Attachment 191741 Details for Bug 248532 – Raceless testcase from Comment 16.

Raceless testcase from Comment 16.

stopped-sigcont-hang-4.c (text/plain), 4.47 KB, created by Jan Kratochvil on 2007-09-10 15:57:53 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Jan Kratochvil

Created: 2007-09-10 15:57:53 UTC

Size: 4.47 KB

patch

obsolete

>/* Copyright 2007 Free Software Foundation, Inc.
>
>   This program is free software; you can redistribute it and/or modify
>   it under the terms of the GNU General Public License as published by
>   the Free Software Foundation; either version 2 of the License, or
>   (at your option) any later version.
>
>   This program is distributed in the hope that it will be useful,
>   but WITHOUT ANY WARRANTY; without even the implied warranty of
>   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>   GNU General Public License for more details.
>
>   You should have received a copy of the GNU General Public License
>   along with this program; if not, write to the Free Software
>   Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
>
>#define _GNU_SOURCE
>#include <signal.h>
>#include <unistd.h>
>#include <string.h>
>#include <stdio.h>
>#include <stdlib.h>
>#include <errno.h>
>#include <sys/wait.h>
>#include <sys/ptrace.h>
>#include <assert.h>
>#include <fcntl.h>
>#include <limits.h>
>#include <ctype.h>
>#include <linux/ptrace.h>
>#include <error.h>
>
>#include <asm/unistd.h>
>#include <unistd.h>
>#define tkill(tid, sig) syscall (__NR_tkill, (tid), (sig))
>
>/* Failure occurs either immediately or in about 20 runs.  */
>#define LOOPS 10000
>
>static pid_t child;
>
>static void
>cleanup (void)
>{
>  if (child != 0)
>    kill (child, SIGKILL);
>}
>
>static void
>handler_fail (int signo)
>{
>  cleanup ();
>
>  signal (SIGABRT, SIG_DFL);
>  abort ();
>}
>
>int main (void)
>{
>  pid_t got_pid;
>  int status;
>  int loops;
>
>  setbuf (stdout, NULL);
>  atexit (cleanup);
>  signal (SIGABRT, handler_fail);
>  signal (SIGALRM, handler_fail);
>  /* Be careful with SIGALRM / ALARM due to false positives under load.  */
>
>  int bad = 0;
>  for (loops = 0; loops < LOOPS; loops++)
>    {
>      child = fork ();
>      switch (child)
>	{
>	case -1:
>	  abort ();
>	case 0:
>	  raise (SIGSTOP);
>	  for (;;)
>	    pause ();
>	  /* NOTREACHED */
>	default:
>	  break;
>	}
>
>      /* We must never WAITPID/WAITID on the child for this testcase.  */
>
>      /* We need to wait till CHILD becomes stopped.  Any DELAY may be
>       * race-prone, check its `/proc/${child}/status'.  */
>      alarm (60);
>      for (;;)
>	{
>	  char buf[0x1000];
>	  ssize_t got;
>	  int fd, i;
>
>	  snprintf (buf, sizeof buf, "/proc/%d/status", (int) child);
>	  fd = open (buf, O_RDONLY);
>	  assert (fd != -1);
>	  got = read (fd, buf, sizeof buf);
>	  assert (got > 0);
>	  assert (got < sizeof buf);
>	  buf[got] = 0;
>	  i = close (fd);
>	  assert (i == 0);
>	  if (strstr (buf, "\nState:\tT (stopped)\n"))
>	    break;
>	}
>      alarm (0);
>
>      /* Here is a point where we - as a debugger - start to attach.  */
>      errno = 0;
>      ptrace (PTRACE_ATTACH, child, NULL, NULL);
>      assert_perror (errno);
>
>      /* Any DELAY makes no difference here.  */
>
>      /* `PTRACE_CONT, SIGSTOP' does not work in 100% cases as sometimes SIGSTOP
>	 gets remembered by kernel during the first PTRACE_CONT later and we get
>	 spurious SIGSTOP event.  Expecting the signal may get delivered to
>	 a different task of the thread group.
>	 `tkill (SIGSTOP)' has no effect in this moment (it is already stopped).  */
>      errno = 0;
>      tkill (child, SIGCONT);
>      assert_perror (errno);
>
>      /* Here must not be any DELAY() to properly test utrace.  The bug was
>	 that WAITPID reported a previous job-control-stop condition even
>	 though SIGCONT has cleared the TASK_STOPPED state.  */
>
>      got_pid = waitpid (child, &status, 0);
>      assert (got_pid == child);
>      assert (WIFSTOPPED (status));
>      /* If we got SIGSTOP here it is a bug of the patch
>	 https://bugzilla.redhat.com/show_bug.cgi?id=248532#c5
>	 tkill (SIGCONT) above should clear all the pending SIGSTOPs.  */
>      if (WSTOPSIG (status) != SIGCONT)
>	{
>	  error (0, 0, "%s after %d iterations",
>		 strsignal (WSTOPSIG (status)), loops);
>	  ++bad;
>	}
>      else
>	{
>	  /* This point is no longer a part of the test, if we want to validate
>	     ptrace(2) below here also must be no DELAY().  */
>
>	  /* At this moment we should be already able to ptrace(2) the inferior.
>	     After the fix of Bug 248532 Comment 5 above there should be no problem
>	     here.  */
>	  errno = 0;
>	  ptrace (PTRACE_PEEKUSR, child, (void *) 0UL, NULL);
>	  assert_perror (errno);
>	}
>
>      /* Cleanup.  */
>      errno = 0;
>      kill (child, SIGKILL);
>      assert_perror (errno);
>
>      got_pid = waitpid (child, &status, 0);
>      assert (got_pid == child);
>    }
>
>  printf ("%d bad in %d iterations: %.2f%%\n",
>	  bad, LOOPS, 100.0 * bad / LOOPS);
>  if (bad)
>    return 1;
>
>  puts ("PASS");
>  return 0;
>}

Actions: View

Attachments on bug 248532: 159411 | 179961 | 181161 | 182321 | 182381 | 185931 | 187231 | 191741