On 10/16, Oleg Nesterov wrote:
>
> On 10/15, Roland McGrath wrote:
> >
> > > Roland, do you see other problems with attach/detach which
> > > should be fixed before v1?
> >
> > It depends what v1 means, on which I am not entirely clear.
>
> The definition of v1 is very simple and absolutely precise.
> The code is v1 when Roland thinks we can make the RFC patch and
> send it for review.
As I see it, we need the following changes.
utrace:
- utrace_control(SINGLESTEP) calls get_user_pages() under
spin_lock()
ptrace:
- stacked VFORK_DONE event, finish ptrace_report_clone().
The fix must be trivial, but every time I am trying to
complete this I understand I should do this with a fresh
head ;)
- PTRACE_SYSEMU/PTRACE_SYSEMU_SINGLESTEP
should be simple...
- stacked JCTL events. Oh. I hope this will not be too hard
to do something simple and "not too buggy" for now.
- ptrace_check_attch().
Must be simple, but I am starting to think I have a blind
spot here.
- simple cleanups, may be more comments.
Anything else?
Oleg.
