LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Marc Lehmann <linux-kernel@tst.eu>
To: Eric Dumazet <dada1@cosmosbay.com>
Cc: linux-kernel@vger.kernel.org, Davide Libenzi <davidel@xmailserver.org>
Subject: Re: epoll design problems with common fork/exec patterns
Date: Sat, 27 Oct 2007 10:51:25 +0200	[thread overview]
Message-ID: <20071027085125.GC12326@schmorp.de> (raw)
In-Reply-To: <4722F575.8080204@cosmosbay.com>

On Sat, Oct 27, 2007 at 10:23:17AM +0200, Eric Dumazet <dada1@cosmosbay.com> wrote:
> >  In this case, the parent process works fine until the child closes fds,
> >  after which the fds become unarmed in the parent too. This works as
> 
> I have no idea what exact problem you have. 

Well, I explained it rather succinctly, I think. If you tell me whats unclear
I can explain...

> But if the child closes some 
> file descriptor that were 'cloned' at fork() time, this only decrements a 
> refcount, and definitely should not close it for the 'parent'.

It doesn't. It removes it from the epoll set, though, so the parent will not
receive events for that fd anymore.

> I have some apps that are happily using epoll() and fork()/exec() and have 

The problem I described is fork/close/exec. close being the explicit
syscall.

> no problem at all. I usually use O_CLOEXEC so that all close() are done at 
> exec() time without having to do it in a loop. epoll continues to work as 
> expected in the parent process.

This is because epoll doesn't behave like documented: It removes the fd
from the parents epoll set only on an explicit close() syscall, not on an
implicit close from exec.

> >fd sets. This would explain the behaviour above. Unfortunately (or
> >fortunately?) this is not what happens: when the fds are being closed by
> >exec or exit, the fds do not get removed from the epoll set.
> 
> at exec() (granted CLOEXEC is asserted) or exit() time, only the refcount 
> of each file is decremented. Only if their refcount becomes NULL, files are 
> then removed from epoll set.

Yes. But thats obviously not the only way to close fds.

> >Is epoll really designed to be so incompatible with the most commno fork
> >patterns? Shouldn't epoll do refcounting, as is commonly done under
> >Unix? As the fd space is not shared between rpocesses, why does epoll
> >try? Shouldn't the epoll information be copied just like the fd table
> >itself, memory, and other resources?
> 
> Too many questions here, showing lack of understanding.

You already said you don't the problem. No need to get insulting :(

> epoll definitly is not useless. It is used on major and critical apps.
> You certainly missed something.

Well, it behaves like documented, which is the problem. You admit you
don't understand the problem or the documentation, so again, no need to
insult me.

> Please provide some code to illustrate one exact problem you have.

   // assume there is an open epoll set that listens for events on fd 5
   if (fork () = 0)
     {
       close (5);
       // fd 5 is now removed from the epoll set of the parent.
       _exit (0);
     }

-- 
                The choice of a
      -----==-     _GNU_
      ----==-- _       generation     Marc Lehmann
      ---==---(_)__  __ ____  __      pcg@goof.com
      --==---/ / _ \/ // /\ \/ /      http://schmorp.de/
      -=====/_/_//_/\_,_/ /_/\_\      XX11-RIPE

  reply	other threads:[~2007-10-27  8:49 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-27  6:22 Marc Lehmann
2007-10-27  8:23 ` Eric Dumazet
2007-10-27  8:51   ` Marc Lehmann [this message]
2007-10-27  9:22     ` Eric Dumazet
2007-10-27  9:34       ` Marc Lehmann
2007-10-27 10:23         ` Eric Dumazet
2007-10-27 10:46           ` Marc Lehmann
2007-10-27 16:59     ` Davide Libenzi
2007-10-27 17:38       ` Willy Tarreau
2007-10-27 18:01         ` Davide Libenzi
2007-10-29 22:36         ` Mark Lord
2007-10-28  4:47       ` David Schwartz
2007-10-28  9:33         ` Eric Dumazet
2007-10-28 21:04           ` David Schwartz
2007-10-29 18:55             ` Davide Libenzi
2008-02-26 15:13               ` Michael Kerrisk
2008-02-26 18:51                 ` Davide Libenzi
2008-02-27  1:30                   ` Chris "ク" Heath
2008-02-27 19:35                     ` Davide Libenzi
2008-02-28 13:12                       ` Michael Kerrisk
2008-02-28 13:23                         ` Michael Kerrisk
2008-02-28 19:34                           ` Davide Libenzi
2008-02-28 19:23                         ` Davide Libenzi
2008-02-29 15:46                           ` Michael Kerrisk
2008-02-29 19:19                             ` Davide Libenzi
2008-02-29 19:54                               ` Michael Kerrisk
2008-03-02 15:11                                 ` Sam Varshavchik
2008-03-02 21:44                                   ` Davide Libenzi
2007-10-28 18:48         ` Davide Libenzi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20071027085125.GC12326@schmorp.de \
    --to=linux-kernel@tst.eu \
    --cc=dada1@cosmosbay.com \
    --cc=davidel@xmailserver.org \
    --cc=linux-kernel@vger.kernel.org \
    --subject='Re: epoll design problems with common fork/exec patterns' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

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 NNTP newsgroup(s).