From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal Hocko Subject: Re: cgroups iptables-restor: vmalloc: allocation failure Date: Thu, 26 Jul 2018 10:13:43 +0200 Message-ID: <20180726081343.GZ28386@dhcp22.suse.cz> References: Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="iso-8859-1" To: Georgi Nikolov Cc: David Rientjes , Tejun Heo , cgroups@vger.kernel.org, linux-kernel@vger.kernel.org For more context this is a follow up for http://lkml.kernel.org/r/df6f501c-= 8546-1f55-40b1-7e3a8f54d872@icdsoft.com On Thu 26-07-18 09:53:20, Georgi Nikolov wrote: > Hello, >=20 > Servers are with 64bit kernel and 32bit userland. >=20 > here is the log: >=20 > [=A0 485.467359] iptables-restor: vmalloc: allocation failure, allocated > 2367488 of 3608576 bytes, mode:0x14010c0(GFP_KERNEL|__GFP_NORETRY), > nodemask=3D(null) > [=A0 485.467871] iptables-restor cpuset=3D/ mems_allowed=3D0 > [=A0 485.468089] CPU: 0 PID: 18093 Comm: iptables-restor Not tainted > 4.14.50 #1 > [=A0 485.468399] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), > BIOS 1.10.2-1 04/01/2014 > [=A0 485.468725]=A0 0000000000000000 ffffffff81ed2b8b ffffffff8333a330 > bb5cfe6a5a821e59 > [=A0 485.469114]=A0 0000000000000000 ffffc90004aaba80 ffffffff811fa3ea > ffffc90004aaba90 > [=A0 485.469422]=A0 ffffffff825e11d0 014010c000000000 ffffffff825e11d0 > ffffc90004aaba18 > [=A0 485.469735] Call Trace: > [=A0 485.469941]=A0 [] dump_stack+0x9a/0x10f > [=A0 485.470159]=A0 [] warn_alloc+0x18a/0x270 > [=A0 485.470379]=A0 [] __vmalloc_node_range+0x251/0x470 > [=A0 485.470614]=A0 [] ? __vmalloc_node_flags_caller+0x= 3e/0x70 > [=A0 485.470881]=A0 [] __vmalloc_node_flags_caller+0x3e= /0x70 > [=A0 485.471160]=A0 [] ? xt_alloc_table_info+0x4e/0xa0 > [=A0 485.471395]=A0 [] kvmalloc_node+0x7a/0xd0 > [=A0 485.471618]=A0 [] xt_alloc_table_info+0x4e/0xa0 > [=A0 485.471849]=A0 [] ? > translate_compat_table.constprop.19+0x447/0x870 > [=A0 485.472197]=A0 [] > translate_compat_table.constprop.19+0x447/0x870 > [=A0 485.472514]=A0 [] ? > compat_do_replace.isra.10.constprop.18+0x197/0x370 > [=A0 485.472833]=A0 [] > compat_do_replace.isra.10.constprop.18+0x197/0x370 > [=A0 485.473230]=A0 [] compat_do_ipt_set_ctl+0x8b/0xb0 > [=A0 485.473465]=A0 [] compat_nf_setsockopt+0x6a/0xd0 > [=A0 485.473701]=A0 [] compat_ip_setsockopt+0x73/0xd0 > [=A0 485.473936]=A0 [] compat_raw_setsockopt+0x7c/0xd0 > [=A0 485.474208]=A0 [] ? handle_mm_fault+0xec/0x240 > [=A0 485.474441]=A0 [] compat_sock_common_setsockopt+0x= 32/0x80 > [=A0 485.474689]=A0 [] compat_sys_setsockopt+0x9a/0x2e0 > [=A0 485.474926]=A0 [] compat_sys_socketcall+0x415/0x570 > [=A0 485.475197]=A0 [] ? page_fault+0x5c/0x90 > [=A0 485.475416]=A0 [] ? retint_user+0x2b/0x2b > [=A0 485.475638]=A0 [] ? rap_compat_sys_socketcall+0x14= /0x40 > [=A0 485.475884]=A0 [] ? do_fast_syscall_32+0xe1/0x2f0 > [=A0 485.476170]=A0 [] ? entry_SYSENTER_compat+0x93/0xa5 > [=A0 485.476414] Mem-Info: > [=A0 485.476579] active_anon:79884 inactive_anon:103007 isolated_anon:0 > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 active_file:54010 inactive_= file:219014 