Date: Tue, 17 Feb 2015 14:26:13 +0900
From: Joonsoo Kim <iamjoonsoo.kim@lge.com>
To: Christoph Lameter <cl@linux.com>
Cc: akpm@linuxfoundation.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        penberg@kernel.org, iamjoonsoo@lge.com,
        Jesper Dangaard Brouer <brouer@redhat.com>
Subject: Re: [PATCH 2/3] slub: Support for array operations
Message-ID: <20150217052612.GD15413@js1304-P5Q-DELUXE>
References: <20150210194804.288708936@linux.com>
 <20150210194811.902155759@linux.com>
 <20150213024515.GB6592@js1304-P5Q-DELUXE>
 <alpine.DEB.2.11.1502130948120.9442@gentwo.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <alpine.DEB.2.11.1502130948120.9442@gentwo.org>
User-Agent: Mutt/1.5.21 (2010-09-15)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 983
Lines: 27

On Fri, Feb 13, 2015 at 09:49:24AM -0600, Christoph Lameter wrote:
> On Fri, 13 Feb 2015, Joonsoo Kim wrote:
> 
> > > +			*p++ = freelist;
> > > +			freelist = get_freepointer(s, freelist);
> > > +			allocated++;
> > > +		}
> >
> > Fetching all objects with holding node lock could result in enomourous
> > lock contention. How about getting free ojbect pointer without holding
> > the node lock? We can temporarilly store all head of freelists in
> > array p and can fetch each object pointer without holding node lock.
> 
> 
> Could do that but lets first see if there is really an issue. The other
> cpu sharing the same partial lists presumaly have cpu local objects to get
> through first before they hit this lock.

Okay.

Thanks.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/