Re: bin/53996: tip(1) fails to handle EOF on stdin

[Andreas Gustafsson <gson%gson.org@localhost>]

The following reply was made to PR bin/53996; it has been noted by GNATS.

From: Andreas Gustafsson <gson%gson.org@localhost>
To: Valery Ushakov <uwe%stderr.spb.ru@localhost>
Cc: gnats-bugs%NetBSD.org@localhost
Subject: Re: bin/53996: tip(1) fails to handle EOF on stdin
Date: Fri, 22 Feb 2019 19:37:08 +0200

 Valery,
 
 You didn't answer my question: what purpose do you need the test case for?
 
 > >From your description it sounded like the minimized test case should
 > be like a couple lines of python, like, in pseudocode
 > 
 >     spawn("tip")
 >     close_pty()
 > 
 > and since you work with pexpect you presumably already know how these
 > "spawn" and "close_pty" are really spelled, so it shouldn't be that
 > much of a trouble.
 > 
 > Is it more involved then that?
 
 Yes, it is more involved than that.
 
 Assuming "dty0c" in your /etc/remote is pointing to a working serial
 port (which it happens to do on my system), the following test case
 should work.  Save this as test.py:
 
 import pexpect
 import sys
 import time
 import os
 child = pexpect.spawn("sh", ["-c", "trap '' HUP && ktrace tip dty0c"], logfile = sys.stdout)
 child.expect("onnected");
 time.sleep(1)
 os.close(child.child_fd)
 time.sleep(1)
 
 Then run the following as root:
 
 python2.7 test.py
 kdump
 
 kdump will show an a large number of system calls writing 0xFF
 characters to the serial port:
 
  19467      1 tip      CALL  write(3,0x7f7fffffdc0f,1)
  19467      1 tip      GIO   fd 3 wrote 1 bytes
        "\M^?"
  19467      1 tip      RET   write 1
  19467      1 tip      CALL  write(3,0x7f7fffffdc0f,1)
  19467      1 tip      GIO   fd 3 wrote 1 bytes
        "\M^?"
  19467      1 tip      RET   write 1
 
 and ktrace.out will keep growing until you kill the tip
 processes.
 -- 
 Andreas Gustafsson, gson%gson.org@localhost