OK CONFIG_DEBUG_BUGVERBOSE enabled kernel BUG at mm/rmap.c:399! this one line added with merged commit b3d574ae after remove it: ---- diff --git a/mm/rmap.c b/mm/rmap.c index 71cd5bd..68d115f 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -396,7 +396,6 @@ void unlink_anon_vmas(struct vm_area_struct *vma) list_for_each_entry_safe(avc, next, &vma->anon_vma_chain, same_vma) { struct anon_vma *anon_vma = avc->anon_vma; - BUG_ON(anon_vma->degree); put_anon_vma(anon_vma); list_del(&avc->same_vma); ---- system working stable and not generate oops. On 11.01.2015 11:56, Vlastimil Babka wrote: > Re: subject > > why do you think it was this commit? bisection? The commit you point to is > "Merge branch 'akpm' (patches from Andrew)" > It's possible that bisection pointed you to this, instead of the proper commit > though. Luckily, there are only 12 of them and ... > > On 01/10/2015 01:15 PM, Krzysztof Kolasa wrote: >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320920] ------------[ cut >> here ]------------ >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320927] Kernel BUG at >> ffffffff81187f1f [verbose debug info unavailable] > Could you enable CONFIG_DEBUG_BUGVERBOSE please? This should make the stacktrace > readable. > >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320929] invalid opcode: 0000 >> [#3] SMP >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320932] Modules linked in: >> dm_crypt(E) pci_stub(E) vboxpci(OE) vboxnetadp(OE) vboxnetflt(OE) >> vboxdrv(OE) c >> use(E) arc4(E) md4(E) bnep(E) rfcomm(E) bluetooth(E) nls_utf8(E) cifs(E) >> fscache(E) binfmt_misc(E) fglrx(OE) uvcvideo(E) videobuf2_vmalloc(E) >> x86_pkg_ >> temp_thermal(E) videobuf2_memops(E) videobuf2_core(E) hp_wmi(E) >> v4l2_common(E) wl(POE) kvm_intel(E) sparse_keymap(E) videodev(E) kvm(E) >> ghash_clmulni_ >> intel(E) aesni_intel(E) aes_x86_64(E) lrw(E) gf128mul(E) >> snd_hda_codec_idt(E) snd_hda_codec_hdmi(E) snd_hda_codec_generic(E) >> glue_helper(E) ablk_helpe >> r(E) snd_hda_intel(E) cryptd(E) snd_hda_controller(E) snd_hda_codec(E) >> snd_hwdep(E) snd_pcm(E) snd_seq_midi(E) snd_seq_midi_event(E) >> microcode(E) snd_ >> rawmidi(E) snd_seq(E) snd_seq_device(E) snd_timer(E) snd(E) cfg80211(E) >> soundcore(E) joydev(E) lpc_ich(E) hp_accel(E) wmi(E) serio_raw(E) >> tpm_infineon >> (E) lis3lv02d(E) amd_iommu_v2(E) input_polldev(E) video(E) tpm_tis(E) >> mac_hid(E) parport_pc(E) ppdev(E) coretemp(E) lp(E) parport(E) >> hid_generic(E) us >> bhid(E) hid(E) mmc_block(E) psmouse(E) firewire_ohci(E) ahci(E) >> libahci(E) sdhci_pci(E) e1000e(E) firewire_core(E) sdhci(E) crc_itu_t(E) >> ptp(E) pps_co >> re(E) >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320979] CPU: 1 PID: 6871 >> Comm: Sweeper thread Tainted: P D OE 3.19.0-rc3-winsoft-x64+ #10 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320981] Hardware name: >> Hewlett-Packard HP ProBook 6560b/1619, BIOS 68SCE Ver. F.50 08/04/2014 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320983] task: >> ffff88005d95b1c0 ti: ffff88018ee18000 task.ti: ffff88018ee18000 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320985] RIP: >> 0010:[] [] unlink_anon_vmas+0x1af/0x200 > Nevertheless, here we see unlink_anon_vmas, which would strongly point to > 7a3ef208e662f4b63d ("mm: prevent endless growth of anon_vma hierarchy") > and we also have another report for that. > > Still your stracktrace could be useful to help determining the cause, if you > enabled the verbose debugging. > > Thanks. > > >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320993] RSP: >> 0018:ffff88018ee1bba8 EFLAGS: 00010286 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320994] RAX: >> ffff880182a47c50 RBX: ffff880182a47c40 RCX: ffff8801b9e1d638 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320996] RDX: >> 00000000ffffffff RSI: ffff8801abb60630 RDI: ffff8801abb605f0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.320998] RBP: >> ffff88018ee1bbe8 R08: 00000000f70a2000 R09: 0000000000000000 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321000] R10: >> ffff880182a47c60 R11: ffffea00060a91c0 R12: ffff8801b9e1d628 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321001] R13: >> ffff8801b9e1d638 R14: ffff8801abb605f0 R15: ffff8801abb605f0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321003] FS: >> 0000000000000000(0000) GS:ffff88023f440000(0000) knlGS:0000000000000000 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321005] CS: 0010 DS: 002b >> ES: 002b CR0: 0000000080050033 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321007] CR2: >> 00000000f653e70c CR3: 0000000001c12000 CR4: 00000000000407e0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321008] Stack: >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321010] ffff8801b9e1d5c0 >> ffff8801abb605f0 ffff88018ee1bbe8 ffff88006a1d70b8 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321013] 00000000a3255000 >> 0000000000000000 ffff88018ee1bc58 ffff8801b9e1d5c0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321016] ffff88018ee1bc38 >> ffffffff81179b58 ffff88018ee1bc38 0000000000000000 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321019] Call Trace: >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321025] [] >> free_pgtables+0xa8/0x120 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321029] [] >> exit_mmap+0xdf/0x170 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321033] [] >> mmput+0x64/0x130 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321036] [] >> do_exit+0x26f/0xb10 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321039] [] >> do_group_exit+0x3f/0xa0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321044] [] >> get_signal+0x1d8/0x5f0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321047] [] >> do_signal+0x20/0x120 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321052] [] >> ? compat_SyS_futex+0x71/0x140 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321055] [] >> do_notify_resume+0x70/0xa0 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321059] [] >> int_signal+0x12/0x17 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321061] Code: 44 24 10 48 8d >> 50 f0 49 8d 44 24 10 49 39 c5 75 9b 48 83 c4 18 5b 41 5c 41 5d 41 5e 41 >> 5f 5d >> c3 0f 1f 40 00 e8 c3 fa ff ff eb 99 <0f> 0b 80 3d ea 8e b6 00 00 74 16 >> 49 8d 7e 08 48 89 55 c8 e8 09 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321088] RIP >> [] unlink_anon_vmas+0x1af/0x200 >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321091] RSP >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321094] ---[ end trace >> 1c9e464233c6be56 ]--- >> Jan 10 12:33:57 krzysiek-hp1 kernel: [ 3872.321096] Fixing recursive >> fault but reboot is needed! >> >> -- >> 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 >> > -- Przedsiębiorstwo Handlowo-Usługowe "WINSOFT" Krzysztof Kolasa ul. Libelta 15/3, 68-100 Żagań, NIP: 924-142-33-91, REGON: 970700380 Rachunek bankowy: PL 26 2490 0005 0000 4500 2204 2675 Alior Bank S.A. tel. +48 602-31-78-77, +48 68-414-14-47, fax +48 68-414-14-48