[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#949171: linux: Please, enable COMPACTION for armel/marvell to mitigate OOMs



Source: linux
Version: 4.19.67-2+deb10u2
Severity: important
X-Debbugs-CC: debian-arm@lists.debian.org

Hi, Ben and others.

I've been getting some OOM errors on my armel/kirkwood device running Debian
stable. One of the reasons for this is, according to the dmesg logs, that
COMPACT (memory compaction) is tried, but not available:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
$ grep -i compact /boot/config-4.19.0-6-marvell 
# CONFIG_COMPACTION is not set
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Here is a snippet of the relevant part of the dmesg logs from an earlier
kernel, but the problem still manifests:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(...)
[3558478.729941] kswapd0: page allocation failure: order:0, mode:0x480020(GFP_ATOMIC), nodemask=(null)
[3558478.738990] CPU: 0 PID: 20 Comm: kswapd0 Not tainted 4.19.0-5-marvell #1 Debian 4.19.37-5
[3558478.747317] Hardware name: Marvell Orion5x (Flattened Device Tree)
[3558478.753697] [<c0016c6c>] (unwind_backtrace) from [<c001370c>] (show_stack+0x18/0x1c)
[3558478.761620] [<c001370c>] (show_stack) from [<c00f5af0>] (warn_alloc+0x90/0x13c)
[3558478.769110] [<c00f5af0>] (warn_alloc) from [<c00f61f8>] (__alloc_pages_nodemask+0x5c4/0xa30)
[3558478.777725] [<c00f61f8>] (__alloc_pages_nodemask) from [<c00f6744>] (page_frag_alloc+0x64/0x150)
[3558478.786693] [<c00f6744>] (page_frag_alloc) from [<c02f9e84>] (__netdev_alloc_skb+0x84/0xfc)
[3558478.795268] [<c02f9e84>] (__netdev_alloc_skb) from [<bf3f12a0>] (rxq_refill+0x74/0x174 [mv643xx_eth])
[3558478.804739] [<bf3f12a0>] (rxq_refill [mv643xx_eth]) from [<bf3f222c>] (mv643xx_eth_poll+0x524/0x558 [mv643xx_eth])
[3558478.815289] [<bf3f222c>] (mv643xx_eth_poll [mv643xx_eth]) from [<c030f01c>] (net_rx_action+0x124/0x380)
[3558478.824857] [<c030f01c>] (net_rx_action) from [<c000a1c8>] (__do_softirq+0x1c0/0x2b0)
[3558478.832862] [<c000a1c8>] (__do_softirq) from [<c0025ae0>] (irq_exit+0x80/0xd4)
[3558478.840273] [<c0025ae0>] (irq_exit) from [<c00567c8>] (__handle_domain_irq+0x7c/0xa8)
[3558478.848298] [<c00567c8>] (__handle_domain_irq) from [<c023b22c>] (orion_handle_irq+0x7c/0x88)
[3558478.857002] [<c023b22c>] (orion_handle_irq) from [<c00099e8>] (__irq_svc+0x68/0x84)
[3558478.864818] Exception stack(0xc79a9d10 to 0xc79a9d58)
[3558478.870045] 9d00:                                     c0731338 b4ace000 c795f5a0 c23fe858
[3558478.878392] 9d20: c79a9d74 c0aad000 00000001 c059e4fa 00000319 c04829bd 00000001 b4ace000
[3558478.886733] 9d40: 00cd914f c79a9d60 c011dcc8 c011dbb0 a0000013 ffffffff
[3558478.893548] [<c00099e8>] (__irq_svc) from [<c011dbb0>] (page_vma_mapped_walk+0x40/0x168)
[3558478.901830] [<c011dbb0>] (page_vma_mapped_walk) from [<c011e400>] (page_referenced_one+0x54/0x164)
[3558478.910969] [<c011e400>] (page_referenced_one) from [<c011e778>] (rmap_walk_anon+0x94/0x12c)
[3558478.919588] [<c011e778>] (rmap_walk_anon) from [<c011fccc>] (page_referenced+0x140/0x174)
[3558478.927937] [<c011fccc>] (page_referenced) from [<c0103268>] (shrink_active_list+0x2c0/0x408)
[3558478.936636] [<c0103268>] (shrink_active_list) from [<c0103700>] (shrink_node+0x350/0x764)
[3558478.944981] [<c0103700>] (shrink_node) from [<c01044d0>] (kswapd+0x41c/0x5d4)
[3558478.952293] [<c01044d0>] (kswapd) from [<c003b7bc>] (kthread+0x120/0x138)
[3558478.959264] [<c003b7bc>] (kthread) from [<c00090e0>] (ret_from_fork+0x14/0x34)
[3558478.966640] Exception stack(0xc79a9fb0 to 0xc79a9ff8)
[3558478.971864] 9fa0:                                     00000000 00000000 00000000 00000000
[3558478.980211] 9fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[3558478.988552] 9fe0: 00000000 00000000 00000000 00000000 00000013 00000000
[3558478.995319] Mem-Info:
[3558478.997805] active_anon:3221 inactive_anon:2017 isolated_anon:32
                  active_file:9147 inactive_file:8585 isolated_file:0
                  unevictable:0 dirty:199 writeback:0 unstable:0
                  slab_reclaimable:3447 slab_unreclaimable:1480
                  mapped:3434 shmem:1 pagetables:422 bounce:0
                  free:985 free_pcp:6 free_cma:0
[3558479.031259] Node 0 active_anon:12884kB inactive_anon:8068kB active_file:36588kB inactive_file:34340kB unevictable:0kB isolated(anon):128kB isolated(file):0kB mapped:13736kB dirty:796kB writeback:0kB shmem:4kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
[3558479.054496] Normal free:3940kB min:2048kB low:2560kB high:3072kB active_anon:12884kB inactive_anon:8068kB active_file:36588kB inactive_file:34340kB unevictable:0kB writepending:796kB present:131072kB managed:124152kB mlocked:0kB kernel_stack:600kB pagetables:1688kB bounce:0kB free_pcp:24kB local_pcp:24kB free_cma:0kB
[3558479.082604] lowmem_reserve[]: 0 0 0
[3558479.086277] Normal: 13*4kB (H) 8*8kB (H) 9*16kB (H) 5*32kB (H) 7*64kB (H) 4*128kB (H) 0*256kB 1*512kB (H) 0*1024kB 1*2048kB (H) 0*4096kB = 3940kB
[3558479.099586] 18154 total pagecache pages
[3558479.103589] 421 pages in swap cache
[3558479.107247] Swap cache stats: add 184262, delete 183841, find 531277/573632
[3558479.114360] Free swap  = 462844kB
[3558479.117841] Total swap = 524284kB
[3558479.121323] 32768 pages RAM
[3558479.124286] 0 pages HighMem/MovableOnly
[3558479.128287] 1730 pages reserved
[3558479.234279] apt: page allocation failure: order:0, mode:0x480020(GFP_ATOMIC), nodemask=(null)
[3558479.242980] CPU: 0 PID: 25580 Comm: apt Not tainted 4.19.0-5-marvell #1 Debian 4.19.37-5
[3558479.251220] Hardware name: Marvell Orion5x (Flattened Device Tree)
[3558479.257599] [<c0016c6c>] (unwind_backtrace) from [<c001370c>] (show_stack+0x18/0x1c)
[3558479.265521] [<c001370c>] (show_stack) from [<c00f5af0>] (warn_alloc+0x90/0x13c)
[3558479.273013] [<c00f5af0>] (warn_alloc) from [<c00f61f8>] (__alloc_pages_nodemask+0x5c4/0xa30)
[3558479.281627] [<c00f61f8>] (__alloc_pages_nodemask) from [<c00f6744>] (page_frag_alloc+0x64/0x150)
[3558479.290595] [<c00f6744>] (page_frag_alloc) from [<c02f9e84>] (__netdev_alloc_skb+0x84/0xfc)
[3558479.299171] [<c02f9e84>] (__netdev_alloc_skb) from [<bf3f12a0>] (rxq_refill+0x74/0x174 [mv643xx_eth])
[3558479.308625] [<bf3f12a0>] (rxq_refill [mv643xx_eth]) from [<bf3f222c>] (mv643xx_eth_poll+0x524/0x558 [mv643xx_eth])
[3558479.319175] [<bf3f222c>] (mv643xx_eth_poll [mv643xx_eth]) from [<c030f01c>] (net_rx_action+0x124/0x380)
[3558479.328741] [<c030f01c>] (net_rx_action) from [<c000a1c8>] (__do_softirq+0x1c0/0x2b0)
[3558479.336746] [<c000a1c8>] (__do_softirq) from [<c0025ae0>] (irq_exit+0x80/0xd4)
[3558479.344147] [<c0025ae0>] (irq_exit) from [<c00567c8>] (__handle_domain_irq+0x7c/0xa8)
[3558479.352169] [<c00567c8>] (__handle_domain_irq) from [<c023b22c>] (orion_handle_irq+0x7c/0x88)
[3558479.360876] [<c023b22c>] (orion_handle_irq) from [<c00099e8>] (__irq_svc+0x68/0x84)
[3558479.368684] Exception stack(0xc0b1fb28 to 0xc0b1fb70)
