www.delorie.com/archives/browse.cgi   search  
Mail Archives: cygwin-developers/1998/09/29/05:41:18

From: cgf AT cygnus DOT com (Christopher Faylor)
Subject: Re: b20 alpha
29 Sep 1998 05:41:18 -0700 :
Message-ID: <19980929082135.B25527.cygnus.cygwin32.developers@cygnus.com>
References: <3610A3BB DOT 77F10FDC AT cityweb DOT de>
Mime-Version: 1.0
To: Corinna Vinschen <corinna DOT vinschen AT cityweb DOT de>,
cygwin32-developers AT cygnus DOT com

I believe that Sergey posted a patch for this a while ago.  I don't
have the patch around anymore.  Maybe he could repost it?

On Tue, Sep 29, 1998 at 11:09:15AM +0200, Corinna Vinschen wrote:
>I have a funny effect with b20 cygwin.dll.
>It only happens in bash, not in tcsh and it
>only comes to pass, if CYGWIN32=tty.
>
>First try:
>    bash-2.01$ echo x | cat
>    x
>
>Second try:
>    bash-2.01$ echo x > foo
>    bash-2.01$ echo x | cat
>    cat: standard output: Bad file number
>
>This doesn't take place in the following example:
>    bash-2.01$ cat file > foo
>    bash-2.01$ cat file | cat
>
>But here:
>    bash-2.01$ echo x > foo
>    bash-2.01$ cat file | cat
>    cat: standard output: Bad file number
>
>After the first stdout redirection of a shell builtin,
>the pipes doesn't work anymore. I have tried a
>	STRACE='0xff:0,xxx'
>and I have found the following strange lines in the output:
>    ...
>    [main] cat 1037 (0) hinfo::de_linearize_fd_array: closed fd 1
>    ...
>    [main] cat 1037 (0) _fstat: -1 = fstat (1, 0x240FF04)
>    ...

-- 
cgf AT cygnus DOT com
http://www.cygnus.com/

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019