Messages in this thread | | | Date | Sun, 13 Jun 2010 17:29:18 +0200 | From | Oleg Nesterov <> | Subject | Re: [PATCH 1/1] signals: introduce send_sigkill() helper |
| |
Andrew, please drop
signals-introduce-send_sigkill-helper.patch
I am stupid.
On 06/10, Oleg Nesterov wrote: > > Cleanup, no functional changes. > > There are a lot of buggy SIGKILL users in kernel. For example, almost > every force_sig(SIGKILL) is wrong. force_sig() is not safe, it assumes > that the task has the valid ->sighand, and in general it should be used > only for synchronous signals. send_sig(SIGKILL, p, 1) or > send_xxx(SEND_SIG_FORCED/SEND_SIG_PRIV) is not right too but this is not > immediately obvious. > > The only way to correctly send SIGKILL is send_sig_info(SEND_SIG_NOINFO)
No, SEND_SIG_NOINFO doesn't work too. Oh, can't understand what I was thinking about. current is the random task, but send_signal() checks if the caller is from-parent-ns.
> Note: we need more cleanups here, this is only the first change.
We need the cleanups first. Until then oom-killer has to use force_sig() if we want to kill the SIGNAL_UNKILLABLE tasks too.
Oleg.
| |