From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Wed, 16 May 2007 10:03:10 +0100 (IST) From: Mel Gorman Subject: Re: [PATCH 1/2] Have kswapd keep a minimum order free other than order-0 In-Reply-To: <464AC00E.10704@yahoo.com.au> Message-ID: References: <20070514173218.6787.56089.sendpatchset@skynet.skynet.ie> <20070514173238.6787.57003.sendpatchset@skynet.skynet.ie> <20070514182456.GA9006@skynet.ie> <1179218576.25205.1.camel@rousalka.dyndns.org> <464AC00E.10704@yahoo.com.au> MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="29444707-1639998271-1179306096=:7139" Content-ID: Sender: owner-linux-mm@kvack.org Return-Path: To: Nick Piggin Cc: Nicolas Mailhot , Christoph Lameter , Andy Whitcroft , akpm@linux-foundation.org, Linux Memory Management List List-ID: --29444707-1639998271-1179306096=:7139 Content-Type: TEXT/PLAIN; CHARSET=X-UNKNOWN; FORMAT=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Content-ID: On Wed, 16 May 2007, Nick Piggin wrote: > Mel Gorman wrote: >> On Tue, 15 May 2007, Nicolas Mailhot wrote: >>=20 >>> Le lundi 14 mai 2007 =E0 19:24 +0100, Mel Gorman a =E9crit : >>>=20 >>>> On (14/05/07 11:13), Christoph Lameter didst pronounce: >>>>=20 >>>>> I think the slub fragment may have to be this way? This calls >>>>> raise_kswapd_order on each kmem_cache_create with the order of the ca= che >>>>> that was created thus insuring that the min_order is correctly. >>>>>=20 >>>>> Signed-off-by: Christoph Lameter >>>>>=20 >>>>=20 >>>> Good plan. Revised patch as follows; >>>=20 >>>=20 >>> Kernel with this patch and the other one survives testing. I'll stop >>> heavy testing now and consider the issue closed. >>>=20 >>=20 >> That is good news, thanks for the report. >>=20 >>> Thanks for looking at my bug report. >>>=20 >>=20 >> Thank you very much for your testing. I know it was a lot to ask to tie = a=20 >> machine up for a few days. > > Hmm, so we require higher order pages be kept free even if nothing is > using them? That's not very nice :( > Not quite. We are already required to keep a minimum number of pages free= =20 even though nothing is using them. The difference is that if it is known=20 high-order allocations are frequently required, the freed pages will be=20 contiguous. If no one calls raise_kswapd_order(), kswapd will continue=20 reclaiming at order-0. Arguably, e1000 should also be calling=20 raise_kswapd_order() when it is using jumbo frames. --=20 Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab --29444707-1639998271-1179306096=:7139-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org