Messages in this thread | | | Date | Wed, 13 Jul 2011 14:10:38 +0200 | From | Tejun Heo <> | Subject | Re: [RFC PATCH 3/3] ptrace: dont send SIGSTOP on auto-attach if PT_SEIZED |
| |
Hello,
On Fri, Jul 08, 2011 at 07:14:17PM +0200, Oleg Nesterov wrote: > The fake SIGSTOP during attach has numerous problems. PTRACE_SEIZE > is already fine, but we have basically the same problems is SIGSTOP > is sent on auto-attach, the tracer can't know if this signal signal > should be cancelled or not. > > Change ptrace_event() to set JOBCTL_TRAP_STOP if the new child is > PT_SEIZED, this triggers the PTRACE_EVENT_STOP report. > > Thereafter a PT_SEIZED task can never report the bogus SIGSTOP. > > Test-case: ... > Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Looks good to me. For 2 and 3,
Acked-by: Tejun Heo <tj@kernel.org>
So, yeah, with this change, we're almost there.
Thanks.
-- tejun
| |