isolated_file:0 > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 unevictable:1822 dirty:22 w= riteback:0 unstable:0 > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 slab_reclaimable:8637 slab_= unreclaimable:25015 > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 mapped:16094 shmem:2271 pag= etables:1481 bounce:0 > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 free:3459 free_pcp:180 free= _cma:0 > [=A0 485.477691] Node 0 active_anon:320376kB inactive_anon:412028kB > active_file:216040kB inactive_file:875844kB unevictable:7288kB > isolated(anon):0kB isolated(file):0kB mapped:64376kB dirty:88kB > writeback:0kB shmem:9084kB writeback_tmp:0kB unstable:0kB > all_unreclaimable? no > [=A0 485.478457] Node 0 DMA free:7896kB min:44kB low:56kB high:68kB > active_anon:2324kB inactive_anon:0kB active_file:12kB > inactive_file:5228kB unevictable:0kB writepending:0kB present:15992kB > managed:15892kB mlocked:0kB kernel_stack:0kB pagetables:0kB bounce:0kB > free_pcp:0kB local_pcp:0kB free_cma:0kB > [=A0 485.479233] lowmem_reserve[]: 0 1963 1963 1963 > [=A0 485.479483] Node 0 DMA32 free:5364kB min:5644kB low:7652kB > high:9660kB active_anon:317672kB inactive_anon:412028kB > active_file:216028kB inactive_file:871340kB unevictable:7288kB > writepending:88kB present:2080624kB managed:2015900kB mlocked:7288kB > kernel_stack:5584kB pagetables:5924kB bounce:0kB free_pcp:1052kB > local_pcp:720kB free_cma:0kB > [=A0 485.480425] lowmem_reserve[]: 0 0 0 0 > [=A0 485.480612] Node 0 DMA: 38*4kB (UMEH) 54*8kB (MEH) 51*16kB (UMEH) > 35*32kB (UMEH) 16*64kB (MEH) 12*128kB (MH) 7*256kB (UMEH) 0*512kB > 1*1024kB (E) 0*2048kB 0*4096kB =3D 7896kB > [=A0 485.481173] Node 0 DMA32: 12*4kB (MH) 16*8kB (H) 264*16kB (UH) 7*32kB > (H) 6*64kB (H) 1*128kB (H) 0*256kB 1*512kB (H) 0*1024kB 0*2048kB > 0*4096kB =3D 5648kB > [=A0 485.481642] Node 0 hugepages_total=3D0 hugepages_free=3D0 > hugepages_surp=3D0 hugepages_size=3D1048576kB > [=A0 485.481975] Node 0 hugepages_total=3D0 hugepages_free=3D0 > hugepages_surp=3D0 hugepages_size=3D2048kB > [=A0 485.482309] 275972 total pagecache pages > [=A0 485.482535] 0 pages in swap cache > [=A0 485.482717] Swap cache stats: add 0, delete 0, find 0/0 > [=A0 485.482937] Free swap=A0 =3D 1976316kB > [=A0 485.483118] Total swap =3D 1976316kB > [=A0 485.483300] 524154 pages RAM > [=A0 485.483492] 0 pages HighMem/MovableOnly > [=A0 485.483711] 16206 pages reserved > [=A0 485.483889] 0 pages hwpoisoned >=20 > Regards, >=20 > --=20 > Georgi Nikolov >=20 > On 07/25/2018 10:42 PM, David Rientjes wrote: > > On Wed, 25 Jul 2018, Georgi Nikolov wrote: > > > >> Hello, > >> > >> I posted a kernel bug https://bugzilla.kernel.org/show_bug.cgi?id=3D20= 0651 and > >> i hope this is the correct place to discuss this. > >> > > Could you post the full allocation failure from the kernel log? It's n= ot=20 > > possible to vmalloc any additional memory due to the memory pressure, s= o=20 > > this is expected to fail but I'm curious as to the actual allocation th= at=20 > > is triggering it. >=20 --=20 Michal Hocko SUSE Labs