... This patch avoid taking tasklist_lock in kill_something_info() when > the pid is -1. It can convert to rcu_read_lock() for this case because > group_send_sig_info() doesn't ... group_send_sig_info() drops ->siglock. Except: We don't send the signal to ...

linux-kernel - Oleg Nesterov - Mar 25 2008 - 09:56

... This patch avoid taking tasklist_lock in kill_something_info() when >> >> the pid is -1. It can convert to rcu_read_lock() for this case because >> >> group_send_sig_info() doesn't ... per task siglock for protection. >> >> > Except: We don't send the ...

linux-kernel - Eric W. Biederman - May 31 2008 - 19:55

... This patch avoid taking tasklist_lock in kill_something_info() when > >> the pid is -1. It can convert to rcu_read_lock() for this case because > >> group_send_sig_info() doesn't ... the per task siglock for protection. > > > Except: We don't send the ...

linux-kernel - Oleg Nesterov - May 31 2008 - 12:55

... have a look? This patch avoid taking tasklist_lock in kill_something_info() when the pid is -1. It can convert to rcu_read_lock() for this case because group_send_sig_info() doesn't need tasklist_lock. This patch is for 2.6.25-rc5-mm1. ...

linux-kernel - Atsushi Tsuji - Mar 25 2008 - 00:27

... >> This patch avoid taking tasklist_lock in kill_something_info() when >> the pid is -1. It can convert to rcu_read_lock() for this case because >> group_send_sig_info() doesn't need ... of the current process, because we don't also send the signal to ...

linux-kernel - Atsushi Tsuji - May 20 2008 - 21:48

... under the rcu > lock. So I don't believe we are guaranteed to ... case it is possible that (without tasklist_lock) kill_something_info() > > sends the signal to the ... > > Note the (broken) "p->pid > 1" check, kill_something_info() skips init. > > Not that ...

linux-kernel - Oleg Nesterov - Jun 1 2008 - 12:29

... This patch avoid taking tasklist_lock in kill_something_info() when >> the pid is -1. It can convert to rcu_read_lock() for this case because >> group_send_sig_info() doesn't ... the per task siglock for protection. > Except: We don't send the signal ...

linux-kernel - Eric W. Biederman - May 20 2008 - 23:47

... >>> This patch avoid taking tasklist_lock in kill_something_info() when >>> the pid is -1. It can convert to rcu_read_lock() for this case because >>> group_send_sig_info() doesn't need ... > Call me paranoid but I don't think there is any guarantee ...

linux-kernel - Atsushi TSUJI - May 26 2008 - 03:03

... jp.nec.com> writes: > Sorry for late reply and thank you ... the current process, because we don't also > send the signal ... in > kill_something_info() can cause stall when some processes execute > kill(-1 ... problem. I think my patch doesn't make the new problem ...

linux-kernel - Eric W. Biederman - May 20 2008 - 22:53

... I think the change to convert tasklist_lock to > rcu_read_lock is reasonable way to avoid the local DOS for kill(-1,sig) case. We can ... rarely hit it will take ages for someone to trace back ... proposing will always work I don't have a problem. Eric ...

linux-kernel - Eric W. Biederman - May 28 2008 - 11:03