linux-kernel - Re: [patch v2]numa: add a sysctl to control interleave allocation granularity from each node

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20111207013754.GA23364@sli10-conroe.sh.intel.com>
Date:	Wed, 7 Dec 2011 09:37:54 +0800
From:	Shaohua Li <shaohua.li@...el.com>
To:	David Rientjes <rientjes@...gle.com>
Cc:	lkml <linux-kernel@...r.kernel.org>, linux-mm <linux-mm@...ck.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	"ak@...ux.intel.com" <ak@...ux.intel.com>,
	Jens Axboe <axboe@...nel.dk>, Christoph Lameter <cl@...ux.com>,
	"lee.schermerhorn@...com" <lee.schermerhorn@...com>
Subject: Re: [patch v2]numa: add a sysctl to control interleave allocation
 granularity from each node

On Wed, Dec 07, 2011 at 04:52:56AM +0800, David Rientjes wrote:
> On Mon, 5 Dec 2011, Shaohua Li wrote:
> 
> > If mem plicy is interleaves, we will allocated pages from nodes in a round
> > robin way. This surely can do interleave fairly, but not optimal.
> > 
> > Say the pages will be used for I/O later. Interleave allocation for two pages
> > are allocated from two nodes, so the pages are not physically continuous. Later
> > each page needs one segment for DMA scatter-gathering. But maxium hardware
> > segment number is limited. The non-continuous pages will use up maxium
> > hardware segment number soon and we can't merge I/O to bigger DMA. Allocating
> > pages from one node hasn't such issue. The memory allocator pcp list makes
> > we can get physically continuous pages in several alloc quite likely.
> > 
> > Below patch adds a sysctl to control the allocation granularity from each node.
> > 
> > Run a sequential read workload which accesses disk sdc - sdf. The test uses
> > a LSI SAS1068E card. iostat -x -m 5 shows:
> > 
> > without numactl --interleave=0,1:
> > Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz avgqu-sz   await  svctm  %util
> > sdc              13.40     0.00  259.00    0.00    67.05     0.00   530.19     5.00   19.38   3.86 100.00
> > sdd              13.00     0.00  249.00    0.00    64.95     0.00   534.21     5.05   19.73   4.02 100.00
> > sde              13.60     0.00  258.60    0.00    67.40     0.00   533.78     4.96   18.98   3.87 100.00
> > sdf              13.00     0.00  261.60    0.00    67.50     0.00   528.44     5.24   19.77   3.82 100.00
> > 
> > with numactl --interleave=0,1:
> > sdc               6.80     0.00  419.60    0.00    64.90     0.00   316.77    14.17   34.04   2.38 100.00
> > sdd               6.00     0.00  423.40    0.00    65.58     0.00   317.23    17.33   41.14   2.36 100.00
> > sde               5.60     0.00  419.60    0.00    64.90     0.00   316.77    17.29   40.94   2.38 100.00
> > sdf               5.20     0.00  417.80    0.00    64.17     0.00   314.55    16.69   39.42   2.39 100.00
> > 
> > with numactl --interleave=0,1 and below patch, setting numa_interleave_granularity to 8
> > (setting it to 2 gives similar result, I only recorded the data with 8):
> > sdc              13.00     0.00  261.20    0.00    68.20     0.00   534.74     5.05   19.19   3.83 100.00
> > sde              13.40     0.00  259.00    0.00    67.85     0.00   536.52     4.85   18.80   3.86 100.00
> > sdf              13.00     0.00  260.60    0.00    68.20     0.00   535.97     4.85   18.61   3.84 100.00
> > sdd              13.20     0.00  251.60    0.00    66.00     0.00   537.23     4.95   19.45   3.97 100.00
> > 
> > The avgrq-sz is increased a lot. performance boost a little too.
> > 
> 
> I really like being able to control the interleave granularity, but I 
> think it can be done even better: instead of having a strict count on the 
> number of allocations (slab or otherwise) to allocate on a single node 
> before moving on to another, which could result in large asymmetries 
> between nodes which is the antagonist of any interleaved mempolicy, have 
> you considered basing the granularity on size instead?  interleave_nodes() 
> would then only move onto the next node when a size threshold has been 
> reached.
based on the allocation size, right? I did consider it. It would be easy to
implement this. Note even without my patch we have the issue if allocation
from one node is big order and small order from other node. And nobody
complains the imbalance. This makes me think maybe people didn't care
about the imbalance too much.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/