[PATCH v2 0/2] exit/pid_ns: comments + simple fix

From: Oleg Nesterov
Date: Wed Nov 26 2014 - 18:54:32 EST


To remind, this doesn't depend on other changes in -mm.

It is not clear to me if I addressed Eric's objections or not, but let
me send V2 (I reordered these patches).

I updated the changelog/comments in 2/2, hopefully now it is more clear
that this patch only tries to explain how the current code works, and why
do we need to wait for nr_hashed==init_pids from the kernel perspective.
Otherwise it is basically the same patch, except I added another note
about sys_wait4().

And, Eric, let me repeat. In any case we do not need for EXIT_DEAD tasks.
They are stealth. In fact this state should die, it only complicates
do_wait/reparenting/etc but we can't avoid it right now (mostly because
of the locking in wait paths).

IOW. If we could somehow detect that this pid_ns has only EXIT_DEAD tasks
we could return and user-space won't notice this change. No, it is not
that I think we should do this, but this should be documented.

Please review.

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/