clone: prepare to recycle CLONE_STOPPED
authorAndrew Morton <akpm@linux-foundation.org>
Tue, 5 Feb 2008 06:27:22 +0000 (22:27 -0800)
committerLinus Torvalds <torvalds@woody.linux-foundation.org>
Tue, 5 Feb 2008 17:44:07 +0000 (09:44 -0800)
Ulrich says that we never used this clone flags and that nothing should be
using it.

As we're down to only a single bit left in clone's flags argument, let's add a
warning to check that no userspace is actually using it.  Hopefully we will
be able to recycle it.

Roland said:

  CLONE_STOPPED was previously used by some NTPL versions when under
  thread_db (i.e.  only when being actively debugged by gdb), but not for a
  long time now, and it never worked reliably when it was used.  Removing it
  seems fine to me.

[akpm@linux-foundation.org: it looks like CLONE_DETACHED is being used]
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kernel/fork.c

index 05e0b6f4365bfc5fca5ef1cabd400c3894c013e3..6caf4f23206b4e0490a9fc57535e99646cc37a70 100644 (file)
@@ -1450,6 +1450,23 @@ long do_fork(unsigned long clone_flags,
        int trace = 0;
        long nr;
 
+       /*
+        * We hope to recycle these flags after 2.6.26
+        */
+       if (unlikely(clone_flags & CLONE_STOPPED)) {
+               static int __read_mostly count = 100;
+
+               if (count > 0 && printk_ratelimit()) {
+                       char comm[TASK_COMM_LEN];
+
+                       count--;
+                       printk(KERN_INFO "fork(): process `%s' used deprecated "
+                                       "clone flags 0x%lx\n",
+                               get_task_comm(comm, current),
+                               clone_flags & CLONE_STOPPED);
+               }
+       }
+
        if (unlikely(current->ptrace)) {
                trace = fork_traceflag (clone_flags);
                if (trace)