From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>, linux-kernel@vger.kernel.org
Subject: Re: [GIT PULL rcu/next] rcu commits for 2.6.40
Date: Sat, 14 May 2011 23:01:43 -0700 [thread overview]
Message-ID: <20110515060143.GE2258@linux.vnet.ibm.com> (raw)
In-Reply-To: <4DCF6789.70701@kernel.org>
On Sat, May 14, 2011 at 10:41:29PM -0700, Yinghai Lu wrote:
> On 05/14/2011 09:14 PM, Yinghai Lu wrote:
> > On 05/14/2011 11:34 AM, Paul E. McKenney wrote:
> >>> and do the inspection afterwards.
> >>
> >> And here is a lightly-tested patch, which applies on tip/core/rcu.
> >>
> >> This problem could account for both the long delays seen with e59fb312
> >> (Decrease memory-barrier usage based on semi-formal proof) and the
> >> shorter delays seen with a26ac245 (move TREE_RCU from softirq to kthread).
> >
> > yes. it fixes the problem.
> >
> > for 1024g system when hotadd mem enabled in kernel config
> >
> > [ 31.814803] cpu_dev_init done
> > [ 35.437163] memory_dev_init done
> >
> > even it is with gcc from opensuse 11.3
>
> got:
>
> [ 86.931217] Switched to NOHz mode on CPU #0
> [ 86.931272] Switched to NOHz mode on CPU #25
> [ 86.931278] ------------[ cut here ]------------
> [ 86.931290] WARNING: at kernel/rcutree.c:364 rcu_enter_nohz+0x44/0x76()
> [ 86.931294] Hardware name: Sun Fire X4800 M2
> [ 86.931297] Modules linked in:
> [ 86.931303] Pid: 0, comm: swapper Not tainted 2.6.39-rc7-tip-yh-04836-g5e42dc2-dirty #3
> [ 86.931307] Call Trace:
> [ 86.931333] [<ffffffff81080280>] warn_slowpath_common+0x85/0x9d
> [ 86.931338] Switched to NOHz mode on CPU #74
> [ 86.931346] [<ffffffff810802b2>] warn_slowpath_null+0x1a/0x1c
> [ 86.931356] [<ffffffff810d3615>] rcu_enter_nohz+0x44/0x76
> [ 86.931370] [<ffffffff810ab3cb>] tick_nohz_stop_sched_tick+0x27d/0x366
> [ 86.931381] [<ffffffff810391bc>] cpu_idle+0x7a/0xcc
> [ 86.931397] [<ffffffff81bd1aa3>] rest_init+0xb7/0xbe
> [ 86.931408] [<ffffffff81bd19ec>] ? csum_partial_copy_generic+0x16c/0x16c
> [ 86.931423] [<ffffffff82738e39>] start_kernel+0x3b2/0x3bd
> [ 86.931428] Switched to NOHz mode on CPU #94
> [ 86.931436] [<ffffffff827382cc>] x86_64_start_reservations+0x9c/0xa0
> [ 86.931446] [<ffffffff827384a8>] x86_64_start_kernel+0x1d8/0x1e3
> [ 86.931463] ---[ end trace 2cfc591bf7de931f ]---
> [ 86.931598] Switched to NOHz mode on CPU #151
> [ 86.931613] Switched to NOHz mode on CPU #152
As I expected!
There is a dyntick entry/exit mismatch somewhere. I haven't yet been able
to find it by inspection, and I cannot reproduce on the systems that I
have access to.
One way that this could happen is if the interrupt-exit code on your
architecture sometimes failed to call irq_exit().
