From: Nikolay Aleksandrov <nikolay-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Thomas Graf <tgraf-G/eBtMaohhA@public.gmane.org>
Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org,
netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org,
davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org,
netfilter-devel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
walpole-sKt6ljEC1JY3uPMLIKxrzw@public.gmane.org,
tklauser-93Khv+1bN0NyDzI6CaY1VQ@public.gmane.org,
paulmck-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org,
kaber-dcUjhNyLwpNeoWH0uzbU5w@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected hash table
Date: Fri, 01 Aug 2014 17:20:41 +0200 [thread overview]
Message-ID: <53DBB049.30902@redhat.com> (raw)
In-Reply-To: <20140801151527.GF7331-FZi0V3Vbi30CUdFEqe4BF2D2FQJk+8+b@public.gmane.org>
On 08/01/2014 05:15 PM, Thomas Graf wrote:
> On 08/01/14 at 04:51pm, Nikolay Aleksandrov wrote:
>> Hmm, in both the rhashtable_insert() and rhashtable_remove() calls in the
>> netlink code you're using GFP_ATOMIC flags but if rhashtable_expand/shring gets
>> called even though the allocation will be with GFP_ATOMIC, they still call
>> synchronize_rcu() which may block. Now I'm not familiar with the netlink code,
>> but I think that in general the flags are useless for GFP_ATOMIC because of the
>> calls to synchronize_rcu() in expand/shrink which can block anyway.
>> Just a thought, I may be missing something of course.
>
> I don't think you are missing anything. The GFP_ATOMIC flag was
> inherited from how the bucket table was allocated prior to the
> convertion to the new hash table but you are right, it can't be
> needed since the table protection was converted to a mutex.
> Using GFP_KERNEL will have a better chance of succeeding.
>
Right, I was wondering why it was atomic in the first place and couldn't find a
good reason from the code :-)
But that explains it.
Cheers,
Nik
WARNING: multiple messages have this Message-ID (diff)
From: Nikolay Aleksandrov <nikolay@redhat.com>
To: Thomas Graf <tgraf@suug.ch>
Cc: davem@davemloft.net, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, kaber@trash.net,
paulmck@linux.vnet.ibm.com, josh@joshtriplett.org,
challa@noironetworks.com, walpole@cs.pdx.edu,
dev@openvswitch.org, tklauser@distanz.ch,
netfilter-devel@vger.kernel.org
Subject: Re: [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected hash table
Date: Fri, 01 Aug 2014 17:20:41 +0200 [thread overview]
Message-ID: <53DBB049.30902@redhat.com> (raw)
In-Reply-To: <20140801151527.GF7331@casper.infradead.org>
On 08/01/2014 05:15 PM, Thomas Graf wrote:
> On 08/01/14 at 04:51pm, Nikolay Aleksandrov wrote:
>> Hmm, in both the rhashtable_insert() and rhashtable_remove() calls in the
>> netlink code you're using GFP_ATOMIC flags but if rhashtable_expand/shring gets
>> called even though the allocation will be with GFP_ATOMIC, they still call
>> synchronize_rcu() which may block. Now I'm not familiar with the netlink code,
>> but I think that in general the flags are useless for GFP_ATOMIC because of the
>> calls to synchronize_rcu() in expand/shrink which can block anyway.
>> Just a thought, I may be missing something of course.
>
> I don't think you are missing anything. The GFP_ATOMIC flag was
> inherited from how the bucket table was allocated prior to the
> convertion to the new hash table but you are right, it can't be
> needed since the table protection was converted to a mutex.
> Using GFP_KERNEL will have a better chance of succeeding.
>
Right, I was wondering why it was atomic in the first place and couldn't find a
good reason from the code :-)
But that explains it.
Cheers,
Nik
next prev parent reply other threads:[~2014-08-01 15:20 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-01 11:58 [PATCH net-next v4 0/3] Lockless netlink_lookup() with new concurrent hash table Thomas Graf
2014-08-01 11:58 ` Thomas Graf
[not found] ` <cover.1406891028.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-01 11:58 ` [PATCH net-next 1/3] lib: Resizable, Scalable, Concurrent Hash Table Thomas Graf
2014-08-01 11:58 ` Thomas Graf
2014-08-01 11:58 ` [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected hash table Thomas Graf
2014-08-01 11:58 ` Thomas Graf
[not found] ` <72a64dfee4f20f2ca596df26f3e4ae543cf4c068.1406891028.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-01 14:51 ` Nikolay Aleksandrov
2014-08-01 14:51 ` Nikolay Aleksandrov
[not found] ` <53DBA976.8030103-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2014-08-01 15:15 ` Thomas Graf
2014-08-01 15:15 ` Thomas Graf
[not found] ` <20140801151527.GF7331-FZi0V3Vbi30CUdFEqe4BF2D2FQJk+8+b@public.gmane.org>
2014-08-01 15:20 ` Nikolay Aleksandrov [this message]
2014-08-01 15:20 ` Nikolay Aleksandrov
2014-08-01 11:58 ` [PATCH net-next 3/3] nftables: Convert nft_hash to use generic rhashtable Thomas Graf
[not found] ` <bd4d9fa542072358cb0bd77e493ec5785a1d9666.1406891028.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-01 12:21 ` Nikolay Aleksandrov
2014-08-01 12:21 ` Nikolay Aleksandrov
2014-08-01 13:49 ` Thomas Graf
2014-08-01 14:12 ` Patrick McHardy
2014-08-01 14:51 ` Thomas Graf
2014-08-01 15:00 ` Patrick McHardy
-- strict thread matches above, loose matches on Subject: below --
2014-08-02 9:47 [PATCH net-next v5 0/3] Lockless netlink_lookup() with new concurrent hash table Thomas Graf
[not found] ` <cover.1406971567.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-02 9:47 ` [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected " Thomas Graf
2014-08-02 9:47 ` Thomas Graf
[not found] ` <a0370889bd1b1bdf097f74df7017d39ae845c12e.1406971567.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-02 10:44 ` Nikolay Aleksandrov
2014-08-02 10:44 ` Nikolay Aleksandrov
2014-08-05 2:10 ` Sasha Levin
2014-08-05 2:10 ` Sasha Levin
[not found] ` <53E03D0B.3000208-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
2014-08-05 2:58 ` David Miller
2014-08-05 2:58 ` David Miller
[not found] ` <20140804.195841.536725238366145203.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2014-08-05 5:50 ` Eric Dumazet
2014-08-05 5:50 ` Eric Dumazet
[not found] ` <1407217807.3178.70.camel-XN9IlZ5yJG9HTL0Zs8A6p/gx64E7kk8eUsxypvmhUTTZJqsBc5GL+g@public.gmane.org>
2014-08-05 5:58 ` David Miller
2014-08-05 5:58 ` David Miller
2014-08-01 8:51 [PATCH net-next v3 0/3] Lockless netlink_lookup() with new concurrent " Thomas Graf
[not found] ` <cover.1406882738.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-08-01 8:51 ` [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected " Thomas Graf
2014-08-01 8:51 ` Thomas Graf
2014-07-31 22:56 [PATCH v2 0/3 net-next] Lockless netlink_lookup() with new concurrent " Thomas Graf
[not found] ` <cover.1406846586.git.tgraf-G/eBtMaohhA@public.gmane.org>
2014-07-31 22:56 ` [PATCH net-next 2/3] netlink: Convert netlink_lookup() to use RCU protected " Thomas Graf
2014-07-31 22:56 ` Thomas Graf
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=53DBB049.30902@redhat.com \
--to=nikolay-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
--cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org \
--cc=dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org \
--cc=josh-iaAMLnmF4UmaiuxdJuQwMA@public.gmane.org \
--cc=kaber-dcUjhNyLwpNeoWH0uzbU5w@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=netfilter-devel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=paulmck-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org \
--cc=tgraf-G/eBtMaohhA@public.gmane.org \
--cc=tklauser-93Khv+1bN0NyDzI6CaY1VQ@public.gmane.org \
--cc=walpole-sKt6ljEC1JY3uPMLIKxrzw@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.