DomainKey-Signature: a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id
	:list-unsubscribe:list-subscribe:list-archive:list-post
	:list-help:sender:date:from:to:subject:message-id:reply-to
	:references:mime-version:content-type:in-reply-to; q=dns; s=
	default; b=NbLZ6EIgDMYXCRU4X1fVngtkZtvNNLYtjL/hUqehF66jTyVqZ2Yqh
	X6wS3l1Np4wIJ9x0iId88EBwwgpY+ACopok0NGqyK4lppf38f6cQWzjH03aVSn0d
	WqAPqJfi6TG6qgigp9hlLgHO6vXbezmkAFcMyFKu9nog31I2Wiy+2c=
Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm
Sender: cygwin-owner@cygwin.com
Mail-Followup-To: cygwin@cygwin.com
Date: Thu, 12 Mar 2015 15:54:23 +0100
From: Corinna Vinschen <corinna-cygwin@cygwin.com>
To: cygwin@cygwin.com
Subject: Re: PTY dies when master in parent process is closed.
Message-ID: <20150312145423.GT5732@calimero.vinschen.de>
Reply-To: cygwin@cygwin.com
Mail-Followup-To: cygwin@cygwin.com
References: <20150305215323.760df4752fdbd6f19a931851@nifty.ne.jp> <20150305133100.GY3213@calimero.vinschen.de> <20150305135839.GZ3213@calimero.vinschen.de> <20150306200710.1264a46aa8dd633943e56212@nifty.ne.jp> <20150308163759.9d3eacd5f3b168a1515a6bc2@nifty.ne.jp> <20150311164829.GJ5732@calimero.vinschen.de> <20150312200343.deab52b09e3126d9cd80ab9c@nifty.ne.jp>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;	protocol="application/pgp-signature"; boundary="8RJ6MkOGpKElHlfH"
Content-Disposition: inline
In-Reply-To: <20150312200343.deab52b09e3126d9cd80ab9c@nifty.ne.jp>
User-Agent: Mutt/1.5.23 (2014-03-12)

--8RJ6MkOGpKElHlfH
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mar 12 20:03, Takashi Yano wrote:
> On Wed, 11 Mar 2015 17:48:29 +0100
> Corinna Vinschen <corinna-cygwin@cygwin.com> wrote:
>=20
> > I'm inclined to apply this patch.  I think using PeekNamedPipe this way
> > is pretty nice, albeit it's not atomic.  It might be a good idea to add
> > an acquire/release_output_mutex bracket, even if that doesn't help for
> > cases in which a process holding a master fd crashes hard.
> >=20
> > Since you suggested it, do you have another idea?
>=20
> Nothing in particular.
>=20
> > Off the top of my head what we could do:
> >=20
> > - Either just set input_available_event always, unconditionally.  But
> >   that probably requires to change fhandler_pty_slave::read as well.
>=20
> The modification of fhandler_pty_slave::read() will be
> probably not so small. To use PeekNamedPipe() or NtQueryObject()
> is better, I suppose.
>=20=20
> > - Or, use NtQueryObject to fetch the handle count of from_master (see
> >   flock.cc function get_obj_handle_count()).  If the handle count is
> >   one, we're the last process holding the handle and then we set
> >   input_available_event.  But I'm not sure if it's really worth it.  The
> >   PeekNamedPipe usage looks rather ok to me.
>=20
> When NtQueryObject() is used, aim is clearer than using
> PeekNamedPipe(). However, either is OK. Please choose one
> which you like.

On second thought, using NtQueryObject introduces a potential new
race condition.  Since for this test to make sense you have to check
the output handle before closing it, the test-and-close action would
have to be made atomically with all other ways to duplicate the handle.
Using PeekNamedPipe after closing the handle is certainly the better
approach.

Patch applied with an additional preceeding comment so we know why
PeekNamedPipe is called here.


Thanks,
Corinna

--=20
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Maintainer                 cygwin AT cygwin DOT com
Red Hat

--8RJ6MkOGpKElHlfH
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2
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=DBQm
-----END PGP SIGNATURE-----

--8RJ6MkOGpKElHlfH--