Re: mmotm 2015-01-22-15-04: qemu failure due to 'mm: memcontrol: remove unnecessary soft limit tree node test'

From: Johannes Weiner <hannes@cmpxchg.org>
To: Christoph Lameter <cl@linux.com>
Cc: Guenter Roeck <linux@roeck-us.net>,
	akpm@linux-foundation.org, mm-commits@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	linux-fsdevel@vger.kernel.org, linux-next@vger.kernel.org,
	sfr@canb.auug.org.au, mhocko@suse.cz
Subject: Re: mmotm 2015-01-22-15-04: qemu failure due to 'mm: memcontrol: remove unnecessary soft limit tree node test'
Date: Sat, 24 Jan 2015 02:16:23 -0500	[thread overview]
Message-ID: <20150124071623.GA17705@phnom.home.cmpxchg.org> (raw)
In-Reply-To: <alpine.DEB.2.11.1501231508020.7871@gentwo.org>

On Fri, Jan 23, 2015 at 03:09:20PM -0600, Christoph Lameter wrote:
> On Fri, 23 Jan 2015, Guenter Roeck wrote:
> 
> > Wouldn't that have unintended consequences ? So far
> > rb tree nodes are allocated even if a node not online;
> > the above would change that. Are you saying it is
> > unnecessary to initialize rb tree nodes if the node
> > is not online ?
> 
> It is not advisable to allocate since an offline node means that the
> structure cannot be allocated on the node where it would be most
> beneficial. Typically subsystems allocate the per node data structures
> when the node is brought online.

I would generally agree, but this code, which implements a userspace
interface, is already grotesquely inefficient and heavyhanded.  It's
also superseded in the next release, so we can just keep this simple
at this point.