[3558479.373917] fb20:                   c7fab600 00000001 00000000 00010408 c7fab600 c7f5eda0
[3558479.382261] fb40: 00005930 c7fab604 c0b1fbf0 c05dadb4 c055d8e0 00000100 00000002 c0b1fb78
[3558479.390596] fb60: c00f49f0 c00f2a70 60000013 ffffffff
[3558479.395838] [<c00099e8>] (__irq_svc) from [<c00f2a70>] (free_pcp_prepare+0x40/0x98)
[3558479.403672] [<c00f2a70>] (free_pcp_prepare) from [<c00f49f0>] (free_unref_page_list+0x84/0x160)
[3558479.412547] [<c00f49f0>] (free_unref_page_list) from [<c01017f4>] (shrink_page_list+0x88/0xee8)
[3558479.421418] [<c01017f4>] (shrink_page_list) from [<c0102e08>] (shrink_inactive_list+0x224/0x3c4)
[3558479.430376] [<c0102e08>] (shrink_inactive_list) from [<c0103a84>] (shrink_node+0x6d4/0x764)
[3558479.438896] [<c0103a84>] (shrink_node) from [<c0103fd8>] (try_to_free_pages+0x4c4/0x5a0)
[3558479.447159] [<c0103fd8>] (try_to_free_pages) from [<c00f6160>] (__alloc_pages_nodemask+0x52c/0xa30)
[3558479.456384] [<c00f6160>] (__alloc_pages_nodemask) from [<c00faa7c>] (__do_page_cache_readahead+0x11c/0x1a0)
[3558479.466305] [<c00faa7c>] (__do_page_cache_readahead) from [<c00ef2e4>] (generic_file_read_iter+0x2c8/0x8c4)
[3558479.476230] [<c00ef2e4>] (generic_file_read_iter) from [<c01326d4>] (new_sync_read+0xec/0x10c)
[3558479.485025] [<c01326d4>] (new_sync_read) from [<c0134744>] (vfs_read+0x98/0x14c)
[3558479.492602] [<c0134744>] (vfs_read) from [<c0134c34>] (ksys_read+0x44/0x7c)
[3558479.499743] [<c0134c34>] (ksys_read) from [<c0009000>] (ret_fast_syscall+0x0/0x50)
[3558479.507468] Exception stack(0xc0b1ffa8 to 0xc0b1fff0)
[3558479.512702] ffa0:                   b6e2f780 be90bf4c 00000004 02510059 00007f7a 00000000
[3558479.521053] ffc0: b6e2f780 be90bf4c 00007f7a 00000003 be90be00 02510059 00007f7a 00000000
[3558479.529387] ffe0: 0000006c be90bd70 b6e29768 b6b847e0
[3558627.671129] git invoked oom-killer: gfp_mask=0x7080c0(GFP_KERNEL_ACCOUNT|__GFP_ZERO), nodemask=(null), order=1, oom_score_adj=0
[3558627.682817] COMPACTION is disabled!!!
[3558627.686762] CPU: 0 PID: 25994 Comm: git Not tainted 4.19.0-5-marvell #1 Debian 4.19.37-5
[3558627.695079] Hardware name: Marvell Orion5x (Flattened Device Tree)
[3558627.701529] [<c0016c6c>] (unwind_backtrace) from [<c001370c>] (show_stack+0x18/0x1c)
[3558627.709530] [<c001370c>] (show_stack) from [<c00f2104>] (dump_header.constprop.4+0x7c/0x1c4)
[3558627.718251] [<c00f2104>] (dump_header.constprop.4) from [<c00f1378>] (oom_kill_process+0x90/0x404)
[3558627.727471] [<c00f1378>] (oom_kill_process) from [<c00f1ff0>] (out_of_memory+0x394/0x3e0)
[3558627.735904] [<c00f1ff0>] (out_of_memory) from [<c00f6474>] (__alloc_pages_nodemask+0x840/0xa30)
[3558627.744862] [<c00f6474>] (__alloc_pages_nodemask) from [<c001f688>] (copy_process.part.4+0x118/0x16d0)
[3558627.754424] [<c001f688>] (copy_process.part.4) from [<c0020de4>] (_do_fork+0xe0/0x3a4)
[3558627.762603] [<c0020de4>] (_do_fork) from [<c0021190>] (sys_clone+0x24/0x2c)
[3558627.769822] [<c0021190>] (sys_clone) from [<c0009000>] (ret_fast_syscall+0x0/0x50)
[3558627.777636] Exception stack(0xc4a11fa8 to 0xc4a11ff0)
[3558627.782956] 1fa0:                   b68654c8 b6865460 003d0f00 b6864f98 b68654c8 b6865920
[3558627.791377] 1fc0: b68654c8 b6865460 be817efc 00000078 be817efa b6865460 be817efc be817f44
[3558627.799792] 1fe0: 003d0f00 be817e80 b6ea2194 b6e0c858
[3558627.805077] Mem-Info:
[3558627.807633] active_anon:2575 inactive_anon:1136 isolated_anon:0
                  active_file:11462 inactive_file:7830 isolated_file:6
                  unevictable:0 dirty:25 writeback:1 unstable:0
