From: Peter Zijlstra <peterz@infradead.org>
To: linux-kernel@vger.kernel.org, torvalds@linux-foundation.org,
manfred@colorfullife.com, dave@stgolabs.net,
paulmck@linux.vnet.ibm.com, will.deacon@arm.com
Cc: boqun.feng@gmail.com, Waiman.Long@hpe.com, tj@kernel.org,
pablo@netfilter.org, kaber@trash.net, davem@davemloft.net,
oleg@redhat.com, netfilter-devel@vger.kernel.org,
sasha.levin@oracle.com, hofrat@osadl.org,
"Peter Zijlstra (Intel)" <peterz@infradead.org>
Subject: [RFC][PATCH 3/3] locking,netfilter: Fix nf_conntrack_lock()
Date: Tue, 24 May 2016 16:27:26 +0200 [thread overview]
Message-ID: <20160524143649.673861121@infradead.org> (raw)
In-Reply-To: 20160524142723.178148277@infradead.org
[-- Attachment #1: peterz-locking-netfilter.patch --]
[-- Type: text/plain, Size: 2037 bytes --]
nf_conntrack_lock{,_all}() is borken as it misses a bunch of memory
barriers to order the whole global vs local locks scheme.
Even x86 (and other TSO archs) are affected.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
net/netfilter/nf_conntrack_core.c | 30 +++++++++++++++++++++++++++++-
1 file changed, 29 insertions(+), 1 deletion(-)
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -74,7 +74,18 @@ void nf_conntrack_lock(spinlock_t *lock)
spin_lock(lock);
while (unlikely(nf_conntrack_locks_all)) {
spin_unlock(lock);
+ /*
+ * Order the nf_contrack_locks_all load vs the spin_unlock_wait()
+ * loads below, to ensure locks_all is indeed held.
+ */
+ smp_rmb(); /* spin_lock(locks_all) */
spin_unlock_wait(&nf_conntrack_locks_all_lock);
+ /*
+ * The control dependency's LOAD->STORE order is enough to
+ * guarantee the spin_lock() is ordered after the above
+ * unlock_wait(). And the ACQUIRE of the lock ensures we are
+ * fully ordered against the spin_unlock() of locks_all.
+ */
spin_lock(lock);
}
}
@@ -119,14 +130,31 @@ static void nf_conntrack_all_lock(void)
spin_lock(&nf_conntrack_locks_all_lock);
nf_conntrack_locks_all = true;
+ /*
+ * Order the above store against the spin_unlock_wait() loads
+ * below, such that if nf_conntrack_lock() observes lock_all
+ * we must observe lock[] held.
+ */
+ smp_mb(); /* spin_lock(locks_all) */
+
for (i = 0; i < CONNTRACK_LOCKS; i++) {
spin_unlock_wait(&nf_conntrack_locks[i]);
}
+ /*
+ * Ensure we observe all state prior to the spin_unlock()s
+ * observed above.
+ */
+ smp_acquire__after_ctrl_dep();
}
static void nf_conntrack_all_unlock(void)
{
- nf_conntrack_locks_all = false;
+ /*
+ * All prior stores must be complete before we clear locks_all.
+ * Otherwise nf_conntrack_lock() might observe the false but not
+ * the entire critical section.
+ */
+ smp_store_release(&nf_conntrack_locks_all, false);
spin_unlock(&nf_conntrack_locks_all_lock);
}
next prev parent reply other threads:[~2016-05-24 14:27 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-24 14:27 [RFC][PATCH 0/3] spin_unlock_wait and assorted borkage Peter Zijlstra
2016-05-24 14:27 ` [RFC][PATCH 1/3] locking: Introduce smp_acquire__after_ctrl_dep Peter Zijlstra
[not found] ` <57451581.6000700@hpe.com>
2016-05-25 4:53 ` Paul E. McKenney
2016-05-25 5:39 ` Boqun Feng
2016-05-25 14:29 ` Paul E. McKenney
2016-05-25 15:20 ` Waiman Long
2016-05-25 15:57 ` Paul E. McKenney
2016-05-25 16:28 ` Peter Zijlstra
2016-05-25 16:54 ` Linus Torvalds
2016-05-25 18:59 ` Paul E. McKenney
2016-06-03 9:18 ` Vineet Gupta
2016-06-03 9:38 ` Peter Zijlstra
2016-06-03 12:08 ` Paul E. McKenney
2016-06-03 12:23 ` Peter Zijlstra
2016-06-03 12:27 ` Peter Zijlstra
2016-06-03 13:33 ` Paul E. McKenney
2016-06-03 13:32 ` Paul E. McKenney
2016-06-03 13:45 ` Will Deacon
2016-06-04 15:29 ` Paul E. McKenney
2016-06-06 17:28 ` Paul E. McKenney
2016-06-07 7:15 ` Peter Zijlstra
2016-06-07 12:41 ` Hannes Frederic Sowa
2016-06-07 13:06 ` Paul E. McKenney
2016-06-07 14:59 ` Hannes Frederic Sowa
2016-06-07 15:23 ` Paul E. McKenney
2016-06-07 17:48 ` Peter Zijlstra
2016-06-07 18:44 ` Paul E. McKenney
2016-06-07 18:01 ` Will Deacon
2016-06-07 18:44 ` Paul E. McKenney
2016-06-07 18:54 ` Paul E. McKenney
2016-06-07 18:37 ` Hannes Frederic Sowa
2016-05-24 14:27 ` [RFC][PATCH 2/3] locking: Annotate spin_unlock_wait() users Peter Zijlstra
2016-05-24 16:17 ` Linus Torvalds
2016-05-24 16:22 ` Tejun Heo
2016-05-24 16:58 ` Peter Zijlstra
2016-05-25 19:28 ` Tejun Heo
2016-05-24 16:57 ` Peter Zijlstra
2016-05-24 14:27 ` Peter Zijlstra [this message]
2016-05-24 14:42 ` [RFC][PATCH 3/3] locking,netfilter: Fix nf_conntrack_lock() Peter Zijlstra
[not found] ` <3e1671fc-be0f-bc95-4fbb-6bfc56e6c15b@colorfullife.com>
2016-05-26 13:54 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160524143649.673861121@infradead.org \
--to=peterz@infradead.org \
--cc=Waiman.Long@hpe.com \
--cc=boqun.feng@gmail.com \
--cc=dave@stgolabs.net \
--cc=davem@davemloft.net \
--cc=hofrat@osadl.org \
--cc=kaber@trash.net \
--cc=linux-kernel@vger.kernel.org \
--cc=manfred@colorfullife.com \
--cc=netfilter-devel@vger.kernel.org \
--cc=oleg@redhat.com \
--cc=pablo@netfilter.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=sasha.levin@oracle.com \
--cc=tj@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).