linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@tv-sign.ru>
To: vatsa@in.ibm.com
Cc: linux-kernel@vger.kernel.org,
	Dipankar Sarma <dipankar@in.ibm.com>,
	Manfred Spraul <manfred@colorfullife.com>,
	Linus Torvalds <torvalds@osdl.org>,
	"Paul E. McKenney" <paulmck@us.ibm.com>,
	Andrew Morton <akpm@osdl.org>
Subject: Re: [PATCH 2/5] rcu: don't check ->donelist in __rcu_pending()
Date: Mon, 09 Jan 2006 17:51:08 +0300	[thread overview]
Message-ID: <43C2785C.4E937748@tv-sign.ru> (raw)
In-Reply-To: 20060109093141.GA10811@in.ibm.com

Srivatsa Vaddagiri wrote:
> On Sun, Jan 08, 2006 at 10:19:24PM +0300, Oleg Nesterov wrote:
> > ->donelist becomes != NULL only in rcu_process_callbacks().
> >
> > rcu_process_callbacks() always calls rcu_do_batch() when
> > ->donelist != NULL.
> >
> > rcu_do_batch() schedules rcu_process_callbacks() again if
> > ->donelist was not flushed entirely.
> >
> > So ->donelist != NULL means that rcu_tasklet is either
> > TASKLET_STATE_SCHED or TASKLET_STATE_RUN, we don't need to
> > check it in __rcu_pending().
>
> Do I smell a bug wrt CPU Hotplug here? Basically, I see that we do
> a rcu_move_batch of ->curlist and ->nxtlist of the dead CPU. Why not
> ->donelist?

After the quick reading CPU Hotplug code I think you are right, there
is a bug in rcu_offline_cpu(). It should also move ->donelist, it is
lost otherwise.

>             If we have to do a rcu_move_batch of ->donelist also,
> then perhaps the ->donelist != NULL check is required in
> rcu_pending?

rcu_move_batch() always adds entries to the ->nxttail, so I think
this patch is correct.

>               This is considering that the RCU tasklet of the dead
> CPU is killed (rather than moved over to a different CPU).

Yes, it is killed explicitly in rcu_offline_cpu() via tasklet_kill_immediate().

Note that we can't remove this tasklet_kill_immediate() and rely on
takeover_tasklets(). rcu_process_callbacks() does __get_cpu_var(),
so it can't find orphaned rcu_data anyway if the tasklet was moved
to another cpu.

So, do we need something like this (untested, uncompiled) patch or not?

--- 2.6.15/kernel/rcupdate.c~	2006-01-09 20:23:32.000000000 +0300
+++ 2.6.15/kernel/rcupdate.c	2006-01-09 20:26:20.000000000 +0300
@@ -355,8 +355,9 @@ static void __rcu_offline_cpu(struct rcu
 	spin_unlock_bh(&rcp->lock);
 	rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail);
 	rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail);
-
+	rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail);
 }
+
 static void rcu_offline_cpu(int cpu)
 {
 	struct rcu_data *this_rdp = &get_cpu_var(rcu_data);

Oleg.

  reply	other threads:[~2006-01-09 13:34 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-01-08 19:19 [PATCH 2/5] rcu: don't check ->donelist in __rcu_pending() Oleg Nesterov
2006-01-08 22:00 ` Paul E. McKenney
2006-01-09  9:31 ` Srivatsa Vaddagiri
2006-01-09 14:51   ` Oleg Nesterov [this message]
2006-01-09 13:42     ` Srivatsa Vaddagiri
2006-01-09 15:33       ` Oleg Nesterov
2006-01-09 18:59 ` Paul E. McKenney
2006-01-09 20:31   ` Oleg Nesterov
2006-01-09 19:59     ` Paul E. McKenney
2006-01-10  9:58       ` Srivatsa Vaddagiri
2006-01-10 14:24         ` [PATCH] rcu: fix hotplug-cpu ->donelist leak Oleg Nesterov
2006-01-11  5:01           ` Paul E. McKenney
2006-01-11 16:28           ` Paul E. McKenney
2006-01-11 19:25             ` Oleg Nesterov
2006-01-11 18:21               ` Paul E. McKenney
2006-01-10 14:27         ` [PATCH 2/5] rcu: don't check ->donelist in __rcu_pending() Oleg Nesterov
2006-01-10 18:13 ` Dipankar Sarma

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=43C2785C.4E937748@tv-sign.ru \
    --to=oleg@tv-sign.ru \
    --cc=akpm@osdl.org \
    --cc=dipankar@in.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=manfred@colorfullife.com \
    --cc=paulmck@us.ibm.com \
    --cc=torvalds@osdl.org \
    --cc=vatsa@in.ibm.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).