From mboxrd@z Thu Jan  1 00:00:00 1970
From: Eric Dumazet <eric.dumazet@gmail.com>
Subject: Re: [PATCH v2 3/4] kthread: NUMA aware kthread_create_on_cpu()
Date: Fri, 10 Dec 2010 06:59:58 +0100
Message-ID: <1291960798.2803.18.camel@edumazet-laptop>
References: <1290972833.29196.90.camel@edumazet-laptop>
	 <20101128224024.GA12300@basil.fritz.box>
	 <1290984712.29196.100.camel@edumazet-laptop>
	 <20101128230146.GB12300@basil.fritz.box>
	 <1290987424.29196.128.camel@edumazet-laptop>
	 <20101129090510.GA15763@basil.fritz.box>
	 <1291023532.3435.29.camel@edumazet-laptop>
	 <1291043695.3435.980.camel@edumazet-laptop>
	 <20101209164438.fae1ba4c.akpm@linux-foundation.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Andi Kleen <andi@firstfloor.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	netdev <netdev@vger.kernel.org>,
	David Miller <davem@davemloft.net>, Tejun Heo <tj@kernel.org>,
	Rusty Russell <rusty@rustcorp.com.au>,
	Tony Luck <tony.luck@intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>
To: Andrew Morton <akpm@linux-foundation.org>
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <20101209164438.fae1ba4c.akpm@linux-foundation.org>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: netdev.vger.kernel.org

Le jeudi 09 d=C3=A9cembre 2010 =C3=A0 16:44 -0800, Andrew Morton a =C3=A9=
crit :

> The name "kthread_create_on_cpu" is pretty misleading.
>=20
> One would expect such a function to create a kthread which is bound t=
o
> that CPU.  But what it in fact does is to create a kthread which is
> bound to all CPUs and whose stack, task_struct and thread_info were
> allocated from the node which contains `cpu'.
>=20
> Also, a saner interface would be one which takes the numa_node_id, no=
t
> the cpu number.
>=20


> >
> > ...
> >
> >  /**
> > - * kthread_create - create a kthread.
> > + * kthread_create_on_cpu - create a kthread.
> >   * @threadfn: the function to run until signal_pending(current).
> >   * @data: data ptr for @threadfn.
> > + * @cpu: cpu number.
> >   * @namefmt: printf-style name for the thread.
> >   *
> >   * Description: This helper function creates and names a kernel
> >   * thread.  The thread will be stopped: use wake_up_process() to s=
tart
> >   * it.  See also kthread_run().
> >   *
> > + * If thread is going to be bound on a particular cpu, give its nu=
mber
> > + * in @cpu, to get NUMA affinity for kthread stack, or else give -=
1.
>=20
> This is a bit presumptuous.  The caller might wish to later bind this
> thread to some or all of the CPUs on the node, rather than to a singl=
e
> CPU (eg, kswapd()).
>=20
>=20
> So what to do?  Maybe add a new kthread_create_node() which prepares =
a
> kthread whose memory is bound to that node, then add a
> kthread_create_cpu() convenience wrapper around that?
>=20

We probably can add the "bind to cpu" as a fifth patch, to avoid one=20
kthread_bind(p, cpu);  done by these callers.

My reasoning not including this kthread_bind(p, cpu) in initial patch
series that I was focusing on NUMA properties first, not on scheduling
(this part already runs correctly as far as I know)

Thanks for taking the patch series, I was about to resubmit it today :)