From:	cgf AT bbc DOT com (Christopher Faylor)
Subject:	RE: Occasional hang with CYGWIN32=tty?
20 Feb 1998 10:02:32 -0800 :
Message-ID:	<199802201628.LAA00817.cygnus.cygwin32.developers@hardy.bbc.com>
To:	cygwin32-developers AT cygnus DOT com, sos AT prospect DOT com DOT ru

>From: Sergey Okhapkin <sos AT prospect DOT com DOT ru>
>Date: Fri, 20 Feb 1998 10:31:57 +0300
>
>Christopher Faylor wrote:
>>which closes down the tty thread if it was created in this process but
>>this the function will block waiting for the tty use count to == 0.  I
>>think that something is incrementing the tty use count and not
>>subsequently decrementing it.
>
>The one place I see is a fork code - dup_for_fork() increments use
>count, but if fork fails for some reasons, the count is not
>decremented.

I had gotten an error from gcc because I was trying to use -pipe with
the EGCS 1.0 version of gcc and it apparently didn't like this.  I
didn't get a fork error but possibly this is it.

I think I've seen this when I haven't had any errors at all but I can't
be sure.  If no one else is seeing it then maybe I've just got something
weird in my configuration.  I haven't relinked everything with the new
libcygwin.a so possibly that's it.
--
http://www.bbc.com/	cgf AT bbc DOT com			"Strange how unreal
VMS=>UNIX Solutions	Boston Business Computing	 the real can be."

- Raw text -