From owner-freebsd-hackers  Tue Nov 12 16:04:22 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id QAA05534
          for hackers-outgoing; Tue, 12 Nov 1996 16:04:22 -0800 (PST)
Received: from time.cdrom.com (time.cdrom.com [204.216.27.226])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id QAA05493
          for <hackers@freebsd.org>; Tue, 12 Nov 1996 16:03:59 -0800 (PST)
Received: from time.cdrom.com (localhost [127.0.0.1]) by time.cdrom.com (8.8.2/8.6.9) with ESMTP id QAA29413 for <hackers@freebsd.org>; Tue, 12 Nov 1996 16:04:01 -0800 (PST)
To: hackers@freebsd.org
Subject: Is our ASYNC I/O support for ttys broken?
Date: Tue, 12 Nov 1996 16:04:01 -0800
Message-ID: <29402.847843441@time.cdrom.com>
From: "Jordan K. Hubbard" <jkh@time.cdrom.com>
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Thomas Roell over at X Inside just sent me this little snippet,
complaining that ASYNC I/O was broken in FreeBSD where it worked with
Linux and SVR4.  Any comments?

#include <stdio.h>
#include <fcntl.h>
#include <sys/types.h>
#include <sys/signal.h>

int fd;          

void
signal_handler(
    int        signo
)
{
  char buf[64];

  printf("bazoing %d\n", read(fd, buf, sizeof(buf)));
}

main()
{
  struct sigaction sa;
  sa.sa_handler = signal_handler;
  sa.sa_flags   = SA_RESTART;
  sigemptyset(&sa.sa_mask);
  sigaction(SIGIO,  &sa, NULL);

  fd = open("/dev/ttyd0", O_RDONLY | O_NONBLOCK);

  fcntl(fd, F_SETFL, fcntl(fd, F_GETFL, 0) | FASYNC);
  fcntl(fd, F_SETOWN, getpid());

  for (;;) { sleep(10000); }
}

And yes indeed, the signal handler never goes off when you're trying
to talk to a standard TTY device.  I changed the /dev/ttyd0 in the
example to /dev/cuaa1, which is where my mouse is (and from which I
can provably read data with cat) and the signal handler was never
called.

					Jordan