lkml.org 
[lkml]   [2011]   [May]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 03/10] ptrace: implement PTRACE_SEIZE
Hello,

On Wed, May 18, 2011 at 08:17:56PM +0200, Oleg Nesterov wrote:
> On 05/16, Tejun Heo wrote:
> >
> > @@ -315,6 +338,9 @@ static int ptrace_attach(struct task_struct *task)
> > if (task_is_stopped(task)) {
> > task->jobctl |= JOBCTL_TRAP_STOP | JOBCTL_TRAPPING;
> > signal_wake_up(task, 1);
> > + } else if (seize) {
> > + task->jobctl |= JOBCTL_TRAP_STOP;
>
> So, this can race with do_signal_stop(), it can clear TRAP_STOP and
> JOBCTL_STOP_PENDING can guarantee the tracee will trap later.

Confused. How can do_signal_stop() clear TRAP_STOP?

> > + * When PT_SEIZED, it's used for both group stop and explicit
> > + * SEIZE/INTERRUPT traps. Both generate PTRACE_EVENT_STOP trap
> > + * with accompanying siginfo.
> > + *
> > + * When !PT_SEIZED, it's used only for group stop trap with
> > + * CLD_STOPPED as exit_code and no siginfo.
> > */
> > if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) {
> > - signr = current->jobctl & JOBCTL_STOP_SIGMASK;
> > - WARN_ON_ONCE(!signr);
> > - ptrace_stop(signr, CLD_STOPPED, 0, NULL);
> > - current->exit_code = 0;
> > + if (current->ptrace & PT_SEIZED) {
> > + ptrace_do_notify(SIGTRAP | PTRACE_EVENT_STOP << 8,
> > + CLD_STOPPED);
>
> So. When PT_SEIZED, we always report PTRACE_EVENT_STOP and PTRACE_GET_SIGINFO
> always works.
>
> Personally I agree, this looks more clean and natural.

Yeah, I like it much better. INTERRUPT trap and group stop trap being
separate while sharing some attributes was disturbing.

Thanks.

--
tejun


\
 
 \ /
  Last update: 2011-05-19 12:37    [from the cache]
©2003-2011 Jasper Spaans