle Ctrl-C as a normal character.
  On the contrary, Ctrl-Break always generates a
call to the installed SetConsoleCtrlHandler.
  This patch allows me to interrupt a debuggee
running in another console (with 'set new-console on')
by pressing Ctrl-Break, even, if Ctrl-C has been
transformed by the debuggee as a normal char
using SetConsoleMode API function.



Pierre Muller
Pascal language support maintainer for GDB



2009-09-21  Pierre Muller  <muller@ics.u-strasbg.fr>

	* src/gdb/windows-nat.c (ctrl_c_handler): Also handle
CTRL_BREAK_EVENT.


Index: src/gdb/windows-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/windows-nat.c,v
retrieving revision 1.196
diff -u -p -r1.196 windows-nat.c
--- src/gdb/windows-nat.c	2 Jul 2009 17:21:07 -0000	1.196
+++ src/gdb/windows-nat.c	21 Sep 2009 14:53:38 -0000
@@ -1289,8 +1304,8 @@ ctrl_c_handler (DWORD event_type)
 {
   const int attach_flag = current_inferior ()->attach_flag;
 
-  /* Only handle Ctrl-C event.  Ignore others.  */
-  if (event_type != CTRL_C_EVENT)
+  /* Only handle Ctrl-C and Ctrl-Break events.  Ignore others.  */
+  if (event_type != CTRL_C_EVENT && event_type != CTRL_BREAK_EVENT)
     return FALSE;
 
   /* If the inferior and the debugger share the same console, do nothing as

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Eli Zaretskii]

> From: "Pierre Muller" <muller@ics.u-strasbg.fr>
> Date: Mon, 21 Sep 2009 17:24:35 +0200
> 
>   This patch allows me to interrupt a debuggee
> running in another console (with 'set new-console on')
> by pressing Ctrl-Break, even, if Ctrl-C has been
> transformed by the debuggee as a normal char
> using SetConsoleMode API function.

If this patch is accepted, please submit also a short change to the
manual, for the Windows-specific section, documenting this behavior.

Thanks.

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Daniel Jacobowitz]

On Mon, Sep 21, 2009 at 05:24:35PM +0200, Pierre Muller wrote:
> 2009-09-21  Pierre Muller  <muller@ics.u-strasbg.fr>
> 
> 	* src/gdb/windows-nat.c (ctrl_c_handler): Also handle
> CTRL_BREAK_EVENT.

Stray "src/gdb/" ?

This seems like a good idea to me.

-- 
Daniel Jacobowitz
CodeSourcery

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Christopher Faylor]

On Mon, Sep 21, 2009 at 01:37:34PM -0400, Daniel Jacobowitz wrote:
>On Mon, Sep 21, 2009 at 05:24:35PM +0200, Pierre Muller wrote:
>> 2009-09-21  Pierre Muller  <muller@ics.u-strasbg.fr>
>> 
>> 	* src/gdb/windows-nat.c (ctrl_c_handler): Also handle
>> CTRL_BREAK_EVENT.
>
>Stray "src/gdb/" ?
>
>This seems like a good idea to me.

Ditto, except for the ChangeLog comment, this is approved.

cgf

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

Thanks Christopher,

patch committed to main branch.

Should it go into 7.0 branch too?

Pierre 


> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Christopher Faylor
> Envoyé : Tuesday, September 22, 2009 12:15 AM
> À : gdb-patches@sourceware.org; Pierre Muller
> Objet : Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in
> ctrl_c_handler
> 
> On Mon, Sep 21, 2009 at 01:37:34PM -0400, Daniel Jacobowitz wrote:
> >On Mon, Sep 21, 2009 at 05:24:35PM +0200, Pierre Muller wrote:
> >> 2009-09-21  Pierre Muller  <muller@ics.u-strasbg.fr>
> >>
> >> 	* src/gdb/windows-nat.c (ctrl_c_handler): Also handle
> >> CTRL_BREAK_EVENT.
> >
> >Stray "src/gdb/" ?
> >
> >This seems like a good idea to me.
> 
> Ditto, except for the ChangeLog comment, this is approved.

