From: Jesper Dangaard Brouer <hawk@comx.dk>
To: Johannes Berg <johannes@sipsolutions.net>
Cc: "David S. Miller" <davem@davemloft.net>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
dougthompson@xmission.com, bluesmoke-devel@lists.sourceforge.net,
axboe@kernel.dk, Patrick McHardy <kaber@trash.net>,
christine.caulfield@googlemail.com, Trond.Myklebust@netapp.com,
linux-wireless@vger.kernel.org, yoshfuji@linux-ipv6.org,
shemminger@linux-foundation.org, linux-nfs@vger.kernel.org,
bfields@fieldses.org, neilb@suse.de, linux-ext4@vger.kernel.org,
tytso@mit.edu, adilger@sun.com, netfilter-devel@vger.kernel.org
Subject: Re: [PATCH 03/10] mac80211: Use rcu_barrier() on unload.
Date: Wed, 24 Jun 2009 12:06:05 +0200 [thread overview]
Message-ID: <1245837965.6695.45.camel@localhost.localdomain> (raw)
In-Reply-To: <1245770155.21314.38.camel@johannes.local>
On Tue, 2009-06-23 at 17:15 +0200, Johannes Berg wrote:
> On Tue, 2009-06-23 at 17:04 +0200, Jesper Dangaard Brouer wrote:
> > The mac80211 module uses rcu_call() thus it should use rcu_barrier()
> > on module unload.
> >
> > I'm having a hardtime verifying that no more call_rcu() callbacks can
> > be schedules in the module unload path. Could a maintainer please
> > look into this?
> >
> > Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
> > ---
> >
> > net/mac80211/main.c | 2 ++
> > 1 files changed, 2 insertions(+), 0 deletions(-)
> >
> > diff --git a/net/mac80211/main.c b/net/mac80211/main.c
> > index 092a017..e9f70ce 100644
> > --- a/net/mac80211/main.c
> > +++ b/net/mac80211/main.c
> > @@ -1100,6 +1100,8 @@ static void __exit ieee80211_exit(void)
> > ieee80211s_stop();
> >
> > ieee80211_debugfs_netdev_exit();
> > +
> > + rcu_barrier(); /* Wait for completion of call_rcu()'s */
> > }
>
> I don't think this is correct at all -- note that call_rcu() is done in
> some of the mesh code, so I would think you need to do this in
> ieee80211_stop() since the call_rcu() code requires the interface to
> still be around. And when it's stopped everything should be idle.
Should it then not be in mesh.c ieee80211_stop_mesh(). We can replace
the synchronize_rcu() in this function with a rcu_barrier().
> I can fix it later, but I'm deep in some other stuff right now.
Yes, I noticed you seem quite active :-)
I can also do a repost... what about the patch below?
--
Med venlig hilsen / Best regards
Jesper Brouer
ComX Networks A/S
Linux Network developer
Cand. Scient Datalog / MSc.
Author of http://adsl-optimizer.dk
LinkedIn: http://www.linkedin.com/in/brouer
[PATCH v2 03/10] mac80211: Use rcu_barrier() on unload.
From: Jesper Dangaard Brouer <hawk@comx.dk>
The mac80211 module uses rcu_call() thus it should use rcu_barrier()
on module unload.
The rcu_barrier() is placed in mech.c ieee80211_stop_mesh() which is
invoked from ieee80211_stop() in case vif.type == NL80211_IFTYPE_MESH_POINT.
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
---
net/mac80211/mesh.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index fc712e6..11cf45b 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -494,7 +494,7 @@ void ieee80211_stop_mesh(struct ieee80211_sub_if_data *sdata)
* should it be using the interface and enqueuing
* frames at this very time on another CPU.
*/
- synchronize_rcu();
+ rcu_barrier(); /* Wait for RX path and call_rcu()'s */
skb_queue_purge(&sdata->u.mesh.skb_queue);
}
next prev parent reply other threads:[~2009-06-24 10:06 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-06-23 15:03 [PATCH 00/10] We must use rcu_barrier() on module unload Jesper Dangaard Brouer
2009-06-23 15:03 ` [PATCH 01/10] ext4: Use " Jesper Dangaard Brouer
2009-07-06 2:31 ` Theodore Tso
2009-06-23 15:04 ` [PATCH 02/10] bridge: Use rcu_barrier() instead of syncronize_net() on unload Jesper Dangaard Brouer
2009-06-23 15:04 ` [PATCH 03/10] mac80211: Use rcu_barrier() " Jesper Dangaard Brouer
2009-06-23 15:15 ` Johannes Berg
2009-06-24 10:06 ` Jesper Dangaard Brouer [this message]
2009-06-24 10:21 ` Johannes Berg
2009-06-24 11:32 ` Jesper Dangaard Brouer
2009-06-24 11:39 ` Johannes Berg
2009-06-23 15:04 ` [PATCH 04/10] sunrpc: " Jesper Dangaard Brouer
2009-06-23 16:59 ` Trond Myklebust
2009-06-23 15:04 ` [PATCH 05/10] nfs: Use rcu_barrier() on module unload Jesper Dangaard Brouer
2009-06-23 15:04 ` [PATCH 06/10] ipv6: " Jesper Dangaard Brouer
2009-06-23 15:04 ` [PATCH 07/10] decnet: " Jesper Dangaard Brouer
2009-06-24 6:23 ` Chrissie Caulfield
2009-06-24 11:44 ` Jesper Dangaard Brouer
2009-06-24 12:09 ` Jesper Dangaard Brouer
2009-06-23 15:04 ` [PATCH 08/10] edac_core: Uses call_rcu() and its own wait_for_completion scheme Jesper Dangaard Brouer
2009-06-23 15:04 ` [PATCH 09/10] cfq-iosched: Uses its own open-coded rcu_barrier Jesper Dangaard Brouer
2009-06-24 6:42 ` Jens Axboe
2009-06-24 14:05 ` Paul E. McKenney
2009-06-23 15:04 ` [PATCH 10/10] nf_conntrack: Use rcu_barrier() Jesper Dangaard Brouer
2009-06-23 16:23 ` Patrick McHardy
2009-06-24 9:02 ` Jesper Dangaard Brouer
2009-06-24 9:40 ` [PATCH v2 10/10] nf_conntrack: Use rcu_barrier() and fix kmem_cache_create flags Jesper Dangaard Brouer
2009-06-24 13:58 ` Patrick McHardy
2009-06-25 9:29 ` Jesper Dangaard Brouer
2009-06-25 10:02 ` [PATCH v3 10/10] nf_conntrack: Use rcu_barrier() Jesper Dangaard Brouer
2009-06-25 14:33 ` Patrick McHardy
2009-06-25 13:59 ` [PATCH v2 10/10] nf_conntrack: Use rcu_barrier() and fix kmem_cache_create flags Patrick McHardy
2009-06-25 19:32 ` Paul E. McKenney
2009-06-24 1:44 ` [PATCH 00/10] We must use rcu_barrier() on module unload Paul E. McKenney
2009-06-24 7:02 ` David Miller
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=1245837965.6695.45.camel@localhost.localdomain \
--to=hawk@comx.dk \
--cc=Trond.Myklebust@netapp.com \
--cc=adilger@sun.com \
--cc=axboe@kernel.dk \
--cc=bfields@fieldses.org \
--cc=bluesmoke-devel@lists.sourceforge.net \
--cc=christine.caulfield@googlemail.com \
--cc=davem@davemloft.net \
--cc=dougthompson@xmission.com \
--cc=johannes@sipsolutions.net \
--cc=kaber@trash.net \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-wireless@vger.kernel.org \
--cc=neilb@suse.de \
--cc=netdev@vger.kernel.org \
--cc=netfilter-devel@vger.kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=shemminger@linux-foundation.org \
--cc=tytso@mit.edu \
--cc=yoshfuji@linux-ipv6.org \
/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).