Skip to contentSkip to conversation

malicious filesystems (was Re: [linux-pm] Re: [PATCH] Remove process freezer from suspend to RAM

DISCUSSION Oleg Nesterov 3 participants 3 messages Jul 14, 2007 – Jul 21, 2007

lore
Share & tools
lore
#0 lore raw

I am sorry for being completely off-topic, but

Quoted 2 lines · Miklos Szeredi

SIGKILL won't work on a stopped task. Neither on a traced task.

Why do you think so? It works in both cases (ptracer can use PT_TRACE_EXIT, but the task is killed anyway).

Quoted 2 lines

Neither on a zombie (how many newbies are thoroughly confused about that ;)

This is true. Unless this zombie is a group leader, and we have other threads.

Oleg.

#1 lore raw
Quoted 4 lines
Quoted 1 line

SIGKILL won't work on a stopped task. Neither on a traced task.

Why do you think so? It works in both cases (ptracer can use PT_TRACE_EXIT, but the task is killed anyway).

Just from experience with tasks stuck in "T" state. After for example an UML dies, some of the processes only react to SIGCONT, but not to SIGKILL.

I have no idea why this happens.

Miklos

#2 lore raw
Quoted 10 lines · Miklos Szeredi
Quoted 4 lines
Quoted 1 line

SIGKILL won't work on a stopped task. Neither on a traced task.

Why do you think so? It works in both cases (ptracer can use PT_TRACE_EXIT, but the task is killed anyway).

Just from experience with tasks stuck in "T" state. After for example an UML dies, some of the processes only react to SIGCONT, but not to SIGKILL.

I have no idea why this happens.

Anyway that should be fixed, not used as excuse for fuse misbehaviour.
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html