                  slab_reclaimable:3121 slab_unreclaimable:1482
                  mapped:10690 shmem:2 pagetables:475 bounce:0
                  free:1201 free_pcp:36 free_cma:0
[3558627.841339] Node 0 active_anon:10300kB inactive_anon:4544kB active_file:45848kB inactive_file:31320kB unevictable:0kB isolated(anon):0kB isolated(file):24kB mapped:42760kB dirty:100kB writeback:4kB shmem:8kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
[3558627.864566] Normal free:4804kB min:2048kB low:2560kB high:3072kB active_anon:10300kB inactive_anon:4544kB active_file:45848kB inactive_file:31320kB unevictable:0kB writepending:104kB present:131072kB managed:124152kB mlocked:0kB kernel_stack:656kB pagetables:1900kB bounce:0kB free_pcp:144kB local_pcp:144kB free_cma:0kB
[3558627.892933] lowmem_reserve[]: 0 0 0
[3558627.896686] Normal: 229*4kB (UMEH) 8*8kB (H) 9*16kB (H) 5*32kB (H) 7*64kB (H) 4*128kB (H) 0*256kB 1*512kB (H) 0*1024kB 1*2048kB (H) 0*4096kB = 4804kB
[3558627.910421] 19527 total pagecache pages
[3558627.914496] 230 pages in swap cache
[3558627.918222] Swap cache stats: add 185216, delete 184986, find 531476/573984
[3558627.925407] Free swap  = 461564kB
[3558627.928955] Total swap = 524284kB
[3558627.932505] 32768 pages RAM
[3558627.935538] 0 pages HighMem/MovableOnly
[3558627.939606] 1730 pages reserved
[3558627.942971] Tasks state (memory values in pages):
[3558627.947902] [  pid  ]   uid  tgid total_vm      rss pgtables_bytes swapents oom_score_adj name
[3558627.956758] [    134]     0   134    15558      283    75776      163             0 systemd-journal
[3558627.966037] [    152]     0   152     1180        0    12288       47             0 blkmapd
[3558627.974624] [    160]     0   160     4604       32    18432      140         -1000 systemd-udevd
[3558627.983731] [    165]   108   165     5779      536    20480      138             0 systemd-network
[3558627.993020] [    226]     0   226      850        0    10240       52             0 rpc.idmapd
[3558628.001877] [    228]   107   228     5631      200    24576      149             0 systemd-timesyn
[3558628.011156] [    229]   113   229     1621       51    14336       78             0 rpcbind
[3558628.019743] [    230]     0   230     8335       80    40960     6507             0 rpc.mountd
[3558628.028590] [    240]     0   240     1867      136    16384       39             0 cron
[3558628.036920] [    241]     0   241     3397      501    20480      138             0 systemd-logind
[3558628.046111] [    243]   102   243     1789      422    14336       57          -900 dbus-daemon
[3558628.055045] [    247]   103   247     1615      223    14336       68             0 avahi-daemon
[3558628.064090] [    256]     0   256     1510      376    14336      154             0 smartd
[3558628.072598] [    262]     0   262     2749       27    18432      157         -1000 sshd
[3558628.080931] [    267]   103   267     1583        0    14336       72             0 avahi-daemon
[3558628.089952] [    270]     0   270     1530        0    14336       32             0 agetty
[3558628.098454] [    271]     0   271      590       82    10240       21             0 syslogd
[3558628.107052] [    283]    33   283     2572      163    20480      303             0 lighttpd
[3558628.115727] [  15189]     0 15189    11888      266    55296      445             0 smbd
[3558628.124053] [  15191]     0 15191    11211       69    49152      457             0 smbd-notifyd
