Messages in this thread | | | From | Denys Vlasenko <> | Subject | Re: [PATCH 03/10] ptrace: implement PTRACE_SEIZE | Date | Fri, 20 May 2011 00:42:12 +0200 |
| |
On Thursday 19 May 2011 21:31, Pedro Alves wrote: > On Thursday 19 May 2011 15:17:28, Tejun Heo wrote: > > But making SEIZE not trigger INTERRUPT and SETOPTIONS without > > requiring TRACED don't seem too difficult. Jan, would that be enough? > > Oleg, what do you think? > > UUIC, that opens a race where between SEIZEing and > SETOPTIONS(O_TRACE FORK|VFORK|EXEC...), the tracee can > fork/vfork/clone/exec, without the tracer getting the > nice corresponding PTRACE_EVENT_ events.
SEIZE,fork-in-tracee,INTERRUPT sequence is indistinguishable from SEIZE happening two microseconds later:
fork-in-tracee,SEIZE,INTERRUPT
> In GDBs case, GDB will want to poke at memory > right after attaching
...where "right after attaching" is defined as "when the first ptrace-stop is reported". Which will happen very soon.
-- vda
| |