Mail Archives: cygwin/2001/09/18/11:50:06
On Tue, Sep 18, 2001 at 06:55:13PM +0900, Dmitry Timoshkov wrote:
>"Corinna Vinschen" <cygwin AT cygwin DOT com> wrote:
>
>[skipped]
>
>> > I didn't investigate it further, and probably will not have time in the near
>> > future to do so. Sorry.
>>
>> :-(
>>
>> You could at least send a testcase if you don't want to debug it.
>
>I'll try, when have a bit of spare time, but I would not promise anything.
>The described scenario is used in almost an every client/server application.
>
>Here is a short description of what is going on:
>
>Client forks and execs server. Server creates unix domain socket, does fork()
>and after all preparations calls accept() on it.
>
>Client then creates socket and calls connect() on it. Right after successful
>return from connect(), client does fcntl(fd, F_SETFD, 1). Right after fcntl()
>server gets error from accept(). If client calls fcntl() *before* connect(),
>all goes without any hitch.
>
>Right now I can't help with something more. Sorry again.
This should be fixed in snapshots.
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -