[PATCH] to_slave pipe is full fix

[Christopher Faylor]

On Mon, Dec 16, 2002 at 01:15:54PM -0600, Steve O wrote:
>On Mon, Dec 16, 2002 at 05:01:22PM +0100, Corinna Vinschen wrote:
>> did you perhaps forgot to attach the patch?
>
>My apologies, the patch is enclosed this time.
>-steve
>
>ChangeLog:
>2002-12-15  Steve Osborn  <bub@io.com>
>
>    * fhandler_termios.cc (fhandler_termios::line_edit): Return
>    line_edit_error and remove last char from readahead buffer
>    if accept_input() fails.
>    * fhandler_tty.cc (fhandler_pty_master::accept_input): Return 0
>    and restore readahead buffer when tty slave pipe is full.

-             SetEvent (input_available_event);
-             ReleaseMutex (input_mutex);
-             Sleep (10);
-             rc = WaitForSingleObject (input_mutex, INFINITE);
+             puts_readahead (p, bytes_left);
+             ret = 0;
+             break;

Won't this introduce a spinning situation since you are no longer
sleeping?  Were we really inappropriately waiting for the input_mutex in
this case?

cgf