[3558628.133072] [  15192]     0 15192    11214       73    49152      457             0 cleanupd
[3558628.141748] [  15205]     0 15205     7792      266    36864      313             0 nmbd
[3558628.150079] [  24970]     0 24970     3154       52    18432      203             0 sshd
[3558628.158405] [  24973]  1000 24973     3677       48    20480      223             0 systemd
[3558628.166994] [  24974]  1000 24974     8957       51    30720      433             0 (sd-pam)
[3558628.175667] [  24987]  1000 24987     3239      225    18432      183             0 sshd
[3558628.183998] [  24988]  1000 24988     2084       45    14336       94             0 bash
[3558628.192326] [  24993]  1000 24993     2528       50    18432       96             0 su
[3558628.200482] [  24994]     0 24994     2175      567    14336       47             0 bash
[3558628.208813] [  25056]   104 25056    33975      134    94208     4667             0 minidlnad
[3558628.217584] [  25954]     0 25954    13164    11044    57344        0             0 apt
[3558628.225825] [  25959]     0 25959    13164     3085    30720        0             0 apt
[3558628.234066] [  25960]     0 25960      501      276     8192        0             0 sh
[3558628.242223] [  25961]     0 25961      501      296    10240        0             0 etckeeper
[3558628.250983] [  25986]     0 25986      501      274     8192        0             0 50uncommitted-c
[3558628.260263] [  25987]     0 25987      501      330     8192        0             0 etckeeper
[3558628.269022] [  25993]     0 25993      501      303    10240        0             0 50test
[3558628.277524] [  25994]     0 25994     4599      918    16384        0             0 git
[3558628.285758] Out of memory: Kill process 25954 (apt) score 68 or sacrifice child
[3558628.293325] Killed process 25959 (apt) total-vm:52656kB, anon-rss:10724kB, file-rss:1616kB, shmem-rss:0kB
[3558628.330213] oom_reaper: reaped process 25959 (apt), now anon-rss:0kB, file-rss:0kB, shmem-rss:0kB
[3559321.964279] systemd[1]: Stopping Journal Service...
[3559321.969754] systemd-journald[134]: Received SIGTERM from PID 1 (systemd).
[3559322.475409] systemd-journal: 56 output lines suppressed due to ratelimiting
[3559322.486605] systemd[1]: systemd-journald.service: Succeeded.
[3559322.498942] systemd[1]: Stopped Journal Service.
[3559322.568880] systemd[1]: Starting Journal Service...
[3559323.616592] systemd[1]: Started Journal Service.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Can CONFIG_COMPACT please be enabled for armel (or only marvel, if that's
the case)?


Thanks in advance,

Rogério Brito.

P.S.: Please, disregard the information collected by reportbug after this,
as I am writing this message on an amd64 system.  I will possibly file a bug
report regarding the TAINT_WARN for this system that I am using after I
debug it a little bit.

-- System Information:
Debian Release: bullseye/sid
  APT prefers testing
  APT policy: (500, 'testing'), (200, 'unstable'), (150, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.4.0-2-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_WARN
Locale: LANG=en_US.utf-8, LC_CTYPE=pt_BR.utf-8 (charmap=UTF-8), LANGUAGE=en_US.utf-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

-- 
Rogério Brito : rbrito@{ime.usp.br,gmail.com} : GPG key 4096R/BCFCAAAA
http://cynic.cc/blog/ : github.com/rbrito : profiles.google.com/rbrito
DebianQA: http://qa.debian.org/developer.php?login=rbrito%40ime.usp.br


Reply to: