* 20120523: bug in fork/exec
@ 2012-05-25 9:38 Yaakov (Cygwin/X)
2012-05-25 9:42 ` Corinna Vinschen
0 siblings, 1 reply; 6+ messages in thread
From: Yaakov (Cygwin/X) @ 2012-05-25 9:38 UTC (permalink / raw)
To: cygwin
With the 20120523 snapshot, if a process attempts to fork/exec another
program which depends on a DLL which is missing, the PID created by
fork() spins:
* in taskmgr, the forked process is listed as an instance of the program
which called fork() (as usual);
* this PID ties up one CPU core (e.g. 25% CPU usage on a quad-core machine);
* in ps(1), the same PID is displayed as the program which was attempted
to be execed;
* /bin/kill does not work on this PID, even with -f.
This is on Win7 x64.
Yaakov
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: 20120523: bug in fork/exec
2012-05-25 9:38 20120523: bug in fork/exec Yaakov (Cygwin/X)
@ 2012-05-25 9:42 ` Corinna Vinschen
2012-05-25 20:30 ` Yaakov (Cygwin/X)
0 siblings, 1 reply; 6+ messages in thread
From: Corinna Vinschen @ 2012-05-25 9:42 UTC (permalink / raw)
To: cygwin
On May 25 03:26, Yaakov (Cygwin/X) wrote:
> With the 20120523 snapshot, if a process attempts to fork/exec
Which snapshot shows this effect first?
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: 20120523: bug in fork/exec
2012-05-25 9:42 ` Corinna Vinschen
@ 2012-05-25 20:30 ` Yaakov (Cygwin/X)
2012-05-29 12:18 ` Corinna Vinschen
0 siblings, 1 reply; 6+ messages in thread
From: Yaakov (Cygwin/X) @ 2012-05-25 20:30 UTC (permalink / raw)
To: cygwin
On 2012-05-25 04:40, Corinna Vinschen wrote:
> On May 25 03:26, Yaakov (Cygwin/X) wrote:
>> With the 20120523 snapshot, if a process attempts to fork/exec
>
> Which snapshot shows this effect first?
Ouch, it starts with 20120309.
FWIW, here is the fork() code I discovered this with:
http://git.gnome.org/browse/glib/tree/glib/gspawn.c?id=2.32.3
Yaakov
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: 20120523: bug in fork/exec
2012-05-25 20:30 ` Yaakov (Cygwin/X)
@ 2012-05-29 12:18 ` Corinna Vinschen
2012-05-30 15:53 ` Corinna Vinschen
0 siblings, 1 reply; 6+ messages in thread
From: Corinna Vinschen @ 2012-05-29 12:18 UTC (permalink / raw)
To: cygwin
On May 25 15:10, Yaakov (Cygwin/X) wrote:
> On 2012-05-25 04:40, Corinna Vinschen wrote:
> >On May 25 03:26, Yaakov (Cygwin/X) wrote:
> >>With the 20120523 snapshot, if a process attempts to fork/exec
> >
> >Which snapshot shows this effect first?
>
> Ouch, it starts with 20120309.
Thanks. That would mean a single change is the culprit, but I need
a way to reproduce the problem to be able to fix it (other than just
reverting the patch).
> FWIW, here is the fork() code I discovered this with:
>
> http://git.gnome.org/browse/glib/tree/glib/gspawn.c?id=2.32.3
That's a bit heavy as far as testcases go. Thing is, I can't reproduce
it. If that's a generic problem, shouldn't this also occur when trying
to start a program from bash or tcsh? I deliberately removed a DLL from
my system, but there is no hang at all when trying to start an affected
application.
Can you please provide a simple testcase? If you can reproduce it with
given binaries from the distro, I don't need source code, just the steps
and the names of the binaries to reproduce.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: 20120523: bug in fork/exec
2012-05-29 12:18 ` Corinna Vinschen
@ 2012-05-30 15:53 ` Corinna Vinschen
2012-06-04 0:30 ` Yaakov (Cygwin/X)
0 siblings, 1 reply; 6+ messages in thread
From: Corinna Vinschen @ 2012-05-30 15:53 UTC (permalink / raw)
To: cygwin
On May 29 11:40, Corinna Vinschen wrote:
> On May 25 15:10, Yaakov (Cygwin/X) wrote:
> > On 2012-05-25 04:40, Corinna Vinschen wrote:
> > >On May 25 03:26, Yaakov (Cygwin/X) wrote:
> > >>With the 20120523 snapshot, if a process attempts to fork/exec
> > >
> > >Which snapshot shows this effect first?
> >
> > Ouch, it starts with 20120309.
>
> Thanks. That would mean a single change is the culprit, but I need
> a way to reproduce the problem to be able to fix it (other than just
> reverting the patch).
>
> > FWIW, here is the fork() code I discovered this with:
> >
> > http://git.gnome.org/browse/glib/tree/glib/gspawn.c?id=2.32.3
>
> That's a bit heavy as far as testcases go. Thing is, I can't reproduce
> it. If that's a generic problem, shouldn't this also occur when trying
> to start a program from bash or tcsh? I deliberately removed a DLL from
> my system, but there is no hang at all when trying to start an affected
> application.
>
> Can you please provide a simple testcase? If you can reproduce it with
> given binaries from the distro, I don't need source code, just the steps
> and the names of the binaries to reproduce.
Never mind. I found the problem and just checked in a patch. I'll
create a new snapshot shortly. Please test.
Thanks,
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: 20120523: bug in fork/exec
2012-05-30 15:53 ` Corinna Vinschen
@ 2012-06-04 0:30 ` Yaakov (Cygwin/X)
0 siblings, 0 replies; 6+ messages in thread
From: Yaakov (Cygwin/X) @ 2012-06-04 0:30 UTC (permalink / raw)
To: cygwin
On 2012-05-30 09:38, Corinna Vinschen wrote:
> Never mind. I found the problem and just checked in a patch. I'll
> create a new snapshot shortly. Please test.
Looks fixed in 20120530. Thanks,
Yaakov
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-06-04 0:30 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-05-25 9:38 20120523: bug in fork/exec Yaakov (Cygwin/X)
2012-05-25 9:42 ` Corinna Vinschen
2012-05-25 20:30 ` Yaakov (Cygwin/X)
2012-05-29 12:18 ` Corinna Vinschen
2012-05-30 15:53 ` Corinna Vinschen
2012-06-04 0:30 ` Yaakov (Cygwin/X)
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).