[PATCH 1/3] ptrace_init_task: initialize child->jobctl explicitly

[Oleg Nesterov]

new_child->jobctl is not initialized during the fork, it is copied
from parent->jobctl. Currently this is harmless, the forking task
is running and copy_process() can't succeed if signal_pending() is
true, so only JOBCTL_STOP_DEQUEUED can be copied. Still this is a
bit fragile, it would be more clean to set ->jobctl = 0 explicitly.

Also, check ->ptrace != 0 instead of PT_PTRACED, move the
CONFIG_HAVE_HW_BREAKPOINT code up.

Signed-off-by: Oleg Nesterov <oleg@xxxxxxxxxx>
---

 include/linux/ptrace.h |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

--- ptrace/include/linux/ptrace.h~3_init_jobctl	2011-07-07 18:34:44.000000000 +0200
+++ ptrace/include/linux/ptrace.h	2011-07-08 17:15:20.000000000 +0200
@@ -217,16 +217,17 @@ static inline void ptrace_init_task(stru
 {
 	INIT_LIST_HEAD(&child->ptrace_entry);
 	INIT_LIST_HEAD(&child->ptraced);
-	child->parent = child->real_parent;
+#ifdef CONFIG_HAVE_HW_BREAKPOINT
+	atomic_set(&child->ptrace_bp_refcnt, 1);
+#endif
+	child->jobctl = 0;
 	child->ptrace = 0;
-	if (unlikely(ptrace) && (current->ptrace & PT_PTRACED)) {
+	child->parent = child->real_parent;
+
+	if (unlikely(ptrace) && current->ptrace) {
 		child->ptrace = current->ptrace;
 		__ptrace_link(child, current->parent);
 	}
-
-#ifdef CONFIG_HAVE_HW_BREAKPOINT
-	atomic_set(&child->ptrace_bp_refcnt, 1);
-#endif
 }
 
 /**

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/