From: James Bottomley Subject: Re: [BUG] fatal hang untarring 90GB file, possibly writeback related. Date: Tue, 10 May 2011 10:29:57 -0500 Message-ID: <1305041397.6737.12.camel@mulgrave.site> References: <1304025145.2598.24.camel@mulgrave.site> <1304030629.2598.42.camel@mulgrave.site> <20110503091320.GA4542@novell.com> <1304431982.2576.5.camel@mulgrave.site> <1304432553.2576.10.camel@mulgrave.site> <20110506074224.GB6591@suse.de> <20110506080728.GC6591@suse.de> <1304964980.4865.53.camel@mulgrave.site> <20110510102141.GA4149@novell.com> <1305036064.6737.8.camel@mulgrave.site> <20110510143509.GD4146@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Mel Gorman , Jan Kara , colin.king@canonical.com, Chris Mason , linux-fsdevel , linux-mm , linux-kernel , linux-ext4 To: Mel Gorman Return-path: In-Reply-To: <20110510143509.GD4146@suse.de> Sender: owner-linux-mm@kvack.org List-Id: linux-ext4.vger.kernel.org On Tue, 2011-05-10 at 15:35 +0100, Mel Gorman wrote: > On Tue, May 10, 2011 at 09:01:04AM -0500, James Bottomley wrote: > > On Tue, 2011-05-10 at 11:21 +0100, Mel Gorman wrote: > > > I really would like to hear if the fix makes a big difference or > > > if we need to consider forcing SLUB high-order allocations bailing > > > at the first sign of trouble (e.g. by masking out __GFP_WAIT in > > > allocate_slab). Even with the fix applied, kswapd might be waking up > > > less but processes will still be getting stalled in direct compaction > > > and direct reclaim so it would still be jittery. > > > > "the fix" being this > > > > https://lkml.org/lkml/2011/3/5/121 > > > > Drop this for the moment. It was a long shot at best and there is little > evidence the problem is in this area. > > I'm attaching two patches. The first is the NO_KSWAPD one to stop > kswapd being woken up by SLUB using speculative high-orders. The second > one is more drastic and prevents slub entering direct reclaim or > compaction. It applies on top of patch 1. These are both untested and > afraid are a bit rushed as well :( Preliminary results with both patches applied still show kswapd periodically going up to 99% but it doesn't stay there, it comes back down again (and, obviously, the system doesn't hang). This is sysrq-M from a couple of times when it went up there: [ 426.736958] SysRq : Show Memory [ 426.736974] Mem-Info: [ 426.736977] Node 0 DMA per-cpu: [ 426.736983] CPU 0: hi: 0, btch: 1 usd: 0 [ 426.736986] CPU 1: hi: 0, btch: 1 usd: 0 [ 426.736989] CPU 2: hi: 0, btch: 1 usd: 0 [ 426.736993] CPU 3: hi: 0, btch: 1 usd: 0 [ 426.736996] Node 0 DMA32 per-cpu: [ 426.737002] CPU 0: hi: 186, btch: 31 usd: 169 [ 426.737005] CPU 1: hi: 186, btch: 31 usd: 40 [ 426.737009] CPU 2: hi: 186, btch: 31 usd: 166 [ 426.737012] CPU 3: hi: 186, btch: 31 usd: 168 [ 426.737015] Node 0 Normal per-cpu: [ 426.737020] CPU 0: hi: 0, btch: 1 usd: 0 [ 426.737024] CPU 1: hi: 0, btch: 1 usd: 0 [ 426.737027] CPU 2: hi: 0, btch: 1 usd: 0 [ 426.737030] CPU 3: hi: 0, btch: 1 usd: 0 [ 426.737036] active_anon:108658 inactive_anon:37031 isolated_anon:0 [ 426.737037] active_file:32006 inactive_file:41051 isolated_file:32 [ 426.737038] unevictable:8 dirty:41202 writeback:204 unstable:0 [ 426.737039] free:191520 slab_reclaimable:8490 slab_unreclaimable:27477 [ 426.737039] mapped:9176 shmem:26412 pagetables:5427 bounce:0 [ 426.737051] Node 0 DMA free:8140kB min:548kB low:684kB high:820kB active_anon:0kB inactive_anon:12kB active_file:240kB inactive_file:7280kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15676kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:180kB slab_unreclaimable:16kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no [ 426.737064] lowmem_reserve[]: 0 1856 1862 1862 [ 426.737078] Node 0 DMA32 free:757892kB min:66820kB low:83524kB high:100228kB active_anon:434632kB inactive_anon:148112kB active_file:127784kB inactive_file:156924kB unevictable:32kB isolated(anon):0kB isolated(file):0kB present:1901408kB mlocked:32kB dirty:164808kB writeback:816kB mapped:36704kB shmem:105648kB slab_reclaimable:33676kB slab_unreclaimable:108372kB kernel_stack:2304kB pagetables:21708kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:160 all_unreclaimable? no [ 426.737092] lowmem_reserve[]: 0 0 5 5 [ 