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

I already checked the patch in,
for the docs, here is what I wanted to add:

A running MS-Windows program can be interrupted by pressing @samp{Ctrl-C},
unless that program configured @samp{Ctrl-C} as a normal character by a call
to @code{SetConsoleMode}. In that case, @samp{Ctrl-Break} can still be used
to interrupt the running program.

I have no idea if I should add something like a @kindex...

But my main problem is that I don't know really where we should 
put that inside the "Cygwin native" section.

I would suggest putting it right after the first paragraph,
but in fact it could also be with the general description of
running program interruption.

Please help me out,


Pierre Muller
Pascal language support maintainer for GDB




> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Eli Zaretskii
> Envoyé : Monday, September 21, 2009 7:29 PM
> À : Pierre Muller
> Cc : gdb-patches@sourceware.org
> Objet : Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in
> ctrl_c_handler
> 
> > From: "Pierre Muller" <muller@ics.u-strasbg.fr>
> > Date: Mon, 21 Sep 2009 17:24:35 +0200
> >
> >   This patch allows me to interrupt a debuggee
> > running in another console (with 'set new-console on')
> > by pressing Ctrl-Break, even, if Ctrl-C has been
> > transformed by the debuggee as a normal char
> > using SetConsoleMode API function.
> 
> If this patch is accepted, please submit also a short change to the
> manual, for the Windows-specific section, documenting this behavior.
> 
> Thanks.

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Joel Brobecker]

> patch committed to main branch.
> Should it go into 7.0 branch too?

It seems useful and safe. Any objection to that, Chris?

-- 
Joel

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Eli Zaretskii]

> From: "Pierre Muller" <muller@ics.u-strasbg.fr>
> Cc: <gdb-patches@sourceware.org>
> Date: Tue, 22 Sep 2009 17:34:56 +0200
> 
> I already checked the patch in,

Only the trunk or the branch as well?

> A running MS-Windows program can be interrupted by pressing @samp{Ctrl-C},
> unless that program configured @samp{Ctrl-C} as a normal character by a call
> to @code{SetConsoleMode}. In that case, @samp{Ctrl-Break} can still be used
> to interrupt the running program.
> 
> I have no idea if I should add something like a @kindex...
> 
> But my main problem is that I don't know really where we should 
> put that inside the "Cygwin native" section.
> 
> I would suggest putting it right after the first paragraph,
> but in fact it could also be with the general description of
> running program interruption.
> 
> Please help me out,

Your text is mostly okay, but it needs to use a different markup for
the Ctrl-C etc.

I will fix it and install it in a couple of days.  I hope GDB 7.0 will
not be released with the code but without the docs.

Thanks.

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Joel Brobecker]

> I will fix it and install it in a couple of days.  I hope GDB 7.0 will
> not be released with the code but without the docs.

The patch is still pending approval for 7.0, so it should not happen.
If it does, though, I'll make a corrective release very shortly after,
to limit the damage.

-- 
Joel

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

> > I already checked the patch in,
> 
> Only the trunk or the branch as well?

  As Joel said it, it is in trunk only.
 
> > A running MS-Windows program can be interrupted by pressing
> @samp{Ctrl-C},
> > unless that program configured @samp{Ctrl-C} as a normal character by
> a call
> > to @code{SetConsoleMode}. In that case, @samp{Ctrl-Break} can still
> be used
> > to interrupt the running program.
> >
> > I have no idea if I should add something like a @kindex...
> >
> > But my main problem is that I don't know really where we should
> > put that inside the "Cygwin native" section.
> >
> > I would suggest putting it right after the first paragraph,
> > but in fact it could also be with the general description of
> > running program interruption.
> >
> > Please help me out,
> 
> Your text is mostly okay, but it needs to use a different markup for
> the Ctrl-C etc.

  I took the markup that I already found in gdb.texinfo,
I found several @samp{Ctrl-C}
by there is also @kbd{Ctrl-c} which seems indeed more appropriate here!

> I will fix it and install it in a couple of days.  I hope GDB 7.0 will
> not be released with the code but without the docs.

We can decide to wait for your doc patch before 
porting to 7.0 branch if that gets approved!

Pierre

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Eli Zaretskii]

> Date: Tue, 22 Sep 2009 20:56:43 +0300
> From: Eli Zaretskii <eliz@gnu.org>
> Cc: gdb-patches@sourceware.org
> 
> Your text is mostly okay, but it needs to use a different markup for
> the Ctrl-C etc.
> 
> I will fix it and install it in a couple of days.

Done.  Here's what I committed:

2009-09-26  Pierre Muller  <muller@ics.u-strasbg.fr>

	* gdb.texinfo (Cygwin Native): Mention support for Ctrl-BREAK.

Index: gdb.texinfo
===================================================================
RCS file: /cvs/src/src/gdb/doc/gdb.texinfo,v
retrieving revision 1.628
diff -u -r1.628 gdb.texinfo
--- gdb.texinfo	21 Sep 2009 12:48:36 -0000	1.628
+++ gdb.texinfo	26 Sep 2009 16:45:54 -0000
@@ -15900,10 +15900,20 @@
 @cindex Cygwin-specific commands
 
 @value{GDBN} supports native debugging of MS Windows programs, including
-DLLs with and without symbolic debugging information.  There are various
-additional Cygwin-specific commands, described in this section.
-Working with DLLs that have no debugging symbols is described in
-@ref{Non-debug DLL Symbols}.
+DLLs with and without symbolic debugging information.
+
+@cindex Ctrl-BREAK, MS-Windows
+@cindex interrupt debuggee on MS-Windows
+MS-Windows programs that call @code{SetConsoleMode} to switch off the
+special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
+by typing @kbd{C-c}.  For this reason, @value{GDBN} on MS-Windows
+supports @kbd{C-@key{BREAK}} as an alternative interrupt key
+sequence, which can be used to interrupt the debuggee even if it
+ignores @kbd{C-c}.
+
+There are various additional Cygwin-specific commands, described in
+this section.  Working with DLLs that have no debugging symbols is
+described in @ref{Non-debug DLL Symbols}.
 
 @table @code
 @kindex info w32

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

Could we apply both 
the patch to windows-nat.c and to docs to branch
or is it too late?

Pierre

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Eli Zaretskii
> Envoyé : Saturday, September 26, 2009 6:51 PM
> À : Pierre Muller
> Cc : gdb-patches@sourceware.org
> Objet : Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in
> ctrl_c_handler
> 
> > Date: Tue, 22 Sep 2009 20:56:43 +0300
> > From: Eli Zaretskii <eliz@gnu.org>
> > Cc: gdb-patches@sourceware.org
> >
> > Your text is mostly okay, but it needs to use a different markup for
> > the Ctrl-C etc.
> >
> > I will fix it and install it in a couple of days.
> 
> Done.  Here's what I committed:
> 
> 2009-09-26  Pierre Muller  <muller@ics.u-strasbg.fr>
> 
> 	* gdb.texinfo (Cygwin Native): Mention support for Ctrl-BREAK.
> 
> Index: gdb.texinfo
> ===================================================================
> RCS file: /cvs/src/src/gdb/doc/gdb.texinfo,v
> retrieving revision 1.628
> diff -u -r1.628 gdb.texinfo
> --- gdb.texinfo	21 Sep 2009 12:48:36 -0000	1.628
> +++ gdb.texinfo	26 Sep 2009 16:45:54 -0000
> @@ -15900,10 +15900,20 @@
>  @cindex Cygwin-specific commands
> 
>  @value{GDBN} supports native debugging of MS Windows programs,
> including
> -DLLs with and without symbolic debugging information.  There are
> various
> -additional Cygwin-specific commands, described in this section.
> -Working with DLLs that have no debugging symbols is described in
> -@ref{Non-debug DLL Symbols}.
> +DLLs with and without symbolic debugging information.
> +
> +@cindex Ctrl-BREAK, MS-Windows
> +@cindex interrupt debuggee on MS-Windows
> +MS-Windows programs that call @code{SetConsoleMode} to switch off the
> +special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
> +by typing @kbd{C-c}.  For this reason, @value{GDBN} on MS-Windows
> +supports @kbd{C-@key{BREAK}} as an alternative interrupt key
> +sequence, which can be used to interrupt the debuggee even if it
> +ignores @kbd{C-c}.
> +
> +There are various additional Cygwin-specific commands, described in
> +this section.  Working with DLLs that have no debugging symbols is
> +described in @ref{Non-debug DLL Symbols}.
> 
>  @table @code
>  @kindex info w32

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Joel Brobecker]

> Could we apply both 
> the patch to windows-nat.c and to docs to branch
> or is it too late?

