linux-kernel - [PATCH] sched: allow preempt notifiers to self-unregister.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <1324052151-3714-1-git-send-email-pierre.habouzit@intersec.com>
Date:	Fri, 16 Dec 2011 17:15:50 +0100
From:	Pierre Habouzit <pierre.habouzit@...ersec.com>
To:	linux-kernel@...r.kernel.org
Cc:	Pierre Habouzit <pierre.habouzit@...ersec.com>,
	Avi Kivity <avi@...ranet.com>, Ingo Molnar <mingo@...e.hu>,
	Peter Zijlstra <peterz@...radead.org>
Subject: [PATCH] sched: allow preempt notifiers to self-unregister.

This is very annoying when unlike kvm (the sole in-tree user I know of)
the process is a plain user process the kernel doesn't really handle and
that it wants to unregister the callbacks just at its death.

The code that one would like to write is:

    static my_module_sched_out(struct preempt_notifier *notifier)
    {
        struct my_module_type *mmi =
            container_of(notifier, struct my_module_type, notifier);

        if (unlikely(current->state & TASK_DEAD)) {
            preempt_notifier_unregister(notifier);
            kfree(mmi);
            return;
        }

        /* do our real work */
    }

Sadly this isn't allowed because the htlist walk isn't "safe" against
removals.

Signed-off-by: Pierre Habouzit <pierre.habouzit@...ersec.com>
Cc: Avi Kivity <avi@...ranet.com>
Cc: Ingo Molnar <mingo@...e.hu>
Cc: Peter Zijlstra <peterz@...radead.org>
---
 kernel/sched.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

  As a background, this need is because I have some kind of module code
  that uses this facility to evaluate how many of a group of threads are
  concurrently running (to regulate a pool of threads).

  Hence I install those callbacks for the thread registering themselves
  and want to keep them until the thread dies. Sadly I have no way to
  unregister those callbacks right now, but for horrible hacks (involving
  private delayed queues processed regularly walked to kfree() the
  structures referencing pids that are dead, urgh).

diff --git a/kernel/sched.c b/kernel/sched.c
index d6b149c..2653169 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3106,7 +3106,7 @@ static void fire_sched_in_preempt_notifiers(struct task_struct *curr)
 	struct preempt_notifier *notifier;
 	struct hlist_node *node;
 
-	hlist_for_each_entry(notifier, node, &curr->preempt_notifiers, link)
+	hlist_for_each_entry_safe(notifier, node, &curr->preempt_notifiers, link)
 		notifier->ops->sched_in(notifier, raw_smp_processor_id());
 }
 
@@ -3117,7 +3117,7 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr,
 	struct preempt_notifier *notifier;
 	struct hlist_node *node;
 
-	hlist_for_each_entry(notifier, node, &curr->preempt_notifiers, link)
+	hlist_for_each_entry_safe(notifier, node, &curr->preempt_notifiers, link)
 		notifier->ops->sched_out(notifier, next);
 }
 
-- 
1.7.8.rc3.348.gb51a37

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