RFR : 8205959 : Do not restart close if errno is EINTR

[David Lloyd]

I think because the only two possible outcomes are either that the FD
was not dup'd, in which case things carry on as before, or that it was
dup'd, in which case (at least in the JVM) re-dupping won't really do
anything harmful since the target FD already references the dead
socket FD.

The POSIX manpage doesn't seem to include any other possibilities.

On Mon, Jul 2, 2018 at 7:04 AM David Holmes <david.holmes at oracle.com> wrote:
>
> In reference to 8205959, where is it stated that dup2 is any more
> restartable than close ??
>
> AFAICS both leave things undefined/unspecified if they set EINTR.
>
> David
>
> On 2/07/2018 7:03 PM, Langer, Christoph wrote:
> > Hi Matthias,
> >
> > forwarding to serviceability-dev, because debugging is usually discussed there.
> >
> > Yes, I would think this coding should be fixed, too. Can you open a bug and prepare a change?
> >
> > Thanks
> > Christoph
> >
> >> -----Original Message-----
> >> From: net-dev [mailto:net-dev-bounces at openjdk.java.net] On Behalf Of
> >> Norman Maurer
> >> Sent: Montag, 2. Juli 2018 10:23
> >> To: Baesken, Matthias <matthias.baesken at sap.com>
> >> Cc: Stuefe, Thomas <thomas.stuefe at sap.com>; net-dev at openjdk.java.net
> >> Subject: Re: RFR : 8205959 : Do not restart close if errno is EINTR
> >>
> >> +1 retry a close on EINTR has most likely not the outcome you expect and
> >> may even close a wrong FD if the same FD is reused already (as even if EINTR
> >> is returned it may have closed the FD)
> >>
> >>> Am 02.07.2018 um 10:17 schrieb Baesken, Matthias
> >> <matthias.baesken at sap.com>:
> >>>
> >>> Hello  ,  there is a similar pattern (attempt to restart close in case of EINTR)
> >> in the coding as well   in  socket_md.c   :
> >>>
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-147-    int rv;
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-148-    do {
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-149-        rv =
> >> close(fd);
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c:150:    } while (rv
> >> == -1 && errno == EINTR);
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-151-
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-152-    return rv;
> >>> src/jdk.jdwp.agent/unix/native/libdt_socket/socket_md.c-153-}
> >>>
> >>> Do you think this needs adjustment   (on LINUX)  as well ?
> >>>
> >>> Best regards, Matthias
> >>>
> >>>
> >>>> Message: 2
> >>>> Date: Thu, 28 Jun 2018 18:19:46 +0100
> >>>> From: Alan Bateman <Alan.Bateman at oracle.com>
> >>>> To: David Lloyd <david.lloyd at redhat.com>, ivan.gerasimov at oracle.com
> >>>> Cc: OpenJDK Network Dev list <net-dev at openjdk.java.net>
> >>>> Subject: Re: RFR : 8205959 : Do not restart close if errno is EINTR
> >>>> Message-ID: <3fd1496f-ab83-a2d5-0699-13c8b735d70b at oracle.com>
> >>>> Content-Type: text/plain; charset=utf-8; format=flowed
> >>>>
> >>>>> On 28/06/2018 17:35, David Lloyd wrote:
> >>>>> :
> >>>>> Do you (or Alan) think that this might have accounted for real-world
> >>>>> connection problems?
> >>>>>
> >>>> In the file I/O area, with NFS I think, we had an issue a long time ago
> >>>> where close was retried after EIO. That issue was fixed a long time ago
> >>>> but it's one that comes to mind in this general area.
> >>>>
> >>>> -Alan
> >>>>
> >>>



-- 
- DML