It is not too late, you have until tomorrow for 7.0. IIRC, this patch
was approved for the branch, right?

-- 
Joel

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

  In fact, I double checked,
but Chris never answered about the 7 branch...

Pierre

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Joel Brobecker
> Envoyé : Tuesday, September 29, 2009 4:16 PM
> À : Pierre Muller
> Cc : 'Eli Zaretskii'; gdb-patches@sourceware.org
> Objet : Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in
> ctrl_c_handler
> 
> > Could we apply both
> > the patch to windows-nat.c and to docs to branch
> > or is it too late?
> 
> It is not too late, you have until tomorrow for 7.0. IIRC, this patch
> was approved for the branch, right?
> 
> --
> Joel

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Eli Zaretskii]

> From: "Pierre Muller" <muller@ics.u-strasbg.fr>
> Cc: <gdb-patches@sourceware.org>
> Date: Tue, 29 Sep 2009 09:29:38 +0200
> 
> Could we apply both 
> the patch to windows-nat.c and to docs to branch
> or is it too late?

If/when the patch to windows-nat.c is approved for the branch, you
have my approval to commit the docs.

I don't have a clear idea how safe is it to commit the changes to
windows-nat.c to the branch, sorry.

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Joel Brobecker]

> If/when the patch to windows-nat.c is approved for the branch, you
> have my approval to commit the docs.

Chris did not object, and I think the patch is very safe, so go ahead
and commit to the branch.

-- 
Joel

RE: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Pierre Muller]

Thanks,
I committed together
the patch in windows-nat.c and 
the doc patch to gdb.texinfo
to gdb_7_0-branch.

Pierre

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de Joel Brobecker
> Envoyé : Tuesday, September 29, 2009 11:13 PM
> À : Eli Zaretskii
> Cc : Pierre Muller; gdb-patches@sourceware.org
> Objet : Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in
> ctrl_c_handler
> 
> > If/when the patch to windows-nat.c is approved for the branch, you
> > have my approval to commit the docs.
> 
> Chris did not object, and I think the patch is very safe, so go ahead
> and commit to the branch.
> 
> --
> Joel

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Christopher Faylor]

On Tue, Sep 29, 2009 at 04:31:37PM +0200, Pierre Muller wrote:
>  In fact, I double checked,
>but Chris never answered about the 7 branch...

Huh?  You said you committed the CTRL_BREAK_EVENT patch in this thread
to the gdb_7_0_branch.

If you'r talking about your other patch in another thread, then I'm not
comfortable with that going into either the trunk or the branch.  It
seems too much like a band-aid.

cgf

Re: [RFA] windows-nat.c: Handle CTRL_BREAK_EVENT in ctrl_c_handler

[Christopher Faylor]

On Wed, Sep 30, 2009 at 04:09:27PM -0400, Christopher Faylor wrote:
>On Tue, Sep 29, 2009 at 04:31:37PM +0200, Pierre Muller wrote:
>>  In fact, I double checked,
>>but Chris never answered about the 7 branch...
>
>Huh?  You said you committed the CTRL_BREAK_EVENT patch in this thread
>to the gdb_7_0_branch.
>
>If you'r talking about your other patch in another thread, then I'm not
>comfortable with that going into either the trunk or the branch.  It
>seems too much like a band-aid.

Ah, sorry.  I see.  I missed a query.  I was ignoring this thread after
the initial approval since it strayed into a discussion the exact right
syntax to use for the documentation.

I have no problems with this going into the branch and actually
mistakenly thought that it was being proposed for both branch and trunk.

cgf