Messages in this thread | | | Date | Thu, 08 Nov 2012 13:00:48 +0000 | From | Pedro Alves <> | Subject | Re: PT_EXITKILL (Was: pdeath_signal) |
| |
On 11/08/2012 12:44 PM, Oleg Nesterov wrote: > On 11/08, Pedro Alves wrote:
>> If this isn't inherited by the ptrace child's children, a fork child can >> end up detached if the tracer dies before it had a chance of setting >> the PTRACE_O_EXITKILL on the new auto-attached child. > > It is copied like the other options.
Oh, you're right. I got confused - GDB has code to always set options on the fork children after PTRACE_EVENT_(V)FORK. Dunno where that came from.
>> Which sounds like another argument for PTRACE_O_INHERIT, as in: >> http://sourceware.org/ml/archer/2011-q1/msg00026.html > > The point of PTRACE_O_INHERIT would be to attach newly-created threads and > children without causing an event stop and the attendant overhead. > > this is another thing, I guess.
Yes, yes.
-- Pedro Alves
| |