426.737106] Node 0 Normal free:48kB min:212kB low:264kB high:316kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:6060kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:104kB slab_unreclaimable:1520kB kernel_stack:176kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes [ 426.737119] lowmem_reserve[]: 0 0 0 0 [ 426.737132] Node 0 DMA: 3*4kB 2*8kB 1*16kB 1*32kB 2*64kB 2*128kB 2*256kB 2*512kB 2*1024kB 2*2048kB 0*4096kB = 8140kB [ 426.737174] Node 0 DMA32: 945*4kB 585*8kB 6469*16kB 4871*32kB 3189*64kB 1338*128kB 228*256kB 58*512kB 24*1024kB 1*2048kB 0*4096kB = 757884kB [ 426.737227] Node 0 Normal: 2*4kB 5*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 48kB [ 426.737257] 99595 total pagecache pages [ 426.737260] 0 pages in swap cache [ 426.737263] Swap cache stats: add 0, delete 0, find 0/0 [ 426.737266] Free swap = 3768316kB [ 426.737268] Total swap = 3768316kB [ 426.744603] 525808 pages RAM [ 426.744612] 57618 pages reserved [ 426.744614] 141551 pages shared [ 426.744617] 186065 pages non-shared [ 472.301810] SysRq : Show Memory [ 472.301826] Mem-Info: [ 472.301829] Node 0 DMA per-cpu: [ 472.301835] CPU 0: hi: 0, btch: 1 usd: 0 [ 472.301839] CPU 1: hi: 0, btch: 1 usd: 0 [ 472.301842] CPU 2: hi: 0, btch: 1 usd: 0 [ 472.301845] CPU 3: hi: 0, btch: 1 usd: 0 [ 472.301848] Node 0 DMA32 per-cpu: [ 472.301854] CPU 0: hi: 186, btch: 31 usd: 184 [ 472.301857] CPU 1: hi: 186, btch: 31 usd: 46 [ 472.301860] CPU 2: hi: 186, btch: 31 usd: 158 [ 472.301863] CPU 3: hi: 186, btch: 31 usd: 163 [ 472.301866] Node 0 Normal per-cpu: [ 472.301871] CPU 0: hi: 0, btch: 1 usd: 0 [ 472.301874] CPU 1: hi: 0, btch: 1 usd: 0 [ 472.301878] CPU 2: hi: 0, btch: 1 usd: 0 [ 472.301881] CPU 3: hi: 0, btch: 1 usd: 0 [ 472.301886] active_anon:107673 inactive_anon:37031 isolated_anon:0 [ 472.301887] active_file:31533 inactive_file:33323 isolated_file:32 [ 472.301888] unevictable:8 dirty:26256 writeback:6475 unstable:0 [ 472.301889] free:198742 slab_reclaimable:9347 slab_unreclaimable:28647 [ 472.301889] mapped:8307 shmem:26412 pagetables:5427 bounce:0 [ 472.301901] Node 0 DMA free:8140kB min:548kB low:684kB high:820kB active_anon:0kB inactive_anon:12kB active_file:240kB inactive_file:7280kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15676kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:180kB slab_unreclaimable:16kB kernel_stack:0kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no [ 472.301915] lowmem_reserve[]: 0 1856 1862 1862 [ 472.301928] Node 0 DMA32 free:786780kB min:66820kB low:83524kB high:100228kB active_anon:430692kB inactive_anon:148112kB active_file:125892kB inactive_file:126012kB unevictable:32kB isolated(anon):0kB isolated(file):128kB present:1901408kB mlocked:32kB dirty:105024kB writeback:25900kB mapped:33228kB shmem:105648kB slab_reclaimable:37104kB slab_unreclaimable:113052kB kernel_stack:2288kB pagetables:21708kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:3196 all_unreclaimable? no [ 472.301943] lowmem_reserve[]: 0 0 5 5 [ 472.301956] Node 0 Normal free:48kB min:212kB low:264kB high:316kB active_anon:0kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:6060kB mlocked:0kB dirty:0kB writeback:0kB mapped:0kB shmem:0kB slab_reclaimable:104kB slab_unreclaimable:1520kB kernel_stack:176kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? yes [ 472.301968] lowmem_reserve[]: 0 0 0 0 [ 472.301982] Node 0 DMA: 3*4kB 2*8kB 1*16kB 1*32kB 2*64kB 2*128kB 2*256kB 2*512kB 2*1024kB 2*2048kB 0*4096kB = 8140kB [ 472.302015] Node 0 DMA32: 6121*4kB 1912*8kB 5094*16kB 4920*32kB 3168*64kB 1381*128kB 262*256kB 68*512kB 24*1024kB 1*2048kB 0*4096kB = 786756kB [ 472.302048] Node 0 Normal: 2*4kB 5*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 48kB [ 472.302090] 91312 total pagecache pages [ 472.302093] 0 pages in swap cache [ 472.302096] Swap cache stats: add 0, delete 0, find 0/0 [ 472.302099] Free swap = 3768316kB [ 472.302102] Total swap = 3768316kB [ 472.309521] 525808 pages RAM [ 472.309529] 57618 pages reserved [ 472.309548] 142496 pages shared [ 472.309551] 177182 pages non-shared I'll finish off this verification, and then re-run with the watch-highorder script running. James -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org