Thanx, Paul
next prev parent reply other threads:[~2011-05-15 6:01 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-08 15:18 [GIT PULL rcu/next] rcu commits for 2.6.40 Paul E. McKenney
2011-05-09 7:36 ` Ingo Molnar
2011-05-09 21:09 ` Yinghai Lu
2011-05-10 8:56 ` Paul E. McKenney
2011-05-10 9:37 ` Ingo Molnar
2011-05-10 18:04 ` Yinghai Lu
2011-05-10 19:32 ` Paul E. McKenney
2011-05-10 20:52 ` Yinghai Lu
2011-05-11 4:54 ` Paul E. McKenney
2011-05-11 6:03 ` Yinghai Lu
2011-05-11 6:42 ` Yinghai Lu
2011-05-11 20:13 ` Paul E. McKenney
2011-05-11 16:54 ` Yinghai Lu
2011-05-11 16:56 ` Yinghai Lu
2011-05-11 20:18 ` Paul E. McKenney
2011-05-11 20:59 ` Yinghai Lu
2011-05-11 21:30 ` Yinghai Lu
2011-05-11 23:02 ` Yinghai Lu
2011-05-12 6:03 ` Ingo Molnar
2011-05-12 7:27 ` Yinghai Lu
2011-05-12 7:42 ` Yinghai Lu
2011-05-12 9:20 ` Paul E. McKenney
2011-05-12 17:31 ` Yinghai Lu
2011-05-12 21:36 ` Yinghai Lu
2011-05-13 1:28 ` Yinghai Lu
2011-05-13 8:42 ` Ingo Molnar
2011-05-13 12:19 ` Ingo Molnar
2011-05-13 13:04 ` Ingo Molnar
2011-05-13 13:12 ` Ingo Molnar
2011-05-13 14:14 ` Paul E. McKenney
2011-05-13 15:07 ` Ingo Molnar
2011-05-13 16:26 ` Paul E. McKenney
2011-05-16 7:08 ` Ingo Molnar
2011-05-16 7:48 ` Paul E. McKenney
2011-05-16 11:51 ` Ingo Molnar
2011-05-16 12:23 ` Ingo Molnar
2011-05-16 14:30 ` Ingo Molnar
2011-05-16 21:33 ` Paul E. McKenney
2011-05-16 22:07 ` Paul E. McKenney
2011-05-16 21:24 ` Paul E. McKenney
2011-05-16 23:52 ` Frederic Weisbecker
2011-05-17 2:40 ` Frederic Weisbecker
2011-05-17 7:53 ` Paul E. McKenney
2011-05-17 12:43 ` Frederic Weisbecker
2011-05-17 22:21 ` Paul E. McKenney
2011-05-18 21:10 ` Yinghai Lu
2011-05-18 23:13 ` Frederic Weisbecker
2011-05-19 4:33 ` Yinghai Lu
2011-05-19 14:47 ` Frederic Weisbecker
2011-05-19 19:51 ` Yinghai Lu
2011-05-19 21:15 ` Frederic Weisbecker
2011-05-19 21:45 ` Yinghai Lu
2011-05-20 0:09 ` [PATCH] rcu: Fix unpaired rcu_irq_enter() from locking selftests Frederic Weisbecker
2011-05-20 8:36 ` Ingo Molnar
2011-05-20 15:12 ` Paul E. McKenney
2011-05-20 15:11 ` Paul E. McKenney
2011-05-20 0:14 ` [GIT PULL rcu/next] rcu commits for 2.6.40 Frederic Weisbecker
2011-05-13 14:40 ` Ingo Molnar
2011-05-13 16:38 ` Paul E. McKenney
2011-05-16 7:10 ` Ingo Molnar
2011-05-13 21:08 ` Yinghai Lu
2011-05-14 14:26 ` Paul E. McKenney
2011-05-14 15:31 ` Paul E. McKenney
2011-05-14 18:34 ` Paul E. McKenney
2011-05-15 3:59 ` Yinghai Lu
2011-05-15 4:14 ` Yinghai Lu
2011-05-15 5:41 ` Yinghai Lu
2011-05-15 5:49 ` Yinghai Lu
2011-05-15 6:04 ` Paul E. McKenney
2011-05-15 6:59 ` Paul E. McKenney
2011-05-16 7:08 ` Paul E. McKenney
2011-05-16 7:39 ` Ingo Molnar
2011-05-15 6:01 ` Paul E. McKenney [this message]
2011-05-15 22:01 ` Frederic Weisbecker
2011-05-16 5:56 ` Paul E. McKenney
2011-05-16 22:40 ` Frederic Weisbecker
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=20110515060143.GE2258@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=yinghai@kernel.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.