From: Cong Wang <amwang@redhat.com>
To: David Miller <davem@davemloft.net>
Cc: bridge@lists.linux-foundation.org, nhorman@tuxdriver.com,
netdev@vger.kernel.org, gospo@redhat.com, fubar@us.ibm.com,
linux-kernel@vger.kernel.org, fbl@sysclose.org,
jmoyer@redhat.com, mpm@selenic.com,
bonding-devel@lists.sourceforge.net
Subject: Re: [Bridge] [v5 Patch 1/3] netpoll: add generic support for bridge and bonding devices
Date: Tue, 08 Jun 2010 16:36:06 +0800 [thread overview]
Message-ID: <4C0E00F6.8020409@redhat.com> (raw)
In-Reply-To: <20100607.030108.235696592.davem@davemloft.net>
On 06/07/10 18:01, David Miller wrote:
> From: Cong Wang<amwang@redhat.com>
> Date: Mon, 07 Jun 2010 17:57:49 +0800
>
>> Hmm, I still feel like this way is ugly, although it may work.
>> I guess David doesn't like it either.
>
> Of course I don't like it. :-)
>
> I suspect the locking scheme will need to be changed.
>
> Besides, if we're going to hack this up and do write lock attempts in
> the read locking paths, there is no point in using a rwlock any more.
> And I'm personally in disfavor of all rwlock usage anyways (it dirties
> the cacheline for readers just as equally for writers, and if the
> critically protected code path is short enough, that shared cache
> line atomic operation will be the predominant cost).
>
> So I'd say, 1) make this a spinlock and 2) try to use RCU for the
> read path.
>
> That would fix everything.
Yeah, agreed. Even not talking about netconsole, bonding code
does have locking problems, netconsole just makes this problem
clear.
I will try your suggestions above.
Thanks!
WARNING: multiple messages have this Message-ID (diff)
From: Cong Wang <amwang@redhat.com>
To: David Miller <davem@davemloft.net>
Cc: andy@greyhouse.net, fubar@us.ibm.com, fbl@sysclose.org,
linux-kernel@vger.kernel.org, mpm@selenic.com,
netdev@vger.kernel.org, bridge@lists.linux-foundation.org,
gospo@redhat.com, nhorman@tuxdriver.com, jmoyer@redhat.com,
shemminger@linux-foundation.org,
bonding-devel@lists.sourceforge.net
Subject: Re: [v5 Patch 1/3] netpoll: add generic support for bridge and bonding devices
Date: Tue, 08 Jun 2010 16:36:06 +0800 [thread overview]
Message-ID: <4C0E00F6.8020409@redhat.com> (raw)
In-Reply-To: <20100607.030108.235696592.davem@davemloft.net>
On 06/07/10 18:01, David Miller wrote:
> From: Cong Wang<amwang@redhat.com>
> Date: Mon, 07 Jun 2010 17:57:49 +0800
>
>> Hmm, I still feel like this way is ugly, although it may work.
>> I guess David doesn't like it either.
>
> Of course I don't like it. :-)
>
> I suspect the locking scheme will need to be changed.
>
> Besides, if we're going to hack this up and do write lock attempts in
> the read locking paths, there is no point in using a rwlock any more.
> And I'm personally in disfavor of all rwlock usage anyways (it dirties
> the cacheline for readers just as equally for writers, and if the
> critically protected code path is short enough, that shared cache
> line atomic operation will be the predominant cost).
>
> So I'd say, 1) make this a spinlock and 2) try to use RCU for the
> read path.
>
> That would fix everything.
Yeah, agreed. Even not talking about netconsole, bonding code
does have locking problems, netconsole just makes this problem
clear.
I will try your suggestions above.
Thanks!
next prev parent reply other threads:[~2010-06-08 8:36 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-05-05 8:11 [Bridge] [v5 Patch 1/3] netpoll: add generic support for bridge and bonding devices Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-05 8:11 ` [Bridge] [v5 Patch 2/3] bridge: make bridge support netpoll Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-05 8:11 ` [Bridge] [v5 Patch 3/3] bonding: make bonding " Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-05 8:11 ` Amerigo Wang
2010-05-06 2:05 ` [Bridge] [v5 Patch 1/3] netpoll: add generic support for bridge and bonding devices Matt Mackall
2010-05-06 2:05 ` Matt Mackall
2010-05-06 7:44 ` [Bridge] " David Miller
2010-05-06 7:44 ` David Miller
2010-05-07 3:24 ` [Bridge] " Cong Wang
2010-05-07 3:24 ` Cong Wang
2010-05-27 18:05 ` [Bridge] " Flavio Leitner
2010-05-27 18:05 ` Flavio Leitner
2010-05-27 20:35 ` [Bridge] " David Miller
2010-05-27 20:35 ` David Miller
2010-05-27 21:25 ` [Bridge] " Flavio Leitner
2010-05-27 21:25 ` Flavio Leitner
2010-05-28 2:47 ` [Bridge] " Cong Wang
2010-05-28 2:47 ` Cong Wang
2010-05-28 19:40 ` [Bridge] " Flavio Leitner
2010-05-28 19:40 ` Flavio Leitner
2010-05-31 5:56 ` [Bridge] " Cong Wang
2010-05-31 5:56 ` Cong Wang
2010-05-31 19:08 ` [Bridge] " Flavio Leitner
2010-05-31 19:08 ` Flavio Leitner
2010-06-01 9:57 ` [Bridge] " Cong Wang
2010-06-01 9:57 ` Cong Wang
2010-06-01 18:42 ` [Bridge] " Jay Vosburgh
2010-06-01 18:42 ` Jay Vosburgh
2010-06-02 10:04 ` [Bridge] " Cong Wang
2010-06-02 10:04 ` Cong Wang
2010-06-04 19:18 ` [Bridge] " Andy Gospodarek
2010-06-04 19:18 ` Andy Gospodarek
2010-06-07 9:57 ` [Bridge] " Cong Wang
2010-06-07 9:57 ` Cong Wang
2010-06-07 10:01 ` [Bridge] " David Miller
2010-06-07 10:01 ` David Miller
2010-06-08 8:36 ` Cong Wang [this message]
2010-06-08 8:36 ` Cong Wang
2010-06-07 13:03 ` [Bridge] " Andy Gospodarek
2010-06-07 13:03 ` Andy Gospodarek
2010-06-08 8:38 ` [Bridge] " Cong Wang
2010-06-08 8:38 ` Cong Wang
2010-06-07 19:24 ` [Bridge] [PATCH] netconsole: queue console messages to send later Flavio Leitner
2010-06-07 19:24 ` Flavio Leitner
2010-06-07 19:50 ` [Bridge] " Matt Mackall
2010-06-07 19:50 ` Matt Mackall
2010-06-07 20:00 ` [Bridge] " Stephen Hemminger
2010-06-07 20:00 ` Stephen Hemminger
2010-06-07 20:21 ` [Bridge] " Matt Mackall
2010-06-07 20:21 ` Matt Mackall
2010-06-07 23:52 ` [Bridge] " David Miller
2010-06-07 23:52 ` David Miller
2010-06-07 23:50 ` [Bridge] " David Miller
2010-06-07 23:50 ` David Miller
2010-06-08 0:37 ` [Bridge] " Flavio Leitner
2010-06-08 0:37 ` Flavio Leitner
2010-06-08 8:59 ` [Bridge] " Cong Wang
2010-06-08 8:59 ` Cong Wang
2010-05-28 8:16 ` [Bridge] [v5 Patch 1/3] netpoll: add generic support for bridge and bonding devices Cong Wang
2010-05-28 8:16 ` Cong Wang
2010-05-28 20:42 ` [Bridge] " Flavio Leitner
2010-05-28 20:42 ` Flavio Leitner
2010-05-28 21:03 ` [Bridge] " Jay Vosburgh
2010-05-28 21:03 ` Jay Vosburgh
2010-05-31 5:29 ` [Bridge] " Cong Wang
2010-05-31 5:29 ` Cong Wang
2010-05-31 5:37 ` [Bridge] " Cong Wang
2010-05-31 5:37 ` Cong Wang
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=4C0E00F6.8020409@redhat.com \
--to=amwang@redhat.com \
--cc=bonding-devel@lists.sourceforge.net \
--cc=bridge@lists.linux-foundation.org \
--cc=davem@davemloft.net \
--cc=fbl@sysclose.org \
--cc=fubar@us.ibm.com \
--cc=gospo@redhat.com \
--cc=jmoyer@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mpm@selenic.com \
--cc=netdev@vger.kernel.org \
--cc=nhorman@tuxdriver.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 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.