From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261577AbVFOVT5 (ORCPT ); Wed, 15 Jun 2005 17:19:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261578AbVFOVT5 (ORCPT ); Wed, 15 Jun 2005 17:19:57 -0400 Received: from e6.ny.us.ibm.com ([32.97.182.146]:40392 "EHLO e6.ny.us.ibm.com") by vger.kernel.org with ESMTP id S261577AbVFOVTx (ORCPT ); Wed, 15 Jun 2005 17:19:53 -0400 Subject: Re: 2.6.12-rc6-mm1 & 2K lun testing From: Badari Pulavarty To: Nick Piggin Cc: Linux Kernel Mailing List , linux-mm@kvack.org In-Reply-To: <42B07B44.9040408@yahoo.com.au> References: <1118856977.4301.406.camel@dyn9047017072.beaverton.ibm.com> <42B073C1.3010908@yahoo.com.au> <1118860223.4301.449.camel@dyn9047017072.beaverton.ibm.com> <42B07B44.9040408@yahoo.com.au> Content-Type: text/plain Organization: Message-Id: <1118868979.4301.458.camel@dyn9047017072.beaverton.ibm.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 (1.2.2-5) Date: 15 Jun 2005 13:56:19 -0700 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2005-06-15 at 12:02, Nick Piggin wrote: > Badari Pulavarty wrote: > > On Wed, 2005-06-15 at 11:30, Nick Piggin wrote: > > > >>Badari Pulavarty wrote: > >> > >> > >>>------------------------------------------------------------------------ > >>> > >>>elm3b29 login: dd: page allocation failure. order:0, mode:0x20 > >>> > >>>Call Trace: {__alloc_pages+990} {cache_grow+314} > >>> {cache_alloc_refill+543} {kmem_cache_alloc+54} > >>> {scsi_get_command+81} {scsi_prep_fn+301} > >> > >>They look like they're all in scsi_get_command. > >>I would consider masking off __GFP_HIGH in the gfp_mask of that > >>function, and setting __GFP_NOWARN. It looks like it has a mempoolish > >>thingy in there, so perhaps it shouldn't delve so far into reserves. > > > > > > You want me to take off GFP_HIGH ? or just set GFP_NOWARN with GFP_HIGH > > ? > > > > Yeah, take off GFP_HIGH and set GFP_NOWARN (always). I would be > interested to see how that goes. > > Obviously it won't eliminate your failures there (it will probably > produce more of them), however it might help the scsi command > allocation from overwhelming the system. Hmm.. seems to help little. IO rate is not great (compared to 90MB/sec with "raw") - but machine is making progress. But again, its pretty unresponsive. Thanks, Badari procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 131 254 34896 31328 2540 4982740 0 0 29 101877 1086 11220 0 100 0 0 149 268 34896 32824 2536 4983712 13 0 42 39505 439 4454 0 100 0 0 135 254 34896 31112 2536 4984768 11 0 20 36233 373 4078 0 100 0 0 130 242 34896 32600 2536 4987364 6 0 161 33626 377 3957 0 100 0 0 153 263 34896 32592 2532 4993560 0 0 14 37124 385 4468 0 100 0 0 144 236 34896 32668 2548 5013148 6 0 154 220366 2360 27530 0 100 0 0 112 243 34896 34636 2544 5011112 5 0 62 79160 850 10540 0 100 0 0 103 234 34896 31980 2544 5014744 0 0 135 33814 363 4511 0 100 0 0 112 230 34896 32204 2552 5012156 0 0 140 33200 378 4812 0 100 0 0 139 212 34896 32832 2528 5020928 31 0 542 142834 1536 18007 0 100 0 0 144 215 34896 32896 2528 5019872 17 0 74 41957 449 4781 0 100 0 0 184 252 34896 33252 2504 5024564 0 0 19 34506 374 4616 0 100 0 0 141 240 34896 31624 2516 5026616 0 0 153 31896 378 4904 0 100 0 0