Mail Archives: cygwin/2004/01/16/15:05:55
Christopher Faylor wrote:
>
> On Sat, Jan 03, 2004 at 06:16:28PM -0500, Pierre A. Humblet wrote:
> >1) With the snapshot I can start cygwin programs fine and I have not
> >observed any popup.
>
> I haven't seen a popup in Windows Me for some time with the latest
> snapshot and I've been able to configure/make cygwin under ME with no
> problem.
>
> >2) The regression repeated before (see original mail for more details)
> >is still there: "However when using sh, setsid sh -c "echo hello >
> >/dev/tty1; /bin/sleep 10; /bin/echo there > /dev/tty" sh and the final
> >echo hang forever (with echo in the "O" state in ps), and sh cannot be
> >terminated with kill (OK from the task manager)."
>
> This should be ok now.
>
> >3) There is another regression that I had also observed with the
> >previous snapshot, but after sending my previous message: the Console
> >is not properly freed after setsid when CYGWIN=notty. Programs such as
> >inetd do not go in the background when starting from a command.com
> >window or from the Windows "run" menu. Things are fine with CYGWIN=tty
>
> This also should be working now.
>
> >By the way, I don't understand why open_fhs is modified in the slave
> >tty and ctty code. The only purpose I see in open_fhs is to free the
> >console at the right time. Why should the presence of a slave tty,
> >possibly connected to an entirely different console window, impact on
> >the local console? (this is not a new issue).
>
> I was hoping that Corinna would chime in here but maybe she isn't
> following this discussion. IIRC, I resisted adding the open_fhs checks
> to the pty/tty handling early on but eventually was convinced that they
> were needed. Previously, it was pretty hard to have a program with
> handles to more than one console but, with the existence of the "new"
> windows station stuff it is conceivable, if unlikely that a program
> could have two open. In that context the open_fhs stuff is being
> overused, but that was part of the motivation for implementing the new
> archetype stuff in the DLL -- then I can eventually get rid of the
> open_fhs stuff in favor of setting up console archetypes. There is
> still a lot of genericizing to do to accommodate that but it is a future
> goal.
Thanks Chris.
I have run into problem with another test, running from cmd.exe.
H:\>set CYGWIN
CYGWIN=tty
H:\>uname -a
CYGWIN_NT-4.0 xxx 1.5.6s(0.108/3/2) 20040116 13:48:00 i686 unknown unknown Cygwin
H:\>cat try_forksetsid.c
#include <unistd.h>
#include <signal.h>
main()
{
signal(SIGHUP, SIG_IGN);
close(0);
close(1);
close(2);
if (!fork()) {
// setsid();
sleep(10);
}
}
H:\>gcc try_forksetsid.c
H:\>sh -c "./a"
waiting for children using tty0 to terminate
waiting for children using tty0 to terminate
It's motivated by a weirdness I saw last night with exim,
where the setsid() is present.